HARVARD UNIVERSITY

Ernst Mayr Library
of the Museum of
Comparative Zoology

•t

s?

I

Volume 122y Number 1, March 2010

Published by the
Wilson Ornithological Society

THE WILSON ORNITHOLOGICAL SOCIETY
FOUNDED 3 DECEMBER 1888

Named after ALEXANDER WILSON, the first American ornithologist.

President E. Dale Kennedy, Biology Department, Albion College, Albion, MI 49224, USA; e-mail:
dkennedy@albion.edu

First Vice-President— Robert C. Beason, P. O. Box 737, Sandusky, OH 44871, USA; e-mail:
Robert.C.Beason@gmail.com

Second Vice-President — Robert L. Curry, Department of Biology, Villanova University, 800 Lancaster
Avenue, Villanova, PA 19085, USA; e-mail: robert.curry@villanova.edu

Editor — Clait E. Braun, 5572 North Ventana Vista Road, Tucson, AZ 85750, USA; e-mail: TWILSONJO@
comcast.net

Secretary — John A. Smallwood, Department of Biology and Molecular Biology, Montclair State University,
Montclair, NJ 07043, USA; e-mail: smallwoodj@montclair.edu

Treasurer— Melinda M. Clark, 52684 HighlandDrive, South Bend, IN 46635, USA; e-mail: MClark@tcservices.biz

Elected Council Members— Robert S. Mulvihill, Timothy J. O’Connell, and Mia R. Revels (terms expire
2010); Jameson F. Chace, Sara R. Morris, and Margaret A. Voss (terms expire 2011); Mary Bomberger
Brown, Mary Garvin, and Mark S. Woodrey (terms expire 2012).

Living Past-Presidents — Pershing B. Hofslund, Douglas A. James, Jerome A. Jackson, Clait E. Braun,
Mary H. Clench, Richard C. Banks, Richard N. Conner, Keith L. Bildstein, Edward H. Burtt Jr., John
C. Kricher, William E. Davis Jr., Charles R. Blem, Doris J. W’att, and James D. Rising.

Membership dues per calendar year are: Regular, $40.00; Student, $20.00; Family, $50.00; Sustaining,
$100.00; Life memberships, $1,000.00 (payable in four installments).

The Wilson Journal of Ornithology is sent to all members not in arrears for dues.

THE WILSON JOURNAL OF ORNITHOLOGY
(formerly The Wilson Bulletin)

THE WILSON JOURNAL OF ORNITHOLOGY (ISSN 1559-4491) is published quarterly in March, June,
September, and December by the Wilson Ornithological Society, 810 East 1 0th Street, Lawrence, KS 66044-8897. The
subscription price, both in the United States and elsewhere, is $40.00 per year. Periodicals postage paid at Lawrence, KS.
POSTMASTER: Send address changes to OSNA, 5400 Bosque Boulevard, Suite 680, Waco, TX 76710.

All articles and communications for publication should be addressed to the Editor. Exchanges should be addressed
to The Josselyn Van Tyne Memorial Library, Mu.seum of Zoology, Ann Arbor, Ml 48109, USA.

Subscriptions, changes of address, and claims for undelivered copies should be sent to OSNA, 5400 Bosque
Boulevard, Suite 680, Waco, TX 76710, USA. Phone: (254) 399-9636; e-mail: business@osnabirds.org. Back issues or
single copies are available for $ 12.00 each. Most back issues of the Journal are available and may be ordered from OSNA.
Special prices will be quoted for quantity orders. All issues of the journal published before 2000 are accessible on a free
web site at the University of New Mexico library (http://elibrary.unm.edu/sora/). The site is fully searchable, and full-text
reproductions of all papers (including illustrations) are available as either PDF or DjVu files.

© Copyright 2010 by the Wilson Ornithological Society
Printed by Allen Press Inc., Lawrence, KS 66044, USA.

COVER: Wilson’s Snipe (Gallinago clelicata). Illustration by Scott Rashid.

@ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).

MCZ

LIBRARY

JUL 1 3 2010

HARVARD

UNIVERSITY

FRONTISPIECE. Evennann's Rock Ptarmigan {Lagopiis nnita evcrnianni), endemic to the Near Islands
in the western Aleutians, were reintroduced to Agattu from Attu to reestablish a breeding population.
Translocations during 2003-2006 were successful as translocated birds survived, nested, and produced
young which survived until recruitment in successive breeding seasons. Photograph of a male during the
breeding period by Steve E. Ebbert.

Wilson Journal

of Ornithology

Published by the Wilson Oniitbologiccil Society

VOL. 122, NO. 1 March 2010 PAGES 1-206

The Wilson Journal of Ornithology 122(1): 1-14, 2010

DEMOGRAPHY OF A REINTRODUCED POPULATION OF
EVERMANN’S ROCK PTARMIGAN IN THE ALEUTIAN ISLANDS

ROBB S. A. KALER,' STEVE E. EBBERT,"

CLAIT E. BRAUN,-" AND BRETT K. SANDERCOCK'-^

ABSTRACT. — We report results of a 4-year translocation effort to reestablish a breeding population of Evermann's
Rock Ptarmigan (Lagopus niuta evermanni) in the Near Islands group of the western Aleutian Archipelago. Habitat
restoration was completed by eradication of introduced foxes from Agattu Island by 1979. We captuied and moved 75
ptarmigan from Attu Island to Agattu Island during 2003-2006, and monitored 29 radio-marked females in the last 2 years
of the study. We compared the demography of newly translocated birds (n = 13) with resident birds established from
translocations in previous years (/? = 16). Mortality risk was increased by translocation and 15% of females died within 2
weeks of release at Agattu Island. All surviving females attempted to nest but initiated clutches 8 days later in the breeding
season and laid 1.5 fewer eggs per clutch than resident females. Probability of nest survival (x ± SE) was good tor both
translocated (0.72 ± 0.17) and resident females (0.50 ± 0.16), and renests were rare. Probability of brood survival was
higher among translocated (0.85 ± 0.14) than resident females (0.25 ± 0.12), partly as a result of inclement weather in
2006. Fecundity, estimated as female fledglings per breeding female, was relatively low for both translocated (0.9 ± 0.3)
and resident females (0.3 ± 0.2). No mortalities occurred among radio-marked female ptarmigan during the 10- week
breeding season, and the probability of annual survival for females in 2005-2006 was between 0.38 and 0.75.
Translocations were successful because females survived, successfully nested, and recruited offspring during the
establishment stage. Post-release monitoring provided useful demographic data in this study and should be a key component
of translocation programs for wildlife restoration. Future population surveys and additional translocations may be recjuiied
to ensure long-term viability of the reintroduced population of ptarmigan at Agattu Island. Received 31 July 2008. Accepted
16 July 2009.

Reintroductions and other translocations are an
important tool for restoring and enhancing plant
and animal populations within their native

'Division of Biology, 116 Ackert Hall, Kansas .Stale
University, Manhattan, KS 66506, USA.

-U.S. Fish and Wildlife Service, Alaska Maritime
National Wildlife Refuge, 95 Sterling Highway. Suite 1.
Homer, AK 99603, USA.

■’Grouse Inc., 5572 North Ventana Vista Road, Tucson.
AZ 85750, USA.

■‘Corresponding author: e-mail: bsanderc@ksu.edu

geographic range (Griffith et al. 1989, Snyder et
al. 1999, Ewen and Armstrong 2007). Greater
effort may be allocated to moving animals if
resources are limited, rather than to post-release
monitoring needed to evaluate project success.
However, post-release monitoring is critical
because it allows estimation of the demographic
parameters needed to evaluate status of reintro-
duced populations. Population models can then be
used to evaluate the effects of removals on source
populations, and management strategies and

7

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 1, March 2010

environmental factors which may affect the
eventual success or failure of newly established
populations (SaiTazin and Legendre 2000, Arm-
strong et al. 2007, Dimond and Armstrong 2007,
Reynolds et al. 2008).

Translocations have been widely used with
continental populations of upland gamebirds.
Release of wild-caught birds has been used to
establish new breeding populations of White-
tailed Ptarmigan {Lugopus leiiciira) (Braun et al.
1978, Hoffman and Giesen 1983, Clarke and
Johnson 1990) and Ruffed Grouse (Bonasa
umhelliis) (Moran and Palmer 1963, Kurzejeski
and Root 1988). Reintroductions have been used
to bolster declining or isolated populations of
prairie grouse {Tympanuchiis spp.) (Snyder et al.
1999, Coates et al. 2006) and Greater Sage-
Grouse (Centrocercus urophasianus) (Reese and
Connelly 1997, Baxter et al. 2008). Translocations
have also been an essential component of applied
con.servation for populations of birds on oceanic
islands, including waterfowl (Reynolds et al.
2008), flightless rails and parrots (Clout and
Craig 1995, Jamieson and Wilson 2003), and
songbirds (Komdeur 1994, Armstrong et al. 2002,
Robertson et al. 2006, Ewen and Armstrong
2007). Restoration of insular populations poses
particular challenges because life-history traits of
island and mainland vertebrates can be highly
divergent with differences in demography, behav-
ior, and morphology (Stamps and Buechner 1985,
Blondel 2000). For example, island populations of
waterfowl, grouse, and .songbirds tend to breed
later, lay smaller clutches of larger eggs, and have
higher adult survival than mainland populations
(Mercer 1967, Blondel 1985, Rohwer 1988,
Atwood et al. 1990, Blondel et al. 1992, Wiggins
et al. 1998).

The Aleutian Archipelago of western Alaska
consists of >450 isolated islands with large
populations of breeding seabirds. Historically,
the Aleutian Islands had no native terrestrial
mammals west of Umnak Island (Murie 1959).
Island populations of birds were negatively
impacted by deliberate introductions of arctic
fox (Alopex lagopus) by fur trappers between
1750 and 1940 (Maron et al. 2006). Depredation
of eggs, young, and breeding birds led to steep
population declines and local extirpation of
waterfowl, seabirds, and terrestrial birds on
islands where foxes were present (Bailey 1993).
The U.S. Fish and Wildlife Service (USFWS)
began systematic removal of foxes in 1949, as

part of recovery efforts for Aleutian Cackling
Geese {Branta hutchinsii leucopareia), as well as
seabirds of conservation concern (Springer et al.
1977, Byrd et al. 1997, Williams et al. 2003).

Rock Ptarmigan {Lagopus miita) are arctic-
breeding grouse with a Holarctic distribution
(Holder and Montgomerie 1993); eight subspecies
have been described from the Aleutian Islands
based on allopatric distributions and phenotypic
differences in plumage coloration (Holder et al.
2000, 2004). Evermann’s Rock Ptarmigan (L. m.
evermanni) is an endemic subspecies confined to
the Near Islands group in the western Aleutian
Archipelago. It is the only subspecies where males
have a black nuptial plumage (Frontispiece), and
has unique population genetics that distinguish it
from all other subspecies of Rock Ptarmigan in
the Aleutians and elsewhere (Holder et al. 2000).
Evermann’s Rock Ptarmigan were historically
present on all islands in the Near Islands group
but introduced foxes reduced their range to Attu
Island and an estimated population size of ~ 1 ,000
birds (Ebbert and Byrd 2002). Arctic fox are still
present on the islands of Nizki-Alaid and Shemya
but were eradicated from Agattu Island and Attu
Island by 1979 and 1999, respectively (Bailey
1993, Ebbert and Byrd 2002). Evermann’s Rock
Ptarmigan, due to their limited geographic range
and small population size, have been designated
birds of special management concern by the
USFWS.

We report results from a field study of a newly
established island population of Evermann’s Rock
Ptarmigan. Prior to our project, comprehensive
surveys by the Alaska Maritime National Wildlife
Refuge indicated that ptarmigan and foxes were
absent on Agattu Island. Ptarmigan may disperse
across marine waters (Zimmerman et al. 2005),
but natural recolonization across the 28 km strait
between the islands of Attu and Agattu had not
occurred in the 25-year period since foxes were
eradicated. Thus, we translocated wild-caught
ptarmigan over a 4-year period and conducted
intensive post-release monitoring at Agattu Island
in the last 2 years of the project. Ecological
studies of Rock Ptarmigan in the Aleutians have
been limited to studies of molt and diet of L. m.
gahrielsoni at Amchitka Island (Jacobsen et al.
1983, Emison and White 1988), and our project is
the first demographic study of an island popula-
tion of ptarmigan in Alaska. Our main goal was to
compare the demographic rates of translocated
and resident ptarmigan, and to evaluate the

Killer et ill. • DEMOGRAPHY OF ISLAND PTARMICiAN

3

efficacy of translocations for restoration of insular
populations of landbirds. A secondary goal was to
compare life history traits between insular and
mainland populations of Rock Ptarmigan that
might influence translocation success for subspe-
cies endemic to the Aleutian Islands.

METHODS

Study Area.— Anu (52.85° N, 173.19° E;
89,279 ha) and Agattu (52.43° N, 173.60° E;
22,474 ha) are in the Near Islands group in the
western range of the Aleutian archipelago and part
of the Alaska Maritime National Wildlife Refuge.
Attu is a mountainous island with steep hillsides
rising from sea level to elevations >600 m
(highest point: 861 m). Most of the land mass of
Agattu Island is <230 m in elevation but a
mountain range composed of seven major sub-
massifs lies along the north side and extends from
Armeria Bay eastward to Krugloi Point (highest
point: 693 m). The dominant plant community in
the Near Islands is maritime tundra because the
climate is consistently cool, wet, and windy
(Maron et al. 2006). Mean minimum and
maximum temperatures are 6.5 and 9.2° C, total
precipitation is 6.8 cm per month, and wind
velocities average 42 kph in the 3-month period
from June to August (climate data from Shemya
Island, 30 km northeast of Agattu Island). Glau-
cous-winged Gulls (Larus glaucescens) and Com-
mon Ravens (Corvus cora.x) were potential
predators of eggs and chicks, whereas Peregrine
Falcons (Falco peregrinus) and Snowy Owls
(Nyctea scandiaca) were a threat to adult
ptarmigan (Gibson and Byrd 2007). Agattu Island
does not have introduced species of rats {Rattus
spp.), which are a problem elsewhere in the
Aleutian Islands (Major et al. 2006).

Translocations. — Five translocations from Attu
Island were conducted to reintroduce EvermaniTs
Rock Ptarmigan to Agattu Island: four during
May to June, 2003-2006 and one during Septem-
ber 2003. Translocations could not be conducted
at other times of year in this remote area because
logistical support was not available and environ-
mental conditions were hazardous. Ptarmigan
were live-captured at Attu Island using noose
poles, noose carpets, and ground nets. Birds were
held for up to 48 hrs in fiberglass transport
containers and provided small pieces of melon
as a source of food and water. The 28-km strait
between the islands of Attu and Agattu required
3 hrs to traverse on the research ves.sel M/V

Tiglax. Ptarmigan were immediately released
upon arrival at Agattu Island at one ol three sites:
McDonald Cove on the east coast (2003), Karab
Cove on the south coast (2()()4-2()()5), or Binnacle
Bay on the north coast (2006). Loss of body mass
of translocated ptarmigan was measured by
subtracting mass at release from mass at capture.
We adjusted the difference in body mass by egg
mass (21 g per egg. Holder and Montgomerie
1993) in ca.ses where females laid eggs during the
holding period. Necropsies of birds that died
during handling or after release were conducted
shortly after recovery of the carcass by a wildlife
veterinarian (W. P. Taylor, Alaska Fish and Game
Department).

Field Methods. — We conducted post-release
monitoring at Agattu Island from late May to
mid-August in 2005 and 2006 to estimate the
fecundity and survival of female Evermann’s
Rock Ptarmigan (Kaler 2007). Our study sample
included newly translocated birds and resident
ptarmigan encountered at Agattu Island. The
resident population included marked birds surviv-
ing from previous translocations and their un-
marked offspring. All ptarmigan released were
individually banded and color banded with batch
marks (2003-2004) or individual band combina-
tions (2005-2006) at first capture. Yearlings and
adults were distinguished by patterns of pigmen-
tation and shape of the wing primaries (Weeden
and Watson 1967). Females were fitted with radio
transmitters attached with either a bib (2005:
15 g. Telemetry Solutions, Concord, CA, USA) or
a necklace harness (2006: 6 g, Holohil Ltd., Caip,
ON, Canada). Radio transmitters had a battery life
of 12-18 months and were equipped with
mortality switches to facilitate detection of
dropped transmitters and mortality events. Ra-
dio-marked birds were tracked with a portable
radio receiver (R2000, Advanced Telemetry
Systems, Isanti, MN, USA) and 3-element Yagi
antenna. We started radio-tracking immediately
after release and relocated females daily by radio
triangulation until nests were located. All points
were recorded in Universal Transverse Mercator
(UTM) coordinates using a hand-held Global
Positioning System receiver (Garmin GPSmap
76; Garmin International, Olathe, KS, USA).

Nests of translocated ptarmigan were located
by radiotelemetry whereas nests of resident
ptarmigan were located by searching suitable nest
habitat around roosts of males, which often acted
as sentinels (Holder and Montgomerie 1993).

4

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. I, March 2010

Unmarked females were captured with noose
poles on territories or with dip nets at nest sites,
and were marked with radio transmitters. We did
not examine the eggs if a nest was discovered
during egg-laying and returned after the clutch
had been completed. We counted the eggs, if a
nest was discovered during incubation, measured
(± 0.1 mm) length (L) and width (W) of all eggs,
and floated the eggs in a small cup of lukewarm
water to estimate stage of incubation (Westerskov
1950). Buoyancy of eggs was related to stage of
incubation as; flat on the bottom of the container
= 0 days, 45° = 5 days, 90° = 10 days, floating at
surface = 1 3 days, and a ~ 1 8 mm diameter circle
protruding above the water surface = 18 days.
The onset of hatching was detected by days 19,
20, and 21 of incubation by tapping, raised star-
shaped pips in the egg shell, and hole-pipped
eggs, respectively.

Female nest attendance was checked every 3^
days during incubation by triangulating the radio
signal at distances >30 m from the nest site. Nests
were visited every 1-2 days around the predicted
hatching date to capture young and ascertain nest
fate. We considered nests to be abandoned if a
female ceased nest attendance during egg-laying or
incubation, depredated if the eggs were destroyed
or removed before the expected hatch date, and
successful if egg shells with detached membranes
were left in the ne,st bowl after the chicks had
departed or if we captured at least one chick at or
near the nest. We counted offspring in broods < 15
days of age with binoculars from >50 m and
flushed broods 2-3 times at 15-25 days of age to
calculate survival of young after hatching.

Demographic Parameters. — Demographic per-
formance of ptarmigan is, at times, affected by
female age or annual conditions (Watson 1965,
Steen and Unander 1985, Gardarsson 1988,
Novoa et al. 2008). Our sample sizes were small
and we pooled data across age clas.ses and years to
compare the demography of translocated and
resident Rock Ptarmigan. We calculated 10
demographic parameters for breeding females
(after Sandercock et al. 2005).

(I) Date of clutch initiation was calculated by
backdating from stage of egg-laying, stage
of incubation, or the date of hatching. We
assumed that Rock Ptarmigan laid one egg/
day. started incubation on the penultimate
egg. and had an average incubation period
of 2 1 days (Holder and Montgomerie 1993).

(2) Total clutch laid (TCL) was the total number
of eggs laid in the clutch, recorded during the
first visit to the nest during incubation.

(3) Egg volume (U) was estimated from linear
measurements of ptarmigan eggs using V =
ALW", where A = 0.49 for Willow Ptarmi-
gan (L. lagopus), L = egg length, and W =
egg width (Sandercock and Pedersen 1994).
We calculated average egg volume per
clutch because eggs within a nest were not
independent observations.

(4) Nest survival until hatching (NEST) was
the probability that at least one egg hatched
and produced a chick that departed the nest.
Values of NEST < 1 included total clutch
losses due to abandonment and predation.

(5) Renesting (RENEST) was the probability
that a female laid a replacement clutch,
conditional upon loss of her first clutch.

(6) Chicks/egg laid (C/E) was the percentage of
eggs laid that eventually hatched and
produced chicks that left the nest. C/E was
calculated only for nests that survived
incubation and hatched at least one egg.
Values of C/E < 1 included partial clutch
losses due to eggs that disappeared during
incubation and eggs that survived incuba-
tion but failed to hatch.

(7) Brood survival until fledging (FLED) was
the probability that at least one chick
survived from hatching until fledging at
15-25 days of age. Eledglings completed
growth of Juvenal feathers by 2 weeks and
began to evade observers with short flights
instead of running and hiding. Values of
FLED < 1 were due to total brood failure,
which was readily ascertained by the
behavior and movements of females. If
females lost their brood, they flushed during
approach by an observer or Joined small
groups of failed breeders.

(8) Fledglings/chick hatched (F/C) was the
percentage of hatched chicks that left the
nest and survived until fledging at 15-25
days. F/C was calculated only for broods
that survived the brood-rearing period and
fledged at least one chick.

(9) Seasonal survival of females was calculated
by monitoiing radio-marked females over the
10-week breeding season. We left-censored a
2-week acclimation period to control for
potential effects of higher mortality rates
immediately following stress of capture.

Kalcr et al. • DEMOGRAPHY OF ISLAND PTARMIGAN

5

transport, release in a new environment, and
attachment of the radio transmitter.

(10) Annual survival of females was calculated
from the return rates of radio-marked birds
from 2005 to 2006. We conducted intensive
searches of territories for all radio-marked
females in 2006. Birds relocated alive in
2006 were known survivors whereas radio
transmitters recovered near scattered feath-
ers or carcass remains were considered
known mortalities. Female fate was un-
known if birds were not relocated or if radio
transmitters were recovered without accom-
panying remains.

Estimation of Fecundity. — ^We calculated a
synthetic estimate of fecundity (F) as the expected
number of female fledglings produced per breed-
ing female per year. We controlled for possible
nest losses before discovery by an observer, and
differences in renesting rates, by first calculating
the number of eggs surviving until hatching for
each breeding female (F):

F= (TCLi xNEST) +

([1 -NEST] X RENEST x TCL2 x NEST)

where TCL„ = total clutch laid, subscripts 1 and 2
are values for first nests and renests, NEST =
probability of nest survival, and RENEST =
probability of renesting after clutch loss. We
calculated fecundity (F) as:

F = F X C/E X FLED x F/C x 0.5

where F = the number of surviving eggs, C/E =
percentage of chicks hatched/egg laid, ELED =
probability of brood survival until fledging, E/C =
percentage of fledglings produced/chick leaving
the nest, and 0.5 is the proportion of females
based on a 1:1 sex ratio.

Statistical Analysis. — Comparisons of demo-
graphic parameters between translocated and
resident females were conducted using statistical
procedures in Program SAS (Version 8.1, SAS
Institute Inc., Cary, NC, USA). Sample sizes
differed among some components of fecundity
because we lacked complete information for a few
nests and losses to predators caused attrition over
the breeding season. Continuous data were
compared with two sample r-tests and frequency
data were compared with contingency tests. Post
hoc comparisons of survival rates were conducted

with Program CONTRAST (USGS, Patuxent
Wildlife Research Center, Laurel, MD, USA).
All tests were two-tailed and considered signili-
cant at a < 0.05. Means ± SE are presented.

We estimated daily survival rates of nests and
broods during the incubation and brood-rearing
periods with the nest survival procedure of
Program MARK (Version 4.1, G. C. White,
Colorado State University, Fort Collins, CO,
USA). We collected data for nests for a 63-day
exposure period from 1 June until 2 August, and tor
broods for a 38-day exposure period from 28 June
until 5 August. We created encounter histories for
each nest and brood suiwival analyses with five
types of information (Dinsmore et al. 2002): ( 1 ) the
day the nest or brood was found {k), (2) the last day
it was known to be active (/), (3) the last day it was
checked (/??), (4) the fate (/) where 0 = successful
and 1 = depredated, and (5) the number of nests or
broods with the same encounter history (/?). All
assumptions of the nest survival model were met in
this study (Kaler 2007).

We considered five candidate models in our
analyses of nest and brood survival. We expected
that translocated females might have lower nest
survival rates than resident females, and daily
survival rates might vary during the breeding
season. Survival rates of ground nests of precocial
birds might be expected to vary seasonally if
vegetative cover changes, poorly concealed nests
are selectively destroyed, or predator activity is
affected by nest densities and availability of
alternative prey (Klett and Johnson 1982, Dins-
more et al. 2002, Wilson et al. 2007). Similarly,
brood survival rates might be lowest after nest
departure and improve as chicks begin to
thermoregulate and fly. Thus, our global model
included the effects of female group (translocated
vs. resident), a linear effect of time, and the
interaction of these factors (Sgroup x linear)*
We also fit reduced models with main effects

(“^group + linear)’ sillglc faCtOrS (Sgroup’ '^tinie)' tUld a

model that constrained daily survival to be a
constant (Sconstam)- Global models are saturated in
the nest survival model, and adjustments for
overdispersion are not possible because the
variance inflation factor (r) is not identifiable
(Dinsmore et al. 2002). All models were fit to the
data using design matrices and the logit-link
function in Program MARK.

Model selection was based on an information
theoretic approach (Burnham and Anderson
1998). The model with the lowest Akaike's

6

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. I. March 2010

Information Criterion corrected for small sample
sizes (AIC(.) value was the best fit and models
with AAIC,. < 2 were equally parsimonious. We
based inference on differences in deviance in
cases where the difference in number of param-
eters was one (AA' = 1 ). We used ratios of Akaike
weights to quantify the extent of support among
models (m’,/m7).

We calculated parameter estimates with uncon-
ditional variances with the model-averaging
procedure of Program MARK to examine season-
al variation in daily survival rates. We extrapo-
lated daily survival rates using A"''"', where S =
the daily survival rate of nests or broods, and chir
— the duration of the nesting or brood-rearing
stages to obtain stage-specific estimates of the
probabilities of nest and brood survival. For
example, the duration of an eight-egg nesting
attempt was 28 days, calculated as the duration of
the egg-laying (8 days) and incubation periods (21
days) with incubation starting after laying of the
penultimate egg (—1 day). The duration of the
brood-rearing period was 20 days. Variances of
stage-specific estimates were calculated using the
delta method, following formulae and methods
described by Powell (2007).

We used parametric bootstrapping to obtain
confidence intervals for our synthetic estimates of
fecundity {F) (Gotelli and Ellison 2004). We
generated bootstrap distributions using tools of
Program MATLAB (Version 6.5, MathWorks,
Natick, MA, USA). Estimates of means and
variances were taken directly from our parameter
estimates with the exception of the probability of
renesting where the variance was estimated as
Var(/?) = p{ \ — p)ln, where p is the probability and
// is the total sample size. Total clutch laid was
modeled as draws from a nonnal distribution
whereas probabilities were modeled as draws from
beta distributions to bound draws within the range
of 0 and 1. We drew a set of seven demographic
rates at random for each bootstrap replicate,
combined the estimates to calculate fecundity,
and repeated these steps for 1,()()() bootstrap
iterations. Bootstrapping of fecundity was con-
ducted .separately for translocated, resident, and
pooled females. We generated a di.stribution of
differences between random pairs of bootstrap
values for translocated and resident females to
compare groups, and calculated a P-value by
recording the number of boot.strap differences that
were greater than the ob.served difference between
means (Gotelli and Elli.son 2004).

TABLE 1. Number of Evermann’s Rock Ptarmigan
translocated from Attu Island to Agattu Island, Aleutian
Archipelago, Alaska, in May to June 2003-2006 and
September 2003.

Year

Males

Females

Juveniles

Totals

2003

1 1

13

2

26

2004

1 1

16

0

27

2005

4

10

0

14

2006

5

3

0

8

Totals

31

42

2

75

RESULTS

We translocated 75 Evermann’s Rock Ptarmi-
gan from Attu Island to Agattu Island over a 4-
year period (2003-2006, Table 1). We monitored
reproduction and survival rates of 29 radio-
marked female ptarmigan (translocated: n — 13;
resident: n = 16) from late May to mid-August
during a 2-year radiotelemetry study (2005-
2006). The proportion of females that were adults
was higher among translocated (92%, n = 13)
than resident birds (56%, n = 16, Eisher’s Exact
test, P — 0.044), but age-classes were pooled for
analyses because sample sizes were small. Our
study sample included 28 nests (translocated: n —
10; resident: n = 18, including 2 renests) and 21
broods (translocated: n = 8; resident: // = 13).

Translocations. — ^Female ptarmigan translocated
from Attu to Agattu Island in 2005-2006 {n = 13)
were in breeding condition: 30.8% had partial
brood patches and 46.2% laid an egg in the holding
containers during transport. Female mass at capture
averaged 568 ± 16 g (/? = 31) and translocated
females lost an average of 9% of their total body
mass (x ± SE loss: -57 ± 15 g, /; = 12). Only one
mortality occuired during the holding and transport
periods in the 4-year period. A juvenile male died
during holding but we were unable to assign cause
of death at necropsy. Two mortalities occuired at
Agattu Island during post-relea.se monitoring in the
last 2 years of the .study. In both cases, radio-
marked females died within 2 weeks of release.
One female was recovered 2 days after release
390 m from the release site and necropsy indicated
that she drowned in a small tundra pool. A second
female was recovered 10 days after relea.se 914 m
from the release site; evidence at the recovery site
indicated she was killed by a raptor. Release site
had little effect on translocation success because
surviving birds quickly dispersed to settle in alpine
habitats at the northern part of Agattu Island.

Kalcr cl al. • DEM(X}RA1M 1 Y OF ISLAND PTARMIGAN

7

TABLE 2. Demographic parameters of translocated and resident Evermann's Rock Ptarmigan at Agattu Island.
Aleutian Archipelago. Alaska. 2()()5-2006. The first three parameters were calculated lor lirst nests only. Data are presented
as means ± SE {n).

Parameter

Translocated

Resident

Pooled

Statistic

P £

Date of clutch initiation (days)

16 June ± 1.0 (10)

8 June ± 1 .3 ( 16) 11

June

± 1.2 (26)

u

It

O.OOl

Total clutch laid (eggs)

6.8 ± 0.3 (10)

8.3 ± 0.2 (15)

7.7

± 0.2 (25)

InJ

11

0.001

Mean egg volume (cm’)

23.1 ± 0.6 (8)

23.3 ± 0.2 (16)

23.2

± 0.2 (24)

= 0.41

0.68

Probability of nest survival

0.724 ± 0.166 (10)

0.500 ± 0.155 (18)

0.588

± 0.1 18 (28)

= !•«

0.18

Probability of renesting

0 (2)

0.50 (4)

0.33 (6)

Fisher’s Exact

0.47

Chicks/egg laid (%)

85.7 ± 4.6 (8)

82.3 ± 5.1 (13)

83.6

± 3.5 (21)

My = -0.5

0.66

Probability of brood survival

0.849 ± 0.139 (8)

0.251 ± 0.123 (13)

0.470

± 0.1 19 (21)

f 1 = 10.4

().()()2

Fledglings/chick hatched (%)

48.5 ± LI (7)

53.9 ± 3.8 (4)

50.4

± 0.7 (1 1)

ry = 0.4

0.73

Fecundity (female fledglings/

breeding female/year)

0.9 ± 0.3

0.3 ± 0.2

0.5

± 0.2

Bootstrap

0.48

Timing of Clutch Initiation, Total Clutch Laid,
and Egg Volume. — ^All translocated females that
survived the 2-week acclimation period initiated
clutches on Agattu Island. The interval between
release and onset of egg-laying was 13.1 ± 0.9
days for translocated females (n = 10). The
average date of clutch initiation was 16 June for
translocated females, which was 8 days later than
an average date of 8 June for resident hens
(Table 2). Total clutch laid in first nests by
translocated females (6.8 eggs) was 1.5 eggs
smaller than the average clutch size of resident
hens (8.3 eggs). Mean egg volume did not differ
between resident and translocated females, and
averaged 23.2 cm^ overall.

Nest Survival, Renesting, and Chicks! Egg. — ^We
used encounter histories from 28 nests (10 translo-
cated and 18 resident, 2 females included in both
years) to estimate daily survival rates for nests.
Nests were monitored over a 63-day exposure
period (day 1 = 1 June). The minimum AIC^. model
contained a linear effect of time over the breeding
season (5|inear, Table 3). The minimum AIC,. model
had 1. 3-2.4 times greater support than the other
candidate models but all five models were equally
parsimonious (AAIQ. < 1.8).

The slope coefficient for the linear effect
indicated that nest survival declined over the
breeding season (from model 5iinear: P —0.07 ±
0.04, logit scale), particularly in the latter part of

TABLE 3. Model selection results for daily survival rates (S) of nests and broods of Evermann’s Rock Ptarmigan at
Agattu Island, Aleutian Archipelago, Alaska, 2005-2006. Model fit was described by the number of parameters [K). model
deviance (Dev), Akaike’s Information Criterion corrected for small sample size (AlC^), deviations from the minimum-AlCc
model (AAIQ.), and Akaike weights (w,). Models for daily survival (S) included comparisons of resident vs. translocated
birds (group), a seasonal effect of time (linear), factorial (X) and main effects models (+), and a reduced model with no
effects (constant). Exposure periods were 63 days for nesting (day 1 = 1 June) and 38 days for brood-reaiing (day 1 —

28 June).

Model

K

Dev

AlC,.

AAtC,

W,

Nests

2

51.5

55.5

0.00

0.300

^linear

‘^constant

1

54.0

56.0

0.54

0.229

3

50.2

56.3

0.76

0.205

■-^group + linear
c

4

48.9

57.0

1.52

0.140

"-^group X linear

C

^ group

2

53.2

57.2

1.74

0. 1 26

Broods

e

2

43.2

47.2

0.00

0.443

•-'group

•^group X linear

‘^group + linear

C

‘-'constant

•^linear

4

39.7

47.9

0.68

0.316

3

1

43.1

49.3

2.02

0.162

49.7

51.8

4.52

0.046

2

48.4

52.4

5.20

0.033

8

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. 1, March 2010

1.0

lO

2.

0

cc

03

>

>

0.6

(A) Nests

N

\

\

\

\

Translocated

Resident

\

J

3

1.0 -

(/)

0.9 ■

03

Q

0.8 ■

0.7 ■

0.6 ■

10 20 30 40 50 60

(B) Broods

1

1

\

\

\

\

0 5 10 15 20 25 30 35

Day of exposure period

LIG. 1. Daily survival rates for nests and broods of
translocated and resident female Rock Ptarmigan at Agattu
Island, Alaska, 2005-2006 (n = 27 nests and 21 broods).
Estimates were calculated by model-averaging in Program
MARK. Exposure periods were 63 days for nesting (day 1
= 1 June) and 38 days for brood-rearing (day 1 = 28 June).

the season (Fig. lA). Estimates of daily survival
rates (from model 5group) revealed that nest
survival was similar for translocated {S = 0.988
± 0.008) and resident females (5 = 0.976 ±
().01 1 ), and was high overall (from model 5con,stant:
S = 0.981 ± 0.007). Extrapolation of daily
survival rates over the duration of a nesting
attempt (27 to 28 days for 7- to 8-egg clutches)
yielded estimates of the probability of nest
survival that were 0.724 ± 0.166 for translocated
females, 0.500 ± 0. 1 55 for resident females, and
0.588 ± 0.1 18 for all females combined.

Six first nesting attempts failed because of nest
predation or abandonment. Two resident females
that renested lost their first nests 2 days before
on.set of incubation and at 12 days of incubation,
and initiated renests after intervals of 13 and 1 1
days, respectively. Clutch size of the two renests
was 5 and 6 eggs. Translocated and resident
females that did not renest lost their nests at more
advanced stages of incubation (12-23 days, n =

4). The percentage of viable eggs that hatched and
produced 1-day old chicks did not differ between
resident and translocated females, if the nest
survived incubation, and was 83.6% overall.

Brood Survival and Fledglings per Chick. — We
used encounter histories from 21 radio-marked
females that successfully hatched young (8
translocated and 13 resident, 1 female included
in both years) to estimate daily survival rates for
broods. Broods were monitored over a 34-day
exposure period (day 1 = 28 June). The minimum
AIC^. model included an effect of group with a
difference between translocated and resident birds
(5group' Table 3). Two additional models with
interactive and additive effects between group and

a linear effect of time (‘Sgroup X Unean ‘^group + linear)

were equally parsimonious or nearly so (AAIC^. <
2.02), but the minimum AIC^. model had 1.4-2. 7
times greater support.

The slope coefficient for the effect of group
(from model 5group x linear: P = 6.87 ± 6.71, logit
scale) was larger than the linear effect of time (P
= 0.53 ± 0.51) or the interaction of the two
factors (p = —0.57 ± 0.51). Brood survival of
translocated females was initially low and in-
creased over the exposure period, whereas
resident birds had moderate brood survival
(Eig. IB). Overall estimates of daily survival
rates (from model Sgroup) indicated brood survival
was higher among translocated females [S =
0.992 ± 0.008) than resident hens {S = 0.933 ±
0.022). Daily survival rates of broods for all
females combined (from model ^con.stant: ^ =
0.963 ± 0.012) were not different from the daily
survival rates of nests [S = 0.98 1 ± 0.007, x? =
1.71, P = 0.19). Extrapolation of daily survival
rates over the duration of brood-rearing (20 days)
yielded estimates of the probability of brood
survival that were 0.849 ± 0.139 for translocated
females, 0.251 ± 0.123 for resident females, and
0.470 ± 0.1 19 for all females. The percentage of
chicks that successfully fledged if the brood
survived the brood-rearing stage did not differ
between resident and translocated females, and
was 50.4% overall.

Fecundity. — We combined our estimates of
demographic parameters to calculate a synthetic
estimate of fecundity that controlled for nest
exposure before nest discovery and productivity
from renests. Fecundity tended to be higher
among translocated birds {F = 0.9 ± 0.3 female
fledgling.s/female) than residents [F = 0.3 ± 0.2),
but the difference was not significant (Table 2).

Kaler ct al. • DEMOGRAPHY Ol' ISLAND PTARMIGAN

9

Survival of Females. — We monitored seasonal
survival during the nesting and brood-rearing
period for 26 female ptarmigan (10 translocated
and 16 resident). All females survived the 10-
week breeding season and no mortalities were
observed during either 2005 or 2006. We searched
during the 2006 breeding season for 16 females
that had been radiomarked in 2005 to assess
overwinter female survival. Of the 16 females,
37.5% were detected alive, 31.3% were dead
recoveries, and 31.3% were not relocated. The
frequency of known mortalities was equivalent in
translocated (30%, /; = 10) and resident females
(33%, n = 6, Fisher’s Exact test, P = 1.0).
Surviving females had strong site fidelity and
distance between nest locations in consecutive
years averaged 123 ± 34 m (/; = 5). We
intensively searched female teixitories and our
minimum estimate of annual survival would be
0.38 if all missing females had died. However, it
missing females survived to disperse elsewhere on
the island, annual survival of females could be as
high as 0.75.

DISCUSSION

Our field study is one of the first investigations
of the effects of translocation on an insular
population of an upland gamebird, and we provide
the first estimates of demographic parameters for
any of the subspecies of Rock Ptarmigan that are
endemic to the Aleutian Archipelago. Our major
conclusions were; ( 1 ) translocations had negative
impacts on female performance with short-term
increases in female mortality, delays in the timing
of clutch initiation, and reductions in clutch size;
(2) settlement was rapid and unaffected by location
of the release sites; (3) translocated and resident
females had similar rates of fecundity and survival
during the period of population establishment; and
(4) island and mainland populations appear to have
similar demography, which may have contributed
to the success of our reintroductions.

Effects of Translocation on Fecundity. — ^The
negative impacts of translocation from Attu Island
to Agattu Island were transitory and had little net
effect on productivity. Logistics constrained us to
hold birds for up to 48 hrs, but mortality rates
were low during holding (1.4%) and comparable
to translocations of other grouse (0%, Hot! man
and Giesen 1983; 2.1%, Baxter et al. 2008) and
waterfowl (0%, Reynolds et al. 2008). Rock
Ptarmigan lost an average of 9% of body mass
during transport, and similar rates of loss have

been reported for translocated Rulfed Grouse
(10%, Kurzejeski and Root 1988) and Laysan
Ducks (Anas lay.sanensLs) (6%, Reynolds et al.
2008). The only mortalities among radio-marked
Rock Ptarmigan were two females that died
shortly after release. Mortality rates ol wildlife
are often higher immediately after translocation,
because of the physiological stress of captivity or
elevated predation risk due to lack of familiarity
with a new environment or attachment of a radio
transmitter (Musil et al. 1993, Clout and Craig
1995, Armstrong et al. 1999, Baxter et al. 2008).

Ptarmigan settled quickly into alpine habitats
during the establishment stage and all surviving
females initiated clutches shortly after release.
Future translocations of ptarmigan in the Aleutian
Islands can select release sites based on safety of
coastal landings without concerns about proximity
to suitable habitats. In other species, translocated
birds often range widely after release in search of
suitable habitat or conspecifics (Musil et al. 1993,
Amistrong et al. 1999, Coates et al. 2006), and
long distance movements can contribute to higher
mortality rates during the acclimation period
(Kurzejeski and Root 1988). Our releases may
have had greater success because post-release
movements were constrained by the limited area
of alpine habitat on Agattu Island and habitat
quality was restored to historical conditions by
eradication of introduced predators. Transloca-
tions of birds in breeding condition during the
nesting season facilitated rapid settlement and
early reproduction, similar to previous releases of
ptarmigan (Braun et al. 1978). Translocations of
juveniles during the nonbreeding season might
have less impact on a source population but could
delay establishment because juveniles must sur-
vive to the next breeding season and, at times,
have lower success as first time breeders (Kurze-
jeski and Root 1988, Sarrazin and Legendre 2000,
Robertson et al. 2006, Baxter et al. 2008,
Reynolds et al. 2008).

Translocated females delayed clutch initiation
and laid clutches that were 1 .5 eggs smaller than
nests of resident Evermann’s Rock Ptarmigan.
Reduced clutch size could have been due to
interruptions in egg-laying if translocated females
were determinate egg-layers that failed to replace
missing eggs laid in clutches at Attu Island or in
the handling crates during transit (Sandercock
1993). Alternatively, smaller clutches could have
been due to the physiological stress of transloca-
tion. White-tailed Ptarmigan breeding in years of

10

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 1. March 2010

harsh environmental conditions delay egg-laying
and take more incubation breaks to maintain body
condition (Wiebe and Martin 2000). Only resident
females renested after nest failure in our study.
However, females that failed to renest lost their
clutches at more advanced stages of incubation
when the probability of renesting was presumably
low (Robb et al. 1992). Evermann’s Rock
Ptarmigan laid fewer eggs in renests than first
nests, similar to other populations of ptarmigan
(Brodsky 1986, Cotter 1999, Sandercock et al.

2005) .

Translocated and resident ptarmigan differed
primarily in demographic parameters that were
recorded at the start of the nesting cycle. The
potential demographic impacts of differences in
timing of laying and clutch size were largely
offset by reproductive losses. Nest and brood
survival are expected to improve over the season
if vegetative cover increases or predators locate
offspring in vulnerable sites earlier in the season
(Klett and Johnson 1982, Wilson et al. 2007).
Survival of ptarmigan broods at Agattu Island
increa.sed during the breeding season, but survival
of nests was lowest during the latter part of the
exposure period. The lowest rates of daily survival
for ptarmigan nests and broods coincided with the
hatching period of Aleutian Cackling Geese and
increased foraging activity by Glaucous-winged
Gulls at inland sites. Gulls primarily hunted for
goose eggs and goslings, but incidental predation
could have reduced the fecundity of ptarmigan.
Predation could potentially be a limiting factor for
ptarmigan populations becau.se gull populations
are maintained at high levels in the Aleutian
Islands by access to refu.se from fish processing
facilities (Gib.son and Byrd 2007).

Ptarmigan young have limited ability to ther-
moregulate or evade predators in the week after
hatching (Scherini et al. 2003, Hannon and Martin

2006) . Increa.ses in late .season survival of broods
could have been due to development of homeo-
thermy and mobility among ptarmigan young or to
more favorable climatic conditions during the latter
part of the exposure period (Aulie 1976, Steen and
Unander 1985). Hatching and fledging success
were high, if ptarmigan young survived incubation
or brood-rearing, and there was no evidence of
deleterious effects caused by inbreeding depression
(Briskie and Macintosh 2004).

Synthesis of the major components of repro-
ductive effort showed that fecundity of Ever-
mann's Rock Ptarmigan was relatively low at <1

fledgling per breeding female, and that translo-
cated females did not have lower fecundity than
resident females in our reintroduced population. If
any difference occurred, it was that translocated
birds tended to have higher productivity, similar
to previous reports for Takahe (Porphyria hoch-
stetteri) (Jamieson and Wilson 2003). Some
variation in productivity could have been ex-
plained by female age class and annual variation
in demographic rates. Our sample of translocated
birds included a high proportion of adults, which
at times, have higher reproductive success than
yearling Rock Ptarmigan (Steen and Unander
1985, but see Cotter 1999, Scherini et al. 2003).
Similarly, annual fecundity of Rock Ptarmigan
can be variable and influenced by local environ-
mental factors, including food, predation, and
climate (Watson 1965, Gardarsson 1988, Novoaet
al. 2008). If conditions are favorable for several
consecutive years, ptarmigan numbers should
exhibit the exponential growth often reported for
populations established by island translocations
(Moran and Palmer 1963, Klein 1968, Komdeur
1994, Armstrong et al. 2002).

Survival Rates. — Seasonal survival of female
Evemiann’s Rock Ptarmigan during the breeding
period was 100% as no mortalities were observed.
Mortality rates during the breeding season are
higher in populations of Rock Ptarmigan where
mammalian predators and Gyrfalcons (Falco
riisticohis) are a threat (23%, Gardarsson 1988;
14%, Cotter et al. 1992; 12%, Scherini et al. 2003).
Annual survival of female Evermann’s Rock
Ptamiigan was between 0.38 and 0.75 for one
interval (2005-2006), and all losses occuired
during the fall or winter after we had departed
from Agattu Island. Raptors capable of killing adult
ptarmigan occun'ed at relatively low densities
during the breeding season (Kaler 2007). Seasonal
mortality rates of mainland populations of Willow
Ptarmigan are usually highest in autumn during
raptor migration (Smith and Willebrand 1999,
Hannon et al. 2003). Seasonal mortality rates of
Evermann’s Rock Ptarmigan may have been higher
outside of the breeding .season if Snowy Owls and
Peregrine Ealcons relied on coastal seabirds during
summer but spent more time foraging at inland
sites during fall and winter (Gibson and Byrd

2007). Our con.servative estimate of annual sur-
vival of 0.38 for Evermann’s Rock Ptarmigan is
comparable to estimates of 0.37 to 0.43 for arctic
and alpine populations of Willow Ptarmigan
(Sandercock et al. 2005).

Kidcr a al. • DEMO(:}RAI’I I Y OF ISLAND P FARMICiAN

1 1

TABLE 4. Estimates of five demographie parameters for island (1) and mainland (M) populations of Rock Ptarmigan.

Location

Pop.

t.al

Dale of clinch
initiation

Clutch

si/e

Hrobabilily of
nest survival

Probability of
brood survival

Probability of
adult survival

Source

Alaska. USA

I

5.1°

N

8 June

8.3

0.59

0.47

0.38-0.75

This study

Iceland

1

66

N

>10.4

>0.90

0.26-0.82

Gardars.son 1988

Norway

1

78

N

7.5

0.44

Steen and Unander 1985

Italy

M

46

N

10 June

6.8

0.50

0.18-0.67

Scherini et al. 2003

Scotland, UK

M

57°

N

Mid June

6.6

0.90

0.62

Watson 1965

Alaska, USA

M

65°

N

7.0

0.67

0.81

0.26-0.71

Weeden 1965

NWT, Canada

M

68

N

1 1 June

8.8

Brodsky 1986, 1988

NWT, Canada

M

69

N

9 June

8.7

0.64

0.53

0.43-0.82

Cotter 1991, 1999

Life-history Traits of Ptarmigan. — Evermann’s
Rock Ptarmigan are endemic to oceanic islands
and are a unique insular form with a black nuptial
plumage and distinct population genetics (Holder
et al. 2000, 2004). We found little evidence that
selection on demography has contributed to
development of an insular syndrome in the life-
history traits of this subspecies, despite distinct
differences in morphology. Compared to main-
land and island populations of Rock Ptarmigan
elsewhere in their Holarctic range, Evermann’s
Rock Ptarmigan had similar dates of clutch
initiation in mid-June, intermediate values for
clutch size, and comparable estimates of the
probabilities of survival for nests, broods, and
adults (Table 4). Evolution of an insular syn-
drome among the life-traits of vertebrates is
frequently attributed to differences in predation
and climate regimes between island and mainland
sites (Mercer 1967, Stamps and Buechner 1985,
Blondel 2000). Isolation reduces the diversity of
predator communities and climatic conditions can
be less variable due to proximity to coastal waters.
These two ecological factors may have had less
influence on Evermann’s Rock Ptarmigan because
the Near Island group supports several species of
avian predators and has extreme climatic condi-
tions associated with an exposed location in the
northern Pacific Ocean (Maron et al. 2006,
Gibson and Byrd 2007). Most of the geographic
variation in life-history traits of Rock Ptarmigan
and congeneric species appears to be driven by
latitudinal and altitudinal gradients in ecological
conditions in arctic and alpine environments
(Sandercock et al. 2005, Novoa et al. 2008).

CONSERVATION IMPLICATIONS

Ten features of our project have been identified
as factors predicting translocation success for

teirestrial vertebrates (Griffith et al. 1989, Snyder
et al. 1999, Ewen and Armstrong 2007). Ever-
mann’s Rock Ptarmigan are a herbivorous, native
gamebird with high reproductive potential that
breed as yearlings. We translocated multiple
batches of wild-caught birds over a 4-year period,
and birds were released into high quality habitat in
the core of their historical range which had been
restored by eradication of introduced predators.
High survival during the breeding season, suc-
cessful nesting, and recruitment of young birds
indicated successful reestablishment of an island
population. Small initial populations can risk
extinction because of demographic stochasticity,
loss of genetic diversity, and inbreeding depres-
sion. However, annual surveys in June using
playbacks of male vocalizations recorded 26-27
territorial male Evermann’s Rock Ptarmigan each
year at Agattu Island in 2008-2009, following
completion of our field study. An adaptive
management strategy based on regular surveys
and supplementary translocations will ensure the
long-term population viability of Evermann’s
Rock Ptarmigan at Agattu Island.

ACKNOWLEDGMENTS

R. B. Benter, J. K. Nooker. M. A. Schroeder, and W. P.
Taylor assisted with capture and translocation of ptarmigan
from Attii Island to Agattu Island. L. A. Kenney and G. T.
Wann were dedicated research assistants for two field
seasons at Agattu Island. Captain K. D. Bell and the crew of
the M/V Tiglax provided us with safe passage to the outer
Aleutians, and G. V. Byrd coordinated logistical support for
our field project. The LORAN Station ot the U.S. Coast
Guard at Attu Island provided housing, meals and other
support during our capture effort. D. P. Armstrong and L.
N. Ellison offered constructive reviews of the manuscript.
Funding for field work for this project was provided by the
USFWS, Alaska Maritime National Wildlife Refuge: and
the U.S. Missile Defen.se Agency. The Division of Biology
at Kansas State University provided financial support to R.

12

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. I, March 2010

S. A. Kaler and B. K. Sandercock. Capture and handling of

birds was conducted under protocols approved by the

Institutional Animal Care and Use Committee at Kansas

State University, and capture and transport permits from the

State of Alaska.

LITERATURE CITED

Armstrong. D. P., I. Castro, and R. Griffiths. 2007.
Using adaptive management to determine require-
ments of re-introduced populations: the case of the
New Zealand Hihi. Journal of Applied Ecology
44:953-962.

Armstrong, D. P., I. Castro, J. C. Alley, B. Feenstra,
AND J. K, Perrott. 1999. Mortality and behaviour of
Hihi, an endangered New Zealand honeyeater, in the
establishment phase following translocation. Biologi-
cal Conservation 89:329-339.

Armstrong, D. P., R. S. Davidson, W. J. Dimond, J. K.
Perrott, I. Castro, J. G. Ewen, R. Griffiths, and J.
Taylor. 2002. Population dynamics of reintroduced
forest birds on New Zealand islands. Journal of
Biogeography 29:609-621.

Atw'ood, J. L., M. J. Elpers, and C. T. Collins. 1990.
Survival of breeders in Santa Cruz Island and
mainland California Scrub Jay populations. Condor
92:783-788.

Aulie, A. 1976. The pectoral muscles and the development
of thermoregulation in chicks of Willow Ptarmigan
(Lagopus lagopiis). Comparative Biochemistry and
Physiology 53:343-346.

Bailey. E. 1993. Introduction of foxes to Alaskan Islands:
history, effects on avifauna, and eradication. U.S. Fish
and Wildlife Service Resource Publication 191.
Anchorage. Alaska, USA.

Baxter, R. J., J. T. Flinders, and D. L. Mitchell. 2008.
Survival, movements, and reproduction of translocated
Greater Sage-Grouse in Strawberry Valley, Utah.
Journal of Wildlife Management 72:179-186.

Blondel. j. 1985. Breeding strategies of the Blue Tit and
Coal Tit iPariis) in mainland and island Mediterranean
habitats: a comparison. Journal of Animal Ecology
54:531-556.

Blondel, J. 2000. Evolution and ecology of birds on
islands: trends and prospects. Vie et Milieu-Life and
Environment 50:205-220.

Blondel, J., R. Pradel, and J.-D. Lebreton. 1992. Low
fecundity insular Blue Tits do not survive better as
adults than high fecundity mainland ones. Journal of
Animal Ecology 61:205-213.

Braun, C. E.. D. H. Nish, and K. M. Giesen. 1978.
Release and establishment of White-tailed Ptarmigan
in Utah. Southwest Naturalist 23:661-668.

Briskie. j. V. and M. Mackintosh. 2004. Hatching failure
increases with .severity of population bottlenecks in
birds. Proceedings of the National Academy of
.Science of the United States of America 101 :558-561 .

Brodsky, L. M. 1986. Correlates and consequences of the
mating tactics of male Rock Ptarmigan (Lagopm
niiilii.s). Dissertation. Queen's University, Kingston.
Ontario. Canada.

Brodsky, L. M. 1988. Mating taetics of male Rock
Ptarmigan Lagopus minus', a conditional mating
strategy. Animal Behaviour 36:335-342.

Burnham, K. P. and D. R. Anderson. 1998. Model
selection and inference: a practical information-
theoretic approach. Springer- Verlag, New York, USA.

Byrd, G. V., E. P. Bailey, and W. Stahl. 1997.
Restoration of island populations of Black Oyster-
catchers and Pigeon Guillemots by removing intro-
dueed foxes. Colonial Waterbirds 20:253-260.

Clarke, J. A. and R. E. Johnson. 1990. Biogeography of
White-tailed Ptarmigan (Lagopus leucurus): impliea-
tions from an introdueed population in the Sierra
Nevada. Journal of Biogeography 17:649-656.

Clout, M. N. and J. L. Craig. 1995. The conservation of
critically endangered flightless birds in New Zealand.
Ibis 137 (Supplement): S181-S190.

Coates, P. S., S. J. Stiver, and D, J. Delehanty. 2006.
Using Sharp-tailed Grouse movement patterns to guide
release-site selection. Wildlife Society Bulletin
34:1376-1382.

Cotter, R. C. 1991. Population attributes and reproductive
biology of Rock Ptarmigan (Lagopus minus) in the
central Canadian arctic. Thesis. University of Alberta.
Edmonton, Canada.

Cotter, R. C. 1999. The reproductive biology of Rock
Ptarmigan (Lagopus minus) in the central Canadian
arctic. Arctic 52:23-32.

Cotter, R. C.. D. A. Boag. and C. C. Shank. 1992.
Raptor predation on Rock Ptarmigan (Lagopus minus)
in the eentral Canadian arctic. Journal of Raptor
Research 26:146-151.

Dimond, W. J. and D. P. Armstrong. 2007. Adaptive
harvesting of souree populations for transloeation: a
case study with New Zealand Robins. Conservation
Biology 21:1 14-124.

Dinsmore, S. j. G. C. White, and F. L. Knopf. 2002.
Advanced techniques for modeling avian nest survival.
Ecology 83:3476-3488.

Ebbert, S. E. and G. V. Byrd. 2002. Eradications of
invasive species to restore natural biological diversity
on Alaska Maritime National Wildlife Refuge. Pages
102-109 ill Turning the tide: the eradication of
invasive species (C. R. Veitch and M. N. Clout.
Editors). lUCN Invasive Speeies Speeialist Group,
Gland, Switzerland and Cambridge, United
Kingdom.

Emison, W. B. and C. M. White. 1988. Foods and weights
of the Roek Ptarmigan on Amchilka. Aleutian Islands.
Alaska. Great Basin Naturalist 48:533-540.

Ewen, J. G. and D. P. Armstrong. 2007. Strategic
monitoring of reintroductions in ecological restoration
programmes. Ecoscience 14:401^09.

Gardarsson, a. 1988. Cyclic population changes and
some related events in Rock Ptarmigan in Iceland.
Pages 300-329 in Adaptive strategies and population
ecology of northern grouse (A. T. Bergerud and M. W.
Gratson, Editors). University of Minnesota Press,
Minneapolis, USA.

Gibson. D. D. and G. V. Byrd. 2007. Birds of the Aleutian
Islands, Alaska. Nuttall Ornithological Club and The

Kalei- Cl at. • DEMOGRAPHY OF ISLAND PTARMIGAN

13

American Ornithologists' Union. Washington, D.C.,
USA.

Gotelli, N. J. and A. M. Ei.lison. 2004. A primer of
ecological statistics. Sinauer Associates, Sundeiiancl,
Massachusetts, USA.

Griffith. B., J. M. Scott, J. W. Carpenter, and C. Reed.
1989. Translocation as a species conservation tool:
status and strategy. Science 245:477-480.

Hannon, S. J. and K. Martin. 2006. Ecology of juvenile
grouse during the transition to adulthood. Journal of
Zoology 269:422^33.

Hannon, S. J., R. C. Gruys, and J. O. Schieck. 2003.
Differential seasonal mortality of the sexes in Willow
Ptarmigan Lagopiis lagopiis in northern British
Columbia, Canada. Wildlife Biology 9:317-326.

Hoffman, R. W. and K. M. Giesen. 1983. Demography of
an introduced population of White-tailed Ptarmigan.
Canadian Journal of Zoology 61:1758-1764.

Holder, K. and R. Montgomerie. 1993. Rock Ptarmigan
(Lagopiis miitus). The birds of North America.
Number 5 1 .

Holder, K., R. Montgomerie, and V. L. Friesen. 2000.
Glacial vicariance and historical biogeography of
Rock Ptarmigan (Lagopiis minus) in the Bering region.
Molecular Ecology 9:1265-1278.

Holder, K., R. Montgomerie, and 'V. L. Friesen. 2004.
Genetic diversity and management of Nearctic Rock
Ptarmigan (Lagopiis miitiis). Canadian Journal ot
Zoology 82:564-575.

Jacobsen Jr., E. E., C. M. White, and W. B. Emison.
1983. Molting adaptations of Rock Ptarmigan on
Amchitka Island, Alaska. Condor 85:420-426.

Jamieson, I. G. and G. C, Wilson, 2003. Immediate and
long-term effects of translocations on breeding success
in Takahe Porpliyrio lioclisterteri. Bird Conservation
International 13:299-306.

Kaler, R. S. a. 2007. Demographic measures, habitat use,
and movement of a recently reintroduced island
population of Evermann’s Rock Ptarmigan. Thesis.
Kansas State University, Manhattan, USA,

Klein, D. R. 1968. The introduction, increase, and crash of
reindeer on St. Matthew Island. Journal of Wildlife
Management 32:350-367.

Klett, a. T. and D. H. Johnson, 1982. 'Variability in nest
survival rates and implications to nesting studies. Auk
99:77-87.

Komdeur, j. 1994. Conserving the Seychelles Warbler
Acrocephuliis sechelleiisis by translocation from Cous-
in Island to the islands of Aride and Cousine.
Biological Conservation 67:143-152.

Kurzejeski, E. W. and B. G. Root. 1988. Survival of
reintroduced Ruffed Grouse in north Missouri. Journal
of Wildlife Management 52:248-252.

Major, H. L., I. L. Jone.s, G. V. Byrd, and J. C.
Williams. 2006. As.sessing the effects of introduced
Norway rats (Ralliis norvegiciis) on survival and
productivity of Least Anklets (Aeiliia piisilla). Auk
123:681-694.

Maron, J. L., J. A. Estes, D. A. Croll, E. M. Danner, S.
C, Elmendorf, and S. L, Buckelew. 2006, An
introduced predator alters Aleutian island plant

communities by thwarting nutrient subsidies. Ecolog-
ical Monographs 76:3-24.

Mercer, W. E. 1967. Ecology of an island population ol
Newfoundland Willow Ptarmigan. Technical Bulletin
Number 2. Newfoundland and Labrador Wildlile
Service, St. John's, Canada.

Moran, R. J. and W. L. Palmer. 1963. Ruffed Grouse
introductions and population trends on Michigan
islands. Journal of Wildlife Management 27:606-614.

Murie, O. j. 1959. Fauna of the Aleutian Islands and
Alaska Peninsula. North American Fauna 61:1-406.

Musil, D. D., j. W. Connelly, and K. P. Ree.se. 1993.
Movements, survival, and reproduction of Sage
Grouse translocated into central Idaho. Journal ol
Wildlife Management 57:85-91.

Novoa, C., A. Besnard, j. F. Brenot, and L. N. Ellison.
2008. Effect of weather on the reproductive rate of
Rock Ptarmigan Lagopiis miita in the eastern Pyrenees.
Ibis 150:270-278.

Powell, L. A. 2007. Approximating variance of demo-
graphic parameters using the delta method: a reference
for avian biologists. Condor 109:949-954.

Reese, K. P. and J. W. Connelly. 1997. Translocation of
Sage Grouse Centrocerciis urophasianus in North
America. Wildlife Biology 3:235-241.

Reynolds, M. H., N. E. Seavy, M. S. Vekasy, J. L.
Klavitter, and L. P. Laniawe. 2008. Translocation
and early post-release demography of endangered
Laysan Teal. Animal Conservation 11:160-168.

Robb, L. A., K. Martin, and S. J. Hannon. 1992. Spring
body condition, fecundity and survival in female
Willow Ptarmigan. Journal of Animal Ecology
61:215-233.

Robertson, H. A., 1. Karika, and E. K. Saul. 2006.
Translocation of Rarotonga Monarchs Pomarea dinii-
diata within the southern Cook Islands. Bird Conser-
vation International 16:197-215.

Rohwer, E. C. 1988. Inter- and intraspecific relationships
between egg size and clutch size in waterfowl. Auk
105:161-176.

Sandercock, B. K. 1993. Free-living Willow Ptarmigan
are determinate egg-layers. Condor 95:554-558.

Sandercock, B. K. and H. C. Pedersen. 1994. The effects
of renesting ability and nesting attempts on egg-size
variation in Willow Ptarmigan. Canadian Journal of
Zoology 72:2252-2255.

Sandercock, B. K., K. Martin, and S. J. Hannon. 2005.
Life history strategies in extreme environments:
comparative demography of arctic and alpine ptarmi-
gan. Ecology 86:2176-2186.

Sarrazin, F. and S. Legendre. 2000. Demographic-
approach to releasing adults versus young in reintro-
ductions. Conservation Biology 14:488-500.

SCHERINI, G. C., G. Tosi, and L. A. Walters. 2003. Social
behaviour, reproductive biology and breeding success
of alpine Rock Ptarmigan Lagopiis miitiis lielvericiis in
northern Italy. Ardea 91:1 1-23.

Smith. A. and T. Willebrand. 1999. Mortality causes and
survival rates of hunted and unhunted Willow Grouse.
Journal of Wildlife Management 63:722-730.

Snyder, J. W., E. C. Pelren, and J. A. Crawford. 1999.

14

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. I, March 2010

Translocation histories of prairie grouse in the United
States. Wildlife Society Bulletin 27:428-432.

Springer, P. F., G. V. Byrd, and D. W. Woolington.
1977. Reestablishing Aleutian Canada Geese. Pages
331-338 in Endangered birds: management techniques
for preserving threatened species (S. A. Temple,
Editor). University of Wisconsin Press, Madison, USA.

Stamps, J. A. and M. Buechner. 1985. The territorial
defense hypothesis and the ecology of insular
vertebrates. Quarterly Review of Biology 60:155-181.

Steen, J. B. and S. Unander. 1985. Breeding biology of
the Svalbard Rock Ptarmigan (Lagopits mutus hyper-
horeiis). Ornis Scandinavica 23:366-370.

Watson. A. 1965. A population study of ptarmigan
[Lagopus mutus) in Scotland. Journal of Animal
Ecology 34:135-172.

Weeden, R. B. 1965. Breeding density, reproductive
success, and mortality of Rock Ptarmigan at Eagle
Creek, central Alaska, from 1960 to 1964. Transac-
tions of the North American Wildlife and Natural
Resources Conference 30:336-348.

Weeden, R. B. and A. Watson. 1967. Determining the age
of Rock Ptarmigan in Alaska and Scotland. Journal of
Wildlife Management 31:825-826.

Westerskov, K. 1950. Methods for determining the age of
game bird eggs. Journal of Wildlife Management
14:56-67.

WiEBE. K. L. and K. Martin. 2000. The use of incubation
behavior to adjust avian reproductive costs after egg
laying. Behavioral Ecology and Sociobiology 48:463-
470.

Wiggins, D. A., A. P. Mgller, M. E. L. Sgrenson, and L.
A. Brand. 1998. Island biogeography and the
reproductive ecology of Great Tits Pants major.
Oecologia 115:478^82.

Williams, J. C., G. V. Byrd, and N. B. Konyukhov.
2003. Whiskered Auklets Aethia pygmaea, foxes,
humans and how to right a wrong. Marine Ornithology
31:175-180.

Wilson, S., K. Martin, and S. J. Hannon. 2007. Nest
survival patterns in Willow Ptarmigan: influence of
time, nesting stage, and female characteristics. Condor
109:377-388.

Zimmerman, C. E., N. Hillgruber, S. E. Burril, M. A. St.
Peters, and J. D. Wetzel. 2005. Offshore marine
observation of Willow Ptarmigan, including water
landings, Kuskokwim Bay, Alaska. Wilson Bulletin
117:12-14.

The Wilson Journal of Ornilhology 1 22( 1 ): I 5-22. 20 1 0

APPARENT SURVIVAL OF BREEDING WESTERN SANDPIPERS ON
THE YUKON-KUSKOKWIM RIVER DELTA, ALASKA

MATTHEW JOHNSON,'-" DANIEL R. RUTHRAUFF,^ BRIAN J. McCAFFERY,'
SUSAN M. HAIG,' AND JEFFREY R. WALTERS^

ABSTRACT. — We used 8 years of live recapture data ( 1998-2005) to estimate apparent annual survival lor male (/) =
237) and female (/; = 296) Western Sandpipers (Calidris maitri) breeding on a 36-ha plot on the Yukon-Kuskokwim River
Delta, western Alaska. Apparent annual survival (O) is the product of true survival and site fidelity, and estimates ol d) were
corrected for the probability of encounter. Overall return rates (individual returned to the study site in a subsequent season)
were lower for females (40%) than males (65%), as was th (± SE, females = 0.65 ± 0.05, males = 0.78 ± 0.03), and
encounter rate (females = 0.51 ± 0.07, males = 0.74 ± 0.04). Results differed from previous estimates of O for this
species as our estimates of O were higher for both males and females compared to estimates from another breeding site and
two nonbreeding locations. Disparity among O estimates from breeding and nonbreeding areas highlights the need to
delineate site-specific factors throughout the annual cycle that inlluence population dynamics of the Western Sandpiper.
Received 30 May 2009. Accepted 9 October 2009.

Twenty-one percent of the world’s 135 shore-
bird species are listed as species of conservation
concern by Birdlife International (Piersma et al.
1997, Johnson and Oring 2002, Sandercock 2003).
Several long-term studies have revealed popula-
tion declines among migratory shorebirds (Char-
adriiformes) in Europe (Tucker and Heath 1994,
Hagemeijer and Blair 1997), North America
(Howe et al. 1989, Momson et al. 2001, Haig et
al. 2005), and Asia (Rose and Scott 1997). The
Western Sandpiper (Calidris mauri) is considered
a species of high conservation concern despite
being one of the most abundant shorebirds in the
Western Hemisphere (Bishop et al. 2000, Morri-
son et al. 2000, Brown et al. 2001). Declining
numbers during spring (Butler and Lemon 2001)
and fall migration surveys (Neil 1992, Butler and
Lemon 2001) at staging areas in British Columbia
and Texas coupled with a limited breeding
distribution and threats to nonbreeding habitat
have led to this conservation assessment (Brown
et al. 2001, Fernandez et al. 2006). However,
reported declines in numbers at staging sites may
be caused by shifts in migration patterns and not

' U.S. Geological Survey, Forest and Rangeland Ecosys-
tem Science Center, 3200 Southwest .lefferson Way,
Corvallis, OR 97331, USA.

^U.S. Geological Survey, Alaska Science Center, 4210
University Drive, Anchorage, AK 99508, USA,

^U.S. Fish and Wildlife Service, Yukon Delta National
Wildlife Refuge, P. O. Box 346, Bethel, AK 99559, USA.

■* Department of Biological Sciences, Virginia Polytech-
nic Institute and State University, Blacksburg, VA 24061,
USA,

■‘’Corresponding author;
e-mail: matthewJohnson@usgs.gov

impacts of demographic rates on population
growth (Ydenberg et al. 2004).

The life history of migratory shorebirds is
generally characterized by delayed maturity, low
productivity, and relatively high adult survivor-
ship (Evans and Pienkowski 1984, Piersma and
Baker 2000). Robust estimation of demographic
parameters is central to understanding population
dynamics of avian species (Eberhardt 1985, Lande
1988, Clobert and Lebreton 1991). Estimation of
survival rate is particularly important because the
persistence of any animal population is the result
of a balance between recruitment of new breeders
into the population via reproduction and immi-
gration, and losses due to the emigration and
mortality of established breeders (Lebreton et al.
1993, Lieske et al. 2000). Estimation of these
parameters and identification of ecological factors
that affect avian populations requires long-term
studies of marked individuals (Lebreton and
North 1993).

A better understanding of shorebird demogra-
phy would facilitate conservation efforts; howev-
er, population models have been developed for
only a few shorebird species. Sensitivity analyses
have found that adult survival often has the
highest elasticity value and potentially the greatest
impact on population growth rates (Hill and
Carter 1991, Hitchcock and Gratto-Trevor 1997,
Reed et al. 1998, Plissner and Haig 2000, Larson
et al. 2002, Sandercock 2003). Effective conser-
vation of migratory shorebirds requires reliable
estimates of annual survival rates, but estimates
are available for <10% of shorebird species
worldwide (Sandercock 2003).

15

16

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. I. March 2010

Western Sandpipers breed along the coasts of
western Alaska and eastern Siberia and are
distributed over a large geographic area in
nonbreeding areas ranging on the Pacific Coast
from California to Peru, and on the Atlantic Coast
from New Jersey to Suriname (Wilson 1994).
Male Western Sandpipers spend the nonbreeding
season at more northerly latitudes compared to
females (Page et al. 1972, Harrington and Haase
1994, Buenrostro et al. 1999, Nebel et al. 2002),
and males generally precede females north during
spring migration (Holmes 1971, Senner et al.
1981, Butler et al. 1987) with the opposite pattern
during fall migration (Butler et al. 1987, Yden-
berg et al. 2005). Comparative studies within their
nonbreeding distribution have revealed divergent
life history strategies (O’Hara et al. 2005).
Juveniles and females from the southern portion
of the nonbreeding range are more likely to
remain at nonbreeding sites during their first
breeding sea.son compared to adults and males at
northerly sites. This suggests Western Sandpiper
life history varies with first-year females favoring
increased survival and first-year males emphasiz-
ing initial breeding opportunities (O’Hara et al.
2005). Given the species’ expansive geographic
distribution and age-and gender-specific variation
in life history strategies, it is likely Western
Sandpipers likewise exhibit region-, age-, and
gender-specific variation in survival.

Our objective was to produce estimates of
apparent annual survival at a representative
breeding site in the middle of the breeding range
of Western Sandpipers in western Alaska. We
used 8 years of live recapture data to examine
whether apparent annual survival rate varied
between males and females at our study site,
and whether apparent survival varied significantly
during the course of this study. We compare our
estimates with those from a more northerly
breeding site and two nonbreeding sites.

METHODS

Study Area and Field Methods. — We studied
breeding Western Sandpipers at the Yukon Delta
National Wildlife Refuge’s Kanaryarmiut Field
Station, Yukon-Kuskokwim River Delta, Alaska
(61 22' N, 165 07' W). We u.sed a 36-ha study
plot in dry upland tundra habitat along the
Kuyungsik River (lowland moist low .scrub com-
munity, Jorgenson and Ely 2001) to address our
objectives. The study site supports an average of 80
breeding pairs annually (Johnson et al. 2005). Two

to four observers searched this study plot daily
from early May through late July from 1998 to
2005 for banded birds, nests, and broods. We
banded each adult at the nest with a U.S.
Geological Survey identification band and unique
color combinations (3 UV-stable color bands/bird).
We recaptured all birds when band loss or
discoloration occumed (2% of all banded birds),
indentified the individual by its U.S. Geological
Survey identification band, and re-banded the bird
with new color bands. Adults were classified as
male or female by culmen length during banding
(>93% of all birds; Page and Fearis 1971, Carter
1984). Birds were classified as males or females
via behavior (courtship displays, copulation posi-
tion) and comparison with their mate (females >
males) when culmen length was inconclusive (24-
25 mm). Locations of nests and banded individuals
were mapped using geographic information system
layers in ArcGIS (ESRI 2007) and nests were
monitored through hatch, predation, or abandon-
ment. It is possible to differentiate first-year
Western Sandpipers that hatched in the previous
breeding season from birds that are >2 years old
based on plumage characteri.stics (Sandercock et al.
1999). However, our sample of first-year birds was
limited (24% of females and 17% of males banded
were first-year birds), and previous attempts to
estimate apparent survival in this population
revealed unacceptable overdispersion when fitting
age-specific models to the data set (r > 8; M.
Johnson, unpubl. data). We chose not to estimate
apparent survival by age class (i.e., first-year and
>2 year old birds), and instead estimated apparent
survival for all breeding adults.

Statistical Analyses. — ^We used Cormack-Jolly-
Seber mark-recapture models to estimate apparent
annual survival (4)) coirected for probability of
encounter (p). Apparent survival rate was the
probability that a bird alive in breeding season /
survived until the following breeding season (/ -t-
1 ) and returned to our study plot. Encounter
probability was the probability that we detected a
bird given that it was alive. Losses from the
breeding population may result from permanent
emigration or mortality, and are the complement
ol apparent survival. We conducted mark-recap-
ture analysis following methodologies outlined by
Burnham and Anderson (1998) using Program
MARK (Version 4.2, Gooch and White 2004).

We incorporated gender (,vc.v) and time-depen-
dence (/) as variables in our global model of O and
p because both parameters are known to vary

Johnson cl ul. • WESTHRN SANDPIPER SEIRVIVAI.

17

between males and females and aeross years in
many shorebird speeies (Sandereock 2003). We
also modeled lemporal variation as a linear (7 )
and quadratie (TI) trend across years. Several
nonexckisive factors may lead to lower apparent
survival rates during the first interval after initial
capture in many bird populations (Sandereock and
Jaramillo 2002). Thus, we modeled (1> for the
interval following initial capture (<!)') and all
subsequent intervals (d)-') separately for each
gender (time-since-marking, or 2 age-class mod-
el). We did not expect p to vary between first and
subsequent time intervals and modeled p as a
function of sex and t. Thus, our starting global
model was Pscs*i-

We used two methods to assess the fit of our
global model to the sandpiper survival data; a
parametric bootstrap goodness-of-fit test, and the
median c procedure in Program MARK (Cooch
and White 2004). The parametric bootstrap
goodness-of-fit procedure generated a bootstrap
distribution of expected deviances through simu-
lation (/? = 1,000 replicates) and calculated a
variance inflation factor (r) as the observed
deviance divided by the mean expected deviance
from the bootstrap distribution (Burnham and
Anderson 1998). We simulated sandpiper survival
data in the median c procedure with a range of
deviances (n = 1,000 replicates), and performed a
logistic regression to estimate the median c value
of the simulated data (Cooch and White 2004).
Moderate amounts of overdispersion are common
in mark-recapture analyses with c values of 1-3
indicating an acceptable global model fit (Lebre-
ton et al. 1992).

We fit reduced models with fewer parameters
to the sandpiper data set after examining the fit of
our starting global model. We began the model
selection process by modeling encounter proba-
bilities followed by apparent survival probabili-
ties. We contrasted models with one and two age
classes (d) vs. d)', (P’^) after modeling p. We
modeled probabilities (O and p) by initially
dropping parameters from a factorial model to
create additive models, and finally examined
single factor and constant models. We assessed
model fit with quasi-Akaike’s Information Crite-
rion, adjusted for small sample sizes (QAIC,,
Burnham and Anderson 1998). Models where
AQAIC,. was <2 from the best fit model (AQAIC,
= 0) were considered equally parsimonious, and
Akaike weights (w,) were used to evaluate the
relative likelihood of a model within the set of

candidate models. The ratio ol QAIC, weights
between two candidate models was used to
quantify the fit of a particular model to the data
relative to other models (Burnham and Anderson
1998). We applied the variance components
procedure in Program MARK to obtain overall
estimates of apparent survival and encounter
probability. We used the model averaging proce-
dure in Program MARK to obtain annual
estimates of apparent survival. Finally, we
calculated the distance between annual nest
attempts for site-faithful males, females, and
reuniting pairs, and used analysis of variance by
ranks to contrast inter-annual dispersal distances
for these three groups (Kruskal and Wallis 1952).
We report means ± SE.

RESULTS

We individually marked 533 breeding Western
Sandpipers (296 females, 237 males) in the first 7
of 8 years of study. Overall return rates (individual
returned to the study site in a subsequent season)
were lower for females (40%) than males (65%).
Both goodness-of-fit methods detected minor
overdispersion when we fit our global model to
the sandpiper survival data set (parametric boot-
strap goodness-of-fit procedure c = 2.9; median c
procedure c = 1 .4). These two estimates differ, but
both indicated that our global model was a
satisfactory starting point (c < 3.0), and we chose
the more conservative estimate of overdispersion (c
= 2.9) and continued with model selection. We
systematically varied r between the two estimates
after completing model selection and found no
differences among the top performing models,
regardless of which c value was used.

Model selection based on QAIC, indicated the
best-fit model was where survival and encounter
probabilities differed between males and females
but were constant over time (dCcv- Table I ).
Apparent survival rale of females (0.65 ± 0.05)
was lower than lor males (0.78 — 0.03) as was
encounter rate (females = 0.51 ± 0.07, males =
0.74 ± 0.04). Model selection revealed moderate
support for a negative linear time trend model that
varied between males and females + r- PseP-
AQAIC, = 1.3; iggs!~2()04 ~ ^^-^2 — 0.75,

I99S-2004 = 0.69 - 0.60), but all 95%
confidence intervals surrounding point estimates
of annual apparent survival were overlapping.
There also was support for a time-since-marking
(2 age-class) model with variation between males
and females (tl>' v,.,v /2«-a- AQAIC,. = 1 .6; (I)',

18

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. I. March 2010

TABLE 1. Mark-recapture models for annual estimates
ot apparent survival for adult Western Sandpipers breeding
near Kanaryarmiut Field Station, Yukon-Kuskokwim River
Delta. Alaska. 1998-2005.

Model statistics'*

Model structure^

K

Deviance

AQAIC,.

W,

Psex

4

122.6

0.0

0.32

Psex

5

121.9

1.3

0.16

^ sex ^ sex' Psex

6

120.2

1.6

0.14

+ TT- Psex

6

120.8

2.2

0.1 1

4)'.ve.v Psex

5

123.1

2.5

0.09

Psex

3

127.6

3.0

0.07

4>/ . Psex

4

127.2

4.6

0.03

4’7T- Psex

5

126.5

5.9

0.02

Pc

3

130.8

6.2

0.01

■st'A + / ^ sex + /' Pse.\

, 15

110.9

10.7

0.002

Psex*t

26

105.3

28.1

< 0.001

sex*i ^ sex*h Psex"^!

37

97.2

43.5

< 0.001

We described model fit by the number of parameters {K). deviance, and the
difference in quasi-Akaike's Information Criterion (AQAIC, ) from the best-fit
model. QAIC, values were calculated using a variance infiation factor of 2.9.
We present models with moderate support (Akaike weight w, ^ 0.01) in order
of relative fit to the best-fit model (i.e., AQAIC, = 0). followed by a two age-
cla.ss (time-since-marking) model with additive effects of sex and time, a
standard Cormack-Jolly-Seber model that did not control for time since
marking, and the starting global model.

Model factors included; O' = apparent survival during the first-year post
banding. O*' = apparent .survival during subsequent years. O = annual
apparent .suiwival in models lacking age structure, c = constant, = gender, t
= time or annual variation. T — linear time trend. 7T = quadratic lime trend. *
= a factorial model, and -i- = an additive model.

males = 0.72 ± 0.06, females = 0.57 ± 0.08;

males = 0.81 ± 0.04, females = 0.69 ±
0.06). However, the best-fit model p^ex) had
>2 times the support of both of these models.
Model selection did not support annual variation
(f) in apparent survival rates (AQAIC,. = 7.3);
however, we examined the point estimates of
annual apparent survival from our global model
^^^'^scx*h Psex*i) because time dependence
is often not supported in sparse data .sets. Female
apparent survival was lower in 2001 compared to
other years in the interval after first capture (d)')
and later intervals Fig. 1).

Inter-annual dispersal distances varied among
site-faithful females, males, and reuniting pairs
{X~2 ~ 54.2, P < 0.001). Inter-annual dispersal
distances were smallest among reuniting pairs
(56.0 ± 12.3 m, ii = 47 pairs). Inter-annual
dispersal distances were greater among females
(220.9 ± 27.7 m, n = 66) compared to males
(86.9 ± 8.5 m, n = 174) for site-faithful
individuals that did not reunite with their prior
mate. We surveyed 29 plots (8.3-ha/plot) annually
within a 29 km^ area surrounding our study site
(15-31 May, 2()()4-2()()5). We did not detect any

LJJ

w

B

ro

7c

>

w

c

0)

CO

Q.

Q.

CO

To

3

c

c

<

1.0 1
0.8 -
0.6 ■
0.4 -
0.2 -
0.0 -■

Subsequent intervals

T I "V I T" • " ■ • ' "»

1998-1999 1999-2000 2000-2001 2001-2002 2002-2003 2003-2004

Transition period (year)

FIG. 1. Annual apparent survival rates (mean ± SE) of
adult Western Sandpipers (296 females, 237 males)
breeding near Kanaryarmiut Field Station, Yukon-Kuskok-
wim River Delta, Alaska (1998-2005) in the interval after
first capture and subsequent intervals (males = squares
connected with solid lines, females = circles connected
with broken lines). Annual estimates were calculated via
model averaging, and estimates for interval 2004-2005
were not identifiable because of the time dependence in
apparent annual survival and the probability of encounter.

color-marked birds during 50 hrs of observation
in the sun'ounding region. No banded birds that
went undetected in our surveys were encountered
by concurrent research projects being conducted
in the surrounding area (2003-2005).

DISCUSSION

We used Cormack-Jolly-Seber mark-recapture
models to estimate apparent annual survival of
breeding Western Sandpipers on the Yukon-
Kuskokwim River Delta, Alaska. These estimates
complement those from other parts of the
breeding and nonbreeding range by identifying
potential temporal and spatial patterns in survival
rates. Apparent survival of female Western
Sandpipers was lower than males, and hierarchical
model selection results indicated that apparent
survival did not vary annually. Previous estimates
ot apparent survival for breeding Western Sand-
pipers also were higher for males than females
(Sandercock et al. 2000), hut our estimates were
higher for both males and females (Table 2). Two
studies also have used mark-recapture techniques
to estimate the apparent annual survival of
Western Sandpipers in nonbreeding areas. Juve-
nile and adult males in Baja California had
relatively low apparent annual survival (Fernan-

.Johnson cl a/. • WESTERN SANDPIPER SURVIVAL.

19

TABLE 2. Annual estimates of apparent survival for Western
sandpipers from studies that used mark-recapture statistics for live

(CuUdris nuniri)
encounter data.

and Sent i|)al mated (C. semipalmuiiis)

Species

Range-'

Age"

GendeL

Apparent .survival''

a birds

# yeans'’ Reference

Western Sandpiper

B

A

F/M

0.57

236

4

Sandercock et al. 2000

NB

J

M

0.49

139

4

Fernandez et al. 2003

A

M

0.45

1 17

NB

J/A

F

0.62

1 ,990

3

Fernandez et al. 2004

M

0.54

1,383

B

A

F

0.65

296

8

This study

M

0.78

237

Semipalmated

B

A

F

0.56

249

8

Sandercock and Gratto-Trevor

Sandpiper

M

0.61

237

1997

B

A

F

0.59

122

4

Sandercock et al. 2000

M

0.73

108

NB

A

F/M

0.65

215

4

Rice et al. 2007

J

F/M

0.55

319

4

“ Range: B = breeding area, NB = nonbreeding area.

Age; J = juvenile.s SI year of age. A = adulls >1 year of age.

'■ Gender: F = females, M = males.

^ .Apparent survival: estimate from best-fit model, or average of annual estimates (i.e.. Sandercock and Gratto-Trevor 1997).
^ # years = duration of study in years.

dez et al. 2003, Table 2), and females had higher
rates of apparent annual survival than males in
Chitre, Panama (Fernandez et al. 2004, Table 2).

Our results are comparable to those from
breeding areas for a sympatric Calidris species
with similar life history characteristics. Semipal-
mated Sandpipers (C. pusilla) are small-bodied
shorebirds that breed in the arctic and sub-arctic
that also undertake long distance migrations and
distribute themselves over a relatively large
nonbreeding range (Gratto-Trevor 1992). Male
Semipalmated Sandpipers near Nome, Alaska had
higher apparent survival rates than females (Ta-
ble 2) and both rates were similar to our estimates
for Western Sandpipers. Male Semipalmated
Sandpipers breeding at La Perouse Bay, Manitoba,
Canada had higher apparent survival rates than
females (Table 2). These estimates were lower
than ours for Western Sandpipers; however, there
was considerable annual variation in that study
(1980-1987, males 0.53-0.74, females 0.43-0.71;
Sandercock and Gratto-Trevor 1997).

Differences between estimates of apparent
survival for Western Sandpipers in this study
and those from Nome may reflect temporal and/or
spatial variation in survival rates. Sandercock et
al. (2000) conducted their study from 1993 to
1996 at a site near Nome, Alaska. It is ditlicult to
assess whether differences are due .solely to
temporal variation because the two studies were
not contemporary. Survival rates vary over time in
other species (e.g.. Lesser Snow Goose \Anser

caerulescens], Francis et al. 1992; Mallard {Anas
platyrhynchos], Franklin et al. 2002; Barn Owl
[Tyto alba], Altwegg et al. 2003), and this
phenomenon may occur with Western Sandpipers.
However, model selection results did not indicate
apparent annual survival varied during the course
of our study (Table 1).

Other researchers have demonstrated that
demographic parameters including survival rates
often vary spatially (e.g., Nichols et al. 1990,
Kanyamibwa et al. 1993, Franklin et al. 2002,
Pearce et al. 2005), and higher estimates of
apparent survival at our study site may reflect
site-specific conditions that promote higher sur-
vival compared to Nome, which is ~365 km and
4° latitude further north than our study site.
Western Sandpipers breeding at Nome experience
a shorter breeding season with longer periods of
inclement weather compared to birds at our study
site (Johnson and Walters 2008). A more severe
climate coupled with longer annual migration
distances may predispose birds at Nome to
decreased survival.

Annual variation in apparent survival did not
occur in our study population, even though it has
been reported in other studies of Western
Sandpipers that were of considerably shorter
duration (4 years, Sandercock et al. 2000; 3 years,
Fernandez et al. 2003). Examination of point
estimates from our global model revealed that
female apparent survival was lower in 2001
compared to other years (Fig. 1 ). We are not able

20

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. I, March 2010

to account for this variation, but lower female
survival in 2001 may partially explain why overall
estimates of apparent survival were lower for
temales compared to males.

Whether lower apparent annual survival among
temale Western Sandpipers represents variation in
true survival or sexual variation in breeding site
fidelity is unknown because we are not able to
differentiate the relative influence of mortality
versus permanent emigration in our estimation of
apparent annual survival. Males generally exhibit
greater breeding site fidelity than females among
monogamous sandpipers (Oring and Lank 1982,
Piersma et al. 1996). We cannot conclude that
temale annual survivorship is lower than for
males in our study population because females
may exhibit lower breeding site fidelity than
males. However, average inter-annual dispersal
distances were relatively small in relation to size
of our study site, and similar to those reported
near Nome (median inter-annual dispersal dis-
tance near Nome: reuniting pairs = 38 m, males
= 48 m, females = 157 m). Further, no banded
birds that went undetected in our surveys were
encountered in the surrounding area. We conclude
that local permanent emigration was not common
at our study site.

Greater inter-annual dispersal distances among
site-faithful females may indicate that females also
emigrate from our study site at a higher rate than
males. However, local movement patterns may or
may not be indicative of larger scale dispersal
patterns. Encounter probabilities were lower for
females during our study, which may indicate
temporary emigration (i.e., attempting reproduc-
tion at another location for I or more intervening
years) is more common among females, or there
are behavioral differences that result in lower
detection rates for females. Males are more
conspicuous early in the breeding sea.son becau.se
they engage in conspicuous epigamic displays
daily, whereas females are more cryptic spending
greater amounts of time in concealed locations,
such as wet meadows (Lanctot et al. 2()()()). Further,
female encounter rates may be lower if only a
portion of tho.se individuals that permanently
emigrate still pa.ss through our study site before
settling at other locales, or if .some individuals pa.ss
through quickly and escape detection.

Western Sandpipers are a species of high
conservation concern yet the underlying causes
warranting this classification are unclear. Disparity
between and among apparent survival estimates

from breeding and nonbreeding areas highlights the
need to delineate site-specific factors throughout
the annual cycle that influence population dynam-
ics of Western Sandpipers. The closely related
Semipalmated Sandpiper is likely to have compa-
rable population demographics to the Western
Sandpiper (Fernandez et al. 2003). Our estimates of
apparent annual survival for Western Sandpipers
are similar to true annual survival rates required to
sustain a breeding population of Semipalmated
Sandpipers (—0.80, Hitchcock and Gratto-Trevor
1997). However, observed population declines may
be misleading. Reported declines in numbers at
staging sites may be caused by a shift in migration
patterns (Ydenberg et al. 2004) and not by a
decreasing population, further underscoring the
necessity of additional research to elucidate these
questions. Effective modeling and predicting
population trends requires accurate estimation of
vital rates encompassing the range of natural
variability. Our estimates of adult survival of
Western Sandpipers augment and complement
previous estimates by providing current estimates
from a centrally located site within the species’
breeding range, and serve to better inform future
con.servation efforts.

ACKNOWLEDGMENTS

We thank the .staff of the Yukon Delta National Wildlife
Reluge for supporting this research. Financial support was
received from the U.S. Fish and Wildlife Service (Yukon
Delta National Wildlife Refuge), the Harold F. Bailey Fund
at Virginia Tech, and the U.S. Geological Service (Forest
and Rangeland Ecosystem Science Center). We thank T. L.
Booms, J. R. Conklin, Z. M. Fairbanks, C. E. Fitzpatrick. L.
F. Hamblin. P. N. Laver, B. L. Johnson, Silke Nebel, A. C.
Niehaus. L. W. Oring, D. J. Rizzolo, and M. K. Spies for
assistance in the tield. G. J. Fernandez, D. B. Lank. C. A.
Nicolai, C. M. Taylor, B. K. Sandercock, and two
anonymous reviewers provided assistance and comments
that greatly improved this manuscript. Any use of trade, or
tirm names is tor descriptive purposes only and does not
imply endorsement by the U.S. Government.

LITERATURE CITED

Altwegg, R., a. Rouun, M. Ke.stkniiolz, and L. Jenni.
2003. Variation and covariation in survival, dispersal,
and population size in Barn Owls Tylo alha. Journal of
Animal Ecology 72:391-399.

Bishop, M. A.. P. M. Meyers, and P. E. Mc Neley. 2()()().

A method to estimate migrant shorebird numbers on
the Copper River Delta, Alaska. Journal of Field
Ornithology 71:627-637.

Brown, S., C. Hickey, B. Harrington, and R. Giee.
2001. flic U.S. Shorebird Conservation Plan. Second

Johnson d al. • WESTERN SANDPIPER SURVIVAE

21

Edition. Manomcl Center for Conservation Seienees,
Manomet, Massachusetts. USA.

Buenorostro, M. a., N. Warnock, and H. dela Clieva.
1999. Wintering Western Sandpipers Caliciris nuntri at
Estero de Pnnta Banda, Baja Calilornia, Mexico.
Wader Study Group Bulletin 88:59-63.

Burnham, K, P. and D. R. Anderson. 1998. Model
selection and inl'erence: a practical information-
theoretic approach. Springer-Verlag, New York, USA.

Butler. R. W. and M. J. E. Lemon. 2001. Trends in
abundance of Western and Least sandpipers migrating
through southern British Columbia. Bird Trends 8:36-
38.

Butler, R. W., G. W. Kaiser, and G. E. J. Smith. 1987.
Migration ehronology, length of stay, sex ratio, and
weight of Western Sandpipers (Calidris niauri) on the
south coast of British Columbia. Journal of Field
Ornithology 58:103-1 1 1.

Carter, R. V. 1984. A morphometric comparison of
Western and Semipalmated sandpipers. Wilson Bulle-
tin 96:277-286.

Clobert, j. and J.-D. Lebreton. 1991. Estimation of
demographic parameters in bird populations. Pages
75-104 in Bird population studies (C. M. Perrins, J.-D.
Lebreton, and G. J. M. Hirons, Editors). Oxford
University Press, Oxford, United Kingdom.

COOCH, E. AND G. White, 2004. Program MARK: a gentle
introduction, http://www.phidot.org/software/mark/
docs/book/

Eberhardt, L. L. 1985. Assessing the dynamics of wild
populations. Journal of Wildlife Management 49:997-
1012.

ESRI Inc. 2007. ArcGIS. Version 9.2. Earth Systems
Research Institute Inc., Redlands, California, USA.

Evans, P. R. and M. W. Pienkowski, 1984. Population
dynamics of shorebirds. Pages 83-123 in Behaviour of
marine animals (J. Burger and B. L. Olla, Editors).
Volume 5. Plenum Press, New York, USA.

FernAndez, G.. P. D. O’Hara, and D. B. Lank. 2004.
Tropical and subtropical Western Sandpipers {Calidris
mauri) differ in life history strategies. Ornitologia
Neotropical 15:385-394.

Fernandez, G., H. de la Cueva, N. Warnock, and D. B.
Lank. 2003. Apparent survival rates of Western
Sandpipers {Calidris mauri) wintering in northwest
Baja California, Mexico. Auk 120:55-61.

Fernandez, G., N. Warnock, D. L. Lank, and J. B.
Buchanan. 2006. Version 1.1. Conservation Plan for
the Western Sandpiper. Manomet Center for Conser-
vation Science, Manomet, Massachusetts, USA.

Francis, C. M., M. H. Richards, F. Cooke, and R. F.
Rockwell. 1992. Long-term changes in survival rates
of Lesser Snow Geese. Ecology 73:1346-1362.

Franklin, A. B., D. R. Anderson, and K. P. Burnham.
2002. Estimation of long-term trends and variation in
avian survival probabilities using random effects
models. Journal of Applied Statistics 29:267-287.

Gratto-Trevor, C. L. 1992. Semipalmated Sandpiper
{Calidris pnsilla). The birds of North America.
Number 6.

Haig, S. M., C. L. Ferland, F. J. Cuthbert, .1.

Dingledine, j. P. Goossen, A. Hechi, and N.
McPhillirs. 2005. A complete species census and
evidence for regional declines in Piping Plovers.
Journal of Wildlife Management 69:160-173.

Hagemeijer. E. j. M. and M. J. Blair. 1997. The EBCC
Atlas of European breeding birds: their distribution
and abundance. T. & A. D. Poyser, London, United
Kingdom.

Harrington, B. and B. Haase. 1994. Latitudinal differ-
ences in sex ratios among nonbreeding Western
Sandpipers in Puerto Rico and Ecuador. Southwestern
Naturalist 39:188-189.

Hill, D. and N. Carter. 1991. An empirical simulation of
an Avocet Recurvirostra avosetta population. Ornis
Scandinavica 22:65-72.

Hitchcock, C. and C. L. Gratto-Trevor. 1997. Diagno.s-
ing a shorebird local population decline with a stage-
structured population model. Ecology 78:522-534.

Holmes, R. T. 1971. Density, habitat and the mating
system of the Western Sandpiper {Calidris mauri).
Oecologia 7:191-208.

Howe, M. A., P. H. Geissler, and B. A. Harrington.
1989. Population trends of North Ameriean shorebirds
based on the International Shorebird Survey. Biolog-
ical Conservation 49:185-199.

Johnson, M. and L. W. Oring. 2002. Are nest exclosures
an effective tool in plover conservation? Waterbirds
25:184-190.

Johnson, M. and J. R. Walters. 2008. Effects of mate and
site fidelity on nest survival of Western Sandpipers.
Auk 125:76-86.

Johnson, M., S. E. Jamieson, and B. J. McCaffery. 2005.
Arctic breeding conditions locality report: Kanaryar-
miut Field Station, Yukon Delta National Wildlife
Refuge. Arctic Birds 7:26-28.

Jorgenson, T. and C. Ely. 2001 . Topography and Hooding
of coastal ecosystems on the Yukon-Kuskokwim
Delta, Alaska: implications for sea-level rise. Journal
of Coastal Research 17:124-136.

Kanyamibwa, S., F. Bairlein, and a. Schierer. 1993.
Comparison of survival rates between populations of
the White Stork Ciconia ciconia in central Europe.
Ornis Scandinavica 24:297-302.

Kruskal, W. H. and W. a. Wallis. 1952. Use of ranks in
one-criterion analysis of variance. Journal of the
American Statistical Association 47:583-621.

Lanctot, R. B., B. K. Sandercock, and B. Kempenaers.
2000. Do male breeding displays function to attract
mates or defend territories? The explanatory role of
mate and site fidelity. Waterbirds 23:15.5-164.

Lande, R. 1988. Demographic models of the Northern
Spotted Owl (Stri.s occidenlalis canrina). Oecologia
75:601-607.

Larson, M. A., M. R. Ryan, and R. K. Murphy. 2002.
Population viability of Piping Plovers: effects of
predator exclusion. Journal of Wildlife Management
66:361-371.

Lieske, D. j., I. G. Warkentin, P. C. James, L. W.
Oliphant, and R. H. M. Espie. 2000. Effects of
population density on survival in Merlins. Auk
1 17:184-193.

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 1, March 2010

22

Lebreton, J.-D. and P. M. North. 1993. Marked
individuals in the study of bird populations. Birkhauser
Verlag, Basel, Switzerland.

Lebreton, J.-D., R, Pradel, and J. Clobert. 1993. The
statistical analysis of animal populations. Trends in
Ecology and Evolution 8:91-95.

Lebreton, J.-D., K. P. Burnham, J. Clobert, and D. R.
Anderson. 1992. Modeling survival and testing
biological hypotheses using marked animals: a unified
approach with case studies. Ecological Monographs
62:67-118.

Morrison, R. 1. G, R. E. Gill Jr., B. A. Harrington, S.
Skagen, G. W. Page, C. L. Gratto-Trevor, and S.
M. Haig. 2000. Population estimates of Nearctic
shorebirds. Waterbirds 23:337-352.

Morrison, R. I. G., Y. Aubry, R. W. Butler, G. W.
Beyersbergen, G. M. Donaldson, C. L. Gratto-
Trevor, P. W. Hicklin, V. H. Johnston, and R. K.
Ross. 2001. Declines in North American shorebird
populations. Wader Study Group Bulletin 94: 34-38.

Nebel, S., D. B. Lank, P. D. O’Hara, G. Fernandez, B.
Haase, F. Delgado, F. A. Estela, L. J. Evans
Ogden, B. Harrington, B. E. Kus, J. E. Lyons, F.
Mercier, B. Ortego, J. Y. Takekawa, N. Warnock,
AND S. E. Warnock. 2002. Western Sandpipers
[Catidris maiiri) during the nonbreeding season:
spatial segregations on a hemispheric scale. Auk
1 19:922-928.

Neil, R. L. 1992. Recent trends in shorebird migration for
north-central Texas. Southwestern Naturalist 37:87-88.

Nichols, J. D., M. Williams, and T. Caithness. 1990.
Survival and band recovery rates of Mallards in New
Zealand. Journal of Wildlife Management 54:629-636.

O'Hara, P. D., G. FernAndez, F. Becerril, H. de al
CuEVA, AND D. B. Lank. 2005. Life history varies
with migratory distance in Western Sandpipers
Calidris nuiuri. Journal of Avian Biology 36:191-202.

Oring, L. W. and D. B. Lank. 1982. Sexual selection,
arrival times, philopatry, and site fidelity in the
polyandrous Spotted Sandpiper. Behavioral Ecology
and Sociobiology 10:185-191.

P.vge, G. and B. Fearis. 1971. Sexing Western Sandpipers
by bill length. Bird-Banding 42:297-298.

Page, G., B. Fearis, and R. M. Jurek. 1972. Age and .sex
composition of Western Sandpipers on Bolinas
Lagoon. California Birds 3:79-86.

Pearce, J. M., J. A. Reed, and P. L. Flint. 2005.
Geographic variation in survival and migratory
tendency among North American Common Mergan-
sers. Journal of Field Ornithology 76:109-1 18.

PlERSMA, T. AND A. J. BAKER. 2000. Life history character-
istics and the con.servation of migratory shorebirds. Pages
105-124 in Behaviour and conservation (L. M. Gosling
and W. J. Sutherland. Editors). Cambridge University
Press. Cambridge, United Kingdom.

PlERSMA. T.. J. VAN GiLS. AND P. WiERSMA. 1996. Family
Scolopacidae (Sandpipers. Snipes and Phalaropes).
Pages 444-487 in Handbook of the birds of the world.
Volume 3, Hoatzin to Auks (J. del Hoyo, A. Elliott,
and J. Sargatal, Editors). Lynx Edicions, Barcelona,
Spain.

PtERSMA, T., P. WiERSMA, AND J. VAN GiLS. 1997. The
many unknowns about plovers and sandpipers of the
world: introduction to a wealth of research opportu-
nities highly relevant for .shorebird conservation.
Wader Study Group Bulletin 82:23-33.

Plissner, J. H. AND S. M. Haig. 2000. Viability of Piping
Plover Charadrius melodus metapopulations. Biolog-
ical Conservation 92:163-173.

Reed, J, M., C. S. ELPHtCK, and L. W. Oring. 1998. Life-
history and viability analysis of the endangered
Hawaiian Stilt. Biological Conservation 84:35-45.

Rice, S. M., J. A. Collazo, M. W. Alldredge, B. A.
Harrington, and A. R. Lewis. 2007. Local annual
survival and seasonal residency rates of Semipalmated
Sandpipers {Calidris piisilla) in Puerto Rico. Auk
124:1397-1406.

Rose, P. M. and D. A. Scott. 1997. Waterl’owl population
estimates. Second Edition. Wetlands International
Publication 44. Wageningen, The Netherlands.

Sandercock, B. K. 2003. Estimation of survival rates for
wader populations: a review of mark-recapture
methods. Wader Study Group Bulletin 100:163-174.

Sandercock, B. K. and C. L. Gratto-Trevor. 1997.
Local survival in Semipalmated Sandpipers Calidris
piisilla breeding at La Perouse Bay, Canada. Ibis
139:305-312.

Sandercock, B. K. and A. Jaramillo. 2002. Annual
survival rates of wintering sparrows: assessing demo-
graphic consequences of migration. Auk 119:149-
165.

Sandercock, B. K., D. Lank, and F. Cooke. 1999.
Seasonal declines in the fecundity of two arctic-
breeding sandpipers: different tactics in two species
with an invariant clutch size. Journal of Avian Biology
30:460-468.

Sandercock, B. K., D. B. Lank, R. B. Lanctot, B.
Kempenaers, and F. Cooke. 2000. Ecological
correlates of mate fidelity in two arctic-breeding
sandpipers. Canadian Journal of Zoology 78:1948-
1958.

Senner, S. E., D. W. Norton, and G. C. West. 1981. The
spring migration of Western Sandpipers and Dunlins
in southcentral Alaska: numbers, timing, and sex
ratios. Journal of Field Ornithology 52:271-284.

Tucker, G. M. and M. Heath. 1994. Birds in Europe: their
conservation status. Birdlife Conservation Series
Number 3. Birdlife International. Cambridge, United
Kingdom.

Wilson, W. H. 1994. Western Sandpiper {Calidris niaiiri).
The birds of North America. Number 90.

Ydenberg, R. C., a. C. Niehaus, and D. B. Lank. 2005.
Interannual differences in the relative timing of
southward migration of male and female Western
Sandpipers {Calidris nuutri). Naturwissen.schaften 92:
332-335.

Ydenberg, R. C., R. W. Butler, D. B. Lank, B. D. Smith,
AND J. Ireland. 2004. Western Sandpipers have
altered migration tactics as Peregrine Falcon popula-
tions have recovered. Proceedings of the Royal Society
of London. Series B 271:1263-1269.

The Wilson Joitnial of Oniilhology 1 22( 1 ):23 -28, 2010

NATAL PHILOPATRY AND APPARENT SURVIVAL OF JUVENILE

SEMIPALMATED PLOVERS

ERICA NOLT^ SIMONE WILLIAMS,' AND BRETT K. SANDERCOCK^

ABSTRACT. — Natal philopatry is rare in long-distance migrant shorebirds and requires long-term population studies to
detect. We report on the rate of natal philopatry from a 1 8-year study of Semipalmated Plovers (Charadrius seniipalniatus)
marked as hatchlings to an arctic breeding site near Churchill, Manitoba. About 2% (27/1271) of banded hatchlings
returned to the Churchill area to breed. There was no male/female bias in rates of philopatry: 17 male and 10 female
hatchlings recruited into the local breeding population. The annual rate of recruitment of hatchlings varied between 0 and
10.7%. Age of first encounter on breeding areas ranged from 1 to 8 years (median age 4) suggesting either unusually
delayed age at first breeding, or low detection rates for philopatric hatchlings. The maximum age of a recruited (known-
age) hatchling was 9 years. Natal dispersal distances did not differ between males and females, and averaged 5 km between
hatching and breeding locations. We used a time-since-marking mark-recapture model to calculate apparent survival of
hatchlings. Apparent survival in the interval after first capture was (j)| = 0.0475 (95% Cl: 0.030-0.075), whereas apparent
survival (cj)2+) of birds during subsequent intervals was 0.866 (95% Cl: 0.764-0.927). Low rates of natal philopatry suggest
little advantage to site familiarity for juveniles, and agree with theoretical predictions for migratory species with widespread
habitat availability. Received 15 June 2009. Accepted 7 October 2009.

Natal philopatry is rare in long-distance mi-
grant birds and requires either long-term popula-
tion studies or large sample sizes to detect
(Thompson and Hale 1989, Renner and McCaff-
ery 2008). Using known-age birds in survival
analyses avoids the complications of assuming
age of first breeding in population viability
analyses and, with sufficiently large sample sizes,
allows precise estimates of age-specific survival
(Sidhu et al. 2007), probability of breeding
(Sedinger et al. 2001, 2008), and life-expectancy
(Lavers et al. 2007). Analyzing natal philopatry
can also help evaluate hypotheses about the
adaptive significance of gender bias in natal
dispersal rates and distances (Greenwood and
Harvey 1982, Weatherhead and Forbes 1994,
Sutherland et al. 2000).

Studies of arctic-breeding birds are disadvan-
taged in documenting natal philopatry and
dispersal becau.se of inaccessibility of breeding
areas and low encounter probabilities (Renner and
McCaffery 2008). Possibly as a con.sequence,
studies of temperate breeding shorebirds typically
report higher rates of encounter of breeding
juveniles, especially if the species is of con.serva-
tion concern (e.g., Charadrius alexandrimts and
C. melodus', Larson et al. 2000, Colwell et al.
2007b, Stenzel et al. 2007).

‘Biology Department, Trent University, Peterborough.
ON, K9J 7B8. Canada.

^Division of Biology. Kansas Slate University. Manhat-
tan, KS 66506, USA.

^Corresponding author; e-mail: enol@lrentu.ca

The Semipalmated Plover (C. semipalmatus)
has been the subject of an 18-year population
study near Churchill, Manitoba. This small
shorebird breeds in open gravel, tundra, beaches,
and riverbeds throughout the sub-arctic regions of
North America (Nol and Blanken 1999). Semi-
palmated Plovers migrate long distances from
sub-arctic breeding areas to a large, dispersed
wintering range that spans coastal areas of the
southern United States, Mexico, Central America,
and northern and central South America. Unlike
many Charadrius plovers, there is currently no
evidence to suggest this species is declining
(Morrison et al. 2006, Bart et al. 2007). An earlier
study of this species suggested low rates of natal
philopatry and no gender bias in these rates (Flynn
et al. 1999). Our objectives are to: (1) update
knowledge of these factors, (2) evaluate natal
di,spersal distances, and (3) estimate apparent
juvenile and adult survival based on a known-age
sample. We also estimate the earliest age of first
breeding and report on a longevity record based
on a known-age hatchling.

METHODS

Study Sites and Species. — Field work was
conducted each year from the beginning of June
to mid-August in 18 years including 1988 and
1992 to 2008. We regularly searched 11 major
breeding sites in the area .south and east of
Churchill, Manitoba. Canada (58° 45' N,
95° 04' W), spanning ~ 20 km of coastal beaches
and two inland sites in an area of 34.2 km'
(Fig. 1). Each field .season, researchers walked

23

24

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 1, March 2010

FIG. I. Churchill, Manitoba area showing the 1 1 major study sites (★).

potential nesting areas in search of Semipalmated
Plovers. The nests of Semipalmated Plovers in
1988 and from 1992 to 2000 were assigned
identification (ID) numbers sequentially as they
were found within the 1 1 areas, and locations of
nests were drawn on field maps and aerial photos.
Nests of Semipalmated Plovers were assigned ID
numbers from 2001 to 2008 and the locations
were recorded using a Global Positioning System
(GPS). In early June, observers searched for
previously banded Semipalmated Plovers and
recorded the band combinations of color-marked
birds using 20-40 X spotting .scopes. Any nesting
Semipalmated Plovers that were unhanded, or
needed clarification of identification (e.g., faded
color bands) were captured using circular walk-in
traps with a keyhole entrance, made of hardware
cloth, that were placed over the nest. Once adults
were captured, they were banded with an
aluminum band and given a unique color
combination of bands or faded color and alumi-
num bands were replaced. Gender of adults was
assigned using the number of brown feathers in
the crown and supcrcilium with females having
more brown feathers than males (Teather and Nol

1997). Gender assignments were confirmed by
behavioral postures with males using display
flights or mounting females during copulations.

Chicks were banded in the nest with unique
aluminum bands and, depending on the year of
study, with either brood-specific or individual
color band combinations.

Statistical Analysis. — ^We assigned the location
of the natal site as the UTM of the center of one of
the 1 1 nesting areas for older nest and banding
records without reliable GPS locations (in years
prior to 2000) (Fig. 1 ), and the location of the first
breeding site as the UTM of the center of the site
at which we first encountered them breeding. We
used a binomial test to examine whether there was
a bias in the gender of returning hatchlings, and t-
tests to compare natal dispersal distances of males
and females.

We used a two-stage ‘time-since-marking’
model for live encounter data using Program
MARK to calculate annual probabilities of
apparent survival (cj)) and encounter (p). The
time-since-marking model separates apparent
survival in the first interval after capture ((t)|)
Irom apparent survival in subsequent intervals

No! et ill. • SllRVIVAl. OI’ JUVCNILH SEM I I’ALM ATEI) IM.OVERS

25

((j)2+) (Larson et al. 2(){)()). We modeled apparent
survival as a fLinetion of year of marking and
gender beeanse these factors are important in
other plovers (Larson et al. 2()()()), and treated the
encounter rate as a nuisance parameter. We tested
for possible overdispersion before testing reduced
models by using the median c procedure to
calculate a variance inflation factor for the global
starting model (Cooch and White 2004).

RESULTS

Twenty-seven of 1,271 (2.1%) banded hatch-
lings from 1988 to 2008 returned to breed in the
Churchill area. Rates of return were low in all
years, but reached 10% in 2000 and 2001
(Table 1). Ten local females and 17 males
recruited into the breeding population, a ratio that
was not significantly different than LI (Binomial
test, P = 0.09). The age (±SE) at which birds
were first encountered breeding ranged widely
from 1 to 8 years (median 4 years) and did not
differ between males and females (males: 3.9 ±
0.4 years, females: 4.2 ± 0.6 years; t = 0.38, P =
0.65). The maximum age of a recruited (known-
age) hatchling was 9 years. The distance (±SE)
between natal sites and nesting sites where we first
encountered philopatric individuals did not differ
between males and females (males: 5.8 ± 1.4 km,
median = 4.2 km, range: 0.7 to 12.6 km, n = 11;
females: 3.7 ± 1.8 km, median = 0.8 km, range:
0.3 to 12.7 km, n = l\t = 0.93, P = 0.37).

Our estimate of c for the global model indicated
there was little evidence of overdispersion (c <

1 ), and we set the variance inflation factor to ? =

1 and used AlCc for model selection. The
minimum AICc model tor the survival analysis
from Program MARK included two capture
periods for apparent survival (cj)) and a constant
encounter rate (p). Apparent survival in the first
interval after capture ((t)i) was 0.048 (95% Cl:
0.030-0.075), whereas survival ((t)2+) fo*' birds
during subsequent capture periods was 0.866
(95% CL 0.764-0.927). Encounter probabilities
(p) were relatively low, 0.174 (95% Cl: 0.113-
0.230). The next best model included a time effect
for encounter probabilities but was 28.8 AlCr
units from the top model and received no weight
relative to the top model (u’l < 0.01 ).

DISCUSSION

The frequency of returning Semipalmated
Plovers marked as hatchlings is among the lowest
reported for shorebirds (Thompson and Hale

I'ABLH 1. Number ol'
l^lover chicks per year and
IVom that year’s cohort.

newly banded Semipalmated
number ol' new adult recruits

Year

Newly banded chiek.s Min. niiinber of recruits front cohort (%)

1988

57

1 (1.75)

1992

31

0 (0)

1 993

87

3 (3.44)

1994

92

1 (1.08)

1 995

81

1 (1.23)

1996

106

2 (1.89)

1997

75

2 (2.67)

1998

58

2 (3.45)

1 999

25

0 (0)

20(){)

28

3 (10.70)

2001

38

4 (10.50)

2002

44

1 (2.27)

2003

94

1 (1.06)

2004

50

2 (4.00)

2005

105

2 (1.90)

2006

103

0 (0)

2007

122

0 (0)

2008

75

0 (0)

1989, Flynn et al. 1999). Our estimate was similar
to an earlier estimate from the same study area
using observation data for returning juveniles
from 1988 to 1996 (1.57%, n = 455, Flynn et al.
1999) despite a sample size that has nearly tripled.
Our hatchlings were banded within 48 hrs of
hatching, and low rates may be due to high
mortality in the period after hatching when young
are unable to thermoregulate or fly, and are
vulnerable to inclement weather conditions and
predators (Thompson and Hale 1989, Flynn et al.
1999, Sandercock et al. 2005, Colwell et al.
20()7a). Our survival estimates obtained for the
first sample period were also low compared to that
reported for the temperate congeners: Piping
Plover (C. melodiis) and Snowy/Kentish Plover
(C. a. alexandrinus) (survival from hatching to the
following year, ()) = 0.15, Sandercock et al. 2005;
(j) = 0. 1 8 for true survival of C. a. nivosus, Stenzel
et al. 2007; (j) = 0.32 for C. melodiis banded as
juveniles >16 days of age, Larson et al. 2000).
Predator control can reduce mortality of hatch-
lings for both temperate North American species
(Stenzel et al. 2007). However, large differences
between these rates and tho.se of Semipalmated
Plovers may also be due to greater permanent
emigration (through greater natal dispersal) from
the Churchill study area, rather than lower
survival, because this population of Semipalmated
Plovers appeared to be stable over the study

26

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. 1. March 2010

period (E. Nol, unpiibl. data) and otherwise could
probably not sustain mortality of this magnitude.
The difference between these survival rates
among congeners also mirrors that seen in
comparing natal philopatry and dispersal rates
between migratory and resident passerines
(Weatherhead and Forbes 1994, Sutherland et al.
2000). Lower rates of natal philopatry in long-
distance migrants like Semipalmated Plovers may
be due to dispersal costs accrued when they
migrate south and relinquish familiarity with their
natal area. It may only be important to find
similar, but not necessarily familiar habitat when
they return to breed (Weatherhead and Forbes
1994).

There was substantial variability in number of
recruits into the breeding season among hatchling
cohorts, but a model of survival with time-
dependence was not supported. A similar absence
of time effects was also found for Kentish Plover
in .southern Turkey (Sandercock et al. 2005) and,
in both cases, may have been due to sparse data
with low numbers of philopatric young. Alterna-
tively, variation in the number of recruiting
individuals may be more common when resources
for breeding birds fluctuate widely (e.g.. Black-
throated Blue Warbler, Dendroica caerulescens\
Rodenhouse et al. 2003) or where interference
competition results in exclusion of potential
recruits (Arcese et al. 1992). Low fecundity of
arctic-breeding shorebirds may keep populations
below the carrying capacity of breeding areas
(Piersma and Baker 2000), but as yet we know
little about the possible role of density-depen-
dence in this group of birds.

Territorial shorebirds are often assumed to have
male-biased natal philopatry (reviewed in Colwell
et al. 2007b) because of a re.source-defen.se mating
system (Greenwood 1980). However, bias to-
wards philopatric recruiting juvenile males, where
it occurs, is usually small (Gratto et al. 1985, Reed
and Oring 1993, Colwell et al. 2007b, Stenzel et
al. 2007; but see Redmond and Jenni 1982). The
advantages of familiarity with the natal site should
be approximately equal for males and females of
monogamous, bi-parental incubating species. Fe-
male vSnowy Plovers share in incubation duties but
typically de.sert the brood and, at times, re-nest
with other males (Page et al. 1995). In this
species, where females are presumed to have a
relatively low attachment to the territory, the
return rate of juvenile females to the local study
area to breed is only 12% lower than for juvenile

males, but still 64% (Stenzel el al. 2007). This
difference reversed in winter with juvenile males
more likely to disperse (54%) from wintering
areas than juvenile females (44%). Thus, a re-
evaluation of the theoretical basis for gender-
bia.sed natal dispersal in birds is probably needed
(Redmond and Jenni 1982, Sutherland et al.
2000).

Most Semipalmated Plovers appear to delay
breeding beyond 1 year although, contrary to the
results from the previous study on this species
(Flynn et al. 1999), at least one male bred in its
first year. Delayed age at maturity differs from
other studies of migratory shorebirds, where most
individuals attempt to breed in their first year
(reviewed in Sandercock et al. 2005). This
suggests that a large proportion of the Churchill
population may over-summer on wintering areas
(e.g., McNeil et al. 1994), similar to reports for
three species of calidridine sandpipers in Africa
and Central America (Summers et al. 1995,
O’Hara et al. 2002).

The precise age of first breeding is difficult to
ascertain because numbers of philopatric young
were low and re-encounter probabilities of banded
hatchlings are low (Sandercock et al. 2005). Age-
specific variation in breeding propensity with
higher rates of philopatry can be estimated with
robust design models (Sedinger et al. 2001, 2008),
but our data were not adequate for this approach.
Most hatchlings were banded with an aluminum
band and a single color band. Band loss and color
fading after 3-4 years reduces the probability of
observing these individuals if breeding in the
Churchill study area. That two hatchlings were not
detected in the local breeding population until the
age of 8 years strongly suggests that some birds
are temporarily leaving the population and then
returning. In another study of this species on
Akimiski Island, Nunavut, a breeding male
underwent a long-distance (30 km) dispersal
event. This also suggests, that despite generally
greater nest site tenacity by males than females
(Flynn et al. 1999, Fishman et al. in prep), males
can and will disperse between breeding attempts
in the same or different years. These difficult to
document movements will continue to result in
lower estimates of adult and juvenile survival for
shorebirds breeding in mostly inaccessible habi-
tats across the North American arctic, and
reaffirm the importance of scale of the study area
for documenting dispersal events (Greenwood and
Harvey 1982, Jackson 1994, Sandercock and

No! et cil. • SURVIVAl. OF’ JUVENll.E SEM 1 1'Al.M ATEF) F^L()VEF<S

27

Gratto-Trevor 1997, Colwell et al. 2()()7b, Slenzel
et al. 2007).

Apparent survival estimates tor philopatric
individuals at intervals after the first return were
high, and nuieh higher than previous estimates
from an adult sample from this population
(Badzinski 2000; adult estimate of 0.71, Cl:
0.64-0.78). Estimates of apparent survival for
birds marked as adults may include first-time
breeders of a wider range of ages that have lower
survival rates. Alternatively, philopatric individu-
als may have higher site fidelity as adults than
individuals that immigrate into our study sites as
adults. Variation in individual strategies for space
use have been reported for wintering Grey Plovers
(Pliivialis squatarola), where individuals are
either territorial or non-temtorial, and do not
change strategies during their lifetime (Turpie
1995). Our mean estimate of apparent adult
survival is also higher than reported for the
congeners Piping Plover (Larson et al. 2000, 5
= 0.77 for known-age juveniles). Snowy Plover (s
= 0.69, Stenzel et al. 2007), and Kentish Plover {s
= 0.58, Sandercock et al. 2005). Similarly,
survival rates of known-age Pacific Golden
Plovers (Pliivialis fulva\ territorial: 5 = 0.90,
non-territorial, 5 = 0.82) were 10-15% higher
than those from a sample of unknown-age adults
(territorial, 5 = 0.80, non-territorial, 5 = 0.67,
Johnson et al. 2001). Variation in individual
strategies for space use and apparent differences
in age of first reproduction invite further research
on the demography of short- versus long-distance
migrant shorebirds to identify possible trade-offs
in life history strategies, and to evaluate the
consequences of life-history variation for planning
conservation efforts.

ACKNOWLEDGMENTS

We thank the countless field assistants and students who
helped band Semipalmated Plovers over the years. We also
thank the Churchill Northern Studies Centre (CNSC) for
their continued logistic and financial support. G. P.
Kershaw kindly provided the map. The Science Horizons
Youth Internship Programme of Environment Canada
supported SW while the field work has been largely
supported by NSERC grants to EN.

LITERATURE CITED

Arcese, P., J. N. M. Smith, W. M. Hochachka, C. M.
Rogers, and D. Ludwig. 1992. Stability, regulation,
and the determination of abundance in an insular Song
Sparrow population. Ecology 73:805-822.

Badzinski, D. S. 2000. Population dynamics ot Semipal-

mated Plovers (Charadrius scmipcdiiuilits) breeding at
Churchill, Manitoba. Thesis. Trent University, Peter-
borough, Gniario, Canada.

Bart, J., S. Brown, B. Harrington, and K. 1. G.
Morrison. 2007. Population trends ol North Ameri-
can shorebirds; population declines or shilting distri-
butions? Journal of Avian Biology 38:73-82.

Colwell, M. A., S. J. Hurley, J. N. Hali., and S. J.
Dinsmorh. 2007a. Age-related survival and behavior
of Snowy Plover chicks. Condor 109:638-647.

Colwell, M. A., S. E. McAlllster, C. B. Millett, A. N.
Transou, S. M. Mullen, Z. J. Nelson, C. A.
Wilson, and R. R. LeValley. 2007b. Philopatry
and natal dispersal of the Western Snowy Plover.
Wilson Journal of Ornithology 119:378-385.

CoocH, E. AND G. White. 2004. Using MARK-A gentle
introduction, http://www.phidot.org/software/mark/
docs/book/markbook.exe (accessed 2 June 2009).

Elynn, L., E. Nol, and Y. Zharikov. 1999. Philopatry,
nest-site tenacity, and mate fidelity of Semipalmated
Plovers. Journal of Avian Biology 30:47-55.

Gratto, C. L., R. 1. G. Morrison, and F. Cooke. 1985.
Philopatry, site tenacity, and mate fidelity in the
Semipalmated Sandpiper. Auk 102:16-24

Greenwood, P. J. 1980. Mating systems, philopatry, and
dispersal in birds and mammals. Animal Behaviour
28:1140-1162.

Greenwood, P. J. and P. H. Harvey. 1982. The natal and
breeding dispersal of birds. Annual Review of Ecology
and Systematics 13:1-21.

Jackson, D. B. 1994. Breeding dispersal and site-fidelity in
three monogamous wader species in the Western Isles,
U.K. Ibis 136:463^73.

Johnson, O. W., P. L. Bruner, J. J. Rotella, P. M.
Johnson, and A. E. Bruner. 2001. Long-term study
of apparent survival in Pacific Golden-plovers at a
wintering ground on Oahu, Hawaiian Islands. Auk
118:342-351.

Larson, M. A., M. R. Ryan, and B. G. Root. 2000. Piping
Plover survival in the Great Plains: an updated
analysis. Journal of Field Ornithology 71:721-729.

Lavers, J. L., I. L. Jones, and A. W. Diamond. 2007.
Natal and breeding dispersal of Razorbills (AIca torda)
in eastern North America. Waterbirds 30:588-594.

Lishman, C., E. Nol, K. F. Abraham, and L. P. Nguyen.
In prep. Mate retention, breeding dispersal, return
rates, and apparent survival in a sub-arctic population
of Semipalmated Plovers (Charadrius semipalmalus).
Submitted to Condor.

McNeil, R., M. T. Diaz, and A. Villenueve. 1994. The
mystery of shorebird over-summering; a new hypoth-
esis. Ardea 82:143-152.

Morrison, R. I. G., B. J. McCafff.ry, R. E. Gill, S. K.
Skagen, S. L. Jones. G. W. Page, C. L. Graito-
Trevor, and B. a. Andres. 2006. Population
estimates of North American shorebirds. Wader Study
Group Bulletin 1 1 1:67-85.

Nol, E. and M. S. Blanken. 1999. Semipalmated Plover
(Charadrius semipalmalus). The birds of North
America. Number 444.

O’Hara, P. D., D. B. Lank, and F. S. Delgado. 2002. Is

28

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 1. March 2010

the timing ot molt altered by migration? Evidence
trom a comparison ot age and residency classes of
Western Sandpipers Calidris niaiiri in Panama. Ardea
90:61-70.

Page, G. W.. J. S. Warriner, J. C. Warriner, and P. W.
C. Paton. 1995. Snowy Plover {Characiritis ale.xan-
drinus). The birds of North America. Number 154.

PlERSMA, T. AND A. j. Baker. 2000. Life history
characteristics and the conservation of migratory
shorebirds. Pages 105-124 in Behaviour and conser-
vation (L. M. Gosling and W. J. Sutherland, Editors).
Cambridge University Press, London, United King-
dom.

Redmond, R. L. and D. A. Jenni. 1982. Natal philopatry
and breeding area fidelity of Long-billed Curlews
(.Numenius americanus): patterns and evolutionary
consequences. Behavioral Ecology and Sociobiology
10:277-279.

Reed, M. J. and L. W. Oring. 1993. Philopatry, site
fidelity, dispersal, and survival of Spotted Sandpipers.
Auk 1 10:541-551.

Renner, H. M. and B. J. McCaffery. 2008. Demography
of Eastern Yellow Wagtails at Cape Romanzof,
Alaska. Wilson journal of Ornithology 120:85-91.

Rodenhouse, N. L., T. S. SiLLETP, P. J. Doran, and R. T.
Holmes. 2003. Multiple density-dependence mecha-
nisms regulate a migratory bird population during the
breeding season. Proceedings of the Royal Society of
London, Series B 270:2105-21 10.

Sandercock, B. K. and C. L. Gratto-Trevor. 1997.
Local survival in Semipalmated Sandpipers Calidris
piisiUa breeding at La Perouse Bay, Canada. Ibis
139:305-312.

Sandercock, B. K., T. Szekely, and A. Kostolanyi.
2005. The effects of age and sex on the apparent
survival of Kentish Plovers breeding in southern
Turkey. Condor 107:583-596.

Sedinger, j. S., M. S. Lindberg, and N. D. Chelgren.

2001. Age-specitlc breeding probability in Black
Brant: effects of population density. Journal of Animal
Ecology 70:798-807.

Sedinger, J. S., N. D. Chelgren, D. H. Ward, and M. S.
Lindberg. 2008. Eidelity and breeding probability
related to population density and individual quality in
black Brent Geese Brant hernicla nigricans. Journal of
Animal Ecology 77:702-712.

SiDHU, L. A., E. A. Catchpole, and P. Dann. 2007. Mark-
recapture-recovery modeling and age-related survival
in Little Penguins (Endyptnla minor). Auk 124:815-
827.

Stenzel, L. E., G. W. Page, J. C. Warriner, J. S.
Warriner, D. E. George, C. R. Ey.ster, B. A.
Ramer, and K. K. Neuman. 2007. Survival and natal
dispersal of juvenile Snowy Plovers (Charadrius
ule.xandrinns) in central coastal California. Auk
124:1023-1036.

Summers, R. W., L. G. Underhill, and R. P. Prys-Jones.
1995. Why do young waders in southern Africa delay
their first migration to the breeding grounds? Ardea
83:351-357.

Sutherland, G. D., A. S. Hare.stad, K. Price, and K. P.
Lertzman. 2000. Scaling of natal dispersal in
terrestrial birds and mammals. Conservation Ecology
4:16.

Leather, K. and E. Nol. 1997. Mixed sexual dimorphism
in the Semipalmated Plover. Condor 99:803-806.

Thompson, P. S. and W. G. Hale. 1989. Breeding site
fidelity and natal philopatry in the Redshank Tringa
totanns. Ibis 134:214-224.

Turpie, j. K. 1995. Non-breeding territoriality: causes and
consequences of seasonal and individual variation in
Grey Plover Pliivialis siptataroia behaviour. Journal of
Animal Ecology 64:429^38.

Weatherhead, P. and M. L. Eorbes. 1994. Natal
philopatry in passerine birds: genetic or ecological
intluences? Behavioral Ecology 5:426-433.

llic Wilson Joiiniol oj Oniilliology 122(1):29 38, 2010

BREEDING BIOLOGY AND NESTING SUCCESS OF THE SLATE-
THROATED WHITESTART {MYIOBORUS MINIATUS) IN
MONTEVERDE, COSTA RICA

RONALD L. MUMME'

ABSTRACT. — I examined the breeding biology and nesting success ol' the Slate-throated Whilestail (Myiohorns
miniaiiis). a socially monogamous neotropical warbler. For Five breeding seasons (2000-2004) in Monteverdc, Costa Rica,
near the center oF its broad geographic range. Nesting was strongly seasonal, extending From late March through the end ol
June and coinciding with the end oF the dry season and the onset oF the rainy season in mid-May. Females constructed
domed nests on open steep slopes or in banks along roads and trails. Mean clutch size was 2.9 eggs, and the mean
incubation period was 14.2 days. Females perFormed all incubation oFeggs and brooding oF young nestlings, but both males
and Females Fed nestlings and Fledglings. Mean provisioning rate at nests when young were 5-9 days of age was 20.3
Feedings/hr, and Females Fed young at a signiFicantly higher rate than males (11.7 vs. 8.3 Feeding.s/hr, respectively).
Nestlings reached mean adult body mass oF 9.5 g — day 7—8, and mean age at Fledging was 1 1 .3 days. Parents Fed juveniles
For at least 4 weeks after Fledging; the latest record of adults feeding Fledged young was for 4()-day-old Juveniles. Daily nest
survival rates showed strong annual variation and generally declined as the nesting season progressed. Mean daily survival
rate was 0.968 and estimated overall nest success was 40.3% with a mean of 2.6 young Fledging from successFul nests.
Predation was implicated in ~85% of nest failures. Received 17 February 2009. Accepted 22 July 2009.

The genus Myiohorns comprises 12 species of
sexually monomorphic parulid warblers found in
montane forests throughout the American tropics
and subtropics (Curson et al. 1994). All tTtembers
of the genus have contrasting black-and-white
tails and animated foraging displays that function
to startle potential insect prey, which are then
pursued and captured in flight (Jablotiski 1999,
Mumme 2002, Mumme et al. 2006). Species in
the genus are termed “redstarts” by the AOU
(1998) but “whitestarts” by the International
Ornithological Congress Standing Committee on
English Names (Gill et al. 2009). Recent phylo-
genetic analyses indicate that Myiohorns is
monophyletic (Perez-Eman 2005) and part of a
larger clade that includes five species from three
additional genera: Ergaticns (2 species, Mexico
and Guatemala), Cardellina ( 1 species, Mexico
and the southwestern United States), and Wilsonia
(2 species, northern United States and Canada)
(Lovette and BerminghaFii 2002, Rabosky and
Lovette 2008).

The Slate-throated Whitestart (Myiohorns min-
iatns) is the FTFOst widely distributed Myiohorns
species, ranging from the mountains ot noF'thern
Mexico to the southern Andes of Bolivia and
northern Argentina (Ridgely and Tudor 1989,
Curson et al. 1994, DiGiacoFiio 1995, McCorFiiack
et al. 2005). Genetic analysis suggests it is

' Department of Biology, Allegheny College, 520 North
Main Street. Meativille. PA 16335, USA;
e-mail: rmumme@allegheny.edu

monophyletic, although populations frotit Mexico
northwest of the Isthrnus of Tehuantepec are
genetically distinct frotn those in southern Mex-
ico, Central America, and South America (Perez-
Eman 2005). In Central America, the Slate-
throated Whitestart is tnonogamous, territorial,
and often retTtains paired throughout the year
(Skutch 1954, Shopland 1985). It occurs in Fnid-
elevation montane forest thi-oughout FTFUch of
southern Central AiTFerica and South America
while a second species of Myiohorns replaces it at
higher elevation (Ridgely and Tudor 1989, Stiles
and Skutch 1989, Curson et al. 1994).

In the last decade new data on the bF'eeding
biology, nest predation, and survival of tF'opical
and southern hemisphere biFxls have fueled a
vigorous F'esui'gence of interest in avian life
histoF'y evolution (e.g., MaF'tin et al. 2000, 2007;
Martin 2002, 2004; Ricklefs and Wikelski 2002).
My objective is to contribute to this gi'owing body
of work by pi'esenting data on the breeding
biology and nesting success ol the Slate-throated
Whitestart in Monteverde, Costa Rica. SoFiie data
aF'e available for this species froFii Guatemala,
Costa Rica, and Venezuela (Skutch 1954. Shop-
land 1985, Collins and Ryan 1994), but 1 pre.sent a
much moF'e extensive analysis based on 5 years of
study of a coloF-banded population and recoFds
from 132 separate nests.

METHODS

Study Area and Study Population. — ^The study
was conducted during five consecutive nesting

29

30

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 1. March 2010

seasons (2000-2004) on a color-banded popula-
tion of Slate-throated Whitestarts at the Estacion
Biologica Monteverde (EBM) and surrounding
properties in Monteverde, Costa Rica (Nadkarni
and Wheelwright 2000). The study area
(10° 18' N, 84° 48' W) encompassed 35-45
breeding pairs of Slate-throated Whitestarts on
teiritories established in primary premontane and
lower montane wet forest (Haber 2000), old
secondary forest, and abandoned pastures with
scattered trees and shrubs at elevations of 1,400-
1 ,700 m. Eield work was conducted during
January-June 2000, May-June 2001-2003, and
April-June 2004.

I mist netted and color banded 170 adult Slate-
throated Whitestarts during the study, either by
attracting birds to song playback or by placing
nets near active nests. An additional 172 individ-
uals were color banded as nestlings. Gender of
color-banded adults was assigned from either the
presence of a brood patch in females or by
subsequent behavioral observations, such as
incubation or differences in vocal behavior;
although both males and females produce song
(Shopland 1985), males sing much more vigor-
ously and frequently than females (Mumme
2002).

Data Collection. — Breeding territories were
surveyed regularly to search for signs of nesting.
The date of nest initiation (first egg date) was
ascertained directly for nests found during build-
ing, but estimated indirectly for nests found with
eggs or nestlings by extrapolating backwards using
data on the laying sequence, incubation period, and
growth and development of nestlings. Nests were
usually checked every 3-4 days to ascertain nest
fate, although a smaller sample of nests was
checked once or twice daily at critical periods to
obtain more detailed infonnation about timing of
egg laying, hatching, and Hedging. 1 marked eggs
daily as they were laid at .seven nests and measured
egg size with dial calipers. 1 used a 10-g or 30-g
Pesola scale to measure body mass of nestlings at
regular intervals from hatching (day 0) through day
8 at 1 1 nests where hatching date was known.
Parental feeding rates at ne.sts were calculated from
data collected during 1- to 2-hr nest watches at 24
different nests where at least one member of the
pair was color banded and the nests contained 2-3
nestlings that were 5-9 days of age. Nest watches
were conducted from an observation blind during
mid-morning to early afternoon (()8()()-l 300 hrs) in
dry weather. 1 recorded all feeding visits by males.

females, and birds whose identity could not be
ascertained. Nestlings were handled and color
banded when 6-8 days of age at all nests. I made
regular post-fledging visits to six territories where
pairs nested successfully to record developmental
state of juveniles and duration of post-fledging
parental care.

Statistical Analysis. — Data on egg size were
analyzed with a two-factor ANOVA in which nest
and laying order were used as factors. Statistical
comparisons of male and female provisioning
rates at nests were made with a matched-pairs t
test. Calculations for both tests were performed
using R for Mac OS X (Version 2.8.1). Data on
nestling growth were fitted to a third-order
polynomial curve using Microsoft Excel 2004
for Mac OS X.

Daily Nest Sunhval and Nest Success. — used the
nest survival methods of Version 5.1 of Program
MARK (White and Burnham 1999, Rotella 2008) to
estimate daily survival rate (DSR) of Slate-throated
Whitestart nests. The analysis was based on records
from 128 nests where nest fate could be reliably
ascertained. I evaluated a series of MARK nest
survival models that u.sed combinations of two
categorical covariates, year (2001-2004) and nest
habitat type (road banks, trail banks, and natural
slopes), and two continuously varying covariates,
date (calendar date within year) and nest age
(number of days since first egg date). Covariates
were coded and incorporated into models following
Rotella (2008), allowing evaluation of models
where nest DSR varied with year, nest habitat type,
date, and nest age.

Candidate models were compared to the null
model of constant nest DSR, S(.), based on
Akaike’s information criterion for small samples
(AIC(.). The model with the lowest AIC^. value was
considered to be the model that best fit the nest
survival data, and Akaike weight (m’,) was used as a
measure of the relative support for each candidate
model (White and Burnham 1999). Estimates of
DSR under specific models were obtained from
model beta parameters and back transformation
following Rotella (2008). I estimated the cumula-
tive probability of overall nest success by raising
DSR to the power 27.6, which is the mean duration
of a successful ne.sting cycle in days.

RESULTS

Timing of Breeding. — Slate-throated White-
starts in Monteverde have strong breeding sea-
sonality. First egg dates ranged from 27 March to

Percent of nests

Miminw BREEDING BIOLOGY OE MYIOHORUS MINIATUS

31

FIG. 1 . Estimated first egg dates for 132 Slate-throated
Whitestart nests in 2000-2004 in Monteverde, Costa Rica.
Indistinct bars indicate that during 2001-2003 an unknown
number of nests initiated in late March through early May
either fledged young or failed before they could
be discovered.

17 June with 6 May being the median date of nest
initiation (Fig. 1). However, field work did not
begin until after 10 May in 2001-2003, and
several pairs had already fledged young from
early nests. Thus, the true peak of nesting was
likely earlier than 6 May (Fig. 1).

I found 132 nests during the study: 60 (45.5%)
during building, 28 (21.2%) during egg laying or
incubation, and 44 (33.3%) during the nestling
stage. Pairs frequently renested when initial
nesting attempts failed by making a second (/? =
16; 12.1% of nests) or third (« = 1; 0.8% of nests)
attempt. However, only one certain case of true
double brooding was observed; in 2001, one pair
was feeding fledged young from its first nest on
29 May and initiated a new nest on 31 May. The
young from the second nest fledged successfully
on 26 June.

Nest Building, Egg Laying and Incubation. —
Slate-throated Whitestarts built domed nests in
road banks (n = 70; 53.0%), trail banks (n = 19;
14.4%)^ or on steep slopes and natural banks in
open areas such as pasture edges, ravines,
washouts, and tree fall gaps (n = 43; 32.6%).
Nests away from roads and trails were relatively
more difficult to find, making it likely they were
under-represented in the sample of 132 nests. I
observed nest-building behavior lor 31 pairs
where at least one member of the pair was color
banded and gender was known; in all 31 cases

females alone carried nesl material and construct-
ed the nesl. Males did not directly participate in
nest building, but they Irequently accompanied
their mates to the nest during building.

Clutch size was 1-4 eggs with a mean ± SE of
2.89 ± 0.04 (/; = 82) and a strong modal clutch
size of 3; 87.8% of nests contained three eggs
(Fig. 2A). Laying occurred early in the morning,
generally within 2 hrs of dawn. Eggs were laid on
consecutive days in 13 (76.5%) of 17 nests where
egg laying was monitored daily. Final clutch size
was three eggs in all 13 of those nests. However,
in four nests, a 1-day (/? = 3) or 2-day (n — 1 ) gap
occurred between laying of the first and second
egg, and final clutch size was two eggs in three of
the four nests. Thus, the mean it SE interval
between laying of the first egg and clutch
completion was 2.12 ± 0.08 days (n = 17 nests).

Mean ± SE egg size in seven three-egg
clutches was 16.67 ± 0.18 X 13.36 ± 0.07 mm
(/7 = 21). Length and width varied significantly
among nests (two-factor ANOVA ^6,12 ~ 26.71
and 9.29, respectively, both P < 0.001), but
neither length nor width was significantly related
to laying order (two-factor ANOVA ^2,12 “ 0.50
and 0.19, respectively, both P > 0.6).

Females alone incubated the eggs, and the
incubation period was 14-15 days (mean ± SE =
14.2 ± 0.1 days; Eig. 2B). Hatching was gener-
ally synchronous, and hatching success of eggs
was high; of 160 eggs from 55 nests that survived
through the end of incubation, 150 (93.8%)
hatched.

Nestling and Fledgling Periods. — Nestlings
averaged 1.4 ± 0.1 g (mean ± SE) at hatching
(n = 3 nests) and reached mean adult body mass
of 9.5 g ~ day 7-8 after hatching (Fig. 3). Both
males and females provisioned nestlings, and
female brooding of young nestlings ended when
nestlings reached 5-6 days of age. Overall
provisioning rates at nests with nestlings of age
5-9 days was 20.3 ± 1.0 feedings/hr (n = 24
nests); females provisioned young at a signifi-
cantly higher rate (11.7 ± 0.8 feedings/hr) than
males (8.3 ± 0.5 feedings/hr; two-tailed paired t
test, t = 4.23, df = 23, P < 0.001).

Age at fledging was 1 1.3 ± 0.2 days (mean ±
SE; Eig. 2C). Juveniles began to forage on their
own by day 30, but parents continued to feed
juveniles for at least 4 weeks after fledging; the
latest record of adults feeding dependent young
was of 40-day-old Juveniles that had fledged
29 days earlier.

Number of nests Number of clutches Number of clutches

32

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 1, March 2010

Clutch size

Incubation period (days)

Nestling period (days)

Nestling age (days)

FIG. 3. Growth in body mass of nestling Slate-throated
Whitestarts in 1 1 nests of known hatch date in Monteverde,
Costa Rica, 2000-2003. Each point repre.sents the mean
ma.ss of nestlings from a single nest with either two or
three young.

Daily Nest Survival and Nest Success. — ^Esti-
mated nest daily survival rate (DSR) for the 128
nests used in the Program MARK analysis was
0.968 with a 95% confidence interval (Cl) of
0.958-0.975. Estimated nest success, based on an
average nest exposure period of 27.6 days
(2.1 days between laying of the first egg and
clutch completion, 14.2 days incubation, and
11.3 days during the nestling period), was
40.3% (95% Cl, 30.5^9.9%). An average (±
SE) of 2.61 ± 0.07 young fledged from 74
successful nests.

The MARK model of nest DSR that best fit the
data included interaction terms between covari-
ates year and date (Table 1). This model was 2.3 1
AIC(. units better than the second best model,
which included the same two covariates in an
additive model. These results suggest (1) there is
considerable annual and seasonal variation in nest
DSR, but (2) the precise nature of the seasonal
effect of date is year-dependent. Estimated DSR
varied annually from a low of 0.953 (95% Cl,
0.927-0.970) in 2004 to a high of 0.989 (95% Cl,
0.965-0.996) in 2003, coiresponding to estimated
nesting success of 26.3 and 72.7%, respectively
(Fig. 4A). Nest DSR generally declined over the
course of the nesting season, but the nature of the

FIG. 2. Variation in clutch size (A), length of the
incubation period (B), and nestling period (C) in Slate-
throated Whitestarts in Monteverde, Costa Rica.

Mwuliw liREEDlNCJ BIOLCXJY ()!• MYIOHORUS MINIA'IVS

33

TABLE 1. Models of nest survival for 128 nests ol
Slate-throated Whitestarts in Monteverde, Costa Riea,
2()()0-2()()4, based on analysis with Program MARK.
Models are ranked from best to least supported aeeording
to values of AAIC,, the differenee in Akaike's information
eriterion for small samples between the listed model and the
top model. The eovariates used in eaeh model are shown in
parentheses. Other eolumns show the number of parameters
estimated in eaeh model (K), model devianee, AAIC,, and
Akaike weight (u,), a measure of the relative support for
eaeh model.

Model

K

Deviance

AAIC,

S(year X date) - interaction

10

333.48

O.OOO

0.606

Styear + date) - additive

6

343.88

2.3 1 2

0.191

S(year)

5

347.89

4.307

0.070

S(date)

2

354. 1 1

4.495

0.064

S(.) - constant survival

1

357.23

5.613

0.037

Stnest habitat type)

3

354.73

7.126

0.017

Stnest age)

2

356.94

7.326

0.016

seasonal decline varied from year to year
(Fig. 4B). Neither nest site habitat type (road
banks, trail banks, or natural slopes) nor nest age
appeared to have a major influence on DSR;
models incorporating these covariates received
little support and were ranked lower than the null
model of constant DSR (Table 1). Thus, DSR
appears to be largely independent of both nest site
characteristics and nest stage.

Apparent predation of the entire clutch (/? = 19)
or brood (/; = 26) was the leading cause of nest
failure, accounting for 83.3% of all observed nest
failures (/? = 54). Abandonment of intact nests
with eggs ill = 3) or young in = 2) accounted for
9.3% of nest failures. An additional three nests
(5.6% of failures) were abandoned after apparent
partial nest predation reduced the nest contents to
either one egg in = 2) or one nestling in = 1 ). I
also observed a single case of nest failure that may
have been a result of male egg removal; at one
nest where the banded male had disappeared and
been replaced by an unbanded immigrant, the
three eggs were found intact underneath the empty
nest. The female was observed building a new
nest with her new mate 6 days later.

DISCUSSION

Breeding’ Biology.— My results are generally
comparable to those ol other studies of the
breeding biology of the Slate-throated Whitestart
and other Mviohorns species. The March-June
breeding season coinciding with the end ot the dry

season and onset of the rainy season is consistent
with the seasonality reported lor Slate-throated
Whitestarts in Costa Rica (Skutch 1954, Shopland
1985, Stiles and Skutch 1989) and Venezuela
(Collins and Ryan 1994), and the Collared
Whitestart iM. torqnutns) at higher elevations in
Monteverde (Shopland 1985). The Brown-capped
Whitestart (M. hriinnkeps) in subtropical Argen-
tina (26° S) is also strongly seasonal with a
September-December nesting season coinciding
with a rainy season that begins in November
(Auer et al. 2007). However, recent work on the
Spectacled Whitestart iM. melanocephcdns) sug-
gests this species has a much more extended May-
December breeding season in Ecuador (00° 36' S)
that occurs during relatively dry periods (Greeney
et al. 2008).

Placement of Slate-throated Whitestart nests in
banks along roads and trails, and on open steep
slopes is consistent with other reports (Skutch
1954, Shopland 1985, Collins and Ryan 1994).
Skutch (1954) reported the male participated in
nest building at one of four nests where he
observed nest construction. In contrast, I observed
nest building at 31 nests where at least one
member of the pair was color banded and gender
was known, and males did not carry nest material
or assist in nest construction. Skutch’ s report,
however, is based on observations ot unbanded
birds, and he did not see both members of the pair
arrive at the nest simultaneously with nest
material. He wrote “1 had proof that both birds
worked at the nest, since I saw the second fly up
to it with material in its bill, before the first, which
had already added its contribution, had left the
clearing” (Skutch 1954:360). However, the am-
biguities in his description, and the possibility the
bird building the nest had slipped away unseen by
Skutch before returning with more nest material,
suggest the possibility of nest construction by
males of this species should be viewed skeptically
until more persuasive evidence is presented. Nest
construction in other species of Myioborns is
either by the female alone or, in cases where
gender could not be ascertained, a single adult
(Marshall and Baida 1974. Barber et al. 2000,
Greeney et al. 2008).

The average clutch size of 2.9 eggs is similar to
that reported by others in Costa Rica (Skutch
1954, 2.8 eggs, = 1 1 nests; Shopland 1985, 2.8
eggs, n = 33 nests), but slightly higher than the
mean clutch size of 2.4 eggs in = 1 nests)
reported for Slate-throated Whitestarts at

34

THE WILSON JOURNAL OF ORNITHOLOGY • Vul. 122, No. 1. March 2010

H Estimated daily survival rate H Estimated nest success

2000 2001 2002 2003 2004 All years

Year combined

FIG. 4. Annual and .seasonal variation in daily nest survival rate and overall nesting success of Slate-throated
Whitestarts from Monteverde, Costa Rica, 2000-2004. (A). Annual variation in estimated daily survival rate and estimated
nesting success. Mean it 95% confidence intervals are shown. (B) The effect of date on predicted daily nest survival rates
for 2000-2004, based on the most strongly supported model of nest survival, the S(year X date) interaction model
(Table 1). Predicted daily survival rates are shown only tor dates when active nests were being monitored.

10° 21' N, 67° 41' W in Venezuela (Collins and
Ryan 1994). The clutch size of the congeneric
Painted Whitestart (M. pictiis) in the Chiricahua
Mountains of Arizona is larger, averaging 3.1
eggs (Barber et al. 2000).

Mean egg size, 16.7 X 13.4 mm, was similar to
that reported by Skutch ( 1954) (17.5 X 13.4 mm)
and Collins and Ryan (1994) (16.3 X 12.3 mm).
Egg size was not related to laying sequence, unlike
many species of birds (e.g., Slagsvold et al. 1984,
Litjeld et al. 2005). Incubation periods of 13-

15 days have been reported for Slate-throated
Whitestarts in both Costa Rica (Skutch 1954) and
Venezuela (Collins and Ryan 1994), but 1 observed
incubation periods of 14-15 days. However, the
mean incubation period 1 recorded in Monteverde,
14.2 days, is nearly identical to that reported by
Skutch (1954) (14.4 day.s, // = 5 nests) and by
Collins and Ryan (1994) (14.3 days, /; = 4 nests).

The percentage of eggs surviving through the
end of incubation that hatched successfully was
93.8%, relatively high for a tropical bird (Koenig

Munmw BREEDING BIOLOCJY OB MY/OBORUS MINIAl'US

35

TABLE 2. Ijl'e-hislory traits of the Slate-throated Whitestart and three related north temperate zone warblers, the Red-
faeed Warbler, Canada Warbler, and Wilson's Warbler. The three related species were selected based on the phylogeny of
Lovette and Bermingham (2002). Data for Slate-throated Whitestarts based on this study. Data for other species derived
from Martin and Barber (1995), Conway (1999) and Ammon and Gilbert (1999).

Life-hi.sU)r>’ trail

Slate-tliroaled Whitestart

Red-l'aced Warbler

Canada Warbler

Wilson’s Warbler

Breeding range

Tropical montane forest

Subtropical montane
coniferous forest

Northern boreal forest

Northern boreal and
humid Pacific forest

Nest site

Ground

Ground

Ground

Usually ground, .some
low in vegetation

Mean clutch size

2.9

4.2

4.3

3.9 (Pacific coast)

4.4 (NE US)

5.3 (Alaska)

Incubation period

14-15 days

13 days

11-12 days

11-12 days

Nestling period

10-12 days (mean 1 1.3
days)

11-13 days (mean

12 days)

10-11 days

10-11 days

Post-Hedging

dependence

29 days after Hedging

No data

No data

24-25 days after
Hedging

1982). The low incidence of hatching failure at
Monteverde, a montane cloud-forest site with cool
ambient temperatures during the nesting season
(Clark et al. 2000), is consistent with the
hypothesis that one of the major cau.ses of loss
of egg viability and hatching failure is exposure to
high ambient temperatures during the laying
period (Stoleson and Beissinger 1999, Beissinger
et al. 2005, Cooper et al. 2006).

Duration of the nestling period averaged
1 1 .3 days, somewhat shorter than reported by
Skutch (1954) (12.4 days, /? = 4 nests) elsewhere
in Costa Rica, but similar to that found for
Monteverde by Shopland (1985) (1 1.7 days, n =
16 nests). The relatively greater role of females in
feeding nestlings was also noted by Skutch (1954)
at a nest of this species in Guatemala. Nestling
growth appears to be slightly more rapid than that
of Venezuelan nestlings (Collins and Ryan 1994),
and the average provisioning rate I observed at
nests (20.3/hr) is about twice the rate at which
food is delivered to Brown-capped Whitestart
nests in subtropical Argentina (Auer et al. 2007).

Few comparative data are available on duration
of post-fledgling care of young, but both Skutch
(1954) and Chipley (1976) noted the apparently
long period of juvenile dependence by Slate-
throated Whitestarts. My observation of adults
feeding 40-day-old juveniles, 29 days after Hedg-
ing is comparable to Shopland’s (1985) ob.serva-
tion of Collared Whitestarts feeding young
28 days after Hedging.

Nesting Success. — Daily survival rate (DSR) of
Slate-throated Whitestart nests in Monteverde was
0.968, yielding an estimated overall nesting

success of 40.3%, comparable to the 45% nest
success reported by Skutch (1954) for 11 nests.
Data on overall nest DSR are not available for
other species of Myiohorus, but Auer et al. (2007)
report the daily nest predation rate for Brown-
capped Whitestarts in subtropical Argentina was
0.023. I found that —85% of nest failure was due
to predation, and the comparable figure of daily
nest predation rate for Monteverde Slate-throated
Whitestarts would be 0.028.

Two other studies of passerines in Monteverde
have reported data on nest DSR and nesting
success. Sargent (1993) found that nest DSR for
Yellow-throated Euphonias {Euphonia hiritndina-
cea), a species that nests in many of the same road
banks favored by Slate-throated Whitestarts, was
0.972, yielding an estimated nest success of
33.6%. Nest DSR and estimated nesting success
for cavity-nesting House Wrens {Troglodytes
aedon) in Monteverde were considerably higher,
0.991 and 72.8%, respectively (Young 1994). My
observed nest DSR of 0.968 in the Costa Rican
highlands is greater than found in many studies of
lowland neotropical birds (e.g., Robinson et al.
2000, Ryder et al. 2008) but comparable to an
average nest DSR of 0.977 at 400 m elevation in
Braulio Carrillo National Park, Costa Rica
(Young et al. 2008).

Comparison to North Temperate Zone War-
blers.— Recent phylogenetic analyses of the Par-
Lilidae and Myiohorus (Lovette and Bermingham
2002, Perez-Eman 2005, Rabosky and Lovette
2008) indicate the three North Temperate Zone
warblers most closely related to the Myiohorus
whitestarts are Red-faced Warbler {Cardellina

36

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. I. March 20/0

riihrifrons), Wilson’s Warbler (Wilsonia pitsilla),
and Canada Warbler (W. canadensis). Molecular
evidence suggests the Hooded Warbler (W.
citrina) is not part of the clade that includes
Myiohonis and the other Wilsonia (Rabosky and
Lovette 2008).

The Slate-throated Whitestart in Monteverde, in
contrast to its relatives in the North Temperate
Zone, displays the standard suite of life-history
traits — small clutch size, prolonged incubation,
and perhaps prolonged nestling and post-fledging
care (Table 2) — that characterize tropical birds
(Martin 2004, Russell et al. 2004). Differences in
duration of incubation are particularly striking, as
the incubation period is 1-3 days longer for Slate-
throated Whitestarts than for Cardellina or
Wilsonia (Table 2). Recent studies have shown
that longer incubation periods of tropical birds are
at least partially a result of slower inherent
developmental rate of eggs (Tieleman et al.
2004, Robinson et al. 2008), although proximate
differences in nest attentiveness and incubation
temperature may have a role as well (Martin et al.
2007). The nestling period of Slate-throated
Whitestarts in Monteverde is only slightly longer
than for Wilsonia and apparently shorter than for
Cardellina (Table 2). It also appears to be slightly
shorter than for other Myiohorus whitestarts
(Barber et al. 2000, Auer et al. 2007, Greeney et
al. 2008). The relatively short nestling period in
Monteverde merits further study, but may be
affected by the relatively high rate at which food
is delivered to nests (20.3 feedings/hr), the
relatively rapid rate of nestling development,
and the moderately high daily nest predation rate
(0.028) (Martin et al. 2000).

Several studies have indicated the post-Oedging
period of juvenile dependence on parents is
typically greater in tropical birds than in related
temperate species (Skutch 1949, Russell et al.
2004, Tieleman et al. 2006). Data from Slate-
throated Whitestarts are consistent with this
hypothesis and suggest a slightly longer period
of post-Oedging care than for Wifson’s Warbler
(Table 2). However, additional data are needed to
reach a firm conclusion about the relative duration
of post-Hedging parental care in Myiohorus versus
related temperate-zone warblers.

ACKNOWLEDGMENTS

I thank Pawcl Cygan. Mark Galatowilsch, Piotr .lablohski,
Lydia Pctcll. Sarah Sargent. Tadck Stawarezyk, Ursula
Valdez, and Scott Wissinger for field assistance. Marvin

Hidalgo of the Estacion Biologica Monteverde (EBM)
allowed me to work on EBM properties, and many other
Monteverde landowners, including James Eorrest, Steven
and Betsy Kendall, Sharon Kinsman, Alan and Karen
Masters, Jim and Phoebe Richards, Bruce Young, and
Willow Zuchowski provided me with acce.ss to their land.
Alicia Mora and Erancisco Campos of the San Jose office of
the Organization for Tropical Studies provided generous
assistance with permits. Jack Eitniear and tw'o anonymous
reviewers provided helpful criticisms of an earlier version of
the manuscript. Einancial support was provided by grant
7194-02 from the Committee for Research and Exploration
of the National Geographic Society, and by the Allegheny
College Academic Support Committee.

LITERATURE CITED

American Ornithologists’ Union (AOU). 1998. Check-
list of North American birds. Seventh Edition.
American Ornithologi.sts’ Union, Washington, D.C.,
USA.

Ammon, E. M. and W. M. Gilbert. 1999. Wilson’s
Warbler (Wilsonia pusilla). The birds of North
America. Number 478.

Auer, S. K., R. D. Bassar, J. J. Eontaine, and T. E.
Martin. 2007. Breeding biology of passerines in a
subtropical forest in northwestern Argentina. Condor
109:321-333.

Barber, D. R., P. M. Barber, and P. G. Jabonski. 2000.
Painted Redstart (Myiohonis pictii.f). The birds of
North America. Number 528.

Beissinger, S. R., M. I. Cook, and W. J. Arendt. 2005.
The shelf life of bird eggs: testing egg viability using a
tropical climate gradient. Ecology 86:2164-2175.
Chipley, R. M. 1976. The impact of wintering migrant
wood warblers on resident insectivorous passerines in
a subtropical Colombian oak woods. Living Bird
15:1 19-141.

Clark, K. L., R. O. Lawton, and P. R. Butler. 2000. The
physical environment. Pages 15-38 in Monteverde:
ecology and conservation of a tropical cloud forest (N.
M. Nadkarni and N. T. Wheelwright. Editors). Oxford
University Press, New York, USA.

Collins. C. T. and T. P. Ryan. 1994. Notes on the
breeding biology of the Slate-throated Redstart
(Myiohorus minialn.s) in Venezuela. Ornitologia Neo-
tropical 5:125-128.

Conway, C. J. 1999. Canada Warbler (Wilsonia canaden-
sis). The birds of North America. Number 421.
Cooper, C. B., W. M. Hocuachka, T. B. Phillips, and A.
A. Dhondt. 2006. Geographical and seasonal gradi-
ents in hatching failure in Eastern Bluebirds Sialia
sialis reinforce clutch size trends. Ibis 148:221-230.
CUR.SON, J., D. Quinn, and D. Beadle. 1994. Warblers of
the Americas: an identification guide. Houghton
MifOin, Bo.ston, Ma.ssachusctls, USA.

Digiacomo, a. G. 1995. Two new species for Argentine
avifauna. Hornero 14:77-78.

Gill, E.. M. Wright, and D. Donsker. 2009. IOC World
bird names (Version 2.0). www.worldbirdnames.org/
(accessed 10 February 2009).

Mummc- BREEDING BIOLOCJY OB MYlOliORUS MINIA'IVS

37

Greeney, H. F., P. R. Martin, R. C. Dobhs. R. A. Gelis,

A. D. L. Bucker, and H. Montag. 2008. Nesting
ecology of the Spectacled WhitestaiT in Ecuador.
Ornitologia Neotropical 19:335-344.

Haber, W. A. 2000. Plants and vegetation. Pages 39-94 in
Monteverde: ecology and conservation of a tropical
cloud forest (N. M. Nadkarni and N. T. Wheelwright,
Editors). Oxford University Press, New York, USA.

jABtONSKi, P. G. 1999. A rare predator exploits prey escape
behavior: the role of tail-fanning and plumage contrast
in foraging of the Painted Redstart (Myiohorns pictns).
Behavioral Ecology 10:7-14.

Koenig, W. D. 1982. Ecological and social factors affecting
hatchability of eggs. Auk 99:526-536.

Lifjeld, J. T., a. Johnsen, and T. Petitguyot. 2005. Egg-
size variation in the Bluethroat (Liiscinki s. svecica):
constraints and adaptation. Journal fiir Ornithologie
146:249-256.

Lovette, 1. J. AND E. Bermingham. 2002. What is a wood
warbler? Molecular characterization of a monophyletic
Parulidae. Auk 119:695-714.

Marshall, J. and R. P. Balda. 1974. The breeding
ecology of the Painted Redstart. Condor 76:89-101.

Martin, T. E. 2002. A new view for avian life history
evolution tested on an incubation paradox. Proceed-
ings of the Royal Society of London, Series B
269:309-316.

Martin, T. E. 2004. Avian life-history evolution has an
eminent past: does it have a bright future? Auk
121:289-301.

Martin, T. E. and P. M. Barber. 1995. Red-faced
Warbler iCardellina nihrifrons). The birds of North
America. Number 152.

Martin, T. E., P. R. Martin, C. R. Olson, B. J.
Heidinger, and j. J. Fontaine. 2000. Parental care
and clutch sizes in North and South American birds.
Science 287:1482-1485.

Martin, T. E., S. K. Auer, R. D. Bassar, A. M. Niklison,
and P. Lloyd. 2007. Geographic variation in avian
incubation periods and parental influences on embry-
onic temperature. Evolution 61:2558—2569.

McCormack, J. E., G. Ca,staneda Guyasamin, B. MilA,
and F. Heredia Pinada. 2005. Slate-throated Red-
starts (Myiohorus miniatus) breeding in Maderas del
Carmen, Coahuila, Mexico. Southwestern Naturalist
50:501-503.

Mumme, R. L. 2002. Scare tactics in a neotropical warbler:
white tail feathers enhance flush pursuit foraging
performance in the Slate-throated Redstart (Myiohorus
minialus). Auk 119:1024—1035.

Mumme, R. L., M. L. Galatowitsch, P. G. Jabionski, T.
M. Stawarczyk, and J. P. Cygan. 2006. Evolution-
ary significance of geographic variation in a plumage-
based foraging adaptation: an experimental test in the
Slate-throated Redstart (Myiohorus minialus). Evolu-
tion 60: 1086-1097.

Nadkarni, N. M. and N. T. Wheelwright. 2000.
Monteverde: ecology and conservation of a tropical
cloud forest. Oxford University Press, New York,
USA.

Perez-Eman, j. L. 2005. Molecular phylogenetics and

biogeography ol ihe neotropical redstarts (Myiohoi us,
Aves, Parulinae). Molecular Phylogenetics and Evo-
lution 37:51 1-528.

Rabosky, D. L. and I. J. Ldveite. 2008. Dcnsity-
dependenl diversification in North American wood
warblers. Proceedings of the Royal Society ol London,
Series B 275:2363-2371.

Ricklefs, R. E. and M. Wikedski. 2002. The physiology/
life history nexus. Trends in Ecology and Evolution
17:462^68.

Ridgely, R. S. and G. Tudor. 1989. The birds of South
America. Volume I. The o.scine passerines. University
of Texas Press, Austin. USA.

Robinson, W. D., T. R. Robinson, S. K. Robinson, and J.
D. Brawn. 2000. Nesting success of understory forest
birds in central Panama. Journal of Avian Biology
31:151-164.

Robinson, W. D., J. D. Styrsky, B. J. Payne, R. G.
Harper, and C. F. Thompson. 2008. Why are
incubation periods longer in the tropics? A common-
garden experiment with House Wrens reveals it is all
in the egg. American Naturalist 171:532-535.

Rotella, j. 2008. Nest survival models. In Program
MARK: a gentle introduction, Seventh Edition (E.
Gooch and G. White, Editors), www.phidot.org/
software/mark/docs/book/ (accessed 10 February
2009).

Russell, E. M., Y. Yom-Tov, and E. Geffen. 2004.
Extended parental care and delayed dispersal: north-
ern, tropical and southern passerines compared.
Behavioral Ecology 15:831-838.

Ryder, T. B., R. Duraes, W. P, Tori, J. R. Hidalgo, B. A.
Loiselle, and j. G. Blake. 2008. Nest survival for
two species of manakins (Pipridae) in lowland
Ecuador. Journal of Avian Biology 39:355-358.

Sargent, S. 1993. Nesting biology of the Yellow-throated
Euphonia: large clutch size in a neotropical frugivore.
Wilson Bulletin 105:285-300.

Shopland, j. M. 1985. Facultative following of mixed-
species flocks by two species of neotropical warbler.
Dissertation. University of Chicago, Illinois, USA.

Skutch, a. F. 1949. Do tropical birds rear as many young
as they can nourish? Ibis 91:430-455.

Skutch, A. F. 1954. Life histories of Central American
birds: families Fringillidae, Thraupidae, Icteridae,
Parulidae and Coerebidae. Pacific Coast Avifauna.
Cooper Ornithological Society. Berkeley, California,
USA.

Slagsvold, T., j. Sandvik, G. Rofstad, O Lorentsen,
AND M. Husby. 1984. On the adaptive value of
intraclutch egg-size variation in birds. Auk 101:685-
697

.Stiles, F. G. and A. F. Skutch. 1989. A guide to the birds
of Costa Rica. Cornell University Press. Ithaca. New
York, USA.

.Stole,son. S. H. and S. R. Beissinger. 1999. Egg viability
as a constraint on hatching synchrony at high ambient
temperatures. Journal of Animal Ecology 68:951—962.

Tieleman, B. 1., .1. B. Williams, and R. E. Ricklefs.
2004. Nest attentiveness and egg temperature do not

38

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. I, March 2010

explain the variation in incubation period in tropical
birds. Functional Ecology 18:571-577.

Tieleman, B. I., T. H. Dukstra, J. R. Lasky, R. A.
Mauck, G. H. Visser, and J. B. Williams. 2006.
Physiological and behavioral correlates of life-history
variation: a comparison between tropical and temper-
ate zone House Wrens. Functional Ecology 20:491-
499.

White, G. C. and K. P. Burnham. 1999. Program MARK:

survival estimation from populations of marked
animals. Bird Study 46 (Supplement): 120-138.

Young, B. E. 1994. The effects of food, nest predation and
weather on the timing of breeding in tropical House
Wrens. Condor 96:341-353.

Young, B. E., T. W. Sherry, B. J. Sigel, and S.
WoLTMANN. 2008. Nesting success of Costa Rican
lowland rain forest birds in response to edge and
isolation effects. Biotropica 40:615-622.

The Wilson Joiinial of Ornithology 122(1):39 45, 2010

BREEDING BIOEOGY OF KEEP GUEES ON THE BRAZIEIAN COAST

GISELE PIRES de MENDON^A DANTAS'-^ AND JOAO STENGHEL MORGANTE'

ABSTRACT. — We studied clutch size, hatching and fledging success, and time necessary lor chick Kelp Gulls {Lams
clominicanns) to leave the nest throughout two breeding seasons (2004 and 2005) on Guararitama Island, Sao Paulo, Brazil.
We followed 93 nests in 2004 and 97 nests in 2005. The average (± SD) clutch size was 2.09 ± 0.64 in 2004 and 1 .93 ±
0.59 in 2005. Hatching success was 74% in 2004 and 53% in 2005, and fledging success was 54% in 2004 and 58% in
2005. Chicks grew quickly, following the linear equation y(t) = 61g + 17.03g X age (in days), and began to fly at 40 days.
Received 1 1 August 2008. Accepted 28 August 2009.

The Kelp Gull (Larits clominicamis) (Family
Laridae) is widely distributed in the Southern
Hemisphere, occumng in South America, South
Africa, Australia, New Zealand, the sub-Antarctic
islands, and the Antarctic Peninsula (Burger and
Gochfeld 1996). It is a resident and common
species in Brazil, occurring from Rio de Janeiro to
the coast of Rio Grande do Sul (Sick 1997). This
species usually establishes colonies with few to
thousands of breeding pairs on islets and islands
near the continent and on continental locations
(Burger and Gochfeld 1996). Studies have report-
ed a large increase in Kelp Gull populations in
recent decades (Steele and Hockey 1989, Quin-
tana and Yorio 1998). The Kelp Gull population
at Peninsula Valdez, Argentina, increased from
3,200 to 6,500 pairs over a 10-year period
(Quintana and Yorio 1998). This population
growth has been attributed to enhanced post-
fledging survival of juveniles due to the provision
of supplementary food through anthropomorphic
activities (Steele and Hockey 1989, 1991; Bertel-
lotti and Yorio 1999).

Some authors believe gull population growth
has been involved in displacement of other
seabird species from breeding sites (Thomas
1972; Burger 1979, 1985; Quintana and Yorio
1998). Kelp Gulls have been observed to prey on
eggs and chicks of Magellanic Penguin (Sphenis-
cus magellanicus). Imperial Shag (Leitcocarho
alriceps). Royal Tern (Sterna maxima), Cayenne
Tern (S. eiirygnatha) and Southern Giant Petrel
(Macronectes giganteus) (Yorio et al. 1998).
Effects of gulls on coastal wildlife are not
confined to other seabirds. Thomas (1988) and

' Institute de Biociencias, Universidade de .Sao Paulo,
Brazil, Rua do Matao, 277, Sao Paulo, SP, Brazil, Cep.
05508-090.

^Corresponding author;
e-mail: giseledantas@mail.icav.up.pt,
dantasgpm@gmail.com

Rowntree et al. (1998) observed Kelp Gulls
injuring Right whales (Enhalaena australis) by
picking off skin and fat when the whales emerge
at the surface to breathe. These authors argue that
intense harassment by gulls induces Right whales
to abandon breeding areas before their young are
sufficiently strong for the open sea.

Little is known about the breeding biology of
the Kelp Gull despite its wide distribution and its
great impact on other seabirds and mammals.
(Crawford et al. 1982, Soares and Fonseca
Schielfler 1995, Yorio et al. 1996, Yorio and
Borboroglu 2002). Breeding biology studies are
essential to understand the dynamics of a
population, because changes in population size
may result from processes of reproduction,
survival, and migration (Doherty and Grubb
2002). Reproductive success is associated with
several factors, including ecological conditions
(intra- and inter-specific interactions, food avail-
ability, breeding site, etc.), and weather conditions
(humidity, temperature, and wind) (Paynter 1949,
Borboroglu and Yorio 2004). Studies that seek to
understand long term survival and mortality rates
in a population may be efficient for monitoring
environmental change. An estimate of reproduc-
tive success is also essential for evaluating
population growth. This information is needed to
create future conservation programs for seabird
communities.

We studied the breeding biology of the Kelp
Gull on the south Brazilian coast. Our objectives
were to: (1) measure the incubation period, (2)
ascertain clutch size, (3) measure predation rate,
(4) identify predators, (5) calculate the length ot
time required for chicks to become independent,
and (6) measure the growth rate of chicks.

METHODS

Study Area. — Data were collected on Guarar-
itama Islet, Sao Paulo, Brazil (24° 23' S,

39

40

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. I, March 2010

46°59'W). This islet is 1.9 km from the
continent and has an area of 1 .2 ha with a rocky
substrate, and shrub and grass-herb vegetation.
Campos et al. (2004) estimated the Lams
dominicanus population on Guararitama to be
100 breeding pairs.

Field Procedures. — We visited the islet every 5
to 7 days during 2004 and 2005 from the end of
June to mid-November We recorded egg-laying
dates, clutch sizes, egg measurements (length,
width, and mass), chick measurements (mass,
wing length, tarsus length, and beak length) and
hatching dates. We measured eggs and chicks to
the nearest 0.1 mm using calipers and to the
nearest 1.0 g using a Pesola scale.

We marked all Kelp Gull nests on Guararitama
Islet with satin tape. We marked eggs with a
waterproof pen and chicks with satin tape, until
they reached an appropriate size to receive a metal
ring provided by CEMAVE/IBAMA (Instituto
Brasileiro do Meio Ambiente e Recursos Naturais
Renovaveis). Nests were kept under surveillance
throughout the breeding season to estimate
survival, predation, and abandonment rates. A
nest was considered active when it was well
formed and contained eggs. It was considered to
have been predated when broken eggs were found
inside or near the nests, or when eggs disappeared
without sign of hatching (e.g., excrement of the
chicks). Nests were considered abandoned when
eggs remained for 40 days without hatching. We
defined hatching success as the number of chicks
per eggs laid and fledging success as the number
of chicks alive after 28 days per eggs laid
(Mougeot and Bretagnolle 2006).

Statistical Analysis. — ^Egg volume was calculat-
ed following Hoyt (1979): V = length X width^
X 0.51. We estimated the number of females by
the number of breeding pairs (bx), assuming that
each nest consisted of one pair. The estimated
number of females was used to calculate fecundity
(fx) and fertility (mx) (Pianka 2()()()):

bx = number of nests,

fx = number of eggs laid/number of females,
and

mx = number of surviving eggs/number of
females.

Two distinct peaks were observed in egg laying
in 2004 and 2005. The 2004 breeding sea.son was
subdivided into an early and a later period, and the
2005 breeding season was divided into early,
intermediate, and later periods, because the first

peak was later in 2005. Differences within and
between breeding seasons in egg volume and
clutch size were evaluated with a t-test. Survival
rates of offspring, hatching success, and fledging
success were compared between breeding seasons
using a yj test.

Chicks captured on the day of hatching and at
least one additional time during the breeding
season (/? = 17) were u.sed to estimate the growth
equation. We fitted the von Bertalanffy, Gom-
pertz, and logistic three-parameter growth equa-
tions (Ricklefs 1967) to the data on chick mass
using least squares procedures to analyze chick
growth patterns. We analyzed the variance of the
residuals distribution to identify which models
best fit the data, and chose the model with low
variance. We estimated the age of all other chicks
captured only once during the breeding season
based on the chosen growth equation. Growth
curves that best fit the body size measurements
(beak, tarsus, and wing length) were calculated
using von Bertalanffy, Gompertz, and logistic
three-parameter functions.

RESULTS

The Kelp Gull breeding season on Guararitama
Islet began in mid-June and ended in early
November. The first eggs were found on 24 June
in 2004 and on 30 June in 2005. Egg laying
continued until October with the last eggs being
laid on 31 October in 2004 and 28 October in
2005. We recorded 195 eggs in 93 nests in 2004
and 172 eggs in 97 nests in 2005. Egg laying in
both years showed two peaks during the breeding
sea.son. The first peak in 2004 occuiTed in July (93
eggs) and the second in September (102 eggs).
The first peak in 2005 was in August (53 eggs)
and the second in October (40 eggs) (Fig. 1 ).

Mean (± SD) clutch size was 2.09 ± 0.64 (/; =
93) in 2004 and 1.93 ± 0.59 (/? = 97) in 2005.
The difference between the years was not
significant (/ = 1.77, P = 0.07). The average
(± SD) incubation length was 24.12 ± 4.48 days
in 2004 and 22.78 ± 3.27 days in 2005. The
number of breeding females (bx) was 93 in 2004
and 97 in 2005. Fecundity (fx) was 2.096 in 2004
and 1.773 in 2005, and fertility (mx) was 1.387 in
2004 and 0.422 in 2005.

Estimated egg volume differed between the two
peaks of each breeding season (2004: t = 4.34, P
< O.OOl; 2005: / = 2.24, P = 0.026). This
difference cannot be attributed to the first peak
consisting mainly of the first eggs laid by each

Number of eggs laid

Dantas unci Mor^ranlc • BREEDINCi BIOEOGY Ol KllLP CiUl.ES

41

80

FIG. E Number of eggs laid throughout the Kelp Gull
breeding season in 2004 (gray bar) and 2005 (black bar), on
Guararitama Islet, Sao Paulo, Brazil.

female, and the second peak consisting of the
second and third eggs laid, because all eggs from
the first peak hatched during the period of time
between the peaks. The average volume of eggs
laid during the first peak in 2004 was 88.11 ±
8.28 cm^ while that of eggs laid during the
second peak was 83.80 ± 12.68 cm^ The average
egg volume during the early period in 2005 was
87.01 ± 8.12 cml The average egg volume
during the first peak (intermediate period) was
87.18 ± 9.01 cm-\ and the average egg volume
during the second peak (later period) was 82.43 ±
12.97 cnr\ The difference in egg volume between
2004 and 2005 was significant (t = 1.99, P =
0.046) with a mean (± SD) of 86.32 ± 10.37 cm^
in 2004 and 84.13 ± 10.75 cm^ in 2005 (Fig. 2).

The first peak of the season produced greater
hatching success than the second peak in both
years (2004: f = 10.09, P = 0.006; 2005: f =
7 IP P = 0.03) (Table 1). Hatching succe.ss was
greater in 2004 than in 2005 {yj = 18.13, F <
0.01) (Fig. 3). Predation was the principal cause
of egg loss (25.6% in 2004 and 69.9% in 2005)
and death of chicks (34.9% and 5.8%). The
principal predators of Kelp Gulls were Black
Vultures (Coragyps atratus) on eggs and adult
Kelp Gulls on chicks.

Heat stress also affected loss of eggs and
chicks. Air temperature can reach 33° C during
the late breeding season in Guararitama and, on
hot days, the chicks appear lethargic and had
difficulty breathing. Thus, they become easy prey
for vultures and other raptors. Environmental data
from Iguape Meteorological Station, Cptec/INPE,
(www.cptec.inpe.br) tor the breeding period
indicated that 2005 was hotter than 2004, but
the years did not differ in precipitation (temper-

120,0

86.7 -

53,3 -

A

20,0

120.0

86,7 -

53,3

20.0

B

Early Laler

2004

Early Intermediate Later

2005

120.0

86,7 -

53,3 -

20,0

2004 2005

FIG. 2. Kelp Gull egg volume (x ± SD. cmb through-
out the 2004(A) and 2005 (B) breeding seasons ami tor both
years (C), on Guararitama islet, Sao Paulo. Brazil.

ature: / = 8.78, P = O.OOO; precipitation: t = 0.53,
P = 0.59).

Field observations indicated that chicks grew
rapidly, remaining in the nest until they were
5 days of age. When chicks were 5 to 1 5 days of
age, they moved around the island and become
more visible to predators. Chicks were able to
avoid predators after 30 days of age. They reached

42

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. I. March 2010

TABLE 1,

Hatching success rates

throughout the breeding

season of Lams

dominicaniis in 2004

and 2005, at

Guararitama Islet, Sao Paulo, Brazil.

Year

Period

Successful eggs

Predated eggs

Abandoned eggs

Totals

2004

Early

65 (69.9%)

16 (17.2%)

12 (12.9%)

93

Later

64 (62.7%)

34 (33.3%)

4 (3.9%)

102

Totals

129 (66.1%)

50 (25.6%)

16 (8.2%)

195

2005

Early

14 (35.9%)

23 (58.9%)

2 (5.1%)

39

Intermediate

18 (20.0%)

69 (76.7%)

3 (3.3%)

90

Later

9 (20.5%)

29 (65.9%)

6 (13.6%)

44

Totals

41 (23. 7%)

121 (69.9%)

1 1 (6.4%)

173

an average mass and body size that made them
difficult to catch by predators at this time. Chicks
were flying at 40 days of age. The survival rate to
fledging was 54% in 2004 and 58% in 2005 (X- =
1.79; P = 0.1798).

Three possible growth functions were com-
pared (i.e., von Bertalanffy, Gompertz, and
logistic functions). The variance of the residual
distributions indicated the models did not differ
significantly, and we present results only from the
logistic equation. The logistic equation for body
weight of 17 Kelp Gull chicks is weight (mass) =
A/( l-hB*EXP{ — C*(days)}) (parameter values are
in Table 2). We were able to predict the ages of
chicks from their measured body mass using this
curve. When chick mass increased linearly
between 5 and 35 days of age, we derived a
simplified formula: y = yo + cit, where yo is birth
mass (mean mass on day of hatching = 61 g); a is
the estimated growth rate of 17.03 g/day, and t is
age in days (Fig. 4). We ob.served rapid growth in
all of the measured traits (beak, tarsus, and wing
length) (Fig. 5). These measurements were fitted
to growth models, and there were no significant
differences among models (parameters estimated
for the logistic three-parameter model. Table 2).

FIG. .1. Hatching and fledging succes.s rate.s (%) of
Kelp Gulls during the 2004 and 200.‘i breeding seasons, on
Guararitama Islet. .Sao Paulo. Brazil.

DISCUSSION

The Kelp Gull breeding season on Guararitama
Islet was similar in both years, starting in mid-
June and ending in early November. This same
pattern was reported for the Santa Catarina coast
in southern Brazil (Soares and Fonseca Schielfler
1995, Branco 2003), although Prellvitz et al.
(2009) reported egg laying began on I June on
Moleques do Sul, Santa Catarina. The Kelp Gull
in Argentina reproduces from November to
January in central and northern Patagonia (Bor-
boroglu and Yorio 2004) and in mid-October in
the Valdez Peninsula (Yorio and Borboroglu
2002). Yorio and Borboroglu (2002) report the
Kelp Gull breeding season shows a direct
relationship with latitude with gulls breeding
earlier in more northern colonies. Laying stages
may be related to latitude and to climatic
conditions of the breeding localities (Fordham
1964).

Clutch size of Kelp Gulls on the Brazilian coast
varies from one to three eggs. This clutch size has
been reported for Kelp Gulls in Brazil, Argentina,
and South Africa (Saliva and Burger 1989, Soares
and Fonseca Schielfler 1995, Branco 2003,
Borboroglu and Yorio 2004, Prellvitz et al.
2009), and tor other species oi Lams in temperate
regions (Paynter 1949, Kadlec and Drury 1968).
The average clutch size in our study (2.09 in 2005

TABLE 2. Logistic growth model parameters. A =
asymptotic growth rate; B = constant integration growth; C
= growth rate of Kelp Gull chicks (/; = 17), at Guararitama
Islet. .Sao Paulo, Brazil.

A

B

c

Body mass, g

1,165.34

14.25

0.1 1

Wing length, mm

1.35.. 30

4.86

9.41

Tarsus length, mm

77.88

1.78

0.06

Beak length, mm

47.63

1.41

5.69

Daniels (111(1 Morf^antc • l}REE[)INCj BIOLOGY OL KELP GULLS

43

FIG. 4. Growth (mass) of Kelp Gull chicks on
Guararitama Islet. Sao Paulo, Brazil.

and 1.95 in 2005) is comparable to that of other
studies in Patagonia, Argentina (2.3-2.5 eggs,
Yorio and Borboroglu 2002), southern Brazil (2.3
eggs, Soares and Fonseca Schielfler 1995; 1.98-
2.26 eggs, Branco 2003; 1.97 eggs, Prellvitz et al.
2009), and the Antarctic Peninsula (2.5 eggs,
Fraser 1989). There were two peaks of egg laying
at Guararitama Islet in both years. However,
Yorio and Borboroglu (2002) in Argentina and
Saliva and Burger (1989) in South Africa
observed only one laying peak for Kelp Gulls.
We believe the bimodal egg laying at Guararitama
Islet may be a consequence of environmental
conditions, including temperature and humidity.
Prellvitz et al. (2008) reported Kelp Gull colonies
are smaller in warmer and less productive waters,
suggesting that environmental conditions can influ-
ence size and reproductive success of colonies.

Egg size is related to the ability of the female to
obtain food resources throughout the laying
period (Herbert and Barclay 1988). We observed
significant differences in egg volume between the
two laying peaks during each breeding season
(Fig. 2). Females produced larger eggs during the
early period than during the later period. Skorka et
al. (2005) have shown that food availability is

reduced late in the breeding sea.son. Thus, older
pairs breeding early in the season are expected to
have better success/survival than younger pairs
due to greater experience and provisioning skills.
Additionally, smaller average egg volume was
observed in 2005 than in 2004 (t = 1.99; P =
0.046). These differences may reflect environ-
mental changes that inlluence chick survival.

The optimal temperature for bird embryo
development is 37-38° C; temperatures above

FIG. Growth (tarsus, wing, beak length) of Kelp Gull
chieks on Guararitama Islet, Sao Paulo. Brazil, during the
2004 and 200.5 breeding seasons.

40.5° C are deadly (Gill 1994). The air temper-
ature during the 2004 breeding season ranged
from 17° to 36° C, and during 2005 from 17.5° to
40° C (Iguape Meteorological Station). Borbor-
oglu and Yorio (2004) reported the breeding
season of Kelp Gulls at Isla Vernacci, southwest-

44

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. /, March 2010

ern Argentina, ends in the middle of January,
when the temperature reaches 40° C. By that time,
the chicks are already flying. Salzman (1982)
reported that temperatures above 32° C are
harmful to chicks of Lams occidentalis, often
causing death. Salzman ( 1982) reported that stress
caused by excessive heat is one of the main causes
for egg loss and chick mortality.

The high mortality observed in 2005 may have
been due to high temperatures recorded through-
out the breeding season, which left nests exposed
to predators. The principal cause of egg loss
(25.6% in 2004, 69.9% in 2005) at Guararitama
was predation by vultures, while the principal
cau.se of chick mortality (34.9% in 2004, 25.0% in
2005) was predation by adult Kelp Gulls.
Borboroglu and Yorio (2004) reported egg loss
of 28% and chick morality of 50%. These authors
also reported chick predation by adults of the
same species. They believe that chick predation
by adults of the same species occurs when chicks
invade breeding territories, searching for shade
and protection under vegetation. Predation can be
reduced by parental care and by selection of a
more hidden site for nest building (Borboroglu
and Yorio 2004).

Other predators observed to prey on eggs and
chicks of Kelp Gulls includes Subantartic Skuas
(Stercorariiis antarcticiis) and Southern Crested
Caracara (Caracara plancus) (Yorio et al. 1995,
Sick 1997, Yorio and Borboroglu 2002). Howev-
er, Prellvitz et al. (2009) observed parental
negligence and starvation to be the main mortality
factor for eggs and chick Kelp Gulls, respectively,
on Deserta Island. The number of eggs abandoned
and chick death by starvation was small (<5%) in
our study. Pierotti and Bellrose (1986) reported
that high los.ses from parental negligence or egg
fertility could indicate lower food availability or
inferior food quality. Prellvitz et al. (2009)
reported 51% hatching success on Deserta Island,
similar to that in 2005 on Guararitama Islet.
However, hatching success on Guararitama in
2004 was 74%, similar to that reported by Yorio
and Borboroglu (2002) at Peninsula Valdez,
Argentina. These differences may be related to
environmental variability in temperature and
humidity. Stable years can have great hatching
and fledging success and hotter years can result in
great mortality from predation or starvation.

Kelp Gull chicks grow rapidly and this species
displays parental care during incubation and
feeding of chicks. Borboroglu and Yorio (2004)

reported that parental care gradually decreases
over the breeding season, increasing predation
risk. Kelp Gull chicks start to move around the
island after 5 days of age. This movement, when
they are only a few days of age, seems to be a
behavioral characteristic (Yorio et al. 1996).
Chicks weigh 800 g at 35 days and are already
flying. Capture by predators at this phase is
difficult. Yorio and Borboroglu (2002) also
observed rapid growth of chick Kelp Gulls in
colonies on the coast of Argentina. Prellvitz et al.
(2009) reported that most chick mortality on
Deserta Island, Brazil, occurred in the first week
of life. This rapid growth rate may represent an
evolutionary adaptation for escape from predators.

Kelp Gulls breed on the Brazilian coast during
the winter, probably to avoid higher summer
temperatures and to seek better conditions for
embryo development. Individuals that breed early
in the season have an advantage over pairs
breeding later in the .season. The main predators
of eggs and chick Kelp Gulls are Black Vultures.
who.se impact is greatest at the egg stage, and
adult Kelp Gulls, which prey on young chicks.
Chicks have rapid growth, probably as a means of
avoiding predators. Despite the high predation
rate observed in 2005 (a hotter year). Kelp Gulls
had a high survival rate (54% in 2004 and 58% in
2005). If these chicks survive to the reproductive
stage, this suggests an incremental population
growth rate of 50% each year. This population
growth may pose a serious problem for the
conservation of other seabirds.

ACKNOWLEDGMENTS

This work was funded hy grants from the Funda^ao de
Amparo a Pesquisa de Sao Paulo (Fapesp; processo O.'i/
354.J8-4) and a Ph.D. fellowship grant to Gi.sele Dantas
(processo 03/01599-1). Wc thank the Instituto Florestal de
.Sao Paulo, Esta9ao Ecologica .lurei-ltatins for logistical
support (783/2003), and CEMAVE/IBAMA for metal rings
and for the license to perform the study (permit 1060). We
also thank several friends for help with fieldwork, and M.
A. Aires, J. O. Braneo, C. Y. Miyiaki, M, C, Arias, F. A.
Santos, and anonymous reviewers for helpful comments on
earlier drafts of the manuscript.

LITERATURE CITED

BERTKi.i.orri, M. and P. Yorio. 1999. Spatial and temporal
patterns in the diet of Kelp Gull in Patagonia. Condor
101:790-798.

Borboroglu, P. G. and P. Yorio. 2004. Flabitat
requirements and selection by Kelp Gulls (Lants
dontinicanus) in central and northern Patagonia,
Argentina. Auk 121:243-252.

D<intas and Morganic • BREEDING BlOECXiY OE KELP CJUI.ES

45

Branco. J. O. 2003. ReprocliKj'ao das aves niariiihas nas
ilhas costeiras de Santa Catarina, Brazil. Revista
Brasileira de Zoologia 20:619-623.

Burger, J. 1979, Competition and predation; Herring Gulls
versus Laughing Gulls. Condor 81:269-277.

Burger, J. 1983. Advantages and disadvantages of mixed-
colonies of seabirds. Proceedings of the International
Ornithological Congress 18:905-918.

Burger, J. and M. Gochfeld. 1996. Laridae. Handbook of
the birds of the world. Volume 3, Hoatzin to Auks (J.
del Hoyo, A. Elliott, and J. Sargatal, Editors). Lynx
Edicions, Barcelona, Spain.

Campos, F. P., D. Paludo, P. J. Faria, and P.
Martuscelli. 2002. Aves Insulares, Residentes on
Migratorias das ilhas costeiras do Estado de Sao Paulo.
Pages 57-82 in Aves Marinhas e Insulares Brazileiras;
bioecologia e conserva9ao. (J. O. Branco, Editor).
Univali, Itajaf, Santa Catarina, Brazil.

Crawford, R. J. M., J. Cooper, and P. A. Shelton. 1982.
Distribution, population size, breeding and conserva-
tion of the Kelp Gull in southern Africa. Ostrich 53:
164-177.

Doherty, P. F. and T. C. Grubb. 2002. Survivorship of
permanent resident birds in a fragmented forested
landscape. Ecology 83:844—857.

Fraser, W. R. 1989. Aspects of the ecology of Kelp Gulls
{Lams dominicanits) on Anvers Island, Antarctic
Peninsula. Dissertation. University of Minnesota,
Minneapolis, USA.

Fordham, R. a. 1964. Breeding biology of Kelp Gull. I.
Pre-egg and egg stage. Notornis 1 1:1 10-126.

Gill, F. B. 1994. Ornithology. W. H, Freeman and
Company, New York, USA.

Herbert, P. N. and R. N. Barclay. 1988. Parental
investment in Herring Gulls; clutch apportionment and
chick survival. Condor 90:332-338.

Hoyt, D. F. 1979. Practical methods of estimating volume
and fresh weight of bird eggs. Auk 96:73-77.

Kadlec, J. A. and W. H. Drury. 1968. Structure of the
New England Herring Gull population. Ecology 49:
644-676.

Mougeot, F. and V. Bretagnolle. 2006. Breeding
biology of the Red Kite Milvns milvus in Corsica. Ibis
148;436-f48.

Paynter Jr., R. A, 1949. Clutch size and egg and chick
mortality of Kent Island Herring Gulls. Ecology 30;
146-166.

PiANKA, E. 2000. Evolutionary ecology. Sixth Edition.
Addison Wesley Educational, San Francisco, Califor-
nia, USA.

PiEROTTi. R. AND C. A. Bellrose. 1986. Proximate and
ultimate cau.sation of egg size and the “third chick
disadvantage” in the Western Gull. Auk 103:410^07.

Prellvitz, L. j., R. I. Hogan, and C. M. Vooren. 2008. O
parai'so do gaivotao. Ciencia Hoje 43:65-67.

Prellvitz, L. J., R. I. Hogan, and C. M. Vooren. 2009.
Breeding biology of Kelp Gulls (Larns dorninicaniis)

on Deserta Island, southern Brazil. Ornitologia Neo-
tropical 20:61-72.

Quintana, F. and P. Yorio. 1998. Competition for nest
sites between Kelp Gulls (Larns doniinicanns) and
terns (Sterna maxima and S. eurygnatha) in Patagonia.
Auk 1 15:1068-1071.

Ricklefs, R. E. 1967. A graphical method ol fitting
equations to growth curves. Ecology 48:978-983.

Rowntree, V. J., P. MacGuinne.ss, K. Mar.shall, R.
Payne, M. Sirone, and J. Seger. 1998. Increased
harassment of Right whales (Euhalena australis) by
Kelp Gulls (Larns dominicanns) at Peninsula Valdez,
Argentina. Marine Mammal Science 14:99-1 15.

Saliva, J. E. and J. Burger. 1989. Effects of experimental
manipulation of vegetation density on nest site
selection in Sooty Terns. Condor 91:689-698.

Salzman, a. G. 1982. The selective importance of heat
stress in gull nest location. Ecology 63:742-751.

Sick, H. 1997. Ornitologia Brasileira. Nova Fronteira
Publisher, Rio de Janeiro, Brazil.

Skorka, P., j. D. Wojcik, and R. Nartyka. 2005.
Colonization growth of Yellow-legged Gulls Larns
cachinnans in southeastern Poland; causes and influ-
ence on native species. Ibis 147:471^82.

Soares, M. and A. Fonseca Schielfler. 1995. Reprodu-
9ao de Larns dominicanns (Aves, Laridae), na Ilhota
Galheta, Laguna, SC, Brazil, Arquivos de Biologia e
Tecnologia 38:313-316.

Steele, W. K. and P. A. R. Hockey. 1989. Population
size, distribution and dispersal of Kelp Gulls in the
southwestern Cape, South Africa. Ostrich 61:97-106.

Steele, W. K. and P. A. R. Hockey. 1991. The human
influence on Kelp Gull diet assessed using stable
carbon isotope analysis of bone collagen. South
African Journal of Science 87:273-275.

Thomas, G. J. 1972. A review of gull damage and
management methods at nature reserves. Biological
Conservation 4:1 17-127.

Thomas, P. O. 1988. Kelp Gulls. Larns dominicanns. are
parasites on flesh of Right whale, Enbalacna australis.
Ethology 79:89-103.

Yorio, P. and P. G. Borboroglu. 2002. Breeding biology
of Kelp Gulls (Larns dominicanns) at Golfo San Jorge,
Patagonia. Argentina. Emu 102:257-263.

Yorio, P.. M. Bertellotti, and F. Quintana. 1995.
Preference for covered nest sites and breeding success
in Kelp Gulls (Larns dominicanns). Marine Ornithol-
ogy 23: 1 2 1 — 128.

Yorio, P.. P. D. Boersma, and S. Swan. 1996. Breeding
biology of the Dolphin Gull at Punta Tombo,
Argentina. Condor 98:208-215.

Yorio. P., M. Bertelloti, P. Gandini, and E. Frere.
1998. Kelp Gulls Larns dominicanns breeding in the
Argentina coast: population status and relationship
with coastal managemeni and conservation. Marine
Ornithology 26:1 1-18.

The Wilson Journal of Ornithology 122( 1 ):46-50, 2010

COOPERATIVE BREEDING OF THE SOCIETY KINGFISHER

(TODIRAMPHUS VENERATUS)

DYLAN C. KESLER,'^ THOMAS GHESTEMMEr
EMMANUELLE PORTIER,^ AND ANNE GOUNE

ABSTRACT. — We present evidence of cooperative breeding in Society Kingfishers (Todiramphus veneratus). Groups of
three kingfishers were observed cooperatively excavating nest cavities, incubating, and provisioning young. Three bird
groups also comprised half of visual observations recorded during island-wide point transect surveys that occurred during
the onset of breeding. Society Kingfisher densities were an order of magnitude greater than other Pacific kingfisher
populations, which may lead to resource saturation and evolution of cooperation in this species. Our results lend insight into
cooperative behaviors and facilitate conservation of birds in French Polynesia. Received 30 July 2009. Accepted 12 October
2009.

Cooperative breeding has been a focus of
behavioral ecologists and evolutionary biologists
for decades (Skutch 1935, Brown 1987, Stacey
and Koenig 1990, Koenig and Dickinson 2004).
Broadly defined, cooperation occurs when more
than two individuals contribute to a single brood
(Cockburn 2004, 2006). A range of social
structures are considered to be cooperative (e.g.,
Stacy and Koenig 1990), and they occur in taxa as
diverse as ants (Order Hymenoptera; Reeve and
Ratnieks 1993) and hawks (e.g., Biiteo galapa-
goensis\ Faaborg and Bednarz 1990). Cooperative
breeding among birds can include multiple
parents that lay eggs in a single nest {Crotopliaga
sulcirostris', Vehrencamp 1978), large breeding
collectives that include multiple breeding pairs
and numerous non-breeding individuals (Mela-
nerpes forniicivorous; Koenig et al. 1995), or
colony-nesting species with helpers assisting
related clans (Merops hiillockoides\ Emlen and
Wrege 1989). Cooperation appears to be most
commonly exhibited as groups comprised of
parents and offspring that work together to rear
related young (e.g., Picoides borealis', Walters
1990). This behavior is rare, as ~2.5% of birds
are considered to be cooperative breeders (Brown
1987).

Several life history characteristics are com-
monly as.sociated with cooperative breeding (Du
Plessis et al. 1995). Some cooperatively breeding
birds are year-round territorial species with young
that have extended parental dependence (Brown

' Dcpartmcnl of Fi.shcries and Wildlife Sciences, 302
Anheuser-Itusch Natural Resources Building, University of
Missouri. Columbia, Mf) 0.321 1, USA.

^Societc d'Ornithologic dc Polyncsic “Manu”. B.F^.
8699, 98719 Taravao, Tahiti. Polyncsie Franyaise.
’Corresponding author; e-mail: kcslerd@missouri.edu

1987, Stacey and Koenig 1990). Cooperation can
also occur in situations with limitations in
breeding resources or territories, in which poten-
tial dispersers are considered to be ecologically
constrained from breeding independently (Selan-
der 1971, Brown 1974, Gaston 1978, Stacey 1979,
Emlen 1982). This concept was first presented as
habitat saturation (Brown 1974) and later artic-
ulated as the ecological constraints hypothesis
(Emlen 1982). A conceptual model of the
hypothesis was described as a situation in which
all available habitats already host territories that
are maximally compressed in size (Brown 1987).
Alternatively, some suggested that cooperation
must have originated on the natal teiritory, and
that potential dispersers delayed dispersing be-
cause the benefits of philopatry outweighed those
of immediate dispersal (Stacey and Ligon 1987,
1991).

We present the first observations of cooperative
breeding by the Society Kingfisher {Todiramphus
veneratus). The Society Kingfisher is a member of
a genus that includes 1 1 imperiled species (near
threatened or above, lUCN 2008) of the 21
recognized (Clements 2007). We present three
lines of evidence for cooperation in the Society
Kingfisher. First, we describe direct observations
of helping behaviors at nests. Second, we report
observations of cooperative groups of birds
throughout the study area and during the breeding
sea.son. Finally, we provide population density
estimates and compare those to densities reported
for other Pacific Todiramphus species to illustrate
the potential for habitat saturation.

METHODS

We observed Society Kingfishers on the island
ol Moorea(17° 32' S, 149° 50' W) during surveys

46

Kesler cl al. • COOPERA TIVE BREEDINC} OE SOCIETY KINGEISIIERS

47

FIG. 1. Copulation with three Society Kingfishers present on 31 December 2008 on the island of Moorea, French
Polynesia. (Photograph by T. Ghestemme.)

and visits in November and December 2008, and
January 2009. We described cooperative behaviors
observed on two territories during nest excavation,
incubation, and brood rearing phases (17° 32'

1 1" S, 149° 49' 46" W; and 17° 32' 26" S, 149°
47' 20' W). Each observation period la.sted —30
min, and dates and times associated with each
behavior are reported.

We conducted variable distance point-transect
surveys (Buckland et al. 1993) of Moorea’s
landbirds at 46 stations in undeveloped upland
forests. Stations were separated by 200 m. The
distance was estimated to all landbirds detected by
visual or auditory means during a lO-min
observation period at each station. We also
recorded the number of kingfishers present when
detected outside of prescribed observation peri-
ods. Point-transect data were analyzed with
Program DISTANCE (Thomas et al. 2006) to
estimate the density of Society Kingfishers on the
island of Moorea. We pooled observations into
25-m groups and conducted the analysis with a
half-normal key, truncation at 100 m, and default
model selection adjustment routine (Akaike s
Information Criteria; AlC,.; Burnham and Ander-
son 2002). Limited detections reduced potential
adjustment parameters to two, but the results were

intuitively appealing and Chi-square goodness of
fit tests indicated the approach was reasonable (yj
= 3.66, df = 2, P = 0.16).

RESULTS

On 25 (1510-1612 hrs), 26 ( 1045-1 1 25 hrs),
and 27 (0720-0815 hrs) November 2008 we
observed three Society Kingfishers cooperatively
excavating nest cavities in the central and
southeastern regions of Moorea. Three kingfishers
were perched in close proximity (<2 m) on each
occasion. All group members flew to the trees
(Neonaucleci fasten) of interest and struck the
main stem with their beaks. Nest excavation was
accompanied by loud calling, and quieter growl-
ing and warbling vocalizations typical of Pacific
kingfishers. Two individuals consistently perched
closer together (within 50 cm) and struck the
cavity more often than the third bird, which only
struck the tree once during each observation
period. Two birds were observed excavating a
cavity without a third individual present on two
other teiTitories. On 31 December (0925 hrs). two
birds copulated (Pig. 1) while the third bird
passively perched within 10 cm. No aggressive
behaviors were observed during nest excavations
or copulations. Two of three group members were

48

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 1. March 2010

marked with individual-specific combinations of
colored plastic leg bands on 13 January 2009 on
the central region territory. On 13 (1040-
1125 hrs), 14 ( 1500-1630 hrs), and 15 (1600-
1715 hrs) January, at least three birds (2 color
banded and 1 unbanded) were observed entering
the nest cavity and incubating. The same 2 color-
banded and 1 unbanded group members were also
observed delivering food items to the nest on 19
February (0630-0830 hrs).

Observations of three-bird groups of Society
Kingfishers throughout the island of Moorea
provided further support for cooperation. King-
fishers were detected at 61% of the point-transect
stations during island-wide surveys. Of the birds
visually observed, we encountered seven three-
bird groups and seven two-bird groups. Three-bird
groups were observed on an additional four
territories outside of survey periods. The dis-
tance-based analyses indicated densities of 2.4
kingfishers/ha (95% Cl = 1.6-3. 5). These results
suggest Society Kingfisher territories on Moorea
were 0.82 ha (95% Cl = 0.57-1.19) in size for
pairs or 1.23 ha (95% Cl = 0.85-1.90) for trios,
based on multiplying density estimates by group
size. No groups were observed with more than
three individuals.

DISCUSSION

Our observations of cooperative nest excava-
tions and food delivery by Society Kingfishers are
congment with cooperative behaviors reported for
other birds, and we suggest the species qualifies as
a cooperative breeder. Cooperative breeding has
been described in other Tocliramphus kingfishers of
Pacific Oceania. Beckon (1987) reported three-bird
nest excavation by Collared Kingfishers (T.
chloris), and substantial work was directed at
cooperative behaviors of Micronesian Kingfishers
(T. cinnamominus) (Kesler and Haig 2005, 2007a).
This behavior is relatively common among other
Coraciiformes. Cooperative breeding has been
documented for Rufous Hornbills (Buceros hydro-
cora.x) (Witmer 1993), White-fronted Bee-eaters
{Merops hullockoides) (Emien and Wrege 1989),
Laughing Kookaburras (Dacelo novaeguineae)
(Legge and Cockburn 2000), and Pied Kingfishers
(Ceryle rudis) (Reyer 1980) among others.

Our distance-based results indicated a Society
Kingfisher population density that was higher
than for other Pacific kingfisher species in
similarly structured habitats. Mean territory size
was 5.9 ha for cooperatively breeding Microne-

sian Kingfishers (Kesler and Haig 2007b, c) and
6-10 ha for Tuamotu Kingfishers {Todirampinis
gamhieri', Coulombe and Kesler, in prep). King-
fishers exclude neighboring conspecifics from
their territories, and territory sizes translate to
population densities of 0.5 Micronesian Kingfish-
ers/ha (assuming territories with 3 birds), and 0.2
Tuamotu Kingfishers/ha (assuming a breeding
pair). Our results indicate population densities of
Society Kingfishers on Moorea that are 4.4 times
higher than cooperatively breeding Micronesian
Kingfishers and 6.7 times higher than breeding
pairs of Tuamotu Kingfishers. This high density
of Society Kingfishers on Moorea may over-
whelm breeding resources, saturate habitats with
territories, and lead to the ecological constraints
previously suggested to cause cooperation in other
species (Emien 1982, Walters et al. 1992).

Cooperative breeders provide opportunities to
simultaneously test fundamental theories in be-
havioral ecology and conservation biology be-
cause both management and cooperative breeding
theory often focus on limited resources (Walters
et al. 1992, Komdeur 1994). For example, much
has been learned about conservation and evolution
of cooperative sociality in the endangered Red-
cockaded Woodpecker {Picoides borealis) (Con-
nor et al. 2001 ). Conservation managers in French
Polynesia should consider the importance of
identifying and managing limited kingfisher
resources to maintain robust populations. Coop-
erative breeding also has the potential to influence
population resilience by creating a reserve set of
adult helpers that are poised to breed when
opportunities are presented, which can buffer the
effects of environmental or demographic stochas-
ticity (Walters et al. 2002, Kesler and Haig
2()07a).

Society Kingfishers were abundant in the
undisturbed upland forests of Moorea and Tahiti
(Gouni and Zysman 2007), and it is somewhat
striking that cooperative behaviors were not
previously described. A similar lack of knowledge
persists for many Pacific birds despite the
immense contributions that island species have
made to biological understanding. We advocate
for added focus on the basic natural history of
birds in remote and threatened regions, including
the islands of Pacific Oceania.

ACKNOWLEDGMENTS

Wc are indebted to all who provided financial, material,
and technical .support to the project, including members of

Kc.sler ct cil. • COOPERA I'lVE BREEDING OF SOCIITY KINGI ISIIERS

49

Societe cl’Oniithologie de Polynesie '■Manu'', the govern-
ment of French Polynesia, and ESRl Conservation Grants
Program. We extend special thanks to Stephan Vanderstokt,
Jeannine Gouni, Jean-Marie Gouni, and Christophe Noiret
for their important logistical contributions. Additionally, we
thank Jeff Walters, Allison Cox, and the anonymous
reviewers who helped improve the quality of the manu-
script.

LITERATURE CITED

Beckon, W. N. 1987. Evidence of cooperative nest
excavation by the White-collared Kingfisher in Eiji.
Ibis 129;39 1-392.

Brow'n, J. L. 1974. Alternate routes to sociality in jays-
with a theory for the evolution of altruism and
communal breeding. American Zoologist 14:63-80.
Brown, J. L. 1987. Helping and communal breeding in
birds: ecology and evolution. Princeton University
Press, Princeton, New Jersey, USA.

Buckland, S., D. Anderson, K. Burnham, and J. Laake.
1993. Empirical models for the spatial distribution of
wildlife. Journal of Applied Ecology 30:478-495.
Burnham, K. P. and D. R. Anderson. 2002. Model
selection and multimodel inference: a practical
information-theoretic approach. Springer-Verlag,
New York, USA.

Clements, J. F. 2007. The Clements checklist of birds of
the world. Sixth Edition. Comstock Publishing Asso-
ciates, Ithaca, New York, USA.

CoCKBURN, A. 2004. Mating systems and sexual conflict.
Pages 81-101 in Ecology and evolution of cooperative
breeding in birds (W. D. Koenig and J. L. Dickinson,
Editors). Cambridge University Press, New York,
USA.

CocKBURN, A. 2006. Prevalence of different models of
parental care in birds. Proceedings of the Royal
Society of London, Series B 273:1375-1383.

Conner, R. N., D. C. Rudoi.ph, and J. R. Walters. 2001.
The Red-cockaded Woodpecker: surviving in a fire-
maintained ecosystem. University of Texas Press,
Austin. USA.

COULOMBE, G. AND D. C. Kesler. In prep. Multiscale
resource selection in critically endangered Tuamotu
Kingfishers. Biological Conservation: In press.

Du Plessis, M. a., W. R. Siegfried, and A. J.
Armstrong. 1995. Ecological and life-history corre-
lates of cooperative breeding in South African birds.
Oecologia 102:180-188.

Emlen, S. T. 1982. The evolution of helping. I. An
ecological constraints model. American Naturalist
1 19:29-39.

Emlen. S. T. and P. Wrege. 1989. A test of alternate
hypotheses for helping behavior in White-fronted Bee-
eaters of Kenya. Behavioral Ecology and Sociohiology
25:303-319.

Faaborg, j. and j. C. Bednarz. 1990. Galapagos and
Harris’ hawks: divergent causes of sociality in two
raptors. Pages 357—384 in Cooperative breeding in
birds: long-term studies of ecology and behavior (P. B.

Stacey and W. D. Koenig, Editors). Cambridge
University Press, Cambridge, United Kingdom.

Gaston, A. J. 1978. The evolution of group territorial
behavior and cooperative breeding. American Natu-
ralist 1 12:1091-1 100.

Gouni, A. and T. Zysman. 2007. Oiseaux du Fenua, Tahiti
et ses Ties. Thethys Editions, Paris, France.

International Union for the Conservation of Nature
(lUCN). 2008. Red list of threatened species. Interna-
tional Union for Conservation and Natural Resources,
Gland, Switzerland, www.iucnredlist.org (accessed I
January 2009).

Kesler, D. C. and S. M. Haig. 2005. Selection of arboreal
termitaria for nesting by cooperatively breeding
Pohnpei Micronesian Kingfishers. Ibis 147:188-
196.

Kesler, D. C. and S. M. Haig. 2007a. Conservation
biology for suites of species: population demography
for the Todiramphus kingfishers. Biological Conser-
vation 136:520-530.

Kesler, D. C. and S. M. Haig. 2007b. Multi-scale resource
use and selection in cooperatively breeding Microne-
sian Kingfishers. Journal of Wildlife Management
71:765-772.

Kesler, D. C. and S. M. Haig. 2007c. Territoriality,
prospecting, and dispersal in cooperatively breeding
Micronesian Kingfishers {Todiramphus cinnamominus
reichenhachii). Auk 124:381-395.

Koenig, W. D. and J. L. Dickinson. 2004. Ecology and
evolution of cooperative breeding in birds. Cambridge
University Press, New York, USA.

Koenig, W. D., P. B. Stacey, M. T. Stanback, and R. L.
Mumme. 1995. Acorn Woodpecker {Melanerpes for-
micivonis). The birds of North America. Number 194.

Komdeur, j. 1994. Experimental evidence for helping and
hindering by previous offspring in the cooperative-
breeding Seychelles Warbler (Acrocephaius sechcUen-
sis). Behavioral Ecology and Sociobiology 34:175-
186.

Legge, S. and a. CocKBURN. 2000. Social and mating
system of cooperatively breeding Laughing Kooka-
burras (Dacelo novaeguincae). Behavioral Ecology
and Sociobiology 47:220-229.

Reeve, H. K. and F. L. W. Ratnieks. 1993. Queen-queen
conflict in polygynous societies: mutual tolerance and
reproductive skew. Pages 45-85 in Queen number and
sociality in insects (L. Keller. Editor). Oxford
University Press, Oxford. United Kingdom.

Reyer, H. 1980. Flexible helper structure as an ecological
adaptation in the Pied Kingfisher (Ceryte rudis).
Behavioral Ecology and Sociobiology 6:219-227.

Selander, R. K. 1971. Systcmatics and spcciation in birds.
Pages 57-147 in Avian biology (D. S. Earner and J. R.
King, Editors). Volume 1. Academic Press. New York.
USA.

Sklitch, a. F. 1935. Helpers at the nest. Auk 52:257-273.

Stacey, P. B. 1979. Habitat .saturation and communal
breeding in the Acorn Woodpecker. Animal Behaviour
27:1 153-1 166.

Stacey, P. B. and W. D. Koenig. 1990. Cooperative
breeding in birds: long-term studies of ecology and

50

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. I. March 2010

behavior. Cambridge University Press, New York.
USA.

Stacey, P. B. and J. D. Ligon. 1987. Territory quality and
dispersal options in the Acorn Woodpecker, and a
challenge to the habitat-saturation model of coopera-
tive breeding. American Naturalist 130:654-676.
Stacey, P. B. and J. D. Ligon. 1991. The benefits-of-
philopatry hypothesis for the evolution of cooperative
breeding: variation in territory quality and group size
effects. American Naturalist 137: 831-846.

Thomas, L., J. L. Laake, S. Strindberg, F. F. C.
Marques, S. T. Buckland, D. L. Borchers, D. R.
Anderson, K. P. Burnham, S. L. Medley, J. H.
Pollard, J. R. B. Bishop, and T. A. Marques. 2006.
DISTANCE 5, Release 2. Research Unit for Wildlife
Population Assessment, University of St. Andrews,
United Kingdom.

Vehrencamp, S. 1978. The adaptive significance of
communal nesting in Groove-billed Anis (Crolopluiga

crotophaga sulcirostris). Behavioral Ecology and
Sociobiology 4:1-33.

Walters, J. R. 1990. Red-cockaded Woodpeckers: a
‘primitive’ cooperative breeder. Pages 67-102 in
Cooperative breeding in birds: long-term studies of
ecology and behavior (P. B. Stacey and W. D. Koenig,
Editors). Cambridge University Press, New York,
USA.

Walters, J. R., C. K. Copeyon, and J. H. Carter 111.
1992. Test of the ecological basis of cooperative
breeding in Red-cockaded Woodpeckers. Auk 109:90-
97.

Walters. J. R., L. Crowder, and J. Priddy. 2002.
Population viability analysis for Red-cockaded Wood-
peckers using an individual-based model. Ecological
Applications 12:249-260.

WiTMER, M. C. 1993. Cooperative breeding by Rufous
Hornbills on Mindanao Island, Philippines. Auk
1 10:933-936.

77r' Wilson Jotinuil ql Ornifliology 122(1):5I 59,2010

REPRODUCTIVE SUCCESS OE BURROWING OWLS IN URBAN AND
GRASSLAND HABITATS IN SOUTHERN NEW MEXICO

DANIELE BERARDELLI,' MARTHA J. DESMOND,'”' AND LEIGH MURRAY^

ABSTRACT. — We examined Burrowing Owl (Athene ciinieiiaria) reproductive success at 144 nests in urban in — 80)
and grassland (n = 64) habitats in southern New Mexico in 2000 and 2001 . Nest success was higher in grassland compared
to urban areas (81 vs. 68%) but urban landscapes fledged more young per successful nest (3.85 vs. 3.07). Mean fledging
success per nest was similar between habitats with 2.60 and 2.50 fledglings in urban and grassland habitats, respectively.
Fledging success was categorized as failure, low, and high. Reproductive success in both habitats was associated with
measures of owl density. High success in urban landscapes was associated with fewer surrounding nests, an index of larger
nesting territories, and open space. Failure was associated with solitary nests. This suggests owls benefit from the presence
of other nesting pairs as long as the density is not too high. High success in grassland habitats was associated with fewer
surrounding nests, index of larger nesting territories, edge nests, and lower fledgling success of the nearest nest. Higher nest
success, but fewer fledglings per successful nest, suggests competition for resources in this habitat. Received 23 February
2009. Accepted 30 September 2009.

Few native species are as widely distributed
and appear as adaptable to environmental change
as Bun-owing Owls {Athene cimiciilaria). This
species ranges throughout much of North, Central,
and South America, and the West Indies (Haug et
al. 1993, Desmond et al. 2001). Most research,
however, has focused on the western subspecies
(A. c. Iiypiigaea), which is distributed across
western North America. The western Burrowing
Owl is associated with fossorial mammals and
historically occurred across grasslands, desert
scrub, and sagebrush (Artemisia spp.) habitats
(Haug et al. 1993, DeSante et al. 2004). More
recently, they have occupied urban and agricul-
tural landscapes, often occurring at densities
substantially higher than observed in native
habitats (Millsap and Bear 2000, DeSante et al.
2004, Rosenberg and Haley 2004).

Populations of the western BuiTowing Owl are
declining in many parts of their range despite the
wide distribution and flexibility of the species to
occupy a variety of burrow systems and habitat
types (Sheffield 1997, Holroyd et al. 2001). Long-
term Breeding Bird Survey data (1966-2007)
indicates an annual decline in Burrowing Owl
populations of 4.4% in the northern and central
plains states compared to an annual increase of
1.5% in the southern plains states (Sauer et al.
2007). Causes of declines are many but are

' Department of Fish. Wildlife and Conservation Ecolo
gy, P. O. Box 30003, MSC 4901, New Mexico State
University, Las Cruces. NM 88003, USA.

^Department of Statistics, Kansas State University,
Manhattan. KS 66506. USA.

^Corresponding author; e-mail: mdesmond(®nmsu.edu

thought to be mainly attributed to habitat loss
and degradation. Land conversion, including loss
of grasslands and populations of prairie dogs
(Cynomys spp.), likely is involved. Land conver-
sion may also be contributing to an apparent shift
in owl populations to urban and agricultural
landscapes, particularly in the southern part of
their range (DeSante et al. 2004, Rosenberg and
Haley 2004, Conway et al. 2006; but see Poulin et
al. 2005). The Imperial Valley of California, prior
to agriculture, had low densities of Burrowing
Owls, but now supports among the densest known
populations of this species (DeSante et al. 2004).

Studies in urban environments in Florida have
linked Burrowing Owl density and reproductive
success with low and moderate levels of devel-
opment (Millsap and Bear 2000). Once develop-
ment exceeded 60% of an area, Millsap and Bear
(2000) found owl populations and reproductive
success declined, likely due to increased human
disturbance. Benefits of sodded, landscaped lawns
and parks in urban environments include in-
creased availability of insects and lizards, due to
regular watering and lawn maintenance (Wese-
mann and Rowe 1987), and birds (Mrykalo et al.
2009). Negative effects include increased preda-
tor densities, disturbance, and road fatalities
(We.semann and Rowe 1987, Millsap and Bear
2000).

A comparison of reproductive biology between
native and urban habitats will help identify
management needs for owls in both habitats,
but particularly urban environments. Bun-owing
Owl populations in native habitats across the
Great Plains are most commonly linked to black-
tailed prairie dog (Cynomys liidovicianus) colo-

51

52

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122. No. I. March 2010

nies (Haug et al. 1993, Desmond et al. 2000, Orth
and Kennedy 2001, Griebel and Savidge 2007).
Reproductive success in the Great Plains was
positively associated with density of fossorial
rodents (black-tailed prairie dogs; Desmond et al.
2000, Lantz et al. 2006) and was both positively
and negatively associated with measures of owl
density (Desmond et al. 2000, Griebel and
Savidge 2007). This suggests that Buirowing
Owls may benefit from prairie dog presence (or
other fossorial rodents) through the dilution
effect or by cueing in on alarm calls and the
presence of other owls by alerting each other to
predators.

Increased Burrowing Owl populations in urban
landscapes underscore the importance of identi-
fying threats and conservation needs for urban
populations. A number of studies have examined
Burrowing Owls in different systems including
comparative studies in urban and agricultural
systems (Conway et al. 2006), but there is a lack
of comparative studies on owl reproduction in
urban and native environments. Bun'owing Owls
nesting in urban compared to rural environments
in west Texas exhibited greater vigilance (Chip-
man 2006). Serious concern has been raised
about the sustainability of owl populations in
urban and agricultural environments due to
conflicts with human activities, especially chang-
es in land management (DeSante et al. 2004,
Rosenberg and Haley 2004). Major threats in Las
Cruces, New Mexico are construction and land
management activities, maintenance of irrigation
canals and fields, use of recreational fields with
large audiences on weekends and evenings, nests
in areas of high pedestrian traffic, and vehicle
mortalities. We hypothesized that nest site
characteristics and reproduction would differ
between urban and grassland sites due to
different predation pressure, disturbance, and
re.source availability in these two environments.
Prairie dog colonies are larger in area, contain
higher burrow densities and, although dynamic,
are more stable compared to rock squirrel
{Spermophiliis vurie^atus) colonies. We predict-
ed Hedging success would be higher in grass-
lands due to decreased levels of human distur-
bance, lower rates of predation, and benefits
from higher densities and larger areas of colonial
sciurids. Owls may benefit from the pre.sence of
prairie dogs through the dilution effect and by
cueing in on prairie dog alarm calls (Desmond et
al. 2()()()).

METHODS

Study Area. — ^Research was conducted in urban
(// = 19) and grassland (/? = 7) habitats of the
northern Chihuahuan Desert in southern New
Mexico, USA in two adjacent counties. Dona Ana
and SieiTa. Urban nests were monitored within a
4,533-ha area of Las Cruces (human population
74,267; U.S. Census Bureau 2000). Nests in
grassland habitats were on the 145,750-ha
Armendaris Ranch within the Pedro Armendariz
Land Grant, 122 km north of Las Cruces. Average
annual rainfall for southern New Mexico is
24.7 cm with greater than 56% in June-Septem-
ber (Western Regional Climate Center Desert
Research Institute 2002). Annual rainfall in 2000
was 1 .6 cm above average in Las Cruces ( 1 0. 1 cm
above average for the May-Jul nesting period)
and 5.6 cm below average on the Armendaris
Ranch (average during the May-Jul nesting
period). The following year was drier with annual
precipitation 10.2 and 9.9 cm below average for
urban and grassland sites, respectively, and 1.2
and 3.3 cm below average for the May-July
nesting period.

Dominant vegetation in the Las Cruces area
consisted of honey mesquite (Prosopis glandu-
losa), desert globemallow (Sphaercdcea cocciuea),
snakeweed (Giitierreziu spp.), manicured lawns,
ornamental trees and shrubs, and flood-iirigated
agricultural row crops. Average elevation of the
study area was 1,358 m. BuiTowing Owl nests in
Las Cruces were mainly associated with rock
squirrel burrows; owls were also nesting in
cavities excavated by pocket gophers {Geomys
spp.) as well as human-made (drain pipes) and
artificial burrows. The city of Las Cruces
Community Development Department made the
following distinctions concerning land use: de-
veloped land consisted of residential, office,
commercial, industrial, and community; open
space included agricultural fields, parks, aiToyos,
and vacant lots. Nesting owls occurred in all of
these areas.

Dominant vegetation on and surrounding prai-
rie dog colonies at the Armendaris Ranch
included alkali sacaton (Sporohuliis airoides),
tobosa {Pleuraphis niutica), burrograss (Scleropo-
I’on hrcvifoUus), three-awn (Aristida spp.), black
grama (Bouteloua eriopoda), honey mesquite,
creosote (Larrea tridentata), cholla (Opuntia
imbricate), and yucca (Yucca spp.). Elevation at
the seven colonies on the Armendaris Ranch was

Berarciclli et al. • BURROWING OWL REPRODUCTIVE SUCCESS

53

1,386 m. Cattle were removed from the Armen-
daris Ranch in 1993 and replaced with 1,200 head
of bison (Bison bison) (Hartsough 2002). Bnrrow-
ing Owl nests on the Armendaris Ranch were in
black-tailed prairie dog colonies averaging 4.5 ha
(range 1.7-10.4 ha) re-established within 6 years
prior to the initiation of this study. Historically,
black-tailed prairie dogs were likely important for
Burrowing Owls in southern New Mexico; prairie
dogs were extiipated from most of the region and
are now reduced to a few areas (Oakes 2000).

Data Collection. — Searches for nesting owls
occurred once each week beginning 1 March 2000
and 1 February 2001. Urban areas were searched
by visiting known nesting locations obtained from
local contacts, searching historic locations, and
visiting known rock squiirel colonies. All suitable
habitat and buiTOWs within each location were
searched for nesting owls. Re-established prairie
dog colonies on the Armendaris Ranch were
searched for owls by walking transects through
colonies and examining each buiTOW (Desmond et
al. 2000). Nests were identified based on presence
of a nesting pair on ^ 1 occasion and the presence
of shredded grass and scat at the buiTow entrance
(Garcia and Conway 2009). Indications of nest
initiation in urban areas also included shredded
paper, plastic, and aluminum foil. The suiTound-
ing area was thoroughly and repeatedly searched
for other nesting pairs when a nest was located.

Nests were monitored weekly (0500-1200 and
1630-2130 hrs) until nesting attempts were ter-
minated or young fledged. Time spent at each
location, 15-60 min, depended on our ability to
obtain reliable above-ground brood counts. Each
brood was counted twice a week for 60 min at
dawn (0500-0800 hrs) and dusk (1830-2100 hrs)
to identify the size of the initial brood. A reliable
count was considered to be verification of each
brood member. The site was revisited at dawn or
just prior to dusk within the week to recount the
number of young, if all Juvenile owls were not
accounted for within 60 min. This method is more
intensive than the standardized 20 min counts
recommended by Gorman et al. (2003), and
provides a reliable estimate of the number of
young. Nest success data were limited to nests
initiated before 30 May to avoid including re-
nesting attempts in the data set. Fledging success
was defined as the number of young per nest
attempt surviving until 42 days of age, the age
when juveniles are capable of extended flight
(Haug and Oliphant 1990, Desmond et al. 2000).

A successful nest was defined as one that fledged
a minimum of one chick. The rate ol nest success
for each year was delined as the number ol
successful nests divided by the total number of
attempts within each of the two study areas.

Characteristics quantified around nest sites
included number of breeding pairs within a 75-
m radius of each nest, distance to the nearest-
neighboring nest, fledging success of the nearest-
neighboring nest, burrow type, number of satellite
burrows within a 50-m radius of each nest,
distance to the nearest perch used by adult males
and, for urban nests, the category of urbanization
(developed vs. open space). We chose to count
nests within a 75-m radius of the nest buiTOW due
to observations that chicks will venture up to 75-
m from the nest burrow before fledging (Desmond
and Savidge 1999, Desmond et al. 2000). We also
recorded for the 2001 field season whether a nest
burrow was new or had been used as an owl nest
the previous season.

Statistical Analyses. — A paired-samples Utest
was used to compare the mean number of nests
per site between years for each habitat. Distance
to the nearest-neighboring nest, number of nests
within a 75-m radius, number of satellite burrows
within a 50-m radius, number of nests per ha, and
perch distance were also compared between
habitat types, pooling years, and within habitat
types between years, using Student’s r-test for two
independent samples. The number of fledglings
per nest attempt was not normally distributed and
could not be coiTected using log or square-root
transformations. Thus, nests were separated into
three categories of fledging success and a Chi-
square test for homogeneity was used to test for
differences between habitat types in fledging
success categories: failure (0 fledglings), low
success (1-2 fledglings), and high success (>3
fledglings). These categories were used to de-
crease variability while maintaining the structure
of the relationship between the outcome and the
predictor variables (Hosmer and Lemeshow
2000).

We used multinomial logistic regression to
evaluate variables as predictors of Buirowing Owl
fledging success categories for each habitat type.
Data within habitat types (grassland vs. urban)
were combined for the 2 years, and regression
analysis was conducted separately for each habitat
type. We developed a set of candidate a priori
models for each habitat type that we believed
would be the most biologically meaningful, 16 for

54

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 1, March 2010

TABLE I. Habitat variables (mean ± SE) and t statistics for BuiTowing Owl nest sites in urban (Las Cruces) and
grassland (Armendaris Ranch) habitats pooled for 2000 and 2001 in southern New Mexico.

Sites

Nest site variables

Urban vs. grassland

Grassland by year

Urban

Grassland

041

P

Nearest-neighbor distance, m

159.1 ± 11.7

49.7 ± 6.3

7.66

<0.001

# Nests within 75 m

0.8 ± 0.1

3.6 ± 0.3

-8.99

<0.001

Nest density/ha

0.13 ± 0.01

1.58 ± 0.10

-14.12

<0.001

# Satellite burrows

10.7 ± 1.3

44.3 ± 2.4

-13.14

<0.001

Perch distance, m

184.0 ± 56.3

27.9 ± 3.4

2.46

0.015

2000

2001

^61

p

Nearest-neighbor distance, m

45.0 ± 7.6

52.2 ± 9.1

-0.54

0.589

# Nests within 75 m

2.8 ± 0.5

4.0 ± 0.4

1.81

0.093

Nest density/ha

1.6 ± 0.1

1.9 ± 0.1

-3.78

<0.001

# Satellite burrows

38.2 ± 4.4

47.6 ± 2.6

-1.93

0.058

Perch distance, m

31.4 ± 7.6

26.0 ± 3.2

0.76

0.451

gra.ssland and 18 for urban habitats. We used a
balanced combination of the same five variables
for each habitat type (year, number of nests within
a 75-m radius of each nest, fledging success of the
nearest neighboring nest, number of satellite
burrows within a 50-m radius of each nest, and
distance to nearest perch) but added one addition-
al variable, category of urbanization, to the urban
data set. The city of Las Cruces makes multiple
distinctions concerning land use, but to increase
the number of nests within a land-use group, we
limited land use to two categories: open space
(including agricultural fields, parks, arroyos, and
vacant lots) and developed land (including
residential, commercial, and community lands).
We calculated Akaike’s Information Criterion
(AIC) corrected for small sample size (AIC^.) for
each model using SYSTAT 12.0 (Systat 2007).
All models were ranked according to AAIC^. and
models with a AAIC^. 2 were considered
competing models (Burnham and Andersen
2002). Model weights were calculated for each
model and cumulative weights for all variables
were calculated. All variables were evaluated for
pairwise collinearity using Pearson Correlation.
None of the variables included in either sets of
models was considered collinear (variables corre-
lated < 0.70).

A Chi-.square homogeneity test was u.sed to
evaluate whether a burrow was more likely to be
re-used if the nest had been successful the
previous year. A Mann Whitney U test was u.sed
to compare fledgling success between new nest
burrows and burrows that were re-used as nests
from the previous year. All statistical analyses

were performed using SYSTAT, Version 12.0
(Systat 2007).

RESULTS

Sixty-six nests were located and monitored in
2000, 43 within 19 urban sites and 23 within six
prairie dog colonies. Seventy-eight nests were
located in 2001, 37 within 15 urban sites and 41
within seven prairie dog colonies, including one
newly established colony. There was no differ-
ence in the mean number of nests per site in urban
areas between years {P = 0.470), but the number
of nests established within the original six prairie
dog colonies in grassland habitats was greater in
2001 (Hi = —3.72, P = 0.003). In urban areas,
15% of nests were associated with rock squiiTel
systems. All monitored nests in grassland areas on
the Armendaris Ranch were associated with
black-tailed prairie dog colonies.

Site characteristics and owl population dynam-
ics differed between the two habitats (Table 1).
Mean nearest-neighbor nest distances in grass-
lands were significantly closer, and the mean
number of nests within 75 m and nest densities
were both significantly greater compared to urban
sites (Table I). The mean number of satellite
burrows within a 50 m-radius of the nest was
significantly higher, whereas mean perch distance
of the adult male from the nest burrow was
significantly less in grassland compared to urban
habitats (Table 1 ).

Grassland site characteristics exhibited differ-
ences between years. The number of nests in
prairie dog colonies almost doubled the second
year and, predictably, nest density was greater in

Bcrardelli ct al. • BURROWING OWL REPRODUCTIVE SUCCESS

55

TABLE 2. Burrowing Owl nesi success in urban and
grassland habitats in southern New Mexico pooled for 2000
and 2001. Percent reproductive effort within three Hedging
success categories (failure, low success, high success) was
examined using a Chi-square test for homogeneity (X- =
5.59, df = 2, P = 0.061). Fledging data were not
normally distributed.

Productivity piirameters

Urban

Grassland

% Failure'’ (/;)

.32.5 (26)

18.8 (12)

% Low succes.s*’ (/;)

16.5 (13)

28.0 (18)

% High success'-' {n)

51.0 (41)

53.0 (34)

Totals

100 (80)

100 (64)

Successful nests'*

{% ± SE)

54 (67.5 ± 5.6)

52 (81.3 ± 4.8)

Young fledged

208

160

Fledgling.s/attempt

(X ± SE)

2.6 ± 0.25

2.5 ± 0.22

Fledglings/successful

nest (X ± SE)

3.9 ± 0.22

3.1 ± 0.20

■* Failure = 0 fledglings.

Low success = 1-2 fledglings.

High success = 3+ fledglings.

Successful nests = si chick.

2001 (Table 1). However, no difference was
detected in the mean number of nests within a
75-m radius of a nest, mean nearest-neighbor
distance, or mean perch distance from the nest
between years (Table 1). The mean number of
satellite burrows within a 50-m radius of a nest in
2001 was higher (Table 1). No differences {P >
0.05) in site characteristics were ob.served be-
tween years in urban landscapes.

No differences were detected among the three
reproductive success categories between habitat
types {X- = 5.59, df = 2, P = 0.061); however,
there was a tendency for higher nest failure in

urban (32.5%) compared to grassland (18.8%)
habitats, but more nests in the low-success
category for grassland compared to urban sites
(28 vs. 16.5%; Table 2). Sixty-eight percent of
Burrowing Owl pairs ne.sting in urban areas
compared to 81% in grassland habitats success-
fully fledged at least one chick (Table 2). The
mean number of young fledged per successful
nest was 3.85 and 3.07 for urban and grassland
sites, respectively. On average, nests in urban and
grassland areas fledged 2.60 and 2.50 young per
nest attempt, respectively (Table 2).

Differences were detected in variables influ-
encing fledging success between the two habitats.
The global model containing all five variables in
grassland habitats had overwhelming support
(Table 3). This was the only competing model
for grassland sites with a model weight of 0.99.
Higher success was associated with nests having
fewer satellite burrows within a 50-m radius (40
vs. 57), greater perch distance from the nest (34
vs. 19 m), and lower fledging success of the
nearest Burrowing Owl nest (2.40 vs. 3.60
fledglings), as compared with failure. Owls with
low success had a greater number of owl nests
within a 75-m radius (5.60 compared to 2.70)
when compared with high success owls. A year
effect was also observed, as significantly more
nests were in the low success category the second
year (Table 3). Cumulative model weights sup-
ported the importance of all five variables
(Table 4).

The two top models in urban habitats were
competing models; variables included in the top
model were number of nests within a 75-m radius,
perch distance, and number of satellite buiTOWs

TABLE 3. Top three multinomial logistic regression models for each habitat type ranked in order of increasing AAIC,
influencing Burrowing Owl fledging success in urban and grassland habitats in southern New Mexico in 2000 and 2001. -i-
(or — ) on a predictor variable indicates increased (or decreased) probability ol high-success vs. lailure or high success vs.
low success as the predictor variable increases for the two multinomial submodels.

Habitat

Models

High v.s. failure

High vs. low succe.ss

K

Ate,

AAIC,

U’,

Grassland

1.

NEST, PERCH,

NNFS, SAT, YR

-fNEST, -perch,

-t-NNES, +SAT, -i-YR

-hNEST, -PERCH,

-NNFS, -SAT, -YR

5

98.94

0

0.99

2.

NEST, PERCH, SAT

+NEST, -PERCH, +SAT

+NE,ST, -PERCH, -SAT

3

107.86

8.92

0.01

3.

NEST, NNES, SAT

-fNEST, -fSAT, -fNNFS

-fNEST, -sat, -NNFS

3

112.26

13.32

0

Urban

1.

NEST, PERCH, SAT

-NEST, -PERCH, -t-SAT

-t-NEST, -PERCH, -hSAT

3

157.03

0

0.44

2.

NEST, SAT, URB

-NEST, +SAT, -t-URB

-t-NEST, -SAT, +URB

3

158.63

1.60

0.20

3.

NEST

-NEST

-hNEST

3

160.36

3.33

0.08

56

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. I, March 2010

TABLE 4. Cumutative model weights of variables in a
priori multinomial logistic regression models used to
identify habitat characteristics that predicted Burrowing
Owl fledging success in grassland and urban environments
in southern New Mexico for 2000 and 2001.

Habitat

Variable (j)

W + (j)

Grassland

Year

0.99

Nests

1.00

NNFS

0.99

Satellites

1.00

Perch distance

0.99

Urban

Year

0.18

Nests

0.89

NNFS

0.17

Satellites

0.73

Perch distance

0.61

Urbanization

0.36

within 50-111 of each nest. The second ranking
model contained the number of nests within a 75-
m radius, number of satellite burrows within a 50-
m radius, and measure of urbanization (Table 3).
Nests in the high success category, as compared to
failed nests, had fewer nests within a 75-m radius
(0.40 vs. 0.80) and nests in open space had higher
success than nests in developed areas (Table 3).
The number of satellite burrows was retained in
both models with a higher number associated with
high versus low success. Perch distance was
retained in one of the top two models with greater
perch distance associated with high success
compared to low success and failure. Cumulative
model weights supported the number of nests
within a 75-m radius, number of satellite burrows
within a 50-m radius, and perch distance as
strongly inlluencing fledging success; number of
nests negatively, and number of satellite bun'ows
and perch distance positively influenced produc-
tivity (Table 4).

Owls in urban areas in 2001 re-used 51.4% of
nest burrows from the previous year whereas in
prairie dog colonies 22.0% of the nest burrows
were re-used. There was no significant relation-
ship between burrow re-use and previous nest
success (P > 0.05). There was also no relation-
ship between burrow re-use and nest success in
urban environments (P = 0.373), but there was a
tendency for owls in grasslands occupying a
burrow used the previous year to produce more
fledglings than new nest burrows {V = 90, P =

o.oni).

DISCUSSION

BuiTowing Owl nest densities were approxi-
mately one order of magnitude greater in prairie
dog colonies than in urban areas. The majority of
nests in urban areas were associated with rock
squin'el bun'ows, which resemble prairie dog
colonies as both provide multiple nesting oppor-
tunities. However, rock squirrel colonies are
smaller, more ephemeral, and may be in open
rocky areas, flat ten'ain, imgation canals, em-
bankments, or among rock piles (Findley 1987).
Prairie dog colonies are larger, provide more
nesting opportunities, have higher burrow densi-
ties, occur in grasslands, and are more stable
(King 1955). Prairie dog colonies had a greater
number and density of buirows, covered substan-
tially more area compared to rock squiirel
colonies, and were surrounded by more foraging
habitat.

Owls nesting in urban environments maintained
nearest neighbor distances (X ± SE = 159 ±
12 m) similar to other studies in urban and
agricultural systems (Rosenberg and Haley
2004) whereas spacing of nests in prairie dog
colonies in our study was closer (A" ± SE = 50 ±
6 m) than commonly observed in this system
(Desmond et al. 1995, Griebel and Savidge 2007).
However, available data supports closer spacing
in prairie dog colonies for southern sites (Ross
1974, McNicoll 2005, Teaschner 2005), possibly
attributable to fewer and smaller colonies result-
ing in fewer suitable nesting sites. This is
supported by the rapid occupancy of small
colonies (A ± SE = 4.85 ± 1.33 ha) on our
study area by Burrowing Owls immediately
following establishment (T. E. Waddell, pers.
comm.).

Eledging success did not differ between the two
habitat types as predicted but was low in
comparison to average clutch sizes for Burrowing
Owls (7-9 eggs; Haug et al. 1993) and reproduc-
tive performance from several other studies
(CoLilombe 1971; Martin 1973; Botelho and
Arrowood 1998; Lutz and Plumpton 1999; MJD,
unpubl. data). Different mechanisms appeared to
be influencing the low productivity at these sites.
Sixty-eight percent of all nest failures in our study
occurred in urban sites; nests in these sites tended
to either fail or be highly successful. In compar-
ison, grassland sites had lower rates of failure but
produced fewer young per successful nest. Higher
nest failure in urban areas suggests higher rates of

Bcrardclli cl al. • BURROWING OWL REPRODUCTIVE SUCCESS

57

nest predation and/or abandonment in this envi-
ronment, whereas fewer lledglings per successful
nest in grasslands suggest increased competition
for resources. We were not, however, able to
differentiate between predation and nest abandon-
ment. Owls in west Texas experienced higher
rates of disturbance in urban compared to rural
environments (Chipman 2006), but it was not
known if this influenced productivity. Foraging
data from this study suggested that owls in urban
environments may have higher foraging rates
possibly compensating for increased disturbance.
Studies in Canada (Wellicome 2005) and Califor-
nia (Haley 2002), using food supplementation,
found that Burrowing Owls were food-limited
during the nestling period. York et al. (2002)
attributed low reproductive success of Burrowing
Owls in the Imperial Valley of California to the
lower amounts of vertebrate prey in their diets
compared to other studies. Fledging rates in
grassland habitats appeared to be negatively
impacted by the high density of owls in these
small, recently established colonies. This was
most apparent in 2001 when number of nests in
prairie dog colonies almo.st doubled, resulting in a
higher density of nesting owls compared to the
previous year.

Measures of BuiTowing Owl density appeared
to strongly influence productivity at both sites but
for different reasons. Productivity in our grassland
habitats was strongly negatively impacted by
measures of nest/owl density resulting in fewer
fledglings per successful nest. This suggests that
competition for resources and/or aggressive inter-
actions were factors in this environment. Griebel
and Savidge (2007) in South Dakota reported
higher productivity for nests that were further
from the nearest neighboring nest. The higher
success of nests with fewer satellite burrows in
our study was associated with nests along colony
edges, suggesting that reproductive success was
higher for edge nests. Several studies have
suggested that Burrowing Owls prefer to nest
near colony edges (Eckstein 1999, Desmond et al.
2000, Orth and Kennedy 2001). Possible advan-
tages include reduced competition related to fewer
surrounding nests and easier access to off-colony
foraging. Perching distance is an indication of the
territory size suiTOunding the nest burrow and
sentinel owls closer to the actual nest likely had
smaller territories and increased competitive
interactions. Owl proximity to the nest may also
increase nest vulnerability to predation; however.

predation rates in the incubation and early nestling
stage for this habitat type were low (81% ol nests
were successful). Nest failure rates were higher in
urban habitats where owls perched lurther from
the nest. The combination of increased owl
densities and below normal precipitation in
grassland habitats in 2001 likely contributed to
increased competition for fewer resources on
those sites. Our results from grassland habitats
support the need for larger prairie dog colonies in
the southern plains that have experienced greater
fragmentation compared to northern populations
(Miller et al. 1994). Desmond et al. (1995) found
that BuiTOwing Owls in prairie dog colonies
exhibited clustered distributions with greater
inter-nest distance when colonies were ^35 ha,
and were randomly distributed with smaller inter-
nest distances in colonies <35 ha. These data
suggest owls require larger colonies to exhibit
preferred nest spacing, and our results indicate
that closer nest spacing negatively impacts
productivity.

Nest density in urban areas also appeared to
have the strongest impact on reproduction. Nests
in the high success category were associated with
lower densities of nesting owls compared to the
low success category, suggesting competition for
resources also influenced reproduction at these
sites. However, failed nests had the lowest nest
density and tended to be solitary, suggesting the
presence of other nesting pairs may increase
productivity, possibly through increased predator
detection, as long as sufficient space is available
to reduce competition for nearby resources. Most
unsuccessful nests failed during the incubation or
early nestling .stages, suggesting nest abandon-
ment or predation. Increased development in
Florida resulted in reduced owl densities despite
higher prey populations in these environments
(Millsap and Bear 2000).

The loss of suitable burrows and space for
multiple nesting pairs may increase predation risk
for owls nesting in developed areas. Productivity
in urban environments in our study was also
associated with land use practices with high nest
success associated with open space compared to
developed areas. Open space is associated with
larger habitat patches that provide more nesting
opportunities and may have lower predation
pressure and less disturbance (vehicular or
pedestrian traffic). Density of some of the major
urban predators, including domestic cats {Fclis
catus) and common raccoons (Procyon lotor) are

58

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. I, March 2010

likely higher in residential areas; we observed
owls mobbing domestic cats at urban sites on
several occasions. As Burrowing Owls become
increasingly associated with urban environments,
city planning offices should work with local and
state biologists to identify and preserve areas
important to Burrowing Owls. This should include
identifying areas free from disturbance with
sufficient space for multiple nesting pairs, pres-
ervation and management of fossorial rodent
populations associated with Buirowing Owls,
and installing and maintaining artificial nest
burrows when necessary.

Owls in urban habitats re-used significantly
more nest burrows than those in grassland sites.
Two factors likely contributed: burrow availabil-
ity and resident owls. Fewer suitable bun'ows
were available to owls in urban compared to
grassland habitats. Holmes et al. (2003) reported
re-use rates in Oregon ranging from 57 to 87%
where owls primarily nest in American badger
{Taxidea taxus) burrows. They attributed high re-
use rates to a shortage of suitable bun'ows,
especially in sandy soils. Fewer burrows are
available in rock squirrel colonies, and rock
squirrel populations in urban areas are often
eradicated due to conflict with human activities.
A portion of our urban owl population was
resident year-round (27% of urban owls in 2001
were present the first week in Feb). Bun'owing
Owls in a mainly resident population in California
had 85% nest site fidelity (Rosenberg and Haley
2004). The combination of a partially year-round
resident population and fewer suitable nest
burrows in urban compared to grassland habitats
likely contributed to a higher re-use rate in urban
environments.

ACKNOWLEDGMENTS

T. E. Waddell, J. C. Truett. and T. A. Ostemier of Ihc
Turner Foundation and the Turner Endangered Speeies
Fund granted access to the Armendaris Ranch and
graciously provided logistical support. T. S. Schrader of
New Mexico State University and T. J. Hunteman ot the
city of Las Cruces assisted with nest mapping and AreView.
Field a.ssistance was provided by J. S. Enriquez. L. A.
Murphy, I. R. Murray. E. J. Quintana, and J. P. Waring.
This manuscript was improved by comments from P. C.
Arrowood. C. E. Braun, E. L. Fredrickson. S. R. Sheffield,
B. C. Thompson, and an anonymous reviewer. Funding was
provided by Threatened and Endangered Species Inc. (T&E
Inc.) and New Mexico State University. This is a New
Mexico Agricultural Experiment Station publication, sup-
ported by state funds and the U.S. Hatch Act. This
publication is dedicated to the memory of T. O. Wooten.

a staunch advocate for research and conservation in the

Chihuahuan Desert.

LITERATURE CITED

Botelho, E. S. and P. C. Arrowood. 1998. The effect of
burrow site use on the reproductive success of a
partially migratory population of western Burrowing
Owls. Journal of Raptor Research 32:233-240.

Burnham, K. P. and D. R. Anderson. 2002. Model
selection and multimodel inference: a practical
information-theoretic approach. Springer-Verlag,
New York, USA.

Chipman, E. D. 2006. Behavioral ecology of western
Burrowing Owls {Athene ciinicidaria hypiigaea) in
northwestern Texas. Thesis. Texas Tech University,
Lubbock, USA.

COULOMBE, H. N. 1971. Behavior and population ecology
of the Burrowing Owl {Speotyto cunicularia) in the
Imperial Valley of California. Condor 73:162-176.

Conway, C., V. Garcia, M. D. Smith. L. A. Ellis, and J.
L. Whitney. 2006. Comparative demography of
BuiTowing Owls in agricultural and urban landscapes
in southeastern Washington. Journal of Field Orni-
thology 77:280-290.

Desmond, M. J. and J. A. Savidge. 1999. Satellite burrow
use by Burrowing Owl chicks and its influence on nest
fate. Studies in Avian Biology 19:128-130.

Desmond, M. J., J. A. Savidge, and T. F. Seibert. 1995.
Spatial patterns of Burrowing Owl (Speotyto cunicularia)
ne.sts in black-tailed prairie dog (Cynomys hulovicianus)
towns. Canadian Journal of Zoology 73:1375-1379.

Desmond, M. J., J. A. Savidge, and K. E. Eskridge. 2000.
CoiTelations between Burrowing Owl and black-tailed
prairie dog declines: a 7-year analysis. Journal of
Wildlife Management 64:1067-1075.

Desmond, M. J., T. J. Parsons, T. O. Powers, and J. A.
Savidge. 2001. An initial examination of mitochon-
drial DNA structure in Burrowing Owl populations.
Journal of Raptor Research 35:274-281.

Desante, D. F., E. D. Ruhlen, and D. K. Rosenberg.
2004. Density and abundance of Burrowing Owls in
the agrieultural matrix of the Imperial Valley,
California. Studies in Avian Biology 27:1 16-1 19.

Eckstein, R. T. 1999. Local and landscape factors affecting
nest site selection and nest success of Burrowing Owls
in western Nebraska. Thesis. University of Nebraska.
Lincoln, USA.

Findley, J. S. 1987. The natural history of New Mexico
mammals. University of New Mexieo Press, Albu-
querque. USA.

Garcia. V. and C. J. Conway. 2009. What constitutes a
nesting attempt? Variation in criteria causes bias and
hinders comparisons across studies. Auk 126:31^0.

Gorman, L. R., D. K. Rosenberg, N. A. Ronan, K. L.
Haley, J. A. Gervais, and V. Franke. 2003.
Estimation of reproductive rales of Burrowing Owls.
Journal of Wildlife Management 67:493-500.

Griebel, R. L. and j. a. Savidge. 2007. Factors
influencing Burrowing Owl reproductive performance

Bcrardi’lli et al. • BURROWING OWL REPRODUCTIVE SUCCESS

59

in unfragniented shoitgrass prairie. Journal of Raptor
Research 41:212-221.

H.aley, K. L. 2002. The role of food limitation and
predation on reproductive success of Burrowing Owls
in southern California. Thesis. Oregon State Univer-
sity, Corvallis, USA.

Hartsough, M. J. 2002. Dietary preferences of black-tailed
prairie dogs reintroduced into a Chihuahuan Desert
grassland. Thesis. New Mexico State University, Las
Cruces, USA.

Haug, E. a. and L. W. Oliphant. 1990. Movements,
activity patterns, and habitat use of Burrowing Owls in
Saskatchewan. Journal of Wildlife Management 54:27-35.

Haug, E. A., B. A. Millsap, and M. S. Martell. 1993.
Burrowing Owl. The birds of North America. Number
61.

Holmes, A. L., G. A. Green, R. L. Morgan, and K. B.
Livezey. 2003. Burrowing Owl nest success and
burrow longevity in north central Oregon. Western
North American Naturalist 63:244-250.

Holroyd, G. L., R. Rodriguez-Estrella, and S. R.
Sheffield. 2001. Conservation of the Burrowing Owl
in western North America: issues, challenges, and
recommendations. Journal of Raptor Research 35:399-
407.

Hosmer, D. W. and S. Lemeshow. 2000. Applied logistic
regression. John Wiley and Sons, New York, USA.

King, J. A. 1955. Social behavior, social organization, and
population dynamics in a black-tailed prairie dog town
in the Black Hills of South Dakota. Contributions of
the Laboratory of Vertebrate Biology, Number 67.
University of Michigan, Ann Arbor, USA.

Lantz, S. J., C. Conway, and S. H. Anderson. 2006.
Multiscale habitat selection by Burrowing Owls in
black-tailed prairie dog colonies. Journal of Wildlife
'Management 71:2664—2672.

Lutz, R. S. and D. L. Plumpton. 1999. Philopatry and nest
site reuse by Burrowing Owls: implications for
productivity. Journal of Raptor Research 33:149-153.

Martin, D. J. 1973. Selected aspects of Burrowing Owl
ecology and behavior in central New Mexico. Condor
75:446-456.

McNicoll, j. L. 2005. Burrowing Owl {Athene cunicii-
laria) nest site selection in relation to prairie dog
colony characteristics and surrounding land-use prac-
tices in Janos, Chihuahua, Mexico. Thesis. New
Mexico State University, Las Cruces, USA.

Miller. B., G. Ceballos, and R. Reading. 1994. The
prairie dog and biotic diversity. Conservation Biology
8:677-681.

Millsap. B. A. and C. Bear. 2000. Density and
reproduction of Burrowing Owls along an urban
development gradient. Journal of Wildlife Manage-
ment 64:33-41.

Mrykalo, R. j., M. M. Grigione, and R. J. Sarno. 2009.
A comparison of available prey and diet ol Florida
BuiTOwing Owls in urban and rural environments: a
first study. Condor 111: 556—559.

Oakes. C. L. 2000. History and consequences of keystone
mammal eradication in the desert grasslands: the

Arizona black-tailed prairie dog (Cynontys luclovicia-
niis arizonensis). Dissertation. Texas A&M University,
College Station, USA.

Orth, P. B. and P. L. Ki;nnedy. 2001. Do land-u,se
patterns inlluence nest-site selection by Burrowing
Owls {Athene ciinicniarici hypugueci) in northeastern
Colorado? Canadian Journal of Zoology 79:1038-
1045.

PouLtN, R. G., L. D. Todd. K. M. Dohms, R. M. Brigham,
AND T. 1. Wellicome. 2005. Factors associated with
nest- and roost-burrow selection by Burrowing Owls
{Athene ciinicniaria] on the Canadian prairies. Cana-
dian Journal of Zoology 83:1373-1380.

Rosenberg, D. K. and K. L, Haley. 2004. The ecology of
Burrowing Owls in the agroecosystem of the Imperial
Valley, CA. Studies in Avian Biology 27:120-135.

Ross, P. V. 1974. Ecology and behavior of a dense colony
of Burrowing Owls in the Texas Panhandle. Thesis.
West Texas State University, Canyon, USA.

Sauer, J. R., J. E. Hines, and J. Fallon. 2007. The North
American Breeding Bird Survey, Results and Analysis
1966-2006. USGS, Patuxent Wildlife Research Cen-
ter, Laurel, Maryland, USA.

Sheffield, S. R. 1997. Current status, distribution, and
conservation of the Burrowing Owl {Speotyto cunicii-
laria) in midwestem and western North America.
Pages 399^07 in Biology and conservation of owls of
the northern hemisphere: Second International Sym-
posium (J. R. Duncan, D. H. Johnson, and T. H.
Nicholls, Editors). USDA, Forest Service, General
Technical Report NC-190. Northcentral Forest Exper-
iment Station, St. Paul, Minnesota, USA.

Systat. 2007. Systat for Windows: Systat Version 12.0.
Systat Software Inc., San Jose, California, USA.

Teaschner, a. 2005. Burrowing Owl nest site use and
productivity on prairie dog colonies in the southern
high plains of Texas. Thesis. Texas Tech University,
Lubbock, USA.

U.S. Census Bureau. 2000. Population, housing units,
area, and density. http://factfinder.census.gov/Itome/
en/datanotes/expsfl u.htm (accessed 5 December
2002).

Wellicome, T. 1. 2005. Hatching asynchrony in Burrowing
Owls is inlluenced by clutch size and hatching success
but not by food. Oecologia 142:326-334.

WE.SEMANN, T. AND M. RowE. 1987. Factors influencing
the distribution and abundance of Burrowing Owls in
Cape Coral, Florida. Pages 129-137 in Proceedings of
the national symposium on urban wildlife (L. W.
Adams and D. L. Leedy, Editors). National Institute
for Urban Wildlife, Columbia, Maryland. USA.

WE.STERN Regional Climate Center, Desert Research
Institute. 2002. Monthly climate summary, southern.
NM 1959-2005. Reno. Nevada. USA. htlp://www.
wrcc.dri.edu/cgi-bin/cliMAlN.pl7nmstat (accessed 19
September 2002).

York, M. M., D. K. Rosenberg, and K. K. Sturn. 2002.
Diet and food breadth of Burrowing Owls {Athene
cnniciil(irici) in the Imperial Valley, California.
Western North American Naturalist 62:280-287.

The Wilson Journal of Ornithology 122( 1 );60-67, 2010

DIETARY TRENDS OE BARN OWLS IN AN AGRICULTURAL
ECOSYSTEM IN NORTHERN UTAH

CARL D. MARTI'

ABSTRACT. — Bam Owl {Tyto alha) diets were studied for 15 years in Utah. Ninety-eight percent of 1 1 1,016 prey items
were mammals, heavily dominated by voles (Microtus spp.). Food-niche breadth (FNB) was 3.33 for the entire sample and
varied gradually but significantly among the 15 years and among seasons. Frequency of prey in the diet did not vary
significantly from year to year or among seasons. Mean daily temperatures did not vary significantly among years but
annual precipitation totals and days when deep snow covered the ground varied significantly among years. Irrigation for
agriculture may have partially mitigated annual precipitation fluctuations. Flay, one of the most important crops on the
study area, increased over the study period and other crops decreased slightly in the amount planted. Flectares of hay
planted, hectares of corn planted, and hectares of barley planted were the variables that combined to best predict annual
FNB. Received 9 February 2009. Accepted 31 July 2009.

Long-term studies can reveal how variation in
prey populations, climatic factors, and changes in
the landscape might influence the diet of Barn
Owls (Tyto alha). Love et al. (2000), for example,
found significant changes in diets of Barn Owls in
the United Kingdom that they attributed primarily
to changing agricultural practices. Taylor (1994)
examined Barn Owl diets over 12 continuous
years in a Scottish agricultural area and reported
fluctuations in prey consumed related to cyclic
changes in small mammal populations.

Barn Owls prey mainly on a wide variety of
mammalian species over their cosmopolitan range
(Taylor 1994, Bruce 1999, Marti et al. 2005); they
eat mostly small, nocturnal mammals (Herrera
and Jaksic 1980, Jaksic et al. 1982), only
occasionally capturing birds at high frequencies
(Brosset 1956, Corner 1978). Barn Owl diets have
been studied most intensively in the United States
and Europe (Campbell et al. 1987, Taylor 1994,
Marti et al. 2005), but also in South .America
(Jaksic et al. 1982, Bellocq 2000), Africa (Vernon
1972, Coodman 1986), Asia (Lenton 1984,
Mahmood-Ul-Hassan et al. 2007), and Australia
(Morton and Martin 1979, Palmer 2001). The
broad geographic distribution and the large
quantity of published dietary information make
the Barn Owl a good candidate for understanding
factors that affect diet variation.

Collectively, the large number of published
studies provide an overview of the diet of Barn
Owls, including types of prey taken (Taylor 1994,
Bruce 1999, Marti et al. 2005), size of prey
(Dickman et al. 1991, llle 1991, Marti et al. 1993),
habitats used (Begall 2005, Bond et al. 2004,

' Raptor Research Center. Boi.sc State University, Boi.se,
11) 83712, USA; e-mail: cmarti@q.eom

Askew et al. 2007), effect of season on diet
(Baudvin 1983, Campbell et al. 1987), and the
influence of climate on diet (Avery 1999, Jaksic
and Lazo 1999). However, most studies of Barn
Owl diets have been short-term and at risk of
being biased because data were often collected
during a single season or year and, at times,
during atypical conditions, e.g., abnormal weather
or in areas with unusual prey populations (Hey-
wood and Pavey 2002, Sahores and Trejo 2004,
Altwegg et al. 2006). Short-term studies, thus,
cannot provide insight into the dynamics of
predation.

1 studied Barn Owls in a north-temperate
agricultural area with the objectives to: ( I )
describe the diet over a 15-year period, (2)
quantify the effects of weather variation on the
diet, and (3) quantify the impact of vegetation
changes on the diet.

METHODS

Stiuly Area. — I studied Barn Owls in an area of
~ 1,000 km' in Box Elder, Davis, and Weber
counties, Utah. The area was in a long, naiTow,
and essentially flat valley at an elevation of
1,300 m between the Wasatch Mountains and the
Creat Salt Lake (4I°0I' N, 112° II' W). The
area was formerly shrub-steppe desert, but the
natural plant community has been supplanted by
irrigated agriculture and urban development.
Primary crops were hay, corn, wheat, barley,
and some livestock pasture. Uncultivated land was
limited to narrow strips along roads, water
courses, borders of agricultural fields, and adja-
cent to the Great Salt Lake. Annual precipitation
averaged 35 cm, and mean temperatures for
January and July were —3.5° C and 23.9° C,

60

A/<//7/ • TRENDS IN BARN OWL DIETS IN NORTHERN U'l'All

61

respectively (http://www.ncdc.noaa.gov/oa/ncdc.
html). Bam Owls, resident on the study area,
occasionally nested in old buildings and hay-
stacks, but most used nest boxes (Marti 1997).

Source of Dora. — Collection sites were nest
boxes placed individually in abandoned agricul-
tural silos (Marti et al. 1979); I visited all boxes
and collected regurgitated pellets each month
from 1977 to 1991. Boxes were placed throughout
the study area with locations dictated by avail-
ability of silos. I identified and quantified prey
items in pellets using standard methods (Marti et
al. 2007). The smallest sampling unit for diet
analyses consisted of prey from 1 month at one
collection site. These samples were combined for
analyzing diets by season or year.

Description of Food-niche Breadth. — ^Food-
niche breadth (FNB) is a measure of dietary
diversity incorporating measures of the number of
kinds of prey eaten and the relative proportions of
the kinds of prey in the diet. I prepared dietary
data for describing FNB by identifying mamma-
lian prey to species. Birds were also identified to
species when possible, but insects were identified
only to class. I estimated food-niche breadth
(FNB) using the formula:

FNB=l/^p?,

where /?, is the proportion of prey category , (each
kind of prey identified) in the Barn Owl diet
(Levins 1968). Larger values of this index indicate
a higher dietary diversity.

Environmental Variables. — I obtained climatic
data — air temperature, precipitation, and number
of days with snow on the ground — from National
Climatic Data Center (http://www.ncdc.noaa.gov/
oa/ncdc.html) weather stations in the study area
and calculated daily means for data recorded at
three stations for use as variables that might explain
changes in FNB. Variation in ambient temperature
can influence prey use because changes in
temperature may alter prey activity and affect the
quantity of food needed by Barn Owls. Precipita-
tion may directly influence Barn Owl hunting
activity and indirectly affect prey population size
and availability through its influence on vegetation.
Snow cover may also affect prey use because some
prey species, particularly voles (Microtus spp.),
buiTow under the snow, making them less vulner-
able to owl predation.

Vegetation composition may also affect Barn
Owl diets. My study area was agricultural land

with interspersed housing developments and most
vegetation was not native. 1 obtained annual crop
data — number of hectares of each major crop
planted — from the U.S. Department of Agricul-
ture (http://www.nass.usda.gov) for use as explan-
atory variables of FNB. I calculated an annual
crop-diversity index using the same formula used
for calculating FNB and then compared it among
seasons and years using PROC GLM (SAS
Institute 2000).

Dietary Variation. — I used contingency table
analysis (PROC REGR) (SAS Institute 2000) to
evaluate changes in prey frequencies. Variation in
diet was further evaluated by comparing FNB
among years and seasons (spring [Mar-May],
summer [Jun-Aug], fall [Sep-Nov] and winter
[Dec-Feb]) and years using PROC GLM (SAS
Institute 2000).

Variation in Food-niche Breadth. — I used
multiple linear regression (PROC GLM) (SAS
Institute 2000) to examine the relationship of FNB
with candidate explanatory variables. The avail-
able variables were: mean annual precipitation;
mean annual daily temperature; number of days
with deep snow cover; hectares planted to hay,
wheat, corn, and barley; and the annual diversity
of crops planted. I did not include mean daily
temperature or crop diversity in model selection
because those variables did not vary significantly
among years. The fit between candidate models
and the data was evaluated using Akaike’s
information criterion (Burnham and Anderson
2002). I reported the adjusted R~ to assist in
evaluating the overall fit of the candidate models.

RESULTS

Food-niche Description. — I identified 111,016
prey items from 1 ,426 monthly samples collected
at 31 sites. Owls consumed at least 26 species of
prey, but 93.6% of the prey items were voles
(75.9%; Microtus penn.sylvanicus and M. monta-
nus) and mice (17.7%; Peroniy.scus maniculatus,
Reithrodontomys megalotis, and Mus musculus:
Table 1 ). The number of prey items collected each
year (1977 = 10,043; 1978 = 7,784; 1979 =
13,607; 1980 - 9,069; 1981 = 9,460; 1982 =
7,437; 1983 = 9,306; 1984 = 6,181; 1985 = 2,413;
1986 = 4,247; 1987 = 9,370; 1988 = 6,809; 1989
= 4,055; 1990 = 6,688; 1991 = 4,547) varied
among years in relation to the number of nesting
Barn Owls (/^ = 0.59, P < 0.02).

FNB was 3.33 when calculated with all prey
items combined (/; = 1 1 1,016), whereas the mean

62

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. I, March 2010

TABLE 1. Diet of a
Utah, 1977-1991.

Barn Owl population

in northern

Number in
diet (%)

% Occur-
rence at 3 1
collection
sites

% Occur-
rence in
1,426 pellet
collections

Mammals

109,035 (98.2;

) 100.0

100.0

Sore.x vagrans

4,223 (3.8)

100.0

70.1

Myotis spp.

8 (tr.9

16.1

0.4

Eptesicus fusciis
Sylvilagtis niittalii

7 (tr.)

6.5

0.1

(neonate)

3 (tr.)

9.7

0.2

Thomomys talpoide.s

649 (0.6)

54.8

8.6

Perognathus pan ics
Reithroclontomys

2 (tr.)

6.5

0.1

megaloti.s

Peromysciis

6,157 (5.9)

100.0

74.7

maniculatii.s
Ondatra zihethiciis

6,853 (6.2)

100.0

82.6

(juvenile)

Microtiis

40 (tr.)

35.5

1.7

pennsylvaniciis

42,718 (38.5) 100.0

99.2

M. montaniis

Rattiis norvegiciis

41,527 (37.4) 100.0

99.3

(Juvenile)

308 (0.3)

90.3

13.2

Mas miiscitliis

6,193 (5.6)

100.0

78.2

Mustela frenata

1 (tr.)

3.2

0.1

Birds

2.5 (1.8)

96.4

50.6

Ralliis liniicola

4 (tr.)

12.9

0.3

Porzana Carolina

76 (tr.)

51.6

2.9

C ha rad tins vociferus

1 (tr.)

3.2

0.1

Reciin'irostra

americana

1 (tr.)

3.2

0.1

Gallinago gallinago

16 (tr.)

16.1

0.6

Cohmiha livia

23 (tr.)

35.5

1.4

Tyto alha (nestling)

1 (tr.)

3.2

0.1

C istothonis pahtstris

36 (tr.)

29.0

1.3

Stnrniis vulgaris

382 (0.3)

90.3

15.3

Sturnella neglecta

11 (tr.)

19.4

0.6

Passer domesticiis

146 (0.1)

74.2

5.0

Unidentified blackbird*’ 198 (0.2)
Unidentified small bird

90.3

8.8

(~20 g) 603 (0.5)

Unidentified small bird

93.5

22.3

(—40 g)

455 (0.4)

lOO.O

17.7

Insects, unidentified

14 (tr.)

29.0

0.8

tr. = <0. 1 %.

^ Red-winged (Afiekiius phocniceus) or Yellow-headed {Xanthocephalus
xanthoci'phalus) blackbirds.

FNB of the 1.426 monthly samples was 2.78 (SD
= 0.85, range = 1.07-8.7). FNB varied among
years (T14.3W = 4.1, P = 0.0001), gradually
decreasing during the study (Fig. 1 ). FNB differed
significantly between consecutive years only for
1979 and 1980 (T1.57 = 6.96, P = 0.01 ) and 1987
and 1988 (T1.58 = 12.0, P = 0.001 ). The decrease

5

1977 1979 1981 1983 1985 1987 1989 1991

Year

FIG. 1 . Barn Owl food-niche breadth in northern Utah
by year.

in FNB was coirelated with an increa.se in the
frequency of voles in the diet {r^ = —0.99, P <
0.0001). Overall, FNB varied among seasons
(Fig. 2) and, although FNB was not statistically
different among seasons (F^ 55 = 2.4, P = 0.07),
the pattern observed may be biologically signif-
icant.

Environmental Variability. — ^Total precipitation
varied among years (F14 28 — 1 1.8, F = 0.0001),
but mean daily temperature did not (F14 |f,5 = 0.2,
P = 0.99). The number of days with snow
sufficiently deep (>15 cm) to interfere with prey
detection and capture by Barn Owls also varied
among years (F14J5 = 3.4, P = 0.0003). The
frequency of voles in the diet of Barn Owls during
winters having the greatest number of days with
deep snow was less than in milder winters, but the
relationship was not significant (F|,i4 = 0.16, r
— 0.01, P = 0.69). Winters when the ground was
covered longest with deep snow were followed by

XJ

CD

0

0

-C

o

'c

*6

o

o

LL

Winter

Season

FIG. 2. Barn Owl food-niche breadth in northern Utah
by season, years 1977-1991 combined.

• TRENDS IN BARN OWL DIETS IN NOR THERN UTAH

63

Year

FIG. 3. Number of days with deep snow (>15 cm) on
ground (bars) in northern Utah and frequency of voles in the
diet (line) of Barn Owls during spring.

Dietary Variation. — Frequencies ol' different
prey items consumed did not vary among years
(5^2 = 64.3, df ^ 10, P = 0.67; Fig. 5), among
seasons for all years combined ix~ = 6.4, dl =15,
p = 0.97), or among seasons within years. The
percentage of voles (both species combined) in
the diet of Barn Owls varied by an average of only
2.9% from year to year (x" = 4.6, df =14, Z' =
0.99).

Variation in Food-niche Breadth.— Tht model
that best explained variation in FNB among years
included the variables: hectares of hay planted,
hectares of corn planted, and hectares of barley
planted (Table 2).

DISCUSSION

a significant increase in the frequency of voles in
spring Barn Owl diets (Ti 14 = 7.79, r = 0.32, P
= 0.01; Fig. 3). Summer diets following winters
with deep snow cover also contained a higher
frequency of voles (C1J4 = 4.6, r = 0.26, P =
0.05). The same relationship continued into
autumn but the trend was not significant (F,j4
= 1.5, /- = 0.1, P = 0.25).

The number of hectares planted in different
crops varied among years with the number of
hectares of hay increasing during the study and
the area devoted to other crops decreasing
(Fig. 4). Crop diversity did not vary among years
(6-11.35 = 1-0, r = 0.32, P = 0.45).

Bam Owls exhibited little variation in diet over
a 15-year period in northern Utah where habitat
changes were minimal and variation in precipita-
tion was buffered by crop irrigation, minimizing
the effects of year-to-year variation of precipita-
tion on vegetation. In contrast to my findings,
Taylor (1994) found pronounced fluctuation in
Barn Owl diets corresponding to vole population
cycles during a long-term study in Scotland. The
proportion of voles in the diet of Barn Owls in my
study, unlike several studies in Europe (Bohnsack
1966, de Bruijn 1979, Taylor 1994), did not show
cyclic fluctuations, possibly because irrigated
agriculture produced a consistent, high-quality

Area (ha) planted in crops in northem Utah (data for hay were not available for 1977-1979).

FIG. 4.

64

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. I. March 2010

'■o

o

c

CD

CD

c

0

o

0

CL

90

80

70

60

50

40

30

20

10

0

Sorex

Mus

Peromyscus

Reithrodontomys
■■ Microtus

i

i

£l

i

El

M

M rfiJ EJ ifl

J1

1977 1979 1981 1983 1985 1987 1989 1991

Year

FIG. 5. Frequency of major prey in Bam Owl diets in northern Utah by year.

habitat that damped vole population fluctuations
(Negus et al. 1986). Changes in Barn Owl diets
sampled in two periods separated by a 19-year
interval (1956-1974 and 1993-1997) in the
United Kingdom were mostly attributable to a
change from mixed-crop farms to more homog-
enous farms (Love et al. 2000). The mode of
farming did not change during my study, but
changes in proportion of crops planted did and
those changes were related to the small changes in

TABLE 2. Models of variation in food-niche breadth
among years using .seven explanatory variables: annual
precipitation (AP), number of days with deep snow on
ground (DOS), hectares of corn planted (HC), hectares of
hay planted (FIH), hectares of barley planted (HB), and
hectares of wheat planted (HW).“

Model

A,"

»v'

Adj. R-

HH -b HC -b HB

0

0.962

0.99

HC -b DOS -b AP

6.65

0.035

0.99

HH -b HC -b HW

12.78

0.002

0.96

HH -b HB -b AP

14.1 1

0.001

0.95

HH -b HC -b AP

14.91

0.001

0.94

“ Akaikc’.s information crilcrion (AIC) for Ihe best model was -42.8,1.

A, = difference between the AIC value of a given model and that of the
best-fitting model.

‘ w, = AIC model weight.

the diet — the more hay planted, the more voles in
the diet. Voles are known to be abundant in forage
crops like hay (Getz 1985), and Barn Owls in
Utah may have concentrated their foraging in the
vegetation that served as best vole habitat as they
did in the United Kingdom (Askew et al. 2007).

Voles were the most common prey in the diet of
Barn Owls in my study, and have been the most
numerous prey of these owls in many other
studies conducted in the north-temperate zone
(Campbell et al. 1987, Taylor 1994, Bruce 1999).
In studies where an index of small mammal
population size was available. Barn Owls captured
voles in greater proportion than expected based on
relative abundance (Marti 1974, Colvin 1984,
Askew et al. 2007). The high proportion of voles
in the diet of Barn Owls in my study also suggests
that voles were taken .selectively. I found strong
evidence of selective predation by Barn Owls on
voles inhabiting an island in the Great Salt Lake
10 km from the present study area, probably due
to the owls hunting preferentially in vole habitat
(Marti 1986); Barn Owl diets on that island
contained 81% voles by number despite vole
habitat being extremely limited. Voles were not
known to occur on Antelope Island prior to my
finding them in the Barn Owl diet (Bowers 1982,

A/r/r// • TRENDS IN BARN OWL DIETS IN NORTHERN UI'AII

65

Marti 1986). A similar pattern was seen in
southwestern Idaho where frequency of voles
was higher in Barn Owl diets at nests surrounded
by iirigated agricultural land (Marti 1988). Barn
Owls appeared to concentrate their foraging in
habitats where voles were most numerous both on
Antelope Island and in southwestern Idaho.

Fast and Ambrose (1976) and Derting and
Cranford (1989) found that captive Barn Owls
selected Microtiis over Peromyscus when given a
simultaneous choice. Prey size may be an
important factor in prey selection — Microtiis are
about twice the mass of Peromyscus and Mus, the
next largest among the most commonly taken prey
in my study. Captive Barn Owls, given a
simultaneous choice between two sizes of the
same species, selected larger prey up to the point
where prey became hard to subdue (Ille 1991).

It seems likely that Barn Owls may not
discriminate between the two vole species be-
cause both are found in the same habitats (Getz
1985), and are similar in appearance and mass. A
sample of voles trapped near my study area
(University of Utah Museum collection; H. J.
Egoscue, unpubl. data) revealed no difference in
mass between the species (t = 1.53, df = 63, F =
0.13). Despite these similarities, I separated the
two voles in my analyses because I did not know
whether Bam Owls can or do distinguish between
them.

Tores et al. (2005) concluded that Barn Owls
prefer the kind of prey that voles exemplify, but
readily switch to different prey if the prefemed
species declines below a certain level. Barn Owls
in Utah ate all of the small mammal species
known to occur in the area (Newey 1951, Frost
1970). Some prey were available or vulnerable
only at certain times of the year; seasonal
variation in uncommon prey was most noticeable
for northern pocket gophers (Thomomys tal-
poides), which were numerous in the diet in
May when young animals disperse above ground
(Nowak 1991). Birds were a minor component of
the diet, and the frequency of .some species in the
Bam Owl diet could be attributed to migration.
Sora (Porzana Carolina), for example, occurred in
the diet only in spring and early summer. Others,
like European Starlings (Sturnus vulgaris), were
not migratory and Barn Owls preyed on them
throughout the year. However, starlings reached
their highest numbers in the Barn Owl diet in
winter when communal roosting may have made
them more vulnerable.

1 observed the same pattern ol seasonal
variation in the diet ol Barn Owls in Utah that
Taylor (1994) reported in Scotland — frequency of
voles in the diet was lowest in spring, increased
into summer, and peaked in late lall and early
winter. During winters with more days of deep
snow cover. Bam Owls in my study preyed on
fewer voles. Winter fluctuations were likely
explained by voles under deep snow cover having
reduced vulnerability to predators (Canova 1989,
Halonen et al. 2007).

The number of Barn Owls breeding in the study
area varied 12-fold over the study period (Marti
1994, 1997), primarily due to differences in
severity of winter weather (Marti and Wagner
1985). However, owl population density did not
influence the rate of predation on voles; owl
density as indexed by the number of nest attempts
per year was not correlated with the frequency of
voles in the diet (ly = —0.25, P — 0.37).

Diets of Barn Owls inhabiting this agricultural
ecosystem exhibited remarkable uniformity over
the long term. The measures I applied — FNB
variation and prey frequencie.s — showed the diet
of Barn Owls in my study exhibited little
variation. The owls preyed primarily on voles at
all collection sites and during all seasons and
years, and the small changes that I found from
year to year likely were the result of an increase in
the amount of hay grown and a concomitant
increase in vole numbers. The reverse — changing
from grass-oriented agriculture to corn and
soybeans — has been associated with declines of
Barn Owl populations in Ohio (Colvin 1985).
Variation in annual precipitation in my study area
was mitigated by crop irrigation providing
consistent moisture levels each year, and, in turn,
consistent prey habitat. The only weather variable
strongly associated with diet variation ot Barn
Owls in Utah was the number of days the ground
was covered with deep snow which interfered
with owls ability to prey on voles.

ACKNOWLEDGMENTS

I ihank Weber State University for financial support
through Research and Professional Growth grants, and
equipment to conduct much of this research. P. W. Wagner
helped with data collection during the first 6 years of this
study. 1 thank B. C. Harvey for advice on using Akaike's
information criterion. I also thank Karen Steenhof, M. N.
Kochert, Motti Charter, Marco Restani, and an anonymous
reviewer for their comments which helped refine this
paper.

66

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. /, March 2010

LITERATURE CITED

Altwegg, R., a. Roulin, M. Kestenholz, and L. Jenni.
2006. Demographic effects of extreme winter weather
in the Barn Owl. Oecologia I49;44-51.

Askew, N. P., J. B. Searle, and N. P. Moore. 2007. Agri-
environment schemes and foraging of Barn Owls Tyto
alha. Agriculture, Ecosystems and Environment
1 18:109-1 14.

Avery, D. M. 1999. A preliminary assessment of the
relationship between trophic variability in southern
African Barn Owls Tyto alha and climate. Ostrich
70:179-186.

Baudvin, H. 1983. La regime alimentaire de la Chouette
Effraie (Tyto alha). Le Jean-le-Blanc 22:1-108.

Begall, S. 2005. The relationship of foraging habitat to the
diet ot Barn Owls (Tyto alha) from central Chile.
Journal of Raptor Re.search 39:97-101.

Bellocq, M. I. 2000. A review of the trophic ecology of
the Barn Owl in Argentina. Journal of Raptor Research
34:108-119.

Bohnsack. P. 1966. Uber die Ernahrung der Schleiereule,
Tyto alha, insbesondore auBerhalb der Brutzeit, in
einem wesholsteinischen Massenwechselgebiet der
Feldmaus, Microtiis arvalis. Corax 1:162-172.

Bond, G., N. G. Burnside, D. J. Metcalfe, D. M. Scott,
and j. Blamire. 2004. The effects of land-use and
landscape structure on Barn Owl (Tyto alha) breeding
success in southern England, U.K. Landscape Ecology
20:555-566.

Bowers, M. A. 1982. Insular biogeography of mammals in
the Great Salt Lake. Great Basin Naturalist 42:589-
596.

Brosset, a. 1956. Le regime alimentaire de FEffraye Tyto
alha au Maroc oriental. Alauda 24:303-305.

Bruce, M. D. 1999. Family Tytonidae (Barn-owls). Pages
34-75 in Handbook of the birds of the world. Volume
5. Barn-owls to hummingbirds (J. del Hoyo. A. J.
Elliott, and J. Sargatal, Editors). Lynx Edicions,
Barcelona. Spain.

Burnham, K. P. and D. R. Anderson. 2002. Model
selection and multimodel inference: a practical
information-theoretic approach. Second Edition.
Springer-Verlag, New York, USA.

Campbell, R. W., D. A. Manuwal. and A. S. Harestad.
1987. Food habits of the Common Barn-Owl in British
Columbia. Canadian Journal of Zoology 65:578-586.

Canova, L. 1989. Influence of snow cover on prey
selection by Long-eared Owls (A.sio otits). Ethology,
Ecology and Evolution 1 :367-372.

Colvin, B. A. 1984. Barn Owl foraging behavior and
.secondary poisoning hazard from rodenticide use on
farms. Dissertation. Bowling Green State University,
Bowling Green. Ohio, USA.

Coi.viN. B. A. 1985. Common Barn-Owl population decline
in Ohio and the relationship to agricultural trends.
Journal of Field Ornithology 56:224—235.

DE Bruijn. O. 1979. Voedseloccologie van ide Kerkuil Tyto
alha in Nederland. Limosa 52:91-154.

Derting, R. L. and j. a. Cranford. 1989. Physical and
behavioral correlates of prey vulnerability to Barn Owl

(Tyto alha) predation. American Midland Naturalist
121:11-20.

Dickman, C. R., M. Predavec, and a. J. Lynam. 1991.
Differential predation of size and sex classes of mice
by the Barn Owl, Tyto alha. Oikos 62:67-76.

Fast, S. J. and H. W. Ambrose III. 1976. Prey preference
and hunting habitat selection in the Barn Owl.
American Midland Naturalist 96:503-507.

Frost, J. L. 1970. The mammals of Davis County, Utah.
Thesis. University of Utah, Salt Lake City, USA.

Getz, L. L. 1985. Habitats. Pages 286-309 in Biology of
New World Microtus. (R. H. Tamarin, Editor).
American Society of Mammalogists Special Publica-
tion Number 12. Allen Press Inc., Lawrence. Kansas,
USA.

Goodman, S. M. 1986. The prey of Barn Owls (Tyto alha)
inhabiting the ancient temple complex of Karnak,
Egypt. Ostrich 57:109-1 12.

Gorner, M. 1978. Schleiereule, Tyto alha, als Vogeljiiger.
Beitraege zur Vogelkunde 24:273-275.

Halonen, M., T. Mappes, T. Meri, and J. Suhonen. 2007.
Influence of snow cover on food hoarding in Pygmy
Owls Glaiicidiiini passehnum. Ornis Fennica 84:105-
111.

Herrer.x, C. M. and F. M. Jaksic. 1980. Feeding ecology
of the Barn Owl in central Chile and southern Spain: a
comparative study. Auk 97:760-767.

Heywood, M. R. and C. R. Pavey. 2002. Relative
importance of plague rodents and dasyurids as prey
of Barn Owls in central Australia. Wildlife Research
29:203-207.

Ille, R. 1991. Preference of prey size and profitability in
Barn Owls Tyto alha guttata. Behaviour 1 16:180-189.

Jaksic, F. M. and I. Lazo. 1999. Response of a bird
assemblage in semiarid Chile to the 1997-1998 El
Nino. Wilson Bulletin 1 1 1:527-535.

Jaksic, F. M., R. L Seib, and C. M. Herrera. 1982.
Predation by the Barn Owl (Tyto alha) in Mediterra-
nean habitats of Chile, Spain and California: a
comparative approach. American Midland Naturalist
107:151-162.

Lenton, G. M. 1984. The feeding and breeding ecology of
Barn Owls Tyto alha in peninsular Malaysia. Ibis 126:
551-575.

Levins, R. 1968. Evolution in changing environments:
some theoretical explorations. Princeton University
Press, Princeton. New Jersey, USA.

Love. R. A.. C. Webbon, D. E. Glue, and S. Harris.
2000. Changes in the food of British Barn Owls (Tvto
alha) between 1974 and 1997. Mammal Review
30:107-129.

Mahmood-Ul-Hassan, M., M. A. Beg. and M. Mushtaq-
Ul-Hassan. 2007. Locality related changes in the diet
of the Barn Owl (Tyto alha stertens) in agroecosystems
in central Punjab, Pakistan. Wilson Journal of
Ornithology 119:479^83.

Marii. C. D. 1974. Feeding ecology of four sympatric
owls. Condor 76:45-61.

Marti, C. D. 1986. Barn Owl diet includes mammal
species new to the island fauna of the Great Salt Lake.
Great Basin Naturalist 46:307-309.

A/(//7/ • I RENDS IN BARN OWL DIETS IN NORTHERN UTAH

67

Marti, C. D. 1‘')88. A long-ierm study of food-niche
dynamics in the Common Barn-Owl: comparisons
within and between populations. Canadian Journal of
Zoology 66:1803-1812.

Marti, C. D. 1994. Barn Owl reproduction: patterns and
variation near the limit of the species' distribution.
Condor 96:468—484.

Marti, C. D. 1997. A 20-year study of Barn Owl {Tylo
alba) reproduction in northern Utah. Page 261 in
Biology and conservation of owls of the Northern
Hemisphere. (J. R. Duncan, D. H. Johnson, and T. H.
Nicholls, Editors). USDA, Forest Service, General
Technical Report NC-190. Northcentral Research
Station, St. Paul, Minnesota, USA.

Marti, C. D. and P. W. Wagner. 1985. Winter mortality
in Common Barn-Owls and its effect on population
density and reproduction. Condor 87:1 1 1-1 15.

Marti, C. D., M. Bechard, and F. M. Jaksic. 2007. Food
habits. Pages 129-149 in Raptor research and
management techniques (K. L. Bildstein and D. M.
Bird, Editors). Second Edition. Raptor Research
Foundation, Hancock House Publishers. Blaine,
Washington. USA.

Marti, C. D., A. F. Poole, and L. R. Bevier. 2005. Barn
Owl {Tylo alba). The birds of North America online
(A. Poole, Editor). bna.birds.cornell.edu/BNA/
account/Bam Owl (accessed 1 1 December 2008).

Marti. C. D., P. W. Wagner, and K. W. Denne. 1979.
Nest boxes for the management of Barn Owls. Wildlife
Society Bulletin 7:145-148.

Marti, C, D., K. Steenhof. M. N. Kochert, and J. S.

Marks. 1993. Community trophic structure; the roles
of diet, body size, and activity time in vertebrate
predators. Oikos 67:6-18.

Morton, S. R. and A. A. Martin. 1979. Feeding ecology
of the Barn Owl, Tylo alba, in arid southern Australia.
Australian Wildlife Research 6:191-204.

Negus, N. C., P. J. Berger, and B. W. Brown. 1986.
Microtine population dynamics in a predictable
environment. Canadian Journal of Zoology 64:785—
792.

Newey, j. P. 1951. Mammals of Weber County, Utah.
Thesis. University of Utah. Salt Lake City, USA.

Nowak, R. M. 1991. Walker’s mammals of the world.
Johns Hopkins Press, Baltimore, Maryland, USA.

Pai.mer, R. a. 2001 . Notes on the diet of the Barn Owl Tylo
alba from Mulyungarie Station in north-eastern South
Australia. South Australian Ornithologist 33:137-138.

Sahores, M. and a. Trejo. 2004. Diet shift of Barn Owls
(Tylo alba) after natural fires in Patagonia, Argentina.
Journal of Raptor Research 38:174-177.

SAS Institute. 2000. SAS/STAT software: changes and
enhancements. Release 8.1. SAS Institute Inc., Cary,
North Carolina, USA.

Taylor, I. 1994. Barn Owls: predator-prey relationships
and conservation. Cambridge University Press. Cam-
bridge, United Kingdom.

Tores, M.. Y. Motro, U. Motro, and Y. Yom-Tov. 2005.
The Barn Owl — a selective opportunist predator. Israel
Journal of Zoology 51:349-360.

Vernon, C. J. 1972. An analysis of owl pellets collected in
southern Africa. Ostrich 43:109-124.

The Wilson Journal of Ornithology 122( l);68-74, 2010

REGIONAL VARIATION IN DIETS OF BREEDING
RED-SHOULDERED HAWKS

BRAD N. STROBEL'-^ AND CLINT W. BOAL^

ABSTRACT. — We collected data on breeding season diet composition of Red-shouldered Hawks (Biiteo lineatus) in
south Texas and compared these data, and those reported from studies elsewhere to examine large scale spatial variation in
prey use in eastern North America. Red-shouldered Hawk diets aligned into two significantly different groups, which
appear to correlate with latitude. The diets of Red-shouldered Hawks in group 1, which are of more northern latitudes, had
significantly more mammalian prey and significantly less amphibian prey than those in group 2, which are at more
southerly latitudes. Our meta-analysis demonstrated the dietary flexibility of Red-.shouldered Hawks, which likely accounts
tor their broad distribution by exploiting regional variations in taxon-specific prey availability. Received 27 April 2009.
Accepted 28 August 2009.

Red-shouldered Hawks (Buteo lineatus\ hereaf-
ter RSHA) were once considered the most
abundant raptor in moist woodlands of North
America (Bent 1937), and are thought to prefer
mature contiguous forests, especially those asso-
ciated with wetlands (Craighead and Craighead
1956, Bednarz and Dinsmore 1981, Dykstra et al.
2008). Breeding Bird Survey data indicate RSHA
populations have declined in portions of the
midwestern and northeastern United States, while
populations in other areas appear to be relatively
stable or increasing (Sauer et al. 2008). Population
declines have heightened concern and protection
for RSHA by several state agencies (Jacobs and
Jacobs 2002).

Conversion of forests to agriculture is one
factor thought to have contributed to RSHA
population declines (Bednarz and Dinsmore
1981, Gehring 2003, Dykstra et al. 2008). Greater
amounts of forest edge and smaller forest stand
size caused by agricultural and silvicultural
practices may provide a competitive advantage
for habitat generalist raptors such as Red-tailed
Hawks (Buteo jamaecensis) (Bednarz and Dins-
more 1981). Wetland degradation and drainage
may also have reduced the quality and availability
of foraging areas important to RSHAs (Jacobs and
Jacobs 2002). Other studies have documented
apparently sustainable RSHA populations in
highly human altered landscapes (e.g.. Bloom et

' Texa.s Cooperative Fish and Wildlife Researeh Unit,
Department of Natural Re.sources Management, Texas Tech
University, laibbock, TX 79409, USA.

- U.S. Geological Survey, Texas Cooperative Fish and
Wildlife Research Unit, Department of Natural Resources
Management. Texas Tech University, Lubbock, TX 79409.
USA.

'Corresponding author; e-mail: bradley. .strobel@ttu.edu

al. 1993, Rottenborn 2000, Dykstra et al. 2001),
suggesting that factors other than habitat alter-
ation may be directly or indirectly contributing to
RSHA population trends. For example. Bloom
(1989) suggested that habitat use by RSHAs is
strongly influenced by prey availability. The
abundance and diversity of prey available to
breeding RSHAs is likely associated with, and
influenced by, landscape characteristics within
RSHA foraging habitat.

Previous studies of breeding season diets
indicate RSHAs use a variety of prey items across
their range. Diets documented for RSHAs nesting
in Missouri were primarily amphibians and
reptiles (Parker 1986), while nearly 75% of the
diets of RSHAs nesting in Maryland were
mammals (Portnoy and Dodge 1979). The diets
of breeding RSHAs in Arkansas (Townsend 2006)
and Texas (Strobel 2007), in contrast, were
largely invertebrates. Bednarz and Dinsmore
(1985) documented RSHA dietary differences
between two breeding seasons in Iowa, apparently
associated with annual differences in precipitation
and riparian flooding. Thus, it appears RSHAs
adapt to some natural and anthropogenic land-
scape changes, likely facilitated by their ability to
exploit a diverse suite of prey.

Regional variation in prey use by RSHAs is
likely caused by large-scale ecological differences
such as precipitation patterns, temperature varia-
tion, and composition of vegetation communities
influencing prey communities. Anthropogenic
changes to the landscape have also influenced
population trends of RSHAs in North America
(Brown 1971, Henny et al. 1973, Peterson and
Crocoll 1992, Jacobs and Jacobs 2002). Under-
standing the coarse scale patterns in prey u.se, and
the breadth and plasticity of RSHA diets may

68

Sirohel am! Boa! • RED-SIIOULDERED HAWK DIET

69

offer insights into productivity and ecology of
RSHA populations that would facilitate develop-
ment of sound conservation strategies for the
species. We used breeding season food habits
information from our study in south Texas and
published studies conducted on breeding RSHA
populations east of the Continental Divide to
examine large scale spatial variation in prey use
by eastern populations of Red-shouldered Hawks
in North America.

METHODS

Study Area and Field Procedures. — 'We located
active Red-shouldered Hawk nests on the Rob and
Bessie Welder Wildlife Refuge and Twin Oaks
Hunting Resort in San Patricio and Refugio
counties (respectively), Texas. Nests of breeding
RSHA pairs were located using broadcast survey
methods as described by McLeod and Andersen
(1998) and intensive systematic nest searching
between April and July 2005 and 2006. We
installed color video surveillance cameras (Mod-
els OC-225 and C3320AX, Clover Electronics®,
Cerritos, CA, USA) after eggs hatched and young
were ~1 week of age. We recorded all diurnal
activities using time-lapse VCRs (Piczel video
security products® and Security Labs®, Nobles-
ville, IN, USA) until each nesting attempt was
completed. We based our analysis on prey items
delivered to nestlings prior to fledging to reduce
bias caused by off camera prey deliveries. We
identified prey items to the most specific taxa
possible using regional field guides, museum
specimens, and Welder Wildlife Refuge staff
expertise (Strobel 2007).

Literature Search. — 'We conducted a literature
search for published studies (peer reviewed
publications, books, and theses) of RSHA breed-
ing season diet. Methods used in other studies
included: direct visual observation, video surveil-
lance, identification of uneaten prey remains,
composition of regurgitated pellets, and crop
contents analysis. We classified methods of the
publications as either direct observation (DO) or
prey remains analysis (PR).

Inconsistent methods of data collection are a
common concern in meta-analysis and the poten-
tial biases of different techniques to assess diets of
raptors have been noted previously (Redpath et al.
2001, Rogers et al. 2005, Marti et al. 2007).
Results from studies using prey remains analysis
can be skewed by higher persistence rates of
heavy boned prey species, such as mammals, and

falsely indicate higher frequency of occurrence of
those species (Marti et al. 2007). Correct identi-
fication of prey items in direct observation studies
often depends on prey item size and its distance
from the observer (Marti et al. 2007). We
attempted to reduce these potential confounding
effects by pooling prey types into five taxonomic
categories: mammals, birds, reptiles, amphibians,
and invertebrates. When studies used multiple
methods (i.e., DO and PR) to collect data and
presented those data separately, we only used
those collected through direct observation and
excluded unknown or unidentified prey items
(Table 1). Some studies presented yearly data
separately and, to remain consistent in our
analyses, we pooled all data from multiple years
within those studies.

Statistical Analyses. — ^We calculated the Pear-
son’s correlation coefficient, based on the pro-
portion of prey types reported in each study to
compare RSHA diets among studies (Krebs
1999). Values of Pearson’s correlation coefficient
range from 0 to 1 indicating no dietary overlap (0)
to complete dietary overlap (1). We used the
Pearson’s correlation coefficient as the distance
function within an unweighted pair-group method
using arithmetic means (UPGMA) cluster analysis
to calculate and compare the similarity of diets
reported for breeding RSHAs. We assessed the
significance of clusters using the approximately
unbiased significance level {P ~ \-au) calculated
from 10,000 multiscale bootstrap iterations of the
distance matrix using the PVCLUST package
(Suzuki and Shimodaira 2006). We used Ho-
telling’s T- test, in the ICSNP package, on the
arcsine transformed proportions of each prey
category to identify which prey categories con-
tributed to the formation of significant clusters.
Reported values represent means ± SD. Analyses
were conducted using program R version 2.8.0 (R
Development Core Team 2008).

RESULTS

We monitored prey use at 10 different Red-
shouldered Hawk nests in south Texas in 2005
and 2006, and collected 1,457 hrs of video
footage yielding 1,315 prey items categorized
into the five taxonomic categories (Table 1). We
located 1 1 other studies that reported the diets of
breeding RSHAs in different states in the eastern
United States (Table 1). Eew multi-year studies
identified prey delivered to individual nests or by
individual RSHAs. Thus, the independence of

70

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. I, March 2010

TABLE 1. Composition of diets ot breeding Red-shouldered Hawks as reported in 12 studies across the eastern United
States. Methods of data collection included direct observation (DO) and pellet and/or prey remains examination (PR).
Shading indicates significantly similar groups identified though cluster analyses. Adapted from Dykstra et al. (2008).

Mammal Bird Reptile Amphib. Invert.

State

Year

# of nests

# Prey
items

#

%

#

%

#

%

#

%

#

%

Method

Source

lA

1977-1978

7

115

41

36

3

3

3

3

38

33

30

26

DO

Bednarz and
Dinsmore 1985

IN

1999-2(X)0

8

195

82

42

17

9

48

25

8

4

40

21

PR

Gehring 2003

MA

1974

4

46

33

72

2

4

2

4

9

20

0

0

DO

Portnoy and Dodge
1979

MD

1978-1979

17

30

23

77

1

3

2

7

4

13

0

0

DO

Janik and Moser
1982

MI

1942/1948

29

1,428

583

41

228

16

311

22

68

5

238

17

PR

Craighead and
Craighead 1956

NY

1939-1942

16

—

—

58

—

8

—

3

—

18

—

11

DO/PR

Ernst 1945“'’

OH

1997-2001

21

198

64

32

14

7

46

23

36

18

38

19

DO

Dykstra et al. 2003

WI

1973-1979

9

97

57

59

1

1

14

14

12

12

13

13

DO

Welch 1987

AR

2004-2005

20

1,139

69

6

4

0

171

15

523

46

372

33

DO

Townsend 2006

GA

1994

8

181

34

19

15

8

43

24

46

25

43

24

DO

Howell and

Chapman 1998

MO

1982-1983

3

—

—

17

—

0

—

26

—

46

—

1 1

DO

Parker 1986“

TX

2005-2006

10

1315

197

15

63

5

412

31

107

8

536

41

DO

Strobel 2007

^ Only reported the relative proportion.s of prey items to total diet.

Invertebrates approximate because of small unknown proportion (<3%) of crayfi.sh categorized as miscellaneous.

data within each study is unknown; however, we
assumed that natural and stochastic processes
(pair and territory fidelity, nest accessibility, etc.)
were similar among study populations, which
facilitated comparison among studies. These data,
in combination with our data, describe the diets of
three {B. 1. alleni, B. I. lineatus, B. 1. texanus) of
the five currently recognized subspecies of
RSHA.

The cluster analysis of the Pearson’s correlation
coefficient for the 12 study areas identified
several significant clusters {P < 0.05) with a
primary division creating two distinct groups
(Fig. 1 ). Our results indicate significant diver-
gences at several levels of correlation; however,
we were most interested in the dichotomous
nature of the results and focus further analyses
on the two most distinct groups. We arbitrarily
titled the groups. Group 1 was comprised of data
from studies in Indiana, Iowa. Maryland, Massa-
chu.setts, Michigan. New York, Ohio, and Wis-
consin. Group 2 was comprised of data collected
in Arkansas, Georgia, Missouri, and Texas.
Although grouped by similarity in breeding
.season diet, the latitude of each study site appears
to correlate with group membership with studies
in group 1 occurring in states of higher latitudes
than studies in group 2.

Analyses of those data from the northern and
southern groups indicated substantial differences
existed between diets in the two primary clusters
(^5,7 = 8.1, P < 0.01). Red-shouldered Hawk
diets in group 1 had more mammalian prey than
those in group 2 (group 1. .v ± SD = 53.2 ±
20.3%, group 2 = 17.9 ± 9.6%; t\\ = 3.8, P <
0.01, Fig. 2). Diets of RSHAs in group 2 had a
higher proportion of amphibians (group 1 = 1 1.2
± 6.2%, group 2 = 32.9 ± 20.3%; tu = -2.7, P
= 0.02) than those in group 1 (Fig. 2). Avian,
reptilian, and invertebrate prey items comprise
similar proportions of RSHA diets in both groups.
All studies using prey remains analysis were
assigned to Group 1, suggesting the suspected
biases previously discussed may have skewed
their results; however, the latitudinal pattern still
exists in direct observational studies (Table 1).

DISCUSSION

Our analysis indicated there are two groups of
Red-shouldered Hawks distinguishable by their
breeding sea.son diet. Breeding pairs in group 1
preyed more frequently on mammals while those
in group 2 preyed more frequently on amphibians.
In addition to diet similarities, there also appears
to be a coarse spatial relationship within groups.
Most studies conducted at northern latitudes

Strobe! and Baal - RED-SHOULDERED HAWK DIET

71

Study site/year

FIG. 1. Dendrogram illustrating similarity of diets of Red-shouldered Elawks reported by studies in 12 states across the
eastern USA. ait values indicate approximately unbiased significance level of each specified grouping {P ~ \-aii) based on
10,000 multiscale bootstrap iterations of the Pearson’s correlation coefficient topology matrix of an UPGMA cluster
analysis. Dashed boxes identify the primary significant (P ^ 0.01) groups.

(>38° N) were classified into group 1 and most
southern latitude studies into group 2. We
speculate the pattern between latitude and diet
composition is ultimately caused by climate,
vegetation, and precipitation patterns influencing
prey type availability differently across North
America. Results from Bednarz and Dinsmore’s
(1985) study in Iowa and Strobel’s (2007) in
Texas, support this hypothesis and show diets of
RSHAs were primarily mammalian prey during
drier conditions, but were primarily amphibian
and invertebrate prey under conditions with
abundant moisture. This may also explain why
Bednarz and Dinsmore’s (1985) and Strobel s
(2007) data were the furthest diverged from other
studies in their respective groups (Fig. 1).

The effects of latitude, temperature, and
moisture regimes on species’ distributions are
largely accepted (MacArthur 1972), and clearly
influence prey availability. These effects would
presumably influence productivity and survival of

RSHAs breeding in different areas of North
America. For example, nutritional value varies
widely among prey types and likely influences
reproductive rates of RSHAs breeding in different
regions of their range. Four studies in our analysis
reported estimates of prey biomass and estimated
mammalian prey items taken by RSHAs averaged
30% heavier than amphibian prey items (Howell
and Chapman 1998, Dykstra et al. 2003, Town-
send 2006, Strobel 2007). Strobel (2007) estimat-
ed that RSHAs breeding in south Texas deliver
-6,000 g of prey to a nest to fledge one nestling.
Common Kestrels {Falco tinniinculns, Wiehn and
Korpimaki 1997), European Starlings (Stiinnis
yiilgciris, Wright et al. 1998), and RSHAs (Strobel
2007) increased prey delivery rates to meet the
increased nutritional demands of larger broods or
older nestlings. Red-shouldered Hawks breeding
in areas with larger and more nutrient rich prey
may be able to rear larger broods, fledge higher
quality young, or allocate more time to activities

72

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. I. March 2010

FIG. 2. Composition of Red-shouldered Flawk diets among studies in two significantly different groups (Group I =
northern studies; Group 2 = .southern studies) identified by 10,000 multiscale bootstrap iterations of the Pearson’s
correlation coefficient. Spreads indicate ± 1 SD from the mean percent contributed by each of the five prey types.

such as nestling defense, and increase their
productivity. A post hoc examination of produc-
tivity rates (average number of nestlings produced
by successful nests) indicated little difference in
productivity of RSHAs in group 1 or group 2
(2.42 ± 0.51, 2.38 ± 0.67; respectively). Bednarz
and Dinsmore (1985) noted similar RSHA
productivity on their study site between years
when RSHAs had markedly different diets. This
suggests RSHAs have similar productivity despite
using different prey types. This also suggests
different prey types result in differences in
foraging effort, prey delivery rates, and energetic
demands upon adults.

Recognizing patterns in prey use by RSHAs
throughout North America may provide valuable
insight for conservation planning. Breeding pop-
ulations of RSHAs feeding primarily on mamma-
lian prey items may not be as greatly inHuenced
by wetland degradation as populations feeding
primarily on amphibians, but may be more
susceptible to fluctuations in forest harvest

regimes or natural fluctuations in small mammal
populations. It is unclear how factors like
bioaccumulation of toxins and juvenile survival
rates vary between populations with different
breeding season diets. Few studies have examined
regional specific survival patterns or lifetime
reproductive success in widespread raptor popu-
lations, both of which are likely influenced by diet
composition and prey availability. Evaluating the
importance of different types of prey for RSHAs
throughout North America would be greatly
enhanced if future studies provided accurate
estimates of local prey biomass and availability.

ACKNOWLEDGMENTS

We thank the Rob and Bes.sie Welder Wildlife
Foundation for indispensable financial and logistic support
as well as the U.S. Geological Survey, Texas Cooperative
Fish and Wildlife Research Unit and Texas Tech Univer-
sity. Additional funding was provided by the Houston
Safari Club. We appreciate Terry Blankenship’s assistance
with prey identification and Carey Haralson for field and
technical assistance. We also thank Jeff Rooke and the staff

Sirohcl (iihI Boa/ - RED-SI lOUlJ^ERED HAWK DIET

73

of the Twin Oaks Hunting Resort. The use of trade, product,
industry, or firm names or products is for informative
purposes only and does not constitute an endorsement by
the U.S. Government, U.S. Geological Survey, or Texas
Tech University. This is Welder Wildlife Foundation
Contribution Number 688.

LITERATURE CITED

Bednarz, J. C. and J. J. Dinsmore. 1981. Status, habitat
use, and management of Red-shouldered Hawks in
Iowa. Journal of Wildlife Management 45:236-241.
Bednarz, J. C. and J. J. Dinsmore. 1985. Flexible dietary
response and feeding ecology of the Red-shouldered
Hawk, Buteo lineatiis, in Iowa. Canadian Field-
Naturalist 99:262-264.

Bent, A. C. 1937. Life histories of American birds of prey.
U.S. National Museum Bulletin 167. Washington,
D.C., USA.

Bloom. P. H. 1989. Red-shouldered Hawk home range and
habitat use in southern California. Thesis. California
State University, Long Beach, USA.

Bloom, P. H., M. D. McCrary, and M. J. Gibson. 1993.
Red-shouldered Hawk home-range and habitat use in
southem California. Journal of Wildlife Management
57:258-265.

Brown. W. H. 1971. Winter population trends in the Red-
shouldered Hawk. American Birds 25:813-817.
Craighead, J. J. and F. C. Craighead Jr. 1956. Hawks,
owls and wildlife. Stackpole Company, Harrisburg,
Pennsylvania and Wildlife Management Institute,
Washington, D.C., USA.

Dykstra, C. R.. j. L. Hays, and S. T. Crocoll. 2008.
Red-shouldered Hawk {Buleo lineatiis). The birds of
North America Online (A. Poole, Editor). Cornell
Laboratory of Omithology, Ithaca, New York, USA.
http://bna.birds.cornell.edu/bna/species/l07doi: 10.
2173/bna. 107 (accessed 16 December 2007).

Dykstra, C. R., J. L. Hays, F. B. Daniel, and M. M.
Simon. 2001. Home range and habitat use of suburban
Red-shouldered Hawks in southwestern Ohio. Wilson
Bulletin 113:308-316.

Dykstra, C. R., J. L. Hays, M. M. Simon, and F. B.
Daniel. 2003. Behavior and prey of nesting Red-
shouldered Hawks in southwestern Ohio. Journal of
Raptor Research 37:177-187.

Ernst, S. G. 1945. The food of the Red-shouldered Hawk
in New York State. Auk 62:452-453.

Gehring, j. L. 2003. The ecology of Red-tailed Hawks and
Red-shouldered Hawks in forested landscapes and in
landscapes fragmented by agriculture. Dissertation.
Purdue University, West Lalayette, Indiana, USA.
Henny, C. j., F. C. Schmidt, E. M. Martin, and L. L.
Hood. 1973. Territorial behavior, pesticides, and the
population ecology of Red-shouldered Hawks in
central Maryland. 1943-1971. Ecology 54:545-554.
Howell, D. L. and B. R. Chapman. 1998. Prey brought to
Red-shouldered Hawk nests in the Georgia piedmont.
Journal of Raptor Research 32:257-260.

Jacobs, J. and E. J. Jacobs. 2002. Con.servation assess-
ment for Red-shouldered Hawk (Buteo hneatus) in

national forests of the North Central .States. U.SDA,
Forest Service, Eastern Region, Milwaukee, Wiscon-
sin, USA.

Janik. C. a. and j. a. Mosher. 1982. Breeding biology of
raptors in the central Appalachians. Journal ol Raptor
Re.search 1 6: 1 8-24.

Krebs, C. J. 1999. Ecological methodology. Second
Edition. Addison-Wesley Educational Publishers Inc.,
New York, USA.

MacArthur, R. H. 1972. Geographical ecology: patterns
in the distribution of species. Princeton University
Press, Princeton, New Jersey, USA.

Marti, C. D., M. Bechard, and F. M. Jasik. 2007. Food
habits. Pages 129-149 in Raptor research and
management techniques (D. L. Bird and K. L.
Bildstein. Editors). Hancock House Publishers, Blaine,
Washington, USA.

McLeod, M. A. and D. E. Andersen. 1998. Red-
shouldered Hawk broadcast surveys: factors affecting
detection of responses and population trends. Journal
of Wildlife Management 62:1385-1397.

Parker, M. A. 1986. The foraging behavior and habitat use
of breeding Red-shouldered Hawks Buteo Hneatus in
southeastern Missouri. Thesis. University of Missouri,
Columbia, USA.

Peterson, J. M. C. and S. T. Crocoll. 1992. Red-
shouldered Hawk, Buteo Hneatus. Pages 333-351 in
Migratory nongame birds of management concern in
the Northeast (K. J. Schneider and D. M. Pence,
Editors). USDl, Fish and Wildlife Service, Newton
Corner, Massachusetts, USA.

Portnoy, J. W. and W. E. Dodge. 1979. Red-shouldered
Hawk nesting ecology and behavior. Wilson Bulletin
91:104-1 17.

R Development Core Team. 2008. R: a language and
environment for statistical computing. R Foundation
for Statistical Computing. Vienna, Austria.

Redpath, S. M., R. Clarke. M. Madders, and S. J.
Thirgood. 2001. Assessing raptor diet: comparing
pellets, prey remains, and observational data at Hen
Harrier nests. Condor 103:184-188.

Rogers, A. S., S. Destefano, and M. F. Ingraldi. 2005.
Quantifying Northern Goshawk diets using remote
cameras and observations from blinds. Journal of
Raptor Research 39:303-309.

Rottenborn, S. C. 2000. Nest-site selection and repro-
ductive success of urban Red-shouldered Hawks in
central California. Journal of Raptor Research 34:18-
25.

Sauer. J. R., J. E. Hines, and J. Fallon. 2008. The
North American Breeding Bird Survey, results and
analysis 1966—2007. Version 5.15.2008. USGS. Pa-
tuxent Wildlife Research Center. Laurel. Maryland.
USA.

.Strobel. B. N. 2007. Nest site selection and nestling diet of
the Texas Red-shouldered Hawk Buteo Hneatus
te.xanus in south Texas. Thesis. Texas Tech University.
Lubbock. USA.

Suzuki, R. and H. Shimodaira. 2006. PVCLUST:
hierarchical clustering with P-values via multiscale
bootstrap resampling. R package Version 1.2-0. http://

74

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. I. March 2010

www.is.titech.ac.jp/~shimo/prog/pvclust/ (accessed
16 December 2008).

Townsend, K. A. L. 2006. Nesting ecology and sibling
behavior of Red-shouldered Hawks at the St. Francis
Sunken Lands Wildlife Management Area in north-
eastern Arkansas. Thesis. Arkansas State University,
State University, USA.

WiEHN, J. AND E. Korpimaki. 1997. Food limitation on

brood size: experimental evidence in Eurasian Kestrel.
Ecology 78:2043-2050.

Welch, R. J. 1987. Food habits of the Red-shouldered
Hawk in Wisconsin. Pas.senger Pigeon 49:81-91.
Wright, J., C. Both, P. A. Cotton, and D. Bryant. 1998.
Quality vs. quantity: energetic and nutritional trade-
offs in parental provisioning strategies. Journal of
Animal Ecology 67:620-634.

The Wilson Joiinnil oj Oniitholoi’y 1 22( 1 ):75 -81 , 2010

EFFECTS OF BROWN-HEADED COWBIRD PARASITISM ON
PROVISIONING RATES OF SWAINSON’S WARBLERS

SARA PAPPAS,' THOMAS J. BENSON,' -^ AND JAMES C. BEDNARZ'

ABSTRACT. — We studied the effects of brood parasitism by Brown-headed Cowbirds (Mololhrus ater) and other
factors on food provisioning rates of Swainson’s Warblers (Lunnothlypis swainsoiiii), a secretive and poorly understood
species of conservation concern. We used time-lapse video systems to collect provisioning data at 25 nests, nine of which
were parasitized by Brown-headed Cowbirds. We found strong relationships between feeding rate and brood size with
increases from 2.0 feeding visits/hr for nests with a single Swainson’s Warbler nestling to 3.0/hr for broods of four. We also
found an effect of nestling age with 1.9 visits/hr early and 3.2 visits/hr late in the nestling period. The relationships between
cowbird parasitism and provisioning were complex. Nests with brood size of one or three that contained a single cowbird
nestling had greater provisioning rates (2.5 and 3.3 visits/hr, respectively) than non-parasitized nests (2.0 and 2.6 visits/hr,
respectively). Nests with two cowbirds and no Swainson's Warbler young had greater provisioning rates (3.4 visits/hr) than
those with two warbler young (2.4 visits/hr), or one warbler and one cowbird (2.6 visits/hr). The increase m provisioning
rate with nestling age was more pronounced with two cowbird nestlings present than in nests with zero or one cowbird
nestling. These results suggest that parasitized nests, especially those with multiple cowbird nestlings, can impose greatei
energetic demands on parents. Food limitation may constrain the ability of Swainson s Warblers to adequately care for both
their own nestlings and cowbird young. Received 10 April 2009. Accepted 2 September 2009.

Swainson’s Warbler {Limnothlypis swainsonii)
is a medium-sized wood warbler that occurs in
dense understory vegetation in bottomland hard-
wood forests of the southeastern United States.
Nondescript olive-brown coloration and secretive
behavior make it elusive and difficult to locate
and study. Swainson’s Warblers have specialized
breeding habitat requirements, preferring forest
stands with dense understory vegetation, often of
giant cane {Arunclinaria gigantea) (Brewster
1885, Meanley 1971, Brown et al. 2009).
Conversion of higher elevation bottomland hard-
wood forests and loss of canebrakes has reduced
the availability of breeding habitat for Swainson’s
Warblers (Brown et al. 2009), likely contributing
to the status of this species as among those of
greatest conservation concern nationally (Hunter
et al. 1993, Rich et al. 2004).

Fragmentation of bottomland hardwood forests
has exposed some Swainson’s Warbler popula-
tions to high frequency of brood parasitism by
Brown-headed Cowbirds (Mololhrus ater) (Ben-
son et al. 2010a). Thirty-six percent of Swainson’s
Warbler nests in eastern Arkansas were parasit-
ized (/? = 135 nests), and 10% received multiple
cowbird eggs (Benson et al. 2010a). Brood
parasitism may be a major contributing factor

' Department of Biological Sciences, Environmental
Sciences Program, Arkansas State University, P. O. Box
599, Jonesboro, AR 72467. USA.

-Current address; Illinois Natural Flistory Survey, 1816
South Oak Street, Champaign, IL 61820, USA.

■'Corresponding author; e-mail: tjbenson@gmail.com

for population declines of some songbirds
(Schmidt and Whelan 1999). Cowbirds reduce
the productivity of host species by removing or
damaging eggs, and through competition between
host and cowbird young (Trine et al. 1998,
Hoover 2003). This competition may lead to
starvation of host nestlings in parasitized nests,
especially for smaller host species such as
Swainson’s Warblers (Benson et al. 2010a).
Parasitism of smaller hosts may also reduce the
growth rate of surviving nestlings (Dearborn et al.
1998, Burhans et al. 2000), leading to higher post-
fledgling mortality. Parasitism of Swainson’s
Warblers in eastern Arkansas decreased produc-
tivity of successful nests from 2.75 in non-
parasitized to 0.60 Swainson’s Warbler young in
parasitized nests (Benson et al. 2010a); parasitized
nests with multiple Brown-headed Cowbird young
produced no Swainson’s Warbler young (Benson
2008).

Beyond the direct effects of parasitism, in-
creased food provisioning resulting from parasit-
ism also may have harmful effects on adults.
Presence of cowbird young in nests of relatively
small hosts may lead to increased prey delivery
rates by the host parents (Hoover and Reetz 2006)
and increase food demand from adults (Kilpatrick
2002). This increased parental investment may
lower future reproductive investment (Gustafsson
and Sutherland 1988, Linden and Mpller 1989).
Increased activity at the nest may attract predators
and lead to decreased survival of parasitized nests
(Haskell 1994, Leech and Leonard 1997, Dear-

75

76

THE WILSON JOURNAL OL ORNITHOLOGY • Vul. 122, No. I. March 2010

born et al. 1998, Dearborn 1999, Tewksbury et al.
2002). The stress imposed on the host parents by
increased provisioning rates may ultimately lead
to lower survival or increased emigration rates of
individuals with parasitized nests (Hoover and
Reetz 2006).

We investigated food provisioning rates at
parasitized and non-parasitized nests of Swain-
son’s Warblers to understand why brood parasit-
ism leads to decreased productivity and potential
implications for con.servation. We expected the
food-delivery rate of adults to increase with
parasitism due to the large size of young cowbirds
relative to Swainson’s Warblers. We also predict-
ed multiple parasitism to have a greater impact
than single parasitism, and for larger and older
broods to receive more food deliveries than
smaller and younger broods.

METHODS

Study Area. — We studied Swainson’s Warbler
breeding biology in 2006 and 2007 at White River
National Wildlife Refuge (Brown et al. 2009).
This >60,000-ha refuge is among the largest
remaining areas of bottomland hardwood forest in
the Mississippi Alluvial Valley (Twedt and
Loesch 1999). Areas with Swainson’s Warblers
at White River National Wildlife Refuge were
dominated by sugarberry {Celtis laevigata), sweet-
gum {Liciuidamhar styraciflua), box elder {Acer
negiindo), elm (Ulmiis spp.), oak (Qiiercus spp.),
sycamore (Plataniis occidentalis), and hickory
(Carya spp.). Dominant plants in the forest
understory included greenbrier {Srnila.x spp.),
Virginia creeper (Parthenocissus quinquefolia),
peppervine (Ampelopsis arhorea), grape (Vitis
spp.), spicebush (Lindera benzoin), box elder,
and dense thickets of giant cane.

Nest Searching and Video Monitoring. — We
.searched for Swainson’s Warbler nests from late
April to early August systematically by searching
known territories with one to six observers, and
opportunistically while engaged in other activities
(Benson et al. 2()l()a). We recorded GPS (Global
Positioning System) coordinates once nests were
located, and returned at I- to 4-day intervals to
ascertain the status and fate of each nest (Benson
et al. 2()l()a). We installed time-lapse video
systems at a subset of nests as part ol an effort
to identify nest predators for this species (Benson
et al. 20 1 Ob). These systems were a small color
video camera (models PC5()61R and PCI68-IR,
Supercircuits, Austin, TX, USA) mounted on 2.2-

cm diameter wooden dowels; the camera was
connected to a time-lapse VCR with two deep-
cycle batteries >10 m from the nests in plastic
weatheiproof containers. Camera systems were
camouflaged to blend in with suiToundings; the
presence of cameras appeared to not affect nest
survival (Benson et al. 2010b). We returned to
video-monitored nests at 1- to 4-day intervals to
change batteries and tapes until nests failed or
fledged young.

Video Processing. — We collected data on food
provisioning rates of adult Swainson’s Warblers at
25 nests, nine of which were parasitized by
Brown-headed Cowbirds. We recorded the time of
each feeding event for each nest.

Data Analyses. — We assessed the effect of
Brown-headed Cowbird parasitism on Swainson’s
Warbler provisioning rates (feedings/hr) using
repeated-measures ANOVA (SAS PROC MIXED)
(Littell et al. 2006). We examined differences
among nests containing no, one, and two cowbird
nestlings to document the effect of parasitism.
Other factors, including brood size and age of
nestlings, may affect provisioning rates, and we
tested for the influence of brood size, nest period
(1^, 5-8, and 9-12 days), and interactions among
parasitism and these two factors. We set signifi-
cance at a = 0.05, and followed significant main
effects or interactions with pair-wise contrasts. We
considered nest, hour within nest, and day within
nest as random effects to account for potential non-
independence of observations, and used the Ken-
ward-Roger approximation for denominator de-
grees of freedom (Littell et al. 2006). We
incoiporated brood size as a time-specific factor;
con.sequently, nests were often observed with
multiple brood sizes and fit into more than one
category. We inspected the distribution of residuals
and plots of residual versus predicted values to
confirm our data conformed to the assumptions of
normality and homogeneity of variances.

RESULTS

The mean brood size for the 16 non-parasitized
nests was 2.86 (SE = 0.22; range = 1^), and the
mean brood size for parasitized nests (including
cowbirds) was 2.67 (SE = 0.33; range = 1^).
The parasitized nests consisted of seven with one
cowbird nestling and two with two cowbird
nestlings.

We recorded 6,320 feeding visits (.v = 252.8/
nest, SE = 24.1, range = 28-465) between 17
May and 25 July 2006 and 2007, and data were

Pappas cl al. • SWAINSON’S WARBLER PROVISIONING

77

TABLE 1. Repeated-measures ANOVA examining the
influence of number of Brown-headed Cowbird (BHCO)
nestlings, brood size, nest period (1-4, 5-8, and 9-12 days
= nestling age), and interactions among these factors on
Swainson's Warbler provisioning rates at White River
National Wildlife Refuge, Arkansas, 2006 and 2007.

Effect

df

F

p

BHCO

2, 34

4.08

0.026

Brood size

3, 307

9.89

<0.001

Nest period

2, 291

35.92

<0.001

BHCO X brood size

5, 276

2.53

0.029

BHCO X nest period

4, 235

5.36

<0.001

Brood size X nest period
BHCO X brood size X nest

6, 326

0.72

0.64

period

5, 297

0.64

0.67

collected from 2 to 12 days for each nest (.v = 8.4,
SE = 0.6) representing 2,495 total nest hours {x =
99.8, SE = 7.7, range = 18-160). There was no
significant three-way interaction among number

of cowbird nestlings, brood size, and ne.st period,
nor was there a significant interaction among
brood size and nest period (Table 1 ). There were
significant interactions between number of cow-
bird nestlings and brood size, and number of
cowbird nestlings and nest period (Table 1).

Provisioning rate increased with brood size for
non-parasitized nests (r > 1.88, P ^ 0.06; Fig. 1).
The provisioning rate for nests with one cowbird
nestling was similar to that for nests with a
cowbird and a warbler nestling {t = —0.49, P =
0.63), but the rates for both nest types differed
from nests with a single cowbird and two or three
warbler nestlings {t 2: 3.09, P ^ 0.002). Nests
with two cowbird nestlings and zero or one
warbler nestling had similar provisioning rates {t
= 1.27, P = 0.21). There was a marginally
significant difference in provisioning rate between
nests with a single warbler and a single cowbird
nestling (r = -1.83, P = 0.07), and a significant
difference between nests with three warbler

4.0 -

3.5 -

•& 3.0

c

"a

CD

CD

2.5

2.0 -

•

One nestling

o

Two nestlings

T

Three nestlings

A

Four nestlings

(5)

I

I

(10)

~r

(12)1

(13)X

(4)

(2)

II

(6)

(5)

(2)

io)

1.5

~r"

0

Number of cowbird nestlings

FIG. I . Effects of Brown-headed Cowbird parasitism on provisioning rale (teedmgs/hr ± SE) at Swainson s Warbler
nests for brood sizes of one through four (including cowbird young). Data were collected from I6 non-parasitized and nine
parasitized nests at White River National Wildlife Refuge, Arkansas, 2006 and 2007. The sample size of nests for each
category is in parentheses; most nests were observed at multiple states and do not equal sums of the total number of
nests observed.

78

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. I. March 2010

FIG. 2. Effects of Brown-headed Cowbird parasitism on provisioning rate (feedings/hr ± SE) at Swainson’s Warbler
nests during three different periods of the nestling stage. Values were calculated at the mean for brood size. Data were
collected from 16 non-parasitized and nine parasitized nests at White River National Wildlife Refuge, Arkansas, 2006 and
2007. The sample size of nests for each category is in parentheses; most nests were observed at multiple states and do not
equal sums of the total number of nests observed.

nestlings and those with two warblers and a
cowbird nestling (/ = —2.76, P = 0.008). There
were no differences between nests with two
warblers and tho.se with a warbler and a cowbird
nestling (t = —0.87, P = 0.39), or those with four
warbler nestlings and those with three warblers
and a cowbird nestling (t = —0.35, P = 0.73).
Nests with two warbler nestlings or a warbler and
a cowbird nestling had lower provisioning rates
than those with two cowbird nestlings (/ = -3.75,
P < 0.001, and t — —2.87, P = 0.007,
respectively). Nests with three warbler nestlings
or two warblers and a cowbird nestling had
similar provisioning rates to a nest with one
warbler and two cowbird nestlings it = — 1.19,

= 0.24, and t = 0.60, P = 0.55, respectively).

The provisioning rate in nests that were not
parasitized increased with nestling age (/ ^ 6.65,
P < 0.001; Fig. 2). There was a difference in
provisioning rates between days 1 to 4 and 5 to 8
for nests with one cowbird nestling, (f = — 1 .98, P

= 0.050), but no other combinations of two
periods (/ < 0.92, P > 0.36). A nest with two
cowbird nestlings had a greater provisioning rate
from days 5 to 12 than for days 1 through 4 (r >
5.56, P < 0.001 ), but rates for days 5 to 8 and 9 to
12 were similar (/ = — 1.30, P = 0.19). Nests with
a single cowbird nestling during the early nestling
period had greater provisioning rates than non-
parasitized nests {t = 2.88, P = 0.007), but there
were no differences between rates for nests with
two versus no cowbirds {t = 1.02, P — 0.32), or
for nests with one versus two cowbird nestlings
during this stage (/ = 0.69, P = 0.49). The
frequency of parental feedings from days 5
through 8 was similar for non-parasitized nests
and those with a single cowbird nestling (/ =
-1.34, P = 0.19), but both differed from a nest
with two cowbirds {t — —5.12, P < 0.001, and t
= —4.17, P < 0.001, respectively). The rate of
provisioning during the late nestling period was
marginally greater in non-parasitized when com-

Pappas ct al. • SWAINSON’S WARBLER PROVISIONING

79

pared to single-parasitized nests {t = 1.85, P =
0.07), but both differed from a double-parasitized
nest (t = -2.97, P = 0.004, and t ^ -3.65, P <
0.001, respeetively).

DISCUSSION

The effect of cowbird parasitism on the
provisioning rate of Swainson’s Warblers was
variable and depended on brood size and age of
nestlings. Our observed increases in provisioning
rates with brood size for non-parasitized nests
were consistent with trends observed for other
species (e.g., Johnson and Best 1982, Haggerty
1992, Rosa and Muiphy 1994), as were the results
for increased provisioning to older nestlings (e.g.,
Johnson and Best 1982, Knapton 1984, Goodbred
and Holmes 1996). These results were expected
given the increasing energetic demand of each
nestling as they grow (Williams and Prints 1986).

There were effects of parasitism on provision-
ing rates at nests with one to three nestlings, and
for all three nestling-age categories. Differences
were generally more pronounced for nests with
two cowbird young. However, these results must
be viewed with caution due to small sample sizes.
We collected provisioning data from only two
multiply parasitized nests, only one of those nests
also contained a warbler nestling, and only one of
the nests survived into the middle and late nestling
periods. Similar effects of single and multiple
parasitism have been observed for other species
(Dearborn et al. 1998, Hoover and Reetz 2006).
However, there appeared to be a limit to the
provisioning rate for Swainson’s Warblers. The
mean provisioning rate of non-parasitized nests
with four nestlings was similar to the rates for
single-parasitized nests with three or four total
nestlings (4 warblers vs. 2 warblers and 1
cowbird: /6o — 1.08, P = 0.28) and double-
parasitized nests with two or three total nestlings
(t3o = 1.40, P = 0.17; t57 = 0.08, P -= 0.94,
respectively). This similarity of mean rates among
these brood sizes, given the larger size and
begging intensity of cowbird nestlings (Dearborn
1998), suggests food may be a limiting factor (the
mass of a cowbird at fledging is approximately
twice that of an adult Swainson’s Warbler;
Lowther 1993, Brown and Dickson 1994, Hoover
2003).

The maximum observed clutch size for Swain-
son’s Warblers at our study area was tour eggs
(Benson et al. 2010a). Our provisioning data
support the hypothesis that this maximum is at

least partly affected by food limitation (Lack
1947, 1954). The difference in productivity for
Swainson’s Warblers between successful non-
parasitized (2.75 young/successful nest) and
parasitized nests (0.60 young/successtul nest) is
among the more-extreme reductions in similar-
sized passerines (Payne 1998). This reduction for
Swainson’s Warblers is mostly due to starvation
of host nestlings (Benson et al. 2010a). The
similar-sized Prothonotary Warbler {Protonotaria
citrea), which breeds in lower elevation bottom-
land hardwood forests, is often able to fledge
multiple cowbird and warbler young (Hoover
2003). The observed provisioning rates for
Prothonotary Warblers (.v = 9.54/hr for non-
parasitized, X — 16.96/hr for parasitized nests;
Hoover and Reetz 2006) far exceed the rates we
observed for Swainson’s Warblers (.v < 5/hr).
Further evidence of food limitation of Swainson’s
Warblers in our study area includes food abun-
dance which appears to influence habitat use
(Brown 2008). These birds have larger home
ranges than similar-sized passerines (,v = 9.4 ha;
Anich et al. 2009), suggesting they require large
areas to acquire necessary resources.

The increased sound and activity at parasitized
nests has been observed to increase nest predation
in some cases (Haskell 1994, Leech and Leonard
1997, Dearborn 1999, Tewksbui^ et al. 2002). Nest
predation rates in our study population appeared to
be unaffected by brood size or parasitism, possibly
because predation rates were already relatively
high (Benson et al. 2010a, b). However, the
complex relationships among brood size, nestling
age, parasitism status, and provisioning rate may
lead to increased predation risk only in limited
circumstances and not necessarily as a simple
function of brood size or parasitism status.

We observed effects of parasitism on prey
delivery rates by Swainson’s Warblers, but
several questions remain. For example, the size
or type of prey that adults bring to the nest may
differ among brood sizes or with nestling age in
some species (Biermann and Sealy 1982), and the
relative contributions of males and females to
provisioning young may also vary (Nolan 1978,
Carey 1990); these issues merit further study. The
inlluence of habitat quality on provisioning rate
and interaction between habitat quality and brood
parasitism effects also remain unknown.

Habitat loss and fragmentation present signif-
icant conservation challenges for Swainson’s
Warbler populations. This fragmentation may

80

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 1. March 2010

lead to low productivity due to high predation and
brood parasitism rates in some populations
(Benson et al. 2010a). Efforts should be made to
conserve and improve existing habitats by pro-
viding forested buffers between quality Swain-
son’s Warbler habitat and agricultural areas to
reduce the negative effects of cowbird parasitism
in this and other populations. Cowbird removal
programs may be beneficial in limited circum-
stances, but are not cost-effective, or a permanent
solution, and should be of secondary importance
to habitat restoration and landscape-level man-
agement (Stutchbury 1997, Rothstein and Peer
2005).

ACKNOWLEDGMENTS

Funding was provided by the Arkansas Game and Fish
Commission and the U.S. Fish and Wildlife Service
(USFWS) through a State Wildlife Grant and a separate
cost-share grant from the USFWS. We are grateful to
Catherine Rideout, Laurel Barnhill, Richard Hines, W. C.
Hunter, and Steve Reagan for their support of this research.
We thank the National Science Foundation which provided
funding for Sara Pappas through the Research Experiences
for Undergraduate (REU) program (NSF-DBI 0552608
Dowling) and Carolyn Dowling, the head of Arkansas State
University’s REU program. Sara Pappas thanks Daniel
Albrecht-Mallinger for his support. We are especially
grateful to Jeremy Brown, Eric Huskinson, Nick Anich,
Carolina Roa, Jason Sardell, Mary-Beth Albrechtsen,
Christy McCarroll, and Jessica O’Connell for providing
valuable field a.ssistance and Nick Anich, C. E. Braun, and
two anonymous reviewers for providing comments on the
manuscript.

LITERATURE CITED

Anich, N. M., T. J. Benson, and J. C. Bednarz. 2009.
Estimating territory and home-range sizes: do singing
locations alone provide an accurate estimate of space
use? Auk 126:626-634.

Benson, T. J. 2008. Habitat use and demography of
Swainson’s Warblers in eastern Arkansas. Disserta-
tion. Arkansas State University, Jonesboro, USA.
Benson, T. J., J. D. Brown, and J. C. Bednarz. 2010b.
Identifying predators clarifies predictors of nest
success in a temperate passerine. Journal of Animal
Ecology 79:225-234.

Benson. T. J., N. M. Anich, J. D. Brown, and J. C.
Bednarz. 2010a. Habitat and landscape effects on
brood parasitism, nest survival, and fledgling produc-
tion in Swain.son's Warblers. Journal of Wildlife
Management 74:81-93.

Biermann. G. C. and S. G. Sealy. 1982. Parental feeding
of ne.stling Yellow Warblers in relation to brood size
and prey availability. Auk 99:332-341.

BREW.STRR, W. 1885. Swainson’s Warbler. Auk 2:65-80.
Brown, J. D. 2008. Arthropod communities and habitat
characteristics of sites occupied by Swain.son’s

Warblers in the White River National Wildlife Refuge,
Arkansas. Thesis. Arkansas State University, Jones-
boro, USA.

Brown, J. D., T. J. Benson, and J. C. Bednarz. 2009.
Vegetation characteristics of habitat occupied by
Swainson’s Warblers at the White River National
Wildlife Refuge, Arkansas. Wetlands 29:586-597.

Brown, R. E. and J. G. Dickson. 1994. Swainson’s
Warbler {Limnothlypis swainsonii). The birds of North
America. Number 126.

Burhans, D. E., F. R. Thompson III, and J. Faaborg.
2000. Costs of parasitism incurred by two songbird
species and their quality as cowbird hosts. Condor
102:364-373.

Carey, M. 1990. Effects of brood size and nestling age on
parental care by male Field Sparrows {Spizella
pusilla). Auk 107:580-586.

Dearborn, D. C. 1998. Begging behavior and food
acquisition by Brown-headed Cowbird nestlings.
Behavioral Ecology and Sociobiology 43:259-270.

Dearborn, D. C. 1999. Brown-headed Cowbird nestling
vocalizations and risk of nest predation. Auk 1 1 6:448-
457.

Dearborn, D, C., A. D. Anders, F. R. Thompson III, and
J. Faaborg. 1998. Effects of cowbird parasitism on
parental provisioning and nestling food acquisition and
growth. Condor 100:326-334.

Goodbred, C. and R. T. Holmes. 1996. Factors affecting
food provisioning of nestling Black-throated Blue
Warblers. Wilson Bulletin 108:467^79.

Gustafsson, L. and W. j. Sutherland. 1988. The costs of
reproduction in the Collared Flycatcher Ficediila
alhicollis. Nature 335:813-815.

Haggerty, T. M. 1992. Effects of nestling age and brood
size on nestling care in Bachman’s Sparrow {Aimo-
phila aestivalis). American Midland Naturalist
128:1 15-125.

Haskell, D. 1994 Experimental evidence that nestling
begging behaviour incurs a cost due to nest predation.
Proceedings of the Royal Society of London, Series B
257:161-164.

Hoover, J. P. 2003. Multiple effects of brood parasitism
reduce the reproductive success of Prothonotary
Warblers, Protonotaria citrea. Animal Behaviour
65:923-934

Hoover, J. P. and M. J. Reetz. 2006. Brood parasitism
increases provisioning rate, and reduces offspring
recruitment and adult return rates, in a cowbird host.
Oecologia 149:165-173.

Hunter, W. C., D. N. Pashley, and R. E. F. E.scano.
1993. Neotropical migratory landbird species and their
habitat of .special concern within the Southeast Region.
Pages 159-171 in Status and management of neotrop-
ical migratory birds (D. M. Finch and S. W. Stangel,
Editors). USDA, Forest Service, General Technical
Report RM-229. Rocky Mountain Research Station,
Fort Collins, Colorado, USA.

Johnson, E. J. and L. B. Best. 1982. Factors affecting
feeding and brooding of Gray Catbird nestlings. Auk
99:148-156.

Kilpatrick, M. A. 2002. Variation in growth of Brown-

Pappas cl al. • SWAINSON’S WARBLER PROVISIONING

81

headed Cowbird (Molothrus ater) nestlings and
energetic impacts on their host parents. Canadian
Journal of Zoology 80; 145-153.

Knapton, R. W. 1984. Parental feeding of nestling Nashville
Warblers: the effects of food type, brood-size, nestling
age, and time of day. Wilson Bulletin 96:594-602.
Lack, D. 1947. The significance of clutch size. Ibis
89:302-352.

Lack, D. 1954. The natural regulation of animal numbers.

Clarendon Press, Oxford, United Kingdom.

Leech, S. M. and M. L. Leonard. 1997. Begging and the
risk of predation in nestling birds. Behavioral Ecology
8:644-646.

Linden, M. and A. P. Moller. 1989. Cost of reproduction
and covariation of life history traits in birds. Trends in
Ecology and Evolution 4:367-371.

Littell, R. C., G. a. Milliken, W. W. Stroup, R. D.
WOLFINGER, AND O. SCHABENBERGER. 2006. SAS for
mixed models. SAS Institute Inc., Cary, North
Carolina, USA.

Lowther, P. E. 1993. Brown-headed Cowbird {Molothrus
ater). The birds of North America. Number 47.
Meanley, B. 1971. Natural history of the Swainson’s
Warbler. North American Fauna Number 69. U.S.
Department of the Interior, Washington, D.C., USA.
Nolan Jr., V. 1978. The ecology and behavior of the
Prairie Warbler Dendroica discolor. Ornithological
Monographs 26:1-595.

Payne, R. B. 1998. Brood parasitism in birds; strangers in
the nest. BioScience 48:377-386.

Rich, T. D., C. J. Beardmore, H. Berlanga, P. J.
Blancher, M. S. W. Bradstreet, G. S. Butcher, D.
W. Demarest, E. H. Dunn, W. C. Hunter, E. E.
Inigo-Elias, j. a. Kennedy, A. M. Martell, A. O.
Panjabi, D. N. Pashley, K. V. Rosenberg, C. M.
Rustay, j. S. Wendt, and T. C. Will. 2004. Partners

in Flight North American landbird conservation plan.
Cornell Laboratory of Ornithology, Ithaca, New York,
USA.

Rosa, S. M. and M. T. Murphy. 1994. Trade-offs and
constraints on Eastern Kingbird parental care. Wilson
Bulletin 106:668-678.

Rothstein, S. I. AND B. D. PEER. 2005. Conservation
solutions for threatened and endangered cowbird
{Molothrus spp.) hosts: separating fact from fiction.
Ornithological Monographs 57:98-114.

Schmidt, K. A. and C. J. Whelan. 1999. The relative
impacts of nest predation and brood parasitism on
seasonal fecundity in songbirds. Conservation Biology
13:46-57.

Stutchbury, B. j. M. 1997. Effects of female cowbird
removal on reproductive success of Hooded Warblers.
Wilson Bulletin 109:74-81.

Tewksbury, J. J., T. E. Martin, S. J. Hejl, M. J. Kuehn,
AND J. W. Jenkins. 2002. Parental care of a cowbird
host: caught between the costs of egg-removal and nest
predation. Proceedings of the Royal Society of
London, Series B 269:423^29.

Trine, C. L., W. D. Robinson, and S. K. Robinson. 1998.
Consequences of Brown-headed Cowbird brood par-
asitism for host population dynamics. Pages 273-295
in Parasitic birds and their hosts (S. I. Rothstein. and S.
K. Robinson, Editors). Oxford University Press, New
York, USA.

Twedt, D. j. and C. R. Loesch. 1999. Forest area and
distribution in the Mississippi Alluvial Valley: impli-
cations for breeding bird conservation. Journal of
Biogeography 26:1215-1224.

Williams, J. B. and A. Prints. 1986. Energetics of growth
in nestling Savannah Sparrows: a comparison of
doubly labeled water and laboratory estimates. Condor
88:74-83.

The Wilson Journal of Ornithology 122( 1 ):82-87, 2010

EFFECTS OF FAT AND FEAN BODY MASS ON MIGRATORY
FANDBIRD STOPOVER DURATION

CHAD L. SEEWAGEN' 2 ' AND CHRISTOPHER G. GUGLIELMO'

ABSTRACT. — We used quantitative magnetic resonance body composition analysis and radiotelemetry to examine
whether fat and lean body mass affected stopover durations of 1 1 birds captured during autumn migration in New York City,
USA. Two Swainson’s Thrushes (Catharus ustulatus), two Hermit Thaishes (C. giittatus), and seven Ovenbirds {Seiurus
aurocapilla) were used in the study. Ovenbird stopover duration was significantly and negatively related to fat mass but
unrelated to lean body mass. The same relationships were found when data from all three species were combined to increase
sample sizes. Birds that departed within I day had fat stores upon capture that represented at least 1 1 % of their total body mass
whereas those with fat content <6% of total body mass remained for no fewer than 4 days. Arrival fat mass clearly influenced
time birds spent at the site but lean body mass did not. Conditions for increasing or maintaining fat stores provided by urban
stopover sites may affect the migration timing of birds. Received 26 May 2009. Accepted 16 September 2009.

Migrating landbirds require several stopovers
for rest and refueling, as most are incapable of
completing their migration in a single flight. Time
and energy spent during stopover periods greatly
exceeds that spent aloft (Wikelski et al. 2003,
Bowlin et al. 2005), and behavior of birds at
stopover sites can greatly influence their overall
migration success. Factors that govern a bird’s
decision to terminate a stopover and begin the
next flight have received much attention, partic-
ularly through development of theoretical models
of optimal migration strategies (Alerstam and
Hedenstrdm 1998, Houston 1998). There is
evidence that fuel stores, refueling rate, distance
from final destination, date, predation risk, and
weather conditions can individually, or in some
combination influence a bird’s length of stay at a
given stopover site (Wang and Moore 1997,
Akesson and Hedenstrom 2000, Danhardt and
Lindstrdm 2001, Dierschke and Delingat 2001,
Schaub et al. 2008).

Field studies of the effects of intrinsic and
extrinsic factors on stopover duration are ham-
pered by the difficulty of knowing when birds
arrive at, and depart from, the site of interest.
Several studies have measured stopover duration
using mark-recapture data, while assuming birds
were marked upon arrival and final recapture
occurred on their actual departure date (e.g..

' Department of Biology, University of Western Ontario,
1151 Richmond .Street North, London, ON N6A .5K7,
Canada.

^Department of Ornithology, Wildlife Con.servation
Society. 2300 Southern Boulevard, Bronx, NY 10460,
USA.

'Corresponding author; e-mail: cseewagcn@wc.s.org

Cherry 1982, Wang and Moore 1997, Morris and
Glasgow 2001). The latter assumption can be
especially tenuous, as migrants may learn to avoid
mist nets (Mac Arthur and MacArthur 1974) or
move beyond the coverage area within the
stopover site (Chernetsov and Mukhin 2006,
Tsvey et al. 2007). This method also relies on
condition and behavior of the small proportion of
birds that are recaptured to represent the majority,
when recaptures may be biased towards individ-
uals in poor condition that remain at the site for a
relatively long time (Guglielmo et al. 2005,
Morris et al. 2005, Hays 2008). More recently,
survival analyses and other probabilistic models
have been used to relax these assumptions and
improve accuracy of stopover duration estimates
ba.sed on mark-recapture data (Schaub et al. 2001,
2008), although large sample sizes are generally
needed (Schaub 2006). Radio tracking offers
another approach for measuring stopover duration
that does not rely on recaptures and allows
knowing departure date with near certainty
(Chernetsov and Mukhin 2006, Tsvey et al.
2007). Drawbacks of radiotelemetry include high
equipment costs, possible effects of transmitters
on bird behavior, and the assumption that birds
are captured upon arrival.

Examining the effect of energetic condition on
stopover duration is complicated by the need for
accurate, non-lethal measurements of body com-
position. Commonly used condition indices (e.g.,
body mass, size-corrected body mass, visual fat
scores) can potentially provide reliable estimates
of fat content (Conway et al. 1994, Spengler et al.
1995, Seewagen 2008). However, their accuracy
may be species-specific (Skagen et al. 1993,
Spengler et al. 1995, Seewagen 2008) or weak-

82

Secn'cigen ami Guglielmo • BODY COMPOSITION AND STOPOVER DURATION

83

ened by inter-observer variation (Krementz and
Pendleton 1990), and they do not provide
separate, direct measures of fat and lean body
mass. Quantitative magnetic resonance analysis
(QMR; Taicher et al. 2003, Tinsley et al. 2004) in
contrast provides accurate, objective, and direct
measures of the fat and lean body mass of small
birds (CGG, unpubl. data).

We attached radio transmitters on 1 1 migrant
songbirds at our study site in New York City
during autumn 2008 in a pilot study to test the
feasibility of radio tracking birds in an urban
setting. We used a QMR body composition
analyzer on all captured migrants for a concurrent
study of body composition dynamics during
stopover refueling. The stopover duration and
body composition data obtained by radiotelemetry
and QMR, respectively, afforded an additional
opportunity to examine relationships between
these variables. We coupled telemetry and QMR
data to examine if arrival fat and lean body mass
affected the stopover durations of these 1 1
migrants.

METHODS

Study Site and Animal Capture. — We captured
birds during autumn 2008 on the grounds of the
Bronx Zoo as part of an ongoing study of urban
stopover ecology of migrant songbirds. The Bronx
Zoo is a 107-ha park in Bronx County, New York,
USA. Our study site was a 4.9-ha fragment of
riparian forest on the eastern edge of the zoo that
does not contain exhibits and is not open to
visitors (40° 85' N, 73° 87' W). Red oak {Quer-
cus rubra), sweet gum (Lic/uidambar styraciflua),
swamp dogwood (Cornus foemina), and willows
(SaUx spp.) are the dominant tree species in the
area. Previous research has suggested this site
offers suitable refueling conditions for landbird
migrants (Seewagen and Slayton 2008; CLS and
CGG, unpubl. data).

Birds were captured in mist nets from sunrise
until -1200 hrs during 4 September to 22
October. Captured birds were marked with a
U.S. Geological Survey aluminum leg band,
measured (unflattened wing length to 1 mm),
and weighed on a digital balance to 0.1 g.
Approximately 75 pi of blood was collected by
brachial veinipuncture for a separate study.

Magnetic Resonance Analysis. — Conscious
birds were scanned within 45 min of capture and
banding in a QMR body composition analyzer
(Echo-MRI, Echo Medical Systems, Houston, TX,

USA) housed in a customized mobile laboratory
(Glendale Recreational Vehicles, Strathroy, ON,
Canada) at the study site. Birds were scanned in
duplicate on the “small bird’’ and “two-accumu-
lation” settings of the Echo-MRI software,
yielding measures of fat mass and wet lean body
mass to 0.001 g. Total scanning time was
—4 min/individual and the average coefficient of
variation for individual birds was 1 .1% for fat and
0.4% for wet lean mass. Raw QMR scan data
were transformed using calibration equations for
small birds developed with House Sparrows
(Passer domesticus) and Zebra Finches (Taenio-
pygia guttata) in a laboratory validation study
(CGG, unpubl. data). Validation indicated the
relative error for predictions of fat and wet lean
mass were ± 11% and ± 1.5%, respectively.

Radiotelemetry. — ^We radiomarked each Swain-
son’s Thrush (Catharus ustulatus). Hermit Thrush
(C. guttatus), and Ovenbird (Seiurus aurocapilla)
captured beginning 3 October until all 1 1
transmitters we had were deployed. These species
were used because they are relatively common
migrants in the area, body masses are amenable to
radiotelemetry, and they are the focus of concur-
rent studies in progress at the site. The 0.5-g
transmitters (A2415, Advanced Telemetry Sys-
tems, Isanti, MN, USA) were affixed after QMR
scanning with eyelash adhesive directly to a
cleared area of skin in the interscapular region
(adapted from Raim 1978). Birds were held in
bags for - 15 min to allow the adhesive to dry and
released within 50 m of where originally cap-
tured.

We could not be certain birds were captured
and marked upon arrival. Eight of 1 1 birds were
marked on days when migrant capture rates were
high (-double) relative to the 2 previous days at
this site and at two affiliated banding stations
elsewhere in New York City. This increased our
confidence these individuals were new amvals
and marked during their first day at the stopover
site (sensu Tsvey et al. 2007). However, we use
the term “minimum stopover duration” hereafter
because of this uncertainty.

We checked for presence/absence daily at
sunrise, noon, and sunset using a handheld Yagi
antenna and receiver (R45()0S, Advanced Telem-
etry Systems, Isanti, MN, USA) from within the
study site. We searched for birds from other points
throughout the zoo including roofs of two three-
story buildings if we initially failed to detect a
signal from within the study site. We assumed

84

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 1, March 2010

TABLE 1. Means ± SD and ranges of total body mass, fat mass, lean body mass, and minimum stopover durations of
three species of migratory landbirds radiotracked during autumn stopovers in New York City, USA.

Species

n

Total body mass (g)

Fat mass (g)

Lean body mass (g) Stopover duration (days)

Swainson’s Thrush

2

27.3, 30.1

1.700, 2.410

22.760,26.119 4,5

Hermit Thrush

2

30.2, 34.6

1.570, 4.000

27.103,29.089 2,7

Ovenbird

7

20.0 ± 1.3
18.3-21.7

2.461 ± 1.069
1.068-3.911

17.009 ± 0.795 4.3 ± 5.0

15.713-18.145 1-14

birds had left the area permanently and resumed
migration if they were undetected for at least 4
consecutive days. We considered a stopover of 1
day as the initial capture of a bird at the site in the
morning and departure any time that night.

Statistical Analyses. — ^We used standardized
residuals from linear regressions of lean body
mass and wing length to correct lean body mass
for body size variation. We could not adjust fat
mass in the same manner because fat mass was
not significantly related to wing length. We
instead divided fat mass by total body mass and
arcsine-root transformed the proportions (Zar
1999, Gotelli and Ellison 2004).

We used backwards selection multiple regres-
sion with adjusted fat and lean body mass as
predictor variables, and number of days of
stopover as the response variable to examine the
effect of body composition on stopover duration.
Collinearity did not occur, as fat and lean body
mass were not significantly correlated. We
performed the analyses on Ovenbird data alone
and on data from all three species combined.

Age and gender were not considered because
sample sizes were small and not all birds could be
confidently assigned to age or gender classes. All
variables met normality assumptions. We con-
ducted tests with SPSS 16.0 and interpreted
results as significant when P < 0.05. Mean (±
SD) values are reported.

RESULTS

We radiomarked two Swainson’s Thrushes,
two Hermit Thrushes, and seven Ovenbirds.
Swainson's Thrushes were marked on 2 and 6
October, and Hermit Thrushes were marked on
10 and 11 October. Four Ovenbirds were
marked on 3 October and three were marked
on 6 October.

Minimum stopover durations ranged Irom 1 to
14 days (Table 1). Three individuals departed
within 1 day and the remainder stayed for at least
2 days. Initial fat content ranged from 6 to 8% of

total body mass in Swainson’s Thrushes, from 5 to
12% in Hermit Thrushes, and from 6 to 18% in
Ovenbirds. An Ovenbird was recaptured on the
eighth day of its 14 day stopover; total body mass
(measured by digital balance) increased 4.2 g
(23% of original), and fat and lean body mass
(measured by QMR) increased 2.767 g (259% of
original) and 1.884 g (11% of original), respec-
tively.

The initial fat loads of birds that stayed at the
site for only 1-2 days (/? = 5) averaged 14 ± 3%
of total body mass, whereas the initial fat loads of
birds that stayed for the longest periods (7, 8, and
14 days; n = 3) averaged 6 ± 0.5% of total body
mass. Ovenbird stopover duration was unrelated
to lean body mass (r = 0.02, Fja = 0.36, P =
0.58), and significantly and negatively related to
fat mass {r = 0.74, F,,5 = 13.95, P = 0.013). The
same relationships were found when data from all
three species were combined to increase sample
sizes (lean: r = 0.03, F2,8 = 0.00, P = 0.98; fat:
r = 0.56, F,,9 =11.31, F = 0.008; Fig. 1). The
relationship between fat and stopover duration
appeared non-linear, and we explored the fit of an
exponential model to the data a posteriori. The
exponential model was also highly significant
(r = 0.71, F,.9 = 22.08, P = 0.001; Fig. 1).

DISCUSSION

This is the first combination of magnetic
resonance technology and radiotelemetry to
examine the influence of fat and lean body mass
on stopover duration to our knowledge. We
believe this is also the first description of autumn
stopover durations of migrants within a major
urban center.

Fat mass upon presumed arrival date appeared
to strongly influence how long birds remained at
the stopover site. This is consistent with other
findings that lean birds stop over longer than fatter
birds (Cherry 1982, Loria and Moore 1990, Wang
and Moore 1997, Matthews 2008; but see Safriel
and Lavee 1988, Salewski and Schaub 2007). The

Seewageii and Guglielnio • BODY COMPOSITION AND STOPOVER DURATION

85

FIG. 1. Relationship between arrival fat content (arcsine-root transformed g fat/g total body mass) and minimum
stopover duration of Swainson’s Thrushes (squares), Hermit Thrushes (triangles), and Ovenbirds (circles) captured during
autumn migration in New York City, USA. Solid and dashed lines represent linear (r = 0.56, T,,9 = 1 1.31, P = 0.008) and
exponential (r = 0.71, P,, 9 = 22.08, P = 0.001) models, respectively.

arrival fat content of individuals that stayed for
only 1 day was at least 1 1 % of total body mass,
whereas all individuals with fat content <6% of
total body mass remained for no fewer than 4
days, suggesting a possible threshold fat level for
departure. However, how much additional fat
birds acquired between capture and departure is
unknown.

We did not find any relationship between lean
body mass and minimum stopover duration. Lean
tissues (e.g., flight muscles, digestive organs) may
contribute significantly to total body mass dy-
namics during migration. This is particularly
apparent in shorebirds that routinely make excep-
tionally long non-stop flights and in passerines
when they must cross ecological barriers (Biebach
1998, Karasov and Pinshow 1998, Battley et al.
2000, Guglielmo and Williams 2003). No formi-
dable ecological barriers exist immediately north
of New York City, and it is possible most arriving
autumn passerine migrants have not recently
metabolized substantial non-fat tissue, and re-
building of lean mass is not a significant
component of stopover refueling. This is not
supported by the lean mass gained by the

recaptured Ovenbird which accounted for ~40%
of its total body mass increase. Four of eight other
(non-radio-marked) birds that we recaptured
during the season also had gains in lean mass
(CLS and CGG, unpubl. data). It is unknown from
our small sample sizes of recaptured and radio-
marked birds whether migrants commonly deposit
lean tissue at this site and if lean body mass
affects departure timing. Fusani et al. (2009) did
not find lean mass (muscle scores) to influence
migratory restlessness in two of three species
examined. This also suggests lean mass does not
influence departure decisions. Further study on
the relationship between lean body mass and
stopover duration is needed.

Fuel load is not the only variable that can
influence stopover duration; weather, predation
risk, and date are also factors that possibly govern
a bird’s decision to depart a stopover site. Our
small sample size of 1 1 birds prohibited including
additional predictor variables in the analyses and
we cannot assess any effect they may have had on
stopover duration. However, the birds were
marked on 5 days over a span of only 9 days;
thus, they probably experienced similar predation

86

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. I. March 2010

risk and temporal pressure, and were exposed to
comparable weather conditions. Arrival body
composition likely differed among the individuals
we studied more so than any of these potential
extrinsic influences.

The recaptured Ovenbird may illustrate how
multiple variables can affect stopover duration.
This bird appeared to have stored sufficient fat to
resume migration by its eighth stopover day, yet it
remained at the site for 6 additional days.
Overnight wind direction following recapture
was primarily from the south until the night the
bird departed when winds came from the north,
the preferred direction of southbound passerines
(Gauthreaux 1991). It is possible this bird was
energetically prepared for departure by the eighth
day but waited 6 additional days for more
favorable overnight flying conditions. Six of the
other birds similarly departed on nights with
northern winds; the remaining four departed
during eastern winds.

Our sample size was small, but arrival fat mass
had a clear and strong effect on time birds
remained at the stopover site. Similar relation-
ships between stopover duration and some
measure of energetic condition have been docu-
mented previously (Cherry 1982, Loria and
Moore 1990, Wang and Moore 1997, Matthews
2008), but the contributions of fat and lean mass
were not addressed individually. Quantitative
magnetic resonance scanning allowed us to
.separately examine how each tissue affected
stopover duration. Arrival fat mass at our study
site affected migrants’ decisions to leave whereas
lean body mass did not. Thus, the conditions for
increasing or maintaining fat stores provided by
this site and possibly other similar urban habitats
can affect the migration timing of birds using
them. Our results demonstrate quantitative mag-
netic re.sonance analysis can be useful under field
conditions, and that combining it with telemetry
and other approaches will improve our ability to
understand the stopover biology of birds.

ACKNOWLEDGMENTS

This research would not have been possible without the
hard work and contributions of Rafael Campos. Nancy
Clum. Gary Del ‘Abate. Robert Haupt, Liam McGuire.
Mark .Shaw, Christine Sheppard. Eric Slayton, and
especially Quentin Hays. Funding was provided by the
Canadian Foundation for Innovation. Ontario Ministry of
Re.search and Innovation, and an NSFRC Discovery Grant
to CGG. Part of this research was undertaken as part of an
Environmental Benefit Project funded through the re.solu-

tion of an enforcement action for violations of the
Environmental Conservation Law of New York State and
its implementing regulations. Helpful comments on earlier
drafts of this manuscript were provided by David Cerasale,
Robert DeCandido, and Liam McGuire. Research protocols
were approved by the University of Western Ontario’s
Council on Animal Care and the Wildlife Conservation
Society’s Institutional Animal Care and Use Committee.

LITERATURE CITED

Akesson. S. and A, Hedenstrom. 2000. Wind selectivity
of migratory flight departures in birds. Behavioral
Ecology and Sociobiology 47:140-144.

Alerstam, T. and A. Hedenstrom. 1998. The develop-
ment of bird migration theory. Journal of Avian
Biology 29:343-369.

Battley P. F.. T. Piersma. M. W. Dietz. S. Tang. A.
Dekinga. and K. Hulsman. 2000. Empirical evidence
for differential organ reductions during trans-oceanic
bird flight. Proceedings of the Royal Society of
London, Series B 267:191-195.

Biebach. H. 1998. Phenotypic organ flexibility in Garden
Warblers (Sylvia horin) during long-distance migra-
tion. Journal of Avian Biology 29:529-535.

Bowlin, M. S., W. W. Cochran, and M. C. Wikelski.
2005. Biotelemetry of New World thrushes during
migration: physiology, energetics and orientation in
the wild. Integrative and Comparative Biology
45:295-304.

Chernetsov, N. and A. Mukhin. 2006. Spatial behavior of
European Robins during migratory stopovers: a
telemetiy study. Wilson Bulletin 118:364-373.
Cherry, J. D. 1982. Fat deposition and length of stopover
of migrant White-crowned Sparrows. Auk 99:725-
732.

Conway, C. J., W. R. Eddleman, and K. L. Simpson.
1994. Evaluation of lipid indices of the Wood Thrush.
Condor 96:783-790.

Danhardt, j. and A. Lindstrom. 2001. Optimal departure
decisions of songbirds from an experimental stopover
site and the significance of weather. Animal Behavior
62:235-243.

Dierschke, V. AND J. Delingat. 2001. Stopover behaviour
and departure decision of Northern Wheatears, Oc-
nanlhe oenantlic. facing different onward non-stop
flight distances. Behavioral Ecology and Sociobiology
50:53.5-545.

Fusani, L., M. Cardinale, C. Carere, and W. Goymann.
2009. Stopover decision during migration: physiolog-
ical conditions predict nocturnal restlessness in wild
passerines. Biology Letters 5:302-305.

Gauthreaux Jr., S, A. 1991. The flight behavior of
migrating birds in changing wind fields: radar and
visual analyses. American Zoologist 31:187-204.
Gotelli, N. j. and A. M. Ellison. 2004. A primer of
ecological statistics. Sinauer A.ssociates Inc., Sunder-
land, Massachu.setts, USA.

Guglielmo, C. G. and T. D. Williams. 2003. Phenotypic
flexibility of body composition in relation to migratory
state, age, and .sex in the Western Sandpiper (Calidris

Scewagcn ami Guglielnio •

BODY COMPOSITION AND STOPOVER DURATION

87

maitri). Physiological and Biochemical Zoology
76:84-98.

Guglielmo, C. G., D. J. Cerasale, and C. Eldermire.
2005. A field validation of plasma metabolite profiling
to assess refueling performance of migratory birds.
Physiological and Biochemical Zoology 78:116-125.

Hays, Q. R. 2008. Stopover behavior and refueling
physiology in relation to migration distance in
Wilson's Warblers {Wilsonia piisilla). Thesis. Univer-
sity of Western Ontario, London, Canada.

Houston, A, 1. 1998. Models of avian optimal migration:
state, time and predation. Journal of Avian Biology
29:395-404.

Kar.asov, W. H. and B. Pinshow. 1998. Changes in lean
mass and in organs of nutrient assimilation in a long-
distance passerine migrant at a springtime stopover
site. Physiological Zoology 71:435^48.

Krementz, D. G. and G. W. Pendleton. 1990. Fat
scoring: sources of variability. Condor 92:500-507.

Loria, D. E. and F. R. Moore. 1990. Energy demands of
migration on Red-eyed Vireos, Vireo olivaceus.
Behavioral Ecology 1:24—35.

MacArthur, R. H. and a. T. MacArthur. 1974. On the
use of mist nets for population studies of birds.
Proceedings of the National Academy of Sciences of
the USA 71:3230-3233.

Matthews, S. N. 2008. Stopover habitat utilization by
migratory landbirds within urbanizing landscapes of
central Ohio. Dissertation. Ohio State University,
Columbus, USA.

Morris, S. R. and J. L. Glasgow. 2001. Comparison of
spring and fall migration of American Redstarts on
Appledore Island, Maine. Wilson Bulletin 113:202-
210.

Morris, S. R., D. A. Liebner, A. M. Larracuente, E.
M. Escamilla, and H. D. Sheets. 2005. Multiple-
day constancy as an alternative to pooling for
estimating mark-recapture stopover length in Nearc-
tic-neotropical migrant landbirds. Auk 122:319-
328.

Raim, A. 1978. A radio transmitter attachment for small
passerine birds. Bird-Banding 49:326-332.

Safriel, U. N. and D. Lavee. 1988. Weight changes of
cross-desert migrants at an oasis — do energetic
considerations alone determine the length of stopover?

Oecologia 76:61 1-619.

Salewski, V. AND M. SCHAUB. 2007. Stopover duration of
Palearctic passerine migrants in the western Sahara -
independent of fat stores? Ibis 149:223—236.

ScHAUB, M. 2006. How to study departure decisions of

migrants from stopover sites using capture-recapture
data. Acta Zoologica Sinica 52:602-605.

SCHAUB. M., L. Jenni and F. Bairlein. 2008. Fuel stores,
fuel accumulation, and the decision to depart from a
migration stopover site. Behavioral Ecology 19:657-
666.

SCHAUB, M.. R. Pradel, L. Jenni, and J. Lebreton. 2001.
Migrating birds stop over longer than usually thought:
an improved capture-recapture analysis. Ecology
82:852-859.

Seewagen, C. L. 2008. An evaluation of condition indices
and predictive models for non-invasive estimates of
lipid mass of migrating Common Yellowthroats.
Ovenbirds and Swainson’s Thrushes. Journal of Field
Ornithology 79:80-86.

Seewagen, C. L. and E. J. Slayton. 2008. Mass changes
of migratory landbirds during stopovers in a New York
City park. Wilson Journal of Ornithology 120:296-
303.

Skagen, S. K., F. L. Knopf, and B. S. Cade. 1993.
Estimation of lipids and lean mass of migrating
sandpipers. Condor 95:944-956.

Spengler, T. j., P. L. Leberg, and W. C. Barrow Jr.
1995. Comparison of condition indices in migratory
passerines at a stopover site in coastal Louisiana.
Condor 97:438—444.

Taicher, G. Z., F. C. Tinsley, A. Reiderman, and M. L.
Heiman. 2003. Quantitative magnetic resonance
(QMR) method for bone and whole-body-composition
analysis. Analytical and Bioanalytical Chemistry
377:990-1002.

Tinsley, F. C., G. Z. Taicher, and M. L. Heiman. 2004.
Evaluation of a quantitative magnetic resonance
method for mouse whole body composition analysis.
Obesity Research 12:150-160.

Tsvey, A., V. N. Bulyuk, and V. Kosarev. 2007.
Influence of body condition and weather on departures
of first-year European Robins. Erithaciis mhecula.
from an autumn migratory stopover site. Behavioral
Ecology and Sociobiology 61:1665-1674.

Wang, Y. and F. R. Moore. 1997. Spring stopover of
intercontinental migratory thrushes along the north-
ern coast of the Gulf of Mexico. Auk 114:
263-278.

WiKELSKi. M., E. M. Tarlow, A. Raim, R. H. Diehl, R. P.
Larkin, and G. H, Visser. 2003. Costs of migration
in free-llying songbirds. Nature 423:704.

Zar, j. H. 1999. Biostatistical analysis. Fourth Edition.
Prentice Hall. New Jersey. USA.

The Wilson Journal of Ornithology 122( l);88-94, 2010

SEASONAL FLUCTUATION OF THE ORANGE- WINGED AMAZON AT

A ROOSTING SITE IN AMAZONIA

LEILIANY NEGRAO de MOURAT^ JACQUES M. E. VIELLIARDr
AND MARIA LUISA da SILVA'

STRACT.— We recorded fluctuations in a population of Orange-winged Amazon (Amazona amazonica) during
1 year at a roostmg site on an island near Belem, Para, Brazil. Parrots were counted from a boat by a minimum of three
n ^ observers, each team oriented in different directions. Orange-winged Amazons were observed flying alone
n xcr "’’- a" U5.7%), and small numbers in family groups (pairs with young) of three (8.7%), four (1.2%), or five
. o) individuals. The larger number of groups of three compared with groups of four and five individuals reflects the low
Qnn generally only one surviving offspring per brood. The total number of parrots increased from

pnl (J,899) to July (8,539), and began to decrease in August (5,351). This decrease was presumably due to onset of the
reeding season, when paired individuals leave the roost in search of a nest, where they breed, nest, and rear youn« until the
nestlings can fly. Received 24 January 2009. Accepted 28 August 2009.

Temporal and spatial variations in estimates of
wild animal populations are important indicators
of underlying demographic processes and ecolog-
ical interactions (Nunes and Betini 2002). Popu-
lation estimates are particularly important in the
case of parrots (Aves, Psittacidae) because this
family has a disproportionate number of threat-
ened species (Collar and Juniper 1992). Censusing
over significant periods of time is an efficient tool
to distinguish between natural and anthropogenic
fluctuations (Bibby et al. 1992), which is essential
for future conservation of species.

There are different methods to estimate popu-
lation size of birds. In the case of parrots, most
methods are based on counts of individuals by
visual and/or auditory contacts (Collar and
Juniper 1992). They are registered in high number
early in the morning and late afternoon (Pizo et al.
1997), and counts in the middle of the day should
be avoided because activity of parrots decreases
abruptly at this time (Marsden 1999). Roost
counts may permit targeted assessments of parrot
populations in small areas and on islands where
parrots roost communally (Snyder et al. 1987,
Matuzak and Brightsmith 2007).

The Orange-winged Amazon (Amazona ama-
zonica) gathers overnight in communal roosts
(Sick 1997, Juniper and Parr 1998). Individuals
meet just before or shortly after sunset (Sick
1997), making it relatively easy to obtain

' Laboratory of Ornithology and Bioacoustics, Institute of
Biological Sciences, Federal University of Para, 66075-1 10
Belem, PA, Brazil.

"Laboratory of Bioacoustics, Institute of Biology, State
University of Campinas, 13083-970 Campinas, SP. Brazil.
’Corresponding author; e-mail: leiliany(®ufpa.br

estimates of the minimum number of individuals
(Nunes and Betini 2002). Our objective was to
conduct population estimates of Orange-winged
Amazons for all months of the year at a roosting
site on the Ilha dos Papagaios (Parrot’s Island),
near Belem (Para State, Brazil), discriminating
family groups (single birds, pairs, and groups of
three, four, or five individuals), and measuring
seasonal fluctuation throughout the year.

METHODS

Study Area. — Ilha dos Papagaios is a roosting site
for a significant Orange-winged Amazon popula-
tion, a common species in the region. It is in
Guajara Bay, south ofBelem, Brazil (01° 31' 37" S,
48° 30' 22" W), and has an area of 7.4 ha, eight
houses, and about 36 inhabitants. The island is low
and subject to periodic tidal flooding (Novaes and
Lima 1998). The climate is tropical humid and the
source of the seasonality is the precipitation,
although there is no pronounced diy season. There
are two seasons: a rainy season and a less rainy
season. The rainy season begins in December, peaks
in March, and subsides in May. January, February,
and March are the wettest months with average
monthly rainfalls exceeding 400 mm. The less rainy
sea.son usually corresponds to the period from June
to November, and October is the driest month with
an average precipitation of 86 mm (annual precip-
itation = 2,800 mm; SECT AM 1994).

The vegetation is mostly dense alluvial forest
with emergent and uniform canopy in areas of
lowland floodplain. Flooded forests, and sandy
alluvial deposits (beaches) covered by sparse
vegetation, characteristic of Guajara Bay, are also
present (Novaes and Lima 1998).

88

de Moiira el cd. • ORANGE-WINGED AMAZON ROOST IN AMAZONIA

89

Surveys. — Counts of A. amazonica were con-
ducted at the roost on llha dos Papagaios. We
conducted 96 surveys (54 in the afternoon and 42
in the morning) from September 2004 through
September 2005. Pan'ots were counted from a
single medium-sized boat (12 m long), positioned
in a point (about 100 m southeast of the island)
where it was possible to clearly observe all the
main flight routes of the parrots’ amval and
departure to and from the roost.

Two visits were made per week, one between
1700 and 1900 hrs and one the following day
between 0500 and 0800 hrs, before sunset and
sunrise, respectively. During counts, single birds,
pairs, and groups of three, four, or five individuals
were noted when possible with larger groups
being counted or estimated when it was not
possible to separate individuals.

Counts were made by three teams each with
at least one observer and one auxiliary.
Observers counted all individuals passing in
their field of view, arriving or departing to or
from previously selected directions in relation
to the roost; (1) to or from the north and
northeast of the island; (2) to or from the
southeast and south of the island; and (3)
parrots going to or arriving from the other side
of the island. The teams were careful to not
duplicate individuals already counted by them-
selves or by another team, eventually announc-
ing loudly their counts to the near-by team.
Each observer announced to his auxiliary the
number and group composition of the parrots
seen with the aid of 7 X 50 binoculars; the
auxiliaries recorded these data on standardized
forms, separating the numbers of birds seen per
minute.

Statistical Analyses. — Data were analyzed with
Statistica 7.1 software package (Stasoft Inc. 2005)
and are expressed as means ± SD. Variables were
examined for normality using Kolmogorov-Smir-
nov tests. All data followed a normal distribution
and parametric tests were used (ANOVA and
paired r-test). Differences between means were
considered statistically significant at P < 0.05.
We calculated the percentages of the parrot flocks
flying to or from the roost (single birds, pairs and
groups of three, four, or five individuals) per visit
and averaged the data by month. Cluster analysis
was conducted using the City-block (Manhattan)
distance method and weighted pair-group aveiage
to test for seasonal differences (Sneath and Sokal
1973).

RESULTS

Patterns of Flight Behavior. — Single individu-
als (probably unpaired adults or young) accounted
for 14.2 ± 6.7% of the observed groups, pairs
(sexually mature adults) 75.7 ± 6.6%, and family
groups of three (presumably pairs with 1 young)
8.7 ± 3.3%, groups of four (presumed pairs with
2 young) 1.2 ± 1.6%, and groups of five
(presumed pairs with 3 young) 0.3 ± 0.4%.

Total Number of Individuals and Seasonal
Fluctuation. — The average number of individuals
per visit varied by month (Eig. 1). June (7,084 ±
1,708) and July (8,539 ± 704) had the highest
number of pan'ots counted at the roosting site (max
= 9,603 on 2 July 2005). Total numbers decreased
in August (5,351 ± 1,534) and again in September
(3,819 ± 462 in 2005 but only 2,040 ± 383 in
2004). The lowest counts were recorded from
October (1,545 ± 245) through December (1,364
± 161), followed by a slightly higher level in
January (1,895 ± 223), February (1,778 ± 356),
and March (2,361 ± 453), after which the number
increased steadily (3,899 ± 547 in April and 4,484
± 814 in May) until the July peak.

Size of Flocks and Seasonal Fluctuation.— Iht
number of single individuals, pairs, and groups of
three, four, and five individuals varied over the year
(Fig. 2). These patterns were similar to those for
total numbers, except for groups of four and five
individuals, which were essentially absent from
September until December. In January 2005 they
began to increase, reaching a level of —3.5 and
—2.3%, respectively between March and July 2005,
and then decreased in August and September 2005.

Annual Cycle. — ^Monthly values separated into
three major groups (Fig. 3). The first period
included September 2004 to March 2005, the
second period was April, May, August, and
September 2005, and the third period was June
and July 2005.

There was a significant difference between
these periods in total number of paiTots across all
Hock sizes (ANOVA E2.92totai ~ 339.96, P <
0.01; /^2.92.single ~ 37.15, P < 0.01; E2,92pairs ~
204.51, P < 0.01; F2,92ihree = 152.55, P < 0.01;
p2.92rour = 36.58, P < 0.01; P2.92f.ve = >8.80, P <
0.01) (Table 1).

All comparisons of total number and number of
single individuals, pairs, and groups of three, were
significant; the averages for period 1 were smaller
than those for period 2, which were smaller than
those for period 3. Total numbers of parrots and

90

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. I. March 2010

Tf

VY

WD

IT)

o

o

o

o

O

o

o

o

O

O

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

a

D

00

Ci

o

O

>

o

Z

u

<D

o

*3

fN

X)

LU

<N

ca

s

fN

a.

<

<N

to

2

c

3

04

D

'a

3

<

04

a

00

Months

® Mean
X MeaniSD

FIG. 1. Number of roosting Orange-winged Amazons by month. Number of counts (/;) is inserted in the
respective column.

numbers of single individuals, pairs, and groups of
three in 2005 increased significantly from April with
a new significant increase in June and a significant
decrease after August. Only period 1 versus 2 and
period 1 versus 3 were significantly different in the
case of groups of four and five individuals.

Validation of Counts. — ^The number of parrots
that arrive at the roosting site in the evening should
be related to the number that leave the following
morning. We verified the reliability of our counts
by comparing the average number of parrots
arriving with the average number leaving the roost
using a paired r-test. We used the average total
number of parrots in each of the three significantly
different periods previously identified (Fig. 3).
Differences between afternoon and morning aver-
ages were not significant (from Sep 2004 to Mar
2005: afternoon = 1,662 ± 364 vs. morning =
1,597 ± 438, ^21 = 0.81, P = 0.43; Apr, May, Aug,
and Sep 2005: afternoon = 4,110 ± 824 vs.
morning = 4,57 1 ± 995, tq = 1 .46, P = 0.62; Jun
and Jul 2005: afternoon = 8,080 ± 1,394 vs.
morning = 8,122 ± 1,399. /f, = 1.52, P = 0.18).

DISCUSSION

Counting Method. — ^The method used in this
study of counting parrots from a fixed site as they
arrived at or left the roost is similar to that used by
other parrot researchers who have studied other

parrot species (Rocha et al. 1988, Guedes 1993,
Casagrande and Beissinger 1997, Mabb 1997,
Martinez and Prestes 2002, Harms and Eberhard
2003, Cougill and Marsden 2004, Berg and Angel
2006, Costa 2006, Matuzak and Brightsmith
2007). Some paiTots could have amved at or left
the island by routes not countable from the survey
boat. Thus, our technique provides only a
minimum of the number of individuals at the
roost (Nunes and Betini 2002). We did not find a
significant difference between the number of
parrots that arrived and the number that left the
roosting site and we can consider our counts valid.

Number of Individuals, Seasonal Fluctuation,
and Breeding Season. — ^The breeding season is
apparently a primary factor influencing the fluctu-
ation in number of parrots at the roosting site
throughout the year. Nesting female parrots cannot
roost gregariously if they are to incubate eggs and
take care of offspring (Berg and Angel 2006). It is
common that numbers of pairots at roosting sites
diminish during the period when pairs need to find
nest sites, lay, and incubate eggs (Cannon 1984,
Chapman et al. 1989, Carrillo et al. 2002). These
findings are similar to those reported for other
species of parrots (Mabb 1997, Olmos 1997, Berg
and Angel 2006, Costa 2006, Matuzak and Bright-
smith 2007); however, the period of low attendance
at the roost is particularly long at our study site.

de Maura et al. • ORANGE-WINGED AMAZON ROOST IN AMAZONIA

91

I Singles
0 Pairs

Months

0 Groups of three

1 Groups of four
I Groups of five

FIG. 2. Number of roosting flocks of Orange-winged Amazons by month. (A). Single birds and pairs. (B). Groups of
three, four, and five individuals.

92

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. I. March 2010

Months

LIG. 3. Tree-clustering of monthly average values of total number of Orange-winged Amazons and number in each
grouping (single birds, pairs, groups of three, four, and five individuals) per visit.

Orange-winged Amazons have asynchronous
reproduction that may account for the long time
between the decrease of the roost population and
the return of pairs with young. Parrots began to
leave the roost in August and their dispersal was
completed by October, while pairs started to come
back to the roost with their young in January
(Fig. 2). Other parrots are still breeding at this
same time, and it is only in June, when other
nestlings had become emancipated from their
parents, that number of individuals reached its
yearly maximum.

Patterns of Flight Behavior. — Parrots in the
genus Amazona remain in family groups even
when flying and roosting in larger aggregations.

and such groups are easily recognizable (Martus-
celli 1995). Most parrots flew in pairs during the
counts, which is the primary social unit at least for
Amazons, as they are monogamous and pair for
life (Snyder et al. 1987, Sick 1997, Gilardi and
Munn 1998, Juniper and Parr 1998, Guedes and
Seixas 2002). Single individuals (14.2% of group
occurrences) are probably unpaired adults or
young that have not reached sexual maturity but
are no longer accompanied by their parents.
Young remain with their parents only until the
parents began to nest again (Sick 1997).

Groups of three constituted 8.7% of all groups,
groups of four 1.2%, and groups of five 0.3%,
suggesting small family groups (Gilardi and Munn

TABLE 1.
using Tukey’s

Size of flocks of Orange-winged Amazons at an Amazonia roosting site, Brazil, 2004-2005 (cluster analysis
HSD test).

Comparison

Total number

Single birds

Pairs

Coups of three

Coups of four

Coups of five

Period 1 vs.

1678 ± 422

1 18 ± 53

610 ± 128

66 ± 35

7 ± 14

2 ± 3 ■

Period 2

4411 ± 1035

230 ± 153

1363 ± 443

196 ± 51

36 ± 22

8 ± 7

P - value

<0.01

<0.05

<0.01

<0.01

<0.01

<0.01

Period 1 vs.

1678 ± 422

1 18 ± 53

610 ± 128

66 ± 35

7 ± 14

2 ± 3

Period 3

7755 ± 1495

557 ± 397

2430 ± 532

259 ± 57

43 ± 26

6 ± 3

P - value

<0.01

<0.01

<0.01

<0.01

<0.01

<0.05

Period 2 vs.

4411 ± 1035

230 ± 153

1 363 ± 443

196 ± 51

36 ± 22

8 ± 7

Period 3

7755 ± 1495

557 ± 397

2430 ± 532

259 ± 57

43 ± 26

6 ± 3

P - value

<0.01

<0.01

<0.01

<0.01

>0.05

>0.05

de Maura ct al. • ORANGE-WINGED AMAZON ROOST IN AMAZONIA

93

1998). The higher number of groups of three
(presumably pairs with one young), compared
with groups of four and five individuals (pre-
sumed pairs with 2 and 3 young, respectively),
reflects the characteristically low survival rate of
nestling parrots (Gnam and Rockwell 1991, Munn
1992, Guedes 1993, Lindsey et al. 1994, Carrillo
et al. 2002, Seixas and Mourao 2002). We
observed clutches of 2-4 eggs in five nests, but
only one nestling survived in each of the two nests
that were not visited for the illegal pet trade
(Moura et al. 2008) when we studied reproductive
biology of the Orange-winged Amazon during the
2006/2007 breeding season in Santa Barbara (Para
State, 41 km from the roosting site).

Orange-winged Amazons have a wide distribu-
tion and are usually abundant and, at the moment,
are not at risk (BirdLife International 2008).
However, this species is largely captured from the
wild for the international trade in Brazil, being
fourth on the list of most apprehended species of
parrots by the Brazilian Institute of Environment
and Renewable Natural Resources (IB AM A)
(Oliveira and Caparroz 2007). It is a popular,
although illegal, pet in the region and poaching is
intensive. Thus, it is important to monitor Orange-
winged Amazon populations for effective conser-
vation actions in the future.

llha dos Papagaios should be considered a
priority area for conservation as it is an important
refuge for the parrots. It is close to an urban center
(8 km from Belem, population of 2-million
inhabitants) and is a convenient site for environ-
mental education programs, where people can see
a great number of free parrots and become aware
that the strong culture of having a wild bird as pet
is a crime against nature.

ACKNOWLEDGMENTS

We thank Ana Paula Assump^ao, Camilo Araujo,
Dnilson Ferraz, Eliane Reis, Isabela Brcko, Karine Pereira,
Jose Leonardo Magalhaes, Paulo Costa, Renata Emin-Lima
and Vitor Lima for assistance with field work; Dr. Jonathan
Widdicombe for the revision of an earlier draft; C. E. Braun
and the reviewers for providing constructive comments and
help in improving the contents of this paper; and
PARATUR and CAPES for financial support of this
project.

LITERATURE CITED

Berg, K. S. and R. R. Angel. 2006. Seasonal roosts of
Red-lored Amazons in Ecuador provide information
about population size and structure. Journal of Field
Ornithology 77:95-103.

Bibby, C. J., D. N. Burgess, and D. A. Hill. 1992. Bird
census techniques. Academic Press, London, United
Kingdom.

Birdlife International. 2008. Amazona amazonica.
Version 2009. 1. lUCN Red List of Threatened
Species, www.iucnredlist.org (accessed 5 August
2009).

Cannon, C. E. 1984. Flock size of feeding Eastern and
Pale-headed rosellas (Aves: Psittaciformes). Austra-
lian Wildlife Research 1 1:349-355.

Carrillo, A. C., E. A. B. Sipinski, M. L. Cavalheiro,
and K. L. Oliveira. 2002. Conserva9ao do Papagaio-
de-cara-roxa (Amazona hrasiliensis) no estado do
Parana. Pages 193-213 in Ecologia e conservafao de
psitacideos no Brasil (M. Galetti and M. A. Pizo,
Editors). Melopsittacus Publica96es Cientificas, Belo
Horizonte, Brazil.

Casagrande, D. G. and S. R. Beissinger. 1997.
Evaluation of four methods for estimating parrot
population size. Condor 99:445-457.

Chapman, C. A., L. J. Chapman, and L. Lefebvre. 1989.
Variability in parrot flock size: possible functions of
communal roosts. Condor 91:842-847.

Collar, N. J. and A. T. Juniper. 1992. Dimensions and
causes of the parrot conservation crisis. Pages 1-24 in
New World parrots in crisis (S. R. Beissinger and N. F.
R. Snyder, Editors). Smithsonian Institution Press,
Washington, D.C., USA.

Costa, P. C. R. 2006. Comportamento alimentar e
dinamica populacional do Periquito— de— asa— branca
Brotogeris versicolurus (Aves - Psittacidae) na cidade
de Belem, PA. Monograph. Universidade Federal do
Pari Belem. Brazil.

COUGILL, S. and S. j. Marsden. 2004. Variability in roost
size in an Amazona parrot: implications for roost
monitoring. Journal of Field Ornithology 75:67—73.

Gilardi, j. D. and C. Munn. 1998. Patterns of activity,
flocking, and habitat use in parrots of the Peruvian
Amazon. Condor 100:641-653.

Gnam, R. S. and R. F. Rockwell. 1991. Reproductive
potential and output of the Bahama Parrot Amazona
leiicocephala hahamensis. Ibis 133:400^05.

Guedes, N. M. R. 1993. Biologia reprodutiva de Arara-azul
(Anodorhynchus liyacinihinus) no Pantanal-MS. Brasil.
Dissertation. Universidade de Sao Paulo. Piracicaba.
Brazil.

Guedes, N. M. R. and G. H. F. Seixas. 2002. M^todos para
estudos de reprodu9ao de Psitacideos, Pages 123-139
in Ecologia e conserva9ao de psitacideos no Brasil (M.
Galetti and M. A, Pizo. Editors). Melopsittacus
Publica96cs Cientificas, Belo Horizonte. Brazil.

Harms, K. E. and J. R. Eberhard. 2003. Roosting
behavior of the Brown-throated Parakeet (Aratinga
pertinax) and roost locations on four southern
Caribbean islands. Ornitologia Neotropical 14:79-89.

Juniper, T. and M. Parr. 1998. Parrots: a guide to the
parrots of the world. Pica Press, London, United
Kingdom.

Lindsey, G. D., W. J Arendt, and J. Kalina. 1994.
Survival and causes of mortality in juvenile Puerto
Rican Parrots. Journal of Field Ornithology 65:76-82.

94

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 1, March 2010

Mabb, K. T. 1997. Roosting behavior of naturalized parrots
in the San Gabriel Valley, California. Western Birds
28:202-208.

Marsden, S. J. 1999. Estimation of parrot and hornbill
densities using a point count distance sampling
method. Ibis 141:377-390.

Martinez, J. and N. P. Prestes. 2002. Ecologia e
conservagao do Papagaio-charao Aniazona pretrei.
Pages 173-192 in Ecologia e conservafao de psitaci-
deos no Brasil (M. Galetti and M. A. Pizo, Editors).
Melopsittacus Publica96es CientiTicas, Belo Hori-
zonte, Brazil.

Martuscelli, P. 1995. Ecology and conservation of the
Red-tailed Amazon (Aniazona hrasiliensis) in south-
eastern Brazil. Bird Conservation International 5:405-
420.

Matuzak, G. D. and D. j. Brightsmith. 2007. Roosting of
Yellow-naped Parrots in Costa Rica: estimating the
size and recruitment of threatened populations. Journal
of Field Ornithology 78:159-169.

Moura, L. N., j. M. E. Vielliard, and M. L. Silva. 2008.
Flutuafao populacional e comportamento reprodutivo
do Papagaio-do-mangue (Aniazona aniazonica). Pages
223-238 in Biologia da Conserva9ao: estudo de caso
com o Papagaio-charao e outros papagaios brasileiros
(J. Martinez and N. P. Prestes, Editors). UPF Editora,
Passo Fundo, Brazil.

Munn, C. a. 1992. Macaw biology and ecotourism, or
“When a bird in the bush is worth two in the hand?”
Pages 47-72 in New World parrots in crisis (S. R.
Beissinger and N. F. R. Snyder, Editors). Smith.sonian
Institution Press. Washington, D.C., USA.

Novaes, F. C. and M. F. C. Lima. 1998. Aves da Grande
Belem: Municfpios de Belem e Ananindeua. Museu
Paraense Emilio Goeldi, Belem, Brazil.

Nunes, M. F. C. and G. S. Betini. 2002. Metodos de
estimativa de abundancia de psitacideos. Pages 99-1 12

in Ecologia e conserva9ao de psitacideos no Brasil (M.
Galetti and M. A. Pizo, Editors). Melopsittacus
Publica96es CientiTicas, Belo Horizonte, Brazil.

Oliveira, T. G. and R. Caparroz. 2007. Analise do
impacto do trafico ilegal sobre as popula96es de
Psittacidae no Brasil. Page 215 in Annals of the XV
Brazilian Ornithological Congress. Porto Alegre.
Brazil.

Olmos, F. 1997. Parrots of the Caatinga of Piaui,
northeastern Brazil. Papageienkunde 1:173-182.

Pizo, M. A., 1. Simao, and M. Galetti. 1997. Daily
variation in activity and flock size of two parakeet
species from southeastern Brazil. Wilson Bulletin
109:343-348.

Rocha, C. E. D., H. G. Bergallo, and S. Siciliano. 1988.
Migra9ao circadiana em cinco especies de psitacideos
em Parintins-AM. Acta Amazonica 18:371-374.

SECTAM. 1994. Parque ambiental de Belem: Plano de
manejo. Sofrelec Engenharia, Rio de Janeiro, Brazil.

Seixas, G. H. F. and G. M. Mourao. 2002. Biologia
reprodutiva do Papagaio-verdadeiro (Aniazona aestiva)
no Pantanal sul-mato-grossense, Brasil. Pages 157-
171 in Ecologia e conserva9ao de psitacideos no Brasil
(M. Galetti and M. A. Pizo, Editors). Melopsittacus
Publica96es Cienti'ficas, Belo Horizonte, Brazil.

Sick, H. 1997. Ornitologia Brasileira. Nova Fronteira, Rio
de Janeiro, Brazil.

Sneath, P. H. and R. R. Sokal. 1973. Numerical
taxonomy: the principles and practice of numerical
classification. W. H. Freeman, San Francisco, Cali-
fornia, USA.

Snyder, N. F. R.. J. W. Wiley, and C. B. Kepler. 1987.
The parrots of Luquilo: natural history and conserva-
tion of the Puerto Rican Pan'ot. Western Foundation of
Vertebrate Zoology, Los Angeles. California, USA.

STATSOFT iNC. 2005. Statistica software. Version 7.1.
StatSoft Inc., Tulsa. Oklahoma, USA.

The Wilson Joiinial oj' Ornithology 122( 1 ):95-l()l, 2010

COSTS AND BENEFITS OF FORAGING AFONE OR IN MIXED-
SPECIES AGGREGATIONS FOR FORSTER’S TERNS

LISA SCHREFFLER,' - JOHN K. LEISER,- AND TERRY L. MASTER'"'’

ABSTRACT.— Forster's Terns (Sterna forsteri) often forage alone or within mixed- species toraging aggregations. We
investigated the relative costs and benefits of these two foraging strategies. Metrics used to assess diflerences in strategies
include^d the number of attempts for prey (both successful and missed), capture efficiency, and capture rate. Forster s Terns
foraoinc socially cue on presence and behavior of other aggregating species, especially Snowy Egrets (E^retta tinila).
JoinTng^iggregations significantly increased their capture rate, but not their capture efficiency, compared to solitarily
foraaing terns. However, total aggregation size, influenced primarily by the two most abundant species, Laughing Gu Is
(Laras \itrieilla) and Snowy Egrets, affected tern behavior. For example, diving terns often collided with Laughing Gu s,
causing frequent missed attempts for socially foraging individuals. Thus, large aggregation sizes had a negative effect on
foraaina success, perhaps because of interference and/or prey depletion. In contrast, solitarily foraging terns dove less often
for prey but also missed less often. This may be why terns were observed foraging solitarily as aggregation sizes and
conaestion increased. Received 20 February 2009. Accepted 17 July 2009.

Important among an individual’s behavioral
decisions are those regarding food acquisition.
Each individual must decide when and where to
forage (Chamov 1976), on what prey item(s) to
focus (Schoener 1971, Collier et al. 1991), and
what capture strategy to use (Ydenberg et al.
1994). These decisions are constrained by the
necessity of energy acquisition and will ultimately
influence an animal’s fitness.

One option animals can choose to make food
acquisition decisions easier is to join a group in a
process called local enhancement. Foraging in
groups has been shown to provide benefits,
including reduction of an individual’s energy
intake variance (Caraco 1981), increased prey
capture rate (Pulliam and Millikan 1982), reduced
energy expenditure (Master et al. 1993), increased
vigilance (Caldwell 1986), reduced predation risk
from “many-eyes” (Lima 1995) or dilution
effects (Hamilton 1971), and the imparting of
information about novel foraging techniques
(Webster and Lefebvre 2001). Foraging in a
group also has disadvantages, including increased
conspicuousness to predators (Hamilton 1971,
Bertram 1978, Caldwell 1986) and increased
energy expenditure on aggressive interactions
with competing foragers (Grand and Dill 1999).

Some disadvantages of feeding within a group
may be assuaged if individuals join mixed rathei
than single-species groups to forage. The com-

' Department of Biological Sciences, East Stroudsburg
University, East Stroudsburg, PA 18301, USA.

^Biology Department, Northampton Community Col-
lege, Monroe Campus. Tannersville, PA 18372, USA.
^Corresponding author; e-mail: tmaster(®po-box. esu.edu

petitive costs of food acquisition are, at times,
reduced by species-specific preferences for prey
items in mixed-species feeding aggregations
among wading birds (Moreno et al. 2005).
Competitive costs are also moderated by differ-
ences in preferred feeding locations (Moreno et al.
2005) and by reduced interference due to variation
in prey catching techniques (Latta and Wunderle
1996).

Few studies of Forster’s Tern (Sterna forsteri)
foraging behavior have been conducted (but see
Salt and Willard 1971, Reed 1985). We investi-
gated tern feeding behavior within mixed-species
aggregations of wading birds and while foraging
solitarily in a salt marsh in southern New Jersey,
USA. Our goals were to: ( 1 ) identify the role of
terns in formation and structure of mixed-species
aggregations that progress from an initial for-
mation phase” with few species and individuals
to an active phase with more species and
individuals, and finally to a dispersal phase
(Master et al. 1993); and (2) investigate foraging
tactics used by terns under changing social
conditions by comparing benefits accrued versus
costs incurred by feeding within aggregations and
solitarily.

METHODS

Study Area. — ^This study was conducted over
two summer seasons (2004—2005) in a 2,083-ha
salt marsh in southeastern New Jersey, USA
(~39° 3' 74° 46' W). This marsh provided an

unusually large number of relatively shallow
(~3_60 cm) pools in which terns forage (Master
1992). The majority of fish prey within the pools
were mummichogs {Funduhis heteroclitus) and

95

96

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 1, March 2010

sheepshead minnows {Cyprinodon variegatus);
prey density averaged (± SD) 147.15 ± 57.92
fish/nr in the pools (Master 1992). The popula-
tion of Forster’s Terns in southern New Jersey
was estimated at 403 individuals in the late 1970s
and early 1980s (Spendelow and Patton 1988). At
least eight separate colonies were within Cape
May County more recently although numbers of
birds were not reported (Walsh et al. 1999).

F oraging Observations. — ^Terns were observed
trom June until mid-August during each field
season. Both solitary and aggregated terns were
observed using binoculars (Nikon 10 X 40) or a
spotting scope (KOWA TSN 821, 20-60X zoom),
depending on the subject’s proximity to the
observer. We defined a mixed-species aggregation
as a group of at least four birds from at least two
different species foraging within close proximity
(~1 m) to one another at a single pool. The
species most often composing aggregations, in
addition to terns, were Great Egret {Ardea alba).
Snowy Egret (Egretta thida). Glossy Ibis {Plega-
dis folcinellus). Great Blue Heron (Ardea lier-
odias). Laughing Gull (Larus atricilla), and Least
Tern (Sterna antillarinm). Aggregations ranged in
size from four to 430 individuals (mean aggrega-
tion size ± SE: 93.74 ± 4.70, n = 87).

Observations began each morning at sunrise as
aggregations were forming and continued until
— 1130 hrs EDT, when nearly all aggregations had
dispersed. An aggregation was chosen randomly at
sunrise and the total number of species represented
and number of individuals of each species were
counted. Numbers of individuals were summed to
calculate total aggregation size. A single Eorster’s
Tern was selected for observation, after the initial
count, using the focal animal sampling technique
(Schmitz and Baldassarre 1992). The focal tern was
observed for 7 min (Balph and Romesburg 1986,
Rooney and Sleeman 1998) during which the
number of successful and missed foraging attempts
was recorded. Capture rate and capture efficiency
were calculated from these observations by dividing
the number of successful attempts by the observa-
tion time of 7 min, and by dividing the number of
successful attempts by the total number of attempts,
both successful and missed, respectively.

A 10-min intermission followed observation of
a focal individual prior to the next 7-min
ob.servation to allow terns to re-distribute them-
selves throughout the aggregation. After the
intermission, a .second count of the participants
in the aggregation was recorded, another tern was

selected, and a 7-min focal observation was
begun. Our protocol continued in this manner
for 58 min, providing four separate focal tern
observations and species counts per aggregation
(3 10-min intermissions between 4 7-min obser-
vation periods). Additional aggregations were
located for continued observation until the
— 1130 hrs dispersion time. The Schmitz and
Baldassame (1992) sampling procedure does not
ensure an individual tern is observed only once,
but this procedure is well accepted when studying
aggregating animals and is resistant to potential
sampling biases (Weinstein 1995). There were 38
aggregations studied in 2004 and 49 in 2005.
Observations were conducted on 275 focal terns
(126 in 2004 and 149 in 2005).

Observations were also conducted on terns
foraging solitarily. These observations followed a
similar protocol with two exceptions. Eirst,
aggregation counts were unnecessary. Second,
solitary terns were observed for as long as they
were visible. Nineteen solitary terns were ob-
served with a mean (± SD) observation time of
7.00 ± 1.09 min. Total attempts, capture rates,
and capture efficiencies were calculated on a per
7-min basis for comparison with aggregating
terns. Solitary terns were observed during the
same morning period as aggregations (sunrise to
-1130 hrs).

Data Analyses. — ^Each focal tern and the coire-
sponding aggregation survey were treated as
independent samples throughout the data analyses,
(n — 275), rather than considering each aggrega-
tion with four individual terns as an independent
sample (n = 87). This methodology was justified
as intermission periods between observations
allowed terns to redistribute themselves randomly
throughout the space above a pool providing for
Linbia.sed observation of foraging birds. Observa-
tions of focal individuals can be considered
independent it the time lag between observations
is sullicient for an animal to traverse its range
(Swihart and Slade 1985, Minta 1992). The 10-min
interval between tern observations was sufficient
tor a tern to traverse the space above a pool. This
also provided us with a preci.se time of day and
aggregation size for each focal bird. Analyses using
each tern and corresponding suiwey as independent
samples were compared to analyses in which one
tern and one survey from each aggregation were
randomly .selected for analysis. The results of these
random selections were similar in all cases with
those using all four focal individuals from a single

Schreffler cl at. • AGGREGATIONS OF FORSTER'S TERNS

97

FIG. 1. Numbers of Laughing Gulls (open pyramids, dotted line). Snowy Egrets (shaded diamonds, dashed li.^e) and
Forster's Terns (dark boxes, solid line) by aggregation size. The number ot gulls (slope - 0.48) and egrets (slope 0.17)
increased faster than the number of terns (slope = 0.06) as group size increased.

aggregation (Pearson product-moment coirelation,
y = 0.99). Observations were separated into four
time intervals: dawn until 0630, 0631-0730, 0731-
0830, and 0831 hrs EDT until dispersal. Group
sizes were divided into four categories: solitai^
terns; and aggregations with 4 to 50, 5 1 to 100, and
>100 birds. The data were then analyzed using
two-factor completely randomized ANOVAs.

RESULTS

Aggregation Characteristics. — ^Eorster’s Terns
were often observed foraging within aggregations.
Laughing Gulls (44% of the total birds within a
group). Snowy Egrets (15%), and terns (12%)
were the three most abundant species, respective-
ly, in these groups. Total aggregation size was
most significantly influenced by number of
Laughing Gulls ((3 = 0.60), followed by Snowy
Egrets (P = 0.38), and Forster’s Terns (P = 0.20;
stepwise multiple regression: T3.272 “ 1693.95, P
< 0.0001, r = 0.95; Fig. 1 ). The number of terns
in aggregations increa.sed with total aggregation
size, but number of terns in these groups increased
more slowly compared to the number ot gulls and
egrets. Thus, the proportion of total aggregation
size comprised by terns decreased with increasing
group size (/^i,274 “ 33.93, P < 0.01 , r = 0. 1 1, P
= -0.33).

The number of terns participating in feeding
aggregations was influenced not only by total

aggregation size, but by time of day (one-way
ANOVA on tern number across the 4 time
intervals: F3,272 = 2.964, P = 0.03). Terns were
most numerous in aggregations during the early
morning periods from sunrise until 0730 his
(mean number of terns in aggregations ± SE:
10.58 ± 1.17 from sunrise to 0630, 10.11 ±
1.01 from 0631-0730) with mean numbers
declining significantly throughout the remainder
of the morning to 7.72 ± 0.75 during the 0731-
0830 period, and to 5.73 ± 0.99 during the
0831 -dispersal time period (ad hoc linear
comparison within the ANOVA: Fi,272 “
7.798, P < 0.01).

The contribution that terns made to aggrega-
tions was most substantial during the formative
period of the aggregation but then declined (one-
way ANOVA: ^3,272 = 2.55, P = 0.05). Forster's
Terns comprised a high percentage of aggrega-
tions initially (18.43 ± 2.48%), followed by a
decline for the rest of the aggregation's existence
(12.44 ± 1.44% for 0631-0730; 12.43 ± 1.20%
for 0731-0830; 13.66 ± 0.38% for 083 1 -dispers-
al). Rather than a linear decline (ad hoc linear
comparison within the ANOVA: F\,2ii ~ 188, P
= 0.17), the percentage of terns in aggregations
was initially high and then decreased to a
relatively stable lower level until aggregation
dissolution (comparison of initial vs. other 3
intervals: F| 272 6.38, P = 0.01).

98

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. I. March 2010

Dawn - 0630 0631 - 0730 0731 - 0830 0831 - Dispersal

Time of day

FIG. 2. Mean (± SE) number of (A) missed and (B) successful attempts over time by Forster's Terns foraging solitarily
(dark diamonds, .solid line) and within aggregations of: 4 to 50 birds (open boxes, dashed line), 51 to 100 birds (open
pyramids, dotted line), and >100 birds (open diamonds, variably dashed line).

Foraging Strategies. — Both aggregated and
solitary terns frequently dove into pools to capture
prey; however, many of their attempts mis.sed.
Solitary birds tended to make fewer dives overall
than aggregated birds and therefore missed prey
less often than aggregated birds (aggregation size
effect within the two-way ANOVA: =

4.078, P = 0.007; Fig. 2A). The number of
missed attempts did not change significantly for
all birds over time (time of day effect within the
two-way ANOVA: T3.273 = 0.500, P = 0.687),
but there was a significant interaction between

aggregation size and time of day (F3 273 = 1.928,
P = 0.048). Solitary birds tended to miss capture
of prey less than aggregated birds. Terns in small
(4 to 50 total individuals) to moderate-sized (5 1 to
100) groups mis.sed more often than solitary birds
but less often than birds in large aggregations
(>100 birds) and missed more often early in the
morning than later (linear trend comparison
within the ANOVA: F1.273 = 8.562, P =
0.004). Terns in large congested aggregations
frequently missed prey continuing throughout the
morning. Occasional collisions with competitors,

Schrcfficr cl al. • AGGREGATIONS OF FORSTER’S TERNS

99

especially Laughing Gulls, may have been al least
partially responsible for misses by terns in
aggregations.

The number of times that terns were successful
in capturing prey during dives was also influenced
by group size (aggregation size effect within the
two-way ANOVA: ^3273 ~ 3.239, P = 0.023;
Fig. 2B). Solitary terns overall dove relatively
infrequently and had the fewest successful dives.
Terns in small or moderate-sized aggregations
were most successful, while terns in large
aggregations made comparatively few successful
dives (quadratic trend comparison of group size
within the ANOVA: F] 273 ^ 4.673, P — 0.031).
The number of successful dives by terns in
aggregations appeared to decline throughout the
morning, but the trend was not significant (time of
day effect within the two-way ANOVA: F3273 ~
1.234, P = 0.298), nor was the interaction
between time of day and group size {Fg.m “
0.430, P = 0.918).

Terns in aggregations, as a result of diving
more frequently, had higher capture rates than
solitary terns (aggregation size effect within the
two-way ANOVA: ^3,273 = 3.197, P = 0.024;
Fig. 3A). However, aggregated terns had fewer
successful dives later in the day, causing their
capture rates to decline throughout the morning
(linear trend comparison within the ANOVA:

= 9.341, P = 0.002), while capture rates
for solitary birds did not (time of day effect within
the two-way ANOVA: ^3 273 1.463, P = 0.225;

interaction between time of day and group size:
F9.273 = 0.338, P = 0.962).

Capture rates differed for the two strategies, but
capture efficiency was similar for aggregated and
solitarily foraging terns (aggregation size effect
within the two-way ANOVA: F3 273 ~ 1.414, f* =
0.239; Fig. 3B). The capture efficiency for
solitary birds appeared more variable than for
aggregated birds throughout the morning, but
there were no significant differences in either lime
of day for all birds (F3 273 = 2.225, P = 0.086) or
the interaction between group size and time
(F9.273 = 1.478, P = 0.156).

DISCUSSION

Forster’s Terns were frequent participants
within aggregations, contributing significantly to
the overall size of these mixed-species groups.
Terns were less numerous in aggregations than
either Snowy Egrets or Laughing Gulls and, as

mornings progressed and more ol these hi ids
joined aggregations, the terns left.

These observations likely reflect the foraging
strategy of terns relative to wading biids that
comprised these large, congested aggregations.
The pre.sence of birds at a pool during the early
morning lormative period indicated to seaiching
terns that it was a worthwhile patch to exploit
(Armstrong 1970, Master 1992). Foraging
maneuvers of wading birds, especially neai
the shoreline of pools, serve to flush prey into
open water (Kushlan 1978). This maneuvering,
particularly by the actively foraging Snowy
Egrets, has been shown to enhance capture
opportunities for other wading birds in the
group (Russell 1978), and apparently for
Forster’s Terns.

Terns used a “hover-hunt” technique over the
center of the pool, after prey had been flushed. In
this behavior, terns would actively beat their
wings to remain stationary in the air ~7 to 10 m
above the center of the pool’s surface. A tern
would then dive into the water to capture a fish
and, if successful, would fly across the pool while
swallowing its prey until returning to its hovering
position. If unsuccessful, the bird would return to
its hovering position. Hover hunting was most
successful for terns early in the morning, when
both their number of successful dive attempts and
rates of prey capture were highest. These results
seem intuitive as, during this time of the morning,
prey has been beaten from cover but are still
abundant while competing foragers are still low in
number. The total number and proportion of terns
in aggregations increased during the lormative
period reflecting good foraging conditions. This
technique became less successful as time pro-
gressed and aggregations grew in size, prey
density declined and other birds became obstacles
during diving, reducing both the terns’ number ot
successful attempts and capture rates. Previous
studies have shown that foraging activity of
aggregations does impact pool fish populations
by reducing density by 80% on average (Master
1992). Solitary terns, in contrast to aggregated
terns, used a “circle hunt” behavior consisting of
flying a circle around the perimeter of the pool al
a relatively low height of 1 to 3 m. The tern would
dive into the water after spotting a prey item.
Whether successful or not, the bird would rise
from the water and continue its circling flight
until diving again or leaving the pool. This
behavior likely flushed some prey items, but

100

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 1. March 2010

Time of day

FIG. 3. Mean ( ± SE) (A) capture rate and (B) capture efficiency over time for Forster’s Terns foraging solitarily (dark
diamonds, solid line) and within aggregations of: 4 to 50 birds (open boxes, dashed line), 51 to 100 birds (open pyramids,
dotted line), and >100 birds (open diamonds, variably dashed line).

probably not a.s many as the combined effects of
multiple egrets and gulls and, thus, solitary terns
dove less often for prey items, which ultimately
reduced their capture rates. However, solitary
birds also missed less frequently, likely due to
lack of interference from waders and gulls, and
their capture efficiency was similar to that of
aggregated terns.

Ultimately, the tactics of Forster’s Terns reflect
the apparent advantages and disadvantages of
aggregated versus solitary foraging over the course
of the morning. Terns Hew over the salt marsh.

encountering multiple pools that either were or
were not occupied by an aggregation. Terns could
easily join a group or forage alone,, or switch
between the two tactics. Incentives for Joining a
group were high as long as the aggregation was not
so large that competition was high and prey items
had been depleted, conditions that seemed to exist
during the early morning. Terns foraging alone had
fewer opportunities to capture prey, but they were
equally likely to catch fish throughout the day as
their prey had not been over-exploited and they had
far le.ss competition.

Schrcfflcr ei at. • AGGREGATIONS OF FORSTER’S TERNS

101

ACKNOWLEDGMENTS

We thank Roger Wood and The Wetlands Institute, Stone
Harbor, New Jersey for generously providing logistical
support throughout this project, the Biology Departments ot
East Stroudsburg University of Pennsylvania and North-
ampton Community College for providing technical
support, the East Stroudsburg University Foundation for
providing travel funds, and two anonymous reviewers for
helpful suggestions which greatly improved the manuscript.

LITERATURE CITED

Armstrong, E. A. 1970. Social signaling and white
plumage. Ibis 113:534.

B.^LPH, D. F. AND H. C. Romesburg. 1986. The possible
impact of observer bias on some avian research. Auk
103:831.

Bertram, B. C. R. 1978. Living in groups: predators and
prey. Pages 64-96 in Behavioural ecology: an
evolutionary approach (J. R. Krebs and N. B. Davies,
Editors). Blackwell Scientific Publishers, Oxford,
United Kingdom; and Boston, Massachusetts, USA.
Caldwell, G. S. 1986. Predation as a selective force on
foraging herons: effects of plumage color and flocking.
Auk 103:494-505.

Caraco, T. 1981. Risk-sensitivity and foraging groups.
Ecology 62:527-531.

Charnov, E. L. 1976. Optimal foraging: the marginal value
theorem. Theoretical Population Biology 9:129-136.
Collier, G., D. F. Johnson, and J. Berman. 1991. Patch
choice as a function of procurement cost and encounter
rate. Journal of the Experimental Analysis of Behav-
iour 69:5-16.

Grand, T. C. and L. M. Dill. 1999. The effect of group
size on the foraging behaviour of Juvenile coho
salmon: reduction of predation risk or increased
competition? Animal Behaviour 58:443-451.
Hamilton, W. D. 1971. Geometry for the selfish herd.

Journal of Theoretical Biology 31:295-311.

Kushlan, j. A. 1978. Feeding ecology of wading birds.
Pages 249-297 in Wading birds (A. Sprunt IV, J. C.
Ogden, and S. Winckler, Editors). Research Report
Number 7. National Audubon Society, New York,
USA.

Latta, S. C. and J. M. WUNDERLE. 1996. The composition
and foraging ecology ot mixed-species flocks in pine
forests of Hispaniola. Condor 98:595-607.

Lim.a, S. L. 1995. Back to the basics of anti-predatory
vigilance: the group-size effect. Animal Behaviour
49:11-20.

Master, T. L. 1992. Composition, structure, and dynamics
of mixed-species foraging aggregations in a southern
New Jersey salt marsh. Waterbirds 15:66-74.

Master, T. L., M. Frankel, and M. Russell. 1993.
Benefits of foraging in mixed-species wader aggrega-

tions in a southern New Jersey salt marsh. Waterbirds
16:149-157.

Minta, S. 1992. Tests of spatial and temporal interaction
among animals. Ecological Applications 2:178—188.

Moreno, A. B., A. R. Lagos, and M. A. Alves. 2005.
Water depth selection during foraging and efficiency
in prey capture by the egrets Casmerodiiis alhits and
Egrelta tiuilci (Aves, Ardeidae) in an urban lagoon in
Rio de Janeiro State, Brazil. Iheringia. Serie Zoologia
95:107-109.

Pulliam, H. R. and G. C. Millikan. 1982. Social
organization in the non-reproductive season. Pages
1 17-132 in Avian biology. Volume 6 (D. S. Earner and
J. R. King, Editors). Academic Press, New York, USA.

Reed, J. M. 1985. Relative energetics of two foraging
behaviors of Forster’s Terns. Waterbirds 8:79-82.

Rooney, M. B. and J. Sleeman. 1998. Effects of selected
behavioral enrichment devices on behavior of western
lowland gorillas (Gorilla gorilla gorilla). Journal of
Applied Animal Welfare Science 1:339-351.

Russell, J. 1978. Effects of interspecific dominance among
egrets commensally following Roseate Spoonbills.
Auk 95:608-610.

Salt, G. W. and D. E. Willard. 1971. The hunting
behavior and success of Forster’s Tem. Ecology
52:989-998.

Schmitz, R. A. and G. A. Baldassarre. 1992. Correlates
of flock size and behavior of foraging American
Flamingos following Hurricane Gilbert in Yucatan.
Mexico. Condor 94:260-264.

Schoener, T. W. 1971. Theory of feeding strategies.
Annual Review of Ecology and Systematics 2:368-
404.

Spendelow, j. a. and S. R. Patton. 1988. National atlas
of coastal waterbird colonies in the contiguous United
States: 1976-1982. U.S. Fish and Wildlife Service
Biological Report 88 (5). Washington, D.C., USA.

SwiHART, R. K. AND N. A. Slade. 1985. Testing for
independence of observations in animal movements.
Ecology 66: 1176-1184.

Walsh. J., V. Elia. R. Kane, and T. Halliwell. 1999.
Birds of New Jersey. New Jersey Audubon Society.
Barnardsville, USA.

Webster, S. J. and L. Lefebvre. 2001. Problem solving
and neophobia in a columbiform-passeriform assem-
blage in Barbados. Animal Behaviour 62:23-32.

Weinstein. R. B. 1995. Locomotor behavior of nocturnal
ghost crabs on the beach: focal animal sampling and
instantaneous velocity from three-dimensional motion
analysis. Journal of Experimental Biology 198:989-
999.

Ydenberg. R. C., C. V. J. Welham, R. Schmid-Hempel, P.
Schmid-Hempel, and G. Beauchamp. 1994. Time
and energy constraints and the relationships between
currencies in foraging theory. Behavioral Ecology
5:28-34.

The IVilson Journal of OmUhology 122( 1 ): 102-1 1 5, 2010

ABUNDANCE AND DISTRIBUTION OF WATERBIRDS IN THE

LLANOS OF VENEZUELA

FRANCISCO J. VILELLA'-^ AND GUY A. BALDASSARRE-

ABSTRACT. — The Llanos is a significant waterbird site in the Western Hemisphere, but abundance and distribution of
waterbirds across this vast region are poorly known, which hampers conservation initiatives. We used point counts along
road routes in the Llanos region of Venezuela to examine abundance and distribution of waterbirds during 2000-2002
within five ecoregions across the Llanos. We detected 69 species of waterbirds and recorded 283,566 individuals, of which
10 species accounted for 80% of our observations. Wading birds (Ciconiifomies) represented the largest guild both in
numbers of species (26) and individuals (55%), followed by waterfowl (26%), and shorebirds (11%). Five species
comprised 62% of all individuals: Cattle Egret {Biihiilciis ibis). White-faced Whistling Duck (Dendrocygna vidiiata). Black-
bellied Whistling Duck (D. ainiimnalis). Great Egret (Ardea alba), and Wattled Jacana (Jacana jacana). Wading birds were
particularly ubiquitous with at least 21 ot 26 species recorded in each ot the ecoregions. Species richness (66), proportion of
waterbirds detected (54%), and mean number of birds per route (1,459) were highest in the Banco-Bajio-Estero savanna
ecoregion. Our study provides the most comprehensive data set available on waterbirds in the Llanos of Venezuela and
highlights regions of special conservation concern. Received 20 April 2009. Accepted 7 October 2009.

The Llanos is the second largest savanna
ecosystem in South America after the Cerrado of
Brazil, covering some 451,474 km^ of which two-
thirds occur in Venezuela and the remainder in
Colombia (Bibby et al. 1992, Mittermeier et al.
1998). Wetlands in the Llanos are concentrated in
southwest Venezuela and eastern Colombia and
can cover an estimated 107,530 km^ which
makes the Llanos the second largest wetland
complex in South America after the Brazilian
Pantanal (Hamilton et al. 2002).

This vast expan.se of wetlands, along with poor
soils for agriculture and low human population
density, combine to make the Llanos one of the
most significant waterbird habitats in the world
(Scott and Carbonell 1986, Roca et al. 1997,
Mittermeier et al. 1998). Over 80 species of
waterbirds have been recorded in Llanos wet-
lands, of which the major groups are wading
birds, sandpipers and plovers, waterfowl, and rails
(Hilty 2003). Waterbird occurrence in the Llanos
has been documented and general range maps are
available in field guides (Hilty 2003), but detailed
information on abundance and distribution is
lacking and often dated, sparse, or limited in
scope and duration (e.g., Ramo and Busto 1984,
Frederick and Bildstein 1992, Tamisier and
Dehorter 2000).

' U.S. Geological .Survey. Cooperative Re.search Unit,
Department of Wildlife and Fisheries, Mississippi State
University. Mississippi State. MS 39762. USA.

LSUNY College of Environmental .Science and Fore.stry,
Syracuse. NY 13210, USA.

’Corresponding author: e-mail: fvilella@cfr.msstate.edu

Little is known about abundance and distribu-
tion patterns of waterbirds in the Llanos despite its
huge size and significance as waterbird habitat
(Hilty 2003). This lack of information is note-
worthy because the Llanos, after many years of
relative isolation, is experiencing increased hu-
man population growth and accompanying distur-
bance including road construction, water control
activities, and hunting, all of which can affect
waterbird populations (Bisbal 1988, Mittermeier
et al. 1998). Lack of information on waterbird
populations in the Llanos seriously hampers
informed conservation decisions. Our objective
was to conduct a geographically extensive survey
to examine the abundance and distribution of
waterbirds across this vast wetland savanna
landscape.

METHODS

Study Area. — ^The Llanos is a vast sedimentary
plain bordering the Orinoco River and its tributar-
ies. Average annual temperature exceeds 24° C
and rainfall varies from 500 to 4,000 mm/year
(Sarmiento 1983, Mittermeier et al. 1998, Sar-
miento and Pinillos 2001). Precipitation is highly
sea.sonal (May-Sep) with 23% of the Llanos
floodplain susceptible to .sea.sonal Hooding at
maximum inundation. In contrast, only 2,080 knr
of open water remain during the dry season (Nov-
Mar), principally in rivers and lakes (Hamilton et
al. 2002). This interaction of .seasonal rainfall, soil
quality, topography, high temperatures, and fire
promotes savanna vegetation across much of the
Llanos, regardless of seasonal inundation. Moist
gallery forest, dry tropical forest, permanent

102

Vilcllci and Baldassarrc • LLANOS WA TIiRBIRDS

103

wetlands, and agriculture occur throughout the
Llanos and contribute to a complex mosaic ot
habitats across the landscape (Sarmiento 1983,
Huber and Alarcon 1988).

Vegetation and Ecoregions. — Several studies
have delimited Llanos ecoregions based on
vegetation, flooding, and soil type, but these have
not produced universally accepted detinitions or
boundaries (Blydenstein 1967, Ramia 1967,
Sarmiento 1983, Velasco and Ayarzaguena
1985, Huber and Alarcon 1988, Bulla et al.
1990). We reviewed the existing classifications
and selected three well-recognized ecoregions
containing most of the waterbird biodiversity in
the Llanos: Alluvial Overflow Plains, Western
Llanos, and Central Llanos.

The Alluvial Overflow Plains include 7.45 mil-
lion ha of flat savanna grassland with gallery
forest along permanent rivers (Ramia 1967,
Velasco and Ayarzaguena 1985). The region is
subdivided into two areas based on topography
and vegetation: (1) Paspalum Savanna, and (2)
Banco-Bajio-Estero Savanna (Ramia 1967, Bulla
et al. 1990). Paspalum Savanna occurs in low
terrain subject to extensive flooding by rivers and
is dominated by Paspalum fasciculatum. Banco-
Bajio-Estero Savanna (hereafter, BBE) borders
Paspalum Savanna (hereafter, Paspalum), but
topographical variations of 1—2 m create consis-
tent differences in inundation patterns. This
savanna is recognized as “banco-bajio-estero”;
literally bank, slope, and marsh. Esteros are the
lowest areas, bancos are higher areas of former
riverbanks, and bajios are intermediate sites.
Esteros comprise —80% of the region (Sarmiento
1983, Bulla et al. 1990). The BBE includes an
area of about 1 million ha of man-made impound-
ments (modulos). These impoundments increase
maximum flood level of the savannas an average
of 50 cm and prolong inundation by nearly
3 months (Bulla et al. 1990).

The Western Llanos ecoregion (5.15 million
ha) is characterized by fertile soils and partial
flooding (Huber and Alarcon 1988, Ramo and
Busto 1988). Historically this region was
deciduous tropical forest interspersed by savan-
na; the remaining forest fragments are now
largely in reserves (Veillon 1976). Irrigation has
allowed this area to become the largest rice-
producing region in Venezuela (Ministerio de
Agricultura y Cna 1998). We divided our
sampling in this ecoregion into (1) Ricefields
and (2) Western Llanos.

The Central Llanos ecoregion (7.15 million ha)
consists ot rolling grasslands dominated by
Traclivpogon spp. This ecoregion includes areas
of chaparral dominated by the tree species
Curatella ainericana, Bowdichia virgilioides, and
Byrsonima crassifolia. Gallery lorest, Moriche
palm {Maiuitia flexuosa) swamps, and patches of
the Orinoco palm (Copernicia tectornm) occur
where standing water persists; patches ol tropical
deciduous forest occurs at higher elevation
(Blydenstein 1962, Ramia 1993, Huber and
Alarcon 1988, Ramo and Busto 1988). Rice fields
have expanded into the Central Llanos (Ministerio
de Agricultura y Cria 1998).

Waterbird Surveys. — We used point counts
along roadsides (Eig. 1 ) to survey waterbirds
because roadside surveys can efficiently assess
community composition and habitat associations of
birds in open habitats and habitat mosaics across
large landscapes (Thiollay 1978, Ellis et al. 1990,
Hanowski and Niemi 1995). Detection of waterbird
species in open wetlands may actually be greater
from roads than off-road (Hanowski and Niemi
1995). Conway and Simon (2003) also noted that
point counts along transects increase detection of
less conspicuous species. Disadvantages of road
count methods are recognized (Buckland et al.
2004), but we believe these limitations were not
significant in our study because wetland habitats in
the Llanos occur continuously along both roadsides
and roadless areas.

We sampled each ecoregion based on rainfall
patterns that occur in four distinct seasons, each
about 3 months in duration: (1) Late Wet, July-
September; (2) Early Dry, October-December; (3)
Late Dry, January-March; and (4) Early Wet.
April-June (Bulla et al. 1990). The hydrological
extremes of these seasons were the Late Wet.
which is characterized by heavy raintall and
maximum annual inundation, whereas the Late
Dry is characterized by drought conditions and
minimal inundation. Ecoregions were visited in
the same order each season (Western. Ricefields.
Central, Paspalum, and BBE), and routes were
surveyed within 7 days of the calendar date
surveyed each year.

An initial sample of 30 routes established in
July 2000 was increased to 54 by year's end as our
familiarity with the Llanos improved (Table 1).
Each route was 22.5 km with 16 point-count
locations spaced at L5-km intervals. A road was
considered for route placement if at least 30 km
was accessible year-round, and it was separated

104

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 1. March 2010

Caribbean Sea

Q

River

•oo

s Kilometers

FIG. 1 . Area of the Llanos of Venezuela surveyed for waterbirds. State boundaries, major cities, and route locations as
referenced by GPS are included.

from another route by at least 5 km. The first
point-count location was randomly selected with-
in the first 5 km of a ,selected road. We used a
Garmin 12 Global Positioning System (GPS) unit
to identify our first point location. Road traffic
was light to none and we did not believe traffic
affected bird detection.

Our point count methodology was based on the
North American Breeding Bird Survey (Robbins
et al. 1986, Droege 1990, Peterjohn 1994) as
modified by local conditions. Counts began
shortly after sunrise and were usually completed
by noon. A two-person team recorded all
waterbirds seen or heard for 3 min within a 500-

TABLE 1. Number of route.s sampled per season in each ecoregion of the Llanos, Venezuela.

Ecoregion

Fieldwork periods

Sea.son

Ceniral

Weslem Llanos

Alluvial Overllow Plains

Totals

Central

Ricefields

Western

Paspalum

BBE

14 Aug-3() Sep 2000

Late Wet

3

6

6

8

7

30

1 Nov- 15 Dec 2000

Early Dry

3

6

7

10

7

33

14 Feb-31 Mar 2001

Late Dry

4

6

7

12

14

43

14 May-30 Jun 2001

Early Wet

4

8

14

14

14

54

14 Aug-30 Sep 2001

Wet

4

8

14

14

14

54

1 Nov- 15 Dec 2001

Early Dry

4

8

14

14

14

54

14 Feb-31 Mar 2002

Dry

4

8

14

14

14

54

14 May-30 Jun 2002

Early Wet

4

8

14

14

14

54

I 'ilclla and Balda.s.sarre • LLANOS WAIERBIRDS

105

m radius on each side of the road (6 min/point) as
measured by rangefinder binoculars; we included
birds in flight but tried to minimize double-
counting. We selected the observation duration
based on a January 2000 pilot study, and all
observers were trained in species identification
before surveys began. Observations were made
using 10 X 42 binoculars and 20X spotting
scopes. Common and scientific names follow Gill
et al. (2009).

Data Analysis. — We selected 69 waterbird
species to survey and classified these into six
guilds based on feeding niche, habitat preference,
and taxonomy: (1) wading birds, (2) waterfowl,
(3) shorebirds, (4) open water, (5) forest, and (6)
passerines. The Open Water guild comprised 13
piscivorous species requiring deep water for
feeding: Least Grebe (Tachybaptiis dominicus).
Neotropic Cormorant {Phalacrocorax hrasilia-
nus), Anhinga {Anhinga anhinga), Sungrebe
(Heliovnis fulica), terns (Laridae, 4 species); and
kingfishers (Alcedinidae, 5 species). The Wading
Birds guild (26 species) comprised long-legged
species that occur in or close to water and forage
on fish, amphibians, and/or invertebrates. These
included herons, egrets, and bitterns (Ardeidae, 14
species), ibises (Threskiornithidae, 8 species),
storks (Ciconiidae, 3 species), and Limpkin
(Aramus guaraima). The Waterfowl guild (10
species) included ducks (Anatidae, 8 species) and
gallinules (Rallidae, 2 species). The Shorebird
guild (12 species) included sandpipers (Scolopa-
cidae, 8 species) as well as Double-striped Thick-
knee (Burhinus histriatus). Collared Plover (Char-
adriiis collaris). Black-necked Stilt (Himantopus
mexicanii.s), and Wattled Jacana (Jacana jacaiia).
The Forest guild (four species) included Hoatzin
{Opisthocomus hoazin), Grey-necked Wood Rail
{Aramides cajanea), Sunbittern (Eurypyga helias),
and Homed Screamer (Anhima corniita). The
Passerine guild included species associated with
water: Pied Water Tyrant (Fluvicola pica). White-
headed Marsh Tyrant (Arundinicoia ieitcoce-
phala). Black-capped Donacobius {Donacohius
atricapilia), and Red-capped Cardinal (Paroaria
gidaris).

We assessed general patterns of waterbird
abundance by summarizing the total numbers of
birds recorded by species and guild, ranking each
in order of abundance, and calculating percent
relative abundance for each species. We also
expressed abundance as the percentage of routes
and points where we detected each species. We

TABLE 2. Number of species and total number of birds
recorded for each guild in the Llanos, Venezuela.

Guild

# .species

% lolal species

Total.s

% total

Wading birds

26

37.7

154,450

54.5

Waterfowl

10

14.5

74,335

26.2

Shorebirds

12

17.4

31,277

11.0

Open water

13

18.8

17,286

6.1

Forest

4

5.8

3,280

1.2

Passerines

4

5.8

2,938

1.0

Totals

69

100.0

283,566

100.0

standardized for sampling effort in each ecoregion
by calculating the mean number of each species
per route per field season. We averaged mean
values per sample period to derive a grand mean
per route for each species or guild in each
ecoregion. We used a species accumulation curve
to evaluate the relationship between sampling
effort and species detection over the survey period
(Colwell 2006).

We grouped data from both years and used t-
tests to compare species abundance in each
ecoregion and season between years; only 3.2%
of the tests were significant and differences were
distributed among all seasons and ecoregions. We
calculated the mean number of individuals per
route to combine data between years when routes
were sampled in both fieldwork periods, and
retained those routes added in the second period
or near the end of the first. This combination
standardized sample effort in each ecoregion. We
compared seasonal differences in number of birds
by guild in each ecoregion using analysis of
variance. We used SPSS 11.0.0 for all statistical
analysis, and accepted significance at the 0.05
level (SPSS 2001).

RESULTS

General Patterns. — We recorded 283,566 indi-
viduals from 69 species of waterbirds in 23 families
(Appendix). The waterbird community was dom-
inated by wading birds, which comprised 54.5% of
the individuals and 37.7% oi the species (Table 2).
Waterfowl were second in number of individuals
(26.2%), followed by shorebirds (1 1.0%). Howev-
er, waterfowl only represented 14.5% of all
species. The Open Water guild ranked second in
number of species ( 1 8.8%) but fourth in abundance
(6.1%). The Passerine and Forest guilds contained
the fewest species (4) and individuals (1.0-1. 2%).
The species accumulation curve indicated sam-
pling effort over the 2-year survey period was

106 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. I, March 2010

FIG. 2. Species accumulation curve for waterbird species detected during 2000-2002 surveys in the Llanos of Venezuela.

adequate for detecting most species in the study
area (Fig. 2). Approximately 10 additional routes,
surveyed seasonally for 2 years, would be needed
to detect one additional species.

Waterbird abundance in the Llanos was dom-
inated by a small number of species; most species
occurred infrequently. Only five species com-
prised 62% of all individuals recorded: Cattle
Egret (Buhulcus ihis). White-faced Whistling
Duck (Dencirocygiia viditata). Black-bellied
Whistling Duck (D. autumnalis). Great Egret
{Arclea alha), and Wattled Jacana. Ten species
comprised 80% of all individuals recorded, and 17
species comprised 90%. The Cattle Egret was the
most abundant species (23.7%) and was at 97.5%
of transect sample points. The second most
abundant species was the White-faced Whistling
Duck (14.7%), followed by the Black-bellied
Whistling Duck (10.1%). The three most abun-
dant species collectively accounted for 48.5% of
all detections. Great Egret (8.0%) and Wattled
Jacana (5.9%) were much less common. We
recorded <50 individuals for 14 of the 69 species
detected (20.3%), including four species repre-
sented by a single observation; Sungrebe, Green-
and-rufous Kingfisher (Chloroceryle iiu/a), Amer-
ican Pygmy Kingfisher (C. acnea), and Least
Bittern (l.xohrycluis e.xilis).

Eorty-one (59%) of the 69 species occurred in
every ecoregion and on >50% of routes; 20
species (29%) occurred in 3^ ecoregions, and 8
(12%) occuiTed in only 1-2 ecoregions. Seven-
teen species occurred on 90-100% of all routes,
but 15 occurred on <20% of routes. Wading
birds, including the Double-striped Thick-knee,
were especially ubiquitous with 17 of 26 species
(65.4%) occLiiTing on >70% of all routes.

Routes in the BBE had the greatest water cover,
ranging from 15 to 19% in the Late Dry season to
54-55% in the Late Wet. The Western Llanos had
the lowest water cover, ranging from 4 to 5% in
the Late Dry season to 1 1-12% in the Late Wet
(Table 3).

The BBE contained the highest proportion of
sample points among ecoregions with small,
excavated ponds along the roadside (prestamos,
63%), followed by Paspalum (45%), Western
(24%), Central (34%), and Ricefields (7%).

Species Richness and Abundance by Ecore-
gion.— Most waterbirds were recorded in the BBE
(54%), followed by Ricefields (17%), whereas the
Central had the fewest (3%, Table 4). The mean/
route provides a less biased comparison among
ecoregions because it standardized our sampling
effort for an uneven number of routes per ecoregion
(Table 5). The BBE had the highest mean number

TABLE 3. Mean (%) water cover at each point count during each fieldwork season and in each ecoregion of the
Llanos. Water coverage was estimated within 200 m of each point count. No data were collected during the Late Wet
season 2001.

Ficorogion

l,alc Wci
(2000)

Ciirly Dry
(2000)

I.ale Dry
(2001)

tuirly Wel
(2001)

Late Wcl
(2001)

Early Drv
12001)'

(.ate I2ry
(2002)

Early Wet
(2002)

Central

1 1.7

15.0

9.9

9.0

1 1.7

7.4

12.5

Ricefields

17.4

27.8

12.6

13.8

14.5

9.7

12.9

Western

1 1.1

14.0

5.4

7.8

9.4

4.0

1 1.8

Paspalum

46.9

44.1

1 1.4

22.4

21.2

10.5

25.8

BBE

54.6

44.0

14.8

29.4

30.9

18.5

54.3

MleUa (111(1 liaUlassarrc • LLANOS WAT ERBIROS

107

TABLE 4. Number of wulerbirds within guilds in
surveyed over two field seasons.

each eeoregion

sampled in the Llanos,

Venezuela,

based on all routes

Guild

Cenlnil

RiceHelds

WL-slern

Paspaluni

BBE

Toials

Wadins birds

6,1 15

36,330

23,944

29,554

58,507

154,450

Waterfowl

378

755

2,280

3,574

67,348

74,335

Shorebirds

449

10,085

1,518

6,045

1 ,005

31,227

Open water

44

703

225

4,153

12,161

17,286

Forest

152

23

591

1,1 10

1,401

3,280

Passerines

211

281

536

905

1,005

2,938

Totals

7,349

48,177

29,094

45,341

153,605

283.566

Percent total

2.6

17.0

10.3

16.0

54.2

TABLE 5. Abundance of waterbirds expressed as mean (± SE) number of individuals per route in each guild for each
eeoregion. Relative percentage (Rel. %) of each mean for each guild among (Rel. %-A) and within (Rel. %-W) ecoregions
and species richness are included.

Guild

Central

Ricefietds

Western

Paspalum

BBE

Wading birds

Mean ± SE

200.6 ± 29.0

617.2 ± 78.4

258.9 ± 25.8

289.8 ±

10.2

574.2 ± 50.9

Rel. %-A

10.3

31.8

13.3

14.9

29.6

Rel. %-W

82.4

76.9

82.1

65.9

39.4

Richness

23

21

25

25

26

Waterfowl

Mean ± SE

13.6 ± 4.6

1 1.9 ± 3.1

24.9 ± 4.2

33.4 ±

8.5

611.0 ± 240.6

Rel. %-A

2.0

1.7

3.6

4.8

87.9

Rel. %-W

5.6

1.5

7.9

7.6

41.9

Richness

6

7

7

7

10

Shorebirds

Mean ± SE

14.9 ± 4.6

157.7 ± 48.1

15.9 ± 1.0

57.5 ±

5.6

126.6 ± 14.9

Rel. %-A

4.0

42.3

4.3

15.4

34.0

Rel. %-W

6.1

19.6

5.0

13.1

8.7

Richness

9

9

7

10

12

Open water

Mean ± SE

1.6 ± 0.3

11.1 ±7.0

2.5 ± 0.1

39.2 ±

1.6

122.5 ± 13.8

Rel. %-A

0.9

6.3

1.4

22.2

69.2

Rel. %-W

0.7

1.4

0.8

8.9

8.4

Richness

5

6

10

1 1

12

Forest

Mean ± SE

5.8 ± 1.6

0.4 ± 0.1

7.6 ± 1 .2

10.8 ±

1.9

13.8 ± 4.7

Rel. %-A

15.1

1.0

19.8

28.2

36.0

Rel. %-W

2.4

0.1

2.4

2.5

0.9

Richness

5

6

10

1 1

10

Passerines

Mean ± SE

6.8 ± 1.9

4.9 ± 0.6

5.5 ± 0.7

9.0 ±

1.2

10.5 ± 1.3

Rel. %-A

18.5

13.4

15.0

24.5

28.6

Rel. %-W

2.8

0.6

1.7

2.1

0.7

Richness

4

4

4

4

4

All guilds

Mean ± SE

243.3

803.2

315.3

439.7

1,458.6

Richness

51

49

57

61

66

108

THE WILSON JOURNAL OL ORNITHOLOGY • Voi 122, No. I, March 2010

of waterbirds per route (1,459/route), and the
Central contained the lowest (243/route). Wading
birds were the most abundant guild per route in all
ecoregions (201-617/route) except the BBE, where
waterfowl were most abundant (61 1/route). Wading
birds comprised 39-82% of the mean/route in each
guild. Shorebirds were most abundant in Ricefields
(158/route) and BBE (127/route) but comprised
<20% of all waterbirds detected within an
ecoregion. Open Water species were concentrated
in the BBE (123/route) and Paspalum (39/route),
but were < 1 1/route in other ecoregions and
comprised <9% of all waterbirds within a guild.
Eorest and Passerine species were not abundant
(< 14/route) or proportionally substantial (<3%) in
any ecoregion.

Overall species richness was greatest in the
BBE (66) and least in the Central and Ricefields
(Table 5). Wading birds were well represented in
all ecoregions, but only the BBE included all 26
species. The fewest wading birds (21) were
recorded in Ricefields. Species richness of the
Open Water, Waterfowl, and Shorebird guilds
showed greater variation among ecoregions than
the Wading Bird guild. Eor example, only six of
13 species in the Open Water guild occuired in the
Central and Ricefields ecoregions.

Waterhird Guilds (Wading Birds). — ^We record-
ed 154,450 wading birds with this guild dominat-
ed by herons (14 species), followed by ibises (8),
and storks (3). The herons included two (Cattle
Egret, Rank 1; and Great Egret, Rank 4) of the 10
most abundant species. The herons also included 4
species ranked in the top 20: Snowy Egret
(Egretta thida), Cocoi Heron (Ardea cocoi). Little
Blue Heron (£. caeruleu), and Whistling Heron
(Syrigma sihilaOix). Three species were rarely
recorded: Least Bittern, Green Heron (Biitorides
virescens), and Pinnated Bittern (Botaiirus pinna-
tus). Overall, herons were widespread, as 1 1
species occurred on >50% of routes, 1 2 species
occurred in all five ecoregions, and eight occiured
on >40% of tran.sect sample points.

The ibis group was also widespread and
frequently encountered. Three species ranked in
the top 10 by abundance: Bare-faced Ibis
(Phinwsiis infuscatiis, Rank 6), Glossy Ibis
(Plegadis falciiudliis. Rank 7), and Scarlet Ibis
(Eudocimus ruhei\ Rank 9). Ibi.ses occurred on
57-100% of all routes and in all ecoregions with
the exception of the Sharp-tailed Ibis (Ccrcihis
o.xycerca), which was absent from the Ricefields.

The Wood Stork (Mycteria americana) was

most abundant; it represented only 2.6% of the
wading bird guild but 74.6% of all storks and
occLin'ed on 70% of all routes. The Wood Stork
also ranked 12th in overall abundance. The
Maguari Stork (Ciconia maguari) occurred on
only 43% of routes, and the Jabiru (Jahini
mycteria) occuiTed on only 35%.

Waterhird Guilds (Waterfowl). — We grouped
10 species, including the two species of gallinules
into the waterfowl guild. However, 95.1% of the
74,076 individuals recorded consisted of two
species: White-faced Whistling Duck and Black-
bellied Whistling Duck. Only the Purple Gallinule
(Porphyrula martinica) and Brazilian Teal (Ama-
zonetta hrasiliensis) of the remaining eight species
exceeded 1% of the guild total. Blue-winged Teal
(Anas discors), the only migratory species of
waterfowl to reach the Llanos from North
America, were rarely recorded (196 individuals).

The waterfowl guild generally was not widely
distributed. Only the whistling ducks and galli-
nules were recorded in all ecoregions and on
>50% of routes. Lour species were concentrated
in only one ecoregion (the Alluvial Overflow
Plains): Orinoco Goose (Neochen juhata), Brazi-
lian Teal, White-faced Whistling Duck, and
Black-bellied Whistling Duck.

Waterhird Guilds (Shorehirds). — ^The Wattled
Jacana was the most abundant of the 12
shorebirds, accounting for 53.6% of the 31,277
birds recorded in the guild and ranked 5th in
overall abundance; the Black-necked Stilt was the
second most abundant shorebird (#12). Two
species occurred on >90% of routes (Wattled
Jacana and Double-striped Thick-knee), but most
shorebirds were more patchily distributed (8 of 12
occurred on <50% of routes), and not abundant.
The South American Snipe (Gallinago para-
guaiae) was the least frequently encountered
shorebird with only eight individuals.

Waterhird Guilds (Open Water). — We recorded
13 species (17,286 individuals) in this guild, of
which the Neotropic Cormorant was the most
frequently recorded species (61.1%), followed by
the Large-billed Tern (Pliaetusa simple.v, 18.8,%).
The Ringed Kingfisher (Megaceryle torcpiata) was
the most abundant kingfisher (3.7%). Swimming
species (Neotropic Cormorant, Anhinga, Least
Grebe, Sungrebe) were more than twice as
common as terns and kingfishers combined.

Open water species were primarily recorded in
the BBE (69.2%); 22.2% of all observations
occurred in Paspalum. The Neolropic Cormorant

Vilelhi and BaUlassarre • LLANOS WATERBIRDS

109

accounted for >60% of observations, but all guild
species were more abundant (>75% of observa-
tions) in these two ecoregions with the exception
of the Least Grebe, which was recorded on only
five routes, four of which were in the Western
Llanos. Five species were recorded <100 times:
Sungrebe, Green-and-rufous Kingfisher, Ameri-
can Pygmy Kingfisher, Least Grebe, and Green
Kingfisher {Chloroceryle americana). We only
encountered one Green-and-rufous Kingfisher and
one American Pygmy Kingfisher.

Waterbird Guilds (Forest). — ^The Hoatzin was
the most abundant species in this guild, repre-
senting 94.2% of the 3,280 individuals recorded.
All species were observed in at least four
ecoregions, but distribution was patchy (<50%
of routes) and lowest in the Ricefields ecoregion.

Waterbird Guilds (Passerines). — We recorded
only 2,938 individuals for the four species in this
guild, of which the Pied Water Tyrant was most
abundant (46.6%). All species were widespread
(>80% of routes) and occurred in all ecoregions.

Temporal Patterns Within Ecoregions. — We
recorded fewer waterbirds in the first fieldwork
period (101,146) than the second (182,422), but
combined data by seasons because seasonal
differences between fieldwork periods generally
were not significant (Table 6). Overall, we
recorded more species and individual birds in
the Late Dry season of the second fieldwork
period. The proportions of waterbirds also dif-
fered between years for the Late Wet season in the
BBE (P = 0.031) and the Early Dry season in the
Central {P = 0.015).

Wading birds were the most abundant guild,
representing >50% of all detections in all
ecoregions and all seasons, except for the BBE
in the Late Dry season of the second fieldwork
period. There were seasonal changes in mean/
route in three ecoregions (F = 2.84; df = 4, 49; P
= 0.056), primarily caused by a decrease in Cattle
Egrets. Wading bird numbers in the Western
ecoregion were lowest in the Late Wet season
(230/route) before increasing in the Early Wet
season (F = 8.13; df = 4, 49; P < 0.001).

Wading bird numbers in the BBE were highest
in the Late Dry season (846/route) and lowest in
the Early Wet season (383/route; F = 3.85; df =
4, 49; P = 0.015). This seasonal .shift in wading
bird distribution between the Western and BBE
ecoregion was most obvious for five ibis species:
Scarlet, American White Ibis (Eudocinnis albus).
Sharp-tailed, Glossy, and Bare-faced Ibis.

Waterfowl were most abundant in the BBE (F
= 22.01; df = 4, 49; P < 0.001) where we
recorded >90% of all individuals, but detections
in each ecoregion showed significant seasonal
patterns. Eor example, detections in the BBE in
the Eate Dry season increased to nearly 2,000/
route (F = 8.06; df = 3, 52; P < 0.001), because
of seasonal aggregations of White-faced Whis-
tling Ducks and Black-bellied Whistling Ducks.
Waterfowl detections were lowest (P < 0.003)
outside the BBE in the Late Dry season and
increased in the Early Wet season.

The BBE was most important for shorebird
abundance (F = 9.45; df = 4, 49; P < 0.001 ) with
counts exceeding 70/route in all seasons. Howev-
er, Ricefields exceeded this average in the Late
Wet season, when mean route totals were 352/
route with the airival of Nearctic migrants (mostly
Calidris and Tringa species). Off-route observa-
tions of Calidris species exceeded 2,000 individ-
uals, but large flocks were infrequently encoun-
tered along routes. Migrants accounted for 87% of
all shorebirds recorded during the Late Wet
season, but this proportion decreased to 40% in
the Early Dry season, as migrants dispersed.

The seasonal dynamic of shorebird abundance
in the Alluvial Overflow Plains was driven mainly
by detections of the Wattled Jacana. Eor example,
numbers of Wattled Jacanas recorded in the
Ricefields accounted for <5% of all shorebirds,
but increased to >70% in the Western, Paspalum,
and BBE. Wattled Jacanas (> 100/route) were
recorded in the BBE during the dry season,
although the mean number of detections for
shorebirds increased to > 200/route in the Eate
Dry season with amval of Black-necked Stilts.

Seasonal variation in the mean count per route
for Open Water birds was not significant in any
ecoregion except the Ricefields (F = 3.20; df = 3.
28; P — 0.039), where numbers peaked at 41
birds/route in the Late Wet season. This differ-
ence potentially resulted from flooding of the Rio
Portuguesa, which was at the southern limit of the
agricultural region and sampled with a single
route. Over 500 Neotropic Cormorants were
attracted to these temporary conditions, which
raised the mean route count for the season to
almost lOO/route. Excluding this point reduced the
average abundance of the Open Water guild in the
Ricefields to <1 /route in any season. Several
species within the Open Water guild exhibited
distinct seasonal trends. Most notably, the Yel-
low-billed Tern (Sternula superciliaris). Gull-

110

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. I. March 2010

TABLE 6. Waterbirds (mean/route ± SE) detected across seasons for each guild and ecoregion. Means were calculated
trom combined data over both fieldwork seasons. Means for each guild within ecoregions that differed across seasons (P <
0.05) are denoted by lower case letters (one-way ANOVA and Waller-Duncan post-hoc comparisons).

Central Llanos ecoregion (« = 4)

Late Wet

Early Dry

Late Dry

Early Wet

Guild

Wading birds

148.5 ± 26.9

269.5 ± 130.6

156.9 ± 54.9

227.5 ± 99.0

Waterfowl

21.6 ± 12.6'’

1.6 ± 0.6“

2.0 ± 1.7“

29.3 ± 9.8"

Shorebirds

28.1 ± 7.2'’

9.4 ± 2.7“-'’

14.5 ± 5.5“-'’

8.0 ± 2.4"

Open water

2.1 ± 0.8

1.3 ± 0.6

1.8 ± 0.7

1.0 ± 0.5

Forest

4.5 ± 4.2

5.0 ± 4.0

10.5 ± 10.3

3.3 ± 3.3

Passerines

10.3 ± 2.5'’

4.1 ± 1.3“

3.0 ± 0.5“

9.8 ± 1.7"

Ricefield.s ecoregion (n = 8)

Late Wet

Early Dry

Late Dry

Early Wei

Guild

Wading birds

321.3 ± 1 10.7“

858.7 ± 243.9"

635.0 ± 187.6“'’

653.9 ± 158.0“

Waterfowl

17.8 ± 5.1'’

20.0 ± l.T

0.4 ± 0.3“

9.6 ± 4.8"

Shorebirds

351.9 ± 239.5

147.9 ± 62.3

86.4 ± 37.5

44.7 ± 12.3

Open water

41.0 ± 38.2'’

2.4 ± 1 .5“-'’

0.4 ± 0.2“

0.6 ± 0.2“

Forest

0.2 ± 0.2

0.3 ± 0.3

0.5 ± 0.4

0.5 ± 0.4

Passerines

6.3 ± 1.5'’

3.6 ± 1.0“

2.6 ± 1.4“

7.1 ± 1.9"

Western Llanos ecoregion (/? = 14)

Late Wet

Early Dry

Late Dry

Early Wet

Guild

Wading birds

230.2 ± 24.4'’

238.3 ± 34.5'’

170.6 ± 38.4“

396.4 ±41.2“

Waterfowl

34.5 ± 10.6'’

8.7 ± 2.7“

14.4 ± 10.9“

41.8 ± 6.4"

Shorebirds

13.2 ± 1.9“'’

12.4 ± 3.3“

20.5 ± 3.8“

17.8 ± 3.5"-“

Open water

3. 1 ± 0.7

2.8 ± 0.7

1 .9 ± 0.4

2.2 ± 0.5

Forest

5.1 ± 2.8

4.7 ± 2.1

13.9 ± 7.9

6.9 ± 3.2

Passerines

7.0 ± 0.9'’

3.5 ± 0.6“

3.4 ± 0.7“

8.3 ± 1.3"

Paspalum ecoregion (« = 14)

Late Wet

Early Dry

Late Dry

Early Wet

Guild

Wading birds

240.5 ± 38.7

301.8 ± 58.7

332.2 ± 1 18.3

284.8 ± 53.5

Waterfowl

36.7 ± 13.3'-’

14.2 ± 6.0'’

5.5 ± 4.6“

77.0 ± 32.3“

Shorebirds

41.2 ± 10.7“-'’

38.0 ± 1 1 .5“

76.4 ± 14.0“

74.6 ± 15.0"“

Open water

36.2 ± 10.5

42.5 ± 12.2

45.4 ± 17.3

32.6 ± 10.6

Forest

4.1 ± 1.4

6.3 ± 1.9

20.8 ± 1 1.0

12.0 ± 4.1

Passerines

12.9 ± 1.3'’

4.8 ± 1 .0“

5.4 ± 0.9“

13.0 ± 1.2"

BBE (/I =

14)

Late Wet

Early Dry

Late Dry

Early Wet

Guild

Wading birds

582.5 ± 149.6“'’

640.4 ± 87.8'’'^'

845.8 ± 143.6“

382.6 ± 46.9“ '

Waterfowl

165.5 ± 51.0^

96.9 ± 43.4“

1.958.7 ± 1.254.2“

223.0 ± 42.7"-“

Shorebirds

71.8 ± 13.2“

133.0 ± 33.9“'’

201.2 ± 52.4'’

100.5 ± 18.8“

Open water

93.5 ± 39.7

187.9 ± 89.2

137.6 ± 69.7

71.2 ± 26.6

Forest

2.2 ± 1.2

7.8 ± 3.9

39.6 ± 2 1 .6

5.5 ± 2.9

Passerines

15.3 ± 2.2'’

5.4 ± 0.9“

7.3 ± 1.3“

13.8 ± 1.9"

I 'ilella ciihI Bcilclassanc • LLANOS WATER BIRDS

billed Tern {Gelochelickm nilotica), and Black
Skimmer (Rynchops niger) were dry season
visitors to the BBE. Anhinga and Ringed
Kingfisher also became less abundant in the Late
Dry season (P < 0.05).

The mean per route for the Forest guild was
greatest in all ecoregions at the end of the dry
season, peaking at 40/route in the BBE as flocks of
Hoatzin became more common. Counts during the
dry season also increased for the Sunbittern but
seasonal differences in abundance were not signif-
icant in any ecoregion (P > 0.05). Passerines were
most abundant in the BBE and Paspalum (P = 6.00;
df = 1, 4; P < 0.001) but seasonal abundance was
similar among ecoregions. The fewest passerines
were detected in all ecoregions during the Early and
Late Dry seasons (P < 0.052).

DISCUSSION

The Llanos support a diverse and abundant
community of wetland dependent birds. We
emphasize three major findings: (1) the waterbird
community is dominated by wading birds, (2)
waterbirds are not homogeneously distributed
across the Llanos ecosystem, and (3) compara-
tively few species comprised the majority of
individuals recorded. The savanna wetlands of the
Pantanal of Brazil are similarly dominated by
wading birds (Figueira et al. 2006). Like the
Pantanal, waterbird communities of the Llanos of
Venezuela were characterized by habitat general-
ists. Specialization is a function of the number of
habitats used (Rotenberry and Wiens 1980). The
waterbirds of the Llanos, both residents as well as
seasonal species, generally occurred in a variety
of wetland types within and between ecoregions.
This may reflect the ecological flexibility and/or
similarities among these habitats regarding struc-
ture and food availability. This flexibility may
reflect the spatiotemporal habitat heterogeneity of
the Llanos and may be fundamental to maintain-
ing local diversity of waterbirds.

We found significant spatial and temporal
heterogeneity across this vast savanna ecosystem.
Specifically, 70% of all waterbirds recorded
occurred in the Alluvial Overflow Plains (BBE
and Paspalum); Ricefields ranked a distant second
at 17%. The BBE, within the Alluvial Overflow
Plains, contained the most waterbird detections
per route (1,459) and was especially important
during the dry seasons. The wettest ecoregion
during all seasons was the Alluvial Overflow
Plains averaging 1 1-55% water cover at each

1 1 1

sample point during each field season. The BBE
was especially wet during the dry seasons (11-
44%), even exceeding water cover of the Rice-
fields (10-27%). Overall, 63% of all half-points
sampled in the BBE and 45% in the Paspalum
contained permanent water in the lorm ol
prestamos along the roadside versus 7% in
Ricefields, 24% in the Western Llanos, and 34%
in the Central. Many species, especially in the dry
seasons, shift distributions in response to these
hydrological patterns, concentrating in the Allu-
vial Overflow Plains. The BBE contained the
highest relative percentage of each waterbird
guild, while the Paspalum usually ranked second
or third (Table 5).

Only a few of the 69 waterbird species
dominate the abundance of waterbirds we en-
countered in the Llanos. Nearctic migrants had
marked seasonal patterns of abundance. We
recorded only 196 Blue-winged Teal during our
2-year study, despite Venezuela’s recognition as
an important wintering site for this species
(Botero and Rusch 1988). Most migrant water-
birds travel and winter along the Caribbean coast
of Venezuela with smaller numbers in the inland
wetlands (Rodner 2006). Recent surveys have also
highlighted the importance of the northern
Orinoco Delta of Venezuela as migratory shore-
bird habitat (Rodner 2006). The BBE was the
most important ecoregion for migrant shorebirds
in the Llanos (Thomas 1987). Numbers increased
in the Ricefields during the Late Wet season.

Waterfowl were mainly concentrated in the
BBE and Ricefields ecoregions with decreasing
detections in all other ecoregions sampled. Only
Black-bellied and White-faced whistling ducks
were recorded in all ecoregions. Waterfowl
exhibited a distinct relationship between seasonal
patterns of abundance in different ecoregions.
which generally reflected the transition from the
Late Dry season to the Early Wet season.
Waterfowl hunting is permitted during the breed-
ing season for some species, particularly whistling
ducks, which are considered serious rice crop
pests (Sanz-Agreda 1998). Our surveys recorded
dramatically reduced numbers of whistling ducks
in the Llanos compared to estimates reported by
Gomez-Dallmeier and Cringan (1989).

CONSERVATION IMPLICATIONS

Waterbirds tended to spread across the eco-
regions during wet seasons but contracted to the
Alluvial Overflow Plains in dry seasons. We

112

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. I. March 2010

believe this ecoregion should be a focus of
conservation efforts, because it contains extensive
water coverage during the dry season, and
supports the greatest number and diversity of
Llanos waterbirds. The dominant land use in the
Llanos is cattle ranching, which at current levels
is compatible with sustaining waterbird popula-
tions and habitats (Rios-Uzcategui 1994). How-
ever, cattle ranching and agriculture can require
deforestation. Widespread deforestation has oc-
curred in parts of the Western Llanos (Hamilton et
al. 1976, Veillon 1976). Extensive natural grass-
lands are the dominant habitat in the Alluvial
Overflow Plains. Major modifications to the
environment, such as dams, dikes, and deforesta-
tion have not yet occurred on most of these vast
plains, and the low human density over most of
the area prevents other significant impacts.

Additional studies are needed on the waterbird
communities in the Llanos, particularly in the
Alluvial Overflow Plains where the role of
impoundments as waterbird habitats should be
evaluated. Information on population dynamics
and species-habitat relationships will help better
understand ecological factors involved in regulat-
ing waterbird community structure and population
levels. Additional studies should also survey for
secretive marsh birds such as rails which,
although common in the Llanos, were not
detected during our study.

About 250,000 ha of private property on large
ranches (hatos) operate ecotourism enterprises that
attract tourists from around the world (Lara 1999).
These arrangements in turn provide a strong
incentive to promote wildlife con.servation. Com-
parisons of waterbird abundance and diversity
between private hatos (where wildlife is protected)
versus areas where hunting is likely to occur may
provide an indication of the severity of hunting
pressure (Perez and OJasti 1996) and the role of
hatos in conservation of Llanos waterbirds. Long-
term conservation benefits for waterbird communi-
ties on private lands of the Venezuelan Llanos will
require the continuing support of government and
conservation agencies as well as outreach to
landowners as a major approach toward protecting
the rich waterbird heritage of this region.

ACKNOWLEDGMENTS

This paper is dedicated to the memory of our friend and
collaborator. Prof. Gilberto Ri'os-llzcategui of the Uni-
versidad de los Llanos Occidentales in Venezuela. Funding
for this research was provided by Ducks Unlimited, the

International Programs Office of the USDA Forest Service,
and the Neotropical Migrant Bird Conservation Program of
the U.S. Fish and Wildlife Service. We especially thank
Bruce Batt of Ducks Unlimited and Doug Ryan of the U.S.
Fish and Wildlife Service for logistic and administrative
support. We are indebted to Mark Gregory and Alexis
Araujo for field assistance. We are especially grateful to the
caretakers and landowners of the following ranches for
access to their properties and use of facilities; Hato El Frio,
Hato El Cedral, Hato Pinero, Hato Eemando Corrales, and
Mantecal. The manuscript was greatly improved by
comments from two anonymous referees.

LITERATURE CITED

Bibby, C. J., N. J. Collar, M. J. Crosby, M. E. Heath, C.
H. IMBODEN, T. H. Johnson, A. J. Long, A. J.
Sattersfield, and S. j. Thirgood. 1992. Putting
biodiversity on the map: priority areas for global
conservation. International Council for Bird Preserva-
tion, Cambridge, United Kingdom.

Bisbal, E. j. 1988. Impacto humano sobre los habitats de
Venezuela. Interciencia 13:226-232.

Blydenstein, j. 1962. La sabana de Traebypogon del alto
llano. Boleti'n de la Sociedad Venezolana de Ciencias
Naturales 103:223-238.

Blydenstein, J. 1967. Tropical savanna vegetation of the
Llanos of Colombia. Ecology 48:1-15.

Botero, j. E. and D. H. Rusch. 1988. Recoveries of North
American waterfowl in the Neotropics. Pages 469^82
in Waterfowl in winter (M. W. Weller, Editor).
University of Minnesota Press, Minneapolis, USA.
Buckland, S. T., D. R. Anderson, K. P. Burnham, J. L.
Laake, D. L. Borchers, and L. Thomas. 2004.
Advanced distance sampling: estimating abundance of
biological populations. Oxford University Press Inc.,
New York, USA.

Bulla, L., J. Pacheco, and G. Morales. 1990. Seasonally
flooded neotropical savanna closed by dikes. Pages
172-210 in Ecosystems of the world: managed
gra,sslands (E. Breymeyer, Editor). Elsevier Sciences,
New York, USA.

Colwell, R. K. 2006. Estimates: statistical estimation of
species richness and shared species from samples.
Version 8.0. http://viceroy.eeb.unconn.edu/EstimateS
(accessed 26 October 2006).

Conway, C. J. and J. C. Simon. 2003. Comparison of
detection probability as.sociated with Burrowing Owl
survey methods. Journal of Wildlife Management
67:501-51 1.

Droege, S. 1 990. The North American Breeding Bird Survey.
Pages 1-4 in Survey designs and statistical methods for
the estimation of avian population trends (J. R. Sauer and
S. Droege, Editors). USDl, Eish and Wildlife Service,
Biological Report 90. Washington, D.C., USA.

Eli.i,s, D. H., R. L. Glin.ski, and D. G. Smith. 1990. Raptor
road surveys in South America. Journal of Raptor
Re.search 24:98-106.

Eigueira, j. E., R. Cintra, L. R. Viana, and C.
Yamashita. 2006. Spatial and temporal patterns of
bird .species diversity in the Pantanal of Mato Grosso.

Vilclla and Baldassarrc • LL.ANOS WATERIilRDS

113

Brazil; implications for conservation. Brazilian Journal
of Biology 66:393—104.

Frederick. P. C. and K, L. Bildstein. 1992. Foraging
ecology of seven species of neotropical ibises
(Thrcskiornithidae) tliiring the dry season in the Llanos
of Venezuela. Wilson Bulletin 104:1-21.

Gill, F., M. Wright, and D. Donsker. 2009, IOC World
bird names (Version 2.1). http://www.worldbirdname.s.
org/ (accessed 20 August 2009).

Gomez-Dallmeier, F. and a. T. Cringan. 1989. Water-
fowl in Venezuela. Editorial Ex Libris, Caracas,
Venezuela.

Hamilton, L. S., J. A. Steyermark, .1. P., Veillon, and E.
Mondolfi. 1976. Conservacidn de los Bosques
Humedos de Venezuela. Sierra Club, Consejo de
Bienestar Rural, Caracas, Venezuela.

Hamilton, S. K., S. J. Sippel, and J. M. Melack. 2002.
Comparison of inundation patterns in South American
floodplains. Journal of Geophysical Research-Atmo-
spheres 107: Article 8038.

Hanowski. j. M. and G. j. Niemi. 1995. A comparison of
on- and off-road bird counts: do you need to go off
road to count birds accurately? Journal of Field
Ornithology 66:469^83.

Hilty, S. L. 2003. Birds of Venezuela. Second Edition.
Princeton University Press, Princeton. New Jersey, USA.

Huber, O. and C. Alarcon. 1988. Mapa de vegetacion de
Venezuela. 1:2,000,000. Ministerio del Ambiente y de
los Recursos Naturales Renovables, Caracas, Venezuela.

Lara, L. C. 1999. El ecoturismo: /,una opcion para las
reservas forestales? Caso: unidad experimental de
Caparo. Revista Forestal Venezolana 43:69-78.

Ministerio de Agricultura y Cria. 1998. Anuario
estadfstico agropecuario 1996. Publicaciones del
Ministerio de Agricultura y Cn'a, Caracas, Venezuela.

Mittermeier, R. a., N. Myers, J. B. Thompsen, G. A. B.
DA Fonseca, and S. Olivieri. 1998. Global biodiver-
sity hotspots and major tropical wilderness areas.
Conservation Biology 12:516-520.

Peterjohn, B. G. 1994. The North American Breeding Bird
Survey. Birding 26:386-398.

Perez, E. M. and J. Ojasti. 1996. La utilizacion de la
fauna silvestre en la America Tropical y recomenda-
ciones para su manejo sustentable en las sabanas.
Ecotropicos 9:71-82.

Ramia, M. 1967. Tipos de sabanas en los llanos de
Venezuela. Boletm de la Socicdad Venezolana de
Ciencias Naturales 27:264—288.

Ramia, M. 1993. Ecologi'a de las sabanas del Estado
Cojedes: relaciones vegetacion-suelo en sabanas secas.
Fundacion La Salle de Ciencias Naturales, Serie
Ciencia y Tecnologia Number 4. Caracas, Venezuela.

Ramo, C. and B. Busto. 1984. Censo aerco de corocoros
(Eiulocinns riihher) y otras aves acuaticas en Vene-
zuela. Boletm Sociedad Venezolana de Ciencias
Naturales 39:65-88.

Ramo, C. and B. Busto. 1988. Status of the nesting
population of the Scarlet Ibis (Endocimus rither) in the
Venezuelan Llanos. Colonial Waterbirds 1 1:311-314.

Rios-UzCATEGUi, G. A. 1994. Role of private landowners in

wildlife conservation; Venezuelan case. Thesis. Uni-
versity of Maryland, College Park, USA.

Robbins, C. S., D. Bystrack, and P. H. Geissler. 1986.
The Breeding Bird Survey: its first 15 years, 1965-
1979. USDl, Fish and Wildlife Service Resource
Publication 157. Washington D.C., USA.

Roca, R., L. Adkins, M. C. Wurschy, and K. Skerl.
1997. Transboundary conservation: an ecoregional
approach to protect neotropical migratory birds in
South America. Environmental Management 21:477—
481.

Rodner, C. 2006. Waterbirds in Venezuela. Sociedad
Conservacionista Audubon de Venezuela, Caracas.

Rotenberry, j. T. and j. A. Wiens. 1980. Habitat
structure, patchiness, and avian communities in North
American steppe vegetation: a multivariate analysis.
Ecology 61:1228-1250.

Sanz-Agreda, a. 1998. Evaluacion del subprograma
manejo de anatidos en Venezuela; caso Dendrocygna.
Direccion de Manejo de Fauna Silvestre PROFAUNA,
Ministerio del Ambiente y de los Recursos Naturales
Renovables, Caracas, Venezuela.

Sarmiento, G. 1983. The savannas of tropical America.
Pages 245-288 in Tropical savannas (F. Bourliere.
Editor). Volume 13. Elsevier Sciences, New York,
USA.

Sarmiento, G. and M. Pinillos. 2001. Patterns and
processes in a seasonally flooded tropical plain: the
Apure Llanos, Venezuela. Journal of Biogeography
28:985-996.

Scott, D. A. and M. Carbonell. 1986. Directory of
neotropical wetlands. lUCN Intemational Waterfowl
and Wetland Research Bureau, Slimbridge, United
Kingdom.

SPSS Inc. 2001. SPSS 11.0 for Windows. SPSS Inc.,
Chicago, Illinois, USA.

Tamisier, a. and O. Dehorter. 2000. Eauna of the Llanos
and Pantanal; ecological basis for the sustainable
management of tropical Hooded ecosystems. Case
studies in the Llanos (Venezuela) and the Pantanal
(Brazil). Centre d’Ecologie Eonctionelle et Evolutive
CNRS, Montpellier. Erance.

Thiollay, j. M. 1978. Population structure and seasonal
tluctuations of the Falconiformes in Uganda Na-
tional parks. East African Wildlife Journal 16:145-
151.

Tiioma.s, B. 1987. Spring shorebird migration through
central Venezuela. Wilson Bulletin 99:571-578.

Veillon, J. P. 1976. Las deforcstaciones de los Llanos
Occidcntales de Venezuela desde 1950 hasta 1975.
Pages 67-112 in Conservacion de los bosques
humedos de Venezuela (L. S. Hamilton. J. Steyermark.
.1. P. Veillon, y E. Mondolfi, Editores). Ministerio del
Ambiente y de los Recursos Naturales Renovables,
Caracas, Venezuela.

Velasco, A. and J. Ayarzaguena. 1985. Situacion
actual de las poblaciones de baba (Caiman crocodi-
Ins) sometidas a aprovechamiento comercial en los
Llanos Venezolanos. Publicaciones de la Asociacion
de Amigos de Donana Number 5 Caracas, Vene-
zuela.

114

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. I. March 2010

c o

•— (U

O v;

c. c

C.

E

= h; a

■a O ^

3 .E
CJ <u c

2 'o CL

OJ OJ _

g- «
cL ^ 5

C. D-
v. - C

.2 -E

^ -E o

^ dJ

<U

D

■o

W) ^

c

T>

c

<D

O

<u

O-

1)

jr

« II

Q-

p =
.= c

Xj >-

^ . #.
^ <u
o

^ -g

!- C

«

^ >

■a

5

<u

a.

<u

c -2
o c

i/l <u

0

c/5
<U
Cl I'

1 <

S C2

c/5 ^

o

'oo o
lU o
s- n
o ^
u

’15 ^

c

OJ

CJ

OJ

a-

OJ

c/:

C

‘5

a.

a; -d

Tj 0/

'C o

OO (U

^ a>
w -a

• E

o =5

D. ?
E! ^

O (D
1)
c/:

C

C3

CJ

(U

o.

i: ^

c «

O (U

L.

c/5 4J

.2

CJ &

OJ

Q. c/5
C/5 C

^ .2

o '=‘^'

Q- <L>

5 O

J2 —

8S

CJ

2 E

^ c3
Q o

C

.

X E

5 z

z

w

CL

D-

<f

■2 o

o ^

S 0/

4—1 Xj

c

T3 C
=3

(U c

Of)

2 II

g -

^ 5
u- o

D • —

O-

<u

I

II §

c/5

c e

1 ^
^ OJ

. , Cu

'/"i •£

CA C)/J

c

2 -n

lO ir; in in “ vn rn lo in ia5 lo iA5 in i/Y lo >^5 in Tt f<5 lAi i/--, rY lAi pj- —

r-

(n

C500C — ocNoccNsccnoopfr-vOcxcnin^

O CN <N sc — (N

a^ in ini — in o cn

in in —

(NONOvOCsOCNmOs

— (N Pf

ni

CN

q

GO ri

o r-

m so

'O — —

in — Tt <N r4

cx5r-r-ooor^ —

o pf m (N o o
— ‘ o o ri d d — '

q in n] in — inscoo —
d 00 ini 00
n m (N

(N rt m in so m

r^oooin — mr-oospt
osos sc TtT:}-ininin

^(NqrNpfqr^ooooossoooscmoinsc — 00
n pf ini d pt 00 — ' (N — sd d mi r4 <N r^’ d rn (N
r^cNm — — — m c^,

qqq^CNtOOOC^O
d mi d r4 d d d d — d
r-^^a^r^oo•^ o

i/^Tt----qp};r^qptqmqpj-ocoo(NsO'^ossor-omscrt

oO'^oooddddddododddddinicNidinidsddddd
— aNOsosoovinOinoosor^aNscr^mpfr^mos — 00 r4

— (N — sOr<', oOfNlOO
qr-;qqqsoooo
d d d d d d 00

(N

O — sOrNOOoOOOpJ-Tj-sOmCNpf*— OsooOmrt — sO'^OspfO
cNO'^qmqfNqqmiq*— inr-(NO — m-^O'—or^OOCN
d — dd — dd— ’dmiddddd

OOOO — OOOpJ-OOO

(N 00 OS 00 mj

\C m m in

— pf(npfin(^iin — moopj-oo^oo
SOSD — sOCN — O'! — — pj- m

m (N os mj

<N m Tt

OO o rn m o
m m ^ in m,

(N

SO
iTj m

vO os ^ in os -sO

— in so
m 00 CO
d' — "

On m> Tt

— r-
d d
so <N

OO — soscmr^oo

Os

Os

O'!

— in

Tt — —

r-' q <N

d

O r- rn 00 in
m CN so r- 00

m

O

o

d" — —

in cN m,

— d

Ttooc — msor-oosso
m, (NinmrNsor^osino

c

d

00

(N

B

O <J CJ O) 0)
“O *D "O -O "O -O
:3 eg C3 r3 C3 C3

a)o>o<uiL)<u<uoajaj<uaj

-o-o-o'o-a-o-o'o’o’o-o-o

C3c3c3rtc:3c3c3!:^cdc3(r3?3

o «u

“O -O

C3 c3

(U <L)

-T3 *0

cs ca

^ ^ ^ ^

s

Sj

-2

C‘ ^

u.

^ O

^ i

^ O

^ U

OJ

Sj

^ c
5 c
B' a;

§ I.

§ z

'S

-5: (u

=: c

O

^ o

'5,

o

3 *-

ah V

u.

C

CJ

.£ -Ji
-C jj
c ^

Z

< pa

-2

-I -5
£

£ £

o

fe o

X oij

CL t2
C3 C3

u u

s:

>C ^

■5:

^ C

S 2

W ^

w C
C3 <U
OJ CJ
u. u-

o o

oo 2

■2 5

C3 ^ ~

52

c

o

O id

oa 1)

pa

4, E <u
X " •-"

u CQ
CQ "S

-= =
L c

c
1)
u

C5i c/c

(U

3

J CL PC

■s

§ 5

JL 5

d ^
^ c
— o

(U u.
i- CJ
0^) ^
X

>Y

o
c

oo O
•r- O

^ CL-

o ^

Sj

>,

c

o

u

CJ

X

X
00 .

'S

Cj

s:

'5' §

1 1

5^"§

CJ

X

X5

^ —
^ "O

■a ji

^ c
"5 -2

? ^
Oc 'Ll
CL ^

^ Cj

:§, s
E" .2
^ c
=

kj

c
o
o

CL ^

cn

s* --
s s

Ti

Ҥ

“O ^
CJ (A

•S s

s3 22
d d
2: ^

CJ

5 E

4^

L.

o

-o

w

3

c

c/3

X

'5

1

q

CJ

CJ

CJ

CJ

c

uL

>>

c/3

C/3

X

c

CJ

CJ

3

CJ

3

1

B-

X

CJ

*»««»

s

•c

3

3

Oi)

cn

TD

o

*H

cJ5

X

o

c

U

"cj

>-

u

3

X

4—

3

X

O

a

CJ

u-

a

c/3

O

CC

3

CJ

c/5

3

X

cn

X

X

3

3

S

c

cd

CJ

u

CJ

00

•o

CJ

E

o

*§

§

**»

3

2

00

r^

S

^ •

Q ,

^ ---
D Q

Q ^

^ o

£ =3

d Q

c/3

1 I

-o

CJ J=

2 ^
CJ

q m

-li o

J I
m X

2

2

D,

CL

Oc

&

^ Q

L<2 -02

Q Q

OXj go

C C

2

2

S "§

-= d

CJ

~o

d c
^ D

5 S
U ^

CJ

U m
3 O

d Q
r o

8 8
cn c

S o

CJ

3

-g

CJ 'o
3 3

-■E o ^
-o 8 X

CJ o ^

E- S

U _3

“5 3
o x:

c X

'qj

•£

w -»- 3 d

Cl Cu <

CJ

3

-o

E

o

X

c/;

CJ

u

X

h-

CJ

3

i£

’E

o

CJ

U

CJ

3

•u

E X

3

C

<

yUc'lla anil Baldassarre • LLANOS WATERBIRDS

115

■o

<u

o

U

X

5

z

LU

□_

CL.

<

^inoor4ooosr^i(Nr--r-m —
or- — OCNOOO’^OCNr-OOOOr^rn

O^OsOO'^^Owo^O —

— tJ-

O O
— fNj O

<N ri o o o

O in CJ O Os CN <N r-J

r- — r- — m'^’^ooo

0 — 0

in m

as o r- j

r- 00 in

ooor-inas'^'^i"

moo — r-ooomo-^or-ONOs<Ninrn^ — asr^ in — (Nnjoo—; — rn in ornoc

oodtnom — or^o^'r-'^mmoomjhjnon

— mo
(N m

O — OsOsmcNO'^Os

^ — m r-

^ mm — osin — omin — osos
5 osinod — odoommooodoo — oooo
3 mmos — ONOsmoosmininmr-

r- m m o — os m

— o ^ p p in m

o o o o o c> o

p p ^ r^

mi O

m m

os p ^ m* c

’ r- m r- mi o r-

m in ms in r- OJ

ooosOS'^Opinp
■ — — Tt o o6 — 00
os O 00 00 00

in — r- m o Os
ooop — inp —

o — in o o —

o — ni — mosTfinoso — oomoor-miO
oooo'^o — oo — ooom'^irm^ — — ;
00000000 — 0 — 000000000

os'sor-'^ — os'^mmino —
m* m m so sD — m^ in — ^ m

r- ^ m* o

mi nO so

r-sosDso — r-inoo'^rj-m'?fmm —
mm^ — in — '^msosom^mm'^'^

mooo — os — somm —
Os"^m — oommnJinoo
r:}- oj p \o — m^ r-^

— — in so"

oo os r-
O

in m in m
in

oooommm'^m — osD —

mlnmJmsO'^lno^r-m

m m m — o m

m o so r- in so

rj- r- m so ^ so

so m r- mi

m

— m

m

m

00

m

o

s ^

^ O

O u-

x)"0'0’0'a’0"a"opppp’p

o o
-c x:

cn c/5

o

-C

cn 00

o o
x: xz
cn

o o
x: xz
00 00

o o

H)

c

fU

a.

OO 00 OO 00 00 oo

Ui

Ln

u.

Lh

L-.

L-

O

0)

a

(U

0)

o

cd

3

td

3

cd

0)

o

a

<u

c

'u

c

•C

c

'u

C

'C

c

c

c

c

c

«u

a

HJ

(D

§ H

D- o

a

D.

QJ

CL

a

a.

<u

CL

a

CL

c/2

c/2

cd

c/2

cz

cd

c/2

C/2

Cd

c/2

C/2

Cd

o £

O

O

O

o

o

a.

Q-

CL

CL

Q.

D.

•2 ^5: -2

Q- ■=

C- DC

§■3 -

-i I -i

c -o .E

Uj g

—

LX

3

<D

c

d

a

a

o

Ui

1)

c

1)

D

u.

w

.2

L.

3

<D

e-

OQ -C
3 C

N

L.

3

3

3

<

a

CL

00

00

S-

OJ

a
c

1

zx
u

Lc
H

■o
a

^ S f

OJ

Q,

CS. •

o

Cz; ’

U -

Q.

■a

o

C3

"P

I

\cxC

o

:3

"O

<u

X

<u

a

“O o

60

'S. 'c

>> C
^:3

UJ <

o) 'r-

6f} 1/5 -C;

(U ^0 -ii

c

OJ

C3 c/2 w

1) czi • —

W- (U C

o ^

u

-2 ^

5^ ^ 'c

i .1 c

rv

“'J -a .y
c c ■-

■r n jj

I

c3

3

O O
Cl 00 C/D

S Cl
w O.

6T 1/5

^sj

Sc

5 ^

2

B

t'l

6c

O

Cl ■

«>3

■r:

a '

o-

O

c

i_

(U

S G

G

fe H
H

(U ^
-C L,

5/2 (3J

i:: j=

^ Si

"2 B

G G

k. L.

a o
x: xz

c/2 C/5

6JJ 61j
C C

2 5

C5

Sj

.

TO

c

3

00

E

oo

3

3

£

H

00

“3

—

-E

O

3

5

'oij

3

3

L.

-o

00

TO

c

OO

00

00

I)

iE

x

$

3

3

O

5

3

3

<L)

0>

3

X)

a

Q

’n

N

3

3

“2

jd

o

__

6Ij

3

3

OJ

0/

o

Cl

'E

cl

Jd

~B

u.

3

"E

O

£

1)

L.

<U

L.

C/D

D

QQ

o

-J

>

:n

<

a

o

D

3

p

o

'C

3

p

c/2

O

fU

C3

~a

“O

C3

u.

C3

X

(U

C3

*0

'E

C6

CQ U QJi

cj

c6

D-

O

O

O

OO

(U

63

*o

G

o

o

o

63

"O

3

O 3

3

p

w '*5

c/2 <U

'u

3

■aiii

X

o <

S S =:

r^> ^ ^

B c

w Li r3
13 1-

6c C

H

5 i ^

— c/2

U. — is

C f3

c/2 ^

^ ^ -O
00 ^ a
C L- TO

"S ^ OJ

oxj -Q • —

.E ^

Cl >

o

63

;p

'E

c

3

H

-c ^

C2 ^

s ^

2 ^

^ Ci

c/) *3

-O ^

o _

c .E

o p

Q S

“O ^

o ^
Q- o

^ &
V 3
V

(J p

i3 ^
cQ ac

o

3 (D
TO c3
w “O

■;:3
00 .i:i
o i-

— a

OU Xj

o

c

LLl

The I'Vilsun Journal of Ornithology 122(1):1 16-125, 2010

PHENOLOGY OF SIX MIGRATORY COASTAL BIRDS IN RELATION

TO CLIMATE CHANGE

CHARLES R. FOSTER, ANTHONY F. AMOS,' AND LEE A. FUIMAN'

ABSTRACT.— The migration phenology of six species of coastal birds on Mustang Island, Texas, USA was examined
tor a 27-year period ( 1978-2005). First arrival date, last date of departure, and duration of stay were quantified for three
winter and three summer residents. These three variables were analyzed for changes over time and correlation with local,
regional, and global temperature indices. Mean local summer temperature increased 0,03° C/year (0.74° C overall), while
mean local winter temperature increased 0.10° C/year (2.76° C overall). The three winter residents had a trend for
increasingly later arrival, increasingly earlier departure, and decrea.sed duration of stay over the 27-year period. These
trends reflect a shortening of the winter season for these birds and are consistent with expected responses due to warming
temperatures. The three terns representing summer residents had less homogeneity in temporal trends than the three winter
residents. Correlations of local temperature with arrival and departure dates, and duration of stay yielded few significant
results and no overall pattern. Only Double-crested Cormorant (Phalacrocorax aiiritus; a winter resident) and Least Tern
(Sterna antillarum; a summer resident) had significant correlations between arrival date and arrival temperature. Received
27 March 2009. Accepted 3 September 2009.

The Earth’s climate has warmed by —0.6° C
over the last 100 years according to the Intergov-
ernmental Panel on Climate Change (IPCC 2001 ).
There have been two distinct periods of warming
during this time; one from 1910 to 1945 and the
other from 1976 continuing to the present (IPCC
2001). The rate of warming during the latter
period has been approximately twice the magni-
tude of the first and greater than any other time
during the past 1,000 years (Walther et al. 2002).
Regional changes in temperature are more
significant when examining ecological changes
and, in some cases, the.se climates have exhibited
greater increa.ses than the mean global tempera-
ture over the past 30 years. The largest regional
increases have been over the middle and high
latitudes of the northern hemisphere (IPCC 2001).
There has been a recent paradigm shift from
focusing on climatic consequences of global
climate change to collecting existing evidence of
current alterations in the biological activity of
organisms as the result of warming temperatures
(Wuethrich 2000, McCarty 2001, Walther et al.
2002, Parmesan and Yohe 2003).

Climate change has altered the activities of a
broad range of organisms in a variety of ways
(McCarty 2001. Walther et al. 2002). Organisms
worldwide are experiencing shifts in range.

' Marine .Science Instilulc, University of Texas at Austin,
750 Channel View Drive, Port Aransas, TX 78373, USA.

-Current address: Gulf Coa.st Research Laboratory,
University of Southern Mississippi, 703 Ea.sl Beach Drive,
Ocean Springs, MS 39564. USA.

'Corresponding author; e-mail; charles.foster@usm.edu

breeding and migration phenology, and commu-
nity structure. It should be expected that organ-
isms in the northern hemisphere will shift their
ranges northward or to higher elevations with
increasing temperatures to remain within their
thermal preferences. Parmesan et al. (1999)
reported significant range shifts consistent with
regional climate change for 22 species of non-
migratory European butterflies over the past
century. GrabheiT et al. (1994) measured signif-
icant upward shifts in elevation in nine plant
species over the previous 90 years in the Austrian
Alps.

Phenology, or the sea.sonal timing of biological
events, can also be an important tool for assessing
effects of climate change. An organism may shift
its range and/or adjust aspects of its phenology to
cope with a changing climate. Phenological
changes have been observed for a wide range of
species, including woody and herbaceous plants,
birds, insects and other invertebrates, amphibians,
and fishes (Beebee 1995, Alias 1999, Bradley et
al. 1999, Roy and Sparks 2()()(), Cotton 2003). A
review of phenologies for 677 species by
Parmesan and Yohe (2003) found the shifts in
phenology were overwhelmingly in the direction
expected from climate change (87% of all cases
studied). Phenological events can occur through-
out the year, but most studies of phenology have
focLi.sed on the spring sea, son. This is because
many biologically relevant events occur in spring
that are likely to indicate an imprint of climate
change since it is a transitional period. Examples
of biological events that occur in spring are
blooming of flowers, thawing of ice on frozen

Foster el al. • COASTAL BIRD PHENOLOGY AND CLIMATE CHANGE

117

lakes, breeding of a variety of organisms, and
migrations of butterflies and birds.

Several studies have shown that arrival dates of
numerous species of migratory birds to breeding
or stopover areas have been advancing over the
past century (Bradley et al. 1999, Cotton 2003,
Jonzen et al. 2006). In addition, the advancement
of airival has been linked to recent increases in
global and regional temperatures. These findings
have been documented by several studies of bird
migrations (Sparks 1999, Sparks et al. 2005,
Gordo and Sanz 2006).

Nearly all studies of avian migration phenology
have focused on passerine species, leaving other
bird groups virtually unexamined. We examined
migration phenologies for six bird species on the
Texas coast with diverse life histories. Our
objectives were to investigate the migration
phenology for winter and summer residents
separately by testing the following hypotheses:
(1) that phenological traits have changed over the
past 27 years, and (2) that phenological traits are
not independent of local, regional, and global
temperature indices.

METHODS

Survey Methods. — Surveys (BEACHobs) of
Mustang Island, Texas, USA began on 14 April
1978 and continue to the present. Surveys were
conducted along a 1 1.7-km stretch of barrier island
shoreline between beach Access Road 1
(27° 47.6' N, 97° 05.1' W) and Access Road 2
(27° 42.2' N, 97° 09.0' W), south of Port Aransas,
Texas (Eig. 1). The primary objective of BEA-
CHobs was to census bird species that use the
beach. Other data recorded on the surveys included
numbers of people, dogs, and motor vehicles, as
well as atmospheric and oceanographic parameters,
beach and dune morphology, tidal stage, and
occurrence of natural and anthropogenic debris.

The survey protocol was established from
results of several tests conducted during initial
development of the survey to optimize surveying
techniques with respect to direction of travel,
survey frequency, and diurnal distribution of
birds. The survey was then designed and con-
ducted every other day, starting at sunrise (0600
to 0740 hrs CST) traveling from north to south.
This sampling frequency ensured that bird pres-
ence was consistently represented in the data set.
All surveys were conducted by one of us (AEA)
from a four-wheel-drive vehicle while driving
slowly southward between the two access roads.

The surveyor, who was also the driver, had a clear
view of birds, the great majority of which were
feeding, roosting, or loafing. Vehicular traffic is
permitted along many Texas barrier island
beaches, and the beach is considered a roadway
where all traffic laws pertain. This made con-
ducting the survey possible.

Surveys were done under all weather conditions
and took from 40 min to more than 3 hrs to
complete, depending on beach conditions, abun-
dance and diversity of bird species, and number of
people. The full survey distance was not com-
pleted on rare occasions due to dangerous weather
events, high storm tides, or unsafe beach surfaces.
There were some gaps in the survey especially in
the winter months of some years when the
surveyor was unavailable (Table 1).

Most birds were seen in the swash zone
(averaging 15 m wide) between the shoreline
and the most recent high tide line. Birds in flight
were also counted and occasionally there were
small roosts of shorebirds between the high tide
line and the foredune, which were also counted.
Birds were usually counted individually, but block
counting was used when abundance was high.

Statistical Analyses. — ^Data for complete calen-
dar years from 1979 through 2004 were analyzed
for the current study. Partial year data for 1978
and 2005 were used for applicable species.
Statistical analyses were computed using
SYSTAT 11.0 (SYSTAT Software Inc. 2004).
Migration phenologies of six species (three winter
and three summer residents) were examined.
These six species were selected because they
exhibited a clear annual pattern of occurrence (all
were present one part of the year but completely
absent for the other part). This allowed identifi-
cation of date of anival and date of departure for
each species each year. The three winter residents
were Double-crested Cormorant {Phalacrocorax
auritus). Eared Grebe (Podiceps nigricolli.'i). and
Herring Gull (Larus argentatiis). The three
summer residents were Gull-billed Tern (Sterna
nilotica). Least Tern (S. antillarum). and Sand-
wich Tern (S. sandvicensi.s).

Phenology was described by: first arrival date
(PAD), last date of departure (LDD), and
duration of stay (DUR). PAD was the first day
of each year that a species was observed on
Mustang Island. LDD was the day that a species
was last seen each year. Duration of stay was
calculated as the difference between LDD and
FAD. If a significant (>1 week) gap in the

118

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 1, March 2010

97'30'0"W

97’0'0''W

96"30'0"W

COPANi

ORPUS CHRll
BAY

fAFPIN

-28’0'0"N

■27'30'0"N

•27"0'0"N

•26°30'0"N

-28”30’0"N

26’30'0"NH

28'30’0''N'

Legend

National Seashore Boundary
Stream

— I —

97-0'0"W

Corpus Chrlsti Bay

Gun of Mexic

USTANG ISLAND

JOSE ISLAND

28’0'0"N'

GULF OF MEXICO

N

97’30'0''W

1—

96’30'0”W

Access Road # 1

Access Road U 2

Mustang Island
Beach Study Area

W-^ Acc !

USTAn

ss Road # 1

G ISLAND

Access Rc ad 2

27’30'0"N'

27‘0'0"N'

FIG. I. Beach survey area on Mustang Island, Texas.

surveys existed during the period of arrival or
departure, that year was omitted from the
analysis. Temporal trends in phenological traits
were de.scribed for each species using linear
regression and the slope of each regression line
was used to estimate the trend. The hypothesis on
temporal change in a phenological trait for winter
or summer residents was tested with Student’s t

statistic ba.sed on the trend value for each species
(i.e., /? = 3 trends).

Local temperature data bases were used to
as.sess changing temperature and possible effects
on avian migration phenology. These data bases
included the temperature record taken during the
Mustang Island beach surveys. Temperature
records from Corpus Christi, Texas (—30 km

Foster et al. • COASTAL BIRD PHENOLOGY AND CLIMATE CHANGE

1 19

TABLE 1. Number of surveys completed by month on
Mustang Island, Texas, 1978-2005.

Month

Number of surveys

Jan

208

Feb

227

Mar

323

Apr

322

May

347

Jun

320

Jul

327

Aug

335

Sep

313

Oct

303

Nov

332

Dec

266

from the survey area) were also used. These data,
obtained from the NASA Goddard Institute for
Space Studies (http://data.giss.nasa.gov/cgi-bin/
csci/), served as an inland location less influenced
by the Gulf of Mexico.

Two additional temperature data bases were
used to assess the importance of large-scale
climatological changes for the six species exam-
ined. Global and northern hemispheric tempera-
ture anomaly records were obtained from the
Climatic Research Unit at the University of East
Anglia, Norwich, UK (http;//www.cru. uea.ac.uk/
cru/data/temperature). These temperature records
are those used by the Intergovernmental Panel on
Climate Change (IPCC). The anomaly values are
measures of temperature relative to the average
temperature during the period of best coverage
(1961-1990) and have become the standard for
most studies of climate change. These global and
hemispheric averages are accurate to — h/— 0.05°
C (2 SEs) for the period since 1951. All four
temperature indices were analyzed using linear
regression to test for a significant change over
time, and for correlations between temperature
and migration phenology.

FAD was compared with mean temperature for
the month(s) in which that particular species
arrived on Mustang Island (arrival temperature).
FDD was compared with the mean temperature of
the month(s) during which that species left
Mustang Island (departure temperature). For
instance. Sandwich Terns arrived in March and
departed in November or December. Arrival
temperature used for the Sandwich Tern was the
mean March temperature, while the departure
temperature was the average of the monthly mean
temperatures for November and December.

RESULTS

Local air temperature on Mustang Island, Texas
increased, on average, by 0.04° C/year during
1978-2005. This overall increase of 1.08° C did
not represent a significant linear trend (P = 0.17;
Fig. 2A). However, a highly significant trend of
similar magnitude (0.05° C/year) was present in
the data for Corpus Christi, Texas over the same
period {P < 0.001; Fig. 2B).

All three species that winter in the study area
had positive slopes for airival date, negative
slopes for departure date, and negative slopes for
duration of stay (Table 2). The mean slopes for
FAD and FDD were not significantly different
from zero. However, the mean slope for DUR was
—0.8 days/year and demonstrated a significant
departure from the null model of no change over
time {t — —5.188, n = 3, P = 0.035; Table 2).
This mean slope equates to an average decrease in
duration of stay for winter residents of 21.9 days
over the 27-year study period.

The three terns had less homogeneity in the
temporal pattern of the phenological traits than the
three winter residents. Sandwich and Gull-billed
terns had a non-significant negative slope for
FAD, while Least Terns had an opposite,
significant trend (Table 3). All three species had
negative slopes for FDD and DUR (Sandwich
Tern was non-significant for both). Sandwich and
Gull-billed terns had a trend toward earlier aiTival.
but the departure date for these two species was
also earlier. The mean slope of DUR for all three
species combined was not significantly different
from zero (t = — 1 .750, n = 3. P = 0.22;
Table 2).

Local mean winter (Dec-Feb) and summer
temperatures (Jun-Aug) were calculated separate-
ly to examine the relationship between trends in
phenology for winter and summer residents. Mean
summer temperature increased by 0.03° C/year
over 1978-2005, an overall significant increase of
0.74° C over the past 27 years {P = 0.013;
Fig. 3). Mean winter temperature increased sig-
nificantly by 0.10° C/year or 2.76° C over the
same period {P < 0.001; Fig. 3).

Comparisons of local temperatures with arrival
and departure dates, and duration of stay yielded
few significant results and no consistent pattern.
Only the Double-crested Cormorant, a winter
resident, and Least Tern, a summer resident, had a
significant con-elation between amval date and
aiTival temperatures and both were positive trends

120

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 1, March 2010

(A)

(B)

FIG. 2. Mean annual air temperature for Port Aransas, Texas (A) from 1979 to 2004 (trend not significant; P = 0.17),
and for Corpus Christi, Texas (B) from 1978 to 2005 (trend is significant, P < 0.001).

Foster et al. • COASTAL BIRD PHENOLOGY AND CLIMATE CHANGE

121

TABLE 2. Slopes of regression lines and results ot one-
sample student's r-test for first arrival date (FAD), last date
of departure (LDD), and duration of stay for both winter
and summer species (individually significant values are
in bold).

Species

FAD

Year vs.

LDD

Duration

Winter

Double-crested Cormorant

(Phalacrocorax auritus)

0.409

-0.416

-0.812

Eared Grebe (Podiceps

nigrieolis)

1.262

-0.171

-0.489

Herring Gull

(Larits argentalits)

0.141

-0.644

-0.993

Mean

0.604

-0.410

-0.765

SE

0.585

0.237

0.255

t

1.787

-3.005

-5.188

P

0.216

0.095

0.035

Summer

Gull-billed Tern

(Sterna nilotica)

-0.404

-1.738

-2.087

Least Tern (S. antillanim)

0.462

-0.393

-0.825

Sandwich Tern

(S. sandvicensis)

-0.542

-0.436

-0.115

Mean

-0.161

-0.856

- 1 .009

SE

0.544

0.764

0.999

t

-0.513

-1.939

-1.750

P

0.659

0.192

0.222

(Fig. 4). The Eared Grebe, a winter resident, had a
significant (negative) relationship between depar-
ture date and local departure temperature (Fig. 5).
None of the six species had a significant
relationship between duration of stay and local
temperature. Similar results were found when
temperature data from Corpus Christi, Texas were
used. The temperature anomaly data sets used to
represent the global and northern hemispheric
temperatures over the past 27 years yielded only
two significant results of a possible 36 (Gull-
billed Tern and Least Tern exhibited significant
relationships between DUR and global tempera-
ture) even though significant warming occurred.

DISCUSSION

The general reduction in time spent by six
species of birds on Mustang Island over the past
27 years, along with confirmation that local
temperatures have increased more in winter than
in summer, suggests that warming temperatures
may be influencing the migration phenology of
three winter residents. However, when phenology
variables were related directly to measures of

C

O ON ri

SO 00 OS

Cj

Cl so r-

1— ' r-

Q.

— to

o p 00

odd

d d d

Cl re O

r*-, O tCj

' — '

O ^ O'

Cl ^

O 00 O

O c-

53

UJ

c/3

d d

d d

Ui

+ 1

+1 +1 +i

+1 +1 -fl

o

TD

C

Cl as rc

r- IT) ic)

(U

3

O

— 00 OS

00 Cl —

"o

H

00 "rt OS

0 00 —

c

o

d d d

ci d d

>

1 1 1

1 1 1

o.

«

3

"O

Q

Tt cn c-

Os tC) Os

c

r-

OO c- sO

C3

Q

(Os tn sO

to p

r

c/3

d o6 ci

d rc

tu

Cl d

Tt ^ Cl

CO

+1

C

— Tf 00

r- m

4J

—

O 00 vO

(U

CI — Cl

d — Cl

c/3

c

00 c~

O Os

O rc Cl

Tf OS \0

OX)

re so

o — ^

cO

d d d

odd

C

o

^ 00 00

r- so

c/3

OS to O

O Os 00

<U

SE

rc rc

00 Cl to

Ui

OX)

d d d

d d d

+ 1

+1 +1 +1

+1 +1 +1

C3

T3

c

so — ^

00 so

0)

W

c-

m Os rc

.B

TD

' so

rc Tt

—

(U

d d d

^ d d

1 1 1

o

3

c

1 1 1

1 1 1

CJ

a.

3

O.

43

r-- os so

d — r-

O

00 O Os

Cl so os

c/3

Q

os OO Os

00 rc — ;

X

c/3

iri (N sd

ci ~ ci

o

c

(c — d

!U

u.

Q

c

>. U-

rv

W 4^ CJ

o o <u

CO

C3

4)

S < S

O O Q

+1

C- Os 00

so tC3 —

O — Cl

— o o

c

o;

r"- — '

m ^

c

^c rc d

d O rc

OX)

— OO

^ o

c

d d d

d d d

o

so 00

O ^ d

c

sO — iC

ic, d rf

Cl 00 00

d — rc

c/3

CJ

SE

d d d

d d d

'o

+ 1

+1 +1 +1

+ 1 +1 +!

CJ

Cu

•a

c

os Cl —

•Tt d Cl

c/3

W

rt

u

O so

O so

-p

H

Tt Cl — '

rt lo

•o

d — ^ d

d d d

Xi

"3

1 1

u-

>

CJ

h

1

<

O Os os

^ o tn

d ^ c*

00 — 00

3

Q

OS p Os

so OS r-

c/3

c/3

d ci ci

00 tc —

"O

— (^ rc

C3

u

CJ

c

>

c

3

0)

w >

O O 3

o z <

2 < s

S

r- OS (^i

00 vn

cS

O C4 Cl

d o —

c/3

u

__

c

C3

C3

O

L.

‘oX)

o

o

E

o

>>

o

c

CJ

-C

CJ

o

T3

U

E c

Cu

'35

Ci

d) u

o

43

t/5

:rest(

re be

Gull

Um cj

^ n ^

ua

JD

Cu

i, O 60

— (2 -

CO

<

C

c/3

CJ

C/3

S *o .£
u- ■= OJ h
o ^ h

w C3 CJ

•_ 'i _ 1

<U _C v; "O

P — ra c

^ CJ c3

.E Q cu

o O J

a

>

122

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. I. March 2010

FIG. 3. Average summer (top; P = 0.01 3) and winter (bottom; P < 0.001 ) temperatures for Corpus Christi, Texas from
1978 to 2005.

local temperature, few significant correlations
were found. It is possible that temperature does
not have a direct effect on migration phenology at
this location, but may affect other places/times
along the migration route. Our results suggest that
migration is triggered by conditions in breeding
areas or by changes in the availability of prey
items, both possible products of climate change.
Double-crested Cormorants, Eared Grebes, and
Herring Gulls are all widely distributed across
North America. Thus, identifying whether or not
climate change is affecting the.se birds, and where
on their migration route the effects are greatest, is
difficult. A lack of banding studies of these birds
prohibits inferences on where the birds that winter
on Mustang Island actually spend the rest of the
year. However, it is known that climate change
affects higher latitudes at a more rapid pace than
equatorial locations (IPCC 2001). The Herring

Gull has the northernmost breeding range of the
three winter species examined and had the
strongest trend (but not significant) toward
contraction of the winter season on Mustang
Island.

The three species of terns representing summer
residents did not exhibit a consistent pattern in
aiTival date, departure date, or duration of stay.
This inconsistency may arise from' a smaller
temperature increa.se during summer over the prtst
27 years. All three species winter from Mexico to
South America. If long-term changes in pheno-
logical traits are more common in the “winter
portion” of the migration cycle and temperature
increase is greater in winter than in summer, terns,
which winter in tropical latitudes or move to a
southern hemisphere summer, might not be
inlluenced by a long-term temperature increase.
Their migratory pattern may keep them in

Foster et ai • COASTAL BIRD PHENOLOGY AND CLIMATE CHANGE

123

■o Mean September-October temperature (°C)

"<5

>

'C

CQ

of)

FIG. 4. Mean arrival temperature (° C) v.s. first arrival date (Julian days) on Mustang Island, Texas for Double-crested
Cormorant (top; P < 0.001) and Least Tern (bottom; P < 0.05) from 1978 to 2005.

“summer” temperatures year-round. All three
summer residents exhibited a decrease in duration
of stay on Mustang Island, although the only
significant trend was tor the Least Tern. Contrac-
tion of the summer season is opposite of what
would be expected in a warming climate where
summers should become warmer and longer.
However, variability in migration phenologies

has been observed in previous studies, where
some species exhibit the opposite trend (Parmesan
and Yohe 2003). All three terns are known to
breed locally, and a decrease in amount of time on
Mustang Island may translate to a shortened
breeding season. Ramifications of an abridged
breeding sea.son for these species are unknown
and warrant further investigation. Studies on

124

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. I, March 2010

FIG. 5. Mean April (departure) temperature in Port Aransas, Texas vs. the last date of departure for Eared Grebe from
Mustang Island, Texas from 1978 to 2005 (P < 0.01 ).

breeding success, clutch size, and food resources
should be initiated, especially for the Least Tern
which is listed as either threatened or endangered
throughout its range.

A major concern in conducting phonological
studies is the effect that small population size may
have on the ability to detect arrival and departure
dates (Sparks 1 999, Tryjanowski and Sparks 2001 ).
Conclusions on whether or not population size
influences detection ability have been mixed and
focused on whether a change in population size
affects detection of the first anlval date. The
variability in airival and departure dates (repre-
sented by the standard deviation around the 27-year
mean; Table 3) in our study was compared to the
mean daily abundance of each species. Mean daily
abundance among these six species ranged from
2.77 birds/day for Gull-billed Terns to 81 .84 birds/
day for Least Terns. No significant correlations
were found between mean daily abundance and
arrival date (P — 0.28, n — 6) or departure date {P
— 0.24, n = 6). This suggests that, at least for this
study, population size (mean daily abundance) did
not cau.se variability in arrival and departure dates,
and did not inlluence the ability of the observer to
detect migration phenologies for the.se species.

Most studies of avian migration phenology have
occurred in Britain (Sparks 1999, Cotton 2003) or
elsewhere in Europe (Gordo and Sanz 2006, Jonzen
et al. 2006) and have focused on passerine species.
Information concerning breeding or migration
phenology of North American birds is sparse
(Brown et al. 1999, Butler 2003). Efforts aie needed
to as,sess changes in breeding and migration
phenology of North American bird species to help
quantify the effects of climate change. These data
will allow compari.sons with European studies to
examine if climate change is similarly or dispropor-
tionately affecting birds on different continents.
Studies of coastal birds are also needed to allow
comparisons across a greater breadth of avian fauna.

ACKNOWLEDGMENTS

The authors thank K. H. Dunton, G. .1. Holt, and A. F.
Ojanguren for helpful suggestions and critical review of
earlier drafts of this manuscript. An earlier draft of this
manuscript was also greatly improved through suggestions
by C. E. Braun and two anonymous reviewers. CRF thanks
his wife and family for their support and encouragement.
Funding for analysis of the data was provided by the Sibyl
Ranfranz and Kenneth F. Wells Endowed Excellence Fund
in Marine Science. This is eonlribution 1510 of The
University of Texas Marine Science Institute.

Foster et al. • COASTAL BIRD PHENOLOGY AND CLIMATE CHANGE

125

LITERATURE CITED

Ahas, R. 1999. Long-term phyto-, ornitho-, and ichthyo-
phenologicul tinie-serie.s analyses in Estonia. Interna-
tional Journal of Biometeorology 42:1 19-123.

Beebee, T. J. C. 1995. Amphibian breeding and climate.
Nature 374:219-220.

Bradley, N. L., A. C. Leopold, J. Ross, and W.
Huffaker. 1999. Phenological changes reflect climate
change in Wisconsin. Proceedings of the National
Academy of Sciences of the USA 96:9701-9704.

Brown, J. L., S. Li, and N. Bhagabati. 1999. Long-term
trend toward earlier breeding in an American bird: a
response to global warming? Proceedings of the
National Academy of Sciences of the USA 96:5565-
5569.

Butler, C. J. 2003. The disproportionate effect of global
warming on the arrival dates of short-distance
migratory birds in North America. Ibis 145:484^95.

Cotton, P. 2003. Avian migration phenology and global
climate change. Proceedings of the National Academy
of Sciences of the USA 100:12219-12222.

Gordo, O. and J. J. Sanz. 2006. Climate change and bird
phenology: a long-term study in the Iberian Peninsula.
Global Change Biology 12:1993-2004.

Grabherr, G., M. Gottfried, and H. Paull. 1994.
Climate effects on mountain plants. Nature 369:448.

IPCC. 2001. Climate Change 2001: third assessment of the
Intergovernmental Panel on Climate Change. Cam-
bridge University Press, Cambridge, United Kingdom.

JoNZEN, N., A. Linden, T. Ergon, E. Knudsen, J. O. Vik,
D. Rubolini, D. Piacentini, C. Brinch, F. Spina, L.
Karlsson, M. Stervander, a. Andersson, j.
Waldenstrom, A. Lehikoinen, E. Edvardsen, R.
SoLVANG, AND N. C. Stenseth. 2006. Rapid advance
of spring arrival dates in long-distance migratory birds.
Science 312:1959-1961.

McCarty, J. P. 2001. Ecological consequences of recent
climate change. Conservation Biology 15:320-331.
Parmesan, C. and G. Yohe. 2003. A global ly-coherent
footprint of climate change across natural systems.
Nature 421 :37^2.

Parmesan, C., N. Ryrholm, C. Stefanescus, J. K. Hill,

C. D. Thomas, H. Descimon, B. Huntley, L. Kaila,
j. Killberg, T. Tammaru, W. J. Tennent, J. A.
Thomas, and M. Warren. 1999. Poleward shifts in
geographic ranges of butterfly species associated with
regional warming. Nature 399:579-583.

Roy, D. B. and T. H. Sparks. 2000. Phenology of British
butterflies and climate change. Global Change Biology
6:407^16.

Sparks, T. H. 1999. Phenology and the changing pattern of
bird migration in Britain. International Journal of
Biometeorology 42:134-138.

Sparks, T. H., F. Bairlein, J. G. Bojarinova, O. Huppop,
E. A. Lehikoinen, K. Rainios, L. V. Sokolov, and

D. Walker. 2005. Examining the total arrival
distribution of migratory birds. Global Change Biol-
ogy 1 1 :22-30.

SYSTAT Software Inc. 2004. SYSTAT Version 11.0.

SYSTAT Software Inc., Richmond, California, USA.
Tryjanowski, P. and T. H. Sparks. 2001. Is the detection
of the first arrival date of migratory birds influenced
by population size? A case study of the Red-backed
Shrike Lanins collwio. International Journal of
Biometeorology 45:217-219.

Walther, G. R., E. Post, P. Convey, A. Menzel, C.
Parmesan, T. J. C. Beebee, J.-M. Fromentin, O.
Heogh-Guldberg, and F. Bairlein. 2002. Ecological
responses to recent climate change. Nature 416:389-
385.

WuETHRiCH, B. 2000. How climate change alters the
rhythms of the wild. Science 287:793-795.

The Wilson Journal of Ornithology 122( 1 ): 126-1 34, 2010

NEST-SITE LIMITATION OF SECONDARY CAVITY-NESTING BIRDS
IN EVEN-AGE SOUTHERN PINE FORESTS

KARL E. M1LLER‘-

ABSTRACT. — 1 used a randomized, replicated, controlled study design to test the hypothesis that nest-site availability
limits breeding densities ot secondary cavity nesting species in even-age southern pine (Pinus spp.) forests. Breeding
densities of secondary cavity nesters increased significantly on treated plots after nest boxes were introduced. Total number
of nesting attempts also increased several told post-treatment. These data indicate nest-site availability was a limiting factor
tor breeding densities of secondary cavity nesting species. The response ot individual species to nest boxes ranged from
moderately high (Great Crested Flycatcher [Myuirchus crinitiis]), to low (Tufted Titmouse [Baeolophus hicolor]. Eastern
Bluebird [Siaha sialis]) to no response (Carolina Wren \Thryothonis htdovicianus]). At least three factors accounted for
interspecific ditferences in this study: different levels of reliance on cavities excavated in snags, different body sizes, and
differences in local population densities. The large number of unoccupied nest boxes (only 9% were used for nesting)
suggests secondary cavity nesters were not limited solely by cavity availability but also by habitat quality. Prescribed
burning appeared to facilitate discovery and use of nest boxes by birds in this study, consistent with the hypothesis that nest-
site limitation is mitigated by habitat structure. Received 4 September 2007 . Accepted 3 September 2009.

Nest sites are often assumed to be the primary
factor limiting densities of secondary (i.e., non-
excavating) cavity-nesting birds (e.g., von Haart-
man 1957, 1971; Thomas et al. 1979; Cody 1985).
However, tests of the nest-site limitation hypoth-
esis by manipulative experiments, which include
knowledge of initial conditions, controls, and
replication (Hairston 1989), are still rare. Newton
(1994) reviewed the evidence for nest-site limita-
tion in cavity-nesting birds and found most to be
circumstantial. Newton (1994) found only six
studies that added or subtracted nest sites for
cavity-nesting birds where controls and pre- and
post-manipulation data were included. Only two of
these studies (Enemar and Sjbstrand 1972, Brawn
and Baida 1988) had replicate study plots. Other
common bia.ses in tests of the nest-site limitation
hypothesis include lack of spatial independence of
experimental and control plots (e.g.. East and
Peirins 1988, Beis.senger and Bucher 1992), and
inadequate controls because of placement of study
plots in dissimilar plant communities (e.g., Biaish
1983). Inadequate experimental controls can leave
apparent experimental effects open to alternative
explanations such as temporal and spatial variation
in habitat quality.

Most nest-site limitation studies have been
conducted on a small number of species inhabit-

' Department of Wildlife Ecology and Conservation,
University of Florida. Gainesville, FI. 3261 I, USA.

’Current address: Fish and Wildlife Research Institute,
Florida Fish and Wildlife Conservation Commission, I lO,*)
Southwest Williston Road, Gainesville, Ft. 32601, USA;
e-mail: karl . m i I lcr@ myfwc.com

ing simplified European forests, especially Great
Tit (Pants major) and European Pied Flycatcher
(Ficedula hypoleuca), and often focusing on only
one species at a time (Wesolowski 2007).
Controlled nest box experiments have not been
conducted with cavity-nesting birds in pine (Pinus
spp.) forests of the southeastern United States.

Research from other forest systems indicates
that cavity density is positively con'elated with
tree density and tree age (van Balen et al. 1982),
and natural cavities formed from decay are more
abundant in hardwood trees than in conifers
(Waters et al. 1990). Thus, secondary cavity
nesters should be more likely to be nest-site
limited in younger forests and in forests dominat-
ed by conifers than in older, more structurally
complex forests with abundant natural cavities
(Brawn and Baida 1988, Waters et al. 1990,
Wesolowski 2007).

I predicted that secondary cavity nesting
species would be nest-site limited in pine forests
of the southeastern United States, where silvicul-
tural practices are considered to limit the
availability of dead trees that provide nest sites
for cavity nesters (McComb et al. 1986, Jackson
1988). The objective of my study was to test the
hypothesis that nest-site availability limits breed-
ing densities of secondary cavity nesting species
in even-age managed pine forests, which are the
predominant forest type in the southeastern
coastal plain.

METHODS

Study Plots and Experimental Design. — Field-
work was conducted on 12 study plots in 35-

126

Miller - NEST-SITE LIMITATION IN NON-EXCAVATORS

127

40 year-old even-age slash pine (Finns elliottii)
plantations at Camp Blanding Training Site, Clay
County, Florida. I distributed study plots evenly
between mesic “tlatwoods” sites and drier
“sandhill” sites to control for variation in site
productivity. Flatwoods plantations were charac-
terized by moderately to poorly drained soils and
a dense shrub layer of gallberry (Ilex glabra), saw
palmetto (Serenoa repens), and ericaceous shrubs
(Abrahamson and Hartnett 1990). Sandhill plan-
tations occLiiTed on sites originally dominated by
longleaf pine (P. palustris) (Myers 1990) and
were characterized by well-drained sandy soils
and a more open shrub layer. I measured live
vegetation within each study plot at point-count
stations (Martin et al. 1997) to develop stand
profiles. Flatwoods plantations were densely
stocked with a basal area (T ± SE) of 31.0 ±

2.2 m%a. Sandhill plantations were moderately
stocked with a basal area of 24.5 ± 2.5 mVha. All
plantations had been unburned for >3 years.

I counted snags >10 cm diameter at breast
height (dbh) on two 50 X 400 m transects within
each stand. Snags were abundant (17.7 ± 2.0 [x ±
SE] snags/ha on flatwoods plots, 14.8 ± 2.1
snags/ha on sandhill plots) relative to the average
of 6.2 snags/ha reported for pine plantations on
public land throughout Elorida (McComb et al.
1986). Most snags were too small for excavation
by woodpeckers; <2% were ^25 cm dbh (Miller
2000). More than 97.5% of all snags counted were
slash pines with the remainder comprising turkey
oak (Quercus laevis) and longleaf pine.

Study plots were of equal size (10 ha) and
dimensions (250 X 400 m) and as far as possible
(typically >75 m) from dirt roads and other
openings. Study plots were broadly dispersed to
encourage spatial independence; the northernmost
study plot was 13.1 km distant from the south-
ernmost plot, and the westernmost study plot was

9.2 km distant from the easternmost plot. Nearest-
neighbor distances between study plots averaged

1.2 km (range: 0.5— 2.6 km). Dense wetland forest
separated the two study plots, that were only
0.5 km apart.

I applied treatments randomly in a balanced
design (Table 1 ) after 2 years of breeding bird
surveys to document initial conditions (1995-
1996). Prescribed burns were conducted on two
flatwoods plots and two sandhill plots during
January-February 1997. Winter bums consumed
leaf litter and pruned or killed shrubs and
saplings, but did not cause tree mortality. I

TABLE I. Experimental design used for test of nest-
site limitation. Treatments (O = no manipulation; X =
nest boxes added; Y = nest boxes added after control
burn) were randomly applied within forest type. Clay
County, Florida. 1995-1998.

Plot #

1995

1996

1997

1998

Flatwoods

1

o

o

0

o

2

0

0

0

0

3

o

o

X

X

4

0

0

X

X

5

o

o

Y

Y

6

o

0

Y

Y

Sandhills

7

o

o

o

O

8

o

0

0

O

9

o

0

X

X

10

o

0

X

X

11

o

0

Y

Y

12

o

0

Y

Y

installed nest boxes during the first week of
March 1997 on all burned plots and on four of the
unburned plots, leaving four unburned plots
without nest boxes as controls (Table 1). Two
years of breeding bird surveys were conducted to
assess the effects of treatments (1997-1998).

Nest Boxes. — constructed 320 wood nest
boxes with dimensions that would accommodate
all species of secondary cavity nesters that breed
in Florida pinelands, except raptors and water-
fowl. Nest boxes differed in inside floor dimen-
sions (10.2 X 8.9 cm or 14.0 X 14.0 cm),
diameter of the entrance hole (3.5 or 5.1 cm),
and in height above ground (1.9 or 4.8 m). I
installed 40 nest boxes (4/ha) on each nest-box
plot, at 50-m intervals in a standardized an’ay. The
nest box array was a balanced, three-factor design:
box size (small vs. large) X hole size (naiTOw vs.
wide) X box height (low vs. high). 1 standardized
the orientation of nest boxes, placing them with
the entrance hole oriented east by southeast,
because cavity openings facing east, southeast,
.south, or southwest are typical for many cavity
nesting species (Conner 1975, Pinkowski 1976,
McEllin 1979, Jackson 1994, Rendell and Ro-
bertson 1994).

Nest Searches and Nest Monitoring. — used
standard methods (Martin and Geupel 1993) to
search for nests in tree cavities on study plots
from early April through early July 1996-1998.
Each nest located was monitored regularly at 3-
4 day intervals to assess nesting status (Martin

128

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. I. March 2010

and Geupel 1993, Ralph et al. 1993). I used an
aluminum extension ladder to inspect the contents
ot each nest box once every 10-14 days during
April-May and once every 14-21 days during
June-early July 1997-1998. Nests in nest boxes
also were monitored regularly at 3^ day inter-
vals. A nesting attempt was defined as a nest
where >1 egg was laid.

Southern flying squirrels {Glaucomys volans),
snakes, and other non-avian occupants were not
discouraged from using nest boxes, because I
wanted to study cavities as a limiting factor for
birds under natural conditions. Squirrel “roost
sites” were defined as those nest boxes in which
flying squirrels were observed on >2 occasions
within a season.

Estimation of Bird Densities. — I used the
intensive point count method (Ralph et al. 1993,
Wilson et al. 1995) to estimate densities of
breeding birds in study plots. Intensive point
counts differ from extensive point counts in that
points are established relatively close together
within a nest search plot or mist net plot, and
count data within each plot are pooled for
analyses (Ralph et al. 1993). Point-count survey
methods followed Ralph et al. (1993): counts were
conducted within 3 hrs post-sunrise; no counts
were made during rain or strong winds; birds
counted within a fixed-width circle were not
recounted if they were observed moving to an
adjacent fixed-width circle. 1 conducted all point-
count surveys to eliminate inter-observer bias. 1
.selected a 9-min count period because pilot
studies indicated that 84% of the species in this
system were detected within the first 9 min
(unpubl. data); a longer count period would
increase the likelihood of double counting (Scott
and Ramsey 1981). Data from pilot studies
demonstrated the effective detection distance
was >35m for most species, including all
cavity-nesting species. 1 established six 35-m
radius point-count stations spaced 125 m apart
within each plot (Hutto et al. 1986). Each point-
count station was sampled three times per year al
intervals of 2.5 weeks between 20 April and the
second week of June.

Analyses of Experimental Effects. — I pooled
fixed-radius point count data during each visit to
a study plot because each plot was considered an
experimental unit for analyses (Ralph et al. 1993).

I used the maximum pooled count for each species
within each plot for compari.sons among plots,
treatments, and years. Count data were not

transformed for analyses because of homogeneity
of variances across treatments and years. I
analyzed experimental effects with a mixed model
analysis-of-variance (ANOVA) using generalized
least squares estimation (PROC MIXED, SAS
Institute Inc. 1997). The model included four
factors: treatment (control vs. nest box vs. burned
with nest box), forest (flatwoods vs. sandhills),
year, and a random plot effect. 1 modeled the
differences in count data between pre- (1995,
1996) and post-treatment years (1997, 1998) using
an average of the 2 pre-treatment years as a
covariate (sensitivity analysis indicated that pre-
treatment years did not differ). Post-treatment
years were treated separately in the ANOVA
because I wanted to examine if responses to
treatments changed over time. Goodness-of-fit
criterion (Schwarz’s Bayesian Criterion) indicated
a model including interaction terms (treatment X
forest, treatment X year) best described the data
for each species. Pairwise differences of least-
squares means (PROC MIXED) identified the
relationships between variables for significant
factors in the model, and P-values < 0.10 were
considered significant.

RESULTS

Response of Cavity-nesting Bird Populations to
Nest Box Treatments. — I detected 14 species of
cavity nesters during point-count surveys. Three
species of woodpeckers (Red-bellied Woodpecker
[Melanerpes carolinus]. Downy Woodpecker [Pi-
coides puhescens]. Northern Elicker [Colaptes
auratus]), two species of weak excavators (Car-
olina Chickadee [Poecile carolinensis], and
Brown-headed Nuthatch \Sitta pusilla\), and five
species of non-excavators (Eastern Screech-Owl
\Megascops asio\. Great Crested Elycatcher
\Myiarchus crinitus]. Tufted Titmouse [Baeolo-
plius hicolor], Carolina Wren \Thryotiioriis liidovi-
f77//?//.v|, and Eastern Bluebird {Sialia sialis]) were
documented nesting on study plots. 1 categorized
Carolina Chickadees and Brown-headed Nuthatch-
es as primary (i.e., excavating) cavity nesters
because of their strong affinity for cavity excava-
tion in this locality. The other four species detected
(Barred Owl \Strix varia]. Red-headed Woodpeck-
er [M. erythroceplialiis], Red-cockaded Wood-
pecker \P. borealis], and Pileated Woodpecker
\Dryocopus pileatus]) were rare (recorded on < 1 %
of surveys) and did not nest on study plots.

Counts of primary cavity nesting species did
not vary among treatments, either for all species

Miller - NEST-SITE LIMITATION IN NON-EXCAVATORS

129

combined (P = 0.23) or for any individual species
{P > 0.10). No primary cavity nester used nest
boxes for nesting. Red-bellied and Downy wood-
peckers accounted for 93% of all woodpeckers
counted within 35-m radius point counts.

Counts of secondary cavity nesting species
varied among treatments. The mixed model
ANOVA for all species combined indicated that
treatment was the only significant factor (F —
7.57, df = 2,5; P = 0.03) explaining the change in
counts between pre-treatment (1995, 1996) and
post-treatment years (1997, 1998); secondary
cavity nesters increased on treated plots but
decreased slightly on control plots (Fig. 1). The
treatment X year interaction term approached
statistical significance (F = 2.89, df = 2,9; P =
0.11), reflecting the different rates of response on
unburned and burned plots. Secondary cavity
nesters did not increase on unburned plots with
nest boxes (NB) until 1998 but increased nearly
twofold on burned plots with nest boxes (BNB) in
1997 (Fig. 1). Differences of least-squares means,
post-treatment, indicated that counts {x ± SE)
were greater (F = 0.01) in NB plots (5.5 ± 0.7)
than in control plots (2.9 ± 0.2) and greater (F =
0.02) in BNB plots (5.0 ± 0.4) than in control
plots. Counts did not differ between NB plots and
BNB plots.

Interspecific Variation in Response to Nest
Box Treatments. — ^Treatment was the only signif-
icant factor in the mixed model (F = 4.42, df =
2,5; F = 0.07) for Great Crested Flycatchers. This
species increased on treated plots after treatments
were applied and decreased slightly on control
plots (Fig. 2). Counts in NB plots (2.6 ± 0.4)
post-treatment were greater (F = 0.02) than
counts in control plots (1.3 ± 0.3), and counts
in BNB plots (2.9 ± 0.4) were greater (F = 0.03)
than counts in control plots. Great Crested
Flycatcher numbers reached their peak during
the study in the BNB plots during the first year
after boxes were introduced (Fig. 2).

Tufted Titmou.se response was similar to that of
Great Crested Flycatchers; counts increased three-
fold on treated plots after treatments were applied
(Fig. 2). Tufted Titmice were not recorded in
fixed-radius point counts before nest boxes were
introduced on any BNB plots in 1995 and on only
one BNB plot in 1996 (Fig. 2). Treatment was not
significant (F = 0.37, df = 2,5; F = 0.70), in part
because of the small samples involved.

Eastern Bluebirds were not recorded on study
plots until after nest boxes were introduced

Year

FIG. 1. .Mean number of secondary cavity-nesting (SCN)
birds on study plots before (1995-1996) and after (1997-
1998) treatments (C = control, NB = unburned plots with nest
boxes, BNB = burned plots with nest boxes). Bars equal
one SE.

(Fig. 2), which suggests they responded to the
nest boxes. No factors were significant, but the
statistical analysis had low power because of
small samples.

Carolina Wrens had no response to treatments
(Fig. 2). The Carolina Wren was the only
secondary cavity nester species that did not
exhibit an immediate response to nest boxes in
BNB plots in 1997. Forest type was the only
significant factor (F = 9.62, df = 1,9; F = 0.01)
explaining the differences in wren counts; wrens
were more common in flatwoods plots than in
sandhill plots, regardless of treatment.

Eastern Screech-Owls were not monitored
effectively by point-count surveys and sample
sizes were insufficient for statistical analyses.

Patterns of Nest Box Occupancy. — recorded
57 nesting attempts in nest boxes during 1997-
1998, including 33 (58%) Great Crested Flycatch-
er nests, 13 (23%) Tufted Titmouse nests, six
(11%) Eastern Bluebird nests, three (5%) Car-
olina Wren nests, and two (3%) Eastern Screech-
Owl nests. Screech-Owls were only able to use
those nest boxes that had entrance holes enlarged
by squirrels. Red-bellied Woodpeckers also were
documented roosting in 19 nest boxes; these nest
boxes contained fecal debris and typically were
near (<50 m) an active Red-bellied Woodpecker
nest in a pine snag. Overall avian occupancy of
ne.st boxes, including woodpecker roost sites,
increased from the first year (8%) to the second
year (16%). Flying squirrels used 70 (22%) and
56 (18%) nest boxes as roost sites in 1997 and
1998, respectively.

The total number of secondary cavity nests on
study plots increa.sed twofold in 1997 (28 total
nests) and threefold in 1998 (42 total nests)

130

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 1. March 2010

A

B

1995 1996 1997 1998

c

CO

<

LU

4

3

2

1

0

1995

1996

1997

1998

D

« 4 -1

1995 1996

1997 1998

FIG. 2. Mean number of (A) Great Crested Flycatchers (GCFLs), (B) Tufted Titmice (TUTls), (C) Eastern Bluebirds
(EABLs), and (D) Carolina Wrens (CAWRs) on study plots before (1995-1996) and after (1997-1998) treatments (C =
control, NB = unburncd plots with nest boxes, BNB = burned plots with nest boxes). Bars equal one SE.

Miller - NEST-SITE LIMITATION IN NON-EXCAVATORS

131

because of the addition of nest boxes (Table 2).
Nesting activity on NB plots lagged behind that
on BNB plots; cavity nesters made little use of
nest boxes on NB plots until the second year after
the nest boxes were in place (Table 2).

DISCUSSION

Response of Cavity-nesting Bird Populations to
Nest Box Treatments. — Numbers of secondary
cavity nesters recorded during the breeding season
increased significantly on treated plots after nest
boxes were introduced. The number of nesting
attempts on treated plots also increased several
fold. These data indicate that nest-site availability
was limiting breeding densities of secondary
cavity nesters. It is unlikely that birds moved
from control to treated plots because interplot
distances were many times larger than the
territory sizes of these species (Grubb and
Pravosudov 1994, Lanyon 1997, Gowaty and
Plissner 1998). Even if a few individuals moved
between plots, the slight decline in number of
cavity nests on control plots (from 4 in 1996 to 3
in each of the following years; Table 2) cannot
account for the large increase in the number of
total nests on treated plots during 1997-1998.

Point-count surveys indicated the bird commu-
nity was lacking in large-bodied woodpeckers
(i.e.. Northern Flicker, Pileated Woodpecker) that
excavate cavities used by secondary cavity
nesters. The Northern Flicker is the primary
creator of large cavities used by other species in
longleaf pine forest (Blanc and Walters 2008),
especially larger species such as American Kestrel
(Falco sparverius), which was absent from my
study site.

TABLE 2. Nests of secondary cavity-nesting birds
(Eastern Screech-Owl, Great Crested Flycatcher, Tufted
Titmouse, Carolina Wren, Eastern Bluebird) in tree cavities
and nest boxes, before (1996) and after (1997-1998)
treatments were applied. Clay County, Florida.

Treatment

Pre-

treatmeni

Post-treatment years

19%

1997

1998

Cavity Box

Cavity

Box

Cavity Box

Control plot.s in = 4)

4

3

a

3

Nest box plots [n = 4)

6

3

5

2 16

Burned nest box plots

(n = 4)

4

1

16

1 20

Totals

14

7

21

6 36

Nesl boxes not available.

Interspecific Variation in Nest-site Limitation. —
Population response to nest boxes by individual
species ranged from moderately high (Great
Crested Flycatcher) to low (Tufted Titmouse,
Eastern Bluebird) to no response (Carolina Wren).
At least three factors accounted for interspecific
differences in this study; different levels of
reliance on cavities excavated in snags, different
body sizes, and differences in local population
densities.

Great Crested Flycatchers responded more
favorably to addition of nest boxes in pine
plantations than Tufted Titmice and Eastern
Bluebirds, indicating they were more nest-site
limited. Similarly, Ash-throated Flycatchers
{Myiarchus cinerascens) increased from 0 to three
breeding pairs after 20 nest boxes were added to a
16-ha plot of riparian forest that was devoid of
cavities (Brush 1983). Myiarchus nest-site selec-
tion is sufficiently generalized to allow use of
hollow branches and other natural cavities if they
are of sufficient size (Lanyon 1997), but natural
cavities available in 35--40 year-old slash pine
trees apparently were too small for this species.
Flycatchers relied heavily on woodpecker-exca-
vated cavities; only one (6%) of 18 flycatcher
cavity nests monitored was in a natural cavity in a
living pine tree (Miller 2000).

Tufted Titmice increased in numbers following
introduction of nest boxes but not as dramatically
as Great Crested Flycatchers. Nest monitoring
indicated that three of six (50%) titmouse nests in
slash pine trees were in shallow slit cavities
formed between forked pine trunks (Miller 2000).
The smaller body size of the Tufted Titmouse may
have allowed it to be more opportunistic in its use
of natural cavities than the Great Crested
Flycatcher.

Eastern Bluebirds showed only slight increases
on treated plots after nest boxes were introduced.
However, Eastern Bluebirds were rare on the
study plots and it was difficult to assess whether
they were nest-site limited. Local or regional
population densities for rare species may be
insufficient to provide surplus or floater individ-
uals that would breed if more nest sites were
available (Newton 1998).

Carolina Wrens had no response to treatments
and, despite their abundance, rarely used nest
boxes. Nest searches indicated the Carolina Wren
was a facultative cavity nester at this locality,
typically building its nests in crevices in brush
piles, stumps, and shrubs or weaving them into

132

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. I. March 2010

palmetto fronds (Miller 2000). Only one wren nest
was found in a tree cavity during 1996-1998, a
‘chimney’ nest {sensu Wesolowski 2007) built in
a low hollow stump.

Ejf’ect of Habitat on Nest-site Limitation. — ^The
large number of unoccupied nest boxes (only 9%
were used for nesting) suggests that secondary
cavity nesters were not limited solely by cavity
availability but also by habitat quality (e.g.,
unsuitable habitat structure, low food availabili-
ty). Studies in western forests have demonstrated
a greater response to nest boxes. For example,
bird use of nest boxes in thinned ponderosa pine
(P. ponderosa) forest in Arizona increased from
10% in the first year of their availability to 58%
in the fourth year (Brawn and Baida 1988). Nest
box occupancy rates in mixed oak-pine wood-
lands in California increased from 25% in the first
year to 68% in the sixth year (Purcell et al. 1997).
Extensive tree cutting during 1999-2000 curtailed
my experiment, but I continued monitoring three
plots unaffected by forestry operations and nest
box occupancy rates during 1999-2000 did not
exceed 13%. Thus, low nest box occupancy rates
persisted for several years, indicating that habitat
quality may .set an upper limit on breeding bird
density in these young forests.

Populations of several cavity nester species,
especially Eastern Bluebird, were low in these
densely-stocked slash pine plantations, possibly
because the forest may not match the niche
structure {sensu Grinellian niche, James et al.
1984) prefen'ed by these species. Eor example, the
Eastern Bluebird prefers forest gaps, forest edges,
and other plant communities that have few trees
and sparse ground cover (Gowaty and Plissner
1998).

Birds in flatwoods plantations and sandhill
plantations responded similarly to treatments
despite modest site differences in basal area and
snag availability. Regardless of forest type, the
respon.se to nest boxes was most rapid on burned
plots, probably becau.se birds were attracted to
areas where winter burns had opened up the shrub
layer and iinderstory. Populations of many
secondary cavity nesting species respond favor-
ably to open microhabitats created by burning
(e.g., Bock and Lynch 1970, Pinkowski 1976,
Hutto 1995, Krei.sel and Stein 1999, Provencheret
al. 2002, Allen et al. 2006).

Other Factors Potentially Limiting, Cavity-
nestOip, Bird Populations. — Breeding populations
of cavity-nesting birds potentially can be limited

by other factors including territoriality and
interspecific competition for nest sites (Newton
1979, 1998). However, ten'itorial behavior at
present bird densities is unlikely to have prevent-
ed individuals from breeding. Eor example,
territory mapping on study plots (unpubl. data)
demonstrated that Great Crested Elycatcher terri-
tories frequently were not contiguous. The large
number of nest boxes unused by birds or flying
squirrels ( —two-thirds of the total each year) also
suggested little potential for interference compe-
tition among species. 1 rarely observed any
evidence of inter- or intra-specific usuipation of
nest boxes, although flying squiirels did depredate
some flycatcher nests (Miller 2002). Interference
competition for nest sites was not observed among
secondary cavity nesting species at natural nests
in tree cavities, although 1 did observe several
aggressive interactions between Red-bellied
Woodpeckers and Great Crested Elycatchers at
Red-bellied Woodpecker nest cavities.

Nest-site limitation should not be assumed for
all cavity-nesting bird species, given that some
species are more opportunistic in their choice of
nest sites than others. Reliance on cavities
excavated in dead wood by primary cavity-nesters
and larger body size were factors associated with
nest-site limitation. Supporting evidence comes
from studies in ponderosa pine forests, where
secondary cavity nesting species that nested
almost exclusively in cavities excavated by
primary cavity nesters (i.e., Violet-green Swallow
[Tachycineta thalassina]\ Pygmy Nuthatch [Sitta
pygmaea]-. Western Bluebird [Sialia me.xicana))
exhibited greater population responses to nest
boxes than species that nested primarily in natural
cavities and crevices (Brawn and Baida 1988).

Nest-site limitation cannot be assumed for all
plant communities and locales. Tree cavities are
most abundant in older forest systems, which tend
to be structurally more complex (Aitken and
Martin 2007, Wesolowski 2007). Results of my
study provide additional support for the hypoth-
esis that secondary cavity-nesting birds are most
nest-site limited in younger forests and in forests
dominated by conifers (Brawn and Baida 1988,
Waters et al. 1990, Wesolowski 2007).

Pre.scribed burning appeared to facilitate dis-
covery and Li.se of nest boxes by birds in my study,
which is consistent with the hypothesis that nest-
site limitation is mitigated by habitat structure
(Brawn and Baida 1988). If increasing cavity-
nesting bird populations is a management goal in

Miller - NEST-SITE LIMITATION IN NON-EXCAVATORS

133

t'ire-mainlained habitats, such as southern pine
forests, managers may find that addition of nest
boxes to newly burned sites will result in the
highest local densities of secondary cavity-nesting
birds, at least in the short-term.

ACKNOWLEDGMENTS

My research was suppoited by funding from the Florida
Fish and Wildlife Conservation Commission, Florida Army
National Guard. University of Florida. North American
Bluebird Society, and a Student Research Grant from
Sandpiper Technologies Inc. I thank M. P. Moulton for
advice and encouragement throughout this project. G. A.
Jones, M. I. Williams, S. A. Johnson, and Annemarie van
Doom assisted in the field. K. M. Portier assisted with
statistical analy.ses. L. A. Blanc, D. L. Leonard, D. J. Levey,
M. P. Moulton, K. E. Sieving, G. L. Slater, and an
anonymous reviewer made constructive comments on earlier
versions of the manuscript. I dedicate this paper to my father,
J. H. Miller, who taught me the value of hard work.

LITERATURE CITED

Abrahamson, W. G. and D. C. FIartnett. 1990. Pine
flatwoods and dry prairies. Pages 103-149 in Ecosys-
tems of Florida (R. L. Meyers and J. J. Ewel, Editors).
University of Central Elorida Press, Orlando, USA.
Aitren, K. E. H. and K. Martin. 2007. The importance of
excavators in hole-nesting communities: availability
and use of natural tree holes in old mixed forests of
western Canada. Journal of Ornithology 148 (Supple-
ment 2):S425^34.

Allen. J. C., S. M. Krieger, J. R. Walters, and J. A.
Collazo. 2006. As.sociations of breeding birds with
fire-influenced and riparian-upland gradients in a
longleaf pine ecosystem. Auk 123:1 1 10-1 128.
Beissinger, S. R. and E. FI. Bucher. 1992. Sustainable
harvesting of parrots for conservation. Pages 73-1 15
in New World parrots in crisis: .solutions in conserva-
tion biology (S. R. Beissinger and N. F. R. Snyder,
Editors). Smithsonian Institution Press, Washington,
D.C., USA.

Blanc. L. A. and J. R. Walters. 2008. Cavity-nest webs
in a longleaf pine ecosystem. Condor 1 10:80-92.
Bock, C. E. and J. F. Lynch. 1970. Breeding bird
populations of burned and unburned conifer forests
in the Sierra Nevada, Condor 72:182-189.

Brawn. J. D. and R. P. Bai.da. 1988. Population biology
of cavity nesters in northern Arizona: do nest sites
limit breeding densities? Condor 90:61-71.

Brush. T. 1983. Cavity use by secondary cavity-nesting birds
and response to manipulations. Condor 85:461-466.
Cody, M. L. 1985. An introduction to habitat selection in
birds. Pages 3-56 in Habitat .selection in birds (M. L.
Cody, Editor). Academic Press. Orlando. Florida, USA.
Conner, R. N. 1975. Orientation of entrances to wood-
pecker nest cavities. Auk 92:371-374.

East, M. L. and C. M. Perrins. 1988. The effect of
nestboxes on breeding populations of birds in broad-
leaved temperate woodlands. Ibis 130:393-401.

Enemar, a. and B. Sjostrand. 1972. Effects of the
introduction ol Pied Elycatchers Ficedulu liypolenca
on the composition of a passerine bird community.
Ornis Scandinavica 3:79-89.

Gowaty, P. a. and j. H. Plissner. 1998. Eastern Bluebird
(Sialia sialis). The birds of North America. Number
381.

Grubb, T. C. and V. V. Pravosudov. 1994. Tufted
Titmou.se (Baeolophus hicolor). The birds of North
America. Number 86.

Hairston Sr., N. G. 1989. Ecological experiments:
purpose, design, and execution. Cambridge University
Press, Cambridge, United Kingdom.

Hutto, R. L. 1995. Composition of bird communities
following stand replacement fires in northern Rocky
Mountain conifer forests. Conservation Biology
9:1041-1058.

Hutto, R. L., S. M. Pletschet, and P. Hendricks. 1986.

A fixed-radius point count method for nonbreeding
and breeding season use. Auk 103:593-602.

Jackson, J. A. 1988. The southeastern pine forest
ecosystem and its birds: past, present, and future.
Pages 119-159 in Bird conservation 3 (J. A. Jack.son,
Editor). University of Wisconsin Press, Madison,
USA.

Jackson, J. A. 1994. Red-cockaded Woodpecker (Picoides
borealis). The birds of North America. Number 85.

James, F. C., R. F. Johnston, N. O. Wamer, G. J. Niemi,
and W. j. Boecklen. 1984. The Grinnellian niche of
the Wood Thrush. American Naturalist 124:17-30.

Kreisel, K, j, and S. j. Stein. 1999. Bird use of burned
and unburned coniferous forests during winter. Wilson
Bulletin 1 1 1 :243-250.

Lanyon, W. E. 1997. Great Crested Flycatcher (Myiarchus
crinitm). The birds of North America. Number 300.

Martin, T. E. and G. R. Geupel. 1993. Nest-monitoring
plots: methods for locating nests and monitoring
success. Journal of Eield Ornithology 64:507-519.

Martin, T. E., C. R. Paine, C. J. Conway, W. M.
Hochachka, P. Allen, and W. Jenkins. 1997.
BBIRD field protocol. Montana Cooperative Wildlife
Research Unit. University of Montana, Mis.soula.
USA.

McComb, W. C., S. A. Bonney, R. M. Sheffield, and N.
D. Cost. 1986. Snag resources in Florida — are they
sufficient for average populations of primary cavity-
nesters? Wildlilc Society Bulletin 14:40^8.

McEllin, S. M. 1979. Nest sites and population demog-
raphics of White-breasted and Pigmy nuthatches.
Condor 8 1 :348-352.

Miller, K. E. 2000. Nest-site limitation, nest predation,
and nest-site selection in a cavity-nesting bird
community. Dis.sertation. University of Florida, Gaines-
ville. USA.

Miller, K. E. 2002. Nesting success of the Great Crested
Flycatcher in nest boxes and in tree cavities: are nest
boxes safer from nest predation? Wilson Bulletin
1 14:179-185.

Myers, R. L. 1990. Scrub and high pine. Pages 150-193 in
Ecosystems of Elorida (R. L. Meyers and J. J. Ewel,

134

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 1, March 2010

Editors). University of Central Florida Press, Orlando,
USA.

Newton, I. 1979. Population ecology of raptors. T. and A.
D. Poyser, London, United Kingdom.

Newton, I. 1994. The role of nest sites in limiting the
numbers of hole-nesting birds; a review. Biological
Conservation 70:265-276.

Newton, I. 1998. Population limitation in birds. Academic
Press, San Diego, California, USA.

PlNKOWSKl, B. C. 1976. The use of tree cavities by nesting
Eastern Bluebirds. Journal of Wildlife Management
40:556-563.

Provencher, L., N. M. Gobris, L. A, Brennan, D, R.
Gordon, and J. L. Hardesty. 2002. Breeding bird
response to midstory hardwood reduction in Florida
sandhill longleaf pine forests. Journal of Wildlife
Management 66:641-661.

Purcell, K. L., J. Verner, and L. W. Oring. 1997. A
comparison of the breeding ecology of birds nesting in
boxes and tree cavities. Auk 1 14:646-656.

Ralph, C. J., G. R. Geupel, P. Pyle, T. E. Martin, and D.
F. Desante. 1993. Handbook of field methods for
monitoring landbirds. USDA, Forest Service, General
Technical Report PSW-GTR-144. Pacific Southwest
Re.search Station, Albany, California, USA.

Rendell, W. B. and R. j. Robertson. 1994. Cavity-
entrance orientation and nest-site use by secondary hole-
nesting birds. Journal of Field Ornithology 65:27-35.

SAS Institute Inc. 1997. SAS/STAT guide for personal
computers. Version 6.12. SAS Institute Inc., Cary,
North Carolina, USA.

Scott, J. M. and F. L. Ramsey. 1981. Length of count
period as a possible source of bias in estimating
bird densities. Studies in Avian Biology 6:409-
413.

Thomas, J. W., R. G. Anderson, C. Maser, and E. L.
Bull. 1979. Snags. Pages 60-67 in Wildlife habitats
in managed forest — the Blue Mountains of Oregon
and Washington (J. W, Thomas, Editor). USDA,
Forest Service, Agriculture Handbook Number 553.

VAN Balen, j. H., C. j. H. Booy, j. A. Franeker, and E.
R. OsiECK. 1982. Studies on hole-nesting birds in
natural nest sites. 1. Availability and occupation of
natural nest sites. Ardea 70:1-24.

VON Haartman, L. 1957. Adaptation in hole-nesting birds.
Evolution 11:339-347.

VON Haartman, L. 1971. Population dynamics. Pages
391-459 in Avian biology (D. S. Earner and J. R.
King, Editors). Volume 1. Academic Press, New
York, USA.

Waters, J. R., B. R. Noon, and J. Verner. 1990. Lack
of nest-site limitation in a cavity-nesting bird
community. Journal of Wildlife Management
54:239-245.

Wesolowski, T. 2007. Lessons from long-term hole-nester
studies in a primeval temperate forest. Journal of
Ornithology 148 (Supplement 2):S395-S405.

Wilson, C. W., R. E. Masters, and G. A. Buken-
HOFER. 1995. Breeding bird response to pine-
grassland community restoration for Red-cockaded
Woodpeckers. Journal of Wildlife Management 59:
56-67.

The Wilson Journal of Ornithology 1 22( I ): 1 35- 1 38, 2010

HOUSE SPARROWS ASSOCIATED WITH REDUCED CLIEE SWALLOW

NESTING SUCCESS

DOUGLAS R. LEASURE,' RAGUPATHY KANNAN,' AND DOUGLAS A. JAMES^'

ABSTRACT. — We quantified the impact of nesting and roosting House Spanows (Passer domesticus) on nesting success
of Cliff Swallows (Petroehelidon pyrrhonota) in colonies in western Arkansas in 2007 and 2008. Two sections of a large
swallow colony under a bridge with House Sparrows were compared in 2007 to two sections with little House Sparrow
usage. Nesting success of Cliff Swallows (percent of nests yielding at least 1 chick) was 61% in sections with low House
Sparrow activity, significantly higher than the 30% in sections with high House Sparrow activity. House Sparrows defended
a broad zone surrounding their nests from Cliff Swallow nesting attempts. We compared the proportion of nests used, clutch
sizes, and brood sizes of Cliff Swallows in two colonies in 2008, one with and one without House Sparrow activity. In the
colony without House Sparrow activity, 48% of old and new nests were used by swallows versus only 8% in the colony
with House Sparrows. Swallow clutch sizes were similar in the two colonies, but swallow brood sizes in the colony with no
House Sparrows were significantly higher, mean = 2.3 nestlings per nest (mode = 2; 75th percentile = 3) compared to 0.8
nestlings (mode = 0; 75th percentile = 1) in the colony with House Sparrows. This suggests Cliff Swallows are less
successful when House Spamows are present in colonies. Received 3 April 2009. Accepted 27 September 2009.

Cliff Swallow (Petroehelidon pyrrhonota) nest-
ing colonies can contain up to 3,500 gourd-shaped
mud nests adhering to rock overhangs, cliff faces,
and concrete structures such as bridges and
culverts (Brown and Brown 1995). Their breeding
range was historically limited to mountain ranges
in western North America but urbanization
produced a myriad of suitable Cliff Swallow
nesting sites, which allowed range expansion
eastward to include most of North America.
House Sparrows (Passer domesticus) were intro-
duced from Europe into North America in the
1850s and 1880s (Lowther and Cink 1992). Their
preference for human altered environments has
enabled them to spread throughout North and
Central America.

House Sparrows in North America are known
to have interfered with nesting of several species
of native birds (Barrows 1889, Estabrook 1907,
Jackson and Tate 1974, Weisheit 1989) including
Cliff Swallows (Brewster 1906, Burleigh 1930,
Grinnell 1937, Stoner 1939, Bent 1942, Buss
1942, Samuel 1969, Brown and Brown 1995).
Previous studies did not quantify the impact of
House Sparrows on Cliff Swallow nesting effort
and reproduction, or compare colonies with and
without House Sparrows. Our objective was to
quantify nesting effort and reproduction of Cliff
Swallows in the Arkansas River valley of western

' Department ot Biology, University of Arkansas-Fort
Smith, Fort Smith, AR 72913, USA.

“Department of Biological Sciences, University ot
Arkansas, Fayetteville, AR 72701, USA.

^CoiTesponding author; e-mail: djames@uark.edu

Arkansas. We compared nesting of Cliff Swal-
lows in several portions of a colony, each with
different levels of House Sparrow activity, and
also between two separate colonies, one with
House Sparrows and one without. The three
colonies were in urban or semi urban areas in or
around Fort Smith, Arkansas.

METHODS

We observed an active Cliff Swallow colony in
2007 under the state highway 59 bridge spanning
the Arkansas River near Barling, Arkansas in
Crawford County. The bridge height made
examination of nest contents impossible. Four
clusters each spanning 6 m were chosen for study
from over 2,100 nests: clusters Ah and Bh with
high House Span'ow activity had 125 and 46 old
swallow nests, clusters Cl and Dl with low House
Sparrow activity had 158 and 77 old swallow
nests, respectively. Subscript “L” denotes low
sparrow activity while subscript “H” denotes
high span-ow activity. These clusters were chosen
.so that Ah and Cl as well as Bh and Dl had
similar density of nest aggregations, and were on
opposite sides of the bridge adjacent to one
another. Observations began prior to swallow
nesting. Nesting in all clusters progressed simi-
larly due to breeding synchrony of Cliff Swallows
(Brown and Brown 1996).

The colony was visited twice each week. At
least one cluster was observed for at least 1 hr
during each visit. Extra observations were made
when nestlings began to peer from nests to ensure
documentation of all nests producing at least one
nestling. Observations were recorded on photo-

135

136

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. I. March 2010

graphs with each nest numbered. Nest observa-
tions included identity of bird species entering
nest, any nest building activity, nest defense
behaviors, provisioning trips, and direct nestling
observations. These data allowed quantification of
nesting success of House Sparrows and Cliff
Swallows in each cluster measured by proportion
of available nests that yielded at least one
nestling. Distance to the nearest successful Cliff
Swallow nest was measured for all House
Sparrow, Clift Swallow, and European Starling
(Sturnus vulgaris) nests. Distances between nests
were measured from the center of each nest
opening to the nearest 0.5 mm on photographs
and scaled by measuring a known distance on
each photograph.

Two colonies near Fort Smith, Sebastian Coun-
ty, Arkansas were studied in 2008, one with no
House Sparrow activity (colony El; 252 old
swallow nests prior to nesting) and one with high
House Sparrow activity (colony Fh; 196 old
swallow nests prior to nesting). Both colonies were
accessible by ladder and were on the underside of
highway overpasses over drainage canals. The
level of House Sparrow activity at each colony was
as.sessed prior to swallow nesting by observing
their pre-roost activity, recording House Sparrows
entering nests, and counting nests that were stuffed
with House Sparrow nesting material.

Nest contents were observed several times
throughout the nesting cycle (May-Jun) in
colonies El and Fh using an extension ladder,
dental mirror, and flashlight following Brown and
Brown (1996). Nest contents were observed
during the appropriate period of nesting at
respective colonies to ensure documentation of
eggs and nestlings. One hundred and five (41.7%)
of 252 nests in colony El, and 1 1 8 (60.2%) of 196
nests in Colony Fh were investigated. A ladder
was randomly positioned throughout the colony
and all nests within reach of each ladder position
were sampled. Nests were numbered with chalk
on the concrete bridge substrate and monitored
during visits. We recorded presence of nest
material, eggshells, living and dead nestlings,
and number of eggs for each nest. This informa-
tion was u.sed to quantify clutch size, brood size,
and the proportion of Cliff Swallow eggs yielding
successful nestlings. Active House Sparrow nests
were recorded, as were clutch and brood sizes,
whenever possible. Nest observations ceased
before nestlings were 10 days of age to prevent
premature fledging.

Chi-square contingency table analyses were
peifomied to examine the relationship between Cliff
Swallow nesting success and House Sparrow
activity. Mann-Whitney f/-tests were used to
examine the difference between nesting performance
(clutch size and brood size) in colonies El and Fh-
We used SPSS (2007) in all .statistical analy.ses.

RESULTS

House Sparrows roosting in clusters A through
D gathered every evening on shrubby vegetation
3 m from the bridge for over an hour before
entering their roost nests. The highest number
recorded in these pre-roost congregations was 58
on 9 March 2007. Swallow nests used for roosting
or nesting by sparrows were concentrated on the
side of the bridge that faced nearby shrubby
vegetation, which included clusters Ah and Bh.

Clusters Ah and Bh had high House Sparrow
activity and were grouped together for analyses.
We recorded Hou.se Sparrows entering or defend-
ing 38 nests and successfully nesting in seven
nests in clusters Ah and Bn. Sparrows nesting
outside our sample still defended nests within
sampled clusters. These clusters also supported
two European Starling ne.sts. Cliff Swallows
nested successfully (produced at least 1 nestling)
in 52 of 171 available nest cavities (30.4%).

Clusters Ci^ and Dl were grouped together for
analy.ses because of low House Sparrow activity.
These clusters faced away from nearby shrubs. We
recorded House SpaiTows entering or defending
eight ne.sts and successfully nesting in only one
nest. These clusters supported a single starling nest.
Cliff Swallows nested successfully in 144 of 235
available nest cavities (61.3%). Cliff Swallow
nesting success was significantly higher in clusters
with low sparrow activity when compared to those
with high .sparrow activity (x" = 37.767, df ^ \,P
< 0.001, Table 1). Mean (± SD) distance to the
nearest successful Cliff Swallow nest in clusters A-
D (Table 2) was 33.2 ± 28.6 cm for ne.sts ofHou.se

TABLE 1, ClitI Swallow (CS) ne.sling .success
(producing > 1 neslling) among high or low House
Sparrow activity. Data from 2007 arc presented first and
2008 data are in parentheses.

House Sparrow activity
High Low

Nests containing > 1 CS nestling 52 (.21) 144 (41)

Empty ne.sts 119(112) 91 (56)

Lcasiire et at. • HOUSE SPARROWS AND CLIFF SWALLOW REPROi:)UCTION

137

Sparrows, 16.6 ± 9.2 cm for starlings, and 1 1 .0 ±
4.7 cm for Cliff Swallows.

Colonies Ei, and Fh were observed in 2008.
Colony El was devoid of House Sparrows except
for one observation of a brief sparrow visit after
most swallow eggs had hatched. An Eastern
Bluebird (Sialia sialis) began nesting in an old
Cliff Swallow nest but abandoned the nest soon
after Cliff Swallows airived on 1 April 2008. This
colony had no shrubby vegetation nearby. Fifty
(47.6%) of 105 sampled nests were used by Cliff
Swallows for nesting. The mean (± SD) clutch size
was 2.64 ± 0.12 eggs (mode = 2; 75"’ percentile =
3, Fig. 1) and mean brood size was 2.31 ± 0.12
nestlings (mode = 2; 75"’ percentile = 3, Fig. 1). At
least 74% of eggs yielded nestlings in colony El.

Colony Fh had a high level of sparrow activity.
This colony was near shrubs that House Sparrows
used as pre-roosts throughout nesting. Thirty-one
nests were stuffed with House SpaiTOW nesting
material, but only 12 active House Sparrow nests
were recorded. The 12 active nests produced at
least 37 House Sparrow eggs. Only 10 of 118
sampled nests were active Cliff Swallow nests
(8.5%). Cliff Swallows nesting in this colony
(producing > 1 nestling) had lower success than
in colony El iy} — 50.319, df = \, P < 0.001,
Table 1). The mean (± SD) clutch size among
Cliff Swallows in Fh was 2.60 ± 0.27 eggs (mode
= 3; 75'" percentile = 3, Fig. 1) but the mean
brood size was only 0.83 ± 0.48 nestlings (mode
= 0; 75'" percentile = 1, Fig. 1). Only 20% of
Cliff Swallow eggs yielded nestlings and several
Cliff Swallow clutches were abandoned. Clutch
size of Cliff Swallows did not differ between
colonies E[ and Fh (C = 248, P = 0.966) but
brood size was significantly lower in colony Fh
than in colony El (C = 39, P = 0.005).

DISCUSSION

Cliff Swallow nesting success was low in both
years when House Sparrows were nesting within
the swallow colony. There was a 3 1 % reduction in

TABLE 2. Distance to the nearest Clid' Swallow (CS)
nest for House Sparrows, European Starlings, and ClitI
Swallows nesting in Clusters A-D in 2007.

Mean ± SD distance (cm) to nearest CS nest n

House Sparrow 33.2 ± 28.6 8

European Starling 16.6 ± 9.2 3

Cliff Swallow 1 1.0 ± 4.7 195

Cliff Swallow clutch size

FIG. 1. Cliff Swallow clutch and brood sizes from two
colonies in 2008. Colony F had high sparrow activity
(black) while colony E (white) had low sparrow activity.
Nests with brood size of zero are those in which eggs were
documented but no nestlings were produced.

Cliff Swallow nests producing at least one
nestling in 2007 in sections of a colony with
House Sparrows, and a 54% reduction in
proportion of Cliff Swallow eggs producing
nestlings in 2008 in the colony with House
Sparrows compared to sections without House
Sparrows. House Sparrow activity was greater in
nests facing and in close proximity to shrubs used
as pre-roosts in all colonies.

The nesting season of House SpaiTows in the
southern United States begins in March (Lowther
and Cink 1992), 1 month before that of Cliff
Swallows (Brown and Brown 1995). House
Sparrows in the present study roosted in Cliff
Swallow colonies throughout winter. Our earliest
observation of Cliff Swallows was 1 April 2008.
House Sparrows had established themselves in old
Cliff Swallow nests at this point and some were
incubating eggs, similar to reports by Buss (1942)
and Samuel (1969). House SpaiTOws defended a
broad zone around their nests often preventing
Cliff Swallows from nesting nearby (Table 2),

138

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 1, March 2010

which could explain the significant impact to Cliff
Swallow nesting associated with even a modest
number of House Sparrows. Samuel (1969)
reported that 33.6% of Cliff Swallow nests were
lost to House Sparrows with most of these nests
occupied before swallows began nesting, but
some were taken despite Cliff Swallow eggs or
nestlings being present.

House Sparrows have been observed pecking
eggs and nestlings of Barn Swallows (Hinmdo
rustica) (Weisheit 1989), but the enclosed nests of
Cliff Swallows made direct observation of this
behavior difficult. Our data suggests this occurred
in Cliff Swallow colonies as there was signifi-
cantly reduced Cliff Swallow production despite
similar clutch sizes in a colony with House
Sparrows compared to a colony without House
Sparrows. We observed a House Sparrow entering
an active Cliff Swallow nest despite aggressive
protest from the incubating swallow and we
observed dead Cliff Swallow nestlings and un-
hatched eggs on the ground beneath colonies with
House Sparrows; these observations are insuffi-
cient to implicate House Sparrows directly.

Buss ( 1942) reported successful House Sparrow
management that included shooting House Spar-
rows and removing Cliff Swallow nests in winter
to prevent sparrow roosting. We suggest that
removal of nearby shrubbery could discourage
House Sparrow roosting and nesting in Cliff
Swallow colonies as we observed more Hou.se
Sparrow activity at swallow nests facing and in
close proximity to shrubby vegetation used for
sparrow pre-roost congregations.

Our finding that House Sparrow activity was
negatively related to the number of active Cliff
Swallow nests, nesting success, brood size, and the
proportion of eggs yielding nestlings indicates that
Hou.se SpaiTows can significantly reduce Cliff
Swallow reproductive success in western Arkansas.

ACKNOWLEDGMENTS

This project was funded by a grant from the Student
Undergraduate Research Fellowship (SURF) of the Arkan-

sas Department of Higher Education. Peter Lowther
provided insightful review and suggested the nearest
neighbor analysis. Field work could not have been
completed without the assistance of Colby Marshall and
Mike Fielder. Charles R. Brown was a valuable source of
advice during development of the project.

LITERATURE CITED

Barrows, W, B. 1889. The English Sparrow in North
America, especially in its relations to agriculture. U.S.
Department of Agriculture, Division of Economic
Ornithology and Mammals Bulletin 1:1^05.

Bent, A. C. 1942. Northern Cliff Swallows. U.S. National
Museum Bulletin 179:468.

Brewster, W. 1906. The birds of the Cambridge region of
Massachusetts. Memoirs of the Nuttall Ornithological
Club 4:1.

Brown, C. R. and M. B. Brown. 1995. Cliff Swallow
{Hinmdo pyrrhonota). The birds of North America.
Number 149.

Brown, C. R. and M. B. Brown. 1996. Coloniality in the
Cliff Swallow: the effect of group size on social
behavior. University of Chicago Press, Chicago,
Illinois, USA.

Burleigh, T. D. 1930. Notes on the birdlife of northwest-
ern Washington. Auk 47:48.

Buss, I. O. 1942. A managed Cliff Swallow colony in
southern Wisconsin. Wilson Bulletin 54:153-161.
Estabrook, a. H. 1907. The present status of the English
Sparrow problem in the United States. Auk 24:129-
134.

GRtNNELL, J. 1937. The swallows of the Life Sciences
Building. Condor 39:206-210.

Jackson, J. A. and J. Tate Jr. 1974. An analysis of nest
box use by Purple Martins, House Sparrows and
starlings in eastern North America. Wilson Bulletin
86:435^49.

Lowther, P. E. and C. L. Cink. 1992. House Sparrow
{Passer domesticus). The birds of North America.
Number 12.

Samuel, D. E. 1969. House Sparrow occupancy of Cliff
Swallow nests. Wilson Bulletin 81:103-104.

SPSS Institute Inc. 2007. SPSS 16.0.1 for Windows.

SPSS Institute Inc., Chicago, Illinois, USA.

Stoner, D. 1939. Parasitism of the English Sparrow on the
Northern Cliff Swallow. Wilson Bulletin 51:221-222.
Weisheit, A. S. 1989. Interference by House Sparrows in
nesting activities of Barn Swallows. Journal of Eield
Ornithology 60:323-328.

The li ilson Joiiniiil oj Oniiihology 122( 1 ): 1 39-145, 2010

HOME-RANGE SIZE AND SITE TENACITY OE OVERWINTERING
EE CONTE’S SPARROWS IN A EIRE MANAGED PRAIRIE

HEATHER Q. BALDWIN,' CLINTON W. JESKE,- MELISSA A. POWELL,^
PAUL C. CHADWICK,' AND WYLIE C. BARROW JR.'

ABSTRACT. — We evaluated home-range size and site tenacity of Le Conte’s Sparrows {Ammodramus lecontii) during
winter 2002-2003 at Brazoria National Wildlife Refuge, Texas. Twenty-six wintering Le Conte’s Sparrows were
radiomarked in 1- and 2-year post-burn units, and monitored for ~10 days. Additionally, 1-ha plots on each 1-, 2- and 3-
year (n = 15) post-burn units were flush-netted once monthly. Telemetry results indicated Le Conte’s Sparrows were
sedentary during winter with a 50% probability mean home-range of 2.41 ha (72% < 1 ha) and a 95% probability mean
home range of 10.31 ha (44% < 1 ha and 55% < 1.5 ha). Home-range size did not differ between post-bum year 1 and 2
(P = 0.227). Le Conte’s Sparrows appeared to exhibit a behavioral response to flush-netting {P < 0.001) with estimated
capture probability of 0.462 and recapture probability of 0.056. Our findings suggest Le Conte’s Sparrows remain fairly
sedentary throughout the winter. Received 28 March 2008. Accepted 7 August 2009.

Site tenacity has been reported for many
overwintering grassland birds (Plentovich et al.
1998, Gordon 2000); familiarity with an area may
increase an individual’s ability to acquire resourc-
es and avoid predators. Optimal management of
habitats in wintering areas for high-priority
species that exhibit strong site tenacity may
increase overwinter survival, and enhance breed-
ing (Pulliam and Enders 1971, Fretwell 1972,
Wiens 1974, Raitt and Pimm 1976, Grzybowski
1982).

Le Conte’s Sparrows {Ammodramus lecontii)
have been listed as a “high priority species in
need of conservation attention” by Partners in
Flight (Carter et al. 2000, Shackelford et al. 2001).
They are fairly uncommon over much of their
breeding range with the highest densities in the
western portions (Lowther 1996). Le Conte’s
Sparrows are poorly covered by the North
American Breeding Bird Survey, in part due to
their secretive behavior (Igl and Johnson 1999).
Similarly, during the non-breeding season, the
species may be underestimated or undetected on
many Christmas Bird Counts. Concern for this
species is compounded by severe habitat loss in

' Louisiana Slate University, 227 Renewable Natural
Resources, Baton Rouge, LA 70803, USA.

^U.S. Geological Survey, National Wetlands Research
Center, 700 Caiundome Boulevard, Lafayette, LA 70506,
USA.

^National Park Service. 2282 South West Resource
Boulevard, Moab, UT 84532, USA.

''Current address: lAP World Services at USGS. National
Wetlands Research Center, 700 Cajundome Boulevard.
Lafayette, LA 70506, USA.

■’Corresponding author; e-mail;
heather_baldwin@usgs.gov

wintering areas. Wintering areas, primarily coast-
al prairie, still used by this species and other
grassland-dependent birds are degraded largely
due to fire suppression, agricultural practices and,
more recently, invasive plant species (Grace 1998,
Igl and Ballard 1999, Allain and Grace 2001).
Knowledge of Le Conte’s Sparrow ecology and
space use during winter can facilitate efforts to
provide optimal habitat during the non-breeding
season.

Little is known about the general ecology of Le
Conte’s SpaiTOw (Igl and Johnson 1999), espe-
cially during winter. They reluctantly flush from
observers and spend much of their time on or near
the ground in dense vegetation. Most literature
related to Le Conte’s Sparrow refers to their
ecology and behavior during the breeding season
(Madden et al. 1999, Johnson and Igl 2001,
Winter et al. 2005). Few studies have examined
Le Conte’s Sparrow winter ecology. Igl and
Ballard (1999) evaluated habitat associations of
migrating and overwintering grassland birds in
southern Texas. Their study showed significant
differences in densities of Le Conte’s Sparrows
among habitats during both winter and spring but
not in the fall. Baldwin et al. (2007) examined
habitat relations in coastal prairie and found Le
Conte’s Sparrows are most common in areas
burned ^2 years and characterized by medium
herbaceous density and sparse shrubs. Reynolds
and Krausman (1998) examined relative abun-
dance between unburned and recently burned
(<l-year) plots in a mesquite {Prosopis spp.)
grassland, and found Le Conte’s Sparrow aban-
doned recently burned areas. Grzybowski (1983)
measured frequency of occurrence of individuals

139

140

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. I, March 2010

to describe patterns of space use. His results
suggested that Le Conte’s Spairow are solitary
during winter and appear to inhabit small home-
ranges, as they were only spaced 1 1-42 m from
other individuals (Grzybowski 1983). His findings
provided important information concerning space
use by Le Conte’s Span'ow during winter, but
Grzybowski (1983) examined unmarked birds and
there is uncertainty regarding individual site
tenacity, and their sedentary tendencies during
the winter. Lorenz (2007) examined site tenacity
(within season) and site fidelity (between years)
for Le Conte’s Span'ow during winter, and found
evidence of site tenacity.

Knowledge of movement and spatial patterns of
overwintering Le Conte’s Sparrows is important
for conservation planning and management of
coastal prairies. For example, the optimal burning
period in coastal prairies occurs during winter
when Le Conte’s Sparrows may be sedentary and
site faithful. The objectives of our study were to:
(1) radio-mark Le Conte’s Sparrows to estimate
home range size among burn years, (2) conduct
monthly banding efforts to examine site tenacity,
and (3) collect measurements of individual birds
to compare body condition among burn years.

METHODS

SOidy Area. — ^This study was conducted during
winter 2002-2003 at the Hoskins Mound Prairie,
Brazoria National Wildlife Refuge (NWR), Tex-
as. This area is part of the Texas Mid-Coast NWR
Complex. Prior to its purcha.se in 1990, the land
was used for rice farming and cattle ranching,
which contributed to dissection of the refuge into
numerous well-defined units surrounded by fire
breaks, including levees, roads, and ditches. Units
included in this study were those that appeared to
have recovered from disturbances (J. B. Grace and
L. K. Allain, pers. comm.; USGS, NWRC),
probably due to adjacent seed banks. Dominant
plant species in the study units included little
bluestem {Schizachyriinn scoparium), bushy blue-
stem (Andropogon glonierutus), saltmeadow cord-
grass (Spartina patens), switchgrass (Panicinn
virgatiini), cutover muhly (Miililenhergia e.\-
pansa), marsh elder (Iva angustifoUa), longspike
tridens (Tridens strietiis), paniegrasses (Di-
chanthelium spp.), and Hat.sedges (Cyperiis spp.).
Saltbush (Baccharis halimifolia), a native and
invasive shrub common in coastal areas, occurred
in .scattered clumps throughout much of the
refuge, and dominated portions of many units

included in this study. Other woody species
included blackbemes (Ruhus spp.). Macartney
rose (Rosa hracteata), wax myrtle {Morelia
cerifera), and Chinese tallow tree (Triadica
sehifera).

Approximately 3,200 ha of Hoskins Mound
Prairie are managed with prescribed burning. The
burning program at this area was established in
1997 with emphasis on growing-season fires.
Mowing and haying also have been used for
management in some areas. The overall manage-
ment plan for the entire refuge and Hoskins
Mound includes an average burn rotation of 3-
5 years, depending on site conditions and succes-
sional status. Most of the refuge is in compliance
with this burn cycle, although weather conditions
may impede implementation of scheduled burns
in some years. Management is directed by general
objectives and guided by periodic monitoring of
habitat conditions.

Le Conte’s Sparrow numbers on Christmas
Bird Counts on Brazoria (Freeport) or San
Bernard NWRs have been among the highest in
the country (National Audubon Society 2002).
The area provided an excellent opportunity to
study Le Conte’s Span'ow’s home-range size and
site tenacity during the winter.

We sampled replicate burn units subjected to a
discrete set of burn intervals. Sample units were
.selected from those available using a completely
randomized design. Each unit included was
selected from all possible prairie units on Hoskins
Mound Prairie, and burned within 1, 2 or 3 years
of the initiation of data collection. We collected
data from 15 burn units with five replicate units
from each burn interval. Selected management
units ranged in size from 40 to 520 ha. Banding
sites were randomly selected for each unit in an
area 50-500 m from the edge of the unit. Birds
were radiomarked only on 1- and 2-year burn
units.

Telemetry. — Le Conte’s Sparrows were cap-
tured and radiomarked in five I -year and five 2-
year burn units during January and February 2003.
Searching for birds to radiomark began from a
random location at least 100 m from each unit
boundary. Searchers moved from the edge of the
unit until a bird was located and captured. Later in
the season, due to limited access to some
previously .selected areas and time constraints,
searching for birds to radiomark began from a
non-random location at least 100 m from the unit
boundary. Searches for a second bird on a unit

Baldwin ci a/. • HOME-RANGE SIZE AND EIDELITY OE LE CONTE’S SPARROW 141

began at least 250 m from the first bird to reduce
the chance of observer disturbance.

Each week, four birds, two from a I -year and
two from a 2-year burn unit were captured and
radiomarked. A crew of 3-6 people located birds
by walking in a line and tlushing birds using poles
and drag-lines. When a Le Conte’s Sparrow was
located, a 12-m mist-net was set up downwind
from the bird, and the crew formed a semicircle
behind the bird. The crew then rushed toward the
net in an attempt to direct the bird into the net.
This was repeated until a bird was netted.

Each bird was fitted with a 0.48-g radio
transmitter (Model LB-2; 10-day battery life with
an operating frequency near 165 MHz; Holohil
Systems Ltd., Carp, ON, Canada) by placing it
over the feathers of the synsacrum and attaching it
with an elastic leg-loop harness (Rappole and
Tipton 1991). Extra steps were taken to prepare
transmitters for application due to the wet, salty
conditions and dense herbaceous vegetation of
coastal prairie. A rain event occurred after the
second application of transmitters, and signals
became weak, inconsistent or nonexistent. We
retrieved one of the censored radios, and found
antennas were rusting. Transmitter antennas were
coated with epoxy following this incident in an
attempt to resolve this problem. A bird was found
entangled in the grass on one occasion. The
transmitter was immediately removed and the bird
was released alive. We removed 2.5 cm of the
transmitter antenna after this incident and there-
after before deployment. Change in transmission
signal as a result of this modification was
negligible as predicted by Beeman et al. (2007).

Locations were recorded daily and every other
day for a 1-hr period, until a mortality event
occurred, the signal disappeared, or the bird was
recaptured. Locations were established Irom an
estimated distance of 25-100 m from the bird.
Azimuth directions were recorded from two
Global Positioning System points, 20-40 m apart.
The radio-marked bird was located every 15 min
for a 1-hr period (i.e., 4 localions/hr) to obtain an
average location for that monitoring session.
Recaptures of birds were attempted 6-1 1 days
after capture to remove transmitters.

Banding. — Within-season site tenacity of win-
tering Le Conte’s Sparrows was examined
through monthly banding efforts. Birds were
mist-netted monthly (Dec 2002-Feb 2003) on
each burn unit (n = 15) on a 1-ha plot with the
assistance of 10 or more volunteers. Plots

consisted of a lOO-m net lane with a ().5-ha
Hushing zone on each side of the net. Plots were
selected randomly. A 50-m buffer zone was
maintained from roads and fire breaks. Vegeta-
tion was cut (~1 m wide) to create a path to
allow the nets to .stretch to the ground to deter
birds from escaping under the net. Banding
efforts began at sunrise and continued into early
afternoon. Plots were netted in a different order
each month.

Each captured bird was banded with an
individually numbered aluminum federal leg band
or their band number was recorded if recaptured.
Wing chord (mm), fat score (scale 0-5), and mass
(g) was recorded for each bird captured (Helms
and Drury 1960, Krementz and Pendleton 1990).
The same observer measured fat score on all birds
to avoid observer bias. Birds were released where
captured. Wind speed (km/hr), temperature (°C),
and time of day also were recorded.

Analyses. — ^Telemetry data were analyzed using
the Home Range Extension in ArcView 3.3
Geographic Information System (ESRI 2002) to
estimate home-range size and activity centers
using the kernel method (Worton 1989). Move-
ment patterns were analyzed with the Animal
Tracking Extension in ArcView. Home-range
sizes were estimated at 50 and 95% probabilities.
Differences among estimated home-range sizes
were compared between 1- and 2-year burn
treatments, and by mass of marked sparrows.
Results were compared using PROG MIXED
(SAS Institute 1999).

General information from banding data was
derived and reported using program CAPTURE
(White el al. 1978, Rexslad and Burnham 1991).
Model Mb (behavioral response after capture) was
selected as the most appropriate model for the
data. The three parameters produced under model
Mb were: A, estimated population size; P, the
probability of capture of an unmarked animal on
any trapping occasion; and c, the probability that
an animal is captured on any trapping occasion
subsequent to the occasion on which it was first
captured (Zippin 1956, 1958; Otis et al. 1978). An
index of body condition (IBC) was used to correct
for variation associated with body mass. IBCs are
residuals from a regression line of wing chord
(independent variable) and body mass (dependent
variable). IBC, fat score, and differences in
recapture weight were analyzed using repeated
measures analysis of variance (PROC MIXED;
SAS Institute 1999).

142

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. I, March 2010

RESULTS

Telemetry. — A total of 268 locations was
obtained for birds monitored in 1-year burns and
236 locations in 2-year burns. Home-range
estimates were produced for 18 of 26 birds
radiomarked due to a minimum of 6 days of
observations needed for analysis. Three of the 26
marked sparrows apparently detached from their
transmitters, two were killed by predators (1
suspected avian, 1 unknown), one died of unknown
causes, and nine were censored. Birds censored
were those whose telemetry signal was lost and their
fate was unknown. A radio was designated as
detached if there was no evidence of predation
( feathers, predator tracks or feces, etc. ) or damage to
the transmitter or antenna. Recaptured birds were
tracked for an average of 9.5 days, and lost an
average of 1.4 g (SE = 0.23) over the tracking
period (Table 1). Transmitters were removed from
1 1 sparrows prior to the end of battery life.

Mean home-range estimates with 50 and 95%
probability were 2.41 ha (median = 0.28 ha) and
10.31 ha (median = 1.27 ha), respectively (Ta-

ble 1). Most birds limited their movements to
locations within 200 m of their initial capture site,
which is more clearly shown at the scale of the
refuge (Eig. 1). The bird that moved the greatest
distance was at an outlier location for 1 day.
Removal of the day 8 location for bird # 289
reduced the 50% estimate of home range size
from 20.91 to 0.08 ha, and 95% home range
estimate from 86.3 to 0.51 ha. Two birds from
different capture periods moved long distances
and returned to their initial capture location the
following day. Long-distance movements on day
3 inflated home-range estimates for birds # 280
and # 203. Removal of the single outlier location
for bird #280 reduced the 50% estimate of home-
range size from 4.50 to 1.54 ha, and 95% home-
range estimate from 20.39 to 4.50 ha. The 50%
home-range estimate for bird # 203 was inflated
from 0.06 to 7.94 ha. and the 95% home-range
estimate from 0.45 ha to 32.01 ha. Bird # 294
moved 509 m from its initial location and had
somewhat nomadic behavior by remaining at the
initial area for at least 5 days, then moved

TABLE I. Home-range estimate.s (ha) at 50% and 95% probability for radio-marked Le Conte’s Sparrows and mass
differences (wtdiff) of recaptured birds. Fates are defined as; survived entire tracking period, censored, or loss of signal
before end of tracking period, mortality, and transmitter detached.

Burn rreatmenl Identification #

Exposure days

Location.s

50%

95%

wtdiff

Fate

1 196

4

2

Censored

197

1

2

Transmitter detached

202

7

27

0.03

0.14

Censored

203

9

34

7.94

32.01

-1

Survived

281

6

27

0.1 1

0.51

-1.3

Survived

283

6

24

0.65

2.37

-2.2

Survived

285

10

25

0.02

0.14

-1.3

Survived

287

13

26

0.05

0.53

Censored

291

10

20

0.21

0.93

Censored

292

6

9

1.53

7.21

Censored

318

10

39

0.31

1.51

-0.6

Survived

319

11

33

0.43

2.19

-2.8

Survived

2 192

0

1

Survived

193

10

39

0.4

2.39

-0.8

Survived

194

1

2

Transmitter detached

195

6

29

0.08

0.34

Mortality (unknown)

199

5

8

Censored

200

1

1

Censored

205

2

3

Survived (grass entanglement)

206

8

32

0.18

0.92

Censored

279

8

32

0.16

1.04

-1

Survived

280

12

34

4.5

20.39

Mortality (predation)

288

1

2

Mortality (predation)

289

8

23

20.91

86.3

-1.6

Survived

294

10

21

5.82

26.01

Transmitter detached

295

6

7

0.24

0.64

Censored

Baldwin el al. • HOME-RANGE SIZE AND FIDELITY OF LE CONTE’S SPARROW 143

FIG. 1. Brazoria National Wildlife Refuge with ^5% estimated home-range size of each radio-marked Le Conte’s
Sparrow indicated with gray circles. Black circles indicate monthly banding locations. Interior polygons represent bum unit
boundaries burned within 1, 2, and 3 years.

—250 m for 1-2 days, and then an additional
—250 m from the initial location area for the final
2 days of monitoring (Table 1).

Bum year had no significant effect on home-
range estimates at 50% {P — 0.25), 95% iP =
0.23), or mass loss of recaptured birds (P = 0.46).
Month since bum had no significant effect on
home-range estimates at 50% (P = 0.40), 95% iP
= 0.38), or mass loss of recaptured birds (P =
0.86). Initial mass of birds was not coirelated with
home-range estimates at 50% (P = 0.28) or 95%
{P = 0.30).

Banding. — A total of 103 Le Conte’s Sparrows
was captured on three trapping occasions (Dec,
n = 54; Jan, n = 34; Feb, n = 15). Eight
individuals were recaptured and no bird was
recaptured more than once. Goodness of fit tests
in program CAPTURE revealed Mb as the best fit
model (y^ = 1.55, df = 2, P = 0.46). Model Mb
tests for a behavioral response after initial capture.
Le Conte’s Spamow showed a significant behav-

ioral response to capture {P < 0.001). Overall
capture probability was estimated to be 0.462,
while recapture probability was 0.056. Model Mb
estimated that 122 (SE = 10.4) birds had all or
part of their home ranges within 15 separate 1-ha
plots. Burn treatment did not significantly influ-
ence overall fat score (P — 0.85), IBC (P = 0.31)
or differences in recapture weights (P = 0.83).

DISCUSSION

Le Conte’s Sparrows are primarily sedentary
during winter and inhabit small home-ranges with
most birds confining their movements to an area
<2 ha. The extent to which site tenacity can be
described was limited because of the small
number of recaptures. Banding results contrasted
with those from telemetry, suggesting that low
recapture probability is likely due to avoidance of
recapture and not movement from the area. This
suggests that surveying birds in an area also with

144

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 1, March 2010

banded birds might decrease their detection
estimates due to avoidance behavior. Le Conte’s
Sparrows were studied throughout much of the
winter, but each individual was tracked for a
relatively short period of time, limiting our ability
to draw conclusions about their behavior through-
out the entire winter season.

Previous findings suggest that habitat selection
of grassland birds is related to vegetation structure
(Pulliam and Mills 1977, Igl and Ballard 1999,
Baldwin et al. 2007) and food abundance (Tucker
and Robinson 2003). If birds inhabit an area with
sufficient cover and food, nomadic movements
may be unnecessary, and survival is likely
enhanced by familiarity with the area (Sinclair
1984, Gordon 2000). There also may be a
tendency for solitary behavior. Grzybowski
(1983) found Le Conte’s, Grasshopper (Ammo-
clramus savannarum), and Baird’s {A. hairdii)
sparrows tended to be solitary and avoided other
individuals when flushed from adjacent locations.
Le Conte’s Sparrows also were suspected to
inhabit restricted home ranges (Grzybowski
1983). Gordon (2000) reported Grasshopper,
Baird’s, Savannah (Passerculiis sandwichensis),
and Vesper (Pooecetes grcimineus) sparrows
remained within fixed home ranges in winter
with Grasshopper Sparrows being the most
.sedentary. Le Conte’s Sparrow did not occur in
Gordon’s (2000) study area.

Grzybowski (1983) examined space use of Le
Conte’s Sparrow and estimated that individuals
were spaced 1 1-45 m apart leading him to
speculate the species maintains small home
ranges. Cooper (1984) found that Le Conte’s
Sparrows maintained a territory size of about
1.0 ha. Results from both studies were based on
unmarked Le Conte’s Sparrows. Estimates of
home-range or territory size of Le Conte’s
Sparrows from both studies are similar to our
findings.

Our radio-marked Le Conte’s SpaiTows consis-
tently lo.st mass but we believe movement was not
significantly impacted. Efforts to recapture radio-
marked birds revealed them to be highly mobile.
Loss of mass may be attributed to the induced
stre.ss of carrying the transmitter or a.s.sociated with
capture. The transmitter may have increased the
resources needed for thermoregulation, although
no evidence was recorded to support this hypoth-
esis. Radio-marked birds carried transmitters that
were <4% of their body mass (Federal Bird
Banding Permit # 22207, 09/()7/()6-()7/31//09).

Burn year and/or bird morphology revealed no
relation to home-range size. Le Conte’s Sparrows,
in coastal prairies, are more abundant in areas that
have been burned within 1 to 2 years versus
3 years (Baldwin et al. 2007), and both 1- and 2-
year burn treatments likely provide suitable
habitat for Le Conte’s Spari’ows.

Mark-recapture methods have improved and are
used in numerous studies, but estimates of site
tenacity in our study were hampered by the small
number of recaptures. We were able to perform a
test among general models in program CAP-
TURE, under closed population assumptions, and
the results suggest Le Conte’s Sparrows may
exhibit a significant behavioral response to
capture (Mb). It is difficult to distinguish between
capture avoidance, poor survival and/or emigra-
tion; selection of model Mb may be confounded if
emigration does occur (i.e., in an open population)
but is not detected (Otis et al. 1978). Analysis
using open population models was not successful
due to insufficient data. Telemetry results (home-
range size <2 ha), selection of model Mb, and an
estimated density (± SE) of 4.91 ± 1.49 birds/ha
(Baldwin et. al. 2007) combined suggest that Le
Conte’s Span'ow home-ranges during the winter
may overlap extensively, or that individuals
remain fairly sedentary for an undetermined time
(>I0 days), and periodically move to other areas
with suitable habitat.

Le Conte’s Sparrow may exhibit site fidelity in
wintering areas. Preliminary banding of Le
Conte’s Sparrows was conducted in January
2002 on one burn unit. A mist net was randomly
placed in the same unit in January 2003, and one
of the 84 Le Conte’s Sparrows banded the
previous year was recaptured less than 90 m from
its original capture site.

ACKNOWLEDGMENTS

Funding wa.s provided by the U. S. Geologieal Survey
(USGS), in a cooperative effort with Regions 2 and 4 of the
U.S. Fish and Wildlife Service. We thank the field cre\v: M.
A. Powell, S. W. Stuart. J. A. Umniel, and M. L. Keprta.
We e.specially thank employees and volunteers associated
with the Texas Mid-Coast National Wildlife Refuge
Complex, Gulf Coast Bird Observatory, USGS National
Wetlands Research Center, and Texas Parks and Wildlife
Department who devoted their time to this project. We
thank J. B. Grace, L. K. Allain, W. G. Vermilion, T. C.
Michot. F. C. Rohwer, P. E. Lowther, L. D. Igl, and C. E.
Braun for editorial comments and advice. Any use of trade,
product, or firm names is for descriptive purposes only and
does not imply endorsement by the U.S. Government.

Bakhvin et al. • HOME-RANGE SIZE AND FIDELITY OF EE CONTE'S SPARROW 145

LITERATURE CITED

Allain, L. and J. B. Grace. 2001. Changes in density and
height of the shrub Baccharis halimifoUa following
burning in coastal tallgrass prairie. Proceedings of the
North American Prairie Conference 17:66-72.

Baldwin, H. Q., J. B. Grace, W. C. Barrow Jr., and F.
C. Rohwer. 2007. Habitat relationships of birds over
wintering in a managed coastal prairie. Wilson Journal
of Ornithology 119:189-197.

Beeman, j. W., N. Bower, S. Juhnke, L. Dingmon, M.
VAN den Tillaart, AND T. Thomas. 2007. Effects of
antenna length and material on output power and
detection of miniature radio transmitters. Hydrobiolo-
gia 582:221-229.

Carter, M. F., W. C. Hunter, D. N. Pashley, and K. V.
Rosenberg. 2000. Setting conservation priorities for
landbirds in the United States: the Partners In Flight
approach. Auk 117:541-548.

Cooper, S. 1984. Habitat and size of the Le Conte’s
Sparrow’s territory. Loon 56:162-165.

Environmental Systems Research Institute (ESRI).
2002. ARCVIEW 'Version 3.3. Environmental Systems
Research Institute Inc., Redlands, California, USA.

Fretwell, S. 1972. Populations in a seasonal environment.
Princeton University Press, Princeton, New Jersey,
USA.

Gordon, C. E. 2000. Movement patterns of wintering
grassland sparrows in Arizona. Auk 117:748-759.

Grace, J. B. 1998. Can prescribed fire save the endangered
coastal prairie ecosystem from Chinese tallow inva-
sion? Endangered Species Update 15:70-76.

Grzybowski, J. A. 1982. Population structure in grassland
bird communities during winter. Condor 84:137-152.

Grzybowski, J. A. 1983. Patterns of space use in grassland
bird communities during winter. Wilson Bulletin
95:591-602.

Helms, C. W. and W. H. Drury Jr. 1960. Winter and
migratory weight and fat field studies on some North
American buntings. Bird Banding 31:1-40

IGL, L. D. AND B. M. Ballard. 1999. Habitat associations
of migrating and overwintering grassland birds in
southern Texas. Condor 101: 771-782.

iGL. L. D. AND D. H. Johnson. 1999. Le Conte’s Sparrows
breeding in Conservation Reserve Program fields:
precipitation and patterns of population change.
Studies in Avian Biology 19:178-186.

Johnson, D. H. and L. D. Igl. 2001. Area requirements of
grassland birds: a regional perspective. Auk 1 18:24—34.

Krementz, D. G. and G. W. Pendleton. 1990. Fat
scoring: sources of variability. Condor 92:500—507.

Lorenz, S. 2007. Site fidelity of wintering Le Conte’s
Sparrows in northeast Texas. North American Bird
Bander 32: 153-157.

Lowther, P. E. 1996. Le Conte’s Sparrow (Ammodranms
leconleii). The birds of North America. Number 224.

Madden, E. M., A. J. Hansen, and R. K. Murphy. 1999.
Influence of prescribed fire history on habitat and
abundance of passerine birds in northern mixed— grass
prairie. Canadian Field-Naturalist 113:627-640.

National Audubon Society. 2002. The Chrisimas Bird
Count historical results [Online]. http://www.audubon.
org/bird/cbc (accessed 1 August 2008)

Otis, D. L., K. P. Burnham, G. C. White, and D. R.
Anderson. 1978. Statistical inference from capture data
on closed animal populations. Wildlife Monographs 62.

Plentovich, S. M., N. R. Holler, and G. E. Hill. 1998.
Site fidelity of wintering Henslow’s Sparrows. Journal
of Field Ornithology 69:486^90.

Pulliam, H. R. and F. Enders. 1971. The feeding ecology
of five sympatric finch species. Ecology 52:557-566.

Pulliam, H. R. and G. S. Mills. 1977. The use of space by
wintering sparrows. Ecology 58:1393-1399.

Rappole, j. H. and A. R. Tipton. 1991. New harness
design for attachment of radio transmitters to small
passerines. Journal of Field Ornithology 62:335-337.

Raitt, R. j. and S. L. Pimm. 1976. Dynamics of bird
communities in the Chihuahuan Desert, New Mexico.
Condor 78:427-442.

Rexstad, E. and K. Burnham. 1991. User’s guide for
interactive program CAPTURE. Colorado Cooperative
Fish and Wildlife Research Unit, Fort Collins, USA.

Reynolds, M. C. and P. R. Krausman. 1998. Effects of
winter burning on birds in mesquite grassland.
Wildlife Society Bulletin 26:867-876.

SAS Institute Inc. 1999. SAS/STAT. "Version 8. SAS
Institute Inc., Cary, North Carolina, USA.

Shackelford, C. E., N. R. Carrie, C. M. Riley, and D.
K. Carrie. 2001. Project prairie birds: a citizen
science project for wintering grassland birds. Second
Edition. PWD BK W7000-485 (1/01). Texas Parks and
Wildlife, Austin. USA.

Sinclair, A. R. E. 1984. The function of distance movements
in vertebrates. Pages 240-258 in The ecology ot animal
movement (I. R. Swingland and P. J. Greenwood.
Editors). Clarendon Press, Oxford, United Kingdom.

Tucker Jr., J. W. and W. D. Robinson. 2003. Influence of
season and frequency of fire on Henslow’s Sparrows
(Ammodrainus henslowii) wintering on Gull Coast
pitcher plant bogs. Auk 120:96-106.

White, G. C., K. P. Burnham, D. L. Otis, and D. R.
Anderson. 1978. User’s manual for Program CAP-
TURE. Utah State University Press. Logan. USA.

Wiens, J. A. 1974. Climatic instability and the "ecological
saturation’’ of grassland bird communities in North
American grasslands. Condor 76:385-400.

Winter. M., J. A. Shaffer. D. H. Johnson. T. M.
Donovan, W. D. Svedarsky, P. W. Jones, and B.
R. EULISS. 2005. Habitat and nesting of Le Conte’s
SpaiTows in the northern tallgrass prairie. Journal of
Field Ornithology 76:61-71.

Worton, B. j. 1989. Kernel methods for estimating the
utilization distribution in home-range studies. Ecology
70:164-168.

ZiPPiN, C. 1956. An evaluation of the removal method of
estimating animal populations. Biometrics 12:163-
189.

ZiPPiN, C. 1958. The removal method of population
estimation. Journal of Wildlife Management 22:82-90.

The Wilson Journal of Ornithology 122( 1):146-152, 2010

PATERNAL SONG COMPLEXITY PREDICTS OFFSPRING SEX RATIOS
CLOSE TO FLEDGING, BUT NOT HATCHING, IN SONG SPARROWS

DOMINIQUE A. POTVIN' --^ AND ELIZABETH A. MacDOUGALL-SHACKLETON’

ABSTRACT. — Sex allocation theory predicts that population sex ratios should be generally stable and close to unity, but
individuals may benefit by adjusting the sex ratio of their offspring. For example, females paired with attractive males may
benefit by overproducing sons relative to daughters, as sons inherit their fathers' attractive ornaments (“sexy son”
hypothesis). Similarly, if compatible gene effects on fitness are more pronounced in males than females, genetically
dissimilar mated pairs may enhance fitness by overproducing sons (“outbred son” hypothesis). We tested these hypotheses
in Song Sparrows (Melospiza melodia) by examining offspring sex ratios of 64 complete families shortly after hatching
(“early-stage”) and again shortly before fledging ("late-stage”) in relation to paternal song complexity and the genetic
similarity of social mates. Neither early nor late-stage offspring sex ratio was related to parental genetic similarity. Nests of
males with larger song repertoires contained more male-biased broods by the late-stage nestling period, but not in the early-
stage nestling period. These findings suggest that attractive males may be better able to successfully raise male-biased
broods, but not that females adaptively adjust primary sex ratios in response to their social mate’s attractiveness. Received
20 April 2009. Accepted 24 July 2009.

Sex allocation theory predicts that population
sex ratios tend to be stable and close to 1 : 1 (Fisher
1930), but variation among individuals in circum-
stances and quality may favor parents facultative-
ly adjusting the sex ratios of their offspring
(Trivers and Willard 1973). The “sexy son”
hypothesis (Weatherhead and Robertson 1979),
originally postulated to explain female tolerance
for polygyny, also predicts that attractive males
and their mates should produce predominantly
sons, as sons may inherit elaborate ornaments.
Consistent with this hypothesis, paternal attrac-
tiveness has been linked to offspring sex ratios in
some birds (Ellegren et al. 1996) although not in
others (Westerdahl et al. 1997; see analyses by
Ewen et al. 2004, and also Cassey et al. 2006).
These mixed results may indicate mechanistic
constraints preventing some species from manip-
ulating sex ratios, and/or a relatively weak
selective advantage to doing so (Komdeur and
Pen 2002, Fawcett et al. 2007). The question of
how paternal attractiveness should affect offspring
sex ratios in the case of sexually .selected traits
that are learned and/or highly environment-
dependent, such as bird song, remains almost
entirely unanswered (but see Leitner et al. 2006).

Good genes models of sexual selection have
recently expanded to consider nonadditive, or

' Department of Biology, University of Western Ontario,
London. ON N6A 5B7. Canada.

'Current address: Department of Zoology, University of
Melbourne. Melbourne, VIC 3010, Australia.

'Corresponding author; e-mail:
d.potvin@pgrad.unimelb.edu.au

compatible-gene effects on fitness (Neff and
Pitcher 2005) with particular emphasis on hetero-
zygote advantage. Compatible-gene effects are
often observed in the expression of secondary
sexual traits (Marshall et al. 2003, Reid et al.
2005), raising the possibility that sons may profit
more than daughters from parental genetic
compatibility. These effects are also commonly
implicated in immunocompetence (e.g., Reid et al.
2005); this trait is expressed by male and females,
but may affect male fitness disproportionately due
to the immunosuppressive effects of high levels of
circulating testosterone (Folstad and Karter 1992).
If compatible-gene effects on fitness are stronger
in males, optimal offspring sex ratios may depend
in part upon parental genetic similarity. We refer
to this possibility as the “outbred son” hypoth-
esis.

Our objective was to examine if offspring sex
ratios in Song Sparrows {Melospiza melodia) are
associated with paternal attractiveness based on a
learned trait (song repertoire size) and/or with
parental genetic similarity. The mechanisms by
which birds might manipulate offspring sex ratios
comprise two major classes: before hatching,- for
example through .selectively developing Z- or W-
bearing ova; and after hatching, for example
through preferentially feeding either males or
females (Pike and Petrie 2003). We investigated
sex ratios at two different points in the nesting
cycle, once shortly after hatching and again
shortly before fledging. Female Song Sparrows
prefer males with more complex song repertoires
(Searcy 1984, Reid et al. 2004), and song
repertoire size has been correlated with immuno-

146

Polvili and MacDoiigall-Shackleton • SONG COMPLEXITY AND SEX RATIOS

147

competence, teiritory defense, and other measures
of male fitness (Reid et al. 2005, Pfaff et al. 2007).
Thus, the "sexy son" hypothesis predicts that
males with larger repertoires (more attractive
males) should produce male-biased broods. The
"outbred son” hypothesis predicts that genetical-
ly dissimilar mated pairs should also produce
male-biased broods, because individual genetic
diversity is related both to immunocompetence
and to repertoire size in this species (Reid et al.
2005).

METHODS

Study Subjects and General Field Methods. —
We examined offspring sex ratios in 64 families
of Song Sparrows breeding near Newboro,
Ontario, Canada (44° 38' N, 76° 19' W). We
captured adults in seed-baited treadle traps in
April and May 2006, 2007, and 2008, outfitted
each with a unique combination of colored leg
bands, and collected a small (<25 pL) blood
sample for genetic analysis. We identified mated
pairs and located nests by behavioral observa-
tions. Each adult was included only once in the
analysis; for birds that bred in multiple years we
randomly selected which year’s nest to include in
the analysis. We included only first nesting
attempts to reduce potential seasonal effects on
sex ratio.

We collected a small (15 mL) blood sample
from each nestling via femoral venipuncture
2 days after eggs hatched for subsequent genetic
identification as male or female. We used felt-tip
markers to mark nestlings’ toes to track subse-
quent survival, weighed individuals to the nearest
0.1 g, and collected any unhatched eggs for
genetic identification of inviable embryos as male
or female. We returned to the nest 4 days later
(6 days after hatching), noted which individuals
were still present, and weighed each individual to
the nearest 0.5 g. Day 6 was used to avoid
artificially inducing early fledging, which nor-
mally occurs 8-10 days after hatching (pers.
obs.).

Song Analysis. — ^We recorded songs of adult
males early in the mating season onto Marantz
Professional PMD 671 solid state recorders using
Telinga Twin Science Pro parabolic microphones.
We considered a repertoire to have been sampled
in full after recording 300 consecutive or 450 non-
consecutive songs following the guidelines estab-
lished for this population by Pfaff et al. (2007).
We generated spectrograms in SYRINX (John

Burt, www.syrinxpc.com), and classilied song
types by visual inspection and sorting to identify
each male’s repertoire size. This was done blind
with respect to offspring sex ratios.

Genetic Analysis. — ^We genotyped all adults at
seven microsatellite loci: Escm 1 (Hanotte et al.
1994), Mme 2 and 7 (Jeffrey et al. 2001), Pdom 5
(Griffith et al. 1999), and Sosp 3, 13, and 14 (L. E.
Keller, pers. comm.). We tested for deviations
from Hardy- Weinberg expectations and from
linkage equilibrium using GENEPOP Version
3.3 (Raymond and Rousset 1995), and found no
evidence for either. We used microsatellite
profiles to calculate Wang’s (2002) coefficient
of genetic similarity for each mated pair using
MARK (Kermit Ritland, http://genetics.forestry.
ubc.ca/ritland/programs.html).

We identified 208 nestlings and unhatched
embryos from 55 nests (33 of which had survived
to 6 days after hatching) as males or females over
3 years of study. We used primers P2 and P8
(Griffiths et al. 1998) to amplify portions of the
CHD-W and CHD-Z gene, located on avian sex
chromosomes. Each gel included control amplifi-
cations from a known male and a known female.

Statistical Analysis. — ^We investigated the rela-
tionships between offspring sex ratios and pater-
nal repertoire size and/or parental genetic simi-
larity using generalized linear models (GEM;
PROG GENMOD) (SAS 2004) with a logit-link
function and binomial error distribution. Embryos
collected from unhatched eggs were included in
the early-stage (2 days after hatching) analyses,
while late-stage analyses included only nestlings
that survived until 6 days after hatching.

We constructed two statistical models for both
early and late-stage sex ratios. Initial models
included paternal repertoire size, parental genetic
similarity (using Wang’s [2002] coefficient), year,
and relative hatching date (number of days before
or after the mean hatching date for all first-brood
nests that year) as predictor variables. Einal
models excluded any clearly non-informative
predictor variables {y- < 1). Number of sons
within a nest for all models was the response
variable and number of offspring (sons plus
daughters) was the binomial denominator. All
predictor variables were normally distributed
(Kolmogorov-Smirnov test) and all statistical tests
were two-tailed.

We tested whether the use of color bands might
confound sex ratios in our study population,
because colored leg bands may affect attractive-

148

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 1, March 2010

ness and offspring sex ratios in some species
(Burley 1986; but see Rutstein et al. 2005). We
investigated the relationship between offspring
sex ratio and the number of bands of that color on
each adult male for each of the eight band colors
using Pearson’s correlations with the appropriate
Bonferroni corrections. We did not observe any
relationship between color bands and either
hatching or fledging sex ratios (all P > 0.05),
suggesting our use of colored leg bands was
unlikely to be an important confounding issue in
this study.

RESULTS

Fifty-four (53.8) percent (112 of 208) of
nestlings analyzed 2 days after hatching were
male. Sixty (55%) of the 109 nestlings that
survived until at least 6 days after hatching were
male. Our initial GLM identified a significant
effect of year on early-stage offspring sex ratios
(/■ = 6.01, P = 0.014) with sex ratios somewhat
more male-biased in 2007 than other years. None
of the other predictor variables included in this
model explained a significant amount of variation
in early-stage offspring sex ratios (repertoire size:
/• = 0.34, P = 0.56; parental genetic similarity:
/- — 0.25, P = 0.56; relative hatching date: =
0. 13, P = 0.72). Our initial GLM at 6 days after
hatching identified a near-significant association
between paternal repertoire size and offspring sex
ratios (x~ — 3.28, P = 0.070), and no significant
effects of other predictor variables on late-stage
sex ratios (parental genetic similarity: x' = 0.77,
P = 0.38; relative hatching date: x' ~ P =
0.38; year: x" = P = 0.27). Our final GLMs
included only repertoire size and year as predictor
variables, and indicated a significant positive
relationship between repertoire size and late-stage
iX^ = 4.66, P - 0.030) but not early-stage (x’ =
0.52, P = 0.47) offspring sex ratios (Fig. 1).

Differences in .sex ratios at day 6 could be
explained by one of two possibilities: either there
was increased mortality of female nestlings in
attractive males’ nests (creating a male bias), or
there was increased mortality of male nestlings in
less attractive males’ nests (creating a female
bias). Twelve of the thirty-three nests (36%) that
had at least one nestling survive to day 6
(surviving nests) had partial brood loss. We used
a /-test to investigate differences between the
social fathers’ repertoire sizes for nests in which
male and female nestlings were dying, and found
that daughters were consistently dying in nests of

males with larger repertoire sizes (/ = —3.78, df
= 64, P < 0.001). The total number of offspring
surviving to day 6 was not significantly coirelated
with social father’s repertoire size {r = 0.003, df
= 34, P = 0.763). However, when analyzing
predictors of number of nestlings lost in surviving
nests, social father’s repertoire size had a
significant effect (x" = 20, P = 0.002).

We also compared the average growth rates of
brothers to the average growth rates of sisters in
the same nest to help explain differences in
mortality. Brothers consistently grew faster than
sisters (/ = —2.88, df — 2Q, P — 0.009). The data
indicate, although there was not sufficient statis-
tical power to compare nestling mortality rates
between years, that female mortality in surviving
nests was higher than male mortality in surviving
nests for 2 of the 3 years (Tables 1-2).

DISCUSSION

Offspring sex ratios near the end of the nestling
period were significantly related to paternal song
repertoire size (Fig. IB), consistent with the
predictions of the “.sexy son’’ hypothesis. How-
ever, this relationship did not appear to be
mediated through manipulation of primary sex
ratios, as repertoire size did not predict sex ratios
earlier in the nestling period (Fig. lA). It is
unlikely the lack of a relationship between song
repertoire size and early-stage sex ratios was due
to low statistical power, as we were able to detect
an effect of repertoire on late-stage sex ratios
despite a lower sample size of surviving nests (38
vs. 66). Adjusting primary sex ratios is presum-
ably less costly in terms of fitness than altering
sex ratios after hatching, but the apparent absence
of primary sex-ratio manipulation may reflect an
inability to do so.

The positive relationship between paternal song
repertoire size and brood sex ratio late in the
nestling period may or may not represent adaptive
post-hatching male/female allocation. It remains
to be learned whether adult Song Sparrows
preferentially feed male or female offspring,
much less how such biases might relate to paternal
attractiveness. A growing body of evidence points
to the importance of early developmental condi-
tion in shaping song-learning ability (e.g., Bu-
chanan et al. 2003), suggesting that complex
singers and their mates may benefit from
oveiproducing, or preferentially feeding sons, if
sons stand to gain more than daughters from direct
benefits via increased paternal care. Secondary

Potvin and MacDougall-Sluickleton • SONG COMPLEXITY AND SEX RATIOS

149

O

Male repertoire size

FIG. 1. Relationship between paternal repertoire size and olTspring sex ratio, measured as proportion ot sons, (A)

2 days and (B) 6 days after hatch over 3 years (filled circles

TABLE 1. Population-wide nestling survival of Song
Sparrows near Newboro, Ontario, Canada.

2006 2007 2008

Males

Females

Males

Female.s

Males Females

Day 2

59

35

26

26

27

35

Day 6

35

22

15

18

10

8

Survival rate

59%

63%

58%

69%

37%

23%

2006, open circles = 2007, and inverted triangles = 2008).

sexual traits in many species honestly advertise
paternal ability or effort (Buchanan and Catchpole
2000, Voltura et al. 2002, Siefferman and Hill
2003). Thus, more attractive males would provision
at a higher rate, resulting in a more noticeable
difference between male and female sibling growth
rates in nests of more attractive males. We observed
a difference in growth rates between brother and
sister nestlings, consistent with biased feeding.

150

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 1, March 2010

TABLE 2. Population-wide nestling survival of Song
Sparrows near Newboro, Ontario, Canada, including only
nests in which at least one nestling survived to day 6.

2006 2007 2008

Males

Females

Males

Females

Males

Females

Day 2

36

26

15

18

15

11

Day 6

35

22

15

18

10

8

Survival rate

97%

85%

100%

100%

80%

72%

Comparing mortality rates between female and
male nestlings indicates the observed relationship
between paternal repertoire size and sex ratios
near fledging may be due to disproportionate
mortality of daughters in the nests of attractive
males. The reason for these female-biased deaths
is not completely clear. One possibility is that
male nestlings that are fed more often, perhaps by
an attractive male, may be better able to
OLitcompete their sisters. There may be a positive
feedback effect whereby an increase in the size of
a male nestling may enhance his begging ability,
resulting in further biased feeding from parents.
Consistent with this possibility, female Song
Sparrow nestlings in nests parasitized by Brown-
headed Cowbirds (Molothrus ater) are more
susceptible to being outcompeted by the faster-
growing cowbird nestling (Zanette et al. 2005).
This could be interpreted as an adaptive sex ratio
bias in that sons may be preferentially fed;
however, the deaths of female nestlings could be
an “unintended” result of this feedback effect.

We cannot rule out effects of disease and partial
predation, but starvation resulting from differen-
tial provisioning by parents is likely the main
cause of partial brood reduction, as theorized by
Ricklefs (1969), and supported by experiments by
Arce.se and Smith (1988) and observations by
Hochachka and Smith (1991). Nestling growth
rate in this population is correlated with parental
visits, providing further support to the theory that
nestling success is heavily influenced by parental
care (Potvin and MacDougall-Shackleton 2009).

We observed no relationship between parental
genetic similarity and offspring sex ratio, contrary
to the predictions of the “outbred son” hypoth-
esis. This may reOect low statistical power, an
inability to detect genetic compatibility, and/or a
relatively weak selective advantage to adjusting
sex ratios in response to this variable. Some traits
exhibiting heterozygote advantage are expressed
mainly by males, such as song-learning ability

(Marshall et al. 2003, Reid et al. 2005), but
genetic diversity can also enhance female fitness,
for example through fecundity (Ortego et al.
2007) and hatching success (Mair et al. 2006).
There may be little, if any, selective advantage to
manipulating primary or secondary brood sex
ratios in response to expected offspring heterozy-
gosity.

Adaptive adjustments to primary sex ratios
appear taxonomically widespread among birds,
but our study adds to a growing body of evidence
that the pattern is by no means universal
(Komdeur and Pen 2002, Ewen et al. 2004,
Cassey et al. 2006). Moreover, apparently adap-
tive patterns of variation in secondary sex ratios
may represent a byproduct of environmental
conditions and the quality of care provided by
attractive versus less attractive males, rather than
a facultative response by females to the perceived
attractiveness of their mates.

ACKNOWLEDGMENTS

We thank Scott MacDougall-Shackleton for helpful
comments; Steve Russell, Megan Barclay, Casey Hoggard,
Janet Lapierre, Laura Dindia, Yanina Sarquis-Adamson,
Vishalla Singh, and Erica Tennenhouse for field and
laboratory assistance; Jeremy Pfaff and Kathryn Stewart
for recordings; Lucas Keller for primer sequences; the
Queen’s University Biological Station for logistic support;
and NSERC Canada and the Canada Foundation for
Innovation for funding.

LITERATURE CITED

Arcese, P. and J. N. M. Smith. 1988. Effects of population
density and supplemental food on reproduction in
Song Sparrows. Journal of Animal Ecology 57;119-
136.

Buchanan, K. L. and C. L. Catchpole. 2000. Song as an
indicator of male parental effort in the Sedge Warbler.
Proceedings of the Royal Society of London, Series B
267;32l-326.

Buchanan, K. L., K. A. Spencer, R. A. Goldsmith, and
C. K. Catchpole. 2003. Song as an honest signal of
past developmental stress in the European Starling.
Proceedings of the Royal Society of London, Series B
270:1 149-1 156.

Burley, N. 1986. Sex-ratio manipulation in color-banded
populations of Zebra Finches. Evolution 40:1191-
1206.

Cassey, P., J. G. Ewen, and A. P. M0ller. 2006. Revised
evidence for facultative sex ratio adjustment in birds: a
correction. Proceedings of the Royal Society of
London, Series B 273:3129-3130
Ellegren, H., L. Gustafs.son, and B. C. Sheldon. 1996.
Sex ratio adjustment in relation to paternal attractive-
ness in a wild bird population. Proceedings of the

Polvin aiul A/cicDoiigcill-Slicicklelon •

SONG COMPLEXITY AND SEX RATIOS

151

National Academy of Sciences of the USA 93; 1 1723-
1 1728.

Ewen. J. G.. P. Cassey, and A. P. Mdller. 2004.
Facultative primary sex ratio variation: a lack of
evidence in birds? Proceedings of the Royal Society of
London, Series B 27 1 ; 1277-1282

F.awcett, T. W., B. Kuuper, I. Pen, and F. J. Weissing.
2007. Should attractive males have more sons?
Behavioral Ecology 18:71-80.

Fisher, R. A. 1930. The genetical theory ot natural
selection. Oxford University Press, Oxtoid, United
Kingdom.

Folstad, I. and a. j. Karter. 1992. Parasites, bright
males, and the immunocompetence handicap. Amer-
ican Naturalist 139:603-622.

Griffith, S. C., I. R. K. Stewart, D. A. Dawson, I. P. F.
Owens, and T. Burke. 1999. Contrasting levels of
extra-pair paternity in mainland and island populations
of the House Sparrow (Passer domesticus): is there an
“island effect”? Biological Journal of the Linnean
Society 68:303-316.

Griffiths, R., M. C. Double, K. Orr, and R. J. G.
Dawson. 1998. A DNA test to sex most birds.
Molecular Ecology 7:1071-1075.

H.^NOTTE, O., C. Zanon, a. Pugh, C. Grieg, A. Dixon,
AND T. Burke. 1994. Isolation and characterization of
microsatellite loci in a passerine bird; the Reed
Bunting Emheriza schoenidus. Molecular Ecology
3:529-530.

Hochachka, W. and j. N. M. Smith. 1991. Determinants
and consequences of nestling condition in Song
Sparrows. Journal of Animal Ecology 60:995-1008.

Jeffery, K. J., L. F. Keller, P. Arcese, and M. W.
Bruford. 2001. The development of microsatellite
loci in the Song Sparrow, Melospiza melodia (Aves)
and genotyping errors associated with good quality
DNA. Molecular Ecology Notes 1:11-13.

KOMDEUR, J. AND 1. PEN. 2002. Adaptive sex allocation in
birds: the complexities ot linking theory and practice.
Philosophical Transactions of the Royal Society of
London, Series B 357:373-380.

Leitner, S., R. C. Mar.shall, B. Leisler, and C. K.
Catchpole. 2006. Male song quality, egg size and
offspring sex in captive Canaries (Serinus camiria).
Ethology 112:554-563.

Marr, A. B., P. Arcese, W. M. Hochachka, J. M. Reid,
AND L. F. Keller. 2006. Interactive effects of
environmental stress and inbreeding on reproductive
traits in a wild bird population. Journal ot Animal
Ecology 75:1406-1415.

Marshall, R. C., K. L. Buchanan, and C. K. Catchpole.
2003. Sexual selection and individual genetic diversity
in a songbird. Proceedings of the Royal Society ot
London, Series B 270:S248-S250.

Neff, B. D. and T. E. Pitcher. 2005. Genetic quality and
sexual selection; an integrated framework for good
genes and compatible genes. Molecular Ecology
14:19-38.

Ortego, j., G. Calabuig, P. J. Cordero, and J. M.
Aparicio. 2007. Egg production and individual

genetic diversity in Lesser Kestrels. Molecular Ecol-
ogy 16:2383-2392.

Pfaff, j., L. Zanette, S. A. MacDougall-Shackleton,
AND E. A. MacDougall-Shackleton. 2007. Song
repertoire size varies with HVC volume and is
indicative of male quality in Song Sparrows (Melos-
piza melodia). Proceedings of the Royal Society of
London, Series B 274:2035-2040.

Pike, T. W. and M. Petrie. 2003. Potential mechanisms of
avian sex manipulation. Biological Reviews 78:553-
574.

PoTViN, D. A. AND E. A. MacDougall-Shackleton.
2009. Parental investment amplifies effects of
genetic complementarity on growth rates in Song
Sparrows (Melospiza melodia). Animal Behaviour
78:943-948.

Raymond, M. and F. Rousset. 1995. GENEPOP
(Version 1.2): a population genetics software for
exact test and ecumenicism. Journal of Heredity
86:248-249.

Reid, J. M., P. Arcese, A. Cassidy, A. Marr, J. Smith,
AND L. Keller. 2005. Hamilton and Zuk meet
heterozygosity? Song repertoire size indicates inbreed-
ing and immunity in Song Sparrows (Melospiza
melodia). Proceedings of the Royal Society of London,
Series B 272:481-487.

REtD, J. M., P. Arcese, A. CASStDY, S. M. Hiebert, J.

N. M. Smith, P. K. Stoddard, A. B. Marr. and
L. F. Keller. 2004. Song repertoire size predicts
initial mating success in male Song Sparrows,
Melospiza melodia. Animal Behavior 68. 1055—
1063.

Ricklefs, R. E. 1969. An analysis of nestling mortality in
birds. Smithsonian Institution Press, Washington.
D.C.. USA.

Rutstein. A. N., H. E. Gorman, K. E. Arnold. L.
Gilbert, K. J. Orr, A. Adam, R. Nager. and J.
Graves. 2005. Sex allocation in response to paternal
attractiveness in the Zebra Finch. Behavioral Ecology
16:763-769.

SAS. 2004. Version 9.1 u.ser's guide. SAS Institute Inc..
Cary, North Carolina. USA.

Searcy, W. A. 1984. Song repertoire size and female
preferences in Song Sparrows. Behavioral Ecology
and Sociobiology 14:281-228

Siefferman, L. and G. E. Hill. 2003. Structural and
melanin coloration indicate parental effort and repro-
ductive success in male Eastern Bluebirds. Behavioral
Ecology 14:855-861

Trivers, R. L. and D. E. Willard. 1973. Natural selection
of parental ability to vary the sex ratio of offspring.
Science 1 79:90-92.

VOLTURA. K. M., P. L. SCHWAGMEYER, AND D. W. MOCK.
2002. Parental feeding rates in the Hou.se Sparrow,
Passer domesticus'. are larger-badged males better
fathers? Ethology 108:1011-1022.

Wang, J. 2002. An estimator for pairwise relatedness using
molecular markers. Genetics 160:1203-1215.

Weatherhead, P. and R. Robertson. 1979. Offspring
quality and the polygyny threshold: “the .sexy son
hypothesis.” American Naturalist 113:201-208.

152

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. I, March 2010

Westerdahl, H., S. Bensch, B. Hansson. D. Hassel-
QUIST, AND T. VON SCHANTZ. 1997. Sex ratio variation
among broods of Great Reed Warblers Acrocephaliis
aritndinaceus. Molecular Ecology 6:543-548.

Zanette, L., E.
Clinchy, and J
Cowbirds skew
86:815-820.

. MacDougall-Shackleton, M.
N. M. Smith. 2005. Brown-headed
host offspring sex ratios. Ecology

The ITilson Joiinuil of Oniilliology 1 22( 1 ); 1 53-1 59, 2010

INFLUENCE OF AGE AND SEASON ON MORPHOMETRIC
MEASUREMENTS OF THE BISCUTATE SWIFT
{STREPTOPROCNE BISCUTATA)

MAURO PICHORIM'

ABSTRACT —Little emphasis has been placed on the influence of season and age on morphological measurements of
swifts which makes comparative mensural studies difficult in the Apodidae. This study provides information about the
imporlance of moiphological variation in measurements to the biology of the Biscutate Swift (Streptoprocne h,sculata)A
studied individuals captured at four colonies in southern Brazil. Mass varied with age, time of day, and season. Birds
captured at dawn weighed 1 15.5 ± 9.9 g {n = 1,320). Mass was greater in fall and spring and lower m summer^and winter
Subadults weighed less than adults during summer, fall, and winter. The wing measured 206.3 ±5.2 mrn in „), an
the tail measured 69.9 ± 4.6 mm (n = 802); both showed seasonal variation in length. The tarsus measured 25.9 _ 0.6 mni
(„ = 695) and the exposed culmen measured 9.94 ± 0.36 mm (n = 658). The Biscutate Swift has significant seasonal
variation in several morphological measurements. This variation should be considered m studies comparing populations of
this species as well as those of other apodids. Received 6 April 2009. Accepted 4 September 2009.

The morphology of the Apodidae is conserva-
tive and they are remarkably uniform in many
respects, such that differences among them are
subtle and may even be undetectable in prepared
skins. Some species of the genera Cypseloides,
Aeroclramus, Chaetiira, and Apiis are not easily
distinguished even when observed in the hand
(Wetmore 1957; Sims 1961; Medway 1966;
Chantler and Driessens 1995; Chantler 1995,
1999) Swifts do not vary greatly in size
interspecifically, and only minor differences are
apparent in the shape of the beak, tail, wing, and
tarsus. Morphological measurements, in some
cases, contribute to diagnoses of species and
subspecies (e.g.. Sick 1991, Mann and Stiles
1992, Browning 1993, Parkes 1993, Chantler and
Driessens 1995). Clinal variations, at times, lead
to high numbers of subspecies, although not all
variations justify taxonomic separation (Chantler
and Driessens 1995).

The body mass of many swifts also depends
upon age and season, and variations are much
greater than those in passerine birds of the same
size (Lack and Lack 1951, Naik and Naik 1966).
Subadults in some species weigh less than adults
during part of the first year of life, and this feature
is used to estimate the age of individuals (Coffey
1958, Collins 1968, Telia et al. 1995). However,
several studies that made morphological compar-
isons among species did not consider possible
variation in measurements caused by environmen-

' Departamento de Botanica, Ecologia e Zoologia.
Univer.sidade Federal do Rio Grande do Norte, Campus
Universitario, Lagoa Nova, Natal, RN, Brazil, 59072-970,

e-mail: mauropichorim@yahoo.com.br

tal and age-related factors. Sick (1991) based the
diagnosis of Streptoprocne hisciitata seridoensis
on differences in body mass and wing and tail
length, but did not consider seasonal variation.
Some descriptions of new Cypseloides species
also ignored variation in measurements through-
out the year (Eisenmann and Lehmann 1962,
Collins 1972, NavaiTO et al. 1992). Taxonomists
have used the data available and, considering the
scarcity of museum material and the difficulty of
studying swifts, those surveys have helped us to
understand species relationships. However, little
emphasis has been placed on the seasonal and age
dependency of some measurements. This tenden-
cy can make comparative population studies
difficult. 1 investigated age-related and seasonal
variation in moiphological measurements at four
colonies of the Biscutate Swift (Streptoprocne
hisciitata) in southern Brazil, and 1 discuss the
results with regard to the use of those parameters
in taxonomic studies.

METHODS

Four Biscutate Swift colonies were studied in
the eastern part of the State of Parana, southern
Brazil (Anhangava colony, municipality of Quatro
Barras - 25° 22' S, 48° 58' W, 1,250 m asl; Seira
do Capivari colony, municipality of Campina
Grande do Sul - 25° 1 1' S, 48° 51' W, 1,050 m
asl; and Arenitos and Furna 1 colonies, munici-
pality of Ponta Grossa - 25° 15' S, 50° 00' W,
-1,000 m asl; Fig. 1). Only measurements from
captured individuals were used. Mist nets and
other nets were used to capture swifts near the
most frequently used exit of each cave or roost

153

154

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. I. March 2010

FIG. 1. Biscutate Swift colonies studied in southern
Brazil. (A) Anhangava, municipality of Quatro Barras. (B)
Capivari, municipality of Campina Grande do Sul, (C)
Arenitos, and (D) Furna 1, both in the Vila Velha State
Park, municipality of Ponta Grossa.

site (Pichorim and Monteiro-Filho 2010). Metallic
numbered bands, supplied by CEMAVE/ICMBio,
(4.5 mm inner diam) were used to mark captured
swifts. The following measurements of each
captured individual were taken: body mass, wing
length (tlattened and straightened), tail length
(measuring with a metal ruler inserted between
the central pair of rectrices), tarsus length,
expo.sed culmen length (measured from the beak
tip in a straight line to the edge of the feathers in
the central part of the forehead), and length of
beak from nostril (hereafter nostril beak tip;
measured from the anterior end of the maxilla in
a straight line to the anterior edge of the nostril).
Pesola spring scales (300 g capacity; estimated to
the nearest Ig), calipers (15 cm), and rulers
(30 cm) were used to measure swifts. All birds
were relea.sed near the place of capture.

Analysis of variance (ANOVA) and Mests
(alpha = 0.05) were u.sed to test the hypothesis

ot the existence of only one group of means for
measurements taken during different seasons, age
classes, and colonies following Zar (1984).
Normality of distribution was tested for each
comparison, and variances were considered anal-
ogous if one did not exceed double of the other.
Measurement sets with samples of similar sizes
were compared by ANOVA and tested for normal
distribution and homogeneity of the variances
using Bartlett, Brown-Eorsythe, and Levene tests
(JMP 4.0.4, SAS 2001). Data are presented as
means ± SD. Individuals younger than 1 year of
age were considered subadults. Age could be
ascertained for marked nestlings captured after
leaving the nest, and for individuals captured
without molt of remiges and rectrices between
January and March, the months in which all adults
molt (pers. obs.).

RESULTS

Adult swifts weighed 115.7 ± 9.9 g (n =
1,340); considering only birds captured at dawn,
the mass was 115.5 ± 9.9 g (u = 1,320). The
heaviest individual (153 g) was captured in
Arenitos on the afternoon of 2 September 2001.
This individual had been captured the previous
year at dawn on 14 October 2000 and weighed
125 g. The lightest individuals (87 g, n = 2) were
captured in Arenitos on 2 December 2000. The
plumage of these birds did not allow estimation of
their age. However, subadults weighed less than
adults captured in the same period. Adults in
summer (Jan-Mar) weighed 110.9 ± 6.8 g (n =
292), while subadults weighed 102.5 ± 9.3 g (// =
8) (t — 3.43, df = 298, P < 0.001). Adults in fall
(Apr-Jun) weighed 1 18.9 ± 6.7 g (/? = 128), and
subadults weighed 1 1 1.0 ± 10.4 g (n = 5) (A =
2.54, df = 131, P = 0.012). Adults in winter (Jul-
Sep) weighed 116.1 ± 10.5 g (// = 456), and
subadults weighed 97.7 ± 1 1 .5 g (/; = 6) (r =
4.28, df = 460, P < 0.001). Adults in spring
(Oct-Dec) weighed 116.9 ± 10.4 g (/; = 444),
and subadults weighed 108.8 ± 4.2 g (n = 6) (/.=
1.88, df = 448, P — 0.061). The lightest
confirmed subadult (86.0 g) was captured at dawn
on 9 July 2002 in Arenitos.

No differences in mass were apparent between
males and females, as measurement of adults by
season had a normal distribution. The mass of
birds captured at dawn varied during the year;
birds weighed more in fall and spring and less in
summer and winter. Analyzing random samples of
the same size from each season confirmed these

Piclwrim . MORPHOMETRICS OF THE BISCUTATE SWIFT

155

differences (ANOVA F,,39(, = 19.8, P < 0.001).
Adults were heavier in September and lighter in
February (124.0 ± 8.7 g, n = 105 and 108.0 ±
5.2 g, n = 35; Table 1). Seasonal variation in
mass was noted in some individuals captured
repeatedly. An individual weighing 139 g was
captured in the Anhangava colony on 1 1 October
1994, and the same individual weighed 104 g on
24 February 1995. This variation coiTesponds to a
loss in mass of 25% in a 4-month period. Two
other individuals, both captured in Arenitos, had
mass changes of 30 g in a few months. The first
individual weighed 138 g on 13 October 2000 and
108 g on 12 December (a 22% decrease). The
second individual weighed 101 g on 12 December
2000 and 131 g on 3 September 2001 (a 30%
increase). These values were not related to time of
capture, as all individuals were weighed at dawn
as they were leaving the night roost.

Variations in mass were also observed in
individuals captured at day’s end. The maximum
variation was 15 g, for an individual from the
Capivari colony, captured on 24 November 2000
weighing 131 g and recaptured the following day

weighing 1 16 g (a 12% decrease). Six birds
captured at day’s end had mass decrease of 9-
13 g (a 7-10% decrease). Sixteen of 25 birds
captured at day’s end had mass decreases, six had
mass increases (not more than 5 g), and three had
no variation. An individual captured in Capivari at
0615 hrs on 27 October 2001 weighed 1 1 1 g, and
109 g at 1000 hrs the same day, a 2 g decrease
during a period of 3 hrs and 45 min. A nocturnal
decrease of 17 g (13%) was observed for an
individual captured on 24 November 2000 at
Capivari colony; it weighed 131 g in late
afternoon and 114 g the following morning.

The mean annual mass of adults captured at
dawn varied among the colonies (Capivari =
120.2 ± 9.0 g, n = 169; Arenitos = 115.9 ±
10.1 g, n = 610; Anhangava = 113.7 ± 8.7 g, n
= 360; and Furna 1 = 1 13.2 ± 7.8 g, /? = 181).
However, these data are not totally comparable
because of seasonal variations of individuals
captured in each colony. Considering only March,
April, August, October, and November, when I
obtained random samples of the same size from
each colony, the Furna 1 average was smaller than

TABLE 1. Measurements of body mass,
at four colonies in southern Brazil (Mean ±

wing

SD,

length, and tail length in adult Streptoprocne hisciitata captured at dawn
n; minimum - median - maximum).

.Month

Mass (g)

Flattened wing (mm)

Tail (mm)

Jan

108.1 ± 5.4, 134

203.6 ± 5.2, 134

66.6 ± 3.8, 99

93.0-108.0-120.0

186.0-204.0-215.0

51.0-67.0-78.0

Feb

108.0 ± 5.2, 35

203.1 ± 6.9, 34

72.7 ± 2.5, 35-''

98.0-109.0-117.0

189.0-204.0-218.0

66.0-73.0-77.0

Mar

1 15.0 ± 6.6, 119

207.3 ± 7.3, 20

75.0 ± 2.5. 16

98.0-1 15.0-132.0

184.0-209.0-217.0

70,0-75.0-79.0

Apr

1 18.5 ± 5.4, 78

200.7 ± 5.6, 59”

75.7 ± 2.82, 52

106.0-1 19.0-134.0

185.0-201.0-214.0

68.0-76.0-83.0

May

122.8 ± 2.6, 5

207.8 ± 3.0, 5'-'

74.8 ± 1 .9. 5

120.0-122.0-126.0

205.0-207.0-21 1.0

72.0-75.0-77.0

Jun

122.3 ± 8.3, 96

207.8 ±4.1. 96

76.0 ± 3.5, 96

101.0-122.5-139.0

199.0-208,0-219.0

62.0-73.5-80.0

Jul

109.9 ±8.1, 233

208.7 ± 3.9, 1 15

72.0 ± 3.0, 115

90.0-1 10.0-135.0

198.0-209.0-218.0

64.0-72.0-80.0

Aug

1 19,4 ± 8.1. 62

208.2 ± 4.4, 62

70.3 ± 3.9. 62

99.0-120.0-134.0

198.0-209.0-217.0

60.0-71.0-77.0

Sep

124.0 ± 8.7, 105

207.2 ± 4.0. 71

69.6 ± 3.6. 72

107.0-125.0-141.0

197.0-208.0-218.0

60.0-70.0-78.0

Oct

120.7 ± 10.2, 217

206.7 ± 4.9. 132

67.6 ± 4.0, 135

96.0-121.0-144.5

194.0-207.0-219.0

54.0-67.5-78.0

Nov

120.2 ± 7.4. 75

206.9 ± 4.8, 50

68.3 ± 2.6. 52

100.0-122.0-134.0

192.0-207.0-214.0

62.0-68.0-73.0

Dec

109.6 ± 8.0, 161

204.2 ± 5.2, 98

67.7 ± 4.7. 98

87.0-1 10.0-130.0

192.0-205.0-214.0

57.0-68.0-76.0

“ Individual.s molting rectrices (27 in the 1st and 2nd pair. 6 in the l.st pair, and 2 in the 2nd pair).

” Individuals molting primaries (33 in the 9th and lOlh pair, 14 in the 9th pair, and 12 in the lOth pair).
Individuals molting the lOlh pair ol primaries.

156

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. I, March 2010

those of the other colonies (ANOVA ^3,3 14 = 5.6,
P < 0.001; meancap.vari = 120.8 ± 8.3 g, SE =
0.92, /; = 80; meanArenitos = 1 19.9 ± 7.7 g, SE =
0.86, 11 = 80; meanAnhangava = 1 19.5 ± 8.6 g, SE
= 0.96, n = 80; meanpuma = 1 15.9 ± 7.7 g, SE =
0.87, n = 78; variances were homogeneous
according to Bartlett, Brown-Eorsythe, and Le-
vene tests).

Of individuals captured that were not molting
primary remiges 9 and 10, wing length was 206.3
± 5.2 mm (/? = 812). The longest wings
measured 219 mm (bird captured in Anhangava
colony on 1 1 October 1994 and another captured
in Arenitos colony on 17 June 2001). The shortest
wing measured 184 mm for a bird captured in
Capivari on 2 March 2002 (this wing was severely
worn). Generally, individuals without molt of the
outer primary feathers, captured from January to
March, showed wear on the tip of the primaries.
As a result, the mean was smaller than during the
rest of the year (203.9 ± 5.9 mm, « = 188). Wing
length was 207.1 ± 4.7 mm (n = 624) between
June and December. The largest monthly mean
was in July (208.7 ± 3.9 mm, n = 115), and the
smallest was in Eebruary (203.1 ± 6.9 mm;
Table 1). Subadults did not have the same
differences in means between the January-March
period and the rest of the year (meanjan-Mar —
203.2 ± 4.4 mm, n = 9; meanApr-oec “ 201.6 ±
3.7 mm, n = 18) (/ = 0.99, df = 25, P = 0.33).
Wings of subadults measured 202.1 ± 4.0 mm
(range 196.0-209.0 mm, n = 27). This value was
not different from that for adult wings at the
beginning of the year (Jan-Mar) (/ = 1.48, df =
213, P = 0.14), but differed from the June-
December period (t = 5.33, df = 649, P < 0.00 1 ).
Mean wing length of adults in the June-December
period had little variation among colonies (Capi-
vari = 207.5 ± 4.8 mm, n — 82; Arenitos =
206.6 ± 4.7 mm, n = 242; Anhangava = 208.6 ±
4.1 mm, n - 167; Furna 1 = 205.7 ± 4.9 mm, n
— 133). Considering similar random samples of
the same period and size from each colony, wing
length of birds from Furna 1 was smaller than in
the other colonies (ANOVA = 1 1.44, P <

0.001; variances were homogeneous according to
Bartlett, Brown-Eorsythe, and Levene tests).

Tail length of adults not molting the innermost
feathers (first and second pairs of rectrices) was
69.9 ± 4.6 mm (/; = 802). An individual captured
at Furna 1 on 8 April 2001 had the longest tail
(83.0 mm), while an individual captured in
Anhangava on 19 January 1995 had the shortest

tail (51.0 mm) with a worn tip and molt of
rectrices 5-6 (outermost feathers). Individuals
without molt of the first rectrices, captured from
January to March, exhibited worn tail tips. The
tail in this period was shorter than during the rest
of the year (meanjan-Mar = 67.8 ± 4.7 mm, n =

1 16; meanApr-Dec = 70.3 ± 4.5 mm, n = 685) {t
= 5.38, df = 799, P < 0.001). The largest
monthly mean (76.0 ± 3.5 mm, n = 96) was in
June and the smallest (66.6 ± 3.8 mm, >1 = 99)
was in January (Table 1 ).

Tail lengths of subadults varied between 55.0
and 71.0 m,m throughout the year, resulting in a
mean of 65.4 ± 4.0 mm (n = 27). The birds had
longer tails during January-March than during the
rest of the year (meanjan-Mar = 68.3 ± 2.5 mm, n
= 9; meanApr-Dec = 64.0 ± 3.8 mm, n = 18) (r =
3.13, df = 25, P = 0.005). Tails of adults in the
second half of the year were longer than those of
subadults (t = 5.89, df = 701, P < 0.001). Length
of tails of adults in the period without molt of the
rectrices (Apr-Dee) varied among colonies (Cap-
ivari = 68.0 ± 4.0 mm, n = 95; Furna 1 = 68.9
± 5.1 mm, n = 144; Anhangava = 70.5 ±
3.4 mm, /? = 183; Arenitos = 71.7 ± 4.4 mm, n
= 263). Considering similar random samples of
the same period and size from each colony, tail
length of birds from Furna 1 was smaller than in
the other colonies and larger than birds from
Arenitos (ANOVA F3218 = 28.79, P < 0.001).

The tarsus of adults measured 25.9 ± 0.6 mm
(/? = 695). An individual captured in Anhangava
had the longest tarsus (27.5 mm), and two
individuals from this colony and two from Furna
1 had the shortest tarsi (24.3 mm). The tarsus of
subadults was 25.8 ± 0.5 mm (24.7-26.7 mm, n
= 24), which did not differ from measurements of
the adults (l = 0.22, df = 717, P = 0.83). The
mean length of the tarsus of adults was similar
among colonies (Furna 1 = 25.8 ± 0.6 mm, n =
155; Capivari = 25.8 ± 0.6 mm, 11 = 108;
Anhangava = 25.8 ± 0.7 mm, /? = 114; Arenitos
= 25.9 ± 0.6 mm, n = 318) (ANOVA F3.69i.=
1.42, P = 0.24; variances were homogeneous
according to Bartlett, Brown-Forsythe, and Le-
vene tests).

The exposed culmen of adults measured 9.94 ±
0.36 mm (/; = 658). An individual captured in
Arenitos had the longest exposed culmen
(11.1 mm), and at least one specimen from each
colony had the shortest expo.sed culmen (9.0 mm).
Individuals with head feather molt had longer
exposed culmens (meanwith molt = 9.98 ±

Pichorim • MORPHOMETRICS OF THE BISCUTATE SWIFT

157

mm, f\ 376, mctin^id-^oui mou 9.88 —
0.37 mm, n = 282; t = 3.40, df = 656, P < 0.001).
The nostril beak, tip (mean = 6.27 ± 0.29 mm, n =
658) did not vai'y with molt (meanwini mou = 6.29
it 0.28 mni, /? 376, mean^ij).(Qi^ii niou 6.24 —
0.32 mm, n = 282) (/ = 1.95, df = 656, P =
0.052). Siibadults had exposed culmens similar to
adults without molt (meaUsubaduits = ±

0.29 mm, n = 15; / = 1.90, df = 295, P =
0.058), but the nostril beak tip was shorter
(meansubaduit = 6.07 ± 0.31 mm, n = 15; r =
2.09, df = 295, P = 0.037). Considering the nostril
beak tip measurement, which did not vary with
molt, and random samples of similar size from each
colony, individuals from Furna 1 had the shortest
beak (Furna 1 = 6.19 ± 0.34 mm, n — 100;
Capivari = 6.28 ± 0.26 mm, n = 100; Arenitos =
6.31 ± 0.27 mm, n = 100; Anhangava = 6.31 ±
0.30 mm, /? = 97) (ANOVA ^3,393 = 3.66, P =
0.013).

DISCUSSION

Body mass of Biscutate Swifts depended on
time of day, age, and season. There was no
evidence of differences in mass between males
and females, a valid comparison as seasonal
samples had normal distributions. Feeding was
presumably the cause of daily mass variations, as
individuals captured in late afternoon were
heavier due to their digestive systems being full.
Mass of subadults was lower than adults during
the greatest part of the first year of life. The
difference in mass was observed in summer, fall,
and winter. Other subadult swifts have a similar
tendency (e.g.. Chimney Swift [Chaetura pela-
gica]. Chapman’s Swift [C. chapmani], and
Alpine Swift [Tachymarptis melha]) (Coffey
1958, Collins 1968, Telia et al. 1995). Fat storage
in nestlings increases their chances of survival
after leaving the nest. However, young swifts
probably experience .some difficulty obtaining
food during the first months of independent life. It
is known that Biscutate Swifts decrea.se in mass
between the nestling pha.se and the first months as
subadults (Pichorim 2002). However, the new
data demonstrate that mass of subadults is below
that of adults for longer than previously known. It
is likely that subadult survival during the first year
of life is lower than for adults, as it is for some
Apodidae (Chantler and Driessens 1995, Chantler
1999). Mass equalization between age groups
occurs during the spring at the end of the first year
of life. Reserves stored during the nestling phase

appear to be indispensable for the survival of
subadults. Control of brood size in the Biscutate
Swift is achieved by egg ejection, probably during
unfavorable conditions (Pichorim and Monteiro-
Filho 2008). This behavior allows for heavier
fledglings and, consequently, higher subadult
survival during the first year.

Seasonal variation in the mass of adults may be
related to prey availability, reproductive period,
and molt with mass peaking in fall and spring, as
observed for Chimney Swift (Coffey 1958,
Johnston 1958) and Common Swift {Apiis apus)
(Lack and Lack 1951). The first period coincides
with the end of molt when demand for energy is
reduced. The new plumage may also act to
increase mass, as it improves flight efficiency
and facilitates capture of prey. It is also possible
that storing of reserves before winter indicates an
adaptation for low food availability. Increases in
mass occur in spring due to increased resource
availability and breeding. Mass starts decreasing
by December, which is certainly related to the
effort of caring for nestlings. The decrease in
mass in summer is probably caused by the
increased metabolic demand of molt. The maxi-
mum gain and loss of mass for the same
individual in a year (30 and 25%, respectively)
were close to rates for the Common Swift (31.3
and 13.2%, respectively (Lack and Lack 1951).
The 13% overnight decrease in mass detected is
the highest rate reported for Apodidae. The
highest value previously reported for the Common
Swift was 10% (Lack and Lack 1951).

Mass decrease of individuals captured in
morning and at day’s end could be linked to
stress of capture. However, captured individuals
were set free only after other individuals left the
overnight shelter and were subsequently isolated
from other individuals of the colony. Thus, the
mass decrease could be an indication the species
u.ses a technique of hunting for food in flocks and.
when an individual is apart from the Bock, it
forages less efficiently. Schoener (1971) reports
that grouping benefits animals when food avail-
ability fluctuates over time. This technique
enlarges the monitored area and increases the
chances of finding prey when resources are
restricted or infrequent. Individuals may experi-
ence more difficulty in feeding when flying alone
or in small flocks.

Seasonal variation of wing and tail lengths was
mainly due to natural wear. Wings were shorter at
the start of molt because the feathers were worn

158

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 1, March 2010

from use. Mean lengths were greater after molt
with the new plumage. Wings and tails of subadults
were shorter than those of the adults, as subadults
do not molt during the first year of life and have
older feathers. The difference in tail lengths of
subadults between summer and other periods of the
year also appeared to be caused by wear.

There was no difference in tarsus length of
adults and subadults. The tarsus of S. hisciitata
grows quickly during the nestling stage, reaching
maximum development at the 19th day of life;
this serves the nestlings well for hanging onto
their nest (Pichorim 2002). Tarburton (1986)
reported the tarsi of White-rumped Swiftlets
(Aerodramus spocliopydiiis assimilis) in Fiji
reached adult size on day 14, whereas adult
weight was not reached until day 23 and wing-
length sometime after fledging. Marin (1997) also
showed the tarsus of Cypseloides uiger reached
adult length in about 12 days whereas mass took
about 20 days and wing reached adult length post
fledging. The tarsus is a good moiphological
measurement to compare populations due to its
stability and quick development.

Molt of head feathers influenced the exposed
CLilmen, which implied a certain limitation in
using it as a comparison between populations that
were not sampled during the same period. No
plausible cause points to age-related variations of
the nostril beak tip length, but the differences
could be caused by increased cranial ossification
or by the development of the beak.

Means of body mass and lengths of wings, tails,
and nostril beak tip were significantly smaller in
Furna 1. These data suggest that individuals from
this colony experience special conditions, which
may be related to competition for better places for
resting, overnight roosting, and nesting with
White-collared Swifts (Streptopvocne zonaris) (a
large number of which inhabit the same shelter).

This study demonstrates that several morpho-
logical parameters of the Biscutate Swift have
significant seasonal variations. There is evidence
that similar variation occurs in other species of the
Family Apodidae. Thus, studies to compare
Apodidae populations must consider the depen-
dence of certain morphological measurements on
season and age. The existence of morphological
differences over a latitudinal gradient is probable
in some swifts. However, comparison of samples
from different seasons could accentuate the
differences between populations and lead to
misinterpretation.

AC KNOWLEDGMENTS

The author thanks the Conselho Nacional de Desenvol-
vimento CientiTico e Tecnologico (CNPq) for financial
support, the Post-Graduate Program in Zoology at the
Federal University of Parana for institutional support, and
the Instituto Ambiental do Parana (lAP) for access to Vila
Velha State Park. 1 also thank Alexandre Lorenzetto, Aline
DaPMaso Ferreira, Andressa C. Bonat, Arthur Angelo
Bispo de Oliveira, Bianca Luiza Reinert, Deborah Afon.so
Cornelio, Cristina Imaguire, Cristiani Hagar Flores, Diter
Liepsch, Douglas Kajiwara, Juliano Ribeiro, Marcos
Ricardo Bornschein, Marlon Carlos Borba, Nilton Carlos
Caceres, Olivia Isfer, Paula Elbl, Paulo Eduardo de Souza,
Priscila Barbosa, Reginaldo Assencio Machado, and
Ricardo Voiles for help in the field. I appreciate the
improvements in English usage made by James Armacost
through the Association of Eield Ornithologists’ program of
editorial assistance. The study complies with the current
laws of Brazil in which it was performed.

LITERATURE CITED

Browning, M. R. 1993. Species limits of the Cave
Swiftlets {Collocalia) in Micronesia. Avocetta
17:101-106.

Chantler, P. 1995. Identification of three Chaetura swifts;
Band-rumped Swift Chaetura spinicauda, Grey-
rumped Swift Chaetura ciuereiventris and Pale-
rumped Swift Chaetura egregia. Cotinga 4:44-5 1 .
Chantler, P. 1999. Family Apodidae (Swifts). Pages 388-
457 in Handbook of the birds of the world. Volume 5.
Barn-owls to hummingbirds (J. del Hoyo, A. Elliott, and
J. Sargatal, Editors). Lynx Edicions, Barcelona. Spain.
Chantler, P. and G. Driessens. 1995. Swifts, a guide to
the swifts and treeswifts of the world. Pica Press. East
Sussex. United Kingdom.

Coffey, L. C. 1958. Weights of .some Chimney Swifts at
Memphis. Bird Banding 29:98-104.

Collins, C. T. 1968. Notes on the biology of Chapman’s
Swift Chaetura chapmani (Aves: Apodidae). Ameri-
can Museum Novitates 2320:1-15.

Collins, C. T. 1972. A new species of swift of the genus
Cypseloides from northeastern South America. Con-
tributions in Science of the Natural History Museum of
Los Angeles County 229:1-9.

Eisenmann. E. and F. C. Lehmann. 1962. A new species
of swift of the genus Cypseloides from Colombia.
American Museum Novitates 21 17:1-16.

Johnston, D. W. 1958. Sex and age characters and salivary
glands of the Chimney Swift. Condor 60:73-84.
Lack. D. and E. Lack. 1951. The breeding biology of the
swift Apus apus. Ibis 93:501-546.

Marin, A. M. and F. G. .Stiles. 1992. On the biology of
five species of swifts (Apodidae, Cypseloidinae) in
Costa Rica. Western Foundation of Vertebrate Zool-
ogy 4:287-351.

Marin, M. 1997. Some aspects of the breeding biology of
the Black Swift. Wilson Bulletin 109:290-306
Medway, F. L. S. 1966. Field characters as a guide to the
specific relations of swiftlets. Proceedings of the
Linnean Society of London 177:151-172.

Pichorini • MORPHOMETRICS OF THE BISCUTATE SWIFT

159

Naik, S. and R. M. Naik. 1966. Studies on the House Swift,
Apiis iijfhiis (G. E. Gray), body weight. Pavo 4:84-91.

N.avarro. a. G. S., a. T. Peterson, B. P. P. E.scalante,
and H. D. BenItez. 1992. Cypseloides sloreri, a new
species of swift from Mexico. Wilson Bulletin 104:55-
64.

Parkes, K. C. 1993. Taxonomic notes on the White-
collared Swift (Streptoprocne zonaris). Avocetta
17:95-100.

PiCHORiM, M. 2002. The breeding biology of the Biscutate-
swift, Streptoprocne hiscutata (Apodidae) in southern
Brazil. Ornitologia Neotropical 13:61-84.

PiCHORiM, M. AND E. L. A. MoNTEtRO-FiLHO. 2008. Brood
size and its importance for nestling growth in the
Biscutate Swift (Streptoprocne hiscutata, Aves: Apod-
idae). Brazilian Journal of Biology 68:851-857.

PtCHORIM, M. AND E. L. A. MONTEIRO-FlLHO. 2010.
Population size, survival, longevity, and movements
of the Biscutate Swift in southern Brazil. Annales
Zoologici Fennici 47: In press.

SAS. 2001. Program JUMP. SAS In.stitute Inc, Cary, North
Carolina, USA.

Schoener, T. W. 1971. Theory of feeding strategies.
Annual Review of Ecology and Systematics 2:369-
404.

Sick, H. 1991. Distribution and subspeciation of the
Biscutate Swift Streptoprocne hiscutata. Bulletin of
the British Ornithologists’ Club 3:38^0.

Sims, R. W. 1961. The identification of Malaysian species
of swiftlets Coliocaiia. Ibis 103:205-209.

Tarburton, M. K. 1986. Breeding of the White-rumped
Swiftlet in Fiji. Emu 86: 214-227.

Tella, J. L., C. Gortazar, R. Lopez, and U. Osacar.
1995. Age related differences in biometrics and body
condition in a Spanish population of Alpine Swift
(Apus meiha). Journal ftir Ornithologie 136:77-79.

Wetmore, a. 1957. Species limitation in certain groups of
the swift genus Chaetura. Auk 74:383-385.

Zar, J. H. 1984. Biostatistical analysis. Second Edition.
Prentice-Hall, Englewood Cliffs, New Jersey, USA.

Short Communications

The Wilson Journal of Ornithology 122(1): 160-162, 2010

Cooperative Breeding by Red-Headed Woodpeckers

Megan R. Atterberry-Jones' and Brian D. Peer'-^

ABSTRACT. — Red-headed Woodpeckers (Mela-
nerpes eiythrocephalus) are one of the most easily
recognizable bird species, but little is known about their
social system and breeding activity due to their lack of
sexual dimorphism. We observed a dense population of
Red-headed Woodpeckers and found evidence of
cooperative breeding. The habitat at our site may
promote cooperative breeding because it has a high
density of utility poles that woodpeckers use for nesting,
roosting, and caching food. Received 7 April 2009.
Accepted 30 July 2009.

Cooperative and communal breeding in which
some individuals forgo breeding and assist other
breeding pairs have been described for > 300
species of birds (Stacey and Koenig 1990,
Cockburn 2003, Koenig and Dickinson 2004).
Seven of 21 members of Melanerpes woodpeck-
ers, including the Acorn Woodpecker (M. for-
micivoriis) and Hispaniolan Woodpecker (M.
striatus), are known to breed either cooperatively
or communally (Koenig and Mumme 1987, Ligon
and Burt 2004). Relatively little is known of the
breeding behavior and social interactions of the
Red-headed Woodpecker (M. erythrocephalus) as
they are sexually monomorphic and difficult to
capture (Smith et al. 2000). Anecdotal observa-
tions suggest Red-headed Woodpeckers may not
be the monogamous loners as portrayed. They
have been observed migrating in groups of 5-6
individuals suggesting they may remain in family
groups for extended periods of time (Hall 1983),
and three adults have been observed attending the
same nest (Short 1982). These observations raise
the possibility that some populations of Red-
headed Woodpeckers may be cooperative breed-
ers.

The objectives of our study were to: ( 1 ) observe
a locally dense population of Red-headed Wood-
peckers in west-central Illinois to better under-
stand their social behaviors during the breeding
.season, and (2) examine if the den.se population

' Department of Biological Sciences, Western Illinois
University, Macomb, IL 6145.'i, USA.

^Corresponding author; e-mail: BD-Peer@wiu.cdu

was a result of cooperative breeding or non-
breeding floaters in the area.

METHODS

The study site was a 6.75-ha campground at
Spring Lake Park (40° 50' N, 90° 71' W) in
Macomb, Illinois, USA. The area was relatively
open with the dominant vegetation being oak
{Quercus spp.) and hickory [Carya spp.) trees
with snags present throughout the area, in
addition to 31 utility poles. Individual wood-
peckers were captured in mist nets using call
backs and mounted specimens, and banded with
uses bands and a unique combination of color
bands. Fifteen Red-headed Woodpeckers were
captured and banded. Red-headed Woodpeckers
were observed during the 2007 and 2008
breeding seasons (late Mar through early Aug)
to identify: (1) number of territories, (2)
territory boundaries and whether territories
overlapped, (3) color band combinations of
marked individuals in each territory, (4) nest
site locations, (5) number of birds and the
identity of individuals tending a nest, and (6)
number of birds and the identity of individuals
defending territories.

Territories were identified by observing a
mated pair’s movements and then mapped onto
a satellite image of the campground. Each
territory was observed for a 2-hr time interval
between 0800 and 1200 hrs CDT once a week.
Nests in territories became the main focus of
observations, and the identities of individuals
coming and going to the nest were recorded. If a
nest was not located, we followed the individuals
in that territory to locate potential future nest sites.
Additional ob.servations were made if more than
two individuals were found in a territory to learn
if the extra birds were floaters impinging on the
mated pair’s territory, or if they were helpers
assisting the mated pair. Helpers were classified
as individuals that helped defend the territory, but
did not care for the young, or as individuals
observed at the nest caring for the young and
defending the territory.

160

SHORT COMMUNICATIONS

161

RESULTS

Red-headed Woodpeckers occurred in the
campground year around and at least some
individuals maintained year-around territories.
Eleven breeding territories were located during
the 2007 breeding season. Ten territories had two
individuals, while one had three individuals.
Three floaters were observed in at least one
census during the breeding season. Boundaries of
territories overlapped with at least one other
breeding territory. Two breeding pairs used utility
poles as their nesting site and the other nine used
trees.

There were 10 breeding territories in the 2008
breeding season. One contained three individuals
while there were two each in the other nine
territories. All territories in 2008 had boundaries
that overlapped with at least one other breeding
territory. One floater was observed during one
census in the 2008 breeding season. The same two
utility poles were used as nesting sites for two
breeding pairs with the remainder of the breeding
pairs using trees. Fifty-eight percent of the utility
poles (« = 31) were used for nesting, roosting,
and/or caching food.

We observed cooperative breeding in both
years of study. The cooperative group in 2007
consisted of an individual with a white band and
two unbanded individuals that nested in a white
oak (Qiiercus alba) on the east side of the
campground. All three individuals were seen
simultaneously on the tree that contained the nest
and responded with chatter calls when they landed
near another territory member. All three also
responded aggressively with nest attentiveness
and alarm calls to a Red-headed Woodpecker call
played in front of the tree that contained the nest.
All three birds defended the territory, but only the
white-banded individual and one of the unbanded
individuals that had a distinctive stripe on its beak
were observed tending the nest.

The cooperative breeding group during the
2008 breeding season was on the west side of the
campground nesting in a utility pole. The group
consisted of the same white banded individual, an
unbanded bird which retained juvenile coloration
around the eyes, and a third unbanded bird. All
three birds defended the territory from invaders
and tended the nest. All three individuals were
seen on the utility pole with the nest and all flew
to the utility pole when one bird emerged from the
nest, giving nest attentiveness and chatter calls to

one another in the process. All three were
observed on several occasions taking turns
incubating eggs or brooding young. All three
responded aggressively with alarm calls when a
Red-headed Woodpecker call was played near the
utility pole.

DISCUSSION

The Red-headed Woodpecker population we
studied in west-central Illinois appeared to
occasionally have cooperative breeding. One
breeding territory in 2007 had three individuals,
all of which defended the territory, but only two
were observed tending the nest. It is possible the
third bird also tended the nest unobserved. One
breeding territory in 2008 contained three birds,
all of which defended the territory and all three
were observed tending the nest. The relationship
between helpers and the breeding pair was
unknown because we were unable to band all
birds and did not document genetic relationships.
The presence of the after hatching year bird in
2008 suggests it could have been one of
“white’s” young from 2007. It is unclear whether
these were two distinct cases of cooperative
breeding or whether it was the same white-banded
bird and its mate with a different helper in both
years because all of the birds were not banded.
This also may be an example of irregular
cooperative breeding (Ligon and Burt 2004),
although the breeding behavior of these birds
requires further study.

The cooperative breeding we observed in
addition to Short’s (1982) observations of three
adult Red-headed Woodpeckers tending a nest
suggests a low level of cooperative breeding
occurs within this species. What factors are
driving cooperative breeding in our population?

First, most birds in our population maintained
year-around territories which can result in greater
adult survival and increased population densities,
and more breeding adults than teiritories (Arnold
and Owens 1999). Second, the open oak habitat in
the park appears ideal for Red-headed Wood-
peckers (Smith et al. 2000). The Red-headed
Woodpecker population density in the park was
greater than average for Illinois. The average
number of breeding individuals in similar habitats
in Illinois was 0.68/ha (Graber et al. 1977), but at
our site there was an average of 1.6 breeding
individuals/ha. The park is typical of many
midwestern parks and campgrounds in that it is
relatively open and dominated by oaks and

162

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 1, March 2010

hickories (pers. obs.). Parks within 16 km of our
site were similar structurally, but have typical
numbers of Red-headed Woodpeckers (unpubl.
data). The most obvious feature that distinguished
our site was the large number of utility poles, the
majority of which were used by woodpeckers. It is
possible the utility poles make this habitat
productive by providing additional nesting and
food storage sites.

ACKNOWLEDGMENTS

We thank S. H. Abott and Raymond Peterson for
allowing us to study the Red-headed Woodpecker population
at Spring Lake Park, J. D. Jones for helping with field work,
and two anonymous reviewers who provided insightful
comments on the manuscript. This research was supported by
grants from the Illinois Ornithology Society and the School
of Graduate Studies at Western Illinois University.

LITERATURE CITED

Arnold, K. E. and I. P. F. Owens. 1999. Cooperative
breeding in birds: the role of ecology. Behavioral
Ecology 5:465-471.

Cockburn, A. 2003. Cooperative breeding in oscine
passerines: does sociality inhibit speciation? Proceed-
ings of the Royal Society of London, Series B 270:
2207-2214.

Graber, J. W., R. R. Graber, and E. L. Kirk. 1977.
Illinois birds: Picidae. Biological Notes Number 102.
Illinois Natural History Survey, Urbana, USA.

Hall, G. A. 1983. West Virginia birds. Special Publica-
tions Number 7. Carnegie Museum of Natural History,
Pittsburgh, Pennsylvania, USA.

Koenig, W. D. and J. L. Dickinson. 2004. Ecology and
evolution of cooperative breeding in birds. Cambridge
University Press, Cambridge, United Kingdom.

Koenig, W. D. and R. L. Mumme. 1987. Population
ecology of the cooperatively breeding Acorn Wood-
pecker. Monographs in Population Biology Number
24. Princeton University Press, Princeton, New Jersey,
USA.

Ligon, j. D. and D. B. Burt. 2004. Evolutionary origins.
Pages 5-34 in Ecology and evolution of cooperative
breeding in birds (W. D. Koenig and J. L. Dickinson,
Editors). Cambridge University Press, Cambridge,
United Kingdom.

Short, L. L. 1982. Woodpeckers of the world. Monograph
Series Number 4. Delaware Museum of Natural
History, Wilmington, USA.

Smith, K. G., J. H. Withgoti', and P. G, Rodewald. 2000.
Red-headed Woodpecker (Melanerpes erythrocepha-
lus). The birds of North America. Number 518.

Stacey, P. B. and W. D. Koenig. 1990. Cooperative
breeding in birds: long-term studies of ecology and
behavior. Cambridge University Press, Cambridge,
United Kingdom.

The Wilson Journal of Ornithology 122( 1 ): 162-165, 2010

Observations on the Breeding Biology of the Collared Crescentchest

(Melanopareia torquata)

Mieko F. Kanegae,'-^ Marina Telles,“ Severino A. Lucena,^ and Jose Carlos Motta-Junior'

ABSTRACT. — The Collared Crescentchest (Melano-
pareia torquata) is an endemic bird of the Cerrado
(Family: Melanopareiidae), and is listed in the State ol
Sao Paulo, Brazil as “endangered". We studied the
breeding biology of Collared Crescentchest at two nests
in the State of Sao Paulo, southeast Brazil. Males were
identified genetically and equipped with radio-transmit-

' Laboratorio de Ecologia de Aves, Departamento de
Ecologia, Instituto de Biociencias da Univcrsidade de Sao
Paulo, 05508-900 Sao Paulo, SP, Brazil.

- Pos-Graduayao cm Ecologia e Recursos Nalurais pela
Univcrsidade Federal de Sao Carlos, Brazil.

'Laboratorio de Bioqui'mica de Insetos, lOC-FlOCRUZ,
Brazil.

■‘Corresponding author: e-mail: mieko. kanegae@gmail.
com. miekok@hotmail.com

ters. The incubation period was 12-16 days and the
nestling period was 12-14 days. Nestling body mass
was measured every second day for the first 10 days.
Males participated in incubation and helped with
nesting care. Measurements of eggs and nests are
compared to those from the single previously known
nest. These data are the first for any member of the
Family Melanopareiidae. Received 27 March 2009.
Accepted 28 August 2009.

Little is known about the life history and
ecology of tropical species, particularly with
regard to endangered and threatened species of
the Melanopareiidae and Rhinocryptidae (Krabbe
and Schulenberg 2003). Recently, the nest, eggs.

SHORT COMMUNICATIONS

163

and young of the Collared Crescentchest (Mela-
nopareia torqiiata) were first described (Gressler
and Marini 2007). The Collared Crescentchest is
an endemic bird of the Cerrado region (Silva and
Bates 2002). Willis and Oniki (2003) indicated the
population is declining in the State of Sao Paulo,
Brazil. The species is classified as “Endangered”
in the List of Animals Threatened with Extinction
in the State of Sao Paulo (Decree 53.494, 2 Oct
2008).

Conservation of biodiversity requires a refined
knowledge of species and their biology, especially
in the tropics, where life histories of birds have
been little investigated (Stutchbury and Morton
2001). Our study was directed at adding new
natural history information about the Collared
Crescentchest, especially reproductive and behav-
ioral aspects with a conservation focus. We asked
the following questions: (1) are both parents
involved with incubation and nestling care, (2)
what is the nest visitation frequency of adults, and
(3) how long do nestlings stay in the nest?

METHODS

Study Area. — ^This study was conducted in the
2,300-ha Esta9ao Ecologica de Itirapina (EEI) in
the municipalities of Itirapina and Brotas (22° 15'
S, 47° 49' W) in Sao Paulo State, Brazil. This area
is one of the last remnants of natural grasslands
and grassland savannahs in Sao Paulo State
(Gianotti 1988). The Cerrado has a complex of
physiognomies from open grasslands (campo
limpo) to forest (cerradao), including intermediate
types “campo sujo”, “campo cerrado”, and
“cerrado sensu stricto" (Ribeiro and Walter
1998). Climate in this region, according to
Kdppen’s (1948) classification, is defined as
Cwa type, receiving on average 1,376 mm of
precipitation annually. During the dry season
(Apr-Sep), rainfall ranges from 32 to 88 mm
per month and, during the rainy season (Oct-
Mar), from 117 to 257 mm per month (DAEE
Post D4-014, Itirapina, SP).

Field Procedures. — We captured individual
Collared Crescentchest, after hearing them, using
a single mist net (12 X 2 m) and, at times using
playback calls. Males and females were identified
genetically by taking blood samples (0.1 ml) from
the jugular vein with analysis at the Laboratorio de
genetica e evolu^ao de aves in the Universidade de
Sao Paulo following Griffiths et al. (1998).

We attached LD-2 radio transmitters (Holohil
Systems Ltd, Carp, ON, Canada) using a back-

pack harness on 10 individuals. We obtained
radio-telemetry data between October and No-
vember 2007 during the morning (0600-1200 hrs)
and evening (1600—1800 hrs) in areas of “campo
sujo” and “campo cerrado”. We located nests by
following these individuals and observing their
behavior. Nests were monitored every 2 days and
the behavior of parents was observed during each
visit for 30 min at a distance of ~ 15 m from the
nest. Observations were conducted 3 hrs after
sunrise, 3 hrs around noon, and 3 hrs before
sunset. This procedure allowed us to record
frequency and duration of visits to the nest. We
measured body mass using a spring scale (Pesola,
50 g, accurate to 1 g), and size of eggs and
nestlings was measured with a digital calliper
(Mitutoyo, 0.01 mm precision).

RESULTS

We found two nests of M. torquata in 2007.
Both were woven of dry grass leaves and a few
dry leaves of trees or shrubs. One of the nests was
half covered by grass leaves of Loudetiopsis
chrysothrix (Poaceae), a grass commonly ob-
served in the area.

Average egg mass was 1.55 g, and average egg
width and length was 19.05 and 14.95 mm,
respectively (Table 1). Eggs were ovoid-shaped,
unspotted with a greenish-blue coloration. The
incubation period was 12-16 days. The first nest,
with two nestlings, was found in an area of
“campo sujo” on 12 November. Both nestlings
left the nest 12 days later on 24 November.
Considering developmental stage of the nestlings
at the moment of nest discovery, the birds
probably fledged 13-14 days after the eggs
hatched. The second nest, with two eggs, was
found on 19 November in an area of “campo
cerrado”. One egg hatched on 29 November and
the offspring fledged 12 days later. However, this
nestling may have fledged 1-2 days prematurely
due to our visits. We could still detect the parents
around the nest 2 days after the young fledged,
where they gave a “pirrr” call. The other egg did
not hatch and was collected (deposited in the
Museu de Zoologia da Universidade de Sao
Paulo). The growth rates of three nestlings were
similar (Fig. 1).

Both males and females shared incubation and
nestling care. The male made few visits to the
nest, which were brief (1-2 min) and without
periodicity. We recorded two visits during the
nestling period in midday and one during

164

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 1, March 2010

TABLE 1. Measurements of three nests of the Collared Crescentchest obtained by Gressler and Marini (2007) in
Distrito Federal and average measurements of two nests during the present study, in Sao Paulo, Brazil.

Characteristics

Gressler and Marini (2007)

Present study

Width of entry, cm

4.2

6.0

Height of entry, cm

3.7

5.25

Nest length, cm

10

14.75

Nest depth, cm

10.5

14.75

Height from nest’s base to the ground, cm

13.5

15.5

Egg mass, g

1.55

Egg width, mm

21.2/19.5

19.05

Egg length, mm

15.7/15.4

14.95

Incubation period, days

15-18

12-16

Nestling period, days

12-14

afternoon. A male was in the nest during
incubation twice in the morning. It is probable
the males knew of our presence during monitor-
ing, which may have affected their behavior.

DISCUSSION

The proportions of the two Collared Cres-
centchest nests we found were similar, but larger
(Table 1) than those of the nests described by
Gressler and Marini (2007). They also observed
that dry leaves in the nests were used only as
support. Egg color was similar to that described
by Gressler and Marini (2007), but differed from
the descriptions of Sick (1997) for coloration of
Olive-crowned Crescentchest (Melanopareia
maximiliani) eggs by the lack of any speckles.

Eggs in this species were whitish with large dark
gray speckles near the large end (Di Giacomo
2005).

Periodicity of nest visits was low, which may
reflect a strategy against predators, thus reducing
chances of discovery of nests and eggs or
nestlings (Skutch 1976). We also noted a different
“pirr” call from adults near nests and suggest that
one of the functions of this call is to keep the
young near the nest, a behavior that may increase
their survival (Martin 1996, Stutchbury and
Morton 2001).

Our study complements the known life history
information of the Collared Crescentchest, a
species less studied, and contributes to the
knowledge of the behavior and reproductive
ecology of tropical birds.

— F1 -m— F2 -A- F3

FIG. I . Growth rates (body ma.ss) of three (FI , F2, F3) Collared Crescentchest nestlings measured every 2 days. FI and
F2 (Nest I ) were monitored between 1 2 and 24 November 2007. F3 (Nest 2) was monitored from hatching on 29 November
to I I December 2007.

SHORT COMMUNICATIONS

165

ACKNOWLEDGMENTS

We thank CEMAVE and the Institute Florestal do Estado
de Sao Paulo (IF) for authorization to capture birds; CAPES
for the doctoral scholarship; Laboratorio de Genetica e
Evolugao de Aves do Instituto de Biociencias da Uni-
versidade de Sao Paulo for bird gender identification;
Instituto de Biociencias da Universidade de Sao Paulo; Idea
Wild; the E. Alexander Bergstrom Memorial Research
Award; Neotropical Grassland Conservancy; W. J. and
Virsinia W. Moorhouse Memorial; Pamela and Alexander
F. Skutch; and Birders Exchange for the financial support
and equipment donation.

LITERATURE CITED

Di Giacomo, A. G. 2005. Aves de la Reserva El Bagual.
Pages 201^65 in Historia natural y paisaje de la
Reserva El Bagual, Provincia de Formosa, Argentina.
Inventario de la fauna de vertebrados y de la flora
vascular de un area protegida del Chaco Hiimedo. (A.
G. Di Giacomo and S. F. Krapovickas, Editors). Aves
Argentinas. Temas de Naturaleza y Conservacion 4.
Asociacion Ornitologica del Plata, Buenos Aires,
Argentina.

Gianotti, E. 1988. Composi§ao florfstica e estrutura
fitossociologica da vegeta9ao de Cerrado e mata ciliar
da Esta9ao Ecologica de Itirapina (SP). Disserta9ao.
Campinas, Universidade Estadual de Campinas, Sao
Paulo, Brazil.

Gressler, D. R. and M. A. Marini. 2007. Nest, eggs and
nestling of the Collared Crescentchest Melanopareia

torquata in the Cerrado region, Brazil. Revista
Brasileira de Ornitologia 15:574-576.

Griffiths, R., M. Double, K. C. Y. Orr, and R. J. G.
Dawson. 1998. A DNA test to sex most birds.
Molecular Ecology 7:1071-1075.

Koppen, W. 1948. Climatologia: con un estudio de los
climas de la Tierra. Fondo de Cultura Economica,
Mexico.

Krabbe, N. and T. Schulenberg. 2003. Family Rhino-
cryptidae (Tapaculos). Pages 748-787 in Handbook of
the birds of the world. Volume 8, Broadbills to
Tapaculos. (J. del Hoyo, A. Elliott, and D. Christie,
Editors). Lynx Edicions, Barcelona, Spain.

Martin, T. E. 1996. Life history evolution in tropical and
south temperate birds: what do we really know?
Journal of Avian Biology 27:263-271.

Ribeiro, j. F. and B. M. T. Walter. 1998. Fitofisionomias
do Bioma Ceirado. Cerrado: Ambiente e Flora. Pages
89-166 in Embrapa-CPAC (S. M. Sano and S. P.
Almeida, Editors). Planaltina, DF, Brazil.

Silva, J. M. C. and J. M. Bates. 2002. Biogeographic
patterns and conservation in the South American
Cerrado: a tropical savanna hotspot. BioScience
52:225-233.

Skutch, A. F. 1976. Parent birds and their young.
University of Texas Press, Austin, USA.

Stutchbury, B. j. M. and E. S. Morton. 2001.
Behavioral ecology of tropical birds. Academic Press,
San Diego, California, USA.

Willis, E. O. and Y. Oniki. 2003. Aves do Estado de Sao
Paulo. Divisa, Rio Claro, Brazil.

The Wilson Journal of Ornithology 122(1): 165-168, 2010

Effects of a Flood on Foraging Ecology and Population Dynamics of

Swainson’s Warblers

Nicholas M. Anich'-^ and Bryan M. Reiley'-^

ABSTRACT.— Swainson’s Warblers (Limnothlypis
swainsonii) have been described as terrestrial leaf litter
foraging specialists, and are thought to vacate flooded
areas entirely. We report Swainson’s Warblers occupy-
ing a flooded area in Arkansas and foraging on unusual
substrates. We observed Swainson’s Warblers feeding
in novel ways in the tree canopy and on floating debris.
This is the first report of Swainson’s Warblers
occupying and foraging in flooded habitat, and suggests
this species may have more short-term flexibility in

' Department of Biological Sciences, Arkansas State
University, P. O. Box 599, Jonesboro, AR 72467, USA.

^Current address: 2414 Fellman Circle, Ashland, WI
54806, USA.

^Corresponding author; e-mail: bryan.reiley@gmail.com

foraging substrate than previously thought. Return rates
to flooded territories were reduced compared to years
without floods and several marked individuals switched
to an unflooded area protected by man-made levees.
The altered hydrology of rivers constrained by levees
may reduce habitat quality for this species. Received 19
February 2009. Accepted 14 September 2009.

The Swainson’s Warbler {Limnothlypis swain-
soniv. Parulidae) is a medium-sized wood-warbler
that breeds in the southeastern United States,
winters in the Caribbean Basin, and inhabits areas
with dense understory vegetation (Brown and
Dickson 1994). Pairs are socially monogamous.

166

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 1. March 2010

and males exhibit breeding site fidelity and defend
their territory against conspecific males (Brown
and Dickson 1994). Swainson’s Warblers are
uncommon and local throughout their breeding
range, and are ranked among the highest-priority
species for conservation (Partners in Flight 2007).

Swainson’s Warblers forage for arthropods in
terrestrial dead leaf litter, primarily by flipping
dead leaves and poking their beaks into litter on
the ground (Meanley 1970, 1971; Barrow 1990;
Graves 1998). Barrow (1990) classified 82% of
foraging substrates {n = 41) as dead leaves,
ground litter, or fallen debris, and classified the
remainder as live leaves or herbs. Most (99%)
observations of foraging Swainson’s Warblers {n
= 399) by Graves (1998) involved lifting dead
leaves from the ground. More rarely, they have
been reported feeding at shrub level and hawking
insects (Meanley 1970, 1971; Barrow 1990),
although Graves (1998) attributed arboreal forag-
ing observations to instances of birds disturbed by
humans or audio playback. Swainson’s Warblers
at times sing from branches in the mid-story
(Meanley 1968), but they are considered primarily
ground foragers (Brown and Dickson 1994,
Graves 1998).

Meanley (1966) reported that none of four
territorial males returned to an area they had used
the previous year after a spring flood in Georgia,
and characterized the effect of flooding on
Swainson’s Warblers as “devastating.” Benson
and Bednarz (2010) examined Swainson’s War-
bler habitat in Arkansas several years after
Hooding and found only nine of 42 previously
occupied locations were still occupied by Swain-
son’s Warblers. Graves (2001) and Klaus (2004)
also reported abandonment of flooded territories.
It is not unusual for Swainson’s Warblers to occur
adjacent to water or to inhabit areas that can be
seasonally flooded, but flooding during the
breeding season is thought to preclude site
occupancy (Graves 2001, Brown et al. 2009).

Our objective is to report on response of
Swainson’s Warblers to a catastrophic Hood
event. We hypothesized that flooding would make
the leaf litter and other ground debris unavailable
for foraging, and therefore cause individuals to
vacate an inundated area, possibly even resulting
in local population extirpation.

OBSERVATIONS

Our observations were made in the Alligator
Lake area of White River National Wildlife

Refuge (WRNWR), Arkansas (34° 2' N, 91° 5'
W) from 24 April to 6 May 2008. This site was
being studied as part of an ongoing project: ( I ) it
had a continuous Swainson’s Warbler population
since at least 2000, (2) had been intensively
studied since 2004, and (3) most temtorial males
were color-banded. The site is several meters
higher in elevation than most of the refuge, which
protects it from flooding in most years allowing
development of dense understory vegetation
preferred by Swainson’s Warblers. Heavy rains
in spring 2008 caused the White River to rise,
flooding WRNWR, which is within earthen levees
that contain the river. The floodwaters that would
have naturally inundated the much larger flood-
plain were confined within the boundaries of the
levees. High water in the Mississippi River also
prevented normal drainage of the White River.
Additionally, an Army Corps of Engineers
pumping plant was pumping water into WRNWR
from agricultural areas bordering the refuge. The
site became completely inundated in late March
and the water started to recede in late April,
disappearing (except for isolated pools) by 6 May
2008.

The first Swainson’s Warblers returned to this
site on 8 or 9 April in 2005 through 2007 (years
with no flooding). We surveyed all previously
known Swainson’s Warbler temtories in the study
area beginning 24 April 2008 via canoe by
listening for their songs and using audio play-
backs. We encountered Swainson’s Warblers
singing and foraging in trees in areas that were
inundated with 0.91-2.75 m of water.

We observed four Swainson’s Warblers forag-
ing in atypical ways. One unbanded male sally-
hovered (terminology follows Remsen and Rob-
inson 1990) above a pile of floating debris and
appeared to attempt to pick an insect from it. This
bird then perched on a vine [Smila.x spp.) 2 cm
above the water and picked an insect from the
flotsam. A banded male was observed hopping
along horizontal branches of a tree ~ 10 m above
the water level, using typical leaf-lifting motions
to search for arthropods under loose pieces of
bark. This bird methodically hopped along one
horizontal branch, looking for arthropods under
loose bark, then gleaned insects from several
other large horizontal branches, until it flew to the
canopy of another tree. We observed a different
banded male and one banded female on separate
occasions searching for in.sects by lifting leaves
from hanging dead leaf clusters lodged in tree

SHORT COMMUNICATIONS

167

limbs. We observed some of these activities after
we had judiciously used audio playback, and in
other instances, we had not used playback.

We observed 10 different males and four
females in flooded habitat, including four instanc-
es of pairs traveling together. Seven of the 10
males were banded and three were unbanded. One
male that settled in flooded habitat had been
banded at WRNWR as a nestling in 2007 (T. J.
Benson and NMA, unpubl. data). We detected one
male in 2006 and 2007 holding a territory outside
the levee bordering the refuge. This area remained
dry during the flood, and the number of males
detected holding territories outside the levee
increased to six in 2008, including four coloi-
banded males that held territories inside the levee
in previous years.

DISCUSSION

Some Swainson’s Warblers survived the period
of flooding and eventually settled on territories
when the water receded, but we found reduced
return rates of color-banded individuals and
reduced site occupancy in 2008. We detected 25
males in 2008, compared to higher numbers in
previous years (40 in 2005; 43 in 2006; 47 in
2007; Benson 2008, Anich et al. 2009). The return
rate for color-banded males from 2007 to 2008
was 0.36 (17 of 47), compared to higher
proportions in previous years (0.53 from 2005 to
2006; 0.58 from 2006 to 2007; Benson 2008,
Anich et al. 2009).

These are the first observations of Swainson’s
Warblers foraging in the tree canopy. Tree
branches and floating debris were the only
substrate available on which birds could feed
during the prolonged flooding period. We ob-
served birds using previously unreported foraging
substrates, but these birds were using typical
foraging maneuvers supporting Graves (1998)
observation that Swainson’s Warblers use a
limited repertoire of foraging maneuvers. Graves
(1998) classified 99% {n = 399) of foraging
maneuvers as “leaf lifting,’’ which occuired with
associated gleaning, probing, and gaping motions,
and two instances of flush-pursuit of moths
(Lepidoptera) from the leaf litter. Strong (2000)
found 80% of foraging maneuvers in wintering
areas consisted of leaf lifting. Barrow (1990)
reported that 63% of foraging maneuvers (/; — 41 )
were flaking (analogous to leaf lifting), 32%
gleaning, and 5% (2 instances) sally-gleaning

(includes sally-strike and sally-hover-, Remsen and
Robinson 1990).

The literature, to our knowledge, does not
report Swainson’s Warblers occupying flooded
landscapes or feeding above shrub-level height.
Thus, we were suiprised to document their return to
breeding areas during the flood and their use of
novel foraging substrates. It is perhaps not
surprising that Swainson’s Warblers have evolved
the flexibility to survive temporary flooding, as
they often inhabit bottomland forests. However,
anthropogenic changes to natural flooding regimes
appear to have increased the duration and depth of
the high water at our site, and at other sites
occupied by Swainson’s Warblers (Graves 2001,
Thompson 2005). Some Swainson’s Warblers
persisted at the site thi'ough the flood and settled
on ten-itories when the flooding subsided. Howev-
er, lower return rates of color-banded individuals
and reduced site occupancy in 2008 suggest that
other individuals either vacated their territories
when confronted with flooding or perished.

Several authors have noted abandonment of
territories after flooding (Meanley 1966, Graves
2001, Klaus 2004). Previously we assumed the
mechanism by which flooding prevented Swain-
son’s Warbler occupancy was through direct
removal of the primary foraging substrate. Our
observations suggest that abandonment may result
from the initial effects of the flood (i.e.,
elimination of ground for foraging) but also
longer-term changes that reduce habitat quality
for Swainson’s Warblers. Benson and Bednarz
(2010) examined Swainson’s Warbler habitat at
WRNWR and three other sites within 3 years of
flooding and found only 21% of sites occupied
after the flood. They also found that locations
exhibiting more flood-related habitat changes,
such as reduced leaf litter and shrub cover, were
more likely to be unoccupied. We observed that
Swainson’s Warblers have the flexibility to
survive short-term flooding, but flooded and
recently flooded habitats appear to be suboptimal
for this species.

ACKNOWLEDGMENTS

Funding for this research was provided hy the U.S. Fish
and Wildlife Service and Arkansas Slate University. T. J.
Benson provided unpublished banding records. J. C.
Bednarz. T. J. Benson. S. K. Robinson. P. A. Spaeth, and
an anonymous reviewer provided insightful comments on
this manuscript. We thank J. C. Bednarz. T. J. Benson,
Carolina Roa, and Alex Zachary for assistance in the field.

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 1. March 2010
LITERATURE CITED Swainson’s Warbler along a hydrological gradient in

Anich, N. M., T. J. Benson, and J. C. Bednarz. 2009.
Effect of radio transmitters on return rates of
Swainson’s Warblers. Journal of Field Ornithology
80:206-211.

Barrow Jr., W. C. 1990. Ecology of small insectivorous
birds in a bottomland hardwood forest. Dissertation.
Louisiana State University, Baton Rouge, USA.

Benson, T. J. 2008. Habitat use and demography of
Swainson’s Warblers in eastern Arkansas. Disserta-
tion. Arkansas State University, Jonesboro, USA.

Benson, T. J. and J. C. Bednarz. 2010. Short-term effects
of flooding on understory habitat and presence of
Swainson’s Warblers. Wetlands 30: In press.

Brown, J. D., T. J. Benson, and J. C. Bednarz. 2009.
Vegetation characteristics of Swainson’s Warbler
habitat at the White River National Wildlife Refuge,
Arkansas. Wetlands 29:586-597.

Brown, R. E. and J. G. Dickson. 1994. Swainson’s
Warbler (Limnothlypis swainsonii). The birds of North
America. Number 126.

Graves, G. R. 1998. Stereotyped foraging behavior of the
Swainson’s Warbler. Journal of Field Ornithology
69:121-127.

Graves, G. R. 2001. Factors governing the distribution of

Great Dismal Swamp. Auk 118:650-664.

Klaus, N. A. 2004. Swainson’s Warblers may shift
territories in response to spring flooding. Oriole 69: 19.

Meanley, B. 1966. Some observations on habitats of the
Swainson’s Warbler. Living Bird 5:151-165.

Meanley, B. 1968. Singing behavior of the Swainson’s
Warbler. Wilson Bulletin 80:72-77.

Meanley, B. 1970. Method of searching for food by the
Swainson’s Warbler. Wilson Bulletin 82:228.

Meanley, B. 1971. Natural history of the Swainson’s
Warbler. North American Fauna Number 69.

Partners in Flight. 2007. Partners in Flight species
assessment database, http://www.rmbo.org/pif/pifdb.
html (accessed 10 November 2008).

Remsen Jr., j. V. and S. K. Robinson. 1990. A
classification scheme for foraging behavior of birds
in terrestrial habitats. Studies in Avian Biology
13:144-160.

Strong, A. M. 2000. Divergent foraging strategies of two
neotropical migrant warblers: implications for winter
habitat use. Auk 117:381-392.

Thompson, J. L. 2005. Breeding biology of Swainson’s
Warblers in a managed South Carolina bottomland
forest. Dissertation. North Carolina State University,
Raleigh, USA.

The Wilson Journal of Ornithology 122(1): 1 68-1 73, 2010

Preiiator Vocalizations Affect Foraging Trade-offs of Northern Cardinals

Mark T. Stanback'-^ and Emily M. Powell'

ABSTRACT. — We investigated the roles of food
quality and predator avoidance in shaping foraging
behavior of Northern Cardinals (Cardinalis cardinalis).
Cardinals visited feeders near cover and feeders with
sunflower seed significantly more than they visited
those in the open and those with commercial wild bird
seed mix. Cardinals with a choice between their
preferred food and preferred location visited the
sunflower seed feeder in the open significantly more
than the mixed seed feeder near cover, despite their
prior preference for feeding near cover. We recorded
visits to these feeders over a 2-day period during which
we played Cooper’s Hawk (Accipiter cooperi) calls
every 2 hrs. The overall visitation rate did not decrea.se
significantly, nor did the visitation rate to feeders near
cover increase significantly. However, the decrease in
the visitation rate to feeders in the open approached
significance. Overall, we observed a significant shift in
the proportion of visits to feeders near cover (containing

' Department of Biology, David.son College, Davidson,
NC 28035, USA.

^Corresponding author; c-mail: ma.stanback@davidson.edu

less preferred seed). We conclude Northern Cardinals
are sensitive to both food quality and predation risk, and
adjust foraging to prevailing perceptions of risk.
Received 23 March 2009. Accepted 23 July 2009.

Natural selection should favor organisms that
make optimal foraging decisions when presented
with conflicting options (Lima and Dill 1990,
Krebs and Kacelnik 1991, Cuthill and Houston
1997). The efficiency of energy/nutrient intake
and avoidance of predation are particularly
important factors affecting foraging animals (Dill
and Fraser 1984, Lima 1985, Lima and Valone
1986, Valone and Lima 1987). Many passerines
prefer to feed near cover due to higher risk of
predation in the open by Accipiter hawks (Morse
1973, Barnard 1979). Feeding in open areas
increases a bird’s visibility and the time it takes
to reach safety (Pulliam and Mills 1977); small
birds spend less time scanning for predators when

SHORT COMMUNICATIONS

169

near cover (Caraco et al. 1980). Given a particular
probability of the arrival of a predator, birds
should be able to perceive the relative danger of
visiting feeders that differ in distance to cover
(Lima 1985). Granivorous birds present an
excellent system for investigating trade-offs. Not
only can the quality of presented food be both
quantified and controlled, but researchers can also
manipulate predation risk by varying the distance
to the nearest cover.

Predation risk may vary seasonally or even
daily (Fenn and Macdonald 1995, Lucas et al.
1996, Lima and Bednikoff 1999). Moreover, prey
species can use visual, auditory, and/or chemical
cues to recognize the presence and threat level of
predators (Lima and Dill 1990, McLean and
Rhodes 1991). Consequently, researchers have
used a variety of techniques to manipulate the
perception of predation risk (Blumstein et al.
2008, Schmidt et al. 2008), including model
predators (Lima 1985), playbacks of conspecific
and heterospecific alarm calls (Peake 2005), and
recordings of predator sounds (McLean and
Rhodes 1991).

Our objectives were to: (1) assess how Northern
Cardinals (Cardinalis cardinalis), a common
feeder visitor, respond to variation in food quality
and distance to cover when subject to predation by
both migrant and resident Cooper’s Hawks
{Accipiter cooperi), and (2) examine it playing
calls of a Cooper’s Hawk near our test site would
alter the foraging decisions of Northern Cardinals.

METHODS

Field Procedures. — We installed four squirrel
(Sciurus spp.)-proof, hopper-style feeders in a
semi-secluded location on the campus ot David-
son College in Davidson, North Carolina, USA.
Birds were acclimated to these feeders for
1 month prior to the beginning of the experiment.
Two feeders were positioned 1.5 m apart, 15 m
from cover, in the middle of a rectangular
clearing. The other two feeders were also 1.5 m
apart, and placed at the edge of the clearing 1.5 m
from den.se vegetation.

Northern Cardinals were by tar the most
common visitor to these feeders. Their large body
size and aggressive behavior also minimized the
potentially confounding effects of interspecific
competition (Alatalo et al. 1987). The extent to
which Accipiter hawks concentrate their hunting
efforts near bird feeders is disputed (Roth and

Lima 2007), but Cooper’s Hawks are present in
Davidson year-round and have been ob.served
hunting at our feeders.

In our first experiment, adapted from Mumme
(2003), we filled one feeder in each pair with
black-oil sunflower seed and the other with ACE®
brand wild bird seed mix (consisting primarily of
millet, but also with milo and black-oil sunflower
seed). We allowed birds several days to acclimate
to the feeder contents before beginning observa-
tions. We recorded the number of feeding visits
by cardinals at each feeder for four 1-hr
observation periods per day. Observations were
made from a chair ~ 50 m from the feeders. All
observations occurred between 1000 and 1800 hrs
EDT. A feeding visit was one in which the bird
was observed to land and consume at least one
food item. We did not quantify the amount of food
consumed or the length of each feeding visit.
Cardinals were not individually marked and it is
probable that we re-sampled many of the same
individuals, violating strict independence. How-
ever, several factors convinced us to pursue the
study. First, the nutritional state of any bird (as
well as the danger associated with visiting a
particular feeder) can change over the course of a
day. Second, the number of cardinals in the area
was large (up to 10 were observed in the vicinity
of the feeders at any given time). Third, capturing
and color-banding all birds present prior to
conducting this undergraduate experiment was
not feasible.

We recorded feeding visits on 12 and 13
September 2007. We switched the food type in
each feeder pair at 1 800 hrs on 1 3 September to
control for any locality or feeder preferences
within the pair. We allowed birds to acclimate to
the change in seed location on 14 September and
resumed data collection on 15 and 16 September.

We again changed the contents of the feeders at
1800 hrs on 16 September, placing the favored
seed type from 12 to 16 September into the two
feeders in the center of the clearing and the other
seed type into feeders near cover. We allowed
birds to acclimate to the new locations on 17
September and resumed data collection on 18
September, continuing through 21 September. We
continued to use two feeders at both locations to
minimize the chance that subordinate individuals
would be forced to visit undesirable feeders.

We repeated this trial on 30-31 January 2008
with preferred seed in the open and the non-
preferred seed near cover. We collected data from

170

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. I, March 2010

(/)

>

c

g

■c

o

Q.

o

Q_

FIG. 1. Northern Cardinal visitation in winter 2008 to exposed feeders with preferred food (sunflower seed) vs. visits to
less preferred food near cover on days without (/? = 668 visits) and with (/? = 455 visits) Cooper’s Hawk call playback
(** = P < 0.001).

this trial in three 1-hr sessions per day (1 hr
between 0800 and 1 100 hrs EST, another between
1 100 and 1400 hrs EST, and again between 1400
and 1700 hrs EST). We played calls of a Cooper’s
Hawk (Neville 2004) on 1 and 2 Eebruary at 0800,
1000, 1200, 1400, and 1600 hrs. We recorded
feeder visits by cardinals on these days as we had
on 30-3 1 January. Playbacks consisted of a 5-sec
call played 12 times in succession. Calls were
played at a natural volume from a hidden position
80 m from the feeders. We did not have an
explicit control playback because of the indirect
nature of the hawk playback and the richness of
the acoustic environment (corvids, dogs, car
horns, trucks).

Statistical Analyses. — ^We used JMP 7.0 (SAS
2007) to analyze our data. We performed X- tests
to compare relative use of feeders containing
different seeds, feeders in different locations, and
feeders in the presence and absence of Cooper’s
Hawk calls. We used Wilcoxon/Kruskal-Wallis
tests to assess absolute changes in feeder visita-
tion rates resulting from playback of hawk calls.
Data are pre.sented as means ± SE.

RESULTS

Northern Cardinals visited feeders containing
sunlJower seed significantly more often than those
with mixed seed (Y"/ = 22.2, P < 0.001) when
both foods were provided at both locations. Sixty-

four percent of 532 visits were to feeders
containing sunflower seeds. Cardinals visited
feeders near cover significantly more often than
those far from cover (X‘/ = 36.6, P < 0.001).
Sixty-eight percent of 532 visits were to feeders
near cover.

We provided cardinals with a choice between
their prefeired food type (sunflower seed) and
preferred location (near cover) in the next phase
of the experiment. Cardinals visited the exposed
sunflower seed significantly more often than
mixed seed near cover {X- / = 34.3, P < 0.001).
Sixty-seven percent of 571 visits were to the
exposed feeders containing sunflower seed.

We repeated the trial to ensure the cardinals
continued to prefer higher quality seed in a less
safe location, even in winter. Cardinals visited the
exposed feeders significantly more often (X- / =
12.4, P < 0.001), as we found in the fall. Sixty
percent of 668 visits were to the exposed feeders
containing sunflower seed (Eig. 1). We then
moved on to our playback experiment. The
decrease in the mean visitation rate of cardinals
as a result of exposure to hawk calls (47.7 ± 8.6/
hr to 37.8 ± 4.9/hr) was not significant (Wil-
coxon/Kruskal-Wallis Z = -0.54, P = 0.59). The
increase in use of feeders near cover (38.6 ± 5.3/
hr to 48.3 ± 5.3/hr) was not significant (Wil-
coxon/Kruskal-Wallis Z = —1.14, P = 0.25), but
the decrease in the visitation rate to exposed

SHORT COMMUNICATIONS

171

feeders (56.9 ± 16.3/hr to 27.3 ± 5.6/hr)
approached significance (Wilcoxon/Kruskal-
Wallis Z = — 1.79, P = 0.074). Overall, exposure
to Cooper's Hawk calls resulted in a dramatic and
significant shift in feeder use (X-/ = 60.0, P <
0.001; Fig. 1). Cardinals, instead of preferring the
higher quality seed in the exposed location, as
they had in the first 2 days, had a significant
preference for the lower quality seed in the safer
location on playback days (X"/ = 18.1, f <
0.001). Only 36% of 454 visits were to exposed
feeders containing sunflower seed (Fig. 1).

DISCUSSION

A basic tenet of foraging theory is that
organisms seek to maximize energy acquisition
(Krebs and Kacelnik 1991, Cuthill and Houston
1997). One means by which this can be achieved
is by foraging selectively, concentrating on higher
quality foods when opportunities present them-
selves. We presented cardinals with black-oil
sunflower seeds (with shells) versus commercial
wild bird seed mix in our experiment. We were
not surprised that, when presented with both foods
at both locations, cardinals visited feeders with
sunflower seed significantly more often; relative
to commercial wild bird seed, black-oil sunflower
seed has a high energetic content and nutritional
value (Willson 1971, Willson and Harmeson
1973). It is possible that cardinals varied the
length of their stay (and quantity of seed ingested)
at feeders, but our results suggest that visitation
rates alone provide a meaningful comparison.

Foraging organisms are also under pressure to
minimize risk of death from predation (Lima and
Dill 1990, Lima 1998). Cardinals should be
sensitive to risks associated with different feeding
locations and, all things being equal, should spend
more time in safer foraging locations (those near
cover) and less time in dangerous locations (those
in the open). We were not surprised to find that,
when presented with both foods at both locations,
cardinals visited feeders near cover significantly
more often.

Optimizing foraging requires making decisions
when all things are not equal (Sih 1980, Lima and
Dill 1990, Cuthill and Houston 1997). In the first
trial, the strength of the preference for sunflower
seed was ostensibly similar to the preference for
foraging near cover. However, when cardinals had
to choose between their prefen'ed food and their
preferred location, they showed a clear bias
toward higher quality food (sunflower seed) —

even though it was available only in the dangerous
(exposed) location. This suggests that cardinals
can recognize a risk of feeding in exposed
locations, but willingly incur that risk to obtain
high quality food.

Our last experiment demonstrated the cardinals
we observed were not applying a single assess-
ment of risk to foraging activities. Most cardinals
under normal conditions perceived a threat in
foraging at the exposed feeders, but not suffi-
ciently great to forego sunflower seeds when
available only at exposed feeders. However, upon
hearing hawk calls in the vicinity throughout the
day, cardinals adjusted their perception of risk.
Neither the decline in the visitation rate to all
feeders nor the increase in the visitation rate to
feeders near cover was significant, but the decline
in the visitation rate to exposed feeders ap-
proached significance. We observed a significant
change in the relative use of exposed and
protected feeders. Most cardinals were unwilling
to risk predation to obtain higher quality food
when exposed to the calls of a Cooper’s Hawk.
Our experiment lacked an explicit control play-
back, but the strength of the foraging response
suggests the trends we observed were real. This
plasticity is presumably adaptive, allowing indi-
viduals to maximize their fitness (Lima and
Bednikoff 1999).

Birds recognize, assess, and respond to danger
using a variety of cues including visual, auditoiT,
and even olfactory (Roth et al. 2008). The
cardinals we tested altered their foraging strategy
dramatically to the playback of hawk calls, but
these results are not always obtained. Schmidt et
al. (2008) observed that eastern chipmunks
{Tamias striatus) altered their foraging in re-
sponse to the alarm calls of Tufted Titmice
(Baeolophus hicolor), but not to the calls of an
actual predator, the Broad-winged Hawk (Buteo
platypterus). Schmidt et al. (2008) hypothesized
the hawk call provided information upon which
chipmunks could make a foraging decision,
whereas the titmouse call simply alerted them to
an eminent danger for which additional vigilance
and information would be required. Similarly,
Van der Veen (2002) found that Yellowhammers
(Emberizu citronella) exposed to an accipter
model resumed foraging sooner than individuals
with less reliable information (those which did not
see the model but did hear the alarm calls of those
exposed to the model). Our experiment is the first
to demonstrate that Northern Cardinals weigh the

172

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 1, March 2010

importance of food quality differently in response
to auditory cues of the presence of a particularly
dangerous predator.

The plasticity we observed suggests the out-
comes obtained were particular to the circum-
stances of our experiments. Our results may have
been different had we varied certain aspects of our
design (e.g., types of food used in the feeders).
Moreover, had our exposed feeders been placed in
a large field 50 m from the forest edge, rather than
in the center of a small meadow, the perceived
risk involved in visiting them may have differed.
Our findings suggest that Northern Cardinals in
our population tended to respond in an adaptive
way to variation in food quality and differing
levels of predation risk.

ACKNOWLEDGMENTS

We thank the students of BIO 1 12 (David Baker, Austin
Bond, Imani Bryan, Kyri Bye-Nagel, Elizabeth Cannon,
Fareed Cheema, Lauren Childs, Jesse Dimmock, Andrew
Dunn, Lauren Felkel, Francisco Fiallo, Joel Fineman,
Robert Flowers, Rieti Gengo, Sarah Griffith, Courtney
Hart, Jon Huggins, Harriet King, Scott Lester, Nina Mace,
Natasha Meyer, Shane Purvis, Allison Ruhe, Chris Ryan,
Kyle Sanders, Meena Sangar, Jackie Tan, Ellen Thomas,
Amber Townsend, Alysen Wallace, Katie Wallace) and
BIO 322 (Elizabeth Arellano, Mickey Belcher, Laura
Bergner, Lindsay Brownell, Jay Buckingham, Middleton
Chang, Esther Cline, Kara Earle, Emma Garren, Nate
Geigle, Christa Goeke, MacLean Hall, Chris Hampton,
Courtney Hart, Andrew Johnson, Alex Kim, Deanna
Lomax, Peggy McKay, Rachel Miranda, Shannon Pittman,
Erevan Rankin, Philip Ruzycki, Sam Sheline, Ellen
Thomas, Brent Winship) for their help in conducting this
experiment. The manuscript was improved by the com-
ments of Howell Burke. Esther Cline, Austin Mercadante,
Ron Mumme, C. E. Braun, and two anonymous reviewers.

LITERATURE CITED

Alatalo. R. V., D. Erik.s,son, L. Gustafsson, and K.
Larsson. 1987. Exploitation competition influences
the u.se of foraging sites by tits: experimental evidence.
Ecology 68:284-290.

Barnard. C. J. 1979. Interaction between House Sparrows
and Sparrowhawks. British Birds 72:569-.573.
Blumstein. D. T.. L. Cooley, J. Winternitz, and J. C.
Daniel. 2008. Do yellow-bellied marmots respond to
predator vocalizations? Behavioral Ecology and So-
ciobiology 62:457-468.

Caraco. T., S. Martindale, and H. R. Pulliam. 1980.
Avian time budgets and distance to cover. Auk
97:872-875.

CUTHILL, I. C. AND A. 1. HOUSTON. 1997. Managing time
and energy. Pages 97-120 in Behavioural ecology: an
evolutionary approach (J. R. Krebs and N. B. Davies,

Editors). Fourth Edition. Blackwell Scientific Publi-
cations, Oxford, United Kingdom.

Dill, L. M. and A. H. G. Fraser. 1984. Risk of predation
and the feeding behavior of juvenile coho salmon
{Oncorhynchiis kisiitch). Behavioral Ecology and
Sociobiology 16:65-71.

Fenn, M. G. P. and D. W. MacDonald. 1995. U.se of
middens by red foxes: risk reverses rhythms of rats.
Journal of Mammalogy 76:130-136.

Krebs, J. R. and A. Kacelnik. 1991. Decision-making.
Pages 105-136 in Behavioural ecology: an evolution-
ary approach (J. R. Krebs and N. B. Davies, Editors).
Fourth Edition. Blackwell Scientific Publications,
Oxford, United Kingdom.

Lima, S. L. 1985. Maximizing feeding efficiency and
minimizing time exposed to predators: a trade-off in
the Black-capped Chickadee. Oecologia 66:60-67.

Lima, S. L. 1998. Stress and decision-making under the risk
of predation: recent developments from behavioral,
reproductive, and ecological perspectives. Advances in
the Study of Behavior 27:215-290.

Lima, S. L. and P. A. Bednikoff. 1999. Temporal variation
in danger drives antipredator behavior: the predation
risk allocation hypothesis. American Naturalist
153:649-659.

Lima, S. L. and L. M. Dill. 1990. Behavioral decisions
made under the risk of predation: a review and
prospectus. Canadian Journal of Zoology 68:619-640.

Lima, S. L. and T. J. Valone. 1986. IntJuence of predation
risk on diet selection: a simple example in the grey
squirrel. Animal Behaviour 33:155-165.

Lucas, J. R., R. D. Howard, and J. G. Palmer. 1996.
Callers and satellites: chorus behaviour in anurans as a
stochastic dynamic game. Animal Behaviour 51:501-
518.

McLean, I. G. and R. Rhodes. 1991. Enemy recognition
and response in birds. Current Ornithology 8: 173-21 1.

Morse, D. H. 1973. Interaction between tit flocks and
Sparrowhawks Accipiter nisiis. Ibis 1 15:591-593.

Mumme, R. L. 2003. Economic decisions and foraging
trade-offs in chickadees. Pages 231-237 in Exploring
animal behavior in laboratory and field (B. J. Ploger
and K. Yasukawa, Editors). Academic Press, New
York, USA.

Neville, J. 2004. Bird songs — Western Boreal Forest.
Neville Recording. Salt Spring Island. British Colum-
bia. Canada.

Peake, T. M. 2005. Eavesdropping in communication
networks. Pages 13-37 in Animal communication
networks (P. McGregor, Editor). Cambridge Univer-
sity Press, Cambridge, United Kingdom.

Pulliam, H. R. and G. S. Mills. 1977. The use of space" by
wintering sparrows. Ecology 58:1393-1399.

Roth, T. C. and S. L. Lima. 2007. Use of predictable prey
hotspots by an avian predator: purpo.seful randomness?
American Naturalist 169:264-273.

Roth, T. C., J. G. Cox, and S. L. Lima. 2008. Can foraging
birds asse,ss predation risk by scent? Animal Behav-
iour 76:2021-2027.

SAS. 2007. JMP 7.0. SAS Institute Inc., Cary, North
Carolina, USA.

SHORT COMMUNICATIONS

173

Schmidt, K. A., E. Lee, R. S. Ostfeld, and K. Sieving.
2008. Eastern chipmunks increase their perception ot
predation risk in response to titmouse alarm calls.
Behavioral Ecology 19:759-763.

SiH, A. 1980. Optimal behavior: can foragers balance two
conflicting demands? Science 210:1041-1043.
Valone, T. J. and S. L. Lima. 1987. Carrying food items to
cover for consumption: the behavior of ten bird species
feeding under the risk of predation. Oecologia 7 1 :286-294.

Van der Veen, 1. T. 2002. Seeing is believing: infor-
mation about predators influences Yellowhammer
behavior. Behavioral Ecology and Sociobiology 51:
466-47 1 .

Willson, M. F. 1971. Seed selection in .some North
American finches. Condor 73:415-429.

Willson, M. E. and J. C. Harmeson. 1973. Seed
preferences and digestive efficiency of Cardinals and
Song Sparrows. Condor 75:225-234.

The Wilson Journal of Ornithology 122( 1 ): 173-177, 2010

Observations and Predation of a Coral-billed Ground Cuckoo
{Carpococcyx renauldi) Nest in Northeastern Thailand

Korakoch Pobprasert' and Andrew J. Pierce

ABSTRACT. — We found the first documented wild
nest of a Coral-billed Ground Cuckoo {Carpococcyx
renauldi) at Khao Yai National Park, Thailand in June
2007. The large stick nest was monitored for 24 days
including 1095 hrs of video footage; it contained two
eggs and was in dense vegetation 4.85 m above the
ground. Nest attentiveness of adults was almost constant
with both birds taking turns to incubate or brood. Food
delivered to nestlings included lizards, nestlings, a
snake, frogs, earthworms, and other invertebrates. Nest
defense was observed against several known nest
predators but the nest ultimately failed due to predation
by a pig-tailed macaque (Macaca nemistrina). These
observations provide insight into the breeding ecology
of the only two congeners, C. viridis and C. radiceus,
for which little is known of their ecology and both are
endangered. Received 27 March 2009. Accepted 18
September 2009.

There is an extreme paucity of studies on the
breeding ecology of tropical Asian birds (Sodhi
and Brook 2006) and they should not be assumed
to have the same life histories as the more widely
studied neotropical birds. It is essential that
studies are undertaken for the long-term manage-
ment of bird communities as forest cover and
forest quality continues to decline in Asia (FAO
2009).

There are 1 1 species of non-parasitic cuckoos
on mainland Southeast Asia including: six mal-
kohas (Phaenicophaeus), four coucals {Centro-

‘ Conservation Ecology Program, King Mongkut’s
University of Technology Thonburi, 83 Moo 8 Thakham,
Bangkhuntien, Bangkok 10150, Thailand.

^Corresponding author; e-mail: andyp67@gmail.com

pus), and one ground cuckoo (Carpococcyx).
There are two other ground cuckoos which inhabit
the forests of Sumatra and Borneo. The Sumatran
Ground Cuckoo (Carpococcyx viridis) has only
recently been rediscovered (Zetra et al. 2002) and
nothing is known of its breeding ecology; it is
Critically Endangered (lUCN 2008). The Bornean
Ground Cuckoo (C. radiceus) is Near Threatened
(lUCN 2008) and what is known of its breeding
ecology is vague and ambiguous derived from a
few captive birds and questionable museum
specimens (Long and Collar 2002). Most wide-
spread of the three species is the Coral-billed
Ground Cuckoo (C. renauldi) (CBGC), that
occurs in Thailand, Cambodia, Laos, and Tonkin
and Annam (Vietnam). It is an uncommon
resident of broad-leaved evergreen forest and
secondary growth up to 1,000 m asl in Thailand
(Robson 2000); its conservation status is of Least
Concern (lUCN 2008). The only documented
nesting attempts are of captive pairs. A pair at
Metro Toronto Zoo produced 13 fledglings and
clutches of up to five eggs were laid between
early April and mid-August from 1976 to 1979
(Atkinson 1982). Robilleret al. (1992) and Rinke
( 1999) both document nests of captive pairs with
clutches of 3-4 eggs and nestlings reared by
hand.

We present details, obtained through video
recordings, of the first documented wild nest of a
Coral-billed Ground Cuckoo, including its preda-
tion by pig-tailed macaques (Macaca nemistrina
leonina) in Khao Yai National Park, northeastern
Thailand in 2007.

174

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. I. March 2010

METHODS

We initiated a 3-year study of nest predation in
2006 on birds in the Mo-Singto area of Khao Yai
National Park (14° 26' N, 101° 22' E), Nakhon
Nayok Province, Thailand. Small waterproof
Eujiko™ cameras (65 X 65 X 85 mm) with
infrared diodes for 24-hr surveillance were
deployed 3-5 m from nests of ground and
understory nesting birds. The cameras were
attached by cables to a digital video recorder
(DVR) placed 20-25 m from the nest whereby the
power source and DVR could be changed without
disturbing the nesting birds (Pierce and Pobprasert
2007). A small microphone was also fitted within
the surrounding case of the camera. Data from the
DVR were stored on DVDs for later analysis.
Cameras have been shown to have little effect on
nesting success when placed off the ground (King
et al. 2001, Staller et al. 2005, Pierce and
Pobprasert 2007). Measurements of the nest were
taken after it had failed. Male Coral-billed Ground
Cuckoos are slightly larger than females, but they
are monomorphic in plumage and males and
females are indistinguishable in the field.

RESULTS

An adult Coral-billed Ground Cuckoo was
observed to Ily down from a tree within a dense
tangle of woody climbers on 13 June 2007. Closer
inspection on 17 June revealed a nest containing
two white eggs. A camera was placed in an
adjacent tree ~4 m from, and slightly below, the
nest which still contained two eggs at 1400 hrs on
18 June. The sitting bird flushed quickly and
silently as we approached the nest area on each of
these visits despite our attempts to avoid disturb-
ing it. An adult returned to incubate within an
hour of us leaving the area after initial setting of
the camera. Apart from a mechanical failure for
8 hrs on I day during incubation, the cameras
recorded continuously until the nest was predated
on the morning of 7 July. The adults were only
disturbed once following placement of the camera
when a person inadvertently pas.sed close to the
nest.

Ne.st Placement. — ^The nest was in seasonally
wet evergreen forest at 725 m asl. The canopy
was generally 20-30 m high with emergent trees
up to —60 m, but the area within 20 m of the nest
included an old tree fall with most trees less than
10 m in height and covered with masses of woody
climbers. The nest was an untidy pile of dead

twigs in the shape of a shallow cup similar to that
of a large raptor. It was lined with a few fresh
small twigs that still had green leaves. It was
placed at the base of a branch forking from the
leaning trunk of a tree with a diameter at breast
height of ~10cm, and further supported by
woody climbers. The rim of the nest was 4.85 m
above ground and vegetation, in the form of small
branches and woody climbers, extended 50-
60 cm above the nest. The outer diameter of the
nest was 33 cm, and the total height was —25 cm.

Parental Care and Food Provisioning. — Both
adults shared incubation duties and one adult was
on or close to the nest at all times with few
exceptions. There may have been some contact
calls from the returning bird as the sitting bird
became aware of the incoming bird shortly before
it arrived at the nest but, if so, they were too soft
to be picked up by the microphones. Both adults
usually brought a small twig complete with fresh
green leaves and placed it in the nest during the
egg stage before commencing incubation.

The adults changed brooding only twice per
day with changeover times quite variable; the first
between 0921 and 1338 hrs and the second
between 1612 and 1754 hrs. Changeovers typical-
ly took <30 sec with two of exceptionally longer
duration of 2 min 32 sec and 8 min 32 sec.
Excluding these apparent anomalies, changeovers,
from when the sitting bird left the nest until the
other bird resumed incubation, averaged (± SD)
20 ± II sec (/; = 24). The bird incubating during
the day averaged (± SD) 5 hrs 36 ± 52 min {n =
12) on the nest while the bird that incubated
overnight averaged 18 hrs 43 ± 50 min (// = 12)
on the nest. We were unable to identify gender of
the adults; however, for 9 days where the data are
continuous, and one or both adults were in view at
all times, the same adult incubated over night as
on the previous night. It is likely therefore that
only one male or one female incubated at night
and the other during the day for the entire period.
Gaps in attentiveness after the eggs hatched meant
we could not ascertain the proportion of time
spent brooding by each bird during the nestling
stage.

The first egg hatched in the early morning of 1
July when the adult started pecking at something
beneath it.self and ate what was presumably
eggshell (no eggshells were removed and only
small fragments of eggshell were found in the nest
after it had failed). The second egg hatched on the
morning of 2 July when the same sequence of

SHORT COMMUNICATIONS

175

events occun-ed. Changeovers followed the same
pattern during hatching as in the incubation period
and it was not until 1 147 hrs on 2 July that the
first food item was brought to the nest. Attentive-
ness at the nest continued to be almost constant
throughout the nestling period although visitation
rates increased to provision the nestlings. There
were two visits with food on 2 July followed by
16, 24. 33, and 29 visits on the sub.sequent 4 days
and a further seven visits in the morning of 7 July.
The incoming bird during the nestling stage
usually perched on the rim of the nest for a few
seconds before the sitting bird stood up and left.
The incoming bird would then hop into the nest
and feed the nestlings before brooding them.

Nestling Diet. — Only a limited number of prey
items could be positively identified. These
included lizards (3), nestlings of small birds (2),
frogs (2), a blind snake (Typhlops or Rampho-
typhlops), Orthoptera (2), and earthwomis (2).
Prey was too large for the nestlings to swallow on
two occasions and it was either taken away or
eaten by the adult. Other prey items tentatively
identified were worms, caterpillars, arthropods,
and slugs or snails but many of the small food
items appeared to be regurgitated or were out of
view of the camera.

Nest Defense. — Variable squirrels {Calloscinrus
finlaysonii) and northern tree shrews (Tupaia
helangeri) visited the nest nine times. The adults
usually turned to face these relatively small
predators with crown feathers depressed and tail
raised and the intruder quickly withdrew. Two
visits to the nest by small snakes (—50-100 cm
long; probably Boiga spp.) were the only potential
nocturnal nest predators, and both soon retreated
rapidly after reaching the sitting bird which did
not react. Pig-tailed macaques passed close by
twice on the afternoon of 4 July. The adult
adopted a threat posture with this no doubt more
serious threat to the nest and, on these occasions,
the macaques left the nest alone. A troop of
macaques could be heard passing the nest at
0854 hrs on 7 July and the sitting CBGC became
alert raising its body and spreading its tail. At
exactly 0858 hrs it raised its crown feathers and
turned to face a macaque with bill open, wings
apart and tail spread over its back. This macaque
dropped down quickly to the nest 17 sec later and
the CGBC flushed to nearby branches. It stayed
within —I m of the nest maintaining its threat
posture and making short darting movements at
the macaque. However, the macaque was not

deterred and grabbed the nestlings one at a time,
put them quickly into its mouth and left. The
whole episode took 3 sec from when the CBGC
left the nest until the macaque took the nestlings
and departed. The CBGC returned to the rim of
the nest; when another (smaller) macaque ap-
proached, <1 min after the first attack, it flew at
the macaque causing the latter to retreat; the
CBGC flew on away from the nest. Two small
macaques came to investigate the nest but left on
finding it empty. The CBGC returned cautiously
at 0918 hrs from below the nest and stood on the
nest for 6 min before leaving. A CBGC entered
the nest carrying a large insect at 0957 hrs but left
again after nearly 1 min. After another visit at
1014 hrs with a different insect there were no
further visits by 1500 hrs the following day when
the camera was removed.

DISCUSSION

Coral-billed Ground Cuckoos have been re-
corded nesting in captivity but ours is the first
observation of a wild nest and provides more
appropriate information about their nesting be-
havior. The CBGC is under no immediate threat
but there is an almost complete lack of informa-
tion on the breeding ecology of the globally
threatened congeners, Sumatran and Bornean
ground cuckoos, the latter of which is known to
be similar to CBGC (Long and Collar 2002).

The CBGC nests at the Toronto Zoo were built
by both adults and were comparable in dimen-
sions to the nest we describe (Atkinson 1982) but.
otherwise quantifying these nests is not appropri-
ate due to necessary restrictions of captive birds.
We found no fecal sacs in or near the nest after it
failed and it is possible the adults ate them;
Robiller et al. (1992) observed the nestlings
defecated on the rim of their nest, but they were
being hand fed with no adult birds present.
Atkinson (1982:170) states “an incubation period
of 28 days seems the normal condition”.
However, the data are ambiguous and could be
interpreted as being anywhere between 17 to 28
days while Payne (2005) reports the eggs hatch in
18-19 days. Eggs in our nest hatched within 15
days of seeing the full clutch and was probably in
the early stages of incubation when found. The
clutch of two eggs is at the lower end of clutch
sizes reported from captive birds. Atkinson ( 1982)
details 10 nests with clutches of 1-5 eggs (mean
3.1, median 4). Rinke (1999) details two nests
with four eggs and Robiller et al. (1992) records

176

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. I, March 2010

two nests, each with three eggs but also states the
clutch size to be 3-4 eggs.

Ground cuckoos are notoriously difficult to
observe in the wild and, apart from a few
observations of birds coming to food waste at
restaurants adjacent to the forest (Payne 2005; AP,
pers. obs.), there are few published data on CBGC
diet or behavior. Delacour and Jabouille (1931)
reported small birds and mammals in their diet.
Based on the stomach contents of a recently dead
adult, found within a kilometer of this nest in
2006, they feed on a wide range of invertebrates;
the stomach included the remains of several
hundred individual prey items, especially Orthop-
tera and Isoptera but also Coleoptera, Lepidop-
tera, Hymenoptera, and a Chilapoda (centipede)
(John Milne, in prep.) We did not definitively
record any of these but it is possible they were
among the many unidentified prey items fed to the
nestlings. There is no evidence that they (or
Bornean Ground Cuckoos [Long and Collar
2002]) eat fruit despite the generic name Carpo-
coccy.x meaning fruit-cuckoo.

Nest success in tropical forests is low; ~8-30%
for most species with predation being the main
cause of nest failure (Martin 1995, Robinson et al.
2000). Our research using video surveillance at
nests in Khao Yai National Park (Pierce and
Pobprasert 2007, and unpubl. data) identified the
main nest predators as pig-tailed macaques,
rodents (Muridae and Sciuridae), tree-shrews
(Tupoia), birds, and snakes. Several of these were
recorded visiting the CBGC nest. We recorded
similar size snakes disturbing sitting Puff-throated
Bulbuls (Alophoxius palUdiis) and trogons (Har-
pactes spp.) and, being able to take the eggs or
nestlings (Pierce and Popbrasert 2007). Rinke
( 1999) shows a CBGC on a nest in captivity in a
defense posture similar that recorded at our nest.
The CBGC in our study was able to deter
macaques on two occasions by adopting this
stance, although ultimately unsuccessful against a
large macaque. A high level of nest attentiveness
has been recorded for other non-parasitic cuckoos
in the region (Payne 2005), and it has been
suggested this may have developed in response to
high predation pressures observed in the tropics
(Auer et al. 2007). The con.stant vigilance of the
adult CBGC at their nest no doubt stopped many
potential predators from taking eggs or nestlings.
However, adults were highly .sensitive to distur-
bance, Hushing from the nest and not returning for
an hour or more afterwards, even when we

approached quietly knowing the nest location.
These results highlight the potential increased
predation risk caused by human disturbance of
nesting birds that might otherwise be able to
defend their nest.

ACKNOWLEDGMENTS

We thank Will Duckworth. Kim McConkey, and two
anonymous reviewers for valuable comments on an earlier
draft. John Milne kindly provided details of the CBGC
stomach contents and Matthias Fehlow found and translated
the two German references. We are grateful to Khao Yai
National Park and the Department of National Parks,
Wildlife and Plants Conservation for facilitating our
research. We thank our fellow researchers for assistance
and support. This project was supported by King Mongkut’s
University of Technology Thonburi, Bangkok with grant
BID-BD-RD-48-02 KMUTT from the National Research
Council of Thailand and the TRF/BIOTEC Special Program
for Biodiversity Research and Training Grant BRT 350002.

LITERATURE CITED

Atkinson, R. W. 1982. Breeding the Renauld’s or Coral-
billed Ground Cuckoo Carpococcyx renauldi at the
Metro Toronto Zoo. International Zoo Yearbook
22:168-171.

Auer, S. K.. R. D. Bassar, J. J. Fonttaine, and T. E.
Martin. 2007. Breeding biology of passerines in a
subtropical montane forest in northwestern Argentina.
Condor 109:321-333.

Delacour. J. and P. J.abouille. 1931. Les oiseaux de
rindochine fran^aise. 14. Exposition Coloniale
Internationale. Paris, France.

Food and Agriculture Organization of the United
Nations (FAO). 2007. State of the world's forests
2009. FAO. Rome, Italy, www.fao.org (accessed 8
September 2009).

lUCN. 2008. The lUCN Red List of threatened species.
http://www.iucnredlist.org/ (accessed 8 September
2009).

King, D. 1., R. M. DeGraaf, P. J. Champlin, and T. B.
Champlin. 2001. A new method for wireless video
monitoring of bird nests. Wildlife Society Bulletin 29:
349-353. ^

Long, A. J. and N. J. Collar. 2002. Distribution, status
and natural history of the Bornean Ground Cuckoo
Carpococcyx radiatiis. Forktail 18:101-119.

Martin, T. E. 1995. Avian life history evolution in relation
to nest sites, nest predation, and food. Ecological
Monographs 65:101-127.

Payne, R. B. 2005. The cuckoos. Oxford University Press,
Oxford. United Kingdom.

Pierce. A. J. and K. Pobprasert. 2007. A portable system
for continuous monitoring of bird nests using digital
video recorders. Journal of Field Ornithology 78:322-
328.

Rinke, D. 1999. Zucht des Renauldkuckticks (Carpococcyx
renaiddi) im Vogelpark. Walsrode 22:212-214.
Robiller, F., H. Michi, and M. Michi. 1992. Uber den

SHORT COMMUNICATIONS

177

Renauld kuckuck (Carpococcyx rcnaiiUli Oustalet,
1896) und seine Zucht in der Forschnngsstalion Ornis
Mallorca. Tropische Vogel 13:92-97.

Robinson, W. D., T. R. Robinson, S. K. Robinson, and J.
D. Brawn. 2000. Nesting success of understory forest
birds in central Panama. Journal of Avian Biology
31:31-164.

Robson, C. 2000. A field guide to the birds of Thailand and
South-East Asia. Asia Book Co. Ltd., Bangkok, Thailand.

SoDHi. N. S. AND B. W. Brook. 2006. Southeast Asian

biodiversity in crisis, Cambridge University Press,
Cambridge, United Kingdom.

Staller, E. L., W. E. Palmer, .1. P. Carroll, R. P.
Thornton, and D. C. Sisson. 2005. Identifying
predators at Northern Bobwhite nests. Journal ot
Wildlife Management 69:124-132.

Zetra, B., A. Rafiastanto, W. M. Rombang, and C. R.
Trainor. 2002. Rediscovery of the critically endan-
gered Sumatran Ground Cuckoo Carpococcyx viridis.
Forktail 18:63-65.

The Wilson Journal of Ornithology 122( 1 ): 177-1 80, 2010

Diet of Chinese Grouse {Tetrastes sewerzowi) during Preincubation

Wang Jie,' - Yang Chen,' - Lu Nan,' - Fang Yun,' and Sun Yue-Hua'-^

ABSTRACT. — The endemic Chinese Grouse (Tetra-
stes sewerzowi) inhabits subalpine coniferous forests.
Little is known about its diet in the breeding season,
especially while birds feed on the ground. Analysis of
crop contents indicated that willow (Salix spp.) was the
primary food of males (>98% wet weight, n — 4),
whereas Dragon spruce (Picea asperata) seeds and
willow were the primary foods of females (both >40%
wet weight, n = 2). Dragon spruce seeds, invertebrates
(mainly ants), and forbs were frequently consumed by
females, but seldom by males, possibly to meet the
nutrient constraints of egg formation. We suggest the
different diets of males and females of this monoga-
mous species may be the result of females allocating
more time to searching for scarce, nutritious food,
whereas males spend more time in vigilance behavior.
Received 2 April 2009. Accepted 5 September 2009.

The Chinese Grouse (Tetrastes sewerzowi),
endemic to the coniferous forests of Qinghai-
Tibet Plateau, is one of the rarest grouse species in
the world (Storch 2000) and is legally considered
an endangered species in China (Zheng and Wang
1998). The small body size of Chinese Grouse
(Sun et al. 2005) may reduce its ability to meet
nutritional requirements when feeding on plants
because gut capacity scales linearly with body
mass, whereas mass-specific nutritional require-
ments increase with decreasing body mass (Dem-

' Key Laboratory of Animal Ecology and Conservation
Biology, Institute of Zoology, Chinese Academy ol
Sciences, Beijing 100101, PR China.

’Graduate School of Chinese Academy of Sciences,
Beijing 100049, PR China.

^Corresponding author; e-mail: sunyh@ioz.ac.cn

ment and Van Soest 1985). Small body size also
precludes storage of energy and nutrients in the
body for egg formation (Swenson et al. 1994).
Chinese Grouse have the greatest nutritional
investment in clutches among tetraonids (Sun et
al. 2005) and longer egg laying intervals (48 hrs)
than other grouse (Sun et al. 2002). Thus, nutrient
constraints possibly exist for egg formation for
this species and female Chinese Grouse may rely
heavily on nutritionally rich food during the
prelaying and laying period to gain mass and
produce eggs when food may not yet be plentiful
(Nager 2006).

The diet of Chinese Grouse during the breeding
season is poorly known and difficult to clarity by
observation because of dense vegetation and birds
not habituated to observers. We present informa-
tion on the diet of this monogamous grouse from
analysis of crop contents to help understand its
foraging strategy to meet nutrient constraints of
egg formation.

METHODS

Study Area. — Our study area was in the
Lianhuashan Natural Reserve, southern Gansu
Province, China (34 57' N, 103° 46' E). Eorests
occur on north-facing, and some northeast and
northwest-facing slopes at elevations of 2,600-
3,200 m; only shrubs and grasses grow on south-
facing slopes. The main cover types are: (1)
coniferous forest dominated by Dragon spruce
(Picea asperata) and Earges fir (Abies fargesii).
(2) coniferous-deciduous forest, including spruce,
fir, Himalayan birch (Betula utilis), and willow

178

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 1. March 2010

TABLE 1. Loods in crops of six Chinese Grouse at Lianhuashan, Gansu, China (Collection dates; Male # 1, 25 Apr;
Male # 2, 9 May; Male # 3, 3 May; Male # 4, 17 May; Lemale #1,16 Apr; Female # 2, 14 May).

Portion of wet weight (%)

Food

Types of food

Male #1

Male #2

Male #3

Male #4

Female #1

Female #2

Invertebrate.s (ant, midge)“

0

0

0

0

3.0

9.9

Willow

Twigs, buds, catkins, leaves

100

100

98.3

100

56.0

43.3

Dragon spruce

Seeds

0

0

1.7

0

40.8

43.3

Lorbs^

Leaves

0

0

0

0

0.2

3.5

Total wet mass, g

2.3

0.3

3.2

0.3

4.9

1.5

^ Most invertebrates were ants (Hymenoptera; Formicidae); only one midge (Diptera) occurred in the crop of Female # 2.
Included Fraf>aria orienuilis, Gcttm atcppicum. and Triosteiim piiuuilifulum.

(Salix spp.), and (3) shrublands, including willow,
sea buckthorn {Hippophae rhamnoides), and
barberry (Berheris spp.). The study area has been
described by Sun et al. (2003).

Field and Analysis Procedures. — We defined
the preincubation period as 20 April-30 May,
which represents 2 weeks before egg laying to
when the last egg is laid. We obtained crops of
four males and two females that died from
predation or during capture while attaching radio
transmitters. One female which died on 16 April
was also included in our analysis. Identification
of gender of grouse carcasses was based on PCR
amplification of CHD genes using P2/P8 primers
(Griffiths et al. 1998) and sexl/sex2 primers
(Wang and Zhang 2009), or based on the
presence of the black chin patch for males
(Bergmann et al. 1996). Age (adult or yearling)
was not assigned, as there is no known reliable
method (Swenson et al. 1996). Crops were
removed and contents were identified, sorted,
and weighed to the nearest 0.001 g. Plants were
identified to species and invertebrates were
identified to Order.

We sampled 1 3 plots (40 X 40 cnr) where
grouse pecked and counted the surface seeds to
assess their availability. Seeds from cones were
collected, dried to constant weight at 50° C and
homogenized to measure nitrogen (Foss Kjeltec™
2100, Hilleroed, Denmark), lipid (Soxhlet), fiber
(ANKOM 200/220 Fiber Analyzer, Fairport, NY,
USA), tannin (Phosphomolybdium Tungstic Acid
colorimetric method, China Entry/Exit Inspection
and Quarantine Association 1999), calcium and
phosphorus (AOAC 1995), ash (combustion at
550 C), and energy (PARR 1281 bomb calorim-
eter, Moline, IE, USA). SPSS 13.0 (SPSS Inc.,
Chicago. IE. USA) was used for all statistical
analysis. Data are expressed as x ± SE.

RESULTS

Willow, including catkins, leaves, twigs, and
buds occun'ed in all crops and comprised 99.6 ±
0.4% and 49.7 ± 6.4% of the wet biomass for
males (/? = 4) and females (n = 2), respectively
(Table 1). Dragon spruce seeds were consumed
intensively by females but seldom by males (42. 1
± 1.3 vs. 0-1.7% of the wet biomass). Inverte-
brates (ants and midges) and forbs (strawberry
[Fragaria orientalis], yellow avens [Geum alep-
piciim], and feverwort [Triosteum pinnatifidiim])
ranged from 0.2 to 9.9% of the wet biomass in
crops of females, but were not found in crops of
males.

Earges fir seeds were not found in crops of
either males or females. Compared to Dragon
spruce seeds, Earges fir seeds occuired in low
densities on the ground (35 ± 67 vs. 296 ± 367
seeds/m", paired t test, P = 0.04, n = 13), and had
lower protein and ash content, and higher fiber,
tannin, phosphorus, and calcium content, whereas
their lipid and energy contents were similar
(Fig. 1).

DISCUSSION

Willow was the primary food of male grouse
(>98% of wet biomass), while Dragon spruce
seeds as well as willow were the primary food of
females. Females ate invertebrates and forbs,
whereas males did not. These findings indicate
the diet of female grouse differed from males
prior to incubation. This is consistent with our
observations that both males and females foraged
in willows intensively at dawn and dusk, but
females allocated more time than their mates to
foraging on the ground during the day (60.8 vs.
15.1%) (Sun 2004).

Egg formation is a demanding process in terms
of energy and nutrient requirements (Nager 2006).

SHORT COMMUNICATIONS

179

S’

O)

E

S

o

S’

TO

E

Food species

FIG. 1. Content (x ± SE) of protein, fiber, tannin, lipid, ash, phosphorous, calcium, and energy in samples of seeds of
Dragon spruce (Picea asperata) (PA) {n = 3) and Farges fir (Abies fargesii) (AF) (/? = 3) collected from the ground.

Some female grouse shift dietary patterns during
spring to select high-nutrient foods, including
invertebrates and forbs, which are high in crude
protein, calcium, and phosphorus (Swenson 1991,
Barnett and Crawford 1994, Gregg et al. 2008).
Female Chinese Grouse in our study similarly
shifted their diet from willow to selected ground
seeds, invertebrates, and forbs during the prelay-
ing period. Female grouse in smaller species with
fewer energy and nutrients reserves are more
dependent on exogenous sources (i.e., nutrition-
ally rich food) for egg production (Moss et al.
1975, Brittas 1988, Swenson et al. 1994). Green
leaves typically contain lower concentrations of
nutrients and are less digestible than animal tissue
(Robbins 1993). Butterfield and Coulson (1975)
estimated that consumption of 8% insect material
in the food of Red Grouse (Lagopiis lagopiis
scoticiis) could increa.se nitrogen and phosphorus
intake by 67 and 51%, respectively. Dragon
spruce seeds had higher protein, energy, and
lipid, and lower fiber and tannin than willow buds
(unpubl. data). Thus, Dragon spruce seeds,
invertebrates, and forbs may be important foods
for Chinese Grouse hens during preincubation.

Dragon spruce seeds rather than Farges fir seeds
were consumed by hens possibly because the
latter are less available on the ground and contain
higher tannin and lower protein contents.

Nutritionally rich foods, including inverte-
brates, forbs, and Dragon spruce seeds are not
highly available for prelaying hens, as it is still
cold (average <10° C) and some areas are
covered by snow until mid-May in the subalpine
forest. These foods are less available in years of
late snowmelt and could affect the breeding
performance of Chinese Grouse. Clutch size was
6.2 ± 0.3 (/? = 9) and number of chicks hatching
was 2.1 ± 0.9 (/; =11) in a year that snowmelt
was delayed by —10 days, whereas the corre-
sponding figures were 7.0 ± 0.3 (/? = 13) and 5.6
± 0.8 (n = 13) in a good year (unpubl. data). Our
sample size was small, but we found females
commonly ate nutritionally rich loods whereas
males seldom foraged on them. Male Chinese
Grouse in this monogamous species (Sun 2004),
spend more time (60.8 vs. 15.1%) in vigilance
behavior than females, allowing their mates more
time to search for nutritious food.

The Chinese Grouse is listed in Category 1 of

180

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 1. March 2010

nationally protected animals and for which
hunting is forbidden (Sun 2000). The crop
samples and diet analysis reported here are
valuable and add to the knowledge of foods used
by Chinese Grouse. This information is also
helpful in understanding the social behavior of
small monogamous birds.

ACKNOWLEDGMENTS

We are grateful to J. E. Swenson. S. J. Hannon, Yi Tao,
and J. F. Bendell for helpful comments, and the staff of
Lianhuashan Natural Reserve for assistance in the field. We
appreciate the constructive comments of M. A. Gregg, C. E.
Braun, and an anonymous reviewer on earlier drafts of the
manuscript. This study was supported by the National
Natural Science Foundation of China (Project 30620-
130110). Licenses to collect the birds were issued by the
Chinese Directorate for Nature Management.

LITERATURE CITED

Association of Analytical Chemists (AOAC). 1995.
Official methods of analysis. Sixteenth Edition. Associ-
ation of Analytical Chemists, Arlington, Virginia, USA.
Barnett, J. K. and J. A. Crawford. 1994. Pre-laying
nutrition of Sage Grouse hens in Oregon. Journal of
Range Management 47:114-1 18.

Bergmann, H.-H., S. Klaus, F. Muller, W. Scherzinger,
J. E. Swenson, and J. Wiesner. 1996. Die haselhuh-
ner, Bonasa honasia und B. sewerzowi. Die Neue
Brehm-Biicherei. Westarp Wissenschaften. Magdeburg,
Germany.

Brittas, R. 1988. Nutrition and reproduction of the Willow
Grouse Lagopiis lagopus in central Sweden. Ornis
Scandinavica 19:49-57.

Butterfield, J. and J. C. Coulson. 1975. Insect food of
adult Red Grouse Lagopus lagopus scoticus (Lath.).
Journal of Animal Ecology 44:601-608.

China Entry/Exit Inspection and Quarantine Associ-
ation. 1999. Cereals and feedstuff for import and
export — Method for the determination of tannin
content [SN/T0800.9-1999]. China Entry/Exit Inspec-
tion and Quarantine Association, Beijing.

Demment, M. W. and P. j. van Soe.st. 1985. A nutritional
explanation for body-size patterns of ruminant and
nonruminant herbivores. American Naturalist 125:
641-672.

Gregg, M. A., J. K. Barnett, and J. A. Crawford. 2008.
Temporal variation in diet and nutrition of preincu-

bating Greater Sage-Grouse. Rangeland Ecology and
Management 61:535-542.

Griffiths R, M. C. Double, K. Orr, and R. J. G.
D.awson. 1998. A DNA test to sex most birds.
Molecular Ecology 7:1071-1075.

Moss. R., A. Watson, and R. Parr. 1975. Maternal nutrition
and breeding success in Red Grouse (Lagopus lagopus
scolicus). Journal of Animal Ecology 44:233-244.

Nager, R. G. 2006. The challenges of making eggs. Ardea
94:323-346.

Robbins, C. T. 1993. Wildlife feeding and nutrition.
Academic Press, San Diego, California, USA.

Storch, I. 2000. Grouse status survey and conservation
action plan 2000-2004. WPA/BirdLife/SSC Grouse
Specialist Group, lUCN, Gland, Switzerland.

Sun. Y.-H. 2000. Distribution and status of the Chinese
Grouse Bonasa sewerzowi. Wildlife Biology 6:271-
275.

Sun, Y.-H. 2004. Distribution, reproduction strategy and
population biology of the Chinese Grouse (Bonasa
sewerzowi). Dissertation. Beijing Normal University,
Beijing, China.

Sun, Y.-H., Y. Fang, S. Klaus, C.-X. Jia, and G.-M.
Zheng. 2002. The application of data logger technique
to the study of incubation rhythms of the Chinese
Grouse. Journal of Beijing Normal University (Natural
Science) 38:260-265.

Sun, Y.-H., Y. Fang, J. E. Swenson, S. Klaus, and G.-M.
Zheng. 2005. Morphometries of the Chinese Grouse
Bonasa sewerzowi. Journal of Ornithology 146:24-26.

Sun, Y.-H., J. E. Swenson, Y. Fang, S. Klaus, and W.
Scherzinger. 2003. Population ecology of the Chi-
nese Grouse, Bonasa sewerzowi, in a fragmented
landscape. Biological Conservation 110:177-184.

Swenson, J. E. 1991. Social organization of Hazel Grouse
and ecological factors influencing it. Dissertation.
University of Alberta, Edmonton, Canada.

Swenson, J. E., L. Saari, and Z. Bonczar. 1994. Effects
of weather on Hazel Grouse reproduction: an allome-
tric perspective. Journal of Avian Biology 25:8-14.

Swenson, J. E., Y.-H. Sun, and N. Liu. 1996. A potential
method for age determination of the Chinese Grouse
Bonasa sewerzowi. Journal fur Ornithologie 137:255-
258.

Wang, N. and Z.-W. Zhang. 2009. The novel primers for
sex identification in the Brown-eared Pheasant and
their application to other species. Molecular Ecology
Resources 9:186-188.

Zheng, G.-M. and Q.-S. Wang. 1998. China Red Data
Book of Endangered animals, Aves. Science Press,
Beijing.

SHORT COMMUNICATIONS

181

The Wilson Joiirncil of Ornithology 122(1): 18 1- 182, 2010

Observations of a Possible Foraging Tool used by Common Ravens

Paul F. Frame

ABSTRACT. — Common Ravens [Coitus cora.x) are
opportunistic generalist foragers. Ravens during winter,
in some portions of their range, rely heavily on meat
scavenged from carcasses. Ravens use a variety of
strategies to find carcasses on the landscape, including
keying on audible cues that suggest the presence of a
food source. I documented Common Ravens investi-
gating a simulated animal distress call on 13 of 17 trials,
suggesting that investigating animal distress vocaliza-
tions may be one tool in the suite of foraging strategies
used by ravens. Received 6 June 2009. Accepted 6
September 2009.

The Common Raven [Coitus cora.x) is an
opportunistic generalist forager and uses a variety
of food sources including carrion, rodents, insects,
nestlings, eggs, seeds, fruit, and garbage (Bent
1946, Heinrich 1988, Engel and Young 1989,
Kelly et al. 2005). Many ravens at northern
latitudes rely heavily on meat from carcasses of
ungulates during winter (Temple 1974, Heinrich
1988, Hayes et al. 2000, Vustich et al. 2004).
These carcasses are an unpredictable food source
that is often difficult to locate on the landscape
(Heinrich 1988, Stabler et al. 2002, Vustich et al.
2004). Stabler et al. (2002) summarized strategies
that carrion feeders, such as ravens, use when
foraging for carcasses as: flying the landscape and
finding them visually by chance, foraging where
conspecifics or other species are seen foraging,
responding to vocalizations from other carcass
feeders, following other ravens that have previ-
ously located food from nocturnal roost sites, and
associating with predators that make carcasses
available. White (2005) found that ravens were
attracted to gunshots as a strategy for finding
ungulate gut piles discarded by successful hunters.
Ravens also cue in on appeasement calls of
conspecifics (Heinrich et al. 1993) and likely the
howling of wolves (Harrington 1978) as a means

' Department of Fisheries, Wildlife, and Conservation
Biology, University of Minnesota. St Paul, MN 55108,
USA.

‘Current address: Carcross Tagish First Nation, P. O.
Box 130, Carcross. YT YOB I BO, Canada; e-mail: pframe@
ualberta.net

of finding food. These examples indicate ravens
respond to audible stimuli as a foraging strategy. 1
present observations made during winter of ravens
responding to imitation prey distress calls and
suggest this response is another tool in the suite of
foraging strategies used by ravens.

OBSERVATIONS

I made these observations from 28 December
through 12 January during winter 2000-2001 on the
Superior National Eorest (SNE) in northeastern
Minnesota (47° 50' N, 91° 25' W) while attempting
to capture gray wolves [Canis lupus) for radio
collai'ing. I located radio-marked wolves from
fixed-wing aircraft and then homed in on the
location with hand-held telemetry immediately
after landing. I used a manual prey distress predator
call to attract wolves sufficiently close to dart them
with anesthesia for capture and radio collaring. I
called wolves at kill sites, non-kill sites, and sites
where kill information was unknown. I approached
wolves from downwind to a distance they would be
able to hear and respond to the predator call. I
concealed myself so that wolves would not
immediately identify me when they approached,
usually by backing into spruce [Picea spp.) or
balsam fir [Abies balsamea) trees, which also
concealed me from above. The predator call was
then blown to imitate a dying rabbit or young deer.
Notes were kept of the change of signal strength and
other occumences including ravens flying over.

The actual distance between my location and
wolves was unknown and varied from site to site.
Typically 6-25 ravens have been reported at wolf
kills (summarized in Vustich et al. 2004) with as
many as 80 being present in some cases (Carbyn
et al. 1993); between 0 and 16 ravens have been
reported in association with wolves away from
kills (Stabler et al. 2002). I observed between one
and five ravens responding to the predator call on
13 of 17 trials (76%) by flying within 100 m or
directly over my location. Those that flew directly
over me approached in a manner that suggested
directed Bight. Those that came within 100 m
Bew directly toward me and then changed course
when they appeared to detect me. Ravens

182

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. I, March 2010

responded during all trials near known kill sites (/?
= 8), once when it was known there was no kill
present {n = 3), and four of six times when kill
information was unknown.

DISCUSSION

I observed ravens investigating a sound that
simulates the distress cry of a dying animal. Ravens
are known to use audible cues as a foraging
technique (Heinrich et al. 1993, White 2005), but
this behavior has not previously been reported
regarding animal distress calls. Ravens congregate
at wolf kill sites (Carbyn et all 993, Stabler et al.
2002) and it can be assumed the eight trials at
known kills were conducted in the proximity of
ravens. The response rate of 100% (// = 8) to trials
in the proximity of ravens suggests this may be an
established foraging technique. Alternatively, ra-
vens may have been investigating the simulated
distre.ss call out of curiosity. My observations are
reported to pre.sent the possibility that ravens key on
prey distress calls as one tool in a suite of foraging
strategies. However, a more standardized study
with controls is required to test for an effect. White
(2005), in an experimental study, fired a rifle, blew
an air horn, a whistle, and did nothing in an attempt
to learn if ravens responded to gunshots. He found
that ravens were attracted to the gunshot, but not the
air horn or whistle. My observations were made
incidental to another project and I did not have a
control. White’s (2005) results suggest the raven
behavior I observed was in response to the distress
call and not a response to a novel loud noise. White
(2005) suggested that ravens might respond to
gun.shots because they associate the sound with
other sounds that indicate potential food. It is
possible prey distress calls are one such sound.

A carcass for ravens is an unpredictable food
source on the landscape (Heinrich 1988, Stabler et
al. 2002, Vustich et al. 2004). Using prey distress
calls to find a carcass would likely increase a
raven’s resource intake and better insure an
individual’s survival.

ACKNOWLEDGMENTS

These observations were made during a project funded
by the Undergraduate Research Opportunities Program in
the College of Natural Re.sources at the University of
Minnesota. C. E. Braun, L. D. Mech, D. R. Stabler, Crow
White, and an anonymous reviewer made useful comments
on earlier drafts of this manuscript. M. E. Nelson provided
field support and many idea generating conversations over
the years.

LITERATURE CITED

Bent, A. C. 1946. Life histories of North American jays,
crows, and titmice. Part 1. U.S. National Museum
Bulletin Number 191.

Carbyn, L. N., S. M. Oosenbrug, and D. W. Anions.
1993. Wolves, bison, and the dynamics related to the
Peace-Athabasca Delta in Canada’s Wood Buffalo
National Park. University of Alberta, Edmonton,
Canada.

Engel, K. A. and L. S. Young. 1989. Spatial and temporal
patterns in the diet of Common Ravens in southwest-
ern Idaho. Condor 91:372-378.

Harrington, E. H. 1978. Ravens attracted to wolf howling.
Condor 80:236-237.

Hayes, R. D., A. M. Baer, U. Wotschikowsky, and A. S.
Harestad. 2000. Kill rate by wolves on moose in the
Yukon. Canadian Journal of Zoology 78:49-59.
Heinrich, B. 1988. Winter foraging at carcasses by three
sympatric corvids with emphasis on recruitment by the
Raven, Corviis corax. Behavioral Ecology and Socio-
biology 23: 141-156.

Heinrich, B., J. M. Marzluff, and C. S. Marzluff. 1993.
Common Ravens are attracted by appeasement calls of
food discoverers when attacked. Auk 110:247-254.
Kelly, J. P., K. L. Etienne, and J. E. Roth. 2005. Factors
influencing the nest predatory behaviors of Common
Ravens in heronries. Condor 107:402-415.

Stahler, D., B. Heinrich, and D. Smith. 2002. Common
Ravens, Corvus corax. preferentially associate with
grey wolves, Canis lupus, as a foraging strategy in
winter. Animal Behaviour 64:283-290.

Temple, S. A. 1974. Winter food habits of raven on the
Arctic Slope of Alaska. Arctic 27:41^6.

Vustich, J. A., R. O. Peterson, and T. A. Waite. Raven
scavenging favours group foraging in wolves. Animal
Behaviour 67:1 17-1 126.

White, C. 2005. Hunters ring dinner bell for ravens:
experimental evidence of a unique foraging strategy.
Ecology 86:1057-1060.

SHORT COMMUNICATIONS

183

The Wilson Jonrmtl of Ornithology 122( 1 ): 1K3-185, 2010

Hanging Behavior of the Hooded Crow {Corviis coi nix)

Mario Melletti' -^ and Marzia Mirabile"

ABSTRACT. — We report observations of Hooded
Crows (Corvns corni.x) hanging upside down in the
wild, for both playing and harvesting acorns. This
behavior was recorded in six different events at the
same site (Villa Chigi’s urban park in Rome, Italy) and,
in two cases, involved two individuals at the same time.
Hanging behavior has been observed mainly in captive
Northern Ravens (Conns cora.x) and in few cases in
wild corvids. Our observations indicate hanging behav-
ior can be used to obtain food. These observations
confirm that corvids have enormous plasticity that can
be adapted to obtain food. Received 5 March 2009.
Accepted 19 July 2009.

Play by birds is difficult to define and is a
behavior that apparently has no purpose (Bekoff
1984, Heinrich 1999). Ficken (1977) noted that
play is much more difficult to define in birds than
in mammals. Birds may play socially, but only a
few species exhibit the full range of play
behaviors, from play chases to reciprocal object
play (Fagen 1981, Ortega and Bekoff 1987). Some
birds, such as corvids and parrots, exhibit more
extensive social play than others (Fagen 1981,
Iwaniuk and Pellis 2001). Corvids, among the
Passeriformes, show the most complex play
behavior including play with objects (e.g., objects
carried into the air, dropped, and then caught with
the beak many times), flight play, bathing play,
vocal play, hanging (upside down posture from a
branch holding with 1 foot, 2 feet or the beak),
allospecific interactions, sliding, and ‘snowromp-
ing’ (e.g., a raven lay on its breast and slid head
forwards downhill in the snow) (Ficken 1977,
Moffett 1984, Heinrich 1990, Ratcliffe 1997,
Heinrich and Smolker 1998, Brazil 2002).

Corvids may use cognitive tools in a way
similar to apes, as reported in a study comparing
mentality in these different animal groups (Emery
and Clayton 2004). In fact, corvids have a brain

' Estacion Biologica De Donana, C.S.l.C. Depaitment ot
Conservation Biology c/Americo Vespucio s/n. 41092
Seville, Spain.

^High Institute for Environmental Protection and Re-
search, Via Brancati 48. 00144 Rome, Italy.

-’Corresponding author; e-mail: mario.melletti@yahoo.it

significantly larger than in other birds and, in
proportion, their brain has similar size as that of
the chimpanzee (Pan spp.). Emery and Clayton
(2004) concluded that cognition in corvids and
apes must have evolved through a process of
divergent brain evolution with convergent mental
evolution. Heinrich and Bugnyar (2005) studied
the acquisition ol problem-solving behavioi in
captive ravens. Their results support the idea that
behavior of ravens in accessing meat on a string is
not only a result of rapid learning, but may
involve the understanding of cause-effect relation
between string and food.

Corvids and parrots can perform upside down
hanging behavior, which can be considered as
play (Elliot 1977, Ortega and Bekoff 1987,
Heinrich and Smolker 1998, Diamond and Bond
2003). Hanging behavior has been observed in
captivity many times in Northern Ravens (Corvus
corax) (e.g., Gwinner 1966, Coombs 1978,
Heinrich and Smolker 1998), but only a few
records have been reported for corvids in the wild
(McIntyre 1953, Gwinner 1966, Elliot 1977,
Heinrich and Smolker 1998, Melletti 1999).
Hanging behavior is primarily a type of play
performed by highly social birds, such as corvids
and parrots, which have the largest brain volume
(Heinrich and Smolker 1998, Diamond and Bond
2003, Emery and Clayton 2004). Hanging behav-
ior can be adopted for purposes other than play,
but most reasons for this behavior remain
unknown (Coombs 1978). Gwinner ( 1966) report-
ed ravens performing hanging behavior to store
food in an aviary and Melletti (1999) recorded a
case of hanging behavior, lasting up to 5 min, in a
wild raven for five consecutively times without
apparent reason.

Corvids have the ability to access food that is
difficult to reach, in particular ravens in captivity
(Heinrich 1995, Heinrich and Bugnyar 2()()5). Our
observations show that Hooded Crows (Corvus
cornix) are able to obtain acorns otherwise
difficult to access from an upright posture. Use
of hanging behavior by Hooded Crows demon-
.strates the ability to solve a problem. To our
knowledge this is the first observation reported of

184

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 1, March 2010

FIG. 1. Hanging behavior of Hooded Crows while feeding. Photograph by Mario Melletti.

hanging behavior in wild Hooded Crows with the
intent to obtain food. Little is still known about
this behavior.

OBSERVATIONS

We made observations of hanging behavior by
Hooded Crows from October 2005 through
October 2006 in the urban park of Villa Chigi in
Rome, Italy. The park is 5 ha in size and it is
characterized by Stone pine (Finns pinea). Holm-
oak (Querciis ilex), and open meadows. Data were
collected weekly for 6 hrs/week divided between
morning and afternoon (total number of hrs =
288). We conducted the study using lOX
binoculars and a digital camera.

Hooded Crows were observed for 96 of 288 hrs
during which hanging in upside down posture was
observed in six different events; two on 7 November
2005 (2 individuals), one on 14 January 2006 (1
individual), one on 6 October 2006 ( 1 individual),
and two on 26 October 2006 (2 different individ-
uals). Hanging behavior lasted from a few seconds
up to 1 min during which birds held the wings
pressed to the body and the head was held both
horizontally and vertically (Fig. I ).

DISCUSSION

Hanging behavior was observed by McIntyre
(1953) for wild Camon Crows (Corviis corone).
Gwinner (1966) and Heinrich and Smolker ( 1998)
for captive Northern Ravens, and Elliot (1977)
and Melletti (1999) for wild Northern Ravens.
Hooded Crows would hang upside down with one
foot and then land upright on the ground. We
observed two Hooded Crows that performed this
behavior alternating one and both feet while
hanging upside down. A Hooded Crow in hanging
posture was imitated by another bird on two
occasions, which performed the same posture at
the same time. Gwinner (1966) reported similar
cases of imitation in captive ravens.

We observed hanging behavior by Hooded
Crows in three events that .seemed as play, as
described by other authors (Gwinner 1966,
Heinrich and Smolker 1998), while in three other
occasions the goal of the crows was to reach
acorns of Holm-oaks, which were eaten or stored
in the ground. Acorns were located on small twigs
and upside down posture might seem to be the
only way for crows to obtain them. The upright
posture probably was not the best way to pick

SHORT COMMUNICATIONS

185

acorns, because the twigs were not sufficiently
strong to support the weight of crows in that
position. By hanging upside down, crows were
able to obtain acorns with their beak, before
coming back to the ground. They remained upside
down for few seconds. These observations
indicate that hanging behavior is not only a type
of play, but can be used to reach food, as also
reported by Gwinner (1966) for captive ravens.
Elliot (1977) suggested the hanging behavior of
ravens was for display by courting males. These
observations confirm that corvids have huge
plasticity that can be adapted to solve different
situations.

ACKNOWLEDGMENTS

We thank in particular Maria del Mar Delgado, Marc
Bekoff, and Igor Krugersberg for useful comments and
criticism.

LITERATURE CITED

Bekoff, M. 1984. Social play behavior. BioScience
34:228-233.

Brazil, M. 2002. Common Raven Conns corcix at play;

records from Japan. Ornithological Science 1 : 150—152.
Coombs, F. 1978. The Crows. Batsford, London, United
Kingdom.

Diamond, J. and A. B. Bond. 2003. A comparative analysis
of social play in birds. Behaviour 140:1091-1 1 15.
Elliot, R. D. 1977. Flanging behavior in Common Ravens.
Auk 94:777-778.

Emery, N. J. and N. S. Clayton. 2004. The mentality of
crows: convergent evolution of intelligence in corvids
and apes. Science 306:1903-1907.

Fagen, R. 1981. Animal play behavior. Oxford University
Press, New York, USA.

Ficken, M. S. 1977. Avian play. Auk 94:573-582.

□winner, E. 1966. U ber einige Bewegungsspiele des
Kolkraben (Conns corax L.). Zeitschrifl fur Tiei-psy-
chologie 23:28-36.

Heinrich, B. 1990. Ravens in winter. Banie and Jenkins,
London, United Kingdom.

Heinrich, B. 1995. An experimental investigation ot
insight in Common Ravens (Corvus corax). Auk
1 12:994-1003.

Heinrich. B. 1999. Mind of the raven: investigating and
adventures with wolf-birds. Harper Collins, New York.
USA.

Heinrich, B. and T. Bugnyar. 2005. Testing problem
solving in ravens: string-pulling to reach food.
Ethology 111:962-976.

Heinrich, B. and R. Smolker. 1998. Play in Common
Ravens (Corvus cora.x). Pages 27-44 in Animal play,
evolutionary, comparative and ecological perspectives
(M. Bekoff and J. A. Byers, Editors). Cambridge
University Press, Cambridge, United Kingdom.

IWANIUK, A. N. AND S. M. Pellis. 2001. Do big-brained
animals play more? Comparative analysis of play and
relative brain size in mammals. Journal of Compara-
tive Psychology 115:29-41.

McIntyre, N. 1953. Curious behaviour of Canion Crow.
British Birds 46:377-378.

Melletti, M. 1999. Un caso di “hanging in un corvo
imperiale Conns corax. Picus 1:112.

Moffett, A. T. 1984. Ravens sliding in snow. British Birds
77:321-322.

Ortega, J. C. and M. Bekoff. 1987. Avian play:
comparative evolutionary and developmental trends.
Auk 104:338-341.

Ratcliffe, D. 1997. The Raven. T. & A. D. Poyser,
London, United Kingdom.

186

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 1. March 2010

The Wilson Journal of Ornithology 122(1): 186- 187, 2010

Conspecific Brood Parasitism by the Dickcissel

Brian D. Peer'

ABSTRACT. — I report the first observation of
conspecific brood parasitism in the Dickcissel (Spiia
americana). I monitored 302 nests during a study of the
interactions between parasitic Brown-headed Cowbirds
(Molothrus ater) and Dickcissels. This and a possible
second case of conspecific brood parasitism may have
resulted from females laying in nests after their original
nests were destroyed in nearby hayfields. Received 9
June 2009. Accepted 4 September 2009.

Avian brood parasites lay their eggs in nests of
other individuals thereby avoiding costs associat-
ed with parental care. Obligate brood parasites
such as the Brown-headed Cowbird (Molothrus
ater) lay their eggs in nests of other species (Peer
et al. 2005), whereas conspecific brood parasites
lay in nests of individuals of the same species
(Yom-Tov 1980). Conspecific brood parasitism
(CBP) has been reported in 236 species (Yom-Tov
1980, 2001) and tends to be most common in
precocial species such as waterfowl, colonial
species, and species with specialized nest sites
such as cavities (Yom-Tov 1980, 2001; Rohwer
and Freeman 1989).

Females may practice CBP to increase their
fecundity by laying extra eggs in nests of other
individuals, in addition to tending their own
clutches (Soren.son 1991, Lyon 1993). Alterna-
tively, females that do not have a nest may
attempt to make the best of a bad situation and lay
their eggs in nests of other females or, if their
nests are destroyed during the laying cycle, they
may lay eggs in nests of nearby neighbors
(Hamilton and Orians 1965, Feare 1991; but see
Rothstein 1993, Peer and Sealy 2000). 1 report the
first known instance of CBP in the Dickcis.sel
(Spiza americana).

METHODS

I monitored Dickcissel nests at a reclaimed
mine from 2006 to 2008 in southern McDonough
County, Illinois, USA during a study of the

' Department of Biological Sciences, Western Illinois
University. Macomb. IL 61455, USA; e-mail: BD-Peer@
wiu.edu

interactions between Brown-headed Cowbirds and
their Dickcissel hosts. Dickcissel nests were
inspected for presence of cowbird eggs and to
record the Dickcissel’ s response to parasitism.
Most nests were checked daily, and each egg was
numbered. Nests were checked beginning at least
3 hrs following sunrise, after the time of day
Dickcissels typically lay (unpubl. data). I checked
nests sufficiently late to avoid missing an egg
being laid 1 day and returning the second day and
finding two additional eggs; this could lead to the
eiToneous conclusion that a second female laid in
the nest. Dickcissels nested in grasslands planted
as part of the mine reclamation and were
suiTounded by agricultural fields that included
hay in which the Dickcissels also nested. Nests in
hayfields were not monitored because the hay was
mowed once a month destroying the nests.

RESULTS

A total of 302 Dickcissel nests was monitored.
Modal clutch size was four eggs with a range of
three to seven eggs. Nest #41 was found in the
afternoon of 24 June 2008 with four eggs and I
numbered all eggs at this time. Seven eggs were in
this nest on 26 June. Because birds lay no more
than one egg per day (Johnson 1999), at least one
additional female laid an egg in this nest. The
hayfields immediately adjacent to this nest were
mowed the afternoon of 24 June. The nest
contents remained the same until 3 July when
five eggs had been removed, and two eggs
remained. The nest was subsequently deserted.

DISCUSSION

This is the first reported observation of
Conspecific Brood Parasitism (CBP) in Dickcis-
sels. The rarity of CBP in this species is not
surprising because Dickcissels do not demonstrate
traits associated with conspecific brood parasites.
What caused the parasitism that I observed'?
Dickcissels at my site nested in grasslands and
in adjacent hayfields. The hayfields were mowed
regularly during the breeding sea.son and in 2008
the hay adjacent to nest #41 was mowed in the
days prior to the parasitic event. It is possible the

SHORT COMMUNICATIONS

187

parasitic female was nesting in the hayfield and
her nest was destroyed by the mowing, after
which she laid one of her remaining eggs in nest
#41. However, I did not monitor nests in hayfields
and I did not ascertain the genetic identities of the
eggs. Thus it is unclear whether nest destruction
precipitated this case of CBP. Birds may make the
best of a bad situation by laying an egg in another
nest if their own nests are destroyed during the
laying stage (Hamilton and Orians 1965). Con-
specific brood parasites such as European Star-
lings {Stiiniis vulgaris) and Wood Ducks (A/.v
sponsa) lay in nests of conspecifics after their
nests have been destroyed or experimentally
removed during the laying cycle (Haramis et al.
1983, Feare 1991). Species not known to practice
CBP (Red-winged Blackbird [Agelaius phoeni-
ceus], Great-tailed Crackle [Quiscalus mexica-
nus]) do not lay parasitically after their nests have
been experimentally removed (Rothstein 1993,
Peer and Sealy 2000). Illinois was mostly
grassland prior to European colonization. Today
<1% of the original grassland habitat remains
(Iverson 1988) and there is little suitable nesting
habitat for Dickcissels and other grasslands birds
(e.g., Herkert 2003). The density of Dickcissels at
my site was high (upubl. data), possibly due to
lack of suitable habitat in the western Illinois
region. Dickcissels may nest in suboptimal habitat
such as hayfields that are mowed causing nest
failure during laying. These females may have
responded by laying in nests of nearby females,
which is a better strategy than dumping physio-
logically committed eggs indiscriminately in the
environment.

The seven-egg clutch at this nest is the largest
reported clutch for the Dickcissel, one egg more
than what had been reported previously (Temple
2002). I also observed a second nest with a seven-
egg clutch. On 26 June 2006 nest #81 was found
with four eggs. On 28 June it contained six eggs,
on 29 June seven eggs, and the contents remained
the same until the eggs hatched. There were no
laying irregularities during the time I monitored
the nest that would suggest CBP, but a second
female could have laid an egg in the nest before I
found it on 26 June. This is only the second time a
seven-egg clutch has been reported tor this
species and the other was a result ol CBP.

ACKNOWLEDGMENTS

1 thank G. K. Arnett anti the hreeman Mining Company

lor allowing me to conduet my studies on their propeity. M.

D. Benson and C. G. Peer assisted with the field work.

LITERATURE CITED

Feare, C. J. 1991. Inlraspecinc ne.st parasitism in Starlings
Sliinuis vulgaris: effects of disturbance on laying
females. Ibis 133:75-79.

Hamilton III, W. J. and G. H. Orians. 1965. Evolution of
brood parasitism in altricial birds. Condor 67:361-382.

Haramis, G. M., W. G. Alliston, and M. E. Richmond.
1983. Dump nesting in the Wood Duck traced by
tetracycline. Auk 100:729-730.

Herkert, J. R., D. L. Reinking, D. A. Wiedenfield, M.
Winter, J. L. Zimmerman, W. E. Jensen, E. J. Finck,
R, R. Koford, D. H. Wolfe, S. K. Sherrod, M. A.
Jenkins, J. Faaborg, and S. K. Robinson. 2003.
Effects of prairie fragmentation on the nest success of
breeding birds in the midcontinental United States.
Conservation Biology 17:587-594.

Iverson, L. R. 1988. Land-use changes in Illinois, USA:
the influence of landscape attributes on current and
historic land use. Landscape Ecology 2:45-61.

Johnson, A. L. 1999. Reproduction in the female. Pages
569-596 in Sturkie’s avian physiology (G. C.
Whittow, Editor). Fifth Edition. Academic Press,
New York, USA.

Lyon, B. E. 1993. Conspecific brood parasitism as a
flexible female reproductive tactic in American Coots.
Animal Behaviour 46:91 1-928.

Peer, B. D. and S. G. Sealy. 2000. Conspecific brood
parasitism and egg rejection in Great-tailed Grackles.
Journal of Avian Biology 31:271-277.

Peer, B. D., S. I. Rothstein, M. J. Kuehn, and R. C.
Fleischer. 2005. Host defenses against cowbird
(Molothrus spp.) parasitism: implications for cow'bird
management. Ornithological Monographs 57:84—97.

Rohwer, F. C. and S. Freeman. 1989. The distribution of
conspecific nest parasitism in birds. Canadian Journal
of Zoology 67:239-253.

Rothstein, S. 1. 1993. An experimental test of the
Hamilton-Orians hypothesis for the origin of avian
brood parasitism. Condor 95:1000-1005.

Sorenson, M. D. 1991. The functional-significance of
parasitic egg-laying and typical nesting in Redhead
ducks: an analysis of individual behavior. Animal
Behaviour 42:771-796.

Temple, S. A. 2002. Dickcissel (Spiza amcricana). The
birds of North America. Number 703.

Yom-Tov, Y. 1980. Intraspecific nest parasitism in birds.
Biological Reviews 55:93-108.

Yom-Tov, Y. 2001. An updated list and some comments on
the occLiiTence of intraspccific nest parasitism in birds.
Ibis 143:133-143.

188

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 1. March 2010

The Wilson Journal of Ornithology 122( 1 ): 1 88-1 89. 2010

First Report of Olrog’s Gull Depredation by Sympatric Kelp Gulls

Luciano La Sala' " and Sergio Martorelli'

ABSTRACT. — We observed adult Kelp Gulls {Larns
clominicanus) capture and eat Olrog’s Gull (L. atlanti-
CHs) chicks in the Bahia Blanca estuary, Argentina. This
estuary holds the largest breeding colony of Olrog's
Gulls. There are no previously published reports of Kelp
Gulls capturing and eating Olrog’s Gull chicks. Our
data support suggestions made by other authors about
the possible existence of conflicts in colonies where
both species breed in close proximity. Received 13
March 2009. Accepted 7 October 2009.

Olrog’s Gull {Lams atkmticus) (Olrog 1958) is
endemic to the coastal wetlands of the Atlantic
Coast in Argentina, Uruguay, and southern Brazil
(Yorio et al. 2005), and is listed as a Vulnerable
species (Birdlife International 2008). Factors that
led to this classification include the limited
geographical range of the species, small estimated
population size (4,000-5,000 pairs), its special-
ized diet, and high susceptibility to anthropogenic
changes in the environment.

Few Olrog’s Gull breeding sites have been
identified, and all are in Argentina (Yorio et al.
2005). Over 80% of the total breeding population
is concentrated in the Bahia Blanca estuary with
the largest colony (—3,600 pairs) on Isla del
Puerto (38° 48' S, 62° 15' W) surrounded by a
colony of Kelp Gulls (L. dominicaniis) (Delhey et
al. 2001). The estimated size of the later was
—5,000 pairs in 2003 but appears to have grown
considerably since then (P. F. Petracci, pers.
comm.).

There is only one previous report of depreda-
tion attempts by Kelp Gulls on Olrog’s Gull eggs
and chicks (Petracci et al. 2004), and no reports of
successful depredation were found during a
literature search. This is the first report of Olrog’s
Gull chicks as prey of adult Kelp Gulls.

' Centro cic E.stuciios Parasitologicos y Vectores (CON-
ICET-UNLP). Calle 2 N 384 CEPAVE, 1900 La Plata.
Buenos Aires, Argentina.

’Corresponding author; e-mail: lucianolasala@yahoo.
com.ar

OBSERVATIONS

We monitored depredation of Olrog’s Gull eggs
and chicks in the Isla del Puerto breeding colony
in 2005, 2006, and 2007. We studied the colony in
2005 during part of the egg-laying phase (29 Sep-
7 Oct) and part of the hatching and chick-rearing
periods (27 Oct- 15 Nov) and, in 2006 and 2007,
during the early hatching and chick-rearing
periods (17 Oct-9 Nov and 1 Oct-1 Nov,
respectively). We spent —8 hrs per day in the
colony between 0800 and 1200 hrs ST, and
between 1500 and 1900 hrs. We did not observe
direct egg depredation in any year but did observe
depredation of Olrog’s Gull chicks in 2006 and
2007. We observed two depredation events in
2006, one on 20 October (1015 hrs) and the other
on 1 November (1945 hrs). We observed six
depredation events in 2007; 21 October (1800
hrs), 22 October (1905 hrs), 27 October (1715
hrs), 28 October (1045 hrs and 1910 hrs), and 30
October (2018 hrs).

Adult Kelp Gulls usually made fast, low-flying
passes over the Olrog’s Gull colony and used their
beaks to snatch chicks from or around their nests.
We observed one Kelp Gull drop the Olrog’s Gull
chick while hovering above the colony, causing
the death of the chick shortly after its impact with
the ground. We calculated all depredated chicks
were —10 days of age based on body size. We
strongly suspect the same Kelp Gull was respon-
sible for four of the six events observed in 2007.
This Kelp Gull returned to the same nest —4 m
from the Olrog’s Gull colony while eating the
chicks. Adult Olrog’s Gulls did not display
defensive behavior during the attacks, nor while
Kelp Gulls fed on chicks a short distance from the
Olrog’s Gull colony.

DISCUSSION

Kelp Gull populations are expanding through-
out most of their geographical range (Yorio et al.
2005), and their generalistic/opportunistic feeding
habits are thought to have an important role in this
expansion (Giccardi el al. 1997). The negative
effects of expanding Kelp Gull populations on

SHORT COMMUNICATIONS

189

other seabird species has been associated with
increased artificial food sources (McGehee and
Eitniear 2007). One study conducted at six
colonies, where both Kelp and Olrog’s gulls
breed, indicated these species share many micro-
habitat characteristics, and suggested the presence
of spatial conflicts between them (Garcia Borbor-
oglu and Yorio 2007a, b).

Isla del Puerto is within an international
deepwater port and 3.2 km from the city of Bahia
Blanca (-450,000 inhabitants). Discarded fish
from a small-scale fishing fleet and presence of
open-sky refuse tips, slaughter houses, and pork
and poultry production systems near these colo-
nies provide readily available and abundant food
sources throughout the year, and might sustain
growth of this colony (Petracci et al. 2004). This
could lead to increased predatory pressure by
Kelp Gulls on Olrog’s Gulls.

Individual specialization as predators of seabird
chicks and eggs has been reported for a number of
marine birds including Kelp Gulls (Yorio and
Quintana 1997, Quintana and Yorio 1998), other
gull species (Harris 1965, Hatch 1970, Parsons
1971, Fuchs 1977, Hulsman 1977), and Great
Skuas {Stercorahiis skua) (Votier et al. 2004,
2007). In our study, it is likely the series of
consecutive predatory attacks observed in 2007
was by a single specialist Kelp Gull.

Our observation of eight chicks depredated over a
3-year period would not suggest substantial pressure
on this Olrog’s Gull colony. However, we provide
sufficient evidence to suggest further studies that
accurately quantify the impact of depredation and
other possible conflicts between these species.

ACKNOWLEDGMENTS

We thank Club Nautico Bahia Blanca, all members of the
Fernandez family and park rangers Martin Sotelo and
Lucrecia Diaz for their support. We also thank Joaquin
Cereguetti and Nicolas Acosta for field assistance. This
study was partly funded by Agenda Nacional de Promocion
CientiTica y Tecnologica (PICT 34412/05).

LITERATURE CITED

BirdLife International. 2008. Species fact sheet: Lants
atlanticus. http://www.birdlife.org (accessed 20 May
2008).

Delhey J. K.. V. P. F. Petracci, and C. M. Grassini.
2001. Hallazgo de una nueva colonia de Gaviota de
Olrog (Lams atlanticus) en la ria de Bahia Blanca.
Argentina. Hornero 16:39-42.

Fuchs, E, 1977. Predation and anti-predator behaviour in a
mixed eolony ol terns Sterna sp. and Blaek-headed
Gulls Lams mdihunchis with speeial rclerenee to the
Sandwieh Tern Sterna sandvieencis. Ornis Seandina-
vica 8: 17-32.

GarcIa Borboroglu, P. and P. Yorio, 20()7a. Breeding
habitat requirements and selection by Olrog’s Gull
(Lams atlanticus), a threatened species. Auk 124:
1201-1212.

GarcIa-Borboroglu, P. and P. Yorio. 2007b. Compar-
ative habitat use by syntopic Kelp Gulls (Lams
donunicanus) and Olrog’s Gulls (L. atlanticus) in
coastal Patagonia. Emu 107:321-326.

Giccardi, M., P. Yorio, and M. E. Lizurme. 1997.
Patrones estacionales de abundancia de la Gaviota
Cocinera (Lams donunicanus) en un basural patago-
nico y si relacion con el manejo de residues urbanos y
pesqueros. Ornitologia Neotropical 8:77-84.

Harris, M. P. 1965. The food of some Lams gulls. Ibis
107:43-53.

Hatch, J. J. 1970. Predation and piracy by gulls at a ternery
in Maine. Auk 87:244-254.

Hulsman, K. 1977. Breeding success and mortality of terns
at One Tree Island, Great Barrier Reef. Emu 77:49-60.

McGehee, S. M. and J. C, Eitniear. 2007. Kleptoparasit-
ism of Magellanic flightless steamer-ducks (Tachyeres
pteneres) by Kelp Gulls (Lams donunicanus). Boletin
SAO 14:143-146.

Olrog, C. C. 1958. Notas ornitologicas sobre la coleccion
del Instituto Muguel Lillo, Tucuman. Acta Zoologica
Lilloana 15:5-18.

Parsons, J. 1971. Cannibalism in Herring Gulls. British
Birds 64:528-537.

Petracci, P., L. La Sala, G. Aguerre, C. Perez, N.
Acosta, M. Sotelo, and C. Pamparana. 2004. Dieta
de la gaviota cocinera (Larus donunicanus) durante el
periodo reproductivo en el estuario de Bahia Blanca.
Buenos Aires, Argentina. Hornero 19:23-28.

Quintana, F. and P. Yorio, 1998. Kelp Gull Lams
donunicanus predation on an Imperial Cormorant
Phcdacrocora-x atriceps colony in Patagonia. Marine
Ornithology 26:84-85.

Votier, S. C., S. Bearhop, N. R.atcliffe, and R. W.
Furness. 2004. Reproductive consequences for Great
Skuas specializing as seabird predators. Condor
106:275-287.

Votier, S. C.. S. Bearhop. J. E. Crane, J. M. Arcos, and
R. W. Furness. 2007. Seabird predation by Great
Skuas Stercorarius skua: intra-specific competition for
food? Journal of Avian Biology 38:234-246.

Yorio. P. and F. Quintana. 1997. Predation by Kelp Gulls
Larus donunicanus at a mixed-speeies colony of Royal
and Cayanne terns Sterna ma.xima and S. euryf’natha in
Patagonia. Ibis 139:536-541,

Yorio, P., M. Bertellotti. and P. Garcia Borboroglu.
2005. Estado poblacional y de conscrvacion de
gaviotas quo se reproducen en el litoral maritimo
Argentino. Hornero 20:5.3-74.

190

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. I, March 2010

The Wilson Journal of Ornithology 122(1): 190- 193, 2010

Second and Third Records of Snares Penguins {Eudyptes rohustus) in the

Falkland Islands

Laurent Demongin, ' Maud Poisbleau,’ “ Georgina Strange, '' and Ian J. Strange^

ABSTRACT. — The Snares Penguin (Eudyptes rohus-
tus) breeds only on the Snares Islands, New Zealand,
and is vagrant throughout the New Zealand region and
southeast Australia. The only previous record outside
this area was one in the Falkland Islands in 1988. We
report the unusual occurrence of two Snares Penguins in
the same colony in the Falkland Islands in 2008, and
discuss identification issues. Vagrant penguins demon-
strate the incredible dispersal ability of these flightless
birds. Received 9 March 2009. Accepted 19 July 2009.

Snares Penguins (Eudyptes rohustus) are en-
demic to New Zealand, breeding only on the
Snares Islands (48° 00' S, 166° 33' E), and
straying to surrounding islands, south-east Aus-
tralia, and Tasmania (Marchant and Higgins 1990,
Williams 1995). They are absent from their
breeding areas during winter, when they are
probably pelagic and migratory. Snares Penguins
are rarely observed away from the Snares Islands
and their movements are poorly known (Marchant
and Higgins 1990). The occurrence of at least five
Snares Penguins in the Chatham Islands in
January and February 2003 was considered
exceptional (Miskelly and Bell 2004, Miskelly et
al. 2006). The only previous record of this species
beyond Australasia was an adult in the Settlement
Rookery on New Island (51° 43' S, 61° 17' W) in
the Falkland Islands on 10 December 1988
(Lamey 1990, Martinez 1992). We discuss the
unusual occurrence of two Snares Penguins in the
same colony in the Falkland Islands 20 years
later.

' Max Planck Institute for Ornithology, Vogelwarte
Radolfzell, Schlossallee 2, 7831.‘i Radolfzell, Germany.

^University of Antwerp, Campus Drie Eiken, Depart-
ment Biology-Ethology, Universiteitsplein 1 , 2610 Antwerp
(Wilrijk), Belgium.

’New Island Conservation Trust, New Island, FIQQ IZZ,
Falkland Islands.

■“Corresponding author; e-mail: laurentdemongin®
gmail.corn

OBSERVATIONS

We visited daily the Settlement Rookery on
New Island during the entire breeding season
during the course of scientific studies of Western
Rockhopper Penguins (Eudyptes chrysocome)
(Poisbleau et al. 2008). An adult Snares Penguin
was found on 30 November and 1 December 2008
(Fig. 1). The rookery was visited every day until
the middle of January 2009, but the bird was not
seen again. However, a second bird was observed
on 24 and 25 December 2008 (Fig. 2). We
captured both birds to take standard measure-
ments (Table 1). The appearance of the birds was
sufficiently different to be certain they were not
the same individual; the second bird was larger,
the black color of its head was duller, it had a
partially gray chin, its crests were very short, the
pink color of its gape and of the skin around its
beak was less obvious, and its eyes were brownish
(not red) even under strong light.

DISCUSSION

The identification of extra-Iimital Snares Pen-
guins requires care. The following characteristics
allow separation from Western Rockhopper Pen-
guins (Shirihai 2007): a wider yellow stripe above
the eye starting close to the beak, an obvious pink
gape and skin along the edge of the lower
mandible, a larger and more bulbous beak, the
absence of a black crest on the rear crown, a
shorter yellow crest less pendulous behind the
eye, a bold black tip on the underside of the
nipper, and an overall larger size. The large beak
and overall size, and the pink gape and skin along
the beak could also resemble a hybrid between a
Rockhopper Penguin and a Macaroni Penguin
(Eudyptes ehrysolophus). Several instances of
such hybridization have been documented in the
Falkland Islands (White and Clausen 2002).
However, the shape of the yellow stripe and crest
of the birds that we found were typical of Snares
Penguin and completely different to those of
Rockhopper or Macaroni penguins or postulated
hybrids between them.

SHORT COMMUNICATIONS

191

FIG. 1. Adult Snares Penguin (right) next to a Western Rockhopper Penguin on New Island, Falkland Islands, 30
November 2008. Photograph by Laurent Demongin.

A Snares Penguin could also be mistaken for an
Erect-crested Penguin (Eiidyptes sclatevi), a spe-
cies that breeds on the Antipodes and Bounty
islands southeast of New Zealand (Marchant and
Higgins 1990) and that occasionally reaches the
Falkland Islands (Strange 1982, Summers 2005),
including one that bred with Rockhopper Pen-
guins on Pebble Island ( 120 km northeast of New
Island) between January 1997 and January 2008
(Morrison et al. 2005, Arnold 2008). Erect-crested
Penguins have crests that are erectile and nearly
parallel when seen from the front (Marchant and
Higgins 1990) while Snares Penguins have crests
that droop behind the eye, and forms a V when
seen from the front. Erect-crested Penguins also
have a different head shape with a domed crown,
and larger chin reaching closer to the beak tip
(Shirihai 2007). Cases of hybridization between
Erect-crested and Rockhopper penguins have been
reported at the Falkland Islands (Napier 1968) but
the descendants were not described. An apparent
Erect-crested x Rockhopper hybrid seen next to

the Erect-crested Penguin on Pebble Island in
January 2006 looked similar to the birds that we
saw, but differed in having the extended chin
profile of an Erect-crested Penguin, and black
feathers in the crest (a Rockhopper characteristic).
Other details of the superciliary stripe and bare
skin at the gape also matched Western Rockhop-
per or Erect-crested penguins rather than Snares
Penguins (C. M. Miskelly, pers. comm.).

The first bird we found, based on measurements
and plumage patterns, was probably an adult female
whereas the second was probably an immature male.
The beak of the second bird was much longer and
was not as deep. The beak shapes were quite
different, the first one being bulbous with parallel
edges while the second was less robust and more
slender. The latter beak shape is typical of imma-
tures; the more massive beak of adults is achieved
during the third or later years (Stonehouse 1971).

The weight of crested penguins varies season-
ally with the heaviest birds being recorded before
the molt (Warham 1975). The weight of the

192

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 1. March 2010

FIG. 2. Immature Snare.s Penguin (right) among Western Rockhopper Penguins on New Island, Falkland Islands, 24
December 2008. Photograph by Laurent Demongin.

second bird we found exceeded the range of
breeding Snares Penguins (Stonehouse 1971) but
fit well with pre-molting immature males (War-
ham 1974a).

Most vagrant crested penguins are sub-adults
recorded before or during their molt (Woehler

1992, Miskelly and Bell 2004), like the bird we
observed on 24-25 December. The two other
Falkland Islands records of Snares Penguin
(including our first bird) were adults during the
breeding season. However, observers rarely visit
the rookeries after the end of the breeding season,

TABLE 1 . Mea.surements (following Warham 1975) of the two Snares Penguins found in the Falkland Islands in 2008,
and comparison with x, (n), and (range] data from (a) Stonehouse (1971) and (b) Warham (1974b).

Characicristic

Bird 1

Adult male

Adult female

Bird 2

(a)

(b)

(a)

(b)

Flipper, mm

181

191

183 (61)

184.0 (1 14)

177.3 (47)

178.8 (82)

[170-193]

(167-187]

Beak length, mm

54.2

57,1

59.2 (68)

59.1 (35)

52.5 (58)

52.0 (12)

(54-691

(49-61]

'

Beak depth, mm

23.3

22.6

28.2 (35)

24.2 (12)

Head and beak length.

122.6

127.9

mm

Mass, g

3.300

4,360

3,320 (41)

2,780 (32)

] 2,450-4,300]

[2,300-3,400]

SHORT COMMUNICATIONS

193

and vagrant birds coming ashore to molt could
easily be missed.

The Chatham Islands are only ~ 1 ,400 km from
the Snares Islands, and yet the presence of several
Snares Penguins in 2003 was considered anoma-
lous, possibly reflecting a changed at-sea distri-
bution (Miskelly and Bell 2004). It is difficult to
assess the reasons for the presence of this species
in the Falkland Islands, —8,000 km from the
Snares Islands. However, the occurrence of
penguins endemic to the New Zealand region in
the Falkland Islands demonstrates their incredible
dispersal ability. The opposite is also true with
Western Rockhopper Penguins reported from the
Snares Islands south of New Zealand (Tennyson
and Miskelly 1989, Miskelly et al. 2001).

ACKNOWLEDGMENTS

We are grateful to the New Island Conservation Trust for
permission to work on the island and for providing
accommodations. We thank Maria Strange, Dan Birch,
Rafael Matias, and Petra Quillfeldt for their support and
advice during the field season. The manuscript benefited
greatly from critical comments by Maria Strange. C. M.
Miskelly, C. E. Braun, and an anonymous referee.

LITERATURE CITED

Arnold, N. 2008. Falkland Islands. 10 to 23 January 2008.
Trip report. The Travelling Naturalist, Limosa Holi-
days. www.naturalist.co.uk/reports2008/falklands08.
pdf (accessed 13 July 2009).

Lamey, T. C. 1990. Snares Crested Penguin in the Falkland
Islands. Notomis 37:78.

Marchant, S. and P. J. Higgins. 1990. Handbook of
Australian. New Zealand and Antarctic birds. Volume
1 . Ratites to ducks. Part A. Ratites to petrels. Oxford
University Press, Melbourne, Australia.

Martinez, I. 1992. Family Sphenicidae (Penguins). Pages
140-160 in Handbook of the birds of the world.
Volume 1. Ostrich to ducks (J. del Hoyo, A. Elliott,
and J. Sargatal, Editors). Lynx Edicions, Barcelona.
Spain.

Miskelly, C. M. and M. Bell. 2004. An unusual influx of
Snares Crested Penguins (Eiidyptes robnstus) on the
Chatham Islands, with a review of other crested penguin
records from the islands. Notomis 5 1 :235-237.
Miskelly, C. M., A. J. Bester. and M. Bell. 2006.

Additions to the Chatham Islands’ bird list, with
further records of vagrant and colonising bird species.
Notomis .63:213-228.

Miskelly, C. M., P. M. Sagar, A. J. D. Tenny.son, and R.
Scofield. 2001. Birds of the Snares Islands, New
Zealand. Notomis 48:1-40.

MoRRtsoN, M., A. Henry, and R. Woods. 2005. Rare and
vagrant birds in the Falkland Islands 2005. Falk-
lands Conservation, London, United Kingdom, www.

falklandsconservation.com/wildlife/birds/Fl-RareBirds-

05.pdf (acces.sed 13 July 2009).

Napier, R. B. 1968. Erect-crested and Rockhopper
penguins interbreeding in the Falkland Islands. British
Antarctic Survey Bulletin 16:71-72.

PoisBLEAU, M., L. Demongin, I. J. Strange, H. Otley,
AND P. Quillfeldt. 2008. Aspects of the breeding
biology of the Southern Rockhopper Penguin Endyptes
c. chrvsocome and new consideration on the intrinsic
capacity of the A-egg. Polar Biology 31:925-932.
Shirihai, H. 2007. A complete guide to Antarctic wildlife.
The birds and marine mammals of the Antarctic
continent and the Southern Ocean. Second Edition.
A&C Black Publishers, London, United Kingdom.
Stonehouse, B. 1971. The Snares Islands Penguin
Endyptes robnstus. Ibis 113:1-7.

Strange, I. J. 1982. Breeding ecology of the Rockhopper
Penguin (Endyptes crestatus) in the Falkland Islands.
Le Gerfaut 72:137-188.

Summers, D. 2005. A visitor’s guide to Falkland Islands.
Second Edition. Falklands Conservation, London.
United Kingdom.

Tennyson, A. J. D. and C. M. Miskelly. 1989. “Dark-
faced” Rockhopper Penguins at the Snares Islands.
Notomis 36:183-189.

Warham, j. 1974a. The breeding biology and behaviour of
the Snares Crested Penguin. Journal of the Royal
Society of New Zealand 4:63-108.

Warham, J. 1974b. The Fiordland Crested Penguin
Endyptes pachyrhynclins. Ibis 1 16:1-27.

Warham, J. 1975. The crested penguins. Pages 189-269 in
The biology of penguins (B. Stonehouse, Editor).
Macmillan Press, London, United Kingdom.

White, R. W. and A. P. Clausen. 2002. Rockhopper
Endyptes chrysocome chrysocome x Macaroni E.
chrysolopbns penguin hybrids apparently breeding in
the Falkland Islands. Marine Ornithology 30:40^2.
Williams, T. D. 1995. The penguins. Oxford University
Press, Oxford, United Kingdom.

WOEHLER. E. J. 1992. Records of vagrant penguins from
Tasmania. Marine Ornithology 20:61-73.

194

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. 1, March 2010

The Wilson Journal of Ornithology 122( 1 ): 194-195, 2010

A New Bird Species for Costa Rica; Sapphire-throated Hummingbird

(Lepidopyga coerideogidahs)

Esteban Biamonte*

ABSTRACT. — I present the first confirmed record of
the Sapphire-throated Hummingbird (Lepidopyga coer-
uleogularis) at Punta Banco, southwestern Costa Rica.
An adult male was seen and photographed perched on a
twig at the border of a second growth forest tract. This
observation extends the geographic distribution of this
.species from northern Colombia to southwestern Costa
Rica. Received 28 Fehriiarv 2009. Accepted 27 August
2009.

The Sapphire-throated Hummingbird {Lepi-
dopyga coenileogidaris) is known to occur
from western Panama to northern Colombia,
along most of the Pacific slope and over the
eastern region of the Caribbean slope of
Panama into Colombia. This hummingbird is
more abundant in coastal forests (Ridgely and
Gwynne 1989), occupying primarily secondary
forests and scrubby clearings, and in lower
frequency, around mangrove patches. It is
usually observed foraging for both nectar and
insects at low strata (Ridgely and Gwynne
1989, Stiles et al. 1989).

This species is known in Costa Rica from an
old and possibly mislabelled specimen and from
an unconfirmed record from 1962 (Stiles et al.
1989). Since then, this species has not been
reported for Costa Rica (Obando-Calderon et al.
2007).

OBSERVATIONS

An adult male Sapphire-Throated Humming-
bird was observed foraging and photographed on
2 April 2008 at Tiskita Jungle Lodge landing
strip at Punta Banco, Costa Rica (8° 21' N, 83°
06' W) (Fig. 1). This area is —10 km from the
Panamanian border. Possibly the same male was
observed the previous day, foraging at the same
spot. The hummingbird was feeding on flowers

' E.scucla (Je Biologi'a, Univensidad de Costa Rica. San
Pedro de Montes dc Oca. Costa Rica; e-mail: cstebanbiamontc@
yahcKi.com

of a vine of the Rubiaceae family growing near
the lodge’s airstrip. The area is an open field
surrounded by patches of secondary forest and
early second growth vegetation, numerous orna-
mental plants in the lodge’s gardens, and an
extensive living fence of flowering Ihiscus rosa-
sinensis, which attracts several other species of
hummingbirds.

On 23 December 2008 a new report of this
species was made by Kevin Easley even further
into Costa Rica, —35 km from the Panamanian
border at la Gamba near Golfito. This region is
comprised of extensive pastures with living fences
and isolated trees suiTounded by mature protected
forest on the slopes that surround the valley. The
species was seen visiting flowers of garden plants
and living fences.

DISCUSSION

This is one of several bird species, typical of
open areas that have been recently reported in
Costa Rica, possibly coming from Panama (Stiles
et al. 1989; Sanchez et al. 1998; EB, unpubl.
data). The range expansion of these species may
be caused by the increasing destruction of the
natural mature forests in the region during the last
decades (Sanchez-Azofeifa et al. 2001, 2003).
Mo,st of the original forest has been eliminated in
the lowlands of Central America and replaced by
pastures or agricultural fields, allowing some
birds to expand their original range. The Sap-
phire-throated Hummingbird is expected to soon
become more common along the Costa Rican
Pacific Coast, given its use of open areas and
behavior.

ACKNOWLEDGMENTS

I thank .Saul Bocian, for being in the right place at
the right time to take the great picture.s and for sharing
them, everyone at Tiskita .lungle Lodge, Onik Morrison,
and Jay Morrison for revising the English, and Luis
Sandoval and Gilbert Barrantes for checking the
manuscript.

SHORT COMMUNICATIONS

195

FIG. 1. Male Sapphire-throated Hummingbird in Costa Rica’s Pacific Coast (photograph by Saul Bocian).

LITERATURE CITED

Obando-Calderon, G., L. Sandoval, J. Chaves-Cam-
pos, J. Villareal-Orias, and W. Alfaro-Cer-
vantes. 2007. Lista Oficial de aves de Costa Rica.
Zeledonia. Numero Especial 1 1:1-76.

Ridgely, R. S. and j. a. Gwynne. 1989. A guide to the
birds of Panama with Costa Rica. Nicaragua and
Honduras. Princeton University Press. Princeton. New
Jersey. USA.

Sanchez-Azofeifa, G. a., R. C. Harrlss, and D. L.
Skole. 2001. Deforestation in Costa Rica: a quanti-

tative analysis using remote sensing imagery. Biotrop-
ica 33:378-384.

Sanchez-Azofeifa, G. A., G. C. Daily, A. S. P. Pfaff,
AND C. Busch. 2003. Integrity and Isolation of Costa
Rica's national parks and biological reserves: exam-
ining the dynamics of land-cover change. Biological
Conservation 109:123-135.

Sanchez. J. E., K. Naoki. and J. Zook. 1998. New
information about Costa Rican birds. Ornitologia
Neotropical 9:99-102.

Stiles. G. F., A. F. Skutch. and D. Gardner. 1989. A
guide to the birds of Costa Rica. Cornell University
Press. Ithaca. New York. USA.

The Wilson Journal of Ornithology 122( 1): 196-205, 2010

Ornithological Literature

Robert B. Payne, Review Editor

THE CURSE OF THE LABRADOR DUCK.
MY OBSESSIVE QUEST TO THE EDGE OF
EXTINCTION. By Glen Chilton. Simon &
Schuster, New York, USA. 2009: x + 305 pages,
10 black and white figures, 1 map. ISBN: 978-1-
4391-0247-3; 978-1-4391-2499-4 (ebook). $25.00
(hardcover). — For those whose interest in birds
extends beyond another tick on their life list,
extinct birds often provide a source of intense
fascination. Each species has its own allure, but
one that is also coupled with an air of unattain-
ability, like a photograph of a beautiful woman on
the obituary page. The Labrador Duck {Campto-
rhynciius lahradorius), the first endemic North
American bird to disappear historically, about
1875, remains one of the most perversely
enigmatic of extinct birds. Although known from
a fair number of specimens (at least 55 still exist),
few have reliable data and virtually nothing is
known about the breeding of the species or its
summer distribution. It has escaped the attention
of monographers such as have labored over the
Great Auk (Pingiiinus impennis). Passenger Pi-
geon (Ectopistes migratoriiis), and Carolina Par-
akeet (Coniiropsis carolinensis).

Enter Glen Chilton, who debuted in ornithology
on the bandwagon of studies of song dialects of
the White-crowned Sparrow, which led to his
being an author on the account of Zonotrichia
leiicophrys in the Birds of North America series.
Lured by the prospect of a gratis set of that costly
work, he chose to write another species account
and what could possibly have been easier than the
Labrador Duck, about which so little was known?
With the account of the Labrador Duck under his
belt, Chilton became its de facto “authority,”
thus setting the stage for his “obsessive” quest to
examine every specimen of the species, perhaps
not unlike the person who patronized every
Starbucks coffee shop in the world to set a record.

This book is about Chilton’s visits to museums
scattered through Europe and North America to
view specimens of Labrador Ducks. The opposite
of a work of .scholarship, it is very much a
travelogue, and a tedious one at that. From the
author’s web site we learn that Chilton wants
people to believe that he is a bon vivant engaged

in “outrageous adventures,” which in this case
means going to a distant city, measuring a
“stuffed duck,” and then wandering about town
in search of alcohol and vegetarian food (to me,
the thought of eating tofu in Paris and Vienna is
more depressing than the fate of the Labrador
Duck). The title, of course, is a gimmick, the only
curse being the book itself. Nor does the subtitle
make much sense considering that the Labrador
Duck is long past the edge of extinction.

Chilton’s background and attitude did little to
prepare him for museum work, as he plainly
knows next to nothing about collecting or
preparing birds. The specimens he looked at met
their ends by being “blasted,” “blown to
kingdom come,” or shot full of “bullets,” which,
if true, would have left little for him to examine.
Although he includes brief instructions on prep-
aration techniques taken from a museum pam-
phlet, he seems not to understand the process and
refers more than once to the incision through
which the “guts were pulled out.” Specimens are
invariably “stuffed,” a word generally avoided
by museum ornithologists but apparently deliber-
ately overused by Chilton to convey his poorly
disguised disdain for museums as places full of
“stuffed” dead things. Not being knowledgeable
of the history of museum ornithology, Chilton is
bound to have overlooked facts that would be
evident to someone with more experience. What
else could one expect from one for whom the
name Verreaux carried no significance? We also
read (page 304) that: “In November of 1844,
Colonel Nicolas Pike shot a drake Labrador Duck
at the mouth of the Ipswich River at the south end
of Plum Island, New York. History doesn’t say
whether or not the duck was doing anything to
provoke the colonel. Perhaps Pike just really,
really hated ducks”. Not only is this gratuitously
stupid, but Chilton shows no evidence of recog-
nizing the identity of Nicolas Pike nor of Pike’s
connection with other extinct birds [Subtropical
Rambles in the Land of the Aphanapteryx, 1873).

Chilton provides a brief description, usually
only a paragraph, mostly detailing defects, of the
various specimens of Labrador Duck. In a few
instances he appears to have unearthed some new

196

ORNITHOLOGICAL LG ERATLJRi-:

197

information on the history of a given specimen,
which is commendable. But most of the book is
simply an ego-indulging procession of irrelevan-
cies. Why do we need to know that he was a breech
birth, a nervous and obsessive child, and that on at
least four occasions in his duck travels he shared a
room, or even a bed with various young women
who were not his wife'.^ Chiiton seems to regard
himself as quite a humorist, but his humor is so
tiresomely puerile as tei make George Gobel and
Dave Barry seem like genuine wits. Pick any
pejorative adjective that describes annoying prose
and it will apply somewhere in this book; arrogant,
facetious, facile, flippant, glib, precious, sopho-
moric, and snotty. Throughout, Chilton manages to
be condescending or even downright insulting to
individuals, institutions, and entire countries. I
daresay that following the appearance of this book
he will be unwelcome in many places where he was
once treated with courtesy and respect.

Errors large and small throughout the book call
into question the accuracy of almost anything
Chilton writes. For example, the Portuguese word
for “cultivator” is lavrador, not llavrador
(incidentally, Chilton seems to revel in his
ignorance of languages). Some older bird speci-
mens were preserved with arsenic, not cyanide.
There are no sea lions, large or small, around
Grand Manan Island or anywhere else in the
North Atlantic. The national mall in Washington
does not run just from the Capitol to the
Washington Monument, which latter he has
apparently confused with the Lincoln Memorial.
It seems intended that the book not be consulted
again after its first reading, as there is no table of
contents, no index, and no bibliography.

As partial recompense for inflicting this
fulsome blot on the literature of extinct birds,
Chilton owes ornithology and the staffs of all the
museums he visited an exceptionally scholarly
and insightful monograph on the Labrador Duck.
Considering that he was ill equipped to undertake
such an endeavor in the first place, I doubt that he
will ever be able to make satisfactory repayment
of his debt. — STORRS L. OLSON, Curator,
Division of Birds, National Museum of Natural
History, Smithsonian Institution, Washington,
D.C. 20560, USA; e-mail: olsons@si.edu

BIRD BANDING IN NORTH AMERICA:
THE FIRST HUNDRED YEARS. Edited by
Jerome A. Jackson, William E. Davis Jr., and

John Tautin. Nuttall Ornithological Club, Cam-
bridge, Massachusetts, USA, 2008; ix and 280
pages, 62 figures and photographs. ISBN: 1-
877973-45-9. $40.00 (cloth).— This volume brings
together contributions presented at a symposium
which occurred at the North American Ornitho-
logical Conference in New Orleans in 2002. The
symposium commemorated 100 years of scientific
bird banding in North America, the birth of which
is designated by the banding of Black-crowned
Night-herons (Nycticorax nycticorax) by the Smith-
sonian Institution’s Paul Bartsch in 1902. The 12
chapters trace the trajectory of bird marking from
modest but enthusiastic beginnings through the
wide variety of scientific pursuits that have used
this important ornithological tool.

The first three chapters provide historical
perspective. Chapter 1 gives the early history of
bird banding in North America. It starts with a bit
of background on marking birds in the 1 6th
century in the Old World, and quickly settles in
North America, relating the well-known story of
John James Audubon’s 1804 experiment of
marking Eastern Phoebes (Sayornis phoehe) with
silver wire. This is followed by an interesting
overview of the factors that led to the popularity
of nature study in general and bird marking in
particular in the early 20"’ century - including
such disparate elements as increased leisure time
due to the availability of electricity and the
sudden accessibility of aluminum. The chapter
goes on to the more specific underpinnings: the
genesis of organized banding, founding of the
American Bird Banding Association and the first
banding stations, and finally the federal govern-
ment’s role in recognizing and controlling band-
ing in the United States

Chapter 2 discusses the role of banding
organizations in education and training of band-
ers; the distribution of data, techniques, and
materials; and in encouraging regional coopera-
tion. The organizations mentioned include nation-
al (including the present North American Banding
Council) and regional groups, as well as major
bird observatories and banding station networks.
They are arranged by type with a brief summary
of each group's history and function.

Chapter 3 focuses on the history of the Bird
Banding Laboratory (BBL). The BBL wasn t
formally designated as such until 1961, but this
chapter begins in 1920, when Frederick Lincoln
was charged by the USDA Bureau ot Biological
Survey with organizing their “banding office.”

198

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. /, March 2010

He remained in this position until 1946 and is
noted as the true founder of the modern bird
banding program. The chapter proceeds to tell the
history of the BBL by decade, ending in 2002,
which coincided with the retirement of the BBL’s
sixth Chief and chapter author John Tautin.

The rest of the book provides overviews of the
role of bird banding in various research areas,
focusing on particular instances where banding
has made important scientific contributions. These
include population monitoring and ecology, rare
species recovery, migratory game bird and
waterbird con.servation, avian ecotoxicology and
disease, habitat use, and behavioral research. The
chapters overall make a fine case for the value of
bird banding, even though, as several authors
point out, the original intent was focused
primarily on bird movements. Most chapters
conclude with recent developments and future
directions in each area, and how banding might fit
into these new horizons.

The chapters in this compendium vary in form
from chronologies and narratives to something
more resembling typical scientific style. Every
chapter ended with literature cited, a convention I
prefer over paging to the end of the entire volume
to look up a reference. I wasn’t able to discern
whether there was any particular order to the
chapters past the first three, which dealt with
history. The volume might have flowed a little
better if the chapters hacj progressed from broad
research themes to more specific studies, or
perhaps followed some chronological order. I
found only one major editing gaff. In Chapter 5,
a concluding paragraph was printed twice, once
at the end of a series of species accounts, and
again three paragraphs later at the end of the
chapter.

There are 62 black-and-white figures in the
book, mostly photographs. Many of them are
“mug shots,’’ generally of historical figures or
researchers whose work is featured in the chapter.

I often don’t notice the.se things, but in this
volume I could not help being struck that over half
of the photos featured white guys, mostly middle-
aged. There were a few women, and no people of
color. Are bird banding and ornithology really that
devoid of ethnic, racial, and gender diversity?
Yikes. We have something to work toward in the
next hundred years. — JULIE A. CRAVES,
Rouge River Bird Observatory, University of
Michigan-Dearborn, Dearborn, MI 48128,
USA; e-mail: )craves@umd. umich.edu

INGESTION OE LEAD FROM SPENT AM-
MUNITION; IMPLICATIONS FOR WILDLIFE
AND HUMANS. By Richard T. Watson, Mark
Fuller, Mark Pokras, and W. Grainger Hunt. The
Peregrine Fund, Boise, Idaho, USA. 2009: 383
pages. ISBN: 0-9619839-5-7. $ None shown,
(paper). — As a rule, conference proceedings are
uninspiring tomes. The ugly stepchildren of the
academic publication family; their contents are
often limited to progress reports of ongoing
research, literature reviews, and the results of
short-term Master’s Degree research projects.
There are some conferences, however, for which
a published proceedings is not only appropriate,
but also provides an exceptionally useful resource
for a wide audience. The proceedings of the
conference Ingestion of Lead from Spent Ammu-
nition: Implications for Wildlife and Humans, held
12-15 May 2008 at Boise State University is one
of these. The book fully addresses the ubiquitous
nature of lead toxicity by including voices from
science, medicine, policy, and business. Whereas
it would be difficult to find one scientific joui'nal
that would accept this breadth of viewpoints, the
proceedings format allows for the cross-disciplin-
ai'y compilation of knowledge necessary to put the
lead issue in context.

Everyone knows something about the toxic
effects of lead but few know everything about it.
For example, most Americans know that in the
1970s, awai'eness of the effects of lead on human
health I'esulted in its removal fr'om house paint
and gasoline. Hunter's know that in 1991,
documented adverse effects of lead ingestion on
eagles and water-fowl species resulted in r'egula-
tions outlawing lead shot for waterfowl hunting.
Ingestion of Lead from Spent Ammunition pr'o-
vides infor'ination of which most people ar'e not
aware, such as: how lead is toxic (in a fascinating
paper by Pokras and Kneeland); effects of low-
dose exposure of lead on humans who consume
hunted game; and the magnitude of the pr'oblem in
ter'rns of the amount of lead released into the
environment by outdoor spor'ts (6-10 thousand
rnetr'ic tons annually in the U.S.) or the number of
species known to have incidences of high lead
levels (over 130 species of ver'tebr'ates).

The unique strength of this book is it’s
“con.servation medicine’’ appr'oach which Pokras
and Kneeland (page 7) define as “examin[ing] the
linkages among the health of people, animals, and
the envir'onment’’. Imagine the value to wildlife
r'ehabilitation veter'inar'ians when they tap into the

ORNITHOLOGICAL LITERATURE

199

rich field of lead exposure in human development,
and vice versa; or the public health official who
becomes familiar with the work on retention ol
lead fragments in hunted wildlife. Proof of this
book’s integrated approach to lead is the diversity
of author affiliations: wildlife research centers,
zoos, environmental agencies and non-protits,
human medicine, and state conservation offices,
to name a few.

The book is composed of 53 papers (25 peer-
reviewed, 10 not peer-reviewed, and 18 extended
abstracts) organized in four main sections: review
of lead uptake and toxicosis in humans and
wildlife; lead exposure in humans from spent
ammunition; lead exposure, sources, and toxicosis
in wildlife; and, remediation of lead exposure
from spent ammunition. There are also nine
expert commentaries transcribed from the meeting
and an introduction and conference summary by
the esteemed raptor biologist Ian Newton. Birds,
especially raptors, are the predominant taxonomic
group in the research. Not surprisingly, work on
the California Condor {Gymnogyps californianus)
is well represented. Other bird species addressed
include Golden Eagles {Aquila chrysaetos) and
fish eagles (Haliaeetus spp.), waterfowl, doves,
and other terrestrial birds.

Geographically, most papers concern the Unit-
ed States, but there is good representation from
Europe. In the one paper from South America,
Saggese et al. review lead toxicosis in raptors
from Argentina. In discussing sources of lead in
the environment, they cite the recent and growing
popularity of dove and pigeon hunting in central
Argentina where, absent any limits on the number
of birds killed, hunters regularly discharge over
1,000 cartridges per day, killing or injuring as
many birds. The barbarism of these hunts is,
therefore, compounded by the 1,600 metric tons
of lead released into the local environment each
year and the fact that crippled and dead birds are
left for scavengers (including humans) to con-
sume, lead shot and all.

There are only two minor flaws with the book.
The first is the repetition of basic information in
many of the papers. One would only notice this if
the book was read cover-to-cover, a method tor
which the book was not designed. The second
flaw concerns the 18 entries that are extended
abstracts. Some of these are quite extensive with
figures and tables, etc. while others aren’t
‘extended’ at all. Either way. no one likes to cite
abstracts and when I searched for full papers that

these abstracts should have evolved into by now I
wasn’t able to find any.

In his closing comments for the meeting, Ian
Newton noted the conference showed that lead
from spent ammunition poses a bigger human
health problem than previously recognized. In-
deed, after reading this book, one cannot help but
be alarmed at the widespread and insidious nature
of lead ingestion by humans and wildlife. These
proceedings are a call for action and, unlike many
environmental problems that seem insurmount-
able; the removal of lead from outdoor sporting
equipment is attainable.

The book is published by the non-profit organi-
zation The Peregrine Eund and the entire proceed-
ings are available at http://www.peregrinefund.org/
Lead_conference/. Other than a few color photos
and figures online that are represented in grayscale
in the print edition, there are no substantial
differences between the book and the online
version. The reader’s personal preference is the
best guide in deciding which version to acquire.
Either way, acquire this book and spread the word! I
strongly recommend this book to anyone who wants
to be more knowledgeable about the threats to
wildlife, humans, and the environment posed by the
release of lead into our fields and wetlands. —
JOHN CURNUTT, Regional Wildlife Ecologist,
USDA, Forest Service, Eastern Region, 626 East
Wisconsin Avenue, Milwaukee, WI 53203,
USA; e-mail; jcurnutt(§>fs.fed.us

THE BIRDS OE THE REPUBLIC OE PANA-
MA. PART 5. GAZETTEER AND BIBLIOG-
RAPHY. By Deborah C. Siegel and Stoms L.
Olson. Buteo Books, Shipman, Virginia, USA.
2008: 516 pages plus 1 inset map. ISBN: 978-0-
931130-17-5. $45.00 (hardcover). — Even in these
modern days of hand-held GPS receivers and
Google Earth, the final volume of Alexander
Wetmore’s magnum opus: The Birds of the
Republic of Panama is a welcome addition to
the library of any serious student of neotropical
birds who will want to make room for it along side
the previous four volumes. It is outstanding in its
primary role, as a 2()th century ornithological
gazetteer, but most modern readers will find it
wanting a few 21st century trappings that were
left out and we can hope will be included in some
future format.

The authors, Deborah Siegel and Stoms Olson,
have done meticulous work in providing geo-

200

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. I, March 2010

graphic precision to ambiguous place names,
which are the norm in much of Latin America.
Those who have done field work in these parts are
tamiliar with the simultaneous tendency for
multiple local names for a single place, the
application of a general name to a region too
large to be considered a single biogeographic
point, and the duplication of common names
across distant districts and provinces. For exam-
ple, this volume lists nine unique locations with
the moniker “San Jose” which the authors
dutifully sort out. Siegel and Olson have also
been mindful of the tendency of place names to
change, which is especially problematic for the
areas surrounding the Panama Canal; in many
cases, a Spanish-language place name replaced
earlier English-language names used during the
U.S. administration of the Canal Zone. I recently
reviewed collecting localities for the niglaris
subspecies of Myrmeciza e.xsul mentioned in The
Birds of Panama. Volume 3. Not surprisingly I
was able to find precise locations and geograph-
ical coordinates for all, but 1 should also note that
1 found the same information on Google; the value
of this volume is in the annotations for each
location. For example, many of the place name
descriptions follow a hyperlink format, whereby
other place names of interest relative to the
location in question are referenced. It was in this
way that 1 was able to make sense of the confusion
surrounding Cerros Colorado, Santiago and
Flores; an important endemic bird area in the
Serrania de Tabasca in western Panama. It should
also be noted the authors were careful with
resolving the many unintentional variants of place
names caused by collectors (including me) whose
command of Spanish was less than perfect.

Equally valuable is the detailed annotated
bibliography of Panamanian ornithology. The
span of this bibliography is vast. Here, topics as
diverse as systematics, paleozoography, and
natural history comingle from both recent (up
until ~ 2005) and historical sources. I am
especially appreciative for the effort the authors
undertook to include works not typically found in
searchable data ba.ses such as museum mono-
graphs and Latin American regional journals
published in Spanish. This volume should foster
the inclusion of the.se works into the cited
literature of future studies. Particularly of interest
is the inclusion of the reference for a multitude of
named taxa (genus, species or subspecies) with a
Panamanian type location; perhaps only the

authors know how close to being an exhaustive
list this may be.

Given this attention to detail, I wonder how
much greater an impact this final volume - in
what is largely a 20th century work - could have
had if only more 21st century informatics were
used. As an active collector of birds in Panama,
one of my favorite features of the book is a map
highlighting every collecting location (as well as
the geographic gaps in our collective efforts).
However, the map appears to be an afterthought,
as it is a separate inset to the book. There is no
way to go from a dot on the map to that place
name in the text to learn more about who
collected there and when. Similarly, there is no
way to find all of the place names in any
geographic region, such as the Pearl Islands or
the Darien Province, despite how useful that
would be. I imagine that many of my colleagues
from the gene Jockey cohort of ornithologists will
appreciate the bibliography’s detailed listing of
19'*’ century taxonomic literature when it comes
time to give proper names to resurrected lineages.
However, without a .searchable index for scientific
names, 1 wonder if this resource will be used to its
fullest capability. Providing a digital PDF of the
text would allow for searching on strings such as
province name or latitude and longitude ranges.
Alternatively, perhaps the basics of the place
names (and the point map) could be created as an
internet-based resource; readers could be directed
to the text for the details pertaining to specific
locations. To be fair, other gazetteers of Latin
American ornithology have these problems, but
the fact that this volume was published in 2008
makes their omission more apparent.

It is worth noting the book begins with a short
bibliography of Dr. Wetmore that includes his
portrait and a detailed timeline of his field
expeditions in Panama. This is fitting for the final
volume of Wetmore’s unparalleled work. Panama
has had a much larger role in the development of
neotropical ornithology than could be predicted
by its geographic size. This is in no small part due
to the generations of ornithologists that refer to
well-thumbed pages of The Birds of Panama to
learn detail after detail about the distribution and
natural history of that country’s birds. We are all
indebted for his effort. — MATTHEW J. MIL-
LER, Postdoctoral Fellow in Molecular Evolu-
tion, Smithsonian Tropical Research Institute,
Apartado Postal 0843-03092 Panama, Repiib-
lica de Panama; e-mail: millerma@si.edu

ORNITHOLOGICAL LITERATURE

201

THE MIGRATION OF BIRDS: SEASONS ON
THE WING. By Janice M. Hughes. Firefly Books,
Buffalo, New York, USA. 2009: 208 pages, 76
color photographs, 27 maps and figures. ISBN:
13-978-1-55407-432-7. $40.00 (hardcover with
jacket). — The biological study of migration,
specifically among birds, is an extremely diverse
field of endeavor. Over the past hundred years the
literature is filled with diverse views and studies,
each attempting to explain one particular facet of
migration. The author of this work attempts to
unify the great number of studies contributing to
the understanding of migration.

The book is divided into six broad chapters,
each with its own strengths and shortfalls-as to be
expected in any attempt to cover such a complex
topic. Ten “Profiles” are presented, two per
chapter. These vignettes deal with specific species
or groups of species that exhibit similar migration
patterns or experience similar migration difficul-
ties.

The opening chapter, “Bird Migration through
Human History,” was perhaps my personal
favorite as it covered a great diversity of bird
lore starting with mythological references, passing
through classical references (such as Aristotle and
Pliny the Elder) on to the treatise on falconry by
the Holy Roman Emperor Frederick II in the 13th
Century. At this point we start to see the origins of
a more scientific study of birds that progresses to
the present century.

The second chapter, “The Five Ws of Avian
Migration,” is the fuel for this amble through
migration. It is here that the author introduces the
reader to the varied aspects of bird migration.
Researchers are occasionally and appropriately
mentioned but there is a complete lack of
reference citations. While this is undoubtedly a
style preference for this type of book, citations
would have made this a powerful chapter, instead
of just an interesting one.

The third and fourth chapters, “The Phenom-
enon of Flight” and “Fueling the Migration,”
provide interesting aspects regarding the anatomy
and energetic needs of birds. Again, citations
would have been appreciated.

“Finding the Way,” the fifth chapter meanders
through the various ways that birds are believed to
navigate. Here the reader is exposed to some
classic experiments and studies in navigation. It is
here that the anecdotal story of Cheri Ami should
hold the reader on edge heading towards the
ultimate chapter of the book. This World War I

carrier pigeon (Columha livia) earned “the Croi.x
de Guerre, one of the country’s | France] highest
honors for distinguished acts ol heroism during
combat.”

The final chapter, “Migratory Birds in Peril,”
is really a one page overview followed by three
birds with contemporary problems, a possibly
extinct species (Eskimo Curlew [Niinienius bor-
ealis]) and “An Indiscriminate Killer,” which
discusses the role of lighthouses and radio towers
as man-made migratory hazards.

A glossary is a welcomed addition to any book
of a highly specific nature. The one in this book
covers the basic terms with which most non-
scientists would need help. While it is quite brief,
it is well done and should be of use to the casual
reader. Following the glossary is a section entitled
“Further Reading.” This two page list of 33
books is inadequate for the topic, especially when
considering three of the texts are reprints of
Aristotle, Frederick II, and Pliny.

The writing throughout is well done and makes
for enjoyable reading. The photographs are first-
rate. This attractive combination draws the reader
interested in migration deeper in, looking for more.
But here it stops. The complete lack of references
to the great number of studies that must have been
consulted for this work will frustrate the reader in
efforts to look elsewhere for greater detail.

While the author has attempted to cover the
topic of migration thoroughly, I found it disturb-
ing that Pliny the Elder garnered a citation in the
index, whereas world-renowned migratory sites
such as Point Pelee (Ontario), Whitefish Point
(Michigan), Cape May (New Jersey), Malmo
(Sweden), and Istanbul (Turkey), among others
were lacking. This book is a cursory overview of a
complex topic.

The appeal of this book will not be to the
ornithological community due to the lack of
referencing or detailed accounts of species or
migration sites. It should, however, be in public
libraries where it will have general audience
appeal. Additionally, it will make a fine coffee-
table book for the casual bird-watcher. Overall
this book is best suited to middle school and high
school libraries where young students will find it
an interesting read, providing enough detail to
stimulate interest. With adequate assistance, a
student should be able to wade through the
“Further Reading” section and continue on in
studying the various aspects of migration. —
MICHAEL A. KIELB, Visiting Lecturer,

202

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122. No. I, March 2010

Eastern Michigan University, Department of
Biology, 316 Mark Jefferson, Ypsilanti, MI
48197, USA; e-mail; mkielb@emich.edu

THE EASTERN SCREECH OWL: LIEE
HISTORY, ECOLOGY, AND BEHAVIOR IN
THE SUBURBS AND COUNTRYSIDE. By
Frederick R. Gehlbach. Texas A&M University
Press, College Station, Texas. USA. First Printing
(1994). Second Printing (2008): 320 pages, 2
color, 36 black and white photos, 23 figures, and
27 tables. ISBN: (1994): 13-978-I-60344-I21-6).
ISBN (2008): 10-1-60344-121-2. $24.95 (paper-
back).— I highly recommend reading this study if
there is anything at all you want to know or think
you might want to know about screech-owls.
There is little here that I can think of that might be
gleaned from natural history observations that
might have been left out. The text contains 10
chapters, 10 appendices, 297 notes, an index, and
about 315 references. This is the story of serious
field research and intimate contact with individual
screech-owls, and with their populations in central
Texas. It is a study spanning 41 years, that
included 1,453 banded owls. The author involved
homeowners, birders, school children, and teach-
ers. It is a comparison of suburban versus nearby
rural populations that were studied concuirently.
The bulk of the study, reported in the first printing
of this book (in 1994) was completed between
1976 and 1987. The (2008) printing contains new
findings. Some things had changed: eggs are now
laid 5 days earlier in the season than reported
earlier. Wooded suburban habitat has decreased
26% due to suburban sprawl in the study plots,
and nesting pairs have decreased 21% although
their nesting success is still high.

In Chapter 1, the Introduction, Gehlbach tells
us how, as a youngster, he was first introduced to
screech-owls. He was riding his bicycle in a city
park after a softball game when he saw three
IJedglings perched out of reach on a branch above
him. He “knocked each bird off’ (page 3) with a
softball. Years later the mystery of the owls he
had discovered that day remained in his memory
as he encountered a pair nesting in a “squirrel
box." (page 3). He .soon made detailed ob.serva-
tions and ended up putting up nest boxes and
monitoring owls in several states (Ohio, New
York, Michigan, New Mexico, Arizona, and also
in Mexico). His close watch of the central Texas
population that he reports started later, in 1967. In

the Introduction, he outlines his study approach
and methods of his statistical analyses. In Chapter
2, he delineates the climate and weather, and the
vegetation of his study sites with correlates of the
owls’ choice of nest sites, as derived from nest
boxes, as well as seasonal choice of roost sites.
Chapters 3 and 4 examine food supplies and
predation, and body weight and molt, respective-
ly. The next three chapters concern details of
nesting from eggs and chicks and their care to
roles of males versus females during nesting. All
of the data culminate in the last three chapters that
concern lifetime reproduction and population
structure, and comparisons of rural versus urban
populations.

The wealth of detail is at times overwhelming,
but each chapter has a summaiY page or two of ‘The
Essentials’. One can thus skim through the book to
areas of individual interest and refer back to the
detailed quantifications and statistics in each
chapter. The coverage is far too wide and too
detailed for me to reiterate in a review, and perhaps
it should not be done, because eveiyone will find
their own field of interest represented. As in any
long-teim study, the goal is to try to find repeating
patterns. This is the exciting game of natural
histoiy, but there are also always unanticipated
SLiiprises that go beyond the original goals.
Although the primary stated goal of this study
was to compare rural versus suburban populations-
and a number of significant differences were
docLimented-for me the most fascinating discover-
ies were the unintended. To cheiry-pick a few
perhaps arbitrary examples, I was most intrigued by
documentation of a 9-year population cycle, which
coincides with a lunar periodicity, and occuiTed in
both rural and suburban owls. It has apparently also
been documented for diurnal birds. Other interest-
ing nuggets were that life-long monogamy was the
rule, but sequential and concurrent polygyny
occurred when food was plentiful and nesting
density high. The rufous phase declined more than
the gray pha.se in hard winters but increased
following rainy years. Birds were the primary
vertebrate prey of these sit-and-wait predators, but
insects and retiles were supplemental .items. Blind
snakes were taken into the nest as prey items but
some survived in nests to live in a novel commensal
association with the owls.

Gehlbach’s three major conclusions, (pages
192-193) are: (I) the .screech-owl is “especially
well-suited to coexistence with humans”, (2)
“modern suburbia is quite munificent toward this

ORNITHOLOGICAL LITERATURE

203

species... and |it] can utilize its resources”, and
(3) "the amenity-pre-adaptation connections cer-
tainly modify some ideas about avian ecology.”
Screech-owls are rather small owls and pre-
adapted to live in the suburban environment as
long as there are cavities and food that will satisfy
a catholic diet. That said, the devil is, as almost
always, in the details. And this book has plenty of
them. As Gehlbach mentions himself (page 198):
”1 have learned much and perhaps even more
from Eastern Screech Owls [Megascops but
long-term events like population cycles need more
time...” He continues to monitor the suburban
population and writes (page 200); “Field exper-
iments are possible now that I know the rules of
the game” so the final phase of the investigation
begins. He proposed that in the next few years he
would use the owls in another study, in another
city around a school curriculum to educate
children (some of the owls were so habituated
they would land on persons and one could reach
under them to take eggs and chicks to examine
and measure without disturbing the birds). These
owls are an excellent study animal for children and
the public to get familiar with wildlife, and through
his birds he hopes to continue to promote an urban
blend of nature and culture. Gehlbach’ s practical
directions for future comparative studies will raise
awareness and help promote the provision of habitat
and nesting sites for these charismatic birds, which
are a barometer of some aspects of the urban
environment. — BERND HEINRICH, Department
of Biology, University of Vermont, Burlington,
VT 05401 , USA; e-mail: bemd.heinrich@uvm.edu

BIRDS OF EAST ASIA: CHINA, TAIWAN,
KOREA, JAPAN, AND RUSSIA. By Mark
Brazil. Princeton University Press, Princeton,
New Jersey, USA. 2009: 528 pages, 236 color
plates, 2 pages line drawings, 2 maps, 950
distribution maps, family key, 2 appendices,
index. ISBN: 978-0-691-13926-5. $39.95 (paper);
ISBN: 978-0-691-13925-8. $79.50 (cloth).—

When I first visited Japan, Korea, and Hong Kong
in 1960 aboard the USS Helena (a heavy cruiser) as
a freshly minted ensign in the U.S. Navy, there were
no field guides covering any part of Asia. My only
guide to the area was Keisuke Kobayashi’s ""Birds
of Japan in Natural Colors."' Although helpful
because of its excellent illustrations, it was difficult
to use as the text was in Japanese. Over the ensuing
years, numerous books on the birds of eastern Asia

were published, including guides to the birds ol
China, Taiwan, Japan, and Korea (but none lor
northeastern Russia). All of these books are useful
and expand the knowledge of the birds in those
areas. However, each has their drawbacks, one ol
the main ones being that they treat only the birds
already known to have been found in the area
covered. The gap that remained was a modern field
guide that tied all of these areas together in a broad
overview.

This new guide fills that void very well. It
covers all the birds known to have occun'ed in
northeastern Asia roughly ea.st of longitude 116°
E, thus all the coastal Chinese provinces north
from Fujian (opposite Taiwan) to the Arctic
Ocean and east to the Bering Sea, including
northeastern Russia, Japan, Korea, Taiwan, and
their satellite islands. Further, it is a competently
executed field identification guide lavishly illus-
trated in the modem fashion with the text for each
species facing its plate.

The use of “East” Asia in the title is both
misleading and inaccurate as the guide is
concerned only with “northeastern” Asia. None
of southeastern Asia is covered. The brief
introduction of 8 pages contains: Aims of this
guide. Geographical scope. Taxonomy, Nomen-
clature, Bird identification. Bird habitats. Migra-
tion, Vagrancy, and How to use this book (which
includes 2 pages of line drawings illustrating
avian topography and terminology). This is
followed by a 16-page key to the families with a
color thumbnail portrait of a representative
species from each family. The brief text for each
family discusses general habits of the family
members which will help place an unidentified
bird in its family.

The bibliography lists only 16 titles. Since a
work such as this requires use of hundreds of
references, listing only a few seems pointless.
However, the entire reference list (and other
information about the book) can be found at:
htlp://sites. google.com/site/birdsofeastasia. Appen-
dix 1 is a useful table giving the status in five broad
areas (Chinese coastal and northeastern provinces.
Taiwan, Korea, Japan, and northeastern Russia) of
each of the species recorded in those areas.
Appendix 2 is a list of 46 species considered likely
to occur eventually in northeastern Asia.

All 969 species known to have occurred in
northeastern Asia are illustrated in color paintings
on 236 color plates. An additional 16 species that
are expected to turn up are illustrated in the plates.

204

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 1. March 2010

as well as eight extinct species. Each plate
displays 2-8 species with 1-1 1 images/species.
The entire text for each species is opposite its
plate and consists of a small (2.2 X 2.1 cm) range
map in seven colors, length, wingspan and weight,
distribution, habitat and habits, identification, bare
parts, and vocalizations. Some have taxonomic
notes or alternate English names. On the same line
with the title ‘English name’ is a useful code
indicating in which of the five geographical areas
covered the bird has occurred. By using tiny print, a
large amount of information was crammed into the
text for each species. However, for some complex
groups such as large gulls and diurnal birds of prey,
the book’s format is crippling as it does not allow a
sufficiently long text. The taxonomy throughout is
up to date and forward-looking.

The color plates were executed by 14 artists,
resulting in uneven quality, most in the good to
very good level with some in the excellent to
superb range. With only a few exceptions, their
goal of enabling identification is attained. Fe-
males and immatures are illustrated where they
differ from the adult male, as are the more
different subspecies, offering mostly complete
coverage of plumage variation. All individual
images are identified by age, male or female, and
often subspecies. Where identification character-
istics are revealed only or best in flight, inserts of
birds in Hight are given. The artists responsible
for each plate are listed only in the copyrights
section on the backside of the title page. 1 believe
the name of the artist responsible for each plate
should be displayed prominently either on the
plate itself or on its facing page. Some of the
plates 1 found most satisfying are: Alan Harris’s
diurnal raptors, Brian Small’s bush-warblers
through Acrocephaliis warblers, and Ren Hath-
away’s large thrushes. 1 am uncomfortable with
the brush strokes in many of Dave Nurney’s birds,
as well as the odd postures and proportions and
too-large heads of some of his birds. Some
comments by plate follow.

Plate 22. While the text coirectly notes the
crown and nape of the adult Pacific Loon (Diver)
(Gavia pacifica) is noticeably paler than the
Black-throated Loon (Diver) {G. airtica), the
plate has this reversed and the difference dimin-
ished, likely due to using specimens rather than
field experience, as the contrast is accentuated in
the brighter light in the field.

Plate 41. The underwing pattern of both adult
and immature Spot-billed Pelican {Felecamts

philippensis) is inaccurate. It should have a rather
banded appearance with white greater and median
underwing coverts contrasting with vinous (adult)
or dusky brown (immature) lesser underwing
coverts and dusky brown (looking paler or darker
depending on light angle) primaries and second-
aries. The outer primaries are blackish.

Plate 46. Adult and immature labels are
switched on Eastern Osprey (Pandion cristatiis).

Plate 48. I had the privilege of seeing an adult
of the "dugeP' subspecies (or moiph) of Steller’s
Sea Eagle {Haliaeetus pelagicus) in South Korea
on 3 January 1962 and aver this entirely black
eagle with its huge yellow bill and white tail could
not possibly be mistaken for anything else, contra
the text.

Plate 52. While the crest of the Crested
Goshawk (Accipiter oivirgatus) in the wind could
possibly look like the plate rendition, it normally
lies flat on the nape with the tip slightly above the
nape (perhaps 2 mm.) and just barely visible.

Plate 54. The complexity of the plumages of the
buzzards (Biiteo) can only be hinted at in the
space available here, e.g., the individual tail
feathers of most Upland Buzzards {B. hemilasiiis)
are white in the center and darker on the edges so
that the tail looks mostly white when spread and
darker when closed.

Plate 59. While the flight picture of Ruddy-
breasted Crake {Porzana fusca) accurately depicts
this bird’s “jizz”, the walking bird is both badly
misshapen and too large (should be smaller than
Band-bellied Crake [P. paykuUii]).

Plate 65. The specific name of European
Golden Plover (Pluvicdis apricaria) is misspelled
{aricaria rather than apricaria).

Plate 69. Neither mentioned in the text nor
pictured properly is the fact the white flank feathers
of the Greater Painted Snipe (Rostratida henglia-
lensis) in flight curve up and over the sides of the
base of the tail to give a distinctive Ruff-like pattern.

Plate 7 1 . The width of the white band on the
tips of the secondaries of the Common Snipe
{Gallinago gallinago) and Wilson’s Snipe (G.
delicata) are different and incorrectly depicted
(switched) in the plate (correct in text).

Plate 90. Missing in text is the fact that Relict
Gull (Ichthyaetus relictus) often forages along
shore like a shorebird.

Plate 104. The Marbled Murrelet (Brachyram-
plius marmoratus) is pictured and listed as a
species that might turn up in northeastern Asia.
However, there is a record for Idlidlya Island in

ORNITHOLOGICAL LITERATURE

205

KoUichia Bay on the northern Chukotka Peninsula
(Dement’ev et al. 1951, Birds of the Soviet Union).

Plate 1 15. Eastern Grass Owl {Tyto longimeni-
hris) text and plates depict only huffy phase.
White phase has white underparts and white
upperside of tail with narrow black bars.

Plate 120. The text account of Brown Hawk-
Owl (Nino.x sentidata) is a bit muddled. My 2002
paper (Bulletin B.O.C. 122: 250-257) on this
species complex showed the song of the migratory
northern race japonica is the same throughout its
range and different from the resident races in
southern Asia. I recommended splitting them
based on vocalizations as well as some moipho-
metric differences. The resident subspecies totogo
(southern Japan and Taiwan) is close to japonica
both vocally and morphometrically with a weak
potential for a split. Brazil chose to split the
japonica subspecies into eastern japonica and
western florensis and suggests that only japonica
might be separate from scutulata, which doesn’t
make sense (japonica and florensis have identical
vocalizations and similar morphometries, which
are consistently distinct from the subspecies of
scutulata). The English name I suggested in my
paper, Northern Boobook, is mentioned, but the
paper is not listed in the online bibliography.

Plates 127, 130. The colors of Rufous (Celeus
hrachyurus). Pale-headed (Gecinulus grantia),
and Bay (Blythipicus pyrrhotis) woodpeckers are
very badly off.

Plate 133. Heads of Tiger Shrikes (Lanins
tigrinus) are much too big.

It should be noted that any new work with as
broad a sweep as this one, and its consequent
massive number of data points, inevitably has
short-comings. That this one has so tew is a
tribute to the care and expertise applied to the
project by the author. It is remarkable indeed that
the book is so far along the route to perfection on
its first try.

This guide covers its area quite thoroughly,
enough so, that I would be quite comfortable
catrying it as my only guide in its geographic
range. Further, it will be quite useful in the
Philippines and Indonesia. Birders throughout
Eurasia and North America will find it helpful
for identifying strays. 1 highly recommend this
book to anyone interested in Asian birds and
expect to be using it in my Asian jaunts. — BEN F.
KING, Ornithology Department, American
Museum of Natural History, Central Park West
at 79th Street, New York, NY 10024, USA;
e-mail; kingbirdtours@earthlink.net

THE WILSON JOURNAL OL ORNITHOLOGY

Editor CLAIT E. BRAUN

5572 North Ventana Vista Road
Tucson, AZ 85750-7204
E-mail: TWILSONJO@comcast.net

Editorial NANCY J. K. BRAUN
Assistant

Editorial Board RICHARD C. BANKS

JACK CLINTON EITNIEAR
SARA J. OYLER-McCANCE
LESLIE A. ROBB

Review Editor ROBERT B. PAYNE

1 306 Granger Avenue
Ann Arbor. MI 48104, USA
E-mail: rbpayne@umich.edu

GUIDELINES FOR AUTHORS

Please consult the detailed ‘'Guidelines for Authors” found on the Wilson Ornithological Society web site
(http://www.wilsonsociety.org). All manuscript submissions and revisions should be sent to Clait E. Braun,
Editor, The Wilson Journal of Ornithology, 5572 North Ventana Vista Road, Tucson, AZ 85750-7204, USA.
The Wilson Journal of Ornithology office and fax telephone number is (520) 529-0365. The e-mail address is
TWilsonJO@comcast.net

NOTICE OF CHANGE OF ADDRESS

Notify the Society immediately if your address changes. Send your complete new address to Ornithological
Societies of North America, 5400 Bosque Boulevard, Suite 680, Waco, TX 76710.

The permanent mailing address of the Wilson Ornithological Society is: 7oThe Museum of Zoology, The
University of Michigan, Ann Arbor, MI 48109, USA. Persons having business with any of the officers may
address them at their various addresses given on the inside of the front cover, and all matters pertaining to the
journal should be sent directly to the Editor.

MEMBERSHIP INQUIRIES

Membership inquiries should be sent to Timothy J. O’Connell, Department of Natural Resources Ecology and
Management, Oklahoma State University, 240 AG Hall, Stillwater, OK 74078; e-mail: oconnet@ okstate.edu

THE JOSSELYN VAN TYNE MEMORIAL LIBRARY

The Josselyn Van Tyne Memorial Library of the Wilson Ornithological Society, housed in the University of
Michigan Museum of Zoology, was e.stabli.shed in concurrence with the University of Michigan in 1930. Until
1947 the Library was maintained entirely by gifts and bequests of books, reprints, and ornithological magazines
from members and friends of tbe Society. Two members have generously established a fund for the purchase
of new books; members and friends are invited to maintain the fund by regular contribution. The fund is
administered by the Library Committee. Jerome A. Jackson, Florida Gulf Coast Univeristy, is Chairman of the
Committee. The Library currently receives over 200 periodicals as gifts and in exchange for The Wilson Journal
of Ornithology. For information on the Library and our holdings, see the Society’s web page at http://
www.wilsonsociety.org. With the usual exception of rare books, any item in the Library may be borrowed by
members of the Society and will be sent prepaid (by the University of Michigan) to any address in the United
States, its pos.sessions, or Canada. Return postage is paid by the borrower. Inquiries and requests by borrowers,
as well as gifts of books, pamphlets, reprints, and magazines, should be addressed to: Josselyn Van Tyne
Memorial Library, Museum of Zoology, The University of Michigan, 1109 Geddes Avenue, Ann Arbor, MI
48109-1079, USA. Contributions to the New Book Fund should be sent to the Treasurer.

This issue of The Wilson Journal of Ornithology was published on 1 March 2010.

Continued from outside back cover

139 Home-range size and site tenacity ol overwintering Le Conte’s Sparrows in a fire managed prairie
Heather Q. Baldivin, Clinton W. Jeske, Melissa A. Powell, Paul C. Chadwick, and Wylie C. Barrow Jr.

146 Paternal song complexity predicts offspring sex ratios close to Hedging, but not hatching, in Song
Sparrows

Dominique A. Potvin and Elizabeth A. MacDougall-Shackleton

153 Influence of age and season on morphometric measurements ol the Biscutate Swift {Streptoprocne
biscutata)

Mauro Pichorim

Short Communications

160 Cooperative breeding by Red-Headed Woodpeckers
Megan R. Atterberry-Jones and Brian D. Peer

162 Observations on the breeding biology of the Collared Crescentchest {Melanopareia torquata)
Mieko F. Kanegae, Marina Telles, Severino A. Lucena, and Jose Carlos Motta-Junior

165 Effects of a flood on foraging ecology and population dynamics of Swainson’s Warblers
Nicholas M. Anich and Bryan M. Reiley

168 Predator vocalizations affect foraging trade-offs of Northern Cardinals
Mark T. Stanback and Emily M. Powell

173 Observations and predation of a Coral-billed Ground Cuckoo {Carpococcyx renauldi) nest in
northeastern Thailand
Korakoch Pobprasert and Andrew J. Pierce

177 Diet of Chinese Grouse {Tetrastes sewerzowi) during preincubation
Wangjie, Yang Chen, Lu Nan, Fang Yun, and Sun Yue-Hua

181 Observations of a possible foraging tool used by Common Ravens
Paul E Frame

183 Hanging behavior of the Hooded Crow {Corvus cornix)

Mario Melletti and Marzia Mirabile

186 Conspecific brood parasitism by the Dickcissel
Brian D. Peer

188 First report of Olrog’s Gull depredation by sympatric Kelp Gulls
Luciano La Sala and Sergio Martorelli

190 Second and third records of Snares Penguins {Eudyptes robustus) in the Falkland Islands
Laurent Demongin, Maud Poisbleau, Georgina Strange, and LanJ. Strange

194 A new bird species for Costa Rica: Sapphire-throated Hummingbird {Lepidopyga coeruleogularis)
Esteban Biamonte

196 Ornithological Literature
Robert B. Payne, Review Editor

The Wilson Journal of Ornithology

(formerly The Wilson Bulletin)

Volume 122, Number 1 CONTENTS March 2010

Major Articles

1 Demography of a reintroduced population of Evermann’s Rock Ptarmigan in the Aleutian Islands
Robb S. A. Kaler, Steve E. Ebbert, Clait E. Braun, and Brett K Sandercock

15 Apparent survival of breeding Western Sandpipers on the Yukon-Kuskokwim River Delta, Alaska
Matthew Johnson, Daniel R. Ruthrauff, Brian J. McCajfery, Susan M. Haig, and Jejjrey R. Walters

23 Natal philopatry and apparent survival of juvenile Semipalmated Plovers
Erica Nol, Simone Williams, and Brett K Sandercock

29 Breeding biology and nesting success of the Slate-throated Whitestart {Myioborus miniatus) in
Monteverde, Costa Rica
Ronald L. Mumme

39 Breeding biology of Kelp Gulls on the Brazilian coast

Gisele Fires de Mendonga Dantas and Joao Stenghel Morgante

46 Cooperative breeding of the Society Kingfisher {Todiramphus veneratus)

Dylan C. Kesler, Thomas Ghestemme, Emmanuelle Fortier, and Anne Gouni

51 Reproductive success of Burrowing Owls in urban and grassland habitats in southern New Mexico
Daniele Berardelli, Martha ]. Desmond, and Leigh Murray

60 Dietary trends of Barn Owls in an agricultural ecosystem in northern Utah
Carl D. Marti

68 Regional variation in diets of breeding Red-Shouldered Hawks
Brad N. Strobel and Clint W. Boal

75 Effects of Brown-Headed Cowbird parasitism on provisioning rates of Swainson’s Warblers
Sara Fappas, Thomas J. Benson, and James C. Bednarz

82 Effects of fat and lean body mass on migratory landbird stopover duration
Chad L. Seewagen and Christopher G. Guglielmo

88 Seasonal fluctuation of the Orange-winged Amazon at a roosting site in Amazonia
Leiliany Negrao de Moura, Jacques M. E. Vielliard, and Maria Luisa da Silva

95 Costs and benefits of foraging alone or in mixed-species aggregations for Forster’s Terns
Lisa SchreJJler, John K Leiser, and Terry L. Master

102 Abundance and distribution of waterbirds in the Llanos of Venezuela j

Francisco J. Vilella and Guy A. Baldassarre |

1 16 Phenology of six migratory coastal birds in relation to climate change ;

Charles R. Foster, Anthony F. Amos, and Lee A. Fuiman . j

126 Nest-site limitation of secondary cavity-nesting birds in even-age southern pine forests
Karl E. Miller

1 35 House Sparrows associated with reduced Cliff Swallow nesting success
Douglas R. Leasure, Ragupathy Kannan, and Douglas A. James

Continued on inside back cover

Volume 122, Number 2, June 2010

I

Published by the
Wilson Ornithological Society

THE WILSON ORNITHOLOGICAL SOCIETY
FOUNDED 3 DECEMBER 1888

Named after ALEXANDER WILSON, the first American ornithologist.

President — E. Dale Kennedy, Biology Department, Albion College, Albion, MI 49224, USA; e-mail:
dkennedy@albion.edu

First Vice-President — Robert C. Beason, P. O. Box 737, Sandusky, OH 44871, USA; e-mail:
Robert.C.Beason@gmail.com

Second Vice-President — Robert L. Curry, Department of Biology, Villanova University, 800 Lancaster
Avenue, Villanova, PA 19085, USA; e-mail; robert.curry@villanova.edu

Editor — Clait E. Braun, 5572 North Ventana Vista Road, Tucson, AZ 85750, USA; e-mail: TWILSONJO@
comcast.net

Secretary — John A. Smallwood, Department of Biology and Molecular Biology, Montclair State University,
Montclair, NJ 07043, USA; e-mail: smallwoodj@montclair.edu

Treasurer — Melinda M. Clark, 52684 Highland Drive, South Bend, IN 46635, USA; e-mail: MClark@tcservices.biz

Elected Council Members — Robert S. Mulvihill, Timothy J. O’Connell, and Mia R. Revels (terms expire
20 1 0); Jameson F. Chace, Sara R. Morris, and Margaret A. Voss (terms expire 2011); Mary Bomberger
Brown, Mary Garvin, and Mark S. Woodrey (terms expire 2012).

Living Past-Presidents — Pershing B. Hofslund, Douglas A. James, Jerome A. Jackson, Clait E. Braun,
Mary H. Clench, Richard C. Banks, Richard N. Conner, Keith L. Bildstein, Edward H. Burtt Jr., John
C. Kricher, William E. Davis Jr., Charles R. Blem, Doris J. Watt, and James D. Rising.

Membership dues per calendar year are: Regular, $40.00; Student, $20.00; Family, $50.00; Sustaining,
$100.00; Life memberships, $1,000.00 (payable in four installments).

The Wilson Journal of Ornithology is sent to all members not in arrears for dues.

THE WILSON JOURNAL OF ORNITHOLOGY
(formerly The Wilson Bulletin)

THE WILSON JOURNAL OF ORNITHOLOGY (ISSN 1559-4491) is published quarterly in March, June,
September, and December by the Wilson Ornithological Society, 810 East 10th Street, Lawrence, KS 66044-8897. The
subscription price, both in the United States and elsewhere, is $40.00 per year. Periodicals postage paid at Lawrence, KS.
POSTMASTER: Send address changes to OSNA, 5400 Bosque Boulevard, Suite 680, Waco, TX 76710.

All articles and communications for publication should be addressed to the Editor. Exchanges should be addressed
to The Josselyn Van Tyne Memorial Library, Museum of Zoology, Ann Arbor, Ml 48109, USA.

Subscriptions, changes of address, and claims for undelivered copies should be sent to OSNA, 5400 Bosque
Boulevard, Suite 680, Waco, TX 76710, USA. Phone: (254) 399-9636; e-mail: business@osnabirds.org. Back issues or
single copies are available for $12.00 each. Most back issues of the Journal are available and may be ordered from OSNA.
Special prices will be quoted for quantity orders. All issues of the journal published before 2000 are accessible on a free
web site at the University of New Mexico library (http://elibrary.unm.edu/sora/). The site is fully searchable, and full-text
reproduetions of all papers (including illustrations) are available as either PDF or DjVu files.

© Copyright 2010 by the Wilson Ornithological Society
Printed by Allen Press Inc., Lawrence, KS 66044, USA.

COVER: Wilson’s Storm Petrel {Oceanites oceanicus). Illustration by Don Radovich.

@ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).

MCZ

LIBRAf=»Y

JUN 0 3 2010

HARVARO

UNIVERSITY

FRONTISPIECE. We quantified male and female parental care by House Sparrows {Passer domesticiis)
to ascertain food provisioning rates by each parent (page 211) during the 2-week nestling period. Adult
fitness was shaped primarily by the parents’ ability to deliver enormous insect prey items to offspring,
f’arental success (i.e., total recruitment of offspring) was predicted by size and numbers of offspring at
banding (day I 1 ). Water color by Don Radovich.

Wilson Journal

ofO rnithology

Published by the Wilson Ornithological Society

VOL. 122, NO. 2 June 2010 PAGES 207-416

The Wilson Journal of Ornithology 122(2):207-2 16, 2010

NOT THE NICE SPARROW
The 2007 Margaret Morse Nice Lecture

DOUGLAS W. MOCK' - AND P. L. SCHWAGMEYER'

ABSTRACT. — We began our studies of House Sparrow (Passer domesticiis) biparental care in the mid-1990s by
applying the classic Margaret Morse Nice field technique of color-banding individuals. Over the ensuing summers, we
slowly accumulated quantitative provisioning records for 100 broods, even as we commenced a series of experimental
manipulations. Provisioning data showed parental fitness, as expressed by offspring recruitment into the local breeding
population, to be shaped mainly by the adults' ability to deliver enormous insect prey items. It also turned out that
production of robust and competitive fledglings routinely involves losing one or more nestlings (brood reduction in 42% of
1.000 multi-chick families). Recruitment success was compensated for the death of an offspring if the subsequent
reallocation of food enables surviving nest-mates to gain at least 2 g more before fledging. Video samples showed that
parents of day 3 broods favored larger siblings, even though brood reduction typically occurs on —day 4, suggesting that
adults participate actively in promoting some offspring over others. The social dynamic affecting how parents work as a
team during provisioning does not fit the pattern expected if partners negotiate actively with one another, but points more
toward the likelihood of “sealed bids.” Specifically, experimental handicapping of individual parents (tail-weights and
hormone implants) indicates partners operate quite independently during brood-rearing. We are now extending our
experiments into the incubation phase, where parents are probably better-informed about partner activities, thus potentially
able to adjust to fluctuating contributions. Finally, behavioral rules affecting food deliveries seem to differ for females and
males. Females normally increase provisioning as the brood ages, but males do not. However, when broods received
supplemental nutrients, males matched the female upsurge, accelerating their deliveries by 25%. showing they usually work
well below capacity. Received 31 December 2009. Accepted 28 February 2010.

Among the many contributions for which bird
behaviorists are grateful to Margaret Morse Nice,
her pioneering use of color bands is probably the
most universal. Color-banding seems utterly
fundamental to behavioral ornithologists today —
a trifle really — but on 26 March 1928, one pink
celluloid ring provided Mrs. Nice with a means of
recognizing a particular male Song Sparrow

'Department of Zoology. 730 Van Vleet Oval. Univer-
sity of Oklahoma. Norman, OK 73019, USA.
“Corresponding author; e-mail: dmock@ou.edu

(Metospiza meloclia) near her Columbus, Ohio
home. Soon she banded a second male with a
green ring, thereby exposing consistent differ-
ences between the two in song, aggressiveness,
etc. It is hard to overstate what a difference this
innovation has made to behavioral field studies.

Pausing to think about it. much that we take for
granted today was not available to Margaret Nice.
For example, well before today's ornithological
neophyte affixes any color bands, the step of
identifying the local avifauna, or even finding out
which birds live nearby, promotes a trip to the

207

208

THE WILSON JOURNAL OF ORNITHOLOGY ♦ Vol. 122, No. 2. June 2010

bookshelf for a field guide where the only
dilemma is which one to choose. When Margaret
Morse Nice arrived in Oklahoma in 1913 (her
husband Blaine had been hired in what is now the
Zoology Department where we both work), no
bird guides or even checklists existed. Although
she had strong academic credentials (including a
Master’s degree from Clark University), she never
held a position at the University of Oklahoma.
Instead, she stayed physically busy with the
traditional duties of wife and mother, while
fighting boredom and intellectual frustration by
reviving her ornithological interests. Eventually
she wrote Oklahoma’s first state bird book but, by
the time it was published, the family had moved
(with Blaine’s career) to Ohio State, where she
began to focus on avian behavior.

The Oklahoma house that Blaine Nice built
(literally!) at 445 College Avenue is just a block
from where we now live, but nearly everything
else has changed. The streets have been paved,
more than one faculty member owns an automo-
bile (indeed, it seems as if most OU students have
more than one!), and women now comprise a third
of our department’s faculty. Meanwhile, Margaret
Nice’s fields of natural history and ethology have
been infused with theoretical and mathematical
fibers to form modem behavioral ecology (Parker
2006). We all inhabit this new world, but should
pause from time to time and acknowledge the
extraordinary spadework that made our studies
possible.

That said, many Nice-era roots remain clearly
visible, despite being taken for granted. House
Sparrows {Passer domesticus) were already abun-
dant in Norman when she arrived in 1913, but we
have now color-banded .several thousand since
1994. Otherwise, we spend much of our field time
sitting quietly and, hopefully, recording what the
birds would have done in our absence.

The overarching goal of our collaborative
research program is to understand what makes
biparental care evolutionarily stable. Because
cooperation in parenting makes a major contribu-
tion to social monogamy, which predominates in
Class Aves, we .see it as central to understanding
that mating system. House Sparrows are abundant
and willing to use nest boxes, so we study House
Sparrows. This also is consistent with the Margaret
Morse Nice tradition of studying “... what ap-
peared common to .so many” (Nice 1979:263).

We thank the Wilson Ornithological Society for
honoring us with the 2007 Margaret Morse Nice

Medal and offer the following sketch of some
things we have learned about these quintessen-
tially ordinary birds through a series of shared
projects. Our goal is not any sort of autobio-
graphical detail, except as it relates to the
evolution of our sparrow research program, but
to provide an overview of several related studies
that have led us to our current work. We do not
presume to offer this account as a model of how
anyone else should proceed, but students of bird
research need not regard the whole field process
as mysterious. To reduce that mystery, we explain
how we try to extract our guiding questions from
the general picture, use them to design empirical
exercises, and then seek to re-apply the analyzed
results back to the bigger issue from whence we
start. Along the way, it must be confessed, we
often stumble. Space limitation and reader interest
preclude a full chronicling of all our mistakes
(trap designs that were prone to equipment
malfunctions, an attempt to supplement breeding
parents with live mealworms that escaped,
occasional duplications of color band combina-
tions, etc.), but we hope students will forgive
themselves for similar false starts as they press
forward.

We both came of age as scientists more or less
as Darwin’s theory of sexual selection was being
revived after a century of neglect (Campbell
1972) and as several other major pieces of the
evolutionary puzzle, especially inclusive fitness
theory (Hamilton 1964a, b) and optimality
modeling (reviewed in Parker and Maynard Smith
1990), were coming into focus as the foundations
of behavioral ecology. Unlike some previous
awardees of the Margaret Morse Nice Medal,
we had not grown up as bird-watchers per se, but
more generally as animal enthusiasts. Knowing
from childhood that we wanted to have “animals”
in our lives, we had gravitated naturally toward
animal behavior research while in school. Arriv-
ing independently 2 years apart and meeting for
the first time as University of Oklahoma faculty
members, we continued with our separate lines of
study for quite a few years focusing on scramble-
competition polygyny and .sperm competition in
local ground squiirels (e.g., Schwagmeyer 1988,
Foltz and Schwagmeyer 1989, Schwagmeyer and
Foltz 1990), and fatal sibling rivalry and piirent-
off.spring conflict in egrets (e.g.. Mock 1984,
1985; Mock et al. 1987). Realizing that we needed
to learn more about mathematical modeling, we
visited several British universities over the 1983

Mock and Scliwai'nicyer • NOT THE NIC'E SPARROW

209

Christmas break, pausing to get married in
Edinburgh. It was easy to agree on Geoffrey
Parker as our theoretical mentor and collaborator,
as he had discovered sperm competition (e.g.,
Parker 1970) was having a major role in the
development of Evolutionary Game Theory (e.g.,
Maynard Smith and Parker 1976, Parker and
Stuart 1976, Parker 1978, Parker 1983, Parker and
Maynard Smith 1990), and was already deeply
involved in modeling both of our research topics
(e.g., Parker and Macnair 1978; Parker 1982,
1985). We spent fall semester 1984 with him at
the University of Liverpool, taking turns working
on models with Geoff and spending the rest of our
time either pulling numbers out of squin-el and
egret data sets or desperately salvaging our
neglected mathematical training. Over the ensuing
years, we produced two separate series of
publications with Geoff (e.g., Schwagmeyer and
Parker 1987, 1990; Parker et al. 1989; Mock and
Parker 1986, 1997), then began searching for a
field system to explore together. That led first to
two field seasons working on Glaucous-winged
Gulls {Larus glaiicescens) in Puget Sound,
specifically on their pre-hatching acoustic com-
munications (Schwagmeyer et al. 1991), before
we decided to focus on biparental care in House
Sparrows.

CHOICE OE PROBLEM

Biparental care and monogamy are not super-
ficially sexy topics, at least on the surface.
Except for feathered vertebrates, both traits are
exceedingly rare in nature. Thus, the question
naturally arises: what ecological and evolutionary
forces generally promote the evolutionary stabil-
ity of reproductive cooperation between two
adults? Why have birds, in particular, adopted
these as the norm, when other taxa have not?
Einally, although many human cultures (includ-
ing our own) claim both as societal goals, why
are they so elusive? It seems logical to explore
the underlying problems of shared parental care
(seeking what may be broadly general causes) by
analyzing the behavioral dynamics with birds.
Eurthermore, it makes sense to choose a bird that
is common and amenable to (non-destructive)
experimental manipulation of the targeted be-
havioral components.

Biparental investment is most simply ex-
plained as the pattern that emerges when neither
of the sexual partners has anything better to do
with its time. Specifically, if either parent has the

realistic option of alternative pursuits that deliver
higher average fitness returns, those distractions
would likely undermine the joint endeavor
(Maynard Smith 1977). Understanding the fitness
payoffs available to each “player” for its various
alternatives lies at the heart of the matter. In
many taxa (e.g., mammals and plants), dependent
offspring remain physically anchored to the
female, extracting nutrients from her for extend-
ed periods. This arrangement “emancipates” the
other parent to explore additional potential
reproductive opportunities. Similarly, when off-
spring are not physically attached, and especially
when their basic requirements are lessened (e.g.,
by precocial development), it is easy to see how
uniparental care may suffice across a broader
array of taxa. If low-cost offspring are shed soon
after fertilization, the uniparent is less rigidly
pre-ordained and “sex-role reversal” is more
easily evolved (males serving as solo care-givers
in many fishes and shorebirds, for example). In
most birds, combining of egg-laying with altricial
development seems to have created a middle
zone in which more than one adult can make
valuable contributions to offspring welfare. In
short, when each embryo is placed in a neutral
arena (nest) soon after it has been fertilized — and
especially if a hasty departure by the first male
reduces the chance that his sperm will fertilize
the next ovum — the male has inducements for
sticking around at least through the laying period.
Thereafter, if he can further his own selfish
interests better by providing additional services
to his offspring (directly to the eggs and/or by
proxy to his partner), than by seeking additional
mating opportunities, he may remain. Once
hatching begins, the need for provisioning with
vast quantities of exogenous food can extend
male participation on the home front (along with
the continuing parental tasks of defense against
weather and enemies, etc.).

Eollowing this argument, ecological and phy-
logenetic circumstances may provide plausible
reasons why biparental care is evolutionarily
stable across —90% of feathered vertebrates
(Lack 1968, Glutton-Brock 1991), yet rare in all
other life forms. Parental duties .seem sufficiently
extensive to require a team of care-givers and,
perhaps equally important, to preclude either the
option of dumping the whole brood-rearing
exercise on the other (i.e., the “cruel bind” of
Trivers 1972), lest the abandoned parent simply
follow suit and allow the brood to fail.

210

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2. June 2010

To put a face on the overall dynamic, consider
House Sparrow reproduction. After ~11 days of
incubation (some by the male), four or five
hatchlings emerge that will develop from tiny
(1-2 g) naked ectotherms into fully feathered,
warm-blooded, adult-sized (20-25 g) flying
machines within just 2 weeks. During the nestling
period, the two adults must find, capture, and
deliver (typically one per trip) 3,000 to 4,000
unwilling insects to fuel this growth. At the same
time, the parents also provide brooding warmth (at
night and on chilly days), resist certain predators
(and various other intruders, including hostile
conspecifics), and maintain a modicum of nest
sanitation by removing fecal sacs. Although most
House Sparrows are socially monogamous, some
males acquire two partners simultaneously and
many females accept gametes from extra-pair
males; in our Oklahoma population, ~ 17-20% of
young are sired by other males (and 41-45% of
broods contain one or more such extra-pair
offspring) (Whitekiller et al. 2000, Edly-Wright
et al. 2007). Overall, these are pretty ordinary
passerines and, thus, a useful model system.

CHOICE OE SUBJECTS

According to W. L. Dawson (1903:40), “With-
out question the most deplorable event in the
history of American ornithology was the intro-
duction of the English Sparrow.” We may wince
over that sweeping indictment, hastily noting that
the arrival of European hominids was surely more
“deplorable” for native bird species by virtue of
myriad associated ecological nightmares (e.g.,
novel pathogens, deforestation, pollution, and
introductions of numerous destructive exotics,
especially pets, livestock, and all the birds
mentioned in Shakespeare, etc.), but we get
Dawson’s point: this is not the most glamorous
and beloved bird in the New World. We did not
pick it for glamour. As behavioral ecologists, we
chose our topic first and then sought the best local
bird with which to pursue it. Had bluebirds,
chickadees, or wrens piled into our nest boxes, we
might well have studied them. But we were
expecting sparrows and chose two sparsely
occupied (by humans), university-owned tracts
near campus for our study areas, partly because
many House Sparrows were already nesting in the
porous old buildings left from a World War II
naval air station. One of these decaying structures
is still dignified as the university’s Animal
Behavior Laboratory. Our goal was to lure

sparrows from those decrepit buildings to fine,
new nest boxes mounted where we could reach
them conveniently. Surveying possible box de-
signs, we learned that standard bluebird houses
are annoyingly popular with House SpaiTows and
explicit warnings advised us that House Span'ows
like having a perch, prefer being near human-
occupied buildings, etc. We actively catered to
sparrow tastes. In the early seasons, we had
problems with a few avian predators, particularly
Loggerhead Shrikes {Lanins ludovicianus) that
trap-lined our boxes, harvesting sparrow nestlings
simply by sticking their heads through the hole
and seizing the tallest beggars. To thwart this, we
attached a simple 10-cm long corridor of galva-
nized hardware cloth to every box. The sparrows
did not mind entering through these hall-like
structures and mass-predation losses dropped.

TRAPPING, MARKING, AND CENSUSING

We used the usual trapping methods for building
a color-banded population, including walk-in traps
baited with seed and bread, mist nets, and even
sneaking up on brooding males to catch them in
nest boxes. We learned the hard way not to touch
desertion-prone females during incubation. We
applied three color bands (slitting toy Perler Beads
as an inexpensive supply; Hill 1992) and one
aluminum band in unique combinations for each
bird. After much trial-and-error with a variety of in-
box trap designs, we developed a Rube Goldberg
contraption involving a modified wire conndor
(sporting a drop-down door to prevent exit), some
monofilament fishing line, a rat-trap, and remote-
control toy car electronics that empowered us to
catch uncooperative individual adults during their
brief visits to provision chicks (Mock et al. 1997).
The general problem was that House Sparrows are
smart and wary: catching one member of a pair
without disturbing its mate required a trap that was
simultaneously invisible, under our instantaneous
control from considerable distances, and impervi-
ous to strong prairie winds. Once captured, we
carried subjects inside an opaque bag to a car,
processed it there (weighing, banding, etc.), and
then canned it to some stranger’s vehicle for release;
simply releasing it from one’s own car was found to
lead to scolding of that vehicle during subsequent
attempts to observe behavior.

BEHAVIORAL SAMPLING

Reliable quantification of male and female
parental care took advance preparation to opti-

Mock and Schwugnwycr • NOT THE NICE SI^ARROW

211

mize the later research elfort. A chief goal was to
ascertain food provisioning rates by each parent
(Frontispiece) across the 2-week nestling period.
We needed a sampling regime that would generate
representative estimates. Average figures in quite
a few published field studies of avian parental
care are based on rather scant samples (at times
just 15-30 min observations taken on 1-2 days
convenient to the researchers) without any
attention given to whether these capture the
essence of the parents’ activities across daylight
hours, age of the nestlings, shifts in food
availability within and between seasons, etc.
Because we were going to do a lot of monitoring
over several summers, we wanted to know how
much sampling was needed and when this should
be done. For this purpose, we conducted a
preliminary study in which we recorded parental
delivery rates intensively at five nests (covering
all daylight hours and brood ages), then performed
regressions on these over-sized samples to see if
some times of day (and/or combinations thereof)
captured most of the variation. We learned that a
single hour-long sample taken mid-day yielded a
good picture of the whole day’s provisioning
(between-nest comparison), but one taken in mid-
morning did the worst (see Schwagmeyer and
Mock 1997 for details). Combining samples
delivered 91-99.7% (depending on which hours)
of the variation between nests and nearly 80% of
within-nest variation. In the seasons that followed,
we used these findings to guide our deployments,
eventually amassing parenting records for more
than 100 broods based on >13 of these selected
hours per brood. During these observations, we
recorded parents’ arrivals and departures, size of
prey being carried (three categories, scaled
against adult beak), and other activities performed
near the nest (e.g., foraging, singing, defending,
etc.) These data provided a good picture of
parental care in regular sparrow nesting cycles
and in those subjected to experimental manipula-
tions.

BUILDING SUCCESSFUL OFFSPRING

At the heart of any parental care problem lies
the question of what is needed to produce valuable
offspring. There are many components, including
protection from the elements and predators, but
finding and delivering sufficient food to the brood
(provisioning) and then distributing it optimally
once home (allocating) are generally assumed to
supersede most other concems, because of the

phenomenal growth challenge ol oilspring. Once
eggs have hatched, parentally delivered food is
the chief factor limiting growth, survival to
Hedging, and recruitment to breeding age of
House Sparrow offspring; hence it is a major
determinant of parental fitness. Provisioning, as
the central task shared between the parents, was
the initial focus of our exploration into the
stability of biparental care.

If food is limiting, the simplest prediction is
that offspring growth and survival should be
density (brood-size) dependent for a given level of
parental effort. This is usually assessed over the
short span of the nestling period (volant offspring
being necessarily harder to count) as one discrete
exercise. In many field studies, brood size at
hatching is compared with survival and body mass
at fledging (typically defined as banding age, just
before first flight). This requires only a brief
census visit at hatching and measurement of body
size during the banding process. Few field studies
calibrate those relationships with intensive sam-
pling of actual food delivered, presumably
because this feature is tedious and time-consum-
ing. A separate exercise, seldom performed in
tandem with pre-fledging evaluation, extends the
matter from that point by comparing how fledging
production (e.g., size and numbers of banded
chicks) translates into offspring survival to
breeding age. This second step has produced far
greater documentation (for at least 22 avian
species; reviewed by Magrath 1991, Schwag-
meyer and Mock 2008) of the commonsense idea
that parents producing more and larger fledglings
should consequently achieve greater recruitment
of breeding offspring.

We were committed to doing the necessary
observations of provisioning, as an essential part
of analyzing biparental division of labor, and took
the opportunity to combine the two steps (pre-
fledging and post-fledging) on the marked House
SpaiTow population. Mean offspring mass on day
1 I, to our mild initial surpri.se, was not predicted
by total food deliveries, but was strongly
predicted by the rate at which the “enormous”
category of insects (>2 cm total body length,
hereafter, “e-prey”) were provided (Schwag-
meyer and Mock 2008). Once a correction is
made for the much greater volume and mass of e-
prey, it seems likely these grasshoppers, caterpil-
lars, etc. can provide most of the nutrition early in
the season, despite being relatively uncommon
(—14% of all deliveries). Thus, total amount of

212

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2. June 2UI0

tood is important, as expected, but simple counts
ot parental food trips — as might be recorded by
automated devices — would have pointed to the
opposite conclusion.

We used reobservations of 100 banded off-
spring from a sample of 1,028 broods to evaluate
how chick numbers and mean mass relate to local
recaiitment. This measure of parental success was
predicted by both size and numbers of offspring at
banding (Schwagmeyer and Mock 2008).

We found that our data from 99 broods
observed for provisioning rates were sufficient
for a direct assessment of how food deliveries
affect recruitment. This connection had been
established before only once, for Long-tailed
Bushtits (Aegithalos caiidatus), a cooperatively-
breeding British bird (MacColl and Hatchwell
2003). Total delivery counts once again failed to
predict survival to breeding age in our sparrows,
but e-prey deliveries did (Schwagmeyer and
Mock 2008). Overall, it seems clear that e-prey
availability is a key determinant of parental
success and these food items have an especially
strong effect on recruiting for earlier broods.

Considering that parents presumably cannot
simultaneously maximize both the number and
sizes of current brood members, much attention
has been given over the years to short-term
adjustments they may make during the period of
offspring dependency. Early mortality of individ-
ual nestlings has the potential to relax competition
for insufficient food and the idea that brood
reduction might be engineered by parents as part
of an adaptive strategy for tracking unpredictable
food availability dates to the classic writings of
David Lack (1947, 1954). The resolution of acute
sibling rivalry in some taxa includes extreme
forms of nestling aggressiveness and even spe-
cialized weaponry, but in many it results from
low-key processes of scramble competition (re-
viewed in Mock and Parker 1997). Passerine
brood reduction falls overwhelmingly into the
latter category and it is hard to learn unambigu-
ously whether the loss of any given brood member
makes a positive contribution to the inclusive
fitness interests of surviving parents and siblings,
an overall negative, or a matter of little impor-
tance one way or the other (O’Connor 1978).

We found that 42% of 1,000 multi-chick House
Sparrow broods experienced partial-brood mor-
tality between hatching and Hedging (Mock et al.
2009) with greater frequency in larger broods, as
one would expect. Most of the.se losses occurred

early in the nestling period (median day 4), before
parental capacities to provision are likely to be
overtaxed. We found parental provisioning (in the
closely observed sample) increased from broods
of three to broods of five, and per-capita food
remained relatively even. Family size was a
strong predictor of whether brood reduction
occurred (provisioning and family size are both
good predictors of the proportion of the brood
surviving to fledge). Once a single-chick mortality
event had occurred, parents did not trim their
provisioning rate, as has been documented for
siblicidal Cattle Egrets (Buhiilcus ibis'. Mock and
Lamey 1991) and Brown Pelicans (Pelecaniis
occidentalis', Ploger 1997); thus, sparrow survi-
vors tended to receive a food bonus in the
aftermath of brood reduction.

We took two further steps to explore whether
parents might favor or disfavor brood reduction.
We affixed tiny video cameras above false
ceilings inside 1 1 nest boxes containing broods
of four and recorded how food was distributed
when nestlings were sufficiently young (3 days)
that allocation could be attributed safely to
parental decisions (i.e., not to nest-mates’ Jock-
eying effectively for positional advantage). The
idea was to see if fully empowered parents show
overt favoritism toward the smallest chick(s) in
each brood by giving runts more food. These
parents skewed allocation more to larger chicks
(Mock et al. 2009), a pattern that persists (and
often increases) after the stronger chicks usurp
increasing control over allocations. Thus, brood
reduction victims tend to be those that were
smallest initially. Second, we analyzed the trade-
off between size and numbers of fledglings on
recruitment to estimate how much additional body
mass of surviving nest-mates is required to
compensate parental fitness for loss of one
offspring. This exercise suggested that one chick’s
death is fully reimbursed if its surviving siblings
fledge about 2 g heavier (e.g., by virtue of the
food the victim would have consumed). Beneficial
reallocation probably occurs often, apparently
when parental control is at its peak (Mock et al.
2009).

The foraging task shared between two sparrow
parents is something of a malleable abstraction,
perhaps roughly estimated (within a range of
limits) when clutch size is first known to both
partners, but then shifting over the course of
events including hatching failure, seasonal ebb
and How of food availability (insect hatches.

Mock (iiicl Sclnccif'incycr • NOT T1 IE NICE SPARROW

213

periods of inclement weather, etc.), and loss of
nestlings.

DO PARENTS NEGOTIATE WITH
EACH OTHER?

The most influential hypothesis for biparental
cooperation outlines a behavioral dynamic be-
tween two monogamous partners as being the fruit
of a negotiation (Chase 1980, Houston and Davies
1985, McNamara et al. 1999). This assumes that
each parent has a generally self-serving level at
which it is willing to work, but must coordinate
with a like-minded partner. The “self-serving”
qualifier reflects that participants are not genetic
kin, so their confederation is held together by
personal fitness interests (cf. inclusive fitness
incentives). Each individual seeks the best deal
possible, despite needing a sexual partner, within
the framework of its joint reproductive enterprise,
while prefering that its partner provide somewhat
more than half of the work. Both partners cannot
realize this preference simultaneously. Just as
union representatives and owners engage in a
series of offers and counter-offers during a labor
dispute (Maynard Smith 1982), male and female
parents are envisioned as adjusting and re-
adjusting their levels of effort, either in evolu-
tionary time or in real time, until each is satisfied
that it cannot get a better deal. This final pair of
positions is, in the parlance of evolutionary game
theory, the ESS (Evolutionarily Stable Strategy)
solution.

One attractive feature of this Negotiation
Model for biparental care is its testable prediction
of partial compensation, the optimal response that
a parent should make if its partner seems to do
less than its share. Eor illustration purposes,
imagine that an idealized pair has a clean 50-50
split of the family-raising labor, then the male
starts gold-bricking by cutting his share down to
30% of the total, leaving a 20% shortfall. The
model reasons the female should not compensate
fully (by making up the whole difference), but
take up only part of the slack. If the system is
working properly, the male’s interest in offspring
welfare should prompt him to increase his
contributions (whereupon she decreases, etc.)
until equilibrium occurs.

Eield tests of the Negotiation Model had begun
to appear in print as we were starting our sparrow
project. The basic experimental approach involves
unilateral handicapping, that is imposing a burden
on one partner to reduce its contributions and

simulating “gold-bricking.” With one parent
slowed, the unencumbered partner’s behavior is
monitored to see if it provides partial compensa-
tion (vs. full compensation or even no compen-
sation, etc.). We followed the example of Wright
and Cuthill (1989) in crimping four small lead
split-shot fishing weights onto the rectrices of the
handicapped parent (male or female), loading
—5-7% of total body mass away from its center
of gravity. After finding that some birds in the
first season had lost these weights, we painted
them with red nail polish in later years, enabling
us to confirm retention through spotting scopes.
Limiting our study pairs to those with clutches of
either four or five eggs, we rotated among the
experimental and control treatments (minimizing
season effects). Our routine was to sample pre-
treatment provisioning behavior for a total of
—5 hrs prior to treatment (typically by day 5 post-
hatching) and to collect another 9-10 hrs of data
after treatment.

Females in control (unmanipulated) pairs typ-
ically increased provisioning throughout the
nestling period, while males leveled off after
about the fifth day (Mock et al. 2005). Our
experimental treatments showed unexpected re-
sponses. When males were handicapped, they had
only a slight decrease in mean delivery rate, but
they reduced their e-prey deliveries significantly;
their mates showed a non-significant increase.
More unexpectedly, at the nests where females
received tail-weights, burdened females showed
little reduction in their food deliveries, but their
partners substantially increased provisioning any-
way. Consequently, the broods received consid-
erably more food than usual. Thus, if the positive
male response was to his mate’s behavior at all. it
could not have been because she was doing less.
More generally, we found little evidence that
parents are responding to each other's provision-
ing rate (as required by the Negotiation Model),
leading to the conclusion that parental effort is
tied more directly to other factors (perhaps brood
condition) and is more like a “sealed bid” with
independent parental decisions (Schwagmeyer et
al. 2002).

We also handicapped male provisioning per-
formance with a second experimental manipula-
tion, this time by surgically inserting small tubes
containing time-release testosterone beneath the
scapular skin, implanting empty tubes for con-
trols. This produced a more dramatic reduction in
male provisioning, as well as a reduction in male

214

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2. June 2010

incubation time (62% lower than controls for
both measures), but a significant female com-
pensatory increase was observed only during
incubation. At that stage of the cycle, testoster-
one-dosed males appeared to reallocate time and
effort to non-parental activities, as they did more
singing and behaved more aggressively (both to
other males and to their own partners) than
control males. These hormone results showed
that male and female workloads are strongly
inter-related when offspring are still inside eggs,
but not after hatching (Schwagmeyer et al. 2005).
One interpretation of this difference considers
that both parents are likely too busy when
seeking and delivering prey items to keep track
of each other’s contributions (their independent
visits to the brood typically last only 1-5 sec and
encounter each other only by chance). In
contrast, most incubation changeovers are face-
to-face and the other changeovers are likely to
carry information about when the partner depart-
ed (from the lingering tell-tale warmth of the
eggs). We are currently exploring the behavioral
dynamics of shared incubation in a new series of
experiments.

WHEN PARENTING COSTS ARE REDUCED

Parents detecting their offspring are developing
ahead of schedule might be expected to reduce
their costly provisioning activities, either to use
the time and effort for immediate reproductive
advantage (especially males that can fertilize
additional females) or for maintaining higher
personal condition to benefit future offspring.
The approach typically used for such load-
lightening studies is to make food readily
available to the experimental pairs (e.g., a feeder
near the nest) and simply assume that a nontrivial
amount of the extra nutrients is passed along to
the offspring. We took a less trusting route by
hand-feeding nestlings with commercial baby bird
food twice a day. Specifically, we suspended
nutrient powder in warm water to form a thin
gruel and squirted it directly down the esophagus
of each brood member through a modified
syringe. Supplemented broods thus received an
extra 25-30% of their estimated Daily Energy
Budget, over and above whatever their parents
supplied. We sampled parental provisioning as
before, averaging ~I5 hrs at each of the 23
experimental broods.

Parents did not abridge their food deliveries to
these supplemented nests (Mock et al. 2005).

Instead, females showed their usual provisioning
increase after day 5, but this time their males
escalated in lock step, outperforming control
males by 25%. This male rate increase did not
come at the expense of food quality (e.g., the
proportion of e-prey in the diet remained con-
stant). Subsidized nestlings, despite this extra
nutrition, neither gained weight more rapidly nor
achieved higher mass by banding age, but they
recruited into the local breeding population at a
somewhat higher rate (11.8%) than matched
controls (4.9%).

Thus, it appears males do not automatically
divert their attention away from the current brood
when parental burdens are lessened, but may
respond positively to stronger offspring. Some
data on begging call intensities showed that our
supplemented broods begged more loudly, sug-
gesting a possible proximate cue underlying
paternal responses, but we found no evidence
these broods differed visibly from normal (i.e., no
mass effects). The asymmetrical responsiveness
of the parents to subsidized broods may stem
primarily from females already operating near
their maximum (a ceiling effect), while males
have extra reserves that normally remain unspent,
or are channeled elsewhere.

EULL CIRCLE

We close this discussion of our research
program where we started with explicit appreci-
ation of our debt to Margaret Morse Nice for
demonstrating the value of color bands. The huge
amount of time and effort we have invested over
the years in building and maintaining study
populations of recognizable individual span'ows
is a direct rellection on its central role in our
work. Of course, numbered metal bands opened
many research areas, but the advantage of being
able to identify individuals without re-capture and
from a distance truly revolutionized behavioral
ornithology. Every project described would have
been either totally impossible or substantially
weaker without that preliminary step. One cannot
even be certain that a given subject has not been
Li.sed already in either the same study (with loss of
statistical independence) or in a previous exper-
imental manipulation (with the risk of introducing
bias from that earlier treatment) without known
individuals. One small celluloid band placed on
one special sparrow in 1928 gave so much to so
many.

Mock am! Scinvai'oicycr • NOT THE NICE SI^ARROW

215

ACKNOWLEDGMENTS

Our research on sparrows has been funded primarily by

the National Science Foundation (1BN-94()8148. IBN-

9982661, and lOS-0843673).

LITERATURE CITED

Campbell, B. (Editor). 1972. Sexual selection and the
descent of man, 1871-1971. Aldine Atherton, Chi-
cago, Illinois, USA.

Chase, I. D. 1980. Cooperative and noncooperative
behavior in animals. American Naturalist 115: 827-
857.

Clutton-Brock, T. H. 1991. The evolution of parental
care. Princeton University Press, Princeton, New
Jersey, USA.

Dawson, W. L. 1903. Birds of Ohio. Wheaton Publishing
Company, Columbus, Ohio, USA

Edly-Wright, C., P. L. Schwagmeyer, P. G. Parker,
AND D. W. Mock. 2007. Genetic similarity of mates,
offspring health and extrapair fertilization in House
Sparrows. Animal Behaviour 73: 367-378.

Foltz, D. W. and P. L. Schwagmeyer. 1989. Sperm
competition in the thirteen-lined ground squirrel:
differential fertilization success under field conditions.
American Naturalist 133: 257-265.

Hamilton, W. D. 1964a. The genetical evolution of social
behaviour. Journal of Theoretical Biology 7:1-16.

Hamilton, W. D. 1964b. The genetical evolution of social
behaviour. Journal of Theoretical Biology 7:17-52.

Hill, G. E. 1992. An inexpensive source of colored leg
bands. Journal of Field Ornithology 63: 408-410.

Houston, A. I. and N. B. Davies, 1985. The evolution of
co-operation and life history in the Dunnock, Prunella
modiilaris. Pages 471—487 in Behavioral ecology:
ecological consequences of adaptive behaviour (R. M.
Sibly and R. H. Smith, Editors). Blackwell, Oxford,
United Kingdom.

Lack, D. 1947. The significance of clutch-size. Parts 1 and
2. Ibis 89:302-352.

Lack, D. 1954. The natural regulation of animal numbers.
Clarendon Press, Oxford, United Kingdom.

Lack, D. 1968. Ecological adaptations for breeding in
birds. Methuen Press, London, United Kingdom.

MacColl, A. D. C. AND B. J. Hatchwell. 2003.
Determinants of lifetime reproductive success in a
cooperative breeder, the Long-tailed Tit Aegithalos
caudatus. Journal of Animal Ecology 73: 1 137-1 148.

Magrath, R. D. 1991. Nestling weight and juvenile
survival in the Blackbird, Tiirdus merula. Journal ol
Animal Ecology 60:335-351.

Maynard Smith, J. 1977. Parental inve.stment: a prospec-
tive analysis. Animal Behaviour 25:1-9.

Maynard Smith, J. 1982. Evolution and the theory ot
games. Cambridge University Press, Cambridge,
United Kingdom.

Maynard Smith, J. and G. A. Parker. 1976. The logic ot
asymmetric contests. Animal Behaviour 24:159-175.

McNamara, J. M., C. E. Ga.sson, and A. I. Houston.
1999. Incorporating rules for responding into evolu-
tionary games. Nature 401:368-371.

Mock. D. W. 1984. Siblicidal aggression and resource
monopolization in birds. Science 225:731-733.

Mock, D.W. 1985. Siblicidal brood reduction: the prey-size
hypothesis. American Naturalist 125:327-343.

Mock, D. W. and T. C. Lamey. 1991. The role of brood
size in regulating egret sibling aggression. American
Naturalist 138: 1015-1026.

Mock, D. W. and G. A. Parker. 1986. Advantages and
disadvantages of ardeid brood reduction. Evolution 40:
459-470.

Mock, D. W. and G. A. Parker. 1997. The evolution of
sibling rivalry. Oxford University Press, Oxford,
United Kingdom.

Mock, D. W., T. C. Lamey, and B. J. Ploger. 1987.
Proximate and ultimate roles of food amount in
regulating egret sibling aggression. Ecology 68:
1760-1772.

Mock, D. W., P. L. Schwagmeyer, and J. A. Gieg. 1997.
A trap design for capturing individual birds at the nest.
Journal of Field Ornithology 70:276-282.

Mock, D. W., P. L. Schwagmeyer, and G. A. Parker.
2005. Male House Sparrows deliver more food to
experimentally subsidized offspring. Animal Behav-
iour 70: 225-236.

Mock, D. W., P. L. Schwagmeyer, and M. B. Dugas.
2009. Parental provisioning and nestling mortality in
House SpaiTows. Animal Behaviour 78:677-684.

Nice, M. M. 1979. Research is a passion with me.
Consolidated Amethyst Communications Inc., Toron-
to, Ontario. Canada.

O'Connor. R. J. 1978. Brood reduction in birds: selection
for infanticide, fratricide, and suicide? Animal Behav-
iour 26: 79-96.

Parker, G. A. 1970. Sperm competition and its evolution-
ary consequences in the insects. Biological Review 45:
525-567.

Parker, G. A. 1978. Selfish genes, evolutionary games,
and the adaptiveness of behaviour. Nature 274: 849-
855.

Parker, G. A. 1982. Why are there so many tiny sperm?
Sperm competition and the maintenance of two sexes.
Journal of Theoretical Biology 96: 281-294.

Parker, G. A. 1983. Arms races in evolution: an ESS to the
opponent-independent costs game. Journal ot Theo-
retical Biological 101: 619-648.

Parker, G. A. 1985. Models of parent-offspring conHict.
V. Effects of the behaviour of the two parents. Animal
Behaviour 33: 519-533.

Parker, G. A. 2006. Behavioural ecology: the science of
natural history. Pages 23-56 in Essays on animal
behaviour: celebrating 50 years of animal behaviour
(J. R. Lucas, and L. W. Simmons. Editors). Elsevier,
Burlington. Massachusetts, USA.

Parker, G. A. and M. R. Macnair. 1978. Models of
parent-offspring conllict. 1. Monogamy. Animal Be-
haviour 26: 97-1 1 1 .

Parker, G. A. and .1. Maynard Smith. 1990. Optimality
theory in evolutionary biology. Nature 348: 27-33.

Parker, G. A. and R. A. Stuart, 1976. Animal behaviour
as a strategy optimizer: evolution of resource assess-

216

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2. June 2010

merit strategies and optimal emigration thresholds.
American Naturalist 110; 1055-1076.

Parker, G. A., D. W. Mock, and T. C. Lamey. 1989. How
selfish should stronger sibs be? American Naturalist
133: 846-868.

Ploger, B. J. 1997. Does brood reduction provide nestling
survivors a food bonus? Animal Behavior 54:1063-
1076.

SCHWAGMEYER, P. L. 1988. Scramble competition polyg-
yny in a social mammal: male mobility and mating
success. American Naturalist 131: 885-892.

SCHWAGMEYER, P. L. AND D. W. FoLTZ. 1990. Factors
affecting the outcome of sperm competition in
thirteen-lined ground squirrels. Animal Behaviour
39: 156-162.

SCHWAGMEYER, P. L. AND D. W. MOCK. 1997. How to
minimize sample sizes while preserving statistical
power. Animal Behaviour 54: 470^74.

SCHWAGMEYER, P. L. AND D. W. MocK. 2008. Parental
provisioning and offspring fitness: size matters.
Animal Behaviour 75: 291-298.

SCHWAGMEYER, P. L. AND G. A. PARKER. 1987. Queuing
for mates in thirteen-lined ground squirrels. Animal
Behaviour 44: 1015-1025.

SCHWAGMEYER, P. L. AND G. A. PARKER. 1990. Male mate
choice as predicted by sperm competition in thirteen-
lined ground squirrels. Nature 348: 62-64.

SCHWAGMEYER, P. L., D. W. MOCK, AND G. A. PARKER.
2002. Biparental care in House Sparrows: negotiation
or sealed bid? Behavioral Ecology 13: 713-721.

SCHWAGMEYER, P. L., H. G. SCHWABL, AND D. W. MOCK.
2005. Dynamics of biparental care in House Sparrows:
hormonal manipulations of parental contributions.
Animal Behaviour 69: 481^88.

SCHWAGMEYER, P. L., D. W. MOCK, T. C. LaMEY, C. S.
Lamey, and M. D. Beecher. 1991. Effects of sibling
contact on hatch timing in an asynchronously hatching
bird. Animal Behaviour 41: 887-894.

Trivers, R. L. 1972. Parental investment and sexual
selection. Pages 136-179 in Sexual selection and the
descent of man, 1871-1971 (B. Campbell, Editor).
Aldine Atherton, Chicago, Illinois, USA.

Whitekiller, R. R., D. F. Westneat, P. L. Schwagmeyer,
and D. W. Mock. 2000. Badge size and extra-pair
fertilization in the House Sparrow. Condor 102:342-348

Wright, J. and 1. Cuthill. 1989. Manipulation of sex
differences in parental care. Behavioral Ecology and
Sociobiology 25: 171-181.

The Wilson Journal of Ornithology 122(2):217 -227, 2010

THE POSTBREEDING MIGRATION OE EARED GREBES

JOSEPH R. JEHL JR.'-' AND ANNETTE E. HENRY-

ABSTRACT. — Eared Grebes {Podiceps nii>ricollis) in autumn make a postbreeding/molt migration from breeding areas
in western North America to hypersaline lakes in the Great Basin. We studied their biology in 2001-2006 during this phase
of the annual migration near Green River, Wyoming, USA where migrants en route to Great Salt Lake, Utah land on
industrial ponds. Most evidently originate in the Prairie Pothole Region of North Dakota. The main movement extends from
late July to mid-October. Migrants arrive almost daily with the cumulative percentage of transiting birds increasing by
about 1% per day. Adult males and females migrate on the same schedule and precede juveniles by 2-3 weeks. Annual
differences in phenology, abundance, age ratio, and wing molt vary with availability of wetland habitats in the main source
area. Data on mass, body composition, energetics, and stomach contents indicate a typical flight involves a direct 2-3 day
non-feeding migration, which is accomplished at night. Grebes are quiescent during the day and do not resume their
migration until 45 min after sunset. We documented two undescribed vocalizations, a short-range contact note and one
associated with departure. The possibility that Eared Grebe productivity, as inferred from studies of migrants through
Wyoming, can provide insight into the status of waterbirds in the source area is worth further investigation. Received 16
March 2009. Accepted 6 December 2009.

The autumnal migration of the Eared Grebe
(Podiceps nigricoUis) in North America occurs in
two phases. Promptly after the breeding season
adults and young undertake a postbreeding
migration from nesting areas in the interior to
molting/staging sites at hypersaline lakes in the
Great Basin (Fig. 1; Storer and Jehl 1985). Nearly
the entire North American population stages at
Mono Lake, California and Great Salt Lake, Utah
from late summer through early winter and then
undergoes a poststaging migration to wintering
areas in Mexico and southern California (Jehl
1988, Cullen et al. 1999).

Major details of grebe biology during the
nonbreeding season are reasonably well estab-
lished (Cullen et al. 1999; Jehl and McKernan
2002; Jehl et al. 2002, 2003; Ellis and Jehl 2003;
Jehl 2007). Most grebes undertake a 800-1,500 km
flight, which apparently occurs without feeding,
to reach staging lakes (Jehl and Yochem 1986,
Boyd et al. 2000). Few can make the trip on a
single flight because migration occurs only at
night. Migrants passing through western Wyo-
ming toward Great Salt Lake must cross large
spans of steppe desert, where rest stops are few
and unreliable. Near the town of Green River
some encounter an apparent oasis in the form ol
large (up to 400 ha) and sterile evaporation ponds
associated with mining a hydrated form of
soda ash (Na2C03) known as trona (Na2C03/

‘ Smithsonian Ornithology, U.S. National Museum ot
Natural History, Washington, D.C. 20560, USA.

^NOAA Fisheries, Southwe.st Fisheries Science Center,
8604 La Jolla Shores Drive, La Jolla, CA 92037, USA.
^Corresponding author; e-mail: grebe5k@verizon.net

NaHC03-2H20). Sodium salts in the supersatu-
rated ponds can encrust feathers and lead to
mortality. The trona industry has developed a
program to reduce potential losses by monitoring
the migration, and capturing and rehabilitating
afflicted birds through the entire migration season.
We took advantage of this exceptional sample to
document Eared Grebe biology and behavior
during a phase of the annual cycle that had been
largely ignored. The data also generated basic
information on age ratios, timing and pattern of
migration, molt patterns, and annual patterns of
production that are essential for understanding
grebe biology but which, for logistical and
sampling reasons, cannot be obtained at the major
fall staging areas (Jehl and Johansson 2002).

METHODS

We studied grebe migration from 2001 through
2006 at the EMC West Vaco and OCI trona ash
plants, which are ~15 km apart and 30 km
northwest of Green River and 220 km east-
northea.st of Great Salt Lake (Fig. 1). Each year
we visited the sites early in the migration season
and instructed staff on Eared Grebe natural history
and procedures for collecting data. The staff
inspect the ponds several times each day starting
about 0700 hrs over the entire fall migration
season, from late June through November, and
record all species of waterbirds encountered
(captured, not captured, found dead). Encrusted
birds, predominantly grebes, are captured as
quickly as possible, transported to a nearby (1-
2 km) treatment facility, washed in warm water to
remove external salts, and allowed to rest and dry

217

218

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2. June 2010

FIG. 1. The major breeding distribution of the Eared Grebe ir
1999). Great Salt Lake, Utah (2) and Mono Lake, California, (3)
passing through Green River, Wyoming (I) probably originate i
Oregon (4) is evidently used by some migrants en route to
Natural History.

in a large fre.shwater tank. In mid-afternoon they
are released locally along the Green River (fresh
water).

Staff collected data on age, wing molt, and
other characteristics as time permitted. Body mass
was measured in early afternoon, after the birds
had been cleaned of salt deposits and thoroughly
dried. Age was recorded as adult (AHY = after
hatch year) or Juvenile (HY = hatch year) based
on iris color; crimson in adults, brown to orange-
brown in juveniles. Birds in their .second year (SY

North America (outlined in black; from Cullen et al.
hold >1 million staging birds in fall. Most migrants
1 eastern North Dakota (cross-hatched). Abert Lake,
staging lakes. Map by Dan Cole, U.S. National Mu.seum of

= hatched in the previous year) have irides that
are orange or pale red and would be classified as
AHY (Storer and Jehl 1985). Adult males and
females were separated by bill length: ^23.5 mm
= male, <23.5 mm = female, a procedure that is
>90% reliable (Jehl et al. 1998). We made ho
consistent effort to sex juveniles because bill
growth may not be completed by onset of
migration. Grebes were weighed on a digital scale
(±1.0 g) after they were cleaned and dried,
typically 4-6 hrs after capture (1400 hrs). We

Jeh! ami Henry • POSTBREEDING MIGRATION OE EARED CJREBES

219

TABLE 1. Eared Grebe migration schedule by age class, 2001-2006, based on birds captured or lound dead during
autumn migration in southwestern Wyoming.

AdulLs Juveniles

Year

Migration span

Day.s

1S% complete

Migration span

Days

75% complete

2001

12 Jul-16 Oct

97

1 Sep

14 Aug-6 Nov

85

22 Sep

2002

5 Jul-12 Oct

100

7 Aug

10 Aug- 16 Oct

68

25 Sep

2003

23 Jiin-28 Sep

98

3 Sep

19 Aug-24 Oct

67

10 Oct

2004

6 Jul-24 Sep

81

19 Aug

20 Aug-1 Nov

84

23 Sep

2005

12 Jul-30 Sep

81

6 Sep

10-Aug-ll Nov

94

22 Sep

2006"

Combined migrulion
Span Days

l2Jul-6Nov 118

5Jul-16 0ct 104

23 Jun-24 Oct 124

6Jul-01Nov 119

12Jul-llNov 123

18Jul-15 0ct 90

" No data by age class.

used color of the primaries, brown in unmolted
grebes, black in newly-molted, to learn whether
wing molt had been completed before grebes left
breeding locations.

We observed grebe behavior on the evaporation
ponds and in the treatment facility on >40 days
(2001-2005), and rehabilitated birds from release
through departure on over 30 occasions. We
recorded vocalizations in the laboratory and in the
field using a Radio Shack Unidirectional micro-
phone (frequency response 0.01-10 kHz), record-
ed sounds directly to computer hard drive, and
analyzed them using ISHMAEL (Mellinger
2001). Body composition and sex identification
of birds found freshly dead (<24 hrs) were
ascertained by dissection (Jehl 1997). Fat content
of migrants was calculated as: fat (g) = 0.76 body
mass (g) ~ 163.54 g (H. 1. Ellis and JRJ, unpubl.
data).

The source of migrants passing through Green
River can only be inferred. Breeding grebes are
scarce in eastern Wyoming and across the High
Plains. Only a few hundred pairs breed in South
Dakota (Drilling 2007), and several thousand in
western Minnesota (Boe 1993). Grebes departing
Great Salt Lake in early winter fly in a straight
line toward wintering areas (JRJ and S. A.
Gauthreaux, unpubl. data), and we assume post-
breeding migrants behave similarly. Thus, we
infer that most Green River migrants originate in
the Prairie Pothole Region of North Dakota,
where tens of thousands breed (Stewart 1975, Igl
et al. 1999). Those originating farther north or
south would likely by-pass the trona ponds. The
distance from the eastern Dakotas is > 1 ,000 km,
requiring a flight of at least 2 days. We examined
synoptic weather maps from the northcentral
U.S. 2 days prior to 19 large arrivals (36-369
individuals) to investigate if weather influenced

the grebes’ decision to migrate. We used a paired
t-test to compare the number of ponds in the
northcentral U.S. to the number of grebes
encountered near Green River.

RESULTS

Phenology. — ^Eared Grebes usually arrived be-
tween dawn and 1000 hrs. This species avoids
flowing water (JRJ and AEH, unpubl. data) and
only rarely did grebes land on the nearby Green
River, which is only 1 km from some ponds (Julie
Lutz and Lan-y Cherny, pers. comm.). Phenology
was similar at both trona plants in all years and
the data were combined. The earliest arrivals
appeared in late June (Table 1). Small numbers
(1-15) arrived almost daily (Fig. 2A) during the
main influx, 1 August-2 October; the cumulative
increase averaged 1.3% per day (Fig. 2B).
An-ivals slowed to 0.5% per day from 3 to 15
October before ceasing in early November. Yearly
patterns were similar despite an occasional pulse
of >100 birds or variation in the total encounters,
and paralleled those documented in 1992-1995
(Jehl and Johansson 2002).

Age and Se.x Ratios. — Adults migrated earlier
than juveniles (Fig. 3), 75% by the first days of
September (Table 1). Those arriving before 1
August were probably nonbreeders or failed
breeders. Juveniles began to appear in mid- August
and, in most years, 75% passed through by 25
September. The sex ratio of adults (2001. 2002.
2005; n = 396 males, 380 females) did not differ
from 1:1 overall or in any individual year. More
juveniles were classilied as females than males
(2001, 28:15; 2002, 67:58; no data for other years),
probably because growth had not finished, inflating
the apparent proportion of the smaller females.

Wing Molt. — Data on wing molt were obtained
in 2001 and 2002. The pattern was similar in both

220

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2. June 20 JO

c

QJ

O

FIG. 2. Phenology of Eared Grebe autumn migration in .southwestern Wyoming, 2001-2005. A. Individual years;
2001-2005, n = total grebes recorded at FMC and OCI plants. B. All years combined.

years. The first migrants with new remiges arrived
in mid-August and the frequency increased as the
season progressed (Fig. 4); by 20 September all
passage adults had molted. The frequency of molt
in breeding areas was greater in 2001 (21% of 206
adults) than in 2002 (7% of 425).

Body Mass, Body Composition, and Stomach
Contents. — Mass did not vary from year to year.
The mean (± SD) was 297 ± 32 g for adults with
males averaging 7.5% heavier than females, and
295 ± 35 g for juveniles (Table 2). There was no
consistent or significant pattern by males and
females, age, or date in the 3 years for which we

have adequate data. Carcasses showed the en-
larged flight muscles and reduced digestive
organs characteristic of birds that had reconfig-
ured body composition prior to migrating. Body
and component masses were greatly reduced in
comparison to migrants elsewhere known to have
been in migration for only one night (Table 3).
Gizzards lacked a pyloric plug and were nearly
devoid of digestible food; the weight of non-
feather debris averaged about 0.4-0. 5 g, most of
which was finely ground and unidentifiable. Eight
of 10 gizzards examined in detail in 2006
contained unidentified insect material — mostly

Jch! and Henry • POSTBREEDING MIGRATION Ol' EARED (IREBES

221

100
80 -I
60
40 -
20 -
0

♦ AHY=222
oHY=199

♦

♦

2001

U)

FIG. 3. Arrival schedule of adult (AHY) and juvenile
(HY) Eared Grebes in southwestern Wyoming. 2001-2005.

fragments of exoskeleton from surface-dwelling
aquatic beetles and alkali flies (Ephydra spp.);
two contained brine shrimp {Anemia spp.) eggs,
four-miscellaneous seeds, three — fragmentary
plant material, three — jaws of nereid worms (/?
= 3, 6, and 200-300), and two— small bits of grit

or sand. Eight also contained feathers, which were
extremely worn in five cases, but freshly ingested
(1-2 days) in three others.

Behavior Diiriny the Day. — Grebes that landed
on the trona ponds remained largely inactive
throughout the day, as is characteristic of birds
stopping over in natural situations. They did not
attempt to drink or feed, nor did they dive, unless
pursued for capture. Typical of grebes, they
preened continually. If salt crystals began to
form, they increased preening efforts, pulling
several feathers through the bill and using
vigorous head-shaking to flick off adhering
crystals. As encrustation increased they moved
closer to shore, often attempting to haul out.
Unimpaired birds remained on the ponds and
evidently departed successfully after dark.

There was little variation in behavior of
rehabilitated birds upon release. They swam
directly to the middle of the Green River and
engaged in comfort movements (bathing, stretch-
ing, preening, adjusting plumage, drinking) for up
to a half-hour; some also made shallow and brief
(8-15 sec) dives, evidently to readjust plumage.
Subsequently they began to investigate their
suiToundings, swimming up- or downstream until
they encountered rapids or other bamers and then
reversed course. They then moved into secluded
sites near shore, out of any cuiTent, and rested or
slept until evening.

Departure. — Sunset is a key time. Grebes began
to become active a few minutes after the sun
dipped below the horizon, when they returned to
mid-river, and began predeparture activities.
These included ritualized preening of the back
and scapulars, one or two short practice flights
(always upstream), and group diving, submerging,
and surfacing in near-unison. As dusk deepened
the birds became quiescent for several minutes.
Then, 40-45 min after sunset, when it was almost
completely dark, they performed a few intention
movements, consi.sting of pumping the head up
and down and giving soft calls. Then, with no
further signals, they flew upstream ~1 m above
the river for about 100 m or so before climbing
and disappearing westward. Small groups left as a
unit but, when relatively large numbers (>15)
were present, llighls could be spread over 3-
10 min. Any that did not leave swam back to
shallow water and rested without feeding until the
next sunset.

We noted zugunruhe in captive grebes twice.
Six adults captured before noon on 17 August

222

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2, June 2010

(/)

"O

'o

FIG. 4. Timing of remige molt of Eared Grebes migrating though southwestern Wyoming, 2001 and 2002. Data
presented in weekly intervals.

2001 were retained overnight for metabolic
studies. They became agitated at 2100 hrs
(46 min after sunset), despite having had no view
of the sky or access to natural light for 9 hrs,
started to vocalize, and swim actively for about
30 min (H. I. Ellis, pers. comm.). Other birds
released earlier that day departed between 2054
and 2112 hrs. In another case, birds in a dark box
being transported to a release site in the early
evening suddenly became restless and started
calling 27 min before sunset. This behavior was
not observed when birds were released earlier in
the day (Julie Lutz, pers. comm.).

Vocalizations. — ^Two vocalizations were record-
ed (Fig. 5). Both were compound sounds, made
up of more than one unit and within the frequency
range (1-5 kHz) of calls associated with breeding,
but their structures were distinct. The “A” call
consisted of three frequency-modulated down-
swept vocalizations in a 1.2 to 1.5 kHz bandwidth.
We heard it as grebes were preparing to depart
and immediately before the actual takeoff. The
“B” call, reminiscent of a soft amphibian-like
grunt, consisted of a patterned series of 5-1 1
individual pulses with a consistent inteipulse
interval of 0.025 sec. Each series was repeated

TABLE 2. Mass (g), by males
southwestern Wyoming, 2001-2005.

and females, and age of migrating Eared Grebes captured

in autumn migration.

Adult niale.s

Adult females

All adults

Juveniles

Year

n

Range (.v ± SD)

n

Range (jf ± SD)

n

Range (.f ± SD)

n

Range (.f ± SD)

2001

83

249^09 (305 ± 29)

75

210-345 (277 ± 23)

158

210-409 (292 ±

30)

191

197-388 (299 ±

35)

2002

223

223-389 (305 ± 30)

210

200-394 (285 ± 30)

424

200-394 (296 ±

31)

125

222-372 (287 ±

31)

2003“

10

228-332 (288 ±

31)

31

218-360 (291 ±

35)

2004“

29

225-367 (304 ±

33)

25

197-348 (290 ±

35)

2005

68

270-396 (316 ± 28)

79

217-390 (297 ± 32)

147

217-396 (306 ±

32)

162

213-398 (297 ±

36)

All years

374

223^09 (307 ± 30)

364

200-394 (286 ± 30)

768

200^09 (297 ±

32)

534

197-398 (295 ±

35)

No duty by categories of males and females.

Jeh! and Henry • POSTliRIIHDINCJ M ICJRA I'lON ()l liARED GRERES

223

TABLE 3. Body composition (g)
August-September 200 1-2006, and

of Eared Grebes found freshly dead at the end of a migration: Green River, Wyoming,
Iron Mountain, California, .lanuary 1993 (from .lehl 1997: Table 1).

Green River, Wyttming

Iron Mountain. California

Description

n

Riinge (X ± St))

n

R:inge (X ± SD)

Mass

25

240-330 (280.2 ± 24.0)

7

312423 (368 ±41.4)

Breast

24

16.0-23.2 (19.7 ± 1.9)

7

21.7-27.0 (24.0 ± 1.9)

Heart

26

2.7-t.7 (3.6 ± 0.4)

7

4.5-5.4 (5.0 ± 0.3)

Intestine

10

7.5-13.6 (9.7 ± 1.7)

7

7.0-12.4 (10.2 ± 1.9)

Leg

10

1 1.1-14.4 (13.1 ± 1.2)

7

14.7-20-1 (17.1 ± 2.0)

Liver

26

6.1-1 1.8 (8.8 ± 1.2)

7

7.8-13.7 (10.6 ± 1.9)

Stomach

26

6.6-15.5 (10.6 ± 2.1)

7

9.4-13.3 (1 1.2 ± 1.3)

Stomach contents

26

0. 1-5.6 (1.2 ± 1.2)

7

1. 6-7.0 (2.7 ± 1.8)

Estimated time in flight

36-60 hrs

9.3 hrs

Fat stores

20.6-86.4 (49.7 ± 17.7)

72.6-151.4 (1 12.9 ± 30.6)

five times within a call. The interval between the
first and second series was 3.0 sec, while the
interval between the 2nd, 3rd, 4th, and 3th series
was 2.0 sec. The fundamental peak frequency
approximated 1.5 kHz with harmonics. More than
50 calls were recorded in 10 min from a tank
holding 28 individuals.

Weather and Migration. — ^The main source of
migrants at Green River is probably the Prairie
Pothole Region of North Dakota. Synoptic
weather maps showed calm conditions prevailed
there at dusk 2 days prior to all major arrivals at
Green River. In all cases, winds were <5 km/hr,
generally from the NE-SE; in 15 cases a cold front
had passed 1-2 days earlier.

DISCUSSION

Eared Grebes migrating through Wyoming en
route to fall .staging areas exhibited behaviors that
had not been documented elsewhere, including
energy-saving sluggishness by day, ziigunriihe,
strict correspondence of departure with onset ot
near-total darkness, and unique vocalizations.

Our finding that adults migrated earlier than
juveniles had been established in breeding areas
and at Mono Lake (Jehl 1988, Cullen et al. 1999),
but the pattern lacked detailed quantification. That
adult males and females migrate on the same
schedule was not unexpected, because adults split
the brood, each providing a similar level of care
for about 2 weeks, and then leave belore the
young can fly (Cullen 1998). Some adults were
known to complete wing molt in breeding areas
(Storer and Jehl 1985, Jehl 1988). The Wyoming
data, supplemented by >5,000 individuals banded
cit Mono Lake over two decades (JRJ, unpubl.
data), indicate that about 11% of adults molt

before migrating, and that frequency of molt
increases as the season progresses and is greater in
years when grebes remain longer in breeding
areas.

Eared Grebes are generally silent during
migration. The two undescribed calls we recorded
were within the frequency range (1-5 kHz) of
calls associated with breeding, but their structures
were distinct (spectrograms in Cullen et al. 1999).
We consider the “B” call a short-range contact
note. The “A” call was associated with departure.
Eared Grebes migrate in the dark, depart staging
areas in flocks, and are susceptible to mid-air
collisions (Jehl 1998). Whether vocalizations or
other sounds are used to minimize that risk
remains to be investigated.

From Breeding Areas to Staging Lakes: Evi-
dence for a Direct, Non-feeding Flight. — ^Eared
Grebes reorganize their body composition before
departing staging and breeding areas (Jehl 1997).
Departure mass approximates 400-450 g, which
decreases to 350-370 g after an overnight flight
(Jehl 1993, 1994, 1997, 1998). Wyoming migrants
averaged —300 g, >50 g lighter, indicating they
had been en route for at least 2 days. Their
stomachs and intestines were small and essentially
empty, confirming a nonfeeding migration and
indicating they had ceased feeding and cleared
their gut several days before departing (Jehl
1997). Meteorological data support earlier find-
ings (Jehl and Johansson 2002) that grebes depart
breeding areas in periods of calm weather. There
is no time for a leisurely migration, as south-
bound adults begin to pass through Green River as
early as late July-the same time that departures
are noted in breeding areas and migrants appear at
the major staging lakes (Jehl 1988, Cullen et al.

224

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2, June 2010

^ I I i I i r

0.1 0,2 0.3 0.4

Time (s)

FIG. 5. Vocalizations of migrating Eared Grebes,
Green River, Wyoming, 1 5 September 2005. A. Type
“A” call (sample rate 20 kHz, 512 FFT). B. Type “B” call
(.sample rate 20 kHz, 512 FFT) showing the entire call,
which consists of five series of pulses. C. Fig. 5B expanded
to show individual pulses.

1999). They do not feed during this movement,
and migrants leave stopover locations the evening
after they arrive. Thus, weather conditions appear
to have little effect once the decision to migrate
has been made.

Energetic.s. — Migrating across arid lands is a
challenge (Jehl 1993, 1994; Cullen et al. 1999;

Jehl et al. 1999, 2003). Grebes leaving Green
River after rehabilitation averaged about 293 g
and would be expected to reach Great Salt Lake
weighing 271 g with 41 g of fat remaining (H. I.
Ellis and JRJ, unpubl. data). This estimate is
based on flight costs of 6 g fat/hr (vs. 7. 1 g/hr for
400^50 g migrants; Jehl 1994) and a flight time
of 3.7 hrs at 60 km/hr (Cullen et al. 1999). The
minimal weight for healthy adults is ~240 g (Jehl
1988, Anderson et al. 2007; JRJ, unpubl. data),
and migrants need to depart Green River at
^262 g, which includes 85% of our sample
(Fig. 6). Laying over 1 day would require
catabolizing at least an additional 17 g of fat
(0.7 g/hr resting rate X 24 hrs; JRJ and AEH,
unpubl. data). Those delayed for 2 days would
deplete nearly all remaining fat and require an
additional stop to rebuild the digestive organs and
refuel before continuing (Hume and Biebach
1996). The rate of weight gain in staging grebes
is 3-4 g/day (Jehl 1988). Consequently, migrants
originating more than a 3-night flight from the
staging areas (—1,500-1,600 km) probably re-
quire a stopover of several weeks or more to
accumulate sufficient fat stores to continue. The
only present candidate locality for a regular
stopover is saline Lake Abert, Oregon where
5,000-7,000 migrants have been reported in
September and early October (Jehl 1988).

Annual Differences. — ^Timing and pattern of
migration differed from year to year. Adults and
young in 2001 passed through steadily in their
respective periods; 75% of adults migrated by 1
September, 21% of adults underwent wing molt
before migrating, and juveniles comprised 47% of
the migrants. In 2002, 75% of adults migrated
3 weeks earlier, only 7% had molted, and
juveniles were few (23%) and concentrated in
the last half of September. We consider 2001 a
year of successful breeding, whereas in 2002 most
adults left too early to have bred successfully, and
the relatively few young evidently resulted from
late nesting. These differences are probably
indicative of habitat loss resulting from the
eastward shift into the Dakotas of the severe
drought that had beset the western U.S. since the
late I99()s (U.S. Drought Monitor maps for 29
May 2001 and 28 May 2002, www.drought.unl.
edu/dm/archive.html). Our view that North Da-
kota is a main source of Green River migrants is
supported by the significant relationship (( =
4.33, P < 0.001) between the number of grebes
encountered near Green River in autumn and the

Jehl and Henry • POSTBREEDING MIGRATION OP EARED GREBES

225

FIG. 6. Mass of Eared Grebes (adults and juveniles combined) captured at Green River, Wyoming, 2001-2005. Arrow
indicates estimated minimum mass needed to reach Great Salt Lake without laying over or making an additional stop.

number of May ponds in the northcentral U.S. by
year (Fig. 7).

Patterns from other years are less clear. There
was an early exodus of adults in 2003, yet nearly
half remained in breeding areas into September;
juveniles migrated much later than in other years,
but comprised a high percentage (65%) of the
population. Evidently conditions were poor early,
but breeders able to take advantage of late-

summer rains fared well, a pattern also inferred
from studies of waterfowl (USDl 2003). Data in
2004 were too few for analysis. There was a
steady passage of adults in 2005 when juveniles
were plentiful (52%), but concentrated late in the
season.

Are the Data Representative? — One might
question whether a sample from industrial ponds
fairly represented the population passing through

3000

2500

C/D

“D

C

0

-§■ 2000

c

CD

C/D

1 1500

OD

05 1000

,Q

E

13

500

0

CM

CO

in

CD

t^

CXD

<J3

o

T-

CM

CO

m

OD

OD

<JD

<JD

03

03

03

o

o

O

o

o

o

O)

CJD

CJD

OD

OD

03

03

03

o

o

O

o

o

o

■r“

T”

T“

T —

T—

CM

CM

CM

CM

CM

CM

FIG. 7. Number of Eared Grebes encountered at trona evaporation ponds in autumn in southwestern Wyoming and
number of spring ponds (in thousands) in the northcentral U.S., 1991-2005. Pond data from USD! (2005).

226

THE WILSON JOURNAL OL ORNITHOLOGY • Voi 122, No. 2. June 2010

the region or, perhaps, was biased in favor of
inexperienced or unfit birds that had dropped out
before they could reach Great Salt Lake. In the
latter case, one would expect juveniles to predom-
inate and average lighter than adults. However,
adults were as numerous as juveniles, body masses
of the two groups were identical, and most birds
had sufficient reserves to reach staging areas. In
addition, our data on migration timing and intensity
parallel those gathered at another series of
industrial ponds near Kemmerer, Wyoming,
100 km west-northwest of Green River (Jehl and
Johansson 2002). We conclude the Green River
sample was derived from healthy birds on a normal
migration that interrupted their flight to seek
shelter at dawn. Our sample was not biased by
variation in timing of the breeding season, because
it was compiled over the entire migration period.
Thus, our encounter rates and age ratios reflect
survivorship to migration in the source area. The
wetland habitats occupied by Eared Grebes are also
used by waterbirds whose breeding success is
difficult to assess (Johnson et al. 2009). The
possibility that data from migrating grebes might
provide an independent index to annual variations
in production of other marsh and water birds in the
source area warrants investigation.

ACKNOWLEDGMENTS

The following plants and personnel associated with the
trona industry made this study possible; FMC WestVaco
(Julie Lutz, Ron McNalley, Simon Lee), FMC Granger
(Brad Kroll), OCI of Wyoming (Larry Cherny), and Solvay
Minerals (Tim Brown, Dolly Potter). We especially
acknowledge Julie Lutz for organizing and coordinating
industry efforts. Shannon Rankin analyzed vocalizations.
Field work in 2003 was supported in part by the U. S. Fish
and Wildlife Service. Our long-term studies on the avifauna
of .saline lakes have been supported by the Los Angeles
Department of Water and Power, Hubbs-SeaWorld Re-
■search Institute, and the National Geographic Society
(Grants 7408-03, 7609-04). We thank Bruce Eichhorst, H.
I. Ellis, S. I. Bond, D. H. John.son. Mark Otto. John Solberg,
Nancy Drilling, L. D. Igl. and Brian Schmidt for additional
help in field work, manuscript preparation, manu.script
review, and acce.ss to unpublished information..

LITERATURE CITED

Ander.son. T. W., M, a. Tiffany, and S. H. Huri.bert.
2007. Stratification sulfide, worms, and decline of the
Eared Grebe (Podiceps nigricolli.s) at the Salton Sea.
California. Lake and Reservoir Management 23:500-
517.

Bob. J. 1993. Colony site selection by Eared Grebes in
Minnesota. Colonial Waterbirds 16:28-38.

Boyd, W. S.. S. D. .Schnfidkr, and S. A. Cui.fen. 2000.

Using radio telemetry to describe the fall migration of
Eared Grebes. Journal of Field Ornithology 71:702-
707.

Cullen, S. A. 1998. Population biology of Eared Grebes in
a naturally fragmented habitat. Thesis. Simon Fraser
University, Burnaby, British Columbia, Canada.

Cullen, S. A., J. R. Jehl, and G. L. Nuechterlein. 1999.
Eared Grebe (Podiceps nigricoHis). The birds of North
America. Number 433.

Drilling, N. 2007. South Dakota Colonial Waterbird
Project. Draft final report. Part. 1 . Rocky Mountain
Bird Observatory, Brighton, Colorado, USA.

Ellis, H. 1. and J. R. Jehl. 2003. Temperature regulation
and the constraints of climate in the Eared Grebe.
Waterbirds 26:275-279.

Hume, I. D. and H. Biebach. 1996. Digestive tract function
in the long-distance migratory Garden Warblers,
Sylvia horin. Journal of Comparative Physiology,
Series B 166:388-395.

iGL, L. D., D. H. Johnson, and H. A. Kantrud. 1999.
Uncommon breeding birds in North Dakota: popula-
tion estimates and frequencies of occurrence. Canadian
Field-Naturalist 113:646-651.

Jehl, J. R. 1988. Biology of the Eared Grebe and Wilson's
Phalarope in the nonbreeding season: a study of
adaptations to saline lakes. Studies in Avian Biology
12.

Jehl, J. R. 1993. Observations on the fall migration of
Eared Grebes, based on evidence from a mass downing
in Utah. Condor 95:470-473.

Jehl, J. R. 1994. Field estimates of energetics in migrating
and downed Black-necked Grebes. Journal of Avian
Biology 25:63-68.

Jehl, J. R. 1997. Cyclical changes in body composition in
the annual cycle and migration of the Eared Grebe
Podiceps nighcoUis. Journal of Avian Biology 28:
132-142.

Jehl, J. R. 1998. Conspecific collision can precipitate
mortality in migrating Eared Grebes. Wilson Bulletin
I 10:409-41 1.

Jehl, J. R. 2007. Why do Eared Grebes leave hypersaline
lakes in autumn? Waterbirds 30:1 12-1 15.

Jehl, J. R. and C. Johansson. 2002. The autumnal
migration of Eared Grebes through southwestern
Wyoming: a key to assessing the size of the North
American population. Western North American Nat-
uralist 62:335-340.

Jehl, J. R. and R. McKernan. 2002. Biology and
migration of Eared Grebes at the Salton Sea.
Hydrobiologia 473:245-253.

Jehl, J. R. and P. K. Yochem, 1986. Movements of Eared
Grebes indicated by banding recoveries. Journal of
Field Ornithology 57:208-212.

Jehl, J. R. H. I. Ellis, and A. E. Henry. 2003. Optimizing
migration in a reluctant and inefficieni flier: the Eared
Grebe. Pages 199-209 in Avian migration (P. Bert-
hold, E. Gwinner, and E. Sonncnschein, Editors).
Springer-Vcrlag. Heidelberg, Germany.

Jehl, J. R., A. E. Henry, and S. I. Bond. 1998. Sexing
Eared Grebes by bill measurements. Colonial Water-
birds 21:98-100.

Jehl and Henry • POSTBREEDING MIGRATION OE EARED GREBES

227

Jehl, J. R., A. E. Henry, and S. I, Bond. 1999. Flying the
gantlet: population characteristics, sampling bias, and
migration routes of Eared Grebes downed in the Utah
desert. Auk 116:178-183.

Jehl, J. R., W. S. Boyd, D. S. Paul, and D. W. Anderson.
2002. Massive collapse rapid rebound: population
dynamics of Eared Grebes {Podiceps nigricollis)
during an ENSO event. Auk 119:1 162-1 166.
Johnson, D. H., J. P. Gibbs, M. Herzog, S. Lor, N. D.
Niemuth, C. A. RiBic, M. Seamans, T. L. Shaffer,
W. G. Shriver, S. V. Stehman, and W. L. Thompson.
2009. A sampling design framework for monitoring
secretive marshbirds. Waterbirds 32:203-215.
Mellinger, D. K. 2001. ISHMAEL 1.0 User’s guide.
NOAA Technical Memorandum OAR PMEL-120.

NOAA/PMEL, 7600 Sand Point Way NE, Seattle,
Washington, USA.

Stewart, R. E. 1975. Breeding birds of North Dakota. Tri-
eollege Center for Environmental Studies, Fargo,
North Dakota, USA.

Storer, R. W. and j. R. Jehl. 1985. Moult patterns and
molt migration in the Black-necked Grebe Podiceps
nigricollis. Ornis Scandinaviea 16:253-260.

U.S. Department of the Interior (USDI). 2003. Water-
fowl population status, 2003. USDI, Fish and Wildlile
Service, Washington, D.C., USA.

U.S. Department of Interior (USDI). 2005. Waterfowl
population status, 2005. USDI, Fish and Wildlife
Service, Washington, D.C., USA. http//migratorybirds.
fws.gov (aecessed 5 February 2007).

The Wilson Jourjuil of Ornithology 122(2):228-235, 2010

POSTFLEDGING AND NATAL DISPERSAL OF CRESTED IBIS IN THE

QINLING MOUNTAINS, CHINA

YU XIAO-PING,"’ XI YONG-MEI,2 LU BAO-ZHONG," LI XIA,^ GONG MING-HAO,'

SHI LIANG,' AND DONG RONG'

ABSTRACT. — Natal dispersal is common among many bird species with females usually dispersing farther than males.
We examined postfledging and natal dispersal ecology of Crested Ibis (Nipponia nippon) in the Qinling Mountains, China
by marking nestlings with either color bands or radio transmitters. Family groups, consisting of juveniles after fledging and
their parents, dispersed progressively farther from nest sites after fledging in a southerly direction with a mean (± SD)
direction of 189.2 ± 46.5° and a mean distance of 20.3 ± 7.0 km. Mean natal dispersal distance of females (9.6 km) was
significantly greater than for males (5.9 km) (/ = 2.14, P = 0.03). Most individuals moved southeasterly from natal areas
and dispersal movements of individual ibis were concentrated at the mean direction (p = 185.6°, s = 86.1°). First
reproduction by females (3.64 ± 1.36 years) was not significantly earlier than for males (4.17 ± 1.47 years). The results of
this study will be useful for reintroduction programs for critically endangered species. Received 20 January 2009. Accepted
20 November 2009.

Dispersal, a common response of many birds to
habitat diversity, is of substantial interest to
animal ecologists due to its importance in
population ecology and management (Clobert et
al. 2001). Despite its importance, the study of
animal dispersal was primarily descriptive until
the 1990s (Kendall and Nichols 2004). Walters
(2000) suggested that ornithologists should focus
on studying dispersal and dispersal biology of
birds is gradually becoming better understood
(Powell and Frasch 2000, Hansson et al. 2002,
Paradis et al. 2002). Recent studies have used
mark-recapture techniques (Paradis et al. 1998),
intensive observations of marked individuals
(Koenig et al. 2000, Pyle 2001), and radiotelem-
etry (Walls and Kenward 1998, Lang et al. 2002).
Indirect techniques, including genetic markers
(Horn et al. 1996, Wang and Trost 2001) and
stable isotope analysis (Hobson et al. 2001), have
also been u.sed.

The Crested Ibis {Nipponkt nippon) was once
one of eastern Asia’s most widespread and
familiar birds. However, deforestation and de-
struction of their habitat led to successive
extinction of Russian, Korean, and Japanese

' College of Life .Sciences. Shaanxi Normal University.
Xi'an. 710062. China.

^College of Life Sciences, Zhejiang University, Hang-
zhou. 310058, China.

LShaanxi Crested Ibis Conservation Station, Yangxian
County, Shaanxi. 723300. China.

■* Crested Ibis Reproduction Center, Shaanxi Ningshan
Forestry Bureau. Ningshan, 725000, China.

Academy of Fore.st Inventory and Planning, State
Forestry Administration. Beijing, 100714, China.
'’Corresponding author; e-mail: yuxp64@163.com

populations in the 20th century (Yamashina
1967, Archibald et al. 1980). The Chinese Crested
Ibis population, once also thought to be extinct
since the 1960s, was rediscovered in Yangxian
County in Shaanxi Province at the end of a 3-year
(1978-1981) survey by a research group from the
Chine.se Academy. The survey covered nine
provinces including Liaoning, Gansu, Hebei,
Shandong, Henan, Anhui, Jiangsu, Zhejiang, and
Shaanxi (Liu 1981). The single remnant popula-
tion in Yangxian, comprised of only two pairs and
three chicks survived in the remote Qinling
Mountains at an altitude of 1 ,200 m where human
exploitation was relatively limited (Shi et al.
1989, Shi and Cao 2001). The wild population has
increased from .seven to —600 individuals fol-
lowing protection and management of this area,
and its range has expanded to —3,000 km^ to
lower elevations in the past 29 years.

The Crested Ibis population has been inten-
sively studied for nearly three decades including
breeding habits (Shi et al. 1989), gender identi-
fication (Li et al. 2001), habitat use (Liu et al.
2003), genetic diversity (Zhang et al. 2004, He et
al. 2006), and reproductive success (Yu et al.
2006). However, postfledging and natal dispersal
of this population remains unclear. We studied the
dispersal ecology of Crested Ibis using colored
plastic and numbered bands, combined with
radiotelemetry, during 1987-2002.

METHODS

Study Area. — We conducted fieldwork for
16 years (1987-2002) in Yangxian County in
the Qinling Mountains of southern Shaanxi

228

Yii el al. • CRESTED IBIS DISPERSAL

229

FIG. 1 . Dispersal of Crested Ibis (« = 25) from natal site to breeding territory. The insert at the lower right shows the
relative position of the study area.

Province, China (33° 02'-33° 43' N, 107° 11'-
108° 03' E; Fig. 1). This area is used by Crested
Ibis for breeding territories for adults and rearing
sites for young. More than 80% of the areas used
are primarily deciduous broad-leaf forests where
the dominant tree species include cork oak
(Quercus variahilis), Chinese pine (Finns tahu-
laeformis), Chinese red pine (P. massoniana), and
oriental arborvitae (Platycladus orientalis). One
nesting area is in even-aged cork oak forest
characterized by trees with diameter at breast
height (dbh) ranging from 42 to 80 cm, and
canopy heights ranging from 17.2 to 32.6 m,
planted in a graveyard of the Qing Dynasty in the
1800s (Shi and Cao 2001). Areas to which both
adults and young dispersed during August-
October are characterized by hills with many
artificial ponds and reservoirs, and vast farmlands
along the Hanjiang River. Area studied was
expanded each season to include sites where both
young and adults dispersed.

The average annual air temperature was 12-
14° C reaching a minimum in January ( — 10. 1° C)
and a maximum in July (38.7° C) with a frost-free
period of 238 days. Average monthly rainfall
ranged from 4 mm in January to 161 mm in July
with most occurring between July and September
(Youngxian County meteorological station).

Banding and Radiotelemetry. — Crested Ibis
prefer to live near human settlements (Li et al.
2002) making it possible to find all nests and mark
all nestlings (304 individuals) (Table 1). We
banded nestlings with a combination of colored
plastic and numbered bands at 24-26 days of age
when their tarsi were fully developed but before
they were able to leave the nest. We checked and
verified the band color combination and number
of both juveniles and adults at roost sites in
wintering areas, and at nests in breeding ten'itories
each year. We also recorded the location, altitude
of nesting and roosting site, as well as age,
gender, behavior, and color markings of all
individuals present. We made repeated observa-
tions to confirm identity when a banded adult was
encountered on a breeding territory. We placed
15-g radio transmitters on Crested Ibis nestlings
weighing 1,350-1,450 g between days 23 and 25
in the 35-day nestling period. Radio transmitters,
which could transmit for 6 months, were attached
dorsally with the antenna pointing toward the tail
using a harness of nylon elastic string tied under
each wing, leaving room for growth of the wings
and pectoral muscles. Three nestlings from
different broods were radio-marked in 1990,
1992, and 1994, respectively. Family groups
consisting of siblings and their parents formed

230

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2. June 2010

TABLE 1. Observation rates of banded nestlings and
natal sites of Crested Ibis. Values in parentheses are the
number of reobservations of banded vs. all nestlings
banded. Sequence number of natal sites (1-25) is
consistent with Figure 1 .

Year

Ob.servation rate %

Natal sites

1987

33 (1/3)

1

1988

43 (3/7)

2,3

1989

57 (4/7)

1,3,16

1990

50 (3/6)

2,3,4

1991

50 (2/4)

11,16

1992

43 (3/7)

5,7

1993

0 (0/2)

1994

17 (2/12)

17,18

1995

0 (0/10)

1996

7 (1/14)

14

1997

32 (7/22)

5,15,17,19,22

1998

50 (8/16)

5,9,15,16,25

1999

31 (31/39)

6,13,23,25

2000

15 (5/33)

11,14,16,18

2001

12 (22/55)

5,14,17,18,20,21

2002

15 (15/67)

7,12,19,24

Totals

24 (73/304)

25

wandering flocks for nearly 3 months. We were
able to locate radio-marked individuals, as well as
other birds with colored and numbered bands in
the same flock, using programmable receivers
(TRX-IOOOS, Wildlife Materials Inc., Carbondale,
IL, USA) and 3-element, hand-held antennas.

Bird Visits and Surveys. — We considered
marked Crested Ibises as new breeders when they
first performed territorial displays (vocalizing,
perching near, or flying close to the observers)
while eggs or chicks were present in the nest from
early January to mid-February. The frequent and
hoarse calls of the pair, especially the male, were
u.sed to locate breeding birds. We classified males
and females of each pair by observing copulation
behaviors during courtship. Age of each bird was
based on individual identification, using a 40X
telescope, of the colored plastic or numbered
bands on their legs.

We defined natal dispersal as movement from
hatch site to first breeding or potential breeding
site (Greenwood and Harvey 1982), and post-
fledging dispersal as movements of young after
they became independent of their parents and left
the natal territory to move to wintering areas (Bull
et al. 1988). Crested Ibis usually start to breed
when they are 3-years of age (occasionally 2 or 4-
7 years) (Zhai and Wang 1994, Lu et al. 1997).
Relocated individuals were grouped into two data

sets: yearlings and >2 year-old individuals. These
data sets were analyzed separately for differences
in distance moved by yearlings and those
dispersing from natal areas until they became
established on breeding areas. The distance
moved by male and female yearlings was
combined as individuals could not be identified
prior to performing courtship displays.

We recorded the location of birds re-encoun-
tered using Global Positioning System (GPS)
receivers. The distance moved was calculated on
1:50,000 topographic maps as a straight-line
between natal sites and first breeding territories
for natal dispersers or post-fledging areas for
juveniles. We estimated the azimuth from the nest
location to the final known location for each
dispersing ibis (final direction). We calculated the
mean final direction (/i) and the angular deviation
(s) around the mean final direction for each
cohort, and used Rayleigh’s z to test for
directional pattern within cohorts (Zar 1996). All
circular analyses were conducted using Oriana for
Windows (Version 2.0) (Oriana 2003).

Non-parametric statistical treatments (SPSS for
WINDOWS, Version 1 1.5) were used for tests if
dispersal data was not normally distributed. We
report means ± SD unless otherwise noted, and
statistical tests were considered significant at F =
0.05.

RESULTS

Postfledging Dispersal of Young. — ^We followed
36 yearlings of the 304 nestlings banded including
three radio-marked individuals from fledging
through dispersal to areas used for breeding. All
natal home ranges were close to paddy fields near
human settlements at the edge of even-aged,
mature oak-pine forests characterized by 25-
80 cm dbh, and canopy heights ranging from 15
to 33 m. Stands ranged in age from 25 to
183 years.

One movement pattern was used by Crested
Ibis family groups during postfledging dispersal in
all cases. Movements of family groups prior to
dispersal were centered on nest sites, and areas
within the nest territory were visited repeatedly up
to time of dispersal. The distance between family
groups and their natal nest sites was 0.54 ±
0.67 km. Nest territories were occupied for an
average of 21.1 ± 10.2 days (n - 36) from time
of fledging (38-40 days of age) to dispersal.
Family groups, composed of chicks and adults,
moved progressively farther (Fig. 2) along the

}•;/ ('/ ul. • CRESTED IBIS DISPERSAL

231

tributaries of Hanjiang River in a southerly
direction with a mean direction of 189.2 ±
46.5° (/? = 36 ibis). Siblings of each brood
increasingly become independent of their parents
at the beginning of August and all young, as well
as adults from different families, formed wander-
ing flocks in post-dispersal areas characterized by
a sweep of distant hills with many artificial ponds,
reservoirs, and vast farmlands along the Hanjiang
River. Mean dispersal distance was 20.3 ± 7.0 km
(n = 36). Young took 51.4 ± 8.1 days (range =
40-64 days) to disperse from their natal area to
their post-dispersal range, where most remained
for nearly 2-3 years until they became sexually
mature. Most immature individuals (2^ years of
age) spent summers in wintering areas 15.5 ±
4.3 km in = 25) south (191.5 ± 33.6°, n = 36
ibis) of their first breeding sites. The average
elevation of the roosting sites (560.0 ± 55.6 m, n
= 15) in wintering areas was significantly lower
than that in first breeding sites (786.5 ± 56.8 m, n
= 25) (/ = 12.3, P = O.OOO).

Natal Dispersal Distance. — Seventy-three (36
males and 37 females) of the 304 chicks banded
from 1987 to 2002 were followed to their first
reproductive attempt (i.e., were recorded with eggs
or chicks) at 25 natal sites (Table 1 ). Males and
females as first breeders nested at a mean distance
of 5.9 ± 6.3 km (// = 36) and 9.6 ± 10.2 km (n —
37) from their natal sites, respectively. Females
dispersed significantly farther than males (/ = 2. 1 4,
P = 0.03) and the ranges of their dispersal
distances were dissimilar with males dispersing

0.7 to 28.0 km and females 1 .0 to 46.2 km (Fig. 3).
Natal dispersal distances of females by age groups
were significantly greater than for males (Wil-
coxon Test, Z = —2.394, P = 0.003; Table 2).
Mean distance of females and males from natal
areas to breeding tendtories decreased with age of
first reproductive attempt (two-way ANOVA: F =

1 1.381, df = 4, F = 0.019).

Direction of Natal Dispersal. — ^There was a
significant directional pattern to natal dispersal for
males and females, based on directions to final
locations for individuals. Mean final direction was
not random for both cohorts, and we pooled all
cohorts for estimation of overall mean direction (//
= 185.6°, s = 86.1°; Fig. 4). Most individuals
dispersed southeast from natal sites, suggesting
that dispersal movements of individual ibis were
significantly concentrated around the mean direc-
tion (Rayleigh’s ~; all P < 0.05).

Age of First Breeding. — ^There were 73 birds for
which age of first breeding was precisely known
from observations of courtship displays of marked
individuals. The youngest birds participating in
breeding were 2 years of age. The oldest bird
recorded as a first breeder was 7 years of age. The
range of ages of first reproduction was 2 to 6 years
for females and 2 to 7 years for males (Fig. 5).
The mean age of these birds when first recorded
breeding was 3.89 ± 1.40 years (/? = 73). First
reproduction by females was not significantly
earlier than males (females 3.64 ± 1.36 years, n
= 37; males 4.17 ± 1.47 years, n = 36, t = 1.60,
F = 0.1 1).

232

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2U 10

<5 <10 <15 <20 <25 <30 <35 <40 <45 <50

Dutance (km)

FIG. 3. Distribution of natal dispersal distances for Crested Ibis (n = 36 males and 37 females).

DISCUSSION

Measuring dispersal in spatially heterogeneous
environments is essential for testing population
dynamics models (Kareiva 1990), but obtaining
unbiased dispersal data is difficult, because long-
distance dispersers often go undetected (Koenig et
al. 2000). We were able to track 73 individual ibis
until they settled and paired, and our study areas
were sufficiently large (—3,000 km-) to allow
reasonable quantification of long-distance dis-
persal (up to 50 km. Fig. 1). Multi-year obser-
vations allowed for large sample sizes for
juveniles.

Crested Ibises had a skewed dispersal pattern
with many individuals dispersing relatively short
distances, but with some dispersing much
greater distances (Fig. 3). Breeding territories
are typically small with a diameter of 50-100 m
(Shi and Cao 2001) with several nests on the
same tree (Liu et al. 2003). However, all marked
individuals dispersed from their natal home

ranges (100% dispersed >0.7 km). A similar
pattern has been reported for other Ciconii-
formes, including Snowy Egrets (Egretta thula).
Black-crowned Night Herons {Nycticorax nycti-
corax), and Forster’s (Sterna forsteri) and
Caspian terns {S. caspia), in contrast to Great
Blue Herons (Ardea herodias) which have a
tendency to return to natal areas to breed as
adults (Gill and Richard 1979).

Gender-biased dispersal is common among
many bird species with females usually dispersing
farthest (Greenwood 1980). Females settled
farther from natal sites than males in some
Ciconiiformes such as White Storks {Ciconia
ciconia) (Chernetsov et al. 2006), but it is most
often the female that is more likely to leave the
natal area and travel longer distances before
settling to breed (Clarke et al. 1997). Our study
strengthens the conclusions drawn by Greenwood
(1980) regarding this broad, general pattern
(Fig. 3).

TABLE 2. Natal dispersal distances (km) in relation to age and gender of Crested Ibis.

Puramclcr

Age of first reproduction (.v ± SD)

2-yr

.Vyr

4-yr

.S-yr

6-yr

7-yr '

Males

1.3.2 ± 9.7

7.3 ± .3.1

3.9 ± 2.4

3.7 ± 2.3

2.0 ± 0.3

2.6 ±

2.2

(n = 4)

(n = 12)

(/; = 8)

(« = -3)

(n = 3)

in =

4)

Females

14.1 ± 10.2

12.2 ± .3..3

9.4 ± 6.3

7.4 ± 2.7

4.3 ± 1.9

0

(n = 9)

in = 9)

(n = 9)

(/) = 6)

(n = 4)

(n =

0)

Gender combined

14.3 ± 9.9

9.4 ± 10.4

6.6 ± 7.4

.3.9 ± 2.6

2.8 ± 1.6

2.6 ±

2.2

(n = 13)

(n = 21)

(n = 17)

(n = 11)

(n = 1)

(/; =

4)

Yti cl uL • CRESTED IBIS DISPERSAL

233

0

FIG. 4. Graphical depiction showing final direction
from the natal site to first breeding territory for 73 banded
Crested Ibis. Directions are spatially nonrandom (Ray-
leigh’s z: all P < 0.05).

Direction of natal dispersal and post-fledging
movement was likely in response to habitat
factors. Crested Ibis were once commonly seen
in towns and cities in agricultural areas, near
which they could find large trees for nesting and
suitable wetlands for feeding (La Touche 1934).
The wild population that survived in Yangxian
County persisted in remote mountainous districts
at an altitude of 1,200 m, where human exploi-
tation was not as severe as in neighboring areas
(Shi et al. 1989, Shi and Cao 2001). The species

has showed a recent prel'erence to move to lower
areas for breeding (Ding 2004). Crested Ibis prior
to 1990 usually nested on oaks in mountainous
areas above 1,000 m. They have colonized new
nesting sites in recent years where they usually
nest on pines at lower altitudes (700-900 m),
presumably indicating the relict population had
become confined to suboptimal upland habitat but
is now expanding into more typical lowland areas.
More recently, a few birds have been found
breeding in patches of secondary forests close to
human settlements (Liu et al. 2003).

Forty-nine (67.1%) of 73 marked chicks in our
study required 2 to 4 years to become territorial
(Fig. 5). Females bred no more than 1 year earlier
than males on average, similar to Marabou Storks
{Leptoptilos cnimeniferus) (Pomeroy 1978), Scar-
let Ibis {Eudocimus ruber), and Eurasian Spoon-
bill {Platalea leucorodia) (Hancock et al. 1992).
These data may indicate the habitat is saturated or
reflect other breeding constraints on the popula-
tion. Variation in age of first breeding could
reflect individual variation in physical condition
or foraging skills or, alternatively, individuals
selecting different strategies for maximizing
reproductive success (Nelson 1977, Osorio-Ber-
istain and Drummond 1993). Males are at least
1 year older than females in most pairs (Lu et al.
1997), and males were observed to have a
dominant role in territory defense (Shi and Cao
2001). This suggests that greater experience is
necessary for males to have a mate and hold a
territory.

0 35

2 3 4 5 6 7

Age (yeirs)

FIG. 5. Age of first breeding attempt in = 36 males and 37 females).

234

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2, June 2010

Survival, productivity, and dispersal data are
important parameters in population models that
highlight stages in a species’ life cycle where
conservation actions may be most critical. Lim-
ited data on dispersal of both young and yearling
ibis led to imprecise survival estimates, and the
low success of first breeding attempts for each age
class (Yu et al. 2006) suggest low year-to-year
survival of juveniles. Our results indicate that
young individuals are likely to disperse farther
than adults, and females breed nearly 1 year
earlier than males. The current distribution is
progressively expanding to areas at lower eleva-
tion and denser human populations by dispersal of
young individuals. These findings have potential-
ly important implications for reintroduction pro-
grams that are currently being considered (Yu et
al. 2009). We strongly recommend that reintro-
duction programs be conducted with younger (2-
3 yrs of age) individuals which are more prone to
establishing new territories and have low site
fidelity.

ACKNOWLEDGMENTS

We thank Y. M. Zhang, T. Q. Zhai, Y. P. Chen, and Y. J.
Wang for assistance in field observations. The State
Forestry Administration, Shaanxi Nature Reserve and
Wildlife Administration, and Shaanxi Natural Science
Foundation provided administrative support and funding.
We thank the Yangxian County weather bureau for
providing meteorological information. Chris Wood, Daniel
White, and Ren Wei significantly improved earlier versions
of this manuscript.

LITERATURE CITED

Archibald, G. W., S. D. FI. Lantis. and I. Munetchika.
1980. Endangered ibises, Threskiornithinae: their
future in the wild and in captivity. International Zoo
Yearbook 20:6-17.

Bull, E. L., M. G. Henjum, and R. S. Rohweder. 1988.
Home range and dispersal of Great Gray Owls in
northeastern Oregon. Journal of Raptor Research
22:101-106.

Chernetsov. N.. W. Chromik, P. T. Dolata. P. Proeu.s,
AND P. TRYJANOW.SKI. 2006. Sex-relatcd natal dispers-
al of White Storks (Cicouici ciconia) in Poland: how
far and where to? Auk 12.1:1 103-1 109.

Clarke, A . L.. B. E. S/ether. and E. Ro.skaft. 1997. Sex
biases in avian dispersal: a reappraisal. Oikos 79:429-
438.

Cl-OBfiRT. J.. E. Danchin. a. a. Diiondt, and j. D.
Nichols. 2001. Dispersal. Oxford University Pre.ss,
Oxford. United Kingdom.

Ding, C. Q. 2004. Research on the Crested Ibis in China.
Shanghai Scientific and Technological Publishing
House. Shanghai, China.

Gill Jr., R. and L. M. Richard. 1979. Dispersal and
migratory patterns of San Erancisco Bay produced
herons, egrets, and terns. North American Bird Bander
4:4-13.

Greenwood, P. J. 1980. Mating systems, philopatry and
dispersal in birds and mammals. Animal Behaviour
28:1140-1162.

Greenwood, P. J. and P. H. Harvey. 1982. The natal and
breeding dispersal of birds. Annual Review of Ecology
and Systematics 13:1-21.

Hancock, J. A.. J. A. Kushlan, and M. P. Kahl. 1992.
Storks, ibises and spoonbills of the world. Academic
Press, New York, USA.

Hansson. B., B. D. Hasselquist. and D. B. Nielsen.
2002. Restricted dispersal in a long-distance migrant
bird with patchy distribution, the Great Reed Warbler.
Oecologia 130:536-542.

He, L. P., Q. H. Wan, S. G. Fang, and Y. M. Xi. 2006.
Development of novel microsatellite loci and assess-
ment of genetic diversity in the endangered Crested Ibis.
Nipponia nippon. Conservation Genetics 7:157-160.

Hobson, K. A., K. P. McFarland, L. 1. Wassenaar, C. C.
Rimmer, and j. E. Goetz. 2001. Linking breeding and
wintering grounds of Bicknell's Thrushes using stable
isotope analyses of feathers. Auk 1 18:16-23.

Horn, P. L., J. A. Rafalski. and P. J. Whitehead, 1996.
Molecular genetic (RAPD) analysis of breeding
Magpie Geese. Auk 1 13:552-557.

Kareiva, P. 1990. Population dynamics in spatially
complex environments: theory and data. Philosophical
Transactions of the Royal Society of London. Series B
330:175-190.

Kendall, W. L. and J. D. Nichols. 2004. On the
estimation of dispersal and movement of birds. Condor
106:720-731.

Koenig, W. D., P. N. Hooge, M. T. Stanback. and J.
Haydock. 2000. Natal dispersal in the cooperatively
breeding Acorn Woodpecker. Condor 102:492-502.

La Touche, J. D. D. 1934. Page 434 in A handbook of the
birds of eastern China. 11. Part 5. Taylor and Francis,
London, United Kingdom.

Lang, J. D., L. A. Powell, D. G. Krementz, and M. J.
Conroy. 2002. Wood Thrush movements and habitat
use: effects of the forest management for Red-
cockaded Woodpeckers. Auk 1 19:109-124.

Li, M.. C. Q, Ding, F. W. Wei. S. J. Meng, Y. M. Xi, and
B. Z. Lu. 2001 . Sex-related gene and sex identification
of Crested Ibis Nipponia nippon (Ciconiiformes:
Threskiornithidae), Chinese Science Bulletin 46:669-
671.

Li, X. H., Z. .1. Ma. C. Q. Ding, T. Q. Zhai, and D. M. Li.
2002. Relationship between the distribution of Crested
Ibis and local farmers. Acta Zoologica Sinica 48:725-
732. . ■

Liu, D. P., C. Q. Ding, and G. Z. Chu. 2003. Home range
and habitat utilization of the Crested Ibis in the
breeding period. Acta Zoologica Sinica 49:755-763.

Liu, Y. Z. 1981. The re-discovery of the Crested Ibis in
Qinling Mountain. Acta Zoologica Sinica 27:273.

Lu, B. Z., T. Q. Zhai, Y. M. Zhang, and Z. Z. Geng. 1997.
Wild populalion slruclure and relationship of the

Yn el cil. • CRESTED IBIS DISPERSAL

235

Crested Ibis (Nipponia nippon). Journal of Shaanxi
Normal University 25 (Supplement):98-I04.

Nelson, J. B. 1977. Some relationships between food and
breeding in the marine Pelecaniformes. Pages 77-78 in
Evolutionary ecology (B. Stonehouse and C. Perrins,
Editors). University Park Press, Baltimore, Maryland,
USA.

Oriana. 2003. Oriana for Windows, Version 2.0. Kovach
Computing Services, Pentraeth, Wales, United King-
dom.

Osorio-Beristain, M. and H. Drummond. 1993. Natal
dispersal and deferred breeding in the Blue-footed
Booby. Auk 110: 234-239.

Paradis, E., S. R. Baillie, and W. J. Sutherland. 2002.
Modeling large-scale dispersal distances. Ecological
Modelling 151:279-292.

Paradis, E., S. R. Baillie, W. J. Sutherland, and R. D.
Gregory. 1998. Patterns of natal and breeding dispersal
in birds. Journal of Animal Ecology 67:518-536.

Pomeroy, D. 1978. The biology of Marabou Storks in
Uganda. II. Breeding biology and general review.
Ardea 66: 1-23.

Powell, L. A. and L. L. Frasch. 2000. Can nest predation
and predator type explain variation in dispersal of
adult birds during the breeding season? Behavioral
Ecology 11:437^143.

Pyle, P. 2001. Age at first breeding and natal dispersal in a
declining population of Cassin’s Auklet. Auk 118:
996-1007.

Shi, D. C. and Y. H. Cao. 2001. The Crested Ibis in China.
China Forestry Publishing House, Beijing.

Shi, D. C., X. P. Yu, X. Y. Chang, and B. Z. Lu. 1989.
Breeding habits of the Crested Ibis. Zoological
Research 10:327-332.

Walters, J. R. 2000. Dispersal behavior: an ornithological
frontier. Condor 102:479-481.

Walls, S. S. and R. E. Kenward. 1998. Movement of
radio-tagged Buzzards Buleo hiiteo in early life. Ibis
140:561-568.

Wang, X. H. and C. H. Trost. 2001. Dispersal pattern of
Black-billed Magpies (Pica hudsonia) measured by
molecular genetic (RAPD) analysis. Auk 118:137-
146.

Yamashina, Y. 1967. The plight of the Japanese Crested
Ibis. Animals 10:275-277.

Yu, X. P., N. F. Liu, Y. M. Xi, and B. Z. Lu. 2006.
Reproductive success of the Crested Ibis Nipponia
nippon. Bird Conservation International 16:325-
343.

Yu, X. P., X. Y. Chang, X. Li, W. G. Chen, and L. Shi.
2009. Return of the Crested Ibis Nipponia nippon: a
reintroduction programme in Shaanxi Province, China.
BirdingASIA 11: 80-82.

Zar, j. H. 1996. Biostatistical analysis. Third Edition.
Prentice-Hall, Englewood Cliffs, New Jersey, USA.

Zhai, T. Q. and Z. Y. Wang. 1994. On the reproduction of
two-year-old Crested Ibis. Acta Ecologica Sinica
14:99-101.

Zhang, B., S. G. Fang, and Y. M. Xi. 2004. Low genetic
diversity in endangered Crested Ibis Nipponia nippon
and implications for conservation. Bird Conservation
International 14: 183-190.

The n'ilson Journal oj Ornithology 122(2):236-243, 2010

MORPHOLOGICAL AND GENETIC VARIATION BETWEEN
MIGRATORY AND NON-MIGRATORY TROPICAL KINGBIRDS
DURING SPRING MIGRATION IN CENTRAL SOUTH AMERICA

ALEX E. JAHNT' DOUGLAS J. LEVEY,' IZENI FIRES FARIAS,- ANA MARIA MAMANI,^

JULIAN QUILLEN VIDOZ,^ AND BEN FREEMAN^

ABSTRACT. — We attempted to distinguish spring passage migrant Tropical Kingbirds (Tyrannus melancholicus) from
resident conspecifics where they overlap in South America. Migrant males at our Bolivian study site had significantly less
tail feather molt and longer wing chords than resident males. Migrant females had significantly longer wing chords, less
flight feather molt, and less flight feather wear than resident females. We found no evidence of genetic population
differentiation between migrants and residents. We also compared wing chords of migrants and residents to those of
breeding kingbirds in breeding populations further south. Wing chords of migrants were more similar to those of breeders
from further south than to those ot breeders at our study site. An ability to distinguish migrant from resident conspecifics
will be critical to understanding migrant winter ecology, migratory routes, and connectivity of migratory populations in
South America. Received IH May 2009. Accepted 2 December 2009.

Almost 70% of neotropical austral migrant
species (hereafter “austral migrants”), which
migrate wholly within South America, are char-
acterized by having both migratory and non-
migratory (permanent resident) populations (Stotz
et al. 1996). Populations of permanent residents
usually occur within or close to tropical latitudes,
whereas populations of migratory individuals
occur in temperate latitudes during the breeding
season and in tropical latitudes during the non-
breeding season. Ranges of migratory and resident
populations of austral migrants typically overlap
during the non-breeding season. Patterns of
migration within the range of overlap are rarely
documented because distinguishing between mi-
gratory and resident individuals is difficult. Thus,
for the great majority of austral migrants, we
know little about migratory routes, non-breeding
season ecology, population connectivity, and
intraspecific interactions between residents and
migrants.

Zimmer (1938) and Lanyon (1978) used
.seasonal distribution, morphology, and molt data
to study where non-breeding populations of
migrants and residents overlap in South America.
More recently, Ches.ser (1995, 1997) primarily

' Department of Biology, University of Florida, Gaines-
ville. FL 3261 I. U,SA.

^Instituto de Ciencias Biologicas, Departamento de
Biologia, Universidade Federal do Amazonas. Manaus.
Amazonas, Brazil.

'Museo de llistoria Natural Noel Kempff Mercado,
Santa Cruz. Bolivia,

■'8412, 36th Avenue Northeast, Seattle, WA 98 1 1 5, USA.
’Corresponding author; e-mail: alexJahn77@yahoo.com

used seasonal distribution data to identify the
range of winter overlap of migrant and resident
austral migrant populations at the center of the
continent. Significant morphological differences
in other migratory systems exist between migrant
and resident individuals of the same species; those
differences have been used to distinguish between
co-occLining migrants and residents. For example,
migratory European Blackcaps {Sylvia atricapilla)
have longer, more pointed wings and smaller
bodies than resident conspecifics in Iberia (Tell-
eria and Carbonell 1999), allowing accurate
classification of migrant and resident Blackcaps
in winter (Perez-Tris et al. 1999). Similarly, a
migratory subspecies of Common Reed Bunting
{Emheriza schoenidus) in Spain has larger and
more convex wings than a sedentary subspecies
(Copete et al. 1999). Molt schedules may also
differ between co-occurring resident and migrant
individuals (Lundberg and Eriksson 1984).

Range overlap of resident and austral migrant
populations in the non-breeding season occurs for
Tropical Kingbirds (Tyrannus melancholicus),
which are permanent residents in most of tropical
South America. Two subspecies (satrapa and
despotes) appear to be permanent residents in
northern Amazonia (Traylor 1979, Chesser 1995),
whereas the subspecies melancholicus is a resident
from most of Amazonia south to —18° S. South' of
this latitude, melancholicus occurs only in sum-
mer, migrating north after breeding into the range
of permanent resident populations (Chesser 1995;
Fig. 1).

We studied a resident population of Tropical
Kingbirds (hereafter “kingbirds”) at a site in

236

John ct c/I. • TROPICAL KINGBIRDS

237

Tropical Kingbird {Tyra//niis m. mela/\cholicits) in South
America. Gray polygon represents the permanent distribu-
tion of the species and the hatched polygon south of 18° S
represents the seasonal breeding range. Star represents
location of Caparu Biological Station, Bolivia. Adapted
from Chesser (1995).

eastern Bolivia (~15° S) within the range of the
nominate subspecies. We noticed that many
kingbirds captured during September and Oc-
tober were much fatter than others, and that
fewer of those fat individuals were seen again
at the site, compared to those individuals that were
less fat. We hypothesized that fatter individuals
were passage migrants, on their way to breeding
areas further south. Our objective was to assess
different methods of distinguishing between
migrant and resident kingbirds where they co-
occur. We compared; ( 1 ) the molt schedule, wing
chord, and genetic signature between putative
migrants and residents, and (2) the wing chord of
migrants and residents at the study site with wing
chords of kingbirds breeding lurther south.

METHODS

Study Site and Field Procedures. — Our study
site was Caparu Biological Station (CBS), De-
partment of Santa Cruz, eastern Bolivia (14° 49'
S, 61° ir W; 170 m asl). The site is near the
center of the continent, north of the boundary of
migratory kingbirds’ breeding range and south of
a large expanse of potential wintering area (i.e..

Amazonia; Fig. 1). The habitat is primarily
cerrado grassland, humid forest edge, and cattle
pasture. Kingbirds were captured using nylon and
polyester mist nets (12 or 18 m X 2.6 m, 36 or
38 mm mesh size) where they were most abundant
(e.g., ponds and forest edge). Kingbirds were also
banded as nestlings and parents were captured at
their nest by placing a stuffed Purplish Jay
(Cyanocora.x cyanomelas), a common nest pred-
ator, nearby.

Kingbirds were banded with uniquely-
numbered aluminum bands provided by the
Museo de Historia Natural Noel Kempff Mercado
(MHNNKM), Santa Cruz, Bolivia as well as with
up to three celluloid color bands. Banding was
conducted during most months from October 2004
to July 2007. Data recorded for captured kingbirds
included reproductive condition (presence/ab-
sence of cloacal protuberance or brood patch),
fat score (measured on an 8-point scale), body
molt (measured on a 5-point scale), primary
feather wear (measured on a 6-point scale), length
of unflattened wing chord, tail, beak, and tarsus
length, and mass (Ralph et al. 1993, Pyle 1997).
We classified flight feather and tail molt as
present or absent; molt was classified as “pres-
ent” if at least one secondary or primary feather
or retrix was molting. These measurements were
taken by AEJ and AMM, who collected data
together on a regular basis to minimize observer
error. We classified birds in their first year of life
as hatch-year individuals (HY) or after-hatch-year
individuals (AHY) for individuals that were of
unknown age but at least 1 year of age. Each year
began on 1 September because the kingbird
breeding season lasts from mid-September to
February.

Blood was taken for genetic analysis from all
color-banded individuals by piercing the brachial
vein with a sterile needle. Blood was collected in
a 0.5-mL heparinized capillary tube and stored in
DNA lysis buffer (Seutin et al. 1991). We
assigned gender of live birds based upon primary
notch length (Pyle 1997) and, when necessary,
through molecular methods using primers 2550F
and 271 8R (Fridolfsson and Ellegren 1999).
Optimal cycle conditions were 95° C for 5 min,
95° C for 30 sec, an annealing temperature of 50°
C for 1 min, and 72° C for 45 sec. repeated 39X.
and held at 72° C for 10 min, using 15 pL Mg.
Blood samples were deposited at the MHNNKM.

We documented seasonal presence/absence of
color-banded kingbirds after banding to identify

238

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2. June 2010

which individuals were residents. We divided the
~ 700-ha study site into 23 sampling plots and
methodically searched for banded individuals in
each plot at least once/month throughout most of
the annual cycle. These censuses were completed
from February 2005 to August 2007. We did not
visit plots during most of the non-breeding season
of 2005 because we were absent during June-
September. We also visited plots on 28 January to
12 February, 2-19 March, and 15-27 June 2008.
We geo-referenced the location when a color-
banded individual was observed using a Global
Positioning System receiver (Garmin GPS 76,
Olathe, KS, USA), noting the date, time, and color
band combination.

AEJ measured 120 kingbird skins from Argen-
tina and Paraguay housed at Museo Argentine
de Ciencias Naturales Bernardino Rivadavia
(MACN; Buenos Aires, Argentina) and 79 skins
at Louisiana State Museum of Natural Science
(LSUMZ; Baton Rouge, USA) to examine
whether morphological variation among popula-
tions could be useful in identifying breeding
locations of passage migrants at CBS. BF also
measured 60 study skins from Paraguay and
Argentina at the American Museum of Natural
History (AMNH; New York City, USA). Un-
flattened wing chord was measured as on live
kingbirds at CBS, following Ralph et al. (1993).
Wing chord was not measured on specimens with
excessively worn primary feather tips. We
previously standardized our measurements to
minimize inter-observer measurement error.

We isolated DNA from blood samples using the
PUREGENE DNA Purification Kit (Gentra Sys-
tems, Minneapolis, MN, USA) and reviewed the
literature to identify potential primers of poly-
morphic loci developed from related species in
Tyrannidae. Six microsatellite primers were
successfully optimized from Empidonax (GATA5;
Pearson et al. 2006) and Sayornis (SAP22,
SAP32, SAP47, SAP50, SAP 156; Watson et al.
2002, Beheler et al. 2007). Optimal PCR cycling
conditions for all primers were: 95° C for 4 min
followed by 94° C for 1 min, then 55° C for 1 min,
and 72° C for 1 min, repeating the last three steps
35X with a final cycle of 72° C for 10 min. We
used 12.5 pL of master mix for all primers,
800 pM dNTP’s, 2.5 pM MgCl2, 0.26 pM of each
primer, and a concentration of 10-20 ng of sample
DNA.

We combined males and females for genotyp-
ing due to low sample sizes of kingbirds of known

gender for which we had DNA available for
analysis. We analyzed DNA from 15 migrants and
15 residents. DNA was run on an ABI 3730XL, 96
capillary automated sequencer (Biosystems, Fos-
ter City, CA, USA). The program GENEMAP-
PER (Softgenetics, State College, PA, USA) was
used to score the alleles. Population structure was
tested in program STRUCTURE 2.2 (Pritchard et
al. 2000) assuming a model of admixture,
correlated allele frequencies, and K = 1 to 2
(where K is the number of populations) with 10
independent replicates for each K (length of the
run 1,000,000 MCMC re-samples), and 100,000
burn-in. We used program ARLEQUIN 3.01
(Excoffier et al. 2005) to test for departures from
Hardy-Weinberg equilibrium, and to calculate the
amount of linkage disequilibrium between pairs of
loci. Bonferroni correction was applied for
multiple comparisons in both tests. There was
violation of Hardy-Weinberg equilibrium in one
locus (SAP32) for both resident and migrant
populations, and in only migrants for another
locus (SAP50) but, overall, loci adhered to
equilibrium. Population differentiation was tested
by performing a hierarchical analysis of molecular
variance (AMOVA) to calculate global and pair-
wise Fs, in ARLEQUIN 3.01.

We only consider adult kingbirds (i.e., those
with only trace juvenile plumage or 95% or more
of the skull ossified) to avoid potential confound-
ing effects of age with wing chord (Low 2006), fat
content (e.g., Lundberg 1985), and molt (e.g.,
Debruyne et al. 2006). After-hatch-year (AHY)
kingbirds captured in September and October that
had higher subcutaneous lipid scores (i.e., >6 on
an 8-point scale, sensu Ralph et al. 1993) were re-
observed significantly less often than those with
lower lipid scores (i.e., <6 on the same scale;
Fig. 2). This was true of both males (x^ = 1 1 .8, P
= 0.001) and females = 7 A, P = 0.008). We
classified as passage migrants all AHY kingbirds
captured in September or October with lipid
scores of 6 or higher, that were not re-observed
after being color banded, and which had no
incubation patch or cloacal protuberance (to
exclude the possibility they were breeding at the
study site). We classified residents as individuals
that were captured in September or October, had
subcutaneous lipid scores of 1-5, and re-observed
at least once later on the study site. There were
insufficient residents for which DNA was avail-
able for analyses, and we included as residents for
puiposes of DNA analyses individuals with the

.lulm cl al. • TROPICAL KINCBIRDS

239

Low

□ Females
■ Males

Subcutaneous fat score

FIG. 2. Relationship between amount of subcutaneous fat and frequency of being re-observed for AHY Tropica)
Kingbirds color banded during spring migration at Caparti Biological Station. Bolivia; “Low” = subcutaneous fat score of
<6 on an 8-point scale, “High” = subcutaneous fat score >6.

same characteristics captured in November, which
is well into the breeding season (A. E. Jahn,
unpubl. data).

RESULTS

Resident Tropical Kingbirds at CBS tended to
molt during and after the breeding season (Fig. 3).
Spring passage migrants had significantly less flight
feather molt (Kruskal-Wallis ~ 5.96, P — 0.015,
n = 81) and significantly less tail molt (Kruskal-
Wallis x^ = 7.66, P = 0.006, n = 79) than re.sidents
when data for males and females, and those of
unknown gender were combined. However, there
was no significant difference between migrants and
residents in body molt ( Kruskal-Wallis x^ = 1 .07, P
= 0.30, n = 81). Migrants also had less primary
feather wear (Kruskal-Wallis x^ = 10.36, P =
0.001, n = 80) and longer wings (/ = 4. 17, P <
0.0001, n - 79) than residents.

Female passage migrants had significantly less
flight feather molt, less primary feather wear, and
longer wings than resident females (Table 1 ).
Male residents had significantly more tail molt
than male migrants, but there was no signilicant
difference in flight feather molt or primary feather
wear between migrant and resident males. Male
migrants were also characterized by a significant-
ly longer wing chord than male residents
(Table 1 ).

We used discriminant function analysis (DFA)
to examine the level at which we could correctly
classify migrants and residents at the site during
spring migration. We combined data for males
and females due to the relatively low sample sizes
for each. We first examined which of the five
morphological measurements (flight feather molt,
tail molt, primary feather wear, body feather molt,
and wing chord) provided the most explanatory
power within the discriminant function. A test of
the equality of group means revealed that body
molt contributed the least amount of explanatory
power to the discriminant function {F]j2 = 1.09,
P = 0.30); we therefore excluded it from further
analysis. All four remaining variables significant-
ly contributed to the discriminant function and
were retained in the full model. The parameter
with the least explanatory power was flight
feather molt {F\j2 — 4.48, P = 0.038), and the
parameter with the highest explanatory power was
wing chord (F\j2 = 16.84, P < O.OOOl). The
resulting DFA correctly classified 86.0% of
migrants and 71.0% of residents with an overall
success rate of 79.7%.

Wing chord from study skins of males at >18°
S in — 146) was significantly longer than that of
both resident (/ = —6.93, P < 0.0001, n = 15)
and passage migrant males at the study site {t =
-2.71, P = 0.007, n = 21; Fig. 4). Wing chord

240

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2. June 2010

90 n
80 -

■ Body molt

May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr

FIG. 3. Molt schedule of AHY color-marked, re-observed resident Tropical Kingbirds at Caparu Biological Station,
Bolivia. Data represent the percentage of kingbirds per month with each molt type. Percentages may not sum to 100
because individuals can have more than one molt type.

from Study skins of females at >18° S (/? = 83)
was significantly longer than that of female
residents (r = 6.787, P < 0.0001, n = 11) but
not of female passage migrants at the study site {t
= 1.055, P = 0.29, n = 16; Fig. 4).

We found better support using genetic data for
the presence of one population (log likelihood =
—515.2) than for the presence of two populations
(log likelihood = —516.6), based on Bayesian
analyses of population structure, as implemented
in program STRUCTURE 2.2. This suggests there
is no significant genetic structure between migrant

and resident kingbirds in the six loci tested.
AMOVA also gave no indication of genetic
differentiation between migrants and residents
(^st = 0.01576, P = 0.578). There was no
indication of departures from Hardy-Weinberg
equilibrium or of linkage disequilibrium (Ta-
ble 2).

We conducted DFA for migrants and residents
classified as above, but without considering re-
observation data (i.e., using only fat scores) to
check whether using fat content alone to distin-
guish between migrants and residents is as useful

TABLE 1. Morphological and molt comparisons of resident and passage migrant Tropical Kingbirds at Caparu
Biological Station, Bolivia. Values for flight feather and tail molt are percent of migrants and residents with molt present.
Values for body feather molt and primary feather wear represent mean score on a 5- and a 6-point .scale, respectively (high
.scores indicate more molt or wear).

Resident.s (n)

Migrants in)

Test statistic

p

Males

Flight feather molt

12% (17)

0% (21)

2.54“

0.1 1 1

Tail molt

20% (15)

0% (21)

4.46“

0.035

Body feather molt

2.44 ± 1.09 (16)

2.24 ± 1.18 (21)

0.28“

0.595

Primary feather wear

2.06 ± 0.25 (16)

1.86 ± 0.48 (21)

2.42“

0.120

Wing chord, mm

1 1 1.27 ± 2.87 (15)

1 13.9 ± 2.84 (21)

2.735”

. 0.010

Females

Flight feather molt

25% (12)

0% (16)

4.06“

0.044

Tail molt

8% (12)

0% (16)

1.25“

0.264

Body feather molt

2.82 ± 0.98 (11)

2.47 ± 1.19 (15)

0.50“

0.481

Primary feather wear

2.58 ± 0.79 (12)

1.80 ± 0.56 (15)

6.92“

0.009

Wing chord, mm

106.09 ± 1.97 (1 1)

1 10.63 ± 2.63 (16)

4.846”

<0.0001

“ Kruskal-Wallis test.

^ Independent samples /-test.

Juhn Cl al. • TROPICAL KINGBIRDS

241

H Female

lOO-' 1 1 I

Resident at Migrant at Southern
CBS (26) CBS (37) breeder (229)

FIG. 4. Box-and-whisker diagram depicting wing chord
of Tropical Kingbird residents, passage migrants, and
breeders from seasonal breeding range south of 18° S
latitude in South America. Numbers in parentheses on x-
axis labels represent sample sizes. Rectangles depict the
range of the first to third quartiles and the dark horizontal
line within each rectangle depicts the median. Lines from
each rectangle extend to the larger and smaller values, and
circles outside of rectangles represent outliers.

as using fat plus re-observation data. Body molt
within the revised DFA had practically no
explanatory power (Fijss = 0.01, P = 0.94)
and was excluded. The resulting test of equality of
group means revealed that flight feather molt had
the most explanatory power (F| 159 = 4. 10, P =
0.044) and that wing chord had the least (T1J59 =
0.28, P = 0.60). The DFA correctly classified
75% of migrants and 26.5% of residents with an
overall success of 41.0%.

DISCUSSION

Tropical Kingbirds with high lipid scores and
those that were not re-observed after banding (i.e.,
putative migrants) were characterized by traits
typical of migrants in the Northern Hemisphere:

advanced molt .schedules (Lundberg and Eriksson
1984) and longer wings (Haberman et al. 1991,
Tellerfa and Carbonell 1999, Mila et al. 2008)
relative to kingbirds with less fat and which were
re-observed at the site (i.e., residents). The
differences between the two groups in molt,
feather wear, and wing chord from our site at
the middle of the continent, where passage
migrants would be expected to occur if migrating
between low latitudes to the north and temperate
breeding areas to the south, strongly suggest
some Tropical Kingbirds are long-distance mi-
grants that overwinter in central or northern
Amazonia.

Migrant females at CBS had wing chords
similar to breeders from the seasonal breeding
range >18° S, whereas male migrants had
significantly shorter wings than kingbirds from
the seasonal breeding range. Thus, at least some
female spring passage migrants were likely
returning to the seasonal breeding range, while
male migrants may have been returning to the
seasonal breeding range, as well as to latitudes
closer to the study site. Future research on
geographical variation in wing chord and genetic
structure south of the study site may reveal
geographical variation in traits that could be
useful in identifying the breeding location of these
migrants.

Molt of resident kingbirds at CBS tended to
occur primarily just before, during, and after the
breeding season (Oct-Dec). Thus, many resident
kingbirds at CBS likely had not molted prior to
passage of spring migrants. Passage migrant male
kingbirds had a more advanced pre-alternate tail
molt compared to local resident males, and female
migrants had a significantly more advanced flight
feather molt than residents at CBS. This indicates
migratory kingbirds have more advanced pre-
alternate molt schedules than non-migratory
individuals at CBS. Lundberg and Eriksson
(1984) found the post-juvenile molting period of
migratory European Starlings (Sturniis vulgaris)

TABLE 2. Genetic diversity for resident and migrant Tropical Kingbirds. Genetic diversity parameters include mean
observed heterozygosity {Hq), mean expected heterozygosity (//pj, inbreeding coefficient (F,s) based on the average of six
loci, average number of alleles per locus (A), and average number of private alleles {PA).

Hii-He Gs Average gene diversity across loci A PA n

Residents 0.423-0.478 0.020 (Z' = 0.497) 0.422 ± 0.259 6.67 1.8 15

Passage migrants 0.438—0.611 0.107 (P = 0.088) 0.582 ± 0.339 8.00 3.0 15

242

THE WILSON JOURNAL OL ORNITHOLOGY • Voi 122, No. 2, June 2010

was shortened relative to that of non-migratory
individuals with migrants finishing the molt of
primary and secondary feathers significantly ear-
lier than residents. They postulated the shortened
molting period of migrants was constrained by the
need to migrate after breeding. Our study focused
on spring migration and we do not know if
migrants molt their feathers in a shorter span of
time compared to residents. That migrant kingbirds
finish the pre-alternate molt earlier than resident
kingbirds suggests migration can be a constraint on
molt timing. It remains unknown whether migra-
tion constrains the scheduling of molt at tropical
latitudes.

Genetic analysis revealed no significant popu-
lation subdivision between migrant and resident
Tropical Kingbirds. The lack of microsatellite
structure in our study may be due to at least three
reasons: ( 1 ) too few individuals or loci sampled,
(2) relatively high gene flow occurring among
populations, and (3) recent colonization of
southern latitudes by kingbirds.

Our results corroborate previous evidence that
populations of Tropical Kingbirds breeding at
high southern latitudes in South America (i.e., T.
m. melancholiciis) have longer wing chords than
populations breeding closer to the equator (i.e., T.
m. satrapa and T. m. despotes\ Cory and Hellmayr
1927). Likewise, southern breeding populations of
Swainson’s Flycatcher (Myiarchus swainsoni) in
South America have longer wing chords than
populations breeding at tropical latitudes (Lanyon
1978, Joseph et al. 2003).

Distinguishing between migrant and resident
conspecifics when they occur together at tropical
latitudes is important in advancing our under-
standing of the evolution of migratory behavior.
However, migrants may not be obviously different
in morphology from permanent resident individ-
uals in most species of austral migrants because
austral migrants that overwinter in South Amer-
ica’s tropics migrate an average of 9.2° (±8.5)
latitude (Chesser 1994).

ACKNOWLEDGMENTS

We thank Terry Ches.ser. Scott Robinson. Katie Sieving.
C. E. Braun, and an anonymous reviewer for many helpful
comments that improved the manuscript. We are grateful to
Jaime Rozenman and Guillermo Wei.se for their hospitality
and ongoing support of research at Caparu Biological
Station, and to Rebecca Kimball, Ginger Clark. Carolyne
Bardelebcn, Amanda Behelcr. Christy Watson, and Olin
Rhodes for advice and logistical support for molecular
analyses. We th;mk collection managers al the AMNH

(Paul Sweet), LSUMZ (Steve Cardiff), and MACN (Pablo
Tubaro) for access to the specimens under their care. A
large number of assistants helped in the field and
laboratory, especially Noelia Barrios, Silvia Estevez, Betty
Flores, Fabian Hilarion, Jennifer Johnson, Patricia Justin-
iano, John Prather, Ana Maria Saavedra, and Jordan Smith.
Funding was provided by the American Ornithologists'
Union, the National Science Foundation (OISE-03 13429,
0612025), Optics for the Tropics, School of Natural
Resources and Environment-University of Florida, NSF
Southeast Alliance for Graduate Education and the
Professoriate, Western Bird Banding Association, and the
Wilson Ornithological Society. The genetic research was
supported by a grant to I. P. Farias from CNPq/CT-
Amazonia #575603/2008-9.

LITERATURE CITED

Beheler, a. S., J. A. Fike, and O. E. Rhodes Jr. 2007.
Eight new polymorphic microsatellite loci from the
Eastern Phoebe (Sayornis pboehe). Conservation
Genetics 8:1259-1261.

Chesser, R. T. 1994. Migration in South America, an
overview of the austral system. Bird Conservation
International 4:91-107.

Chesser, R. T. 1995. Biogeographic, ecological, and
evolutionary aspects of South American austral
migration, with special reference to the Family
Tyrannidae. Dissertation. Louisiana State University.
Baton Rouge, USA.

Chesser, R. T. 1 997. Patterns of seasonal and geographical
distribution of austral migrant llycatchers (Tyrannidae)
in Bolivia. Ornithological Monographs 48:171-204.
CoPETE, J. L.. R. Marine, and D. Bigas. 1999. Differences in
wing shape between sedentary and migratory Reed
Buntings Emheriza schoenichis. Bird Study 46:100-103.
Cory, C. B. and C. E. Hellmayr. 1927. Catalogue of birds
of the Americas. Field Museum of Natural History
Publications, Zoology Series 13, Part 5.

Debruyne, C. a., C. M. Hughes, and D. J. T. Hussell.
2006, Age-related timing and patterns of pre-basic
molt in wood warblers (Parulidae). Wilson Journal of
Ornithology 118:374-379.

Excoffier, L., G. Laval, and S. Schneider. 2005.
ARLEQUIN Version 3.1: an integrated software
package for population genetics data analysis. Evolu-
tionary Bioinformatics Online 1:47-50.

Fridolfssgn, a. K. and H. Ellegren. 1999. A simple and
universal method for molecular sexing of non-ratite
birds. Journal of Avian Biology 106:1 16-121.
Haberman, K., D. I. Mackenzie, and J. D. Rising. 1991.
Geographic variation in the Gray Kingbird. Journal of
Field Ornithology 62:1 17-131.

Joseph, L., T. Wilke, and D. Alpers. 2003. Independent
evolution of migration on the South American
landscape in a long-distance temperate-tropical migra-
tory bird, Swainson's Flycatcher Myianhus swain.soui.
Journal of Biogeography 30:925-937.

Lanyon, W. E. 1978. Revision of the Myiarchu.s Hycatch-
ers of South America. Bulletin of the American
Museum of Natural History 161:429-627.

Jahu et al. • TROPICAL KINGBIRDS

243

Low, M. 2006. Sex, age and .season influence morphomet-
ries in the New Zealand Stitchbird (or Hihi; Notio-
mystis cincta). Emu 106:297-304.

Lundberg. P. 1985. Dominance behaviour, body weight
and Fat variations, and partial migration in European
Blackbirds Turdus meriila. Behavioral Ecology and
Sociobiology 17:185-187.

Lundberg, P. and L.-O. Eriksson. 1984. Postjuvenile moult
in two northern Scandinavian starling Sturniis vulgaris
populations-evidence for difference in the circannual
time-program. Ornis Scandinavica 15:105-109.

Mila, B., R. K. Wayne, and T. B. Smith. 2008.
Ecomorphology of migratory and sedentary popula-
tions of the Yellow-rumped Warbler (Dendroica
coronata). Condor 1 10:335-344.

Pearson, T., M. J. Whitfield, T. C. Theimer, and P.
Keim. 2006. Polygyny and extra-pair paternity in a
population of Southwestern Willow Flycatchers.
Condor 108:571-578.

Perez-Tris, J., R. Carbonell, and J. L. Telleria. 1999.
A method for differentiating between sedentary and
migratory Blackcaps Sylvia atricapilla in wintering
areas of southern Iberia. Bird Study 46:299-304.

Pritch.ard, j. K., M. Stephens, and P. Donnelly. 2000.
Inference of population structure using multilocus
genotype data. Genetics 155:945-959.

PVXE, P. 1997. Identification guide to North American birds.
Part I. Slate Creek Press, Bolinas, California, USA.

Ralph, C. J., G. R. Guepel, P. Pyle, T. E. Martin, and D.
F. Desante. 1993. Handbook of field methods for
monitoring landbirds. USDA, Forest Service, General
Technical Report PSW-GTR-144. Albany, California,
USA.

Seutin, G., B. N. White, and P. T. Boag. 1991.
Preservation of avian blood and tissue samples for
DNA analyses. Canadian Journal of Zoology 69:82-
90.

Stotz, D. F., j. W. Fitzpatrick, T. A. Parker III, and D.
K. Moskovits. 1996. Neotropical birds: ecology and
conservation. University of Chicago Press, Chicago,
Illinois, USA.

TellerIa, j. L. and R. Carbonell. 1999. Morpho-
metric variation of five Iberian Blackcap Sylvia
atricapilla populations. Journal of Avian Biology
30:63-71.

Traylor Jr., M. A. 1979. Subfamily Tyranninae. Pages
186-229 in Checklist of birds of the world (M. A.
Traylor Jr., Editor). Volume 8. Museum of Compar-
ative Zoology, Cambridge, Massachusetts, USA.

Watson, C. J. W., A. A. Beheler, and O. E. Rhodes Jr.
2002. Development of hypervariable microsatellite
loci for use in Eastern Phoebes {Sayornis phoebe) and
related tyrannids. Molecular Ecology Notes 2:117-
118.

Zimmer, J. T. 1938. Notes on migrations of South
American birds. Auk 55:405-410.

The Wilson Journal of Ornithology 122(2):244-258, 2010

RED-COCKADED WOODPECKER MALE/EEMALE EORAGING
DIEEERENCES IN YOUNG EOREST STANDS

KATHLEEN E. FRANZREB'

ABSTRACT. The Red-cockaded Woodpecker (Picoides borealis) is an endangered species endemic to pine (Finns
spp.) forests ot the southeastern United States. I examined Red-cockaded Woodpecker foraging behavior to learn if there
were male/female differences at the Savannah River Site, South Carolina. The study was conducted in largely young forest
stands (<50 years of age) in contrast to earlier foraging behavior studies that focused on more mature forest. The Red-
cockaded Woodpecker at the Savannah River site is intensively managed including monitoring, translocation, and
installation of artificial cavity inserts for roosting and nesting. Over a 3-year period, 6,407 foraging observations covering
seven woodpecker family groups were recorded during all seasons of the year and all times of day. The most striking
differences occurred in foraging method (males usually scaled [45% of observations] and females mostly probed [47%]),
substrate used (females had a stronger preference [93%] for the trunk than males [79%]), and foraging height from the
ground (mean ± SE foraging height was higher for males [11.1 ± 0.5 m] than females [9.8 ± 0.5 m[). Niche overlap
between males and females was lowest for substrate (85.6%) and foraging height (87.8%), and highest for tree .species
(99.0%), tree condition (98.3%), and tree height (96.4%). Both males and females preferred to forage in older, large pine
trees. The habitat available at the Savannah River Site was considerably younger than at most other locations, but the
pattern of male/female habitat partitioning observed was similar to that documented elsewhere within the range attesting to
the species’ ability to adjust behaviorally. Received 24 March 2009. Accepted 9 November 2009.

The Red-cockaded Woodpecker (Picoides bo-
realis), endemic to the pine (Finns spp.) forests of
the southeastern United States, is strongly associ-
ated with open, mature longleaf pine (P. palnstris)
forest (USFWS 2003). Historically, this preferred
habitat has been maintained by fire without which
the hardwood midstory may develop to make the
area unsuitable for Red-cockaded Woodpeckers.
A reduction in the frequency of fire (Jackson
1986, Ligon et al. 1986, Walters 1990), loss of
longleaf pine forests (Conner et al. 2001), and a
general overall reduction in the age of forest
stands resulting from short-rotation ages (Wahlen-
berg 1960, Frost 1993) have resulted in modifi-
cation or loss of habitat throughout the geographic
range of the species. Red-cockaded Woodpeckers
are cooperative breeders living in social groups
consisting of the mated pair and often one or more
helpers, which are usually male and the offspring
of one or both members of the breeding pair. They
construct their own cavities in mature, living pine
trees, which are used for nesting and night
roosting (Jackson et al. 1979). This is a fairly
long-lived species in which all individuals in the
group cooperate in tetritorial defense, excavation
of cavities, incubation, and feeding nestlings and
fledglings (Hooper and Lennartz 1981 ). There is a
long period of juvenile dependency, which Ligon

' U.SDA .Southern Research Station, Department of
Forestry, Wildlife, and Fisheries, University of Tennessee,
Knoxville, TN 37996, USA; e-mail: fran7.reb@utk.edu

(1968) speculated was the result of the young
having to learn specialized foraging techniques.

The Red-cockaded Woodpecker population at
the Savannah River Site (SRS), South Carolina,
had declined to a low of four individuals by 1986
at which time an intensive management effort was
initiated to prevent extirpation of the species at
the site. Results and details of the management
program that was implemented are available in
Franzreb (1997). Most forest stands on the site
had been replanted with pine species when this
study was initiated in 1992 and —92% of the
pines were <40 years of age (K. E. Franzreb,
unpubl. data). Previous Red-cockaded Woodpeck-
er studies on foraging and habitat use relied on
areas that contained a fairly large proportion of
older trees in contrast to the younger age and
smaller size-class distribution patterns for trees at
SRS where old, large diameter trees were sparse.

The Red-cockaded Woodpecker population at
SRS is intensively managed including: removal of
nest site competitors such as southern flying
.squirrels (Gktncomys volans) from cavities; trans-
locations of woodpeckers from off-site locations
to augment the population; translocations on-site
to replace lost mates or establish new pairs; and
research to examine habitat requirements, home-
range size and configuration, and foraging
behavior. All birds were monitored, nestlings
were banded, and the reproductive rate was
estimated for each group on an annual basis
(Franzreb 1997). The Red-cockaded Woodpecker

244

Franzreh • RED-COCKADED WOODPECKER F'ORAGING IN YOUNG FORESTS

population increased on the site during the course
of this study from 36 to 89 birds, including 21
translocated birds from off-site (Franzreb 1997,
1999).

Previous investigations by others were con-
ducted in habitat largely consisting of stands with
relatively mature trees, offering a more structur-
ally complex vegetative environment with greater
opportunities to partition the habitat than likely
found in much younger forest stands. Further,
most studies of Red-cockaded Woodpecker for-
aging behavior and habitat use have not differen-
tiated observations by males and females; those
that did noted differences in foraging behavior.
Published studies have noted male/female varia-
tion in method of foraging, tree species selection,
use of hardwoods, foraging height from the
ground, foraging substrate, tree height selection,
location, and tree condition (Skorupa 1979,
Hooper and Lennartz 1981, Rudolph et al.
2007). However, Zwicker and Walters (1999)
found no difference in foraging in relation to tree
age, diameter at breast height, or species between
male and female Red-cockaded Woodpeckers in
the Coastal Plain of North Carolina. Porter and
Labisky (1986) suggest that differing foraging
requirements in different habitats necessitate
habitat-specific forest management guidelines
for Red-cockaded Woodpeckers.

The objectives of my study were to: (1)
examine habitat partitioning between foraging
male and female Red-cockaded Woodpeckers in a
habitat that largely consisted of stands of trees that
were considerably younger than most habitat
found elsewhere in the range; (2) compare habitat
availability with foraging use for males and
females; (3) assess differences among groups in
selection and preferences on the SRS; and (4)
discuss the possible role younger forest stands
may have in the range-wide management and
conservation of the species.

METHODS

Study Area. — ^The Savannah River Site, a
National Environmental Research Park, is in the
Upper Coastal Plain Physiographic Region in
Aiken, Allendale, and Barnwell counties. South
Carolina, USA. Most of the land was acquired by
the Department of Energy (DOE) in the late 194()s
and early 1950s to develop into a nuclear
production facility. Most of the site was in
agricultural use or recently had been harvested
for timber prior to acquisition by DOE. The

USDA, Forest Service has managed the natural
resources of the site for DOE since 1952. The area
managed for woodpeckers during this study
contained 31,970 ha of pine forest consisting ot
longleaf (37.7% of the pine), loblolly (P. taeda,
45.4%), slash (P. elliottii, 13.4%), and other pines
(0.2%), in addition to pine-hardwoods (3.3%)
(Glenn Gaines, U.S. Eorest Service, unpubl. data).
Most of the existing pine stands are the result of
replanting efforts undertaken in the 1950s,
although there are some residual older pine
trees. Approximately 92% of the pine trees
were <40 years of age (K. E. Eranzreb, unpubl.
data).

Foraging Data Collection. — ^Each Red-cockad-
ed Woodpecker on the site was banded with a
unique set of colored plastic leg bands for field
identification and also with a numbered aluminum
leg band provided by the U.S. Geological Survey.

I obtained the necessary endangered species
permits and banding permits from the U.S. Eish
and Wildlife Service, U.S. Geological Survey, and
State of South Carolina.

I selected seven groups of Red-cockaded
Woodpeckers for observation on SRS based
largely on their history of successfully breeding,
increasing the likelihood the group would be
present for the entire length of the study. Each
group consisted of a breeding male and breeding
female and often at least one helper. There was no
turnover in either member of the breeding pair for
two groups (groups 3 and 6). The male breeder
also remained the same for three of the other
groups (groups 1, 4, and 7). The breeding female
varied from a group that had no turnover during
the study, to changing every breeding season
(group 2). Group composition during most years
included at least one helper and there was no
group that did not have a helper at least one of the
years. The Red-cockaded Woodpecker population
on the site during this study increased from 36 to
89 birds, including 21 translocated individuals
from off-site (Eranzreb 1997, 1999).

Foraging observations were obtained from 5
May 1992 to 26 July 1995. Each foraging
observation consisted of; individual identification
(band colors), date, time, tree species, tree height
(m), tree condition (alive or dead), diameter at
breast height (dbh) (cm), tree age, foraging
method, foraging substrate (trunk, live limb, dead
limb, cone), foraging height from the ground (m),
timber stand location (compartment and stand
number), and weather conditions (percent cloud

246

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2, June 2010

cover, wind conditions, precipitation). Tree age,
height, and dbh are likely to be positively
correlated. These three variables are included
because forest managers use them in developing
management plans and there is a large amount of
variation depending on site quality and physio-
graphic region. The location of each observation
was defined using a global positioning system
(GPS) Trimble Pathfinder Professional (use of
brand names conveys no recommendation by the
U.S. Forest Service). Observations were obtained
for all members of the Red-cockaded Woodpecker
groups. Data were collected during all seasons of
the year and all times during the day. Birds were
observed from the time of leaving the roost cavity
in the morning and, if possible, until they returned
to the cavity at night. Each bird located was
followed until it made a foraging “strike”
(defined as an actual attempt to collect prey) at
which time the first observation was taken. At
least 15 min separated sequential observations of
the same individual to minimize inter-dependence
of subsequent observations (Hejl et al. 1990).
Porter et al. (1985) reported that a 15-min interval
between sequential observations was sufficient for
independence for Red-cockaded Woodpeckers. I
accounted for possible correlation among obser-
vations of the same individual instead of relying
on the informal 15 min rule.

Vegetation Data. — I used the Continuous In-
ventory of Stand Conditions (CISC) data base for
information on stands such as dominant tree
vegetation, stand age, and stand condition on the
SRS (U.S. Forest Service, unpubl. data). A stand
is a contiguous group of trees sufficiently uniform
in age-class distribution, composition, and struc-
ture growing on a site of sufficiently uniform
quality to be a distinguishable unit (Helms 1998).
More detailed information was obtained by
sampling vegetation in all stands within 800 m
of the nest tree of each of the Red-cockaded
Woodpecker groups studied. Most foraging ob-
servations occurred within 800 m of the nest tree
of each group. Hardwood and pine trees in each
stand were sampled using 0.04-ha fixed-radius
plots established along transect lines at a rate of
one plot per 2 ha of stand size following James
and Shugart (1970). The beginning of each
transect in a stand was established from a random
point. Sample stations were separated by 50 m
and transects were at least 50 m apart. The
following variables were measured in each 0.04-
ha plot for every tree: tree species, tree height (m).

tree condition (alive, dying, or dead), diameter at
breast height (dbh) (cm), age (increment bore
reading of 4,945 trees including 1,061 hardwoods
and 3,884 pines or CISC for even-aged pine
stands), location (compartment/stand/Red-cock-
aded Woodpecker group number), and specific
plot location.

Statistical Analyses. — A randomized block ex-
perimental design was used with repeated obser-
vations where blocks are the Groups and treat-
ment is Gender, which is tested with the
Group*Gender interaction as the error term.
Observations of each individual bird were repre-
sented as BirdID (individual band of a bird) and
nested within each Group*Gender. This is equiv-
alent to a randomized block repeated measures
design where observations of a given bird follow a
compound symmetry covariance matrix (common
variance and common covariance). There were
numerous observations of the same individual
during the 3-year study period, often ranging to
over 100 observations of certain individuals. I
used Proc Mixed (SAS 2004) for analysis that
considered nesting or clustering of repeated
observations for a given bird. This weighted the
observations of individual birds so those with high
numbers of observations would not unduly
influence the outcome and bias the results (Rao
and Scott 1992). This ensured that error terms and
variances would be calculated correctly, yielding
valid tests of hypotheses. 1 obtained least-squares
means (hereafter referred to as means) for all
continuous variables (foraging height, tree height,
tree dbh, and tree age) for males and females in
each group and for all groups combined. I tested
differences between means using an F-statistic
based on the mixed model using Proc Mixed.
Individuals with fewer than five observations over
the 3-year period were deleted from the analysis,
which affected the degrees of freedom. Groups
with the highest number of observations tended to
have the highest number of birds with observa-
tions of more than five and these groups had the
largest degrees of freedom indicated by the
denominator term. I examined each group sepa-
rately as well as combining data for all males and
for all females. The amount and distribution of the
habitat described by the variables studied differed
within the home ranges of each group. Compar-
isons of the data on a per group basis provide a
more complete representation of the range of
foraging behavioral differences for males and
females.

Frcinzreh • RED-COCKADEI) WOODPEC KER EORACJINCi IN YOUNCj l-ORESTS 247

I combined all observations of all pine trees
regardless of species and then grouped them by
age. height, and dbh class interval to allow for
comparison of the distribution of observations for
the continuous tree variables by gender. Tree
height and foraging height data were grouped by
3-m intervals. Age class data were grouped into
10-year intervals. There were few observations in
trees <30 years of age and the age class intervals
<30 were grouped. Tree dbh data were grouped
by 5-cm size class intervals. The distribution of
observations of males versus females was com-
pared for these variables as well as for the other
categorical variables (method, substrate, tree
species, and tree condition) using the T-statistic
in the Rao-Scott modified likelihood ratio test as
obtained from Proc SurveyFreq (SAS 2004) for
each group, and for males and females of all
groups combined. This considered the nesting of
the repeated observations for a bird in much the
same way as the mixed model did using Proc
Mixed for the other continuous variables. The
Rao-Scott modified likelihood ratio test uses the
survey design to adjust the typical Pearson Chi-
square test by dividing the P-value by the survey
design effect. The F-statistic used is an additional
modification that provides a better statistic as
suggested by SAS (2004). Degrees of freedom
varied depending on the number of categories
compared. Any cells with fewer than five
observations were merged for these tests with
the closest adjoining cell until there were at least
five observations. The clustering process in the
Rao-Scott tests also affected degrees of freedom.

I examined tree composition within an 800-m
radius circle with a group’s nest tree as its center.
The value of 800-m was .selected because it was
the same radius used by USFWS in its recom-
mended foraging guidelines for this species. I
estimated how many pine and hardwood trees of
specific ages, heights, and dbh were available
within each of these stands or portions thereof
using the 0.04-ha plot vegetation information. I
then estimated the availability of the resources
(tree species, tree heights, dbh, and tree age)
within the 800-m radius circle. These availability
data were used to derive estimates of how many
observations were expected to occur if birds were
using the foraging resources randomly. Arcinlo
coverages were used to aid in this process. The
percent use by each gender in each Red-cockaded
Woodpecker group of pine trees by tree species
and by tree age, tree height, and dbh class

intervals was compared to expected u.se based
on availability of the.se trees within the 80()-m
radius circle. Males and females in each group,
and all males versus all females combined were
compared with respect to the distribution of actual
observations versus the number ol ob.servations
expected. Expected values were based on resource
availability within 800 m of the nest tree and these
comparisons were used to examine any preference
or avoidance of foraging resources. Any cells that
contained <5 observations were either deleted or
grouped with similar cells so the overall number
of observations in each cell was >5.

I compared means for tree age and dbh for each
pine species for trees that were available in the
habitat of each of the groups and overall for all
groups combined using Proc GEM and the
BonfeiToni adjustment for multiple comparisons
(SAS 2004). All F-tests and comparisons of
means were performed at the 0.05 significance
level. Values reported are means ± SE.

I used Levin’s (1968) measure of niche breadth
where B = l/Ep", and /;, is the proportion of
observations recorded in class I followed

Schoener (1970) to estimate niche overlap for
each variable: % overlap = 100( 1 — 0.5 Z I F^., —
Pyj\) where P^j and F^, are the respective
frequencies for males and females in each class
for a given type of behavior. An overlap of 100%
indicates that males and females acted identically
in regard to the type of behavior examined,
whereas 0% overlap indicates completely non-
overlapping behavior.

RESULTS

There were 6,407 foraging observations of all
seven groups ranging from a low of 794 in group
7 to a high of 1,152 in group 2. Observations of
males (/; = 3,769) ranged from 403 in group 7 to a
high of 576 in group 6. Observations of females (/?
= 2,638) ranged from 269 (group 4) to 502 (group
2).

Method of Foraging. — Scaling was the most
commonly used foraging method for males in all
seven groups, ranging from 40 to 52% of the total
observations (// = 3,769; Eig. 1 ), followed closely
by probing. However, females in five of the seven
groups used probing more commonly than scaling
(/? = 2,638; Fig. 1 ). Foraging method significant-
ly differed between males and females for all
groups combined, primarily because females
tended to probe more {F^ 2')i — 2.89, F = 0.04).

248

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2, June 2010

Group Number

05

c

g

ro

5

0)

o

c

Q)

L

05

CL

FIG. 1. Foraging method of male and female Red-cockaded Woodpeckers at the Savannah River Site, South
Carolina, 1992-1995.

Suhso-ate. — Males strongly preferred the trunk
as the foraging substrate with 73 to 87% of
observations occurring there (Fig. 2). For all
groups combined, 79% of all male foraging was
on the trunk surface, followed by 10% on a live
limb, and 1 1% on a dead limb (Fig. 2). Females in
each group had a stronger preference for foraging
on the trunk than males and used that surface a
minimum of 87% of the time (Fig. 2). Overall,
93% of all female foraging observations were on
the trunk, followed by 4% on a live limb, and 2%
on a dead limb. Occasionally males and females
were observed foraging on cones (19 instances for
males, 7 instances for females), and there was one
instance of a female foraging on the ground.
There was a significant difference in foraging
substrate .selection in every group and for all
groups combined between males and females (all
F values > 3.79, all P < 0.03).

Tree Species Selection. — At least 98% of all
foraging by males and females, regardless of
group, occurred in pine trees. Overall for males
there were 27 (0.7%) instances and up to 1% of
observations for groups foraging in hardwood
trees. Females were observed foraging in hard-
woods on 20 instances (0.8%) and 2% of
observations by group.

There was a decided preference by both males
and females to forage in longleaf pines and this
did not .seem to be the result of tree dbh or age.

The mean overall dbh values for pine species
showed that, on average, longleaf pine trees
available at SRS were 1 .0 cm larger than loblolly
pines and 2.8 cm larger than slash pine trees.
Mean tree age of longleaf pine exceeded that of
loblolly pine by 0.6 to 8.5 years but, in some
groups, the mean age of the available loblolly
pines exceeded that of longleaf by 0.3 to 5.0 years.
For example, in group 1, loblolly pines on average
were bigger (larger dbh) and older than longleaf
pines. The latter species was most frequently used
for foraging by males in all groups (range 69 to
90%) and accounted for 79% of all male
observations (n = 3,743). Females also preferred
to forage in longleaf pine trees, in which 78% of
ob.servations occurred (range 62-94% in the 7
groups). Loblolly and slash pine also were used,
but to a much less extent. There was a significant
difference in tree species selection by males and
females in two of the seven groups (F values ^
4.60, P values < 0.02); however, considering all
male and female observations combined, there
was no significant difference in tree species
selection (F2.196 = 0. 1 1, F = 0.91).

Tree species selection by males differed from
expected on the basis of tree species availability
in all groups and for all males combined (all F >
4.23, all P < 0.04). Females in all but group 2 did
not use tree species in the same proportion as
available (all F > 4.79, all P < 0.03). Males in all

Franzrcb • RED-COCKADEI) WOODPECKER FORAGING IN YOUNG FORESTS 249

Group Number

I I Trunk

v//\ Live limb
I I Dead limb

HTTTTTI Cone

FIG. 2. Substrates used by foraging male and female Red-cockaded Woodpeckers at the Savannah River Site, South
Carolina, 1992-1995.

groups and females in five of the seven groups
foraged more in longleaf pines than expected
based on availability.

Foraging Height. — ^Mean foraging height of
males ranged from 9.5 ± 0.6 m in group 1 to 12.7
± 0.4 m in group 5 (Table 1). Mean foraging
height of females ranged from 7.5 ± 0.5 m in
group 1 to 11.8 ± 0.4 m in group 5 (Table 1).
Mean foraging height for males in two groups was
higher than for females {F > 6.0, P ^ 0.03;
Table 1). Foraging height for males for all groups
combined was higher (11.1 ± 0.5 m) than for
females (9.8 ± 0.5 m; F\ (, = 17.67, P = 0.006;
Table 1).

Tree Height. — ^Mean heights of trees selected
for foraging by males ranged from 17.6 ± 0.6 m
in group 1 to 21.5 ± 0.3 m in group 5. Mean tree
height selection by females ranged from 17.1 ±
0.6 to 21.3 ± 0.3 for groups 1 and 5, respectively.
Mean tree height selection for all male observa-
tions combined (19.6 ± 0.5 m) was not different
than for females (19.6 ± 0.5 m; F\fi = 0.02, P =
0.90). There was no difference in mean tree
heights between males and females in five of the
seven groups (all F 0.42, all P ^ 0.53) or
between males and females within any group [F
> 4.49, p > 0.05).

There was no difference for five of the groups
comparing the distribution of observations for
male versus female tree height selection (all F <
1.39, all P ^ 0.24). Nor was there a difference

between males and females for all groups
combined in distribution of observations by tree
height (^7.693 = 0.28, P = 0.96).

Males and females in every group and for all
groups combined did not randomly select trees by
height (all F > 10.56, all P < 0.001). Both males
and females avoided shorter trees (those less than
12 m), either avoided or used in proportion to
availability mid-sized trees (12-18 m), and
strongly prefen'ed trees >18 m in height.

Tree Condition. — Both males and females
heavily foraged on live trees (average of 90% of
male and 91% of female foraging observations).
There was no difference in selection of trees by
condition (alive, dead, or dying) between males
and females in five of the groups and for all
groups combined (F s 2.33, P ^ 0.12). There was
a difference between males and females in two
groups (F2,34 = 6.06, P = 0.006; Fj 12 = 53.2. P
< 0.0001; groups 2 and 7, respectively) and males
in both groups used dying trees more than
females.

Diameter at Breast Height (dhh). — ^There was a
difference in mean diameter at breast height of
trees selected by males versus females for two of
the seven groups. Females in group 6 (F| 14 =
6.23, P — 0.03) and group 7 (Fi n = 9.22, P =
0.01) selected trees with larger dbh than males.
Overall, there was no difference in mean dbh of
trees selected for foraging by males and females
in all groups combined (Fj ^, = 0.39, P = 0.55).

250

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2, June 2010

TABLE 1. Foraging height

(m) of male

and female

Red-cockaded Woodpeckers by seven

groups and all groups

combined at the Savannah River Site, South Carolina, 1992-1995.

Percent observations

Group 1

Group 2

Group 3

Group 4

Foraging height

M

F

M

F

M

F

M

F

<3

4.9

16.1

5.7

13.0

2.3

9.1

4.7

10.4

>3-6

13.5

23.2

15.1

14.3

8.9

16.5

10.3

17.8

>6-9

25.8

25.7

22.0

16.9

16.9

18.2

20.0

21.6

>9-12

26.4

14.9

21.7

20.5

23.0

16.2

24.7

20.1

>12-15

18.4

13.4

16.6

16.1

27.3

17.6

16.1

15.2

>15-18

7.4

5.5

10.8

1 1.6

16.2

15.6

14.6

7.4

>18-21

2.2

1.0

5.9

6.6

4.6

5.4

6.9

5.6

>21-24

0.8

0.0

1.2

0.2

0.7

1.1

1.7

1.5

>24

0.2

0.0

0.5

0.0

0.0

0.3

0.7

0.4

Ground or no data

0.4

0.3

0.6

0.8

0.0

0.0

0.3

0.0

Sample .size

509

398

646

500

561

.352

589

269

LSMeans ± SE

9.5 ± 0.6

7.5 ± 0.5

10.4 ± 0.3

9.8 ± 0.3

1 1.7 ± 0.7

9.0 ± 0.8 1

1.1 ± 0.5

9.6 ± 0.7

F value”

T,.n

= 6.54

F 1.16 ”

= 1,67

F^.u

= 6.00

T|,,4 =

= 3.00

P

= 0.03

P --

= 0.21

P

= 0.03

P --

= 0.10

Group 5

Group 6

Group 7

All groups combined

M

F

M

F

M

F

M

F

1.9

6.2

7.8

13.4

3.2

12.0

4.5

11.5

3.4

8.5

14.8

22.6

9.2

14.3

1 1.0

16.3

1 1.5

14.7

17.4

14.7

22.3

20.5

19.5

18.8

23.9

23.9

22.7

20.6

26.1

20.0

23.9

19.5

32.1

23.2

20.3

15.1

20.1

15.9

21.3

16.8

18.5

15.6

10.8

9.6

10.7

9.2

12.7

10.9

6.5

5.8

5.4

3.1

6.5

5.9

5.4

5.0

1.1

0.5

0.4

1.0

1.7

2.3

1.1

0.7

0.2

0.0

0.2

0.0

0.3

0.0

0.3

0.1

1.1

1.6

0.4

0.0

0.0

0.0

0.4

0.4

472

428

575

292

402

390

3,754

2,629

12.7 ± 0.4

1 1 .8 ± 0.4

10.5 ± 0.6

9.4 ± 0.8

1 1.9 ± 0.9

10.8 ± 0.84 1

l.l ±0.5

9.8 ± 0.5

T,.,3 ^

= 2.38

F 1.14 =

= 1.22

T,,n ^

= 0.78

T,,6 =

17.67

P ■■

= 0.15

P =

= 0.28

P

= 0.40

P =

0.006

F and df values from Proc Mixed model; Note for dl ot x and y as given in Fy^ y. x represents k — I and _v is calculated by the clustering program

There was a difference between males and
females in two of the groups when comparing
tree selection by distribution of observations by
dbh size class with females using trees with larger
dbh (group 6, = 3.78, P = O.OOl ; and group

7, T7_s4 = 5.39, P < 0.0001 ). There were no male/
female differences in distribution of observations
for dbh size class intervals in the other five groups
and for all groups combined (all F values < 1.31,
P > 0.26).

Males and females did not randomly select
trees on the basis of dbh (all F > 9.68, all P <
O.OOl ). Both males and females avoided foraging
in trees <20 cm dbh and displayed strong
preferences for the moderate and larger dbh size
classes.

Longleat and loblolly pine trees were the two
most common pine species in the habitat. The
mean dbh for longleaf pine trees available to three
groups was higher than for loblolly pine (Ta-
ble 2); however, loblolly pine for two groups had
higher mean dbh values (all P < 0.001). There
was no difference in mean dbh values between
longleaf and loblolly pine trees in the remaining
two groups (Table 2). The available longleaf
pines had a higher mean dbh (longleaf pine •=
17.9 ± 0.2 cm vs. loblolly pine = 16.8 ± 0.2 cm,
P = ().()()4; Table 2) for all groups combined.

Tfcc Age. — Females in group 5 (mean = 39.9
± 0.2 years vs. mean = 39.3 ± 0.2 years; F\ =
5.27, P = 0.04) foraged more on older trees than
males (Table 3). There was no difference between

TABLE 2. Tree species availability (least-squares means ± SE) for tree age (yrs) and diameter at breast height (dbh) (cm) for foraging Red-cockaded Woodpeckers at the
Savannah River Site, South Carolina, 1992-1995.

Franzreh • RED-COCKADED WOODPECKER FORAGING IN YOUNG FORESTS 251

— u
>

OO O
OS O
p

lo o

cN r-

lO Tf so

o — — Os
— (N d ^

so

sd

m

m o —

00
p —
— Tt

so

r-1 o

II

V II

II II II

II V II

11 V II

II V II

II

II II

II

II

II II

k.

a, s:

Ci. a, s:

k. a. =:

k, k. =:

k.

k, C

k.

n

F

<

u

CN

u

p

d

p

<N

(N

+ 1

+1

+ 1

p

p

p

rn

CN

1

m

<

<

ffi

o

p

OO

d

+1

+1

+ 1

00

p

p

i/S

d

d

CQ

<

CQ

CQ

CQ

<

CQ

r-

so

00

p

p

(N

d

d

d

d

d

— -

d

d

+1

+1

+1

+1

+1

+ 1

+ 1

+1

p

—

p

P

p

00

os

00

so

CN

<

<

<

<

<

<

<

<

CN

r--

r--

os

so

(N

d

d

d

d

d

d

d

d

+1

+1

+1

+ 1

+1

+1

+ 1

+1

o

o

o

d

V

0.

00

<N

(N

00

so

(N

o

(N

so

a N

d

O

O

o

so

o

r-

o

os

r-

o

p

so

O)

OS

O

O

o

o

(N

p

o

o

o

vn

O

p

O

O

o

m

in

p

r^

in

o

o

o

r-

o

00

O)

<N

Os

os

CN

Os

o

o

p

o

r-

p

d

00

d

d

—

(N

d

00

d

in

(N

d

CN

d

—

d

(n

d

in

11

V

11

11

A

11

11

V

11

11

V

11

11

V

11

11

V

11

11

A

II

II

V

II

Ck

c:

0.

s:

c

s:

a.

Ck

a.

s:

U

p

(N

■fl

rn

<

r--

U

p

—

(N

+ 1

+ 1

o

m

d

d

<N

Q

<

<

<

cn

p

OS

d

d

+ 1

+ 1

+1

p

p

00

d

in

(M

(N

(N

[-

CQ

<

CQ

CQ

OQ

CQ

<

CQ

o

Os

m

p

(n

m

—

d

d

d

d

d

+1

+1

+ 1

+ 1

+ 1

+ 1

+ 1

+ 1

—

p

P

p

oq

p

00

in

d

d

os

d

d

d

<N

CnI

cn

CN

<

<

<

<

<

<

<

<

in

rn

so

Os

os

00

CN

d

d

d

d

d

d

d

d

+i

+ 1

+ 1

+ 1

+1

+ 1

+ 1

+ 1

p

OO

p

r-

p

p

(N

d

d

d

d

d

d

d

sd

r-i

m

(N

m

(N

(N

g

o

a:

r-i m

lo so

c/:

Q,

3

O

oj:)

<

Xt

f-value based on Proc GLM comparison of least-squares means with Bonferroni adjustment for multiple comparisons.

Significant difference comparing means of individual tree species with Bonferroni adjustment; same letters indicate no significant difference.

252

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

in

d'

S

>

"r3

>

OO

D

r3

>

D

CJJ

r3

■U

GJ

GJ

Cij

>%

GJ

O

Ci)

£

"o

CJ

GJ

GJ

-1^

O

GJ

CL

■o

c

"3

GJ

*CJ

r3

O

o

0

1

•a

G>

c

-3

C

GJ

3^

I

r I

o^

UJ

-J 6
CO E

^ C

U

o

vO

(N O On CN O r<^
CN (N CN (N

On 00 OO
O O

SO

m

r<)

O

m*

ON

NO

r-

o

o

r--

ON

II

p

p

p

p

p

m

q

o

p

d

d

d

II

m

in

mi

00

NO

o

d

o

m.

NO

+ 1

II

II

II

II

(N

rA

[y

o

(N

GN

d

in

1.14

cu

vr,

r^’,

[>

r*\

o

Jj

O

d

p

N<.

-2

r3

V

a

E

II

p

q

o

ON

—

o

o

o

—

E

a.

,GJ

a.

Tf — O
— ^ r-'

vO Tf Ov o ON VY

— On O -—

rj n (N

o o o

(N +1
ON

IT) 00

00

o

d

r- — o
odd —

oo

oo

sD

cn

r^>

00

c^i

c

d

m O
m

O rn
d iri

sOOOsqONr'if^Oir)

o d (N — r-’ — d d

p

On

NO

NO

OO

(N

—

O

r-

(n

00

II

+1

d

d

d

d

II

o

oo

II

II

II

II

X

fNj

o

rr

o

d

in

(N

a.

in

vq

0.

d

o

»n

d

-2

d

(n

ti.

GJ

V

C3

V

"£

E

+ 1

E

a.

cx.

nO Os — ITi so ON
— 00 sd — *

(^1 IT'j

^ o o
d d d

NO

m <N
in

ON

r^

00

00

NO

r-

o o
d d

o o

On (N On

nO d d

r- o o
d d d

r--

o o o
d d d

p

NO

CN

ON

NO

(N

p

o

ON

+ 1

o

d

d

d

II

r^i

11

II

II

II

fN

O

rs

k.

c

mi

■T

a.

c

f*'.

a.

mi

(N

d

k.

GJ

II

a

ni

E

0.

_o o

I II

+ 1

‘T'i

p

p

r^j

—

C^l

p

(n

o

<n

OO

p

p

oo

p

P

m

(N

CN

00

r<'i

—

sd

rd

—

d

d

d

ON

00

mi

mi

rd

00

r^i

d

d

d

c^-

cTi

o

<n

m

CO

00

ON

m

p

o

ON

m

00

‘

m

—

r-

II

o

o

O

c^i

o

p

p

o

o

r“i

rn

d

d

II

Q

m

o

cn

m

O

--

p

o

in

o

d

d

d

ri

>-s 1

mi

—

■ 4

d

d

d

+1

II

II

II

II

rj /— N

(N -—

Q.

3

d

d

d

d

NO

tri

d

d

d

p p o
d d d iri

r-- r^i NO nO

-t — 00 d

— r^i

o

d

lO

+1

o

ON

iT'i

00 0^

u.

Z1 O

Tf O

k.

c

_b'

d

^ V

E

V

E 0,

k.

UJ

00

c

C3

a

£

>

GJ

2 I

o o

? t

GJ

+1

£

C3

E

y>

C

GJ

u.

■3

o

o

o

o

o

-2

Cl

c3

GJ

C3

CL

cn

GJ

in

nO

r-

00

J.

ON

o

ON

£

C3

S

00

E

o

CxC.

in

NO

r--

00

A

m

J

U

T3

G^

O

D

a.

X

GJ

3

ro — — 00
d lO oS 00
(N rj

r-; p p

d d — — d

r-;rsj(^|pON'^(NOro

— dddrsiodrodro

ro — (N —

(N

ro OO

d d

NO r-
rN —

00 ro 00 —

(N

o

*r^o

NoddroodcNrod— •
m ro

pooor-r-(NOO
ddddd — lodd

<N — NO

ppppr-^cNNOOO
ddddddrsidd
(N — NO

^ppGNp'^ON'^O
— ONONdNONOCNGNrO
(N (N ro —

—

d —

ON oo —
in d d NO
fN ro (N

p Tt Tt oo m NO —
ln CN c^i 00 in NO —
— — CO (N

o o
d — ^

O NO

d d

o o o c —
d d d d

NO

in

00 —

O CN o
d d d

TABLE 3. Continued.

Franzrch- RED-COCKADED WOODPECKER EORACJINCi IN YOUNCi FORESTS 253

— O

d d

On

r-

(N

CN

(N

p

+ 1

d

d

d

11

II

ON

ON

\q

0.

p

(N

p

+ 1

IT)

d

ON

ON

O

O

O

d

V

CL.

oo"

00

o

00 ^

ON

o

o

o

d

V

CL.

o o
d d

O

+ 1
CO

o6

00
d d

sc

NO

o

o

o o

o8

V o
a, !l

+1
(N ^

ON

o ^
o .

iri CO
00 o

—

^ " so

O ^ CA

II tc' -

t- .
a> c3

iT^ £

^ G eCJ

in o
d d

o

o

o

d

V

p

O)

o

d

II

o

o

o

+1

II

O.

d

p

d

d

d

rt

II

On

r-

o

II

5C

r^i

d

d

—

d

(N

ti.

o

o

+1

II

d

d

<NJ

CO

1/*,

cz

-T-

E

o

o

o

d

V

CL,

o o

o o
d d

d mi
+1 II

ON

o

o

o

c

d

V

CL.

r-

r-

o o
d d

O'

ri

d

+1

(n

O'*

r*-,

k.

O

u.

00

+1

c

<u

£

D

•o

(U o
2 CL

£ X

qj ^

.

^ >
> _
(u 2

I ^

C cz

cz cz

(U Q.

2 E

CO O

-J U

7.

^ a

^ 'Z
■S S
S .2

2

e i

ji -d

d

JZ

U X)
w c
CL =
X 3
LU Z

Eqj

X

= E

o ■£

r3

Qi

3 C«i x
-j 2

males and females for six of the seven groups and
for all groups combined in tree age classes (all F
< 1.79, all P > 0.13, Table 3). Neither males nor
females randomly selected trees by age class (all
f > 4.25, all P < 0.002). Both males and females
used younger trees (<30 years of age) less than
expected and preferred trees >30 years of age
(Table 3).

Mean age was higher for longleaf pine than
loblolly pine in three groups {P < 0.0001) within
the 800-m radius circles centered on the nest tree
of each group with the reverse true in two groups
{P < 0.0001; Table 2). There was no difference in
mean age between available longleaf and loblolly
pines in groups 2 and 7 (Table 2). Mean tree age
of longleaf pine (26.2 ± 0.2 years) was higher
compared to loblolly pine (23.3 ± 0.3 years, P <
0.0001; Table 2) combining the 800-m circles of
all groups.

Niche Breadth and Overlap. — ^Foraging vari-
ables that had the narrowest niche breadth were
the same for males and females: tree species
selection (1.54 vs. 1.57), tree condition (1.24 vs.
1.20), and substrate (1.56 vs. 1.15). Both males
and females were most generalized in tree dbh
(6.25 vs. 6.54), foraging height (5.78 vs. 6.41),
tree height (4.85 vs. 4.91), and tree age (5.55 vs.
4.55). Niche overlap between males and females
was >90% for foraging method (90.8%), tree
species (99.0%), dbh (95.7%), tree height
(96.4%), tree condition (98.3%), and tree age
(90.9%). The least overlap between males and
females was for substrate (85.6%) and foraging
height (87.8%).

DISCUSSION

Results of foraging ecology studies of male and
female Red-cockaded Woodpeckers in regard to
foraging method have not been consistent (Ramey
1980, Hooper and Lennartz 1981, Rudolph et al.
2007). In my study, males scaled more than
females in two groups, whereas females engaged
more in probing. Red-cockaded Woodpeckers at
SRS did not forage by excavating to the same
extent observed at other locations, all of which
had more mature pine habitat. It may be that
young stands at SRS did not provide the deep,
furrowed bark of older trees that would encourage
more excavation. Culmen length is significantly
longer in males than females (Mengel and Jackson
1977), which may contribute to the differences
that have been observed. However, the Red-
cockaded Woodpecker is one of the least sexually

254

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

dimorphic species of Picoides and, although
difference in beak size could be an adaptation
for differential foraging behavior, culmen length
would seem to be insufficient to obviate the need
to have other means to reduce intraspecific
competition.

It has been reported that both males and
females forage mostly on the trunk (Ligon 1968,
Skorupa 1979, Ramey 1980, Hooper and Lennartz
1981). My results showed that males spent more
time foraging on limbs than females, whereas
females more frequently foraged on trunks.
Ramey (1980) found variation in foraging site
was the main difference between males and
females, and there was substantial foraging in
hardwoods, particularly by males. Pizzoni-Arde-
mani (1990) reported that male Red-cockaded
Woodpeckers have longer beaks, toes, and legs,
and shorter tails than females, which may be
advantageous when clinging to branches and
twigs on limbs. The shorter legs and longer tail
of females may be beneficial when foraging on
trunks (Pizzonni-Ardemani 1990), the common
female foraging substrate.

Red-cockaded Woodpeckers forage on living
pine trees and rarely u.se non-pine substrates
(Ligon 1968, Hooper and Lennartz 1981, De-
Lotelle et al. 1983, Porter and Labisky 1986);
however, DeLotelle et al. (1983) also found that
10% of foraging occurred in pond cypress
(Taxodium ascendens) stands. Both males and
females at SRS heavily used pine trees and at
least 98% of all observations occurred there,
similar to that reported by other studies (Ramey
1980; Zwicker and Walters 1999; Walters et al.
2000, 2002; Rudolph et al. 2007). 1 detected no
difference between males and females in four of
the groups and in all groups combined in u.se of
different pine species. However, based on
availability, both males (all groups) and females
(5 of 7 groups) foraged in longleaf pine trees
more than expected. Previous work on tree
species selection produced conflicting results as
to whether Red-cockaded Woodpeckers prefer a
particular pine species (Nesbitt et al. 1978,
Porter and Labisky 1986). Tree species selection
can be confounded by tree age and size,
presence of a hardwood midstory, and density
of trees in the area.

All studies that examined foraging height of
Red-cockaded Woodpeckers have reported that
males forage higher in trees than females (Hooper
and Lennartz 1981, Eng.strom and Sanders 1997,

Rudolph et al. 2007). However, in my study this
was true of only two of the seven groups and all
groups combined. It may not have been apparent
that for most of the groups there was no difference
in foraging height for males and females if only
combined data had been analyzed. Habitat
partitioning by foraging height could reduce
competition for prey and permit specialization
by gender so that trees are used more efficiently.
Males are dominant to females and may be using
the most productive portions of the foraging
substrate.

Porter and Labisky (1986) reported that when
Red-cockaded Woodpeckers foraged in stands
that were usually avoided, they selected trees of
greatest height and dbh. Ramey (1980) detected
no significant difference in tree heights used by
males and females. I also found there was little
difference in tree height selection between males
and females. Both strongly preferred taller and
avoided shorter trees, probably reflecting the
larger, more deeply fissured bark surface area of
the taller trees.

Dying and recently dead pine trees, especially
those with southern pine beetle {Dendroctomis
frontalis), are an important foraging resource for
Red-cockaded Woodpeckers, but the reported
response by males/females has not been consis-
tent. Hooper and Lennartz (1981), Repasky
(1984), and this study found that male and female
foraging was similar. Five of the groups and all
groups combined in my study had no difference in
foraging .selection between males and females
based on tree condition. Skorupa and MacFarlane
(1976), Skorupa (1979), and Repasky (1984)
reported that males foraged more frequently than
females on dead and dying trees. The differences
in the response may reflect the level and
occurrence of insect infestations.

Hooper and Lennartz (1981) reported niche
overlap between males and females was highest
for tree condition (99%) and dbh (97%). Most
overlap in my study occurred in tree species
(99.0%) and tree condition (98.3%). Foraging
height overlap was low (56%) in the study by
Hooper and Lennartz (1981), but was 87.8% in
my study. Higher niche overlap for foraging
height and also similarity in foraging height for
five of the seven groups in my study reflected the
considerably younger forest than that studied by
Hooper and Lennartz (1981) where there was less
opportunity for vertical divergence in foraging
height between males and females.

Frcmzreh • RED-COCKADED WOODPECKER FORAGING IN YOUNG FORESTS 255

Hanula et al. (2000) sampled 300 living pine
trees at SRS during the same time frame in which
my study was conducted and found that arthropod
biomass/tree increased with increasing stand age
up to —65-70 years, but the arthropod biomass/ha
was highest in the youngest stands. Abundance
and biomass of arthropods on each tree trunk was
positively correlated with bark thickness and tree
diameter, and negatively correlated with basal
area (mVha). There was no correlation of
diversity, abundance, or biomass with arthropods
on the tree trunk with site index, numbers of
herbaceous plant genera in the understory, number
of herbaceous plant stems, or percent ground
cover by herbs. The characteristics of the stands,
including average bark thickness and dbh, asso-
ciated with greater arthropod abundance and
biomass on the bark are positively correlated with
tree age, but are subject to change depending on
how stands are managed (Hanula et al. 2000).
Current management guidelines at SRS include
prescribed burning at least 2-3 times/ 10 years in
active Red-cockaded Woodpecker clusters, mid-
story tree/shrub removal, thinning of overstory
pines, suppression of pine beetle infestations, and
extensive recommendations for regeneration (Ed-
wards et al. 2000).

Hooper and Lennartz (1981) and Zwicker and
Walters (1999) reported Red-cockaded Wood-
peckers preferentially selected the largest trees
with respect to dbh with no differences between
males and females. My results were similar for
five of the groups and for all groups combined.
Both males and females consistently selected
larger dbh trees based on availability, far more
than expected on the basis of their presence in the
habitat. Larger trees have more arboreal surface
area available in which to forage per visit and
probably richer resource patches/unit area than
smaller trees.

All researchers who have investigated individ-
ual tree selection have found that large old trees
are preferred over smaller, younger ones (De-
Lotelle et al. 1987, Jones and Hunt 1996, Hardesty
et al. 1997, Walters et al. 2002). Age and size of
pine trees are highly correlated until about age 80
(Platt et al. 1988). It is not clear whether tree size
(dbh), age, or both is the more important
determinant of tree selection by foraging Red-
cockaded Woodpeckers. There was a significant
difference between males and females in my study
in mean tree age selected in four of the groups and
in all groups combined; however, it was not

always the same gender that had the stronger
selection for older trees. Females tended to select
trees with higher dbh values. Females spend more
time foraging on the trunks than males, and it is
likely this difference reflects females taking
advantage of the additional foraging surface area
provided on larger trees. Both males and females
avoided foraging in trees <30 years of age with
some groups not using them. There was a strong
preference by both males and females to use trees
at least 60 years of age in my study. Males and
females still preferred to forage on older trees
even though a majority of stands at SRS were
considered young.

There was little difference in dbh size among
the pines at the SRS, but longleaf pines on
average were strongly preferred as foraging
substrate by both males and females. Loblolly,
the second most commonly available pine species
within the home ranges next to longleaf pine,
tended to be selected against. The mean age of
longleaf pine in some groups exceeds that of
loblolly pine, but in other groups the opposite is
true. Males (69.4% of observations) and females
(72.7% of observations) strongly selected longleaf
pine trees in relation to the predicted use of
longleaf pine (46.2%). Red-cockaded Woodpeck-
ers evolved in a mixed-age forest containing only
a relatively small proportion of young pine trees
(Frost 1993) and in a landscape dominated by
longleaf pine; thus, they may be more adapted to
foraging on surfaces of old, large pine trees. It is
logical to assume they are best equipped morpho-
logically and behaviorally to use longleaf pine.
The strong preference by both males and females
to use longleaf pine does not appear to be strictly
the result of tree size or age.

Habitat segregation observed in foraging Red-
cockaded Woodpeckers may be designed to
reduce competition between males and females
to allow more efficient use of the available
foraging habitat. The more socially dominant
males may cau.se females to use less productive
areas. It is plausible that partitioning by gender in
this species may be an efficient mechanism to
facilitate cohesion of the group by reducing male/
female competition and aggression. Breeding
males at SRS tended to hold their territories until
they disappeared (and were presumed dead).
Females were more flexible in maintaining their
pair bonds. Thus, it would be beneficial to the
group to minimize aggressive encounters and to
maximize .social cohesion of group members. This

256

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

may be the mechanism that developed differences
between males and females as documented at this
younger growth site and elsewhere.

Red-cockaded Woodpecker mean reproductive
success (fledglings/pair) at SRS ranged from 1.4
to 3.4 with the highest rates (2. 8-3. 4) observed for
groups with at least 10,000 pine stems >25.4 cm
dbh within 800 m of the nest cavity (Franzreb
2004). Reproductive success was significantly
related to number of available pine stems
5:25.4 cm dbh (Franzreb 2004). Typical mean
reproductive rates in other parts of the range were
1.34-1.67 fledglings/pair (Conner et al. 2001).
Higher rates at SRS, rather than due mainly to
habitat quality, are possibly the result of the
unusually intensive management of the species on
the site, including installation of artificial cavity
inserts, removal of cavity competitors, translocat-
ing birds to re-establish a pair where one member
has been lost, frequent monitoring of cavities to
detect damage to artificial inserts, and frequent
monitoring of all individuals in the population
(Franzreb 2004).

Subsequent to initiation of these management
activities, the population level has increased to
— 196 adults and fledglings in 2006 (John Blake,
pers. comm.) from a low of four birds in 1986
(Franzreb 1997). Development of the technique to
install artificial cavity inserts for the Red-
cockaded Woodpecker was pioneered at SRS
(Allen 1991) and installation of the.se devices has
been a widely used management tool in many
parts of the range. Artificial in.serts at SRS were
used by 47% of the birds in comparison to 53%
that used naturally excavated cavities for the nest
location during my study (K. E. Franzreb, unpubl.
data). It is likely that Red-cockaded Woodpeckers
would not have persisted at SRS without the
assistance provided by these management actions.

Current federal guidelines for managing Red-
cockaded Woodpecker foraging habitat are com-
plex and address factors including type of forest
regeneration, rotation age, basal area, and home-
range size needed per group (USFWS 2003).
Management and recovery of the Red-cockaded
Woodpecker have focused on providing suitable
foraging habitat in more mature and older pine
stands than typically found at SRS. The results of
my study indicate that younger forest stands, as
long as they are at least 30 years of age, may
provide acceptable foraging habitat and should
not automatically be regarded as unacceptable or
sub-optimal.

Younger trees (preferably at least 40 cm dbh) in
which managers have either installed artificial
cavity inserts or drilled cavities are necessary to
provide for nesting and night roosting if older
trees are insufficient for natural cavity excavation.
It is preferable to provide trees of sufficient size
so birds can create their own cavities rather than
rely on artificial inserts. Allowing pines to age to
>60 years, preferably even older, provides better
foraging habitat and trees of sufficient size for
birds to excavate their own cavities. Maintaining
the site in just young pines will not maintain a
self-sustaining population without continued in-
tensive management actions as there will be
insufficient trees for natural cavity excavation
nor would there be the larger trees that are
preferred for foraging. The need to provide larger,
older pines has been recognized in the SRS
management plan for the Red-cockaded Wood-
pecker, which calls for a number of proactive
actions including a rotation age of 120 and
100 years for longleaf pine and loblolly pine
stands, respectively, in the main Red-cockaded
Woodpecker management area (Edwards et al.
2000). Older pines will be optimal foraging
substrate for both male and female Red-cockaded
Woodpeckers and will provide the large trees
needed for natural cavity excavation.

ACKNOWLEDGMENTS

This research was funded by the Department of Energy,
SRS, through the USDA Forest Service Savannah River
under Interagency Agreement DE-IA09-76SR00056 and is
gratefully acknowledged. I am especially grateful to Steven
Schulze for field support and for generating the data bases
to handle a large volume of behavioral and vegetation data.
1 also thank my field staff whose support has been crucial to
the success of the Red-cockaded Woodpecker research at
the site. I thank C. H. Greenberg, J. A. Jackson, D. C.
Rudolph, and John Blake for reviewing the manuscript and
Stan Zarnoch for statistical review. The cooperation of the
USDA Forest Service Savannah River is gratefully
acknowledged (John Irwin, Elizabeth LeMaster, William
Jarvis, Glenn Gaine.s, and David Wilson). In particular, 1
thank John Blake, Savannah River Forest Station, for his
long-standing support of my research on the site.

LITERATURE CITED

Allen, D. H. 1991. An insert technique for con.structing
artificial Red-cockaded Woodpecker cavities. USDA,
Forest Service, General Technical Report SE-73.
Southea.stern Forest Experiment Station, Asheville,
North Carolina, USA.

Conner, R. N., D. C. Rudolph, and J. R. Walters. 2001.
The Red-cockaded Woodpecker: surviving in a fire-

Frcnizreh • RED-COCKADED WOODPECKER FORAGING IN YOUNG FORESTS 257

maintained ecosystem. University of Texas Press,
Austin, USA.

DeLotelle, R. S., J. R. Newman, and A. E. Jerauld.
1983. Habitat use by Red-cockaded Woodpeckers in
central Florida. Pages 59-67 in Proceedings of the
Red-cockaded Woodpecker Symposium, 11 (D. A.
Wood, Editor). Florida Game and Fresh Water Fish
Commission, Tallahassee, USA.

DeLotelle, R. S., R. J. Epting, and J. R. Newman. 1987.
Habitat use and territory characteristics of Red-
cockaded Woodpeckers in central Florida. Wilson
Bulletin 99:202-217.

Edwards, J. W., W. M. Smathers Jr., E. T. Lema.ster,
AND W. L. Jarvis. 2000. Savannah River Site Red-
cockaded Woodpecker management plan. USDA,
Forest Service, Unpublished Report. Savannah River
Forest Station, New Ellenton, South Carolina, USA.

Engstrom, R. T. AND F. J. Sanders. 1997. Red-cockaded
Woodpecker foraging ecology in an old-growth
longleaf pine forest. Wilson Bulletin 109:203-217.

Franzreb, K. E. 1997. Success of intensive management of
a critically imperiled population of Red-cockaded
Woodpeckers in South Carolina. Journal of Field
Ornithology 68:458^70.

Franzreb, K. E. 1999. Factors that influence translocation
success in the Red-cockaded Woodpecker. Wilson
Bulletin 111:38^5.

Franzreb, K. E. 2004. Habitat preferences of foraging
Red-cockaded Woodpeckers at the Savannah River
Site, South Carolina. Pages 553-561 in Red-cockaded
Woodpecker: road to recovery (R. Costa and S. J.
Daniels, Editors). Hancock House Publishers, Blaine,
Washington, USA.

Frost, C. C. 1993. Four centuries of changing landscape
patterns in the longleaf pine ecosystem. Proceedings of
the Tall Timbers Fire Ecology Conference 18:17^4.

Hanula, j. L., K. E. Franzreb, and W. D. Pepper. 2000.
Longleaf pine characteristics associated with arthro-
pods available for Red-cockaded Woodpeckers. Jour-
nal of Wildlife Management 64:60-70.

Hardesty, J. L., K. E. Gault, and R. P. Percival. 1997.
Ecological correlates of the Red-cockaded Woodpeck-
er (Picoides borealis) foraging preferences, habitat
use, and home range size in northwest Florida (Eglin
Air Force Base). Final Report. Research Work Order
99. Florida Cooperative Fish and Wildlife Research
Unit. University of Florida, Gainesville, USA.

Hejl. S. j., j. Verner, and G. W. Bell. 1990. Sequential
versus initial observations in studies in avian foraging.
Studies in Avian Biology 13:166-173.

Helms, J. A. (Editor). 1998. The dictionary of forestry. S(x;iety
of American Foresters, Bethesda, Maryland, USA.

Hooper, R. G. and M. L. Lennartz. 1981. Foraging
behavior of the Red-cockaded Woodpecker in South
Carolina. Auk 98:321-334.

Jackson, J. A. 1986. Biopolitics, management of federal
lands, and the con.servation of the Red-cockaded
Woodpecker. American Birds 40:1 162-1 168.

Jackson, J. A., M. R. Lennartz, and R. L. Hooper. 1979.
Tree age and cavity initiation by Red-cockaded
Woodpeckers. Journal of Forestry 77:102-103.

James, F. and H. Shugart. 1970. A quantitative method of
habitat description. Audubon Field Notes 24:727-736.

Jones, C. M. and H. E. Hunt. 1996. Foraging habitat of the
Red-cockaded Woodpecker on the D’Arbonne Nation-
al Wildlife Refuge, Louisiana. Journal of Field
Ornithology 67:51 1-518.

Levins, R. 1968. Evolution in changing environments.
Princeton University Press, Princeton, New Jersey, USA.

Ligon, j. D. 1968. Sexual differences in foraging behavior
in two species of Dendrocopos woodpeckers. Auk
85:203-215.

Ligon, J. D., P. B. Stacey, R. N. Conner, C. E. Bock, and
C. S. Adkisson. 1986. Report of the American
Ornithologists’ Union Committee for the conservation
of the Red-cockaded Woodpecker. Auk 103:848-855.

Mengel, R. M. and j. Jackson. 1977. Geographic
variation of the Red-cockaded Woodpecker. Condor
79:349-355.

Nesbitt, S. A., D. T. Gilbert, and D. B. Barbour. 1978.
Red-cockaded Woodpecker fall movements in a
Florida flatwoods community. Auk 95:145-151.

Pizzoni-Ardemani, a. 1990. Sexual dimorphism and
geographic variation in the Red-cockaded Woodpeck-
er. Thesis. North Carolina State University, Raleigh,
USA.

Platt, W. J., G. W. Evans, and S. L. Rathburn. 1988.
The population dynamics of a long-lived conifer
(Pinus palustris). American Naturalist 131:491-525.

Porter, M. L. and R. F. Labisky. 1986. Home range and
foraging habitat of Red-cockaded Woodpeckers in
northern Florida. Journal of Wildlife Management
50:239-247.

Porter, M. L., M. W. Collopy, R. F. Labisky, and R. C.
Littell. 1985. Foraging behavior of Red-cockaded
Woodpeckers: an evaluation of research methodolo-
gies. Journal of Wildlife Management 49:505-507.

Ramey, P. 1980. Seasonal, sexual, and geographic variation
in the foraging ecology of Red-cockaded Woodpeck-
ers {Picoides borealis). Thesis. Mississippi State
University, Starkville, USA.

Rao, j. N. K. and a. j. Scott. 1992. A simple method for
the analysis of clustered binary data. Biometrics
48:577-585.

Repasky, R. R. 1984. Home range and habitat utilization of
the Red-cockaded Woodpecker. Thesis. North Car-
olina State University, Raleigh, USA.

Rudolph, D. C, R. N. Conner, R. S. Schaefer, and N. E.
Koerth. 2007. Red-cockaded Woodpecker foraging
behavior. Wilson Journal of Ornithology 1 19:170-180.

SAS Institute Inc. 2004. SAS/STAT 9.1. User’s guide.
SAS Inslitute Inc., Cary, North Carolina, USA.

Schoener, T. W. 1970. Non-synchronous spatial overlap of
lizards in patchy environments. Ecology 51:408-418.

Skorupa, j. P. 1979. Foraging ecology of the Red-
cockaded Woodpecker in South Carolina. Thesis.
University of California, Davis, USA.

Skorupa, J. P. and R. W. McFarlane. 1976. Seasonal
variation in foraging territory of Red-cockaded
Woodpeckers. Wilson Bulletin 88:662-665.

U.S. Fish AND Wildlife Service (USFWS). 2003. Recovery
Plan for the Red-cockaded Woodpecker {Picoides

258

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2, June 2010

borealis). Second Revision. USDI, Fish and Wildlife
Service. Southeast Region, Atlanta, Georgia, USA.
Wahlenberg, W. G. 1960. Loblolly pine: its use, ecology,
regeneration, protection, growth and management.
Duke University School of Forestry, Durham, North
Carolina, USA.

Walters, J. R. 1990. Red-cockaded Woodpecker: a
‘primitive’ cooperative breeder. Pages 69-101 in
Cooperative breeding in birds (P. B. Stacey and W.
D. Koenig, Editors). Cambridge University Press,
London, United Kingdom.

Walters, J. R., S. J. Daniels, J. H. Carter III, and P. D.
Doerr. 2002. Defining quality of Red-cockaded

Woodpecker foraging habitat based on habitat use
and fitness. Journal of Wildlife Management 66:1064-
1082.

Walters, J. R., S. J. Daniels, J. H. Carter III, P. D.
Doerr, K. Brust, and J. M. Mitchell. 2000.
Foraging habitat resources, preferences and fitness of
Red-cockaded Woodpeckers in the North Carolina
sandhills. Fort Bragg Project Final Report. Virginia
Polytechnic Institute and State University, Blacksburg,
and North Carolina State University, Raleigh, USA.

ZwiCKER, S. M. AND J. R. WALTERS. 1999. Selection of
pines for foraging red-cockaded woodpeckers. Journal
of Wildlife Management 63:843-852.

The Wilsini Joiinicil of Ornithology 1 22(2):259-272, 2010

FORAGING HABITS AND HABITAT USE BY EDIBLE-NEST AND
GLOSSY SWIETLETS IN THE ANDAMAN ISLANDS, INDIA

SHIRISH S. MANCHl'"' AND RAVI SANKARAN' -

ABSTRACT. — Foraging habits and habitats of exclusive aerial insectivores, the Edible-nest Swiftlet (Aerodramus
fitciphagus inexpectatiis) and Glossy Swiftlet {Collocaliu escidenta affinis), were studied in Andaman Islands, India.
Observations were made during January to June 2004 between 0500 and 1 800 hrs at four locations in the forest and on open
paddy land. Edible-nest and Glossy swiftlets, respectively, spent (.v ± SD) 17.2 ± 1 1.4% and 25.8 ± 15.6% of their time
foraging with significant temporal variations. Glossy Swiftlets had spatial variations in twist, flutter, and tail-wing-open
foraging maneuvers. This species also had diurnal variations in flock size, which were positively correlated with feeding
attempts. Both swiftlets shared all microhabitats except Inside Forest Canopy and Inside Stream Bank Canopy.
Microhabitat use did not vary significantly in Below Stream Bank Canopy, >10 m Above Forest Canopy, >30 m Above
Ground, and Above Forest Canopy for Edible-nest Swiftlets. Inside Forest Canopy and Inside Stream Bank Canopy
categories for Glossy Swiftlets were relatively important in descending order. Deforestation near and distant from caves
used by swiftlets for breeding in the islands can severely affect the wild population of both species. Received 2 September
2009. Accepted If January 2010.

Swiftlets (Apodidae) are exclusively aerial
insectivores and land only on their nesting and
roosting sites in caves or cave-like conditions
(Medway 1962, Charles 1987, Chantler and
Driessens 2000, Nguyen et al. 2002). Swiftlets
have wings well adapted for long and fast flights
(Norberg 1986, Chantler 1999). However, a single
observation of a Glossy Swiftlet {Collocalia
esculent a) perching on a tree was reported by
Spennemann (1928). Detailed information about
aerial and foraging habits, and habitat and
microhabitat requirements of most swiftlet species
is .scarce. Swiftlets have large foraging ranges and
travel far from caves used for breeding. Medway
(1962) observed Aerodramus maximus and A.
salanganus flying >24 km from their breeding
caves. Swiftlet abundance has been mostly studied
in different habitats or air spaces whereas their
behaviors have been briefly studied (Medway
1962, Harrison 1972, Hails and Amirrudin 1981,
Waugh and Hails 1983, Tarburton 1986, Charles
1987, Lim and Cranbrook 2002, Nguyen et al.
2002). Most studies have recorded the occurrenc-
es of swiftlet species in different habitats and
altitudes in relation to breeding seasonality, and
weather conditions with information on diet and
general foraging habits (Medway 1962, Harrison
1972, Langham 1980, Hails and Amirrudin 1981,
Nguyen 1983, Waugh and Hails 1983, Laurie and

' Division of Conservation Ecology, Salim AH Centre for
Ornithology and Natural History, Anaikatty P. O., Coim-
batore- 641 108, India.

^Deceased 17 January 2009.

^Corresponding author; e-mail: ediblenest@gmail.com

Tompkins 2000, Lim and Cranbrook 2002,
Nguyen et al. 2002). The breeding biology and
breeding habitats of the economically valuable
Edible-nest Swiftlet (A. fuciphagus) are well
documented in different parts of their ranges
(Lim and Cranbrook 2002, Nguyen et al. 2002).

India has four species of swiftlets; Indian
Swiftlet (A. unicolor), Himalayan Swiftlet (A.
hrevirostris). Edible-nest Swiftlet, and Glossy
Swiftlet. The later two species are distributed in
the Andaman and Nicobar Islands. The Edible-
nest Swiftlet {A. fuciphagus inexpectatiis), endem-
ic to the Andaman and Nicobar Islands, is known
to build highly commercial white edible nests
exclusively made up of its saliva (Medway 1963,
Lau and Melville 1994, Chantler 1999) inside
limestone caves throughout the island arc. The
Glossy Swiftlet (C. escidenta affinis), also en-
demic to the Andaman and Nicobar Islands,
commonly constructs its nest using saliva as the
cementing material to bind nesting materials
including mosses, leaves, flowers, and Casuarina
needles inside caves, suitable houses and build-
ings, under jetties, etc. (Medway 1963, Chantler
1999). Uncontrolled exploitation of nests caused a
decline of —80% in the Edible-nest Swiftlet
population between 1990 and 2000 (Sankaran
2001). In-situ and the e.x-situ conservation pro-
grams for the species were initiated at selected
sites to aiTest this population decline in the
Andaman and Nicobar Islands. The in-situ
population of the Edible-nest Swiftlet is being
protected during the breeding season. The ability
to breed in diverse habitats made the Glossy
Swiftlet an important part of the ex-situ conser-

259

260

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

vation program for fostering the Edible-nest
Swiftlet (Sankaran and Manchi 2008).

The Edible-nest Swiftlet and Glossy Swiftlet of
the Andaman and Nicobar Islands have not been
studied extensively except for status surveys and
brief information on their breeding seasonality
(Sankaran 1995, 1998, 2001). Our objectives were
to study: (1) aerial and foraging habits, (2) spatial
and temporal variations in foraging habits, and (3)
habitats with the microhabitat requirements of
both species near their breeding sites.

METHODS

Study Area. — Andaman and Nicobar Islands in
the northeastern Indian Ocean are the peaks of a
submerged continuous mountain ridge arching
from Arakan Yoma in the north to Sumatra in the
south, between 06° 45' N and 13° 41' N, and 92°
12' E and 93° 57' E (Manchi and Sankaran 2009).
Data were collected near the breeding sites of
Edible-nest and Glossy swiftlets at Chalis-ek (a
single hillock with a group of inland limestone
caves) and Pattilevel (human habitation area
adjacent to the Chalis-ek hillock) near Ramnagar,
in the southeastern part of North Andaman Island.
The Chalis-ek cave complex had 650 breeding
pairs of Edible-nest Swiftlets. Eive caves were
used by 1 1 7 breeding pairs of the Glossy Swiftlet,
along with pairs of Edible-nest Swiftlets. Chota
tikree, another small hillock within 0.5 km of
Chalis-ek, had 16 breeding pairs of Edible-nest
Swiftlets and seven breeding pairs of Glossy
Swiftlets in four caves. The Chalis-ek and
Pattilevel areas have evergreen, semi-evergreen,
and dry deciduous habitats (Sankaran 1998).
People inhabiting the area altered part of the
forested land into paddy fields. There are two
major open paddy fields (within 1-2 km of the
Chalis-ek hillock. The remaining forest in the area
is under immense pressure of deforestation.

Data Collection. — Scan sampling (Altmann
1974) was used to record foraging behavior of
Edible-nest and Glossy swiftlets between ()5()()
and 1800 hrs (Indian Standard Time). Observa-
tions were made at four points every 2 weeks from
January to June 2004. Ten-minute observations
were made every 15 min. The number of birds
was recorded after each observation. There were
2,086 observations in 624 hrs for both species.
Activity performed, habitat, microhabitat type,
and Hock size were recorded for each encounter.
We developed terminology for describing the
aerial behavior of swiftlets.

Flylglide. — ^The flight with wing beats was
recorded as ‘fly’ and flight without wing beat
was recorded as ‘glide.’ These two flying patterns
are combined as Flylglide.

Feeding Attempts (Foraging Maneuvers). — Ob-
servations were limited to recording feeding
attempts by presuming that different behaviors/
activities were performed to capture different
kinds of prey. Five types of behaviors were
recorded as Foraging Maneuvers.

Twist. — The bird makes a sudden twist while
gliding.

Fly-pause. — A fast-flying individual occasion-
ally makes a sudden pause for a second and then
moves on with a slight twist.

Roll. — The individual catches prey in its beak
and rolls down. Birds were observed rolling down
for about 2 sec.

Flutter. — A hover performed with a rapid wing
beat and a pause in flight.

Tail and wing open. — A pre-planned position to
capture prey once observed. Wing and tail
feathers are stretched while approaching the prey
and a small twist or flutter is performed during
capture. This event takes at most 2 sec.

Call. — ^There are two types of calls: tik-tik-tik
and chirk-chirk-chirk made in flight.

Follow. — Individuals were observed chasing
each other in flight.

Preen. — Preening during flight.

Defecate. — ^Defecating during flight.

Carry Nest Material. — Swiftlets collect and
carry nest materials in flight.

Major studies related to food preferences
(Harrison 1972, Hails and Amin'udin 1981, Laurie
and Tompkins 2000, Nguyen et al. 2002) were
ba.sed on gut content and bolus collection; these
were not attempted in our study to avoid sacrifice
of any individual. We presumed that gut content
and bolus collection during feeding visits by
adults would adversely affect nesting success.

The Point-centered Quarter Method (Mitchell
2001) was used to study tree diversity in the
forests of Chalis-ek. Habitats were classified
following Hails and Amirrudin (1981) with
separation of habitats into microhabitats depen-
dent on foraging heights in air space. Heights
considered for segregating microhabitats differed
from those of Hails and Amirrudin ( 1981 ) per the
suitability of the study area based on canopy
levels. Forest habitat was categorized into four
microhabitats: (1) Below Forest Canopy, (2)
Inside Forest Canopy, (3) 0-10 m Above Forest

Mauchi and Sankaran • SWIITLET FORACHNG IN I'llH ANDAMAN ISLANDS

261

TABLE 1. Activity budgets ( v ± SD, %) of Edible-nest and Glossy swiftlets near breeding caves at Chalis-ek, North
Andaman Island, India.

Activity

Edible-nest Swiftlet

n = 719

Glossy Swil'tlet
« = l,26.S

Mann-Whiiney lesi

z

r

Fly/glide

74.42 ± 16.10

84.60 ± 17.50

-13.609

<0.001

Feeding attempts

14.03 ± 12..30

13.75 ± 17.20

-5.217

<0.001

Call

2.08 ± 5.10

0.67 ± 3. .30

-14.572

<0.001

Follow

9.47 ± 12.10

0.98 ± 4.20

-25.020

<0.001

Canopy, and (4) >10 m Above Forest Canopy.
The open paddy lands developed through defor-
estation had streams with vegetation along the
stream banks. This vegetation was included in the
open land habitat which was divided into six
microhabitats: (1) 0-5 m Above Ground, (2) 5-
30 m Above Ground, (3) >30 m Above Ground,
(4) Below Stream Bank Canopy, (5) Inside Stream
Bank Canopy, and (6) 0-10 m Above Stream
Bank Canopy. Temperature was recorded using a
Zeal thermometer. A rain gauge was placed at the
top of the hill in an open canopy area to record the
amount of rainfall.

Statistical Analyses. — ^Percentage of time spent
on each activity was estimated in each set of
observations from instantaneous scan samples and
the proportion of time spent on different aerial
activities was calculated. Behaviors including
preening, defecating, and carrying nest material
were collectively performed <0.1% of the time
by both species, and were not included in the
statistical analyses. Variations in activities of
Edible-nest and Glossy swiftlets were calculated
with Mann-Whitney C-tests (Z). Days were
divided into four sets of time: 0500-0815 hrs
(early morning), 0815-1130 hrs (late morning),
1130-1445 hrs (afternoon), and 1445-1800 hrs
(late afternoon). The data were arcsine trans-
formed for one-way ANOVAs. Kruskal-Wallis
tests and one-way ANOVAs were performed to
examine temporal and spatial variations in
activities overall, different behaviors separately,
and activities performed in different microhabi-
tats. Identification of the differing means was
achieved by a Tukey’s test for the detailed activity
patterns when significant differences were found.
All statistical analyses were also conducted with
the data on feeding attempts to derive or
understand foraging maneuvering patterns. Pear-
son’s correlation test (/) was performed to
examine the relation between foraging frequency
and foraging flock size. Diurnal variation in size

of the foraging flock was also described. An
Important Value Index (IVI) of trees was
calculated following Mitchell (2001). Chi-.square
tests (Crosstab: contingency table) were used to
confirm distinctness of the species in terms of
microhabitat use. Binary logistic regression anal-
ysis was performed to estimate use and relative
importance of each microhabitat. All statistical
tests were performed using Microsoft Excel 2003
and SPSS software. Version 10.0 (SPSS Institute
1998).

RESULTS

Edible-nest Swiftlets were encountered 41,959
times in 719 observations with an average (.v ±
SD) of 58.4 ±41.6 activities per set. There were
36,906 encounters of Glossy Swiftlets in 1,265
sets of observations with an average of 29.2 ±
36.5 activities per 10-min period.

Aerial Activity Budget. — ^Edible-nest and Glossy
swiftlets in flight performed fly/glide, feeding
attempts, follow, and call in descending order of
their frequencies of occuirence. Preen, defecate,
and nest material transport were collectively
performed for 0.07% of the total activities and
were not included in further analyses. The
proportions of time by the two species for
different activities varied significantly (Table 1).

Spatio-temporal Variation in Aerial Activity'. —
Edible-nest and Glossy swiftlets had significant
temporal and spatial variations in their overall
active periods (Eig. 1). The monthly active period
varied (/‘ = 105, P < O.OOl) for Edible-nest and
Glossy swiftlets (x^ = 23, P < 0.001). The
diurnal active period varied for Edible-nest (x' =
55.896, P < O.OOl) and Glossy swiftlets (x' =
8.464, P = 0.037). Spatial variation in active
period was also shown by Edible-nest (x* =
58.081, P < 0.001) and Glossy swiftlets (x' =
293.654, P < 0.001). Individual aerial activities
of both species, except feeding attempts for
Edible-ne.st Swiftlets, had significant monthly

262

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2, June 2010

Jan Feb Mar Apr May Jun

□ Edible-nest Swiftlet

I

□ Glossy Swiftlet

B

□ Edible-nest Swiftlet

I

□ Glossy Swiftlet

Indian Standard Time

C

i

□ Edible-nest Swiftlet

I

□ Glossy Swiftlet

Hill base Open land-1

Hill top Open land-ll

FIG. i. Spatio-temporal variation (A = monthly, B =
diurnal, and C = location; %) in activities of Edible-nest and
Glossy .swiftlets at Chalis-ek, North Andaman Island, India.

variations (Table 2). All activities had significant
diurnal variations except feeding attempts for both
species, and call behavior for the Edible-nest
Swiftlet (Table 3). All the other activities, unlike

OVJ

70

LU 60
(S)

CM

4- 50

c

03

S 40

> 30

o

<

>5 20

10

0

■ S" '

a

ZBI

fly/glide and call, had significant spatial varia-
tions (Table 4).

Foraging Activity Budget. — ^Edible-nest and
Glossy swiftlets spent (.v ± SD) 17.2 ± 11.4
and 25.8 ± 15.6% of their time in feeding
attempts, respectively. Both species spent sub-
stantial time on twist, flutter, tail-wing-open, fly-
pause, and roll in descending order. The propor-
tion of time spent on different foraging maneuvers
varied significantly (Table 5).

Spatio-temporal Variations in Foraging Activ-
ities.— Both species had significant spatial (Edi-
ble-nest Swiftlet: — 61.236, P < 0.001; Glossy
Swiftlet: x^ — 84.702, P < 0.001) and monthly
(Edible-nest Swiftlet: x~ ~ 19.588, P < 0.001;
Glossy Swiftlet: x^ ~ 53.952, P < 0.001)
variations in time spent in feeding attempts.
However, significant temporal variations across
the day were not apparent (Eig. 2). Eoraging
maneuvers, except flutter by Edible-nest Swiftlets
and roll by Glossy Swiftlets, had significant
monthly variations (Table 6), but roll had signif-
icant diurnal variations by Glossy Swiftlets
(Table 7). Twist alone in both species, and tail-
wing-open in Glossy Swiftlets, had significant
spatial variations (Table 8).

Foraging Flock Size. — ^Edible-nest Swiftlets
foraged in significantly larger flocks (T ± SD,
9.4 ± 3.5 individuals; min = 3, max = 17) before
returning to roosting caves between 1700 and
1800 hrs (x" = 120.944, P < 0.001). Glossy
Swiftlets also formed comparatively large flocks
(J ± SD, 5.7 ± 2.4 individuals; min = 2, max =
II) between 1700 and 1800 hrs without any
significant diurnal variation (Fig. 3). Edible-nest
and Glossy swiftlets had significant coirelations (r
= 0.400, P < 0.001 and r = 0.307, P < 0.001,
respectively), between foraging flock size and
frequency of feeding attempts. This correlation
did not change when partial con-elation was
performed by controlling for month, location,
and hours (Edible-nest Swiftlet: r = 0.420, P <
0.001, Glossy Swiftlet: r = 0.370, P < 0.001).

Quantification of Vegetation. — Chalis-ek had 63
species of trees in the area. The forest community
was dominated by Pterocarpus dalhergioides (IVI
= 35.7) followed by Diospyros criimenata (IVl, =
22.4) and Drypetes andamanica (IVl = 21.3).

Habitats and Microliahitats. — Foraging individ-
uals of both species were encountered significant-
ly more in forest than open land (Edible-nest
Swiftlet: x' = 94.298, P < 0.001; Glossy
Swiftlet: x' = 58.548, P < 0.001; Fig. 4).

TABLE 2. Temporal (monthly) variation (%) in activities of Edible-nest and Glossy swiftlets at Chalis-ek, North Andaman Islands, India.

Manchi and Sankaran • SWIFTLET FORAGING IN THE ANDAMAN ISLANDS 263

m

OJ

X

fN

o

o

o

O

o

O

o

•—

.— (

0.

o

o

o

o

o

o

o

o

O

o

Os

fN

in*

nj

g

O

o

in

O

§

o

o

<

d

V

d

d

V

d

V

d

V

d

V

d

V

d

V

d

V

d

d

d

V

d

V

d

d

V

d

V

>

o

<

>

z

<

o

r--

o

00

so

q

(N

O

z

m

m

r-

so

q

m

Tf

<

ro

q

q

q

q

q

q

O

S

q

in

fN

q

00

o

in

(N

d

in

d

in

in

d

C3

u.

fN

m

m

in

00

q

m

q

q

*■5

c

od

—

—

d

sd

fN

d

d

A

CN

<N

r--

o

-o

q

in

q

os

c

—

in

r-

r--

—

—

d

m

'7a

Jun

+ 1

+ 1

+ 1

+ 1

II

+ 1

+1

+ 1

11

c

(N

—

q

s:

q

q

o

00

s:

ir3

rn

(N

d

d

d

d

E

00

Os

CQ

•o

c

8

00

q

d

q

q

q

r-

in

oo

od

q

od

X

O

OS

<

d

d

fN

d

fN

ON

q

oo

q

00

00

q

in

q

in

q

r—

fS

m

x:

t:

o

T

cr,

1

+ 1

00

+1

q

+ 1

q

+ 1

q

II

+ 1

q

+ 1

+ 1

fN

+1

X

II

— *

d

(N

d

d

(N

d

d

(N

d

in

d

d

o

+1

+ 1

+ 1

+1

II

+ 1

+ 1

+ 1

+ 1

II

o

c/5

so

00

d

SD)

(n

q

9.3

q

in

q

d

0.2

0.3

(D

1

c/5

+1

+ 1

r--

—

00

—

"E

>%

JZ

\'<

U

■o

flj

(A

c

O

S

u

>

D

a.

d

q

d

q

d

q

d

OO

CN

OS

c5

c/5

o

A

± 13.3

± 10.9

Apr

+ 1

os

+ 1

q

fN

+ 1

q

+1

q

CN

II

5:

+ 1

q

+ 1

m

o

+ 1

fN

II

?

c/5

3

o

c

at

m

5

± 2.7

± 9.6

= 142

± 16.1

± I6.C

± 3.3

± 3.2

= 299

at

in

r-

d

00

d

00

d

d

>%

A

1

r<5

q

00

s:

X

q

X

c:

c

A

sd

d

X

od

sd

fN

d

d

O

I— H

00

’>

O

•c

u

a

CN

d

00

d

m

cn

q

od

q

od

q

q

00

m

"O

c

T3

«

—

—

(N

—

—

—

—

in

in

fN

’>

c

c

Mar

+ 1

+1

+ 1

+ 1

II

+ 1

+ 1

+ 1

+ 1

II

A

(U

O

CO

O

sd

q

d

oo

q

os

fN

sd

q

in

q

q

(N

c

a>

so

in

q

CN

q

q

d

q

q

c:

c

t

c

cd

C

T

+ 1

+ 1

in

+ 1

od

+ 1

II

+ 1

+1

d

+ I

d
+ 1

ro

II

<u

P

00

X)

o

IT)

'"B

c

00

q

q

q

q

s:

q

q

q

q

s:

m

at

O-

o

d

in

d

d

in

fN

d

0)

OX)

n

os

r--

o

^ 1

4-

O

A

at

E

00

c

at

d

m

d

m

—2

q

q

XJ

A

O

(N

(N

sd

so

(N

d

in

fN

0)

O.

Feb

+ 1

+ 1

+ 1

+1

11

+1

+ 1

+ 1

+ 1

II

’>

q

q

X

fN

q

fN

in

r-

q

c:

O)

OS

r-

S;

o

CO

od

d

q

d

od

d

q

O

On

so

(N

00

d

—

C13

00

in

fN

d

sd

fO

SO

c

+ 1

+ 1

+1

+ 1

II

+ 1

+ 1

+ 1

+ 1

II

g

»r)

■q

q

fN

s:

X

q

o

os

k:

o

d

d

(N

X

d

in

d

in

q

sd

os

q

00

q

oo"

(n

q

O

c

o

oo

nj

—

00

—

q

—

—

d

d

fN

Jan

+ 1

+ 1

+ 1

+ 1

II

+1

+ 1

4-1

+ 1

II

c5

’C

cd

(n

q

q

(n

q

Os

q

?

>

ri

d

d

d

d

d

r--

00

' — '

"u,

D

A

c/5

'A

c/5

O

sz

s.

£

D,

£

c.

H.

cd

B

B

’>

p

B

£

o

in

sO

II

Lh

O

a.

o

<u

;o

cd

OX)

os

(U

."2

cd

W>

in

X

fN

£

.£

r-«.

c

o

<

<u

;o

Oi)

c

'-B

o

= 719

u

;o

'ob

00

c

'*5

«u

H

E

''B

o

<u

pp

Call

"o

p^

1!

s;

E

'•B

<u

pp

Call

*0

LU

II

5:

o

o

'o

>-*

(U

OJ

*o

d

u.

U-

U

a.

s:

u-

U.

U

m

5:

tu

CQ

<

H

OJ

B

•A

at

on

A.

w

u

g.

(A

<u

at

o.

A

OJ

>

c

cn

c/5

c

00

JJ

dJ

c/5

A

C/5

X

A

O

o

'B

O

pa

a

264

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2. June 2010

TABLE 4. Spatial variation (%) in activities of Edible-nest and Glossy swiftlets at Chalis-ek, North Andaman
Islands, India.

Time .speni on an activity at different locations (.v ± SD)

ANOVA

Species Activity Hill base Hill top Open land-1 Open land-II F P

Edible-nest

Fly/glide

74.8

-h

15.1

72.9

-F

15.8

79.3 ±

20.8

73.9

-F

13.3

3.069

0.027

Swiftlet

Feeding attempts

13.7

-h

10.9

12.4

-F

10.6

20.7 ±

20.8

26.1

4-

13.3

13.390

<0.001

Call

2.4

H-

5.9

2.3

-F

4.6

0

0

5.130

0.002

Follow

9.1

-F

11.4

12.4

-F

13.2

0

0

24.400

<0.001

n = 719

n

=

333

n

=

306

n =

67

n

=

13

Glossy Swiftlet

Fly/glide

90.4

4-

13.6

90.3

-F

14.9

74.00 ±

17.8

79.9

4-

18.6

57.552

<0.001

Feeding attempts

7.4

-h

12.2

6.7

-F

12.9

25.90 ±

17.7

19.8

4-

18.6

152.767

<0.001

Call

1.0

-F

4.6

1.2

-F

3.8

0.03 ±

0.6

0.05

4-

0.5

14.502

<0.001

Follow

1.2

-F

4.9

1.9

-F

5.4

0.10 ±

1.2

0.20

4-

1.5

18.562

<0.001

n = 1,265

n

=

359

n

=

41 1

n —

358

n

=

137

Statistically, monthly variations in feeding at-
tempts could not be shown for Edible-nest
Swiftlets for Inside Stream Bank Canopy, Inside
Forest Canopy, and 0-5 m Above Ground
categories and for Glossy Swiftlets in Below
Stream Bank Canopy. The Edible-nest Swiftlet
had monthly variations in feeding attempts except
in 5-30 m Above Ground. The Glossy Swiftlet
had monthly variations in 0-10 m Above Forest
Canopy, Inside Forest Canopy, 0-10 m Above
Stream Bank Canopy, and Inside Stream Bank
Canopy (Table 9). Statistically, diurnal variations
in feeding attempts could not be shown for
Edible-nest Swiftlets in Inside Forest Canopy,
Inside Stream Bank Canopy, Below Stream Bank
Canopy, and 0-5 m Above Ground, and for
Glossy Swiftlets in Below Stream Bank Canopy.
The Edible-nest Swiftlet had significant diurnal
variations in feeding attempts in 0-10 m Above
Stream Bank Canopy, >30 m Above Ground, and
5-30 m Above Ground microhabitats, whereas the
Glossy Swiftlet had diurnal variations, except in
>10 m Above Forest Canopy, 0-10 m Above

Forest Canopy, and >30 m Above Ground
microhabitats (Table 10).

All microhabitats except Inside Forest Canopy
and Inside Stream Bank Canopy were shared by the
two species. Variations in microhabitat use could
not be calculated for > 10 m Above Forest Canopy,
Inside Forest Canopy, Inside Stream Bank Canopy,
>30 m Above Ground, and 0-5 m Above Ground.
Both species had variations in use of microhabitats
0-10 m Above Forest Canopy, 0-10 m Above
Stream Bank Canopy, and 5-30 m Above Forest
Ground (Table 11). Important microhabitats for
Edible-nest Swiftlets in descending order were
>10 m Above Forest Canopy, >30 m Above
Ground, and Above Forest Canopy. Inside Forest
Canopy and Inside Stream Bank Canopy were
important microhabitats for Glossy Swiftlets.

Foraging activities for both species were not
correlated to mean temperature (Edible-nest
Swiftlet: r = 0.096, P = 0.856; Glossy Swiftlet:
r = 0.007, P = 0.989) and average rainfall
(Edible-nest Swiftlet: r = —0.447, P = 0.375;
Glossy Swiftlet: r = —0.216, P — 0.682).

TABLE 5. Variation.s (.v ± SD, %) in feeding maneuvers by Edible-nest and Glossy swiftlets at Chalis-ek, North
Andaman Island, India.

Feeding maneuver

Ediblc-nesl Swiftlel
n = .S87

Glossy Swiftlet
n = 67.1

Mann-Whitney test
Z P

Twi.st

44.12 ± 47.44

61.29 ±

31.26

-4.347

<0.001

Fly-pause

2.43 ± 8.82

0.17 ±

1.59

-9.149

<0.001

Roll

2.40 ± 9.59

0.15 ±

1.43

-8.053

<0.001

Flutter

25.57 ± 23.26

19.28 ±

23.85

-3.516

<0.001

Tail-wing-open

25.48 ± 22.92

19.11 ±

23.83

-3.619

<0.001

% Foraging attempt (Mean +- 2 SE) % Foraging attempt (Mean +- 2 SE) %Foraiging attempt (Mean +- 2 SE)

Mamhi ciiul Saukaran • SWIFTLET FORAGING IN THE ANDAMAN ISLANDS

265

A

I

□ Edible-nest Swiftlet

I

□ Glossy Swiftlet

B

1

□ Edible-nest Swiftlet

I

□ Glossy Swiftlet

C

I

□ Edible-nest Swiftlet

I

□ Glossy Swiftlet

FIG. 2. Spatio-temporal variation (A = monthly, B =
diurnal, and C = location; %) in feeding attempts of
Edible-nest and Glossy swiftlets at Chalis-ck, North
Andaman Island, India.

DISCUSSION

Aerial and Foraging Behavior. — Edible-nest
and Glossy swiftlets performed fly/glide, foraging
attempts (twist, fly-pause, roll, flutter, and tail-
wing-open), call, follow, preen, defecate, and
carry nest material in decreasing order. When
breeding seasonality of the Edible-nest Swiftlet
based on Sankaran and Manchi (2008), and
Manchi (2009) was considered, temporal varia-
tions in activities of the two species near caves
used for breeding appeared to be related to timing
of breeding. Swiflets at Chalis-ek had chicks and
eggs in the nest during April and May, and were
more active near breeding sites. Edible-nest
Swiftlets near caves used for breeding were
inactive between 0700 and 1700 hrs, and 0800
and 1400 hrs, respectively, during early and late
nest construction periods. Not requiring nest
material might have led Edible-nest Swiftlets to
spend most of the time exploring potential
foraging areas away from breeding sites. Com-
mencement of egg-laying during February fol-
lowed by incubation, and brooding and feeding
nestlings led to increased activity of Edible-nest
Swiftlets near breeding sites. The active period of
Edible-nest Swiftlets declined after May (Fig. 1),
which was attributed to fledging occurring from
most nests and avoidance of extreme weather
conditions (Medway 1962).

Medway (1962) reported wider foraging ranges
of Edible-nest Swiftlets in the non-breeding
season, leading to changes in activity near
breeding sites. Swiftlets at Chalis-ek, traveled
away from the caves even during the pre-
incubation period, possibly due to non-availability
of food near caves. Food availability for swiftlets
depends on the density and diversity of insects
which in turn depend on weather conditions
(Johnson 1969). Glossy Swiftlets were more
active in early and late morning hours, but did
not have significant diurnal variations in activities
(Fig. 2). Asynchronous breeding in the Andaman
and Nicobar Islands by Glossy Swiftlets may have
resulted in no variation in diurnal activities.

Fly/glide was the most frequent behavior
apparently used for scanning foraging areas while
traveling to other foraging areas. Both species
spent substantial time in fly/glide which depends
on availability of in.sects. Insect density varies
chronologically, spatially, and seasonally (John-
■son 1969), thus, affecting fly/glide in the two
species. Feeding attempt was the second major

TABLE 6. Temporal (monthly) variation (%) in foraging maneuvers of Edible-nest and Glossy swiftlets at Chalis-ek, North Andaman Islands, India.

266 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2, June 2010

On

ro

o

o

o

O

o

o

in

o

o

in

CN

00

o

Os

00

SO

cn

a

o

o

o

CN

O

o

o

o

O

cn

00

os

OS

CN

SO

in

o

r-

O

d

d

d

d

d

d

d

d

d

d

CL

o

CN

o

O

O

in

o

O

<

>

V

V

V

V

V

V

V

V

<

d

d

d

d

d

d

d

d

V

d

d

o

>

z

o

<

00

o

_

CN

wn

—

r-

in

00

2

so

o

—

00

so

os

00

k.

— ■

00

OS

p

fO

p

c

—

sO

o

so

cn

o

os

CN

CN

vri

CN

d

—

CN

d

d

d

d

CN

cA

■o

c

u.

so

p

p

m

p

r-*

00

p

00

m

os

sO

O

CN

in

p

O

00

CN

CN

d

CN

d

d

—

d

CN

P

m

Tt

Os

p

C3

CO

ON

d

sd

o

sd

sd

d

in

(N

d

CN

CN

VO

m

cn

CN

Jun

+ 1

+ 1

+ 1

+ 1

+ 1

II

+ 1

o

o

+ 1

+ 1

II

e

03

g

00

sd

in

d

p

p

os

d

os

cn

p

0C3

cn

p

d

p

CN

cn

(N

00

00

p

s:

os

r-

p

c:

XJ

00

CN

—

00

CN

CN

CN

d

CN

CN

sd

sd

CN

m

d

so

d

m

00

c

<:

Q

T

+ 1

+ 1

+ 1

+ 1

+ 1

II

+ 1

+1

O

+1

+ 1

11

s:

00

OS

p

p

—

—

5:

cn

r--

00

:::

t:

+ 1

“

d

d

d

in

00

CN

o

d

in

00

00

p

r-)

d

p

d

m

r-

00

sd

<N

CN

p

d

m

o

Z

>%

Tt

CN

CN

SO

d

CN

S

>>

<N

d

CN

d

CN

CN

CO

-o

+1

CO

+ 1

+ 1

+ 1

+ 1

+ 1

II

+ 1

+ 1

o

+ 1

+ 1

II

0)

1

c/5

U

X

O

—

oo

5;

—

p

p

s;

—

csi

d

sd

d

00

o

d

00

03

00

p

SO

P

p

p

CN

in

(N

(N

d

CN

CN

u

3

in

sd

in

CN

d

CN

CN

p

r-*

sd

d

o

O

X

?

CN

sd

CN

CN

cn

—

CN

CN

CN

—

£

>

_

p

00

CN

r-;

Os

p

in

sd

in

CO

U

£

u

ci

m

+ 1

in

+ 1

00

+ 1

p

+ 1

+ 1

p

II

c:

+ 1

os

+ 1

cn

+ 1

SO

+ 1

OS

+ 1

II

u

8.

u

(N

d

(N

CN

m

CN

CN

“

d

d

d

d

d

sd

d

d

d

d

Api

+ 1

+ 1

+ 1

+ 1

+1

II

+ 1

o

o

+ 1

+ 1

II

S(— (

■Q

01)

CN

CN

in

CN

CN

c

r-;

00

m

p

00

s:

<N

in

CN

=:

CO

•n

tri

d

d

d

d

d

d

T3

V

CN

(N

CN

CO

CO

(A

U

>

3

O

u

>

c

ja

(N

QO

CN

p

p

5

3

lU

c

cn

p

in

"t

p

r-

P

d

d

d

d

(N

r-

d

p

CN

d

m

~o

CO

E

o

d

CN

d

d

d

CN

d

CN

in

d

m

1

CN

CN

d

(N

so

(N

•M

(N

rj

Os

d

•— <

CN

CN

<3

CO

OD

c

~~

oh

CO

a

+ 1

+ 1

+1

+ 1

+ 1

II

+ 1

+ 1

+1

+ 1

+ 1

II

T

m

+ 1

+ 1

+ 1

+ 1

+1

II

+ 1

+ 1

+ 1

+ 1

+ 1

11

(N

C^l

00

o

OJ

00

p

p

p

p

p

•M

00

p

p

”

p

p

p

C

s

o

00

CN

d

d

d

d

o

sd

d

a

d

d

d

d

d

d

in

d

D

CN

CN

in

d

CN

V

C'J

(N

so

d

CN

c

,<u

E

0)

E

1

00

—

SO

d

'•B

o

rS

r-

—

d

p

00

d

CN

o

c

m

—

00

—

CN

— «

CJ

o

0)

X)

+ 1

+1

+ 1

+ t

+1

+1

CO

c

p

p

p

p

CN

r-

os

r-

a.

u

U.

11

+ !

o

+ 1

+1

II

<U

u

m

d

p

d

CN

d

p

CN

d

E

<i>

e

ON

00

p

s:

CN

m

p

o

c:

>

3

1)

00

CN

d

CN

CN

CN

CN

CN

CN

CN

SO

sd

d

c-i

d

d

d

d

sd

0)

E

T

+ 1

+ 1

+1

+ 1

+ 1

li

+ 1

+ 1

o

+ 1

+ 1

II

so

—

c

O

03

<n

CN

cn

p

s:

CN

P

P

c:

E

o

5

d

d

d

d

in

d

od

in

so

Os

lo

p

p

CN

O

Oi)

CN

CN

so

— <

d

Tf

d

-J

rn

d

oo

p

in

d

c

d

(N

m

d

CN

CN

'ob

Jan

+ 1

+ i

o

+ 1

+ 1

II

+ 1

+ 1

+ 1

+ 1

+ 1

II

03

u

cS

<N

p

—

s:

p

os

(n

m

p

s:

ri

d

d

d

d

d

d

C

so

(N

r-

c

o

2

’C

c3

>

in

C

c

2

>

3

c

<D

CL

c

<L>

CL

Im

k.

<V

>

3

05

D

CL

O

(U

CL

O

U

o

o

3

C

i3

£

(U

OJO

<u

00

c

ca

a>

00

<u

OX)

O

CO

3

c

r-

00

CO

3

c

m

£

&o

CO

3

c

00

c/5

3

c

C^j

r-

on

c

CO

C3

a.

u-

w

in

CO

C3

CL

u

■f

SO

c

W)

CO

to

CL

“o

S

3

II

to

%

Cl

"o

S

_3

c5

sO

II

E

u

O

Cl

‘oi)

2

o

LL

?

H

Ph

Roll

E

E

'E

H

11

c:

H

E

Roll

3

E

E

H

II

H

a:

U-

H

c:

H

E

CC

PL

H

E

aj

H

u

ClJ

_)

—

CQ

<

p

t/5

tj

(Z)

H

in

.Si

%

cn

<d

8.

to

5

u

D.

CO

<u

c

%

00

<U

c

CO

CO

cn

JJ

>>

c/5

c/5

c

o

CO

-C

UJ

p

E

PJ

CO

O

O

Mimvhi and Sankamn • SWIFTLET FORAGING IN THE ANDAMAN ISLANDS

267

TABLE 8. Spatial variations in the proportionate activities of Edible-nest and Glossy swiftlets at Chalis-ek, North
Andaman Island, India {P < 0.05 = significant difference).

Time spent on different foraging maneuvers at different locations (,v ± SD) ANOVA

Foraging

Species maneuvers Hill ba.se Hill top Open land-1 Open land-11 F P

Edible-nest

Twist

43.0

-+■

25.5

42.4

-E

27.6

56.6

-E

31.4

63.0

-E

35.0

5.977

<0.001

Swiftlet

Fly-pau,se

2.4

8.8

2.4

-E

9.3

2.8

-E

6.9

2.4

-E

5.0

0.248

0.863

Roll

1. 5

7.8

3.3

“E

11.6

3.2

-E

7.6

0.6

-E

2.1

2.147

0.093

Flutter

26.5

20.7

25.2

H-

25.6

21.7

23.4

25.8

-E

26.4

1.141

0.332

Tail-wing-open

26.6

22.8

26.8

-E

23.3

15.7

+

19.6

8.5

-E

1 1.5

4.844

0.002

n = 587

n

=

276

n

=

256

n

=

42

n

=

13

Glossy Swiftlet

Twist

65.7

-E

36.9

69.3

H-

33.6

56.5

-E

25.7

58.9

-E

32.4

6.407

<0.001

Fly-pause

0.3

-E

2.8

0

0.2

-E

1.4

0.07

-E

0.4

1.127

0.337

Roll

O.l

-E

1.2

0.2

-E

1.4

0.2

1.7

0

0.725

0.537

Flutter

21. 1

-E

31.9

14.4

-E

25.9

20.2

-E

18.6

20.3

-E

21.4

4.553

0.004

Tail-wing-open

12.7

-4-

25.3

16.1

-E

26.0

22.9

-E

20.6

20.7

-E

25.9

11.288

<0.001

n = 673

n

z=

141

n

=

134

n

=

306

n

92

activity after fly/glide. All individual activities,
except feeding attempts, had temporal variations
in their proportions by month. No diurnal
variation in feeding attempts and calls was
recorded. Call and follow were presumed to be
mating/pairing displays as well as a response to
foraging competition. Call and follow activities
by Edible-nest Swiftlets, and follow by Glossy
Swiftlets were higher during nest construction
(Jan-Feb). This suggests that follow must be more
related to mating.

Insect diversity was assumed to affect the
occurrence of different foraging maneuvers with
the presumption that different foraging maneuvers
are used to prey on different type of insects. This

dependence resulted in temporal (monthly and
hourly) variations in the proportions of different
foraging maneuvers.

One factor considered was that Glossy Swiftlets
have been shown to be more maneuverable than
Edible-nest Swiftlets (Tarburton 1986), as it has
both a higher wing and tail maneuverability index.
This could help explain why Glossy Swiftlets
were able to produce greater variability in feeding
and flutter maneuvers. Roll alone by Glossy
Swiftlets had diurnal variations; however, no
diurnal variation was observed in foraging
maneuvers by Edible-nest Swiftlets. Spatial var-
iations were shown by twist and tail-wing-open in
both species, and flutter alone by Glossy Swift-

0600 0800 1000 1200 1400 1600

I

□ Edible-nest Swiftlet

I

□ Glossy Swiftlet

Indian Standard Time

n = 276 141 256 134 42 306 92 13

Edible-nest Swiftlet

Glossy Swiftlet

Hill base Hilltop Openland-l Openland-ll

FIG. 3. Temporal variations in diurnal hock sizes of
Edible-nest and Glossy swiftlets at Chalis-ek, North
Andaman Island, India.

FIG. 4. Variation in abundance of Edible-nest and
Glossy swiftlets at four locations at Chalis-ek, North
Andaman Island. India.

268

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2, June 2010

C

■a

c

c

ra

p

~o

c

<

o

Z

u

ca

U

X)

:a

X

u.

o

£

S

L.

'■B

c

O

■o

c

ra

:/5

<U

c

I

X

UJ

c

ou

c

'U

c

o

ca

>

o

£

c

o.

<u

H

ON

UJ

_]

QD

<

H

Q

C/5

—

—

X

—

_

On

ON

in

0

0

0

0

in

X

0

0

0

0

X

ON

0

0

0

0

0

0

0

0

0

0

CN

0

d

d

d

d

d

d

d

d

d

d

d

d

d

V

V

V

V

V

V

V

V

00

in

•nj-

00

CN

in

X

r-

CN

r-

in

cn

CN

00

r-

r-

o^

in

m

00

r-

X

ON

CN

in

00

q

CN

X

in

0

r-

(N

00

X

d

d

d

d

d

d

d

X

X

'nj*

m

X

in

—

cn

r--

ON

X

«n

CN

q

00

00

0

0

CN

m

r^J

(N

m

— “

m

m

CN

+ 1

+ 1

0

0

0

0

0

0

0

II

+1

+ 1

+ 1

0

0

0

0

0

0

II

CN

00

S;

(N

00

0

2-

00

in

00

d

00

ON

CN

r-

CN

cn

od

00

ON

q

d

q

od

q

On

q

m

<N

CN

—

d

m

CN

CN

— ■

d

+ 1

+ 1

0

0

0

0

0

0

0

11

+1

+ 1

+ 1

+ 1

+ 1

0

0

+ 1

+ 1

II

X

•nf

c:

cn

X

q

--

CN

cn

e:

m

X

d

in

od

od

— 1

X

d

CN

CN

cn

CN

X

X

CN

d

00

d

CN

CN

ON

d

CN

X

od

q

ON

q

in

CN

ON

'ni-

r^.

(n

rn

—

X

—

'nj*

CN

(N

CN

CN

CN

+ 1

+ 1

0

+ 1

0

0

+ 1

+ 1

+ 1

II

0

+ 1

+ 1

+1

+ 1

0

+ 1

+ l

+1

1!

nf

0

cn

X

cn

00

r-

•nj;

rn

q

CN

q

e:

in

d

CN

d

d

d

od

CN

— *

d

d

CN

CN

CN

00

00

q

X

0

(N

O'!

X

in

d

r--

d

d

m

ro

m

ON

CN

(N

CN

— '

+ 1

+ 1

0

0

0

0

0

0

0

II

0

+1

+ 1

+ 1

+1

d

0

+ 1

0

II

00

(N

s:

0

in

—

ON

in

e:

ON

d

ni*

od

od

X

X

X

r^,

m

ON

06

CN

d

nf

X

in

ON

in

CN

CN

d

q

d

q

d

q

CN

•nr

'Tt

—

(n

—

■nr

m

CN

d

<N

d

+1

+ 1

0

+i

0

0

+ 1

+ 1

0

II

0

+ 1

+ 1

+ 1

+ 1

0

+ 1

+ 1

+ 1

II

CN

ON

X

::

O'

00

cn

r-

rj-

c:

d

m

d

d

c^i

d

d

od

d

X

d

cn

CN

O)

00

in

00

rn

cn

rn

q

m

CN

r-*

X

d

d

CN

d

d

CN

X

X

d

q

q

—

C^J

od

cn

cn

—

m

CN

—

in

cn

in

ON

+1

+1

0

+ 1

0

+ 1

+ 1

0

0

II

+ 1

+ 1

+ 1

+ 1

+ 1

+1

+ 1

+ 1

+ 1

II

m

ON

X

C^l

s:

q

q

in

q

in

c:

X

X

d

d

c^i

ON

d

m

d

d

d

—

C^l

—

—

^ CN

& O

c g

c3u

>>

O-

O

c

a

U

c

ca

CQ

o o

fT UL«

°l
^ <
S E
o 2
A <i

oo

ii

c

'■5

UJ

§■

« ^
U -D
>
o

X

<

£

o

5

g: a

§ I

U O

== s

ca ^

QQ 03

S I

iy5 C/5

<U ?

o

^ ij
£ aa

-a ^ _

C C "O
33c

S S g
a o -

OJ ¥

> > OJ

o o >

^ X) o

< < 3

F E r- °°

^ F lo

o

If; rn

A ' '

^ I

g P

3 .“F
U O

4J U

O [?

tu U-

(U ^

< <
E E
o 2

>>

a.

c

c

ca

U

Jii

C

ca

ca

CL

o

c

ca

U

? 2
m O

rn 10 ^ —

>A<i ^ Aili

E

ca

(O

tJ5

o

QJ

CQ

■o

c no
3 C
O 3

r’=: O

>

c

X

E E

E

o

r^i u~j

00

o

O

TABLE 10. Temporal (diurnal) variations in activity of Edible-nest and Glossy swiftlets in different microhabitats at Chalis-ek, North Andaman Islands, India.

Manchi and Saukuran • SWIFTLET FORACIING IN THE ANDAMAN ISLANDS

269

in

00

SO

m

in

m

in

r-

in

rt

in

o

r-

so

r-

00

•—

(N

00

(N

in

p

00

(N

o

00

*—

o

(N

in

CN

00

d>

o

d

d

m

(N

00

sO

00

ON

00

OO

o

m

o

o

m

o

o

o

n

»— •

«—

m

o

o

o

O

sO

o

o

o

00

o

O

d

d

d

d

d

d

d

d

d

d

d

d

d

V

in iri

+ 1 +1
00 (N
— ^ o6
m

00

O

o o o o o o o

o +1 +i

OO 00
(N
m

+ 1

On

sd

rJ —

+10 0+10

OO

00

00

p

m

p

p

in

(N

p

00

m

d

p

d

p

in

?

m

m

os

CN

cn

m

d

d

i

+ 1

+ 1

o

o

o

o

o

o

o

II

+ l

+1

+ 1

+ 1

+ 1

o

+ 1

+ 1

+ 1

11

m

CN

00

s:

p

p

p

so

—

<N

p

m

5:

“

00

d

d

(N

d

d

d

d

in

m

09

o

so

vO

O'!

p

d

(N

in

00

p

p

d

On

On

sd

p

p

d

p

in

so

m

m

m

—

<N

—

r--

d

m

(N

(N

d

(N

od

T

in

+ 1

+ 1

o

+ 1

o

o

+ 1

+ 1

+ 1

11

+ 1

+ 1

+ 1

+ 1

+ 1

o

+ 1

+ 1

+ 1

11

OO

—

p

p

rn

s;

so

p

00

(^1

p

m

00

p

c

O

(N

rn

oi

00

(N

d

o6

d

d

in

d

00

cn

00

p

r-

09

in

p

p

p

p

os

sd

d

p

oq

(N

d

O-

os

d

p

in

OS

m

m

od

d

—

—

ro

m

09

d

d

09

od

+ i

+1 O +1

O +1

+ 1

+ 1 o

II

o +1

+ 1

+ 1

+ 1

+ 1

+ 1

+ 1

+ 1

II

p

o

os

p

p

00

00

p

in

in

os

c;

in

in

d

d

— !

d

od

d

od

d

d

od

d

m

09

09

g; I

2 c

OO

u u

o r?

Cc W-

OJ u

< <
E -
o 2
A ci

CL

o

c

r3

u

C3

QQ

O i-

^ d:
o
c

U ^

CL

o

c

cj

XJ

C X3
3 C
O 3
rt: o

^ o

<u ^

o o

c s

U ^

>>

CL

o

c

c3

u

<v

<D

O r?

U- U-

C3

pa

S' ^

O u

CO

O-

o
c

« r-s

u

'c

c E

CL

o

c

a

"O

c -a

- r- 9^

c

3

O

Ui

a

OJ

to

CL)

o

O)

tz

c3

1)

u.

Jo

0)

3

(U

o

X)

o

X)

>

c

0^

o

X

o

X

(U

,§ §
< s

E ^
1)

03

<U

o

X

o

X

>

o

u

O

li-

eu

<

£

u.

Jo

$

<

e

<

£

X

<

E

583

<

E

<

£

t— 1

o

IL

1)

Jo

<

E

<

E

X

<

£

656

*0

o

!E

o

o

o

o

72.

o •;o

p

o

o

c

5

c

13

PP

ro

A

m

1

in

in

+

II

s:

A

7

o

c

+ £

13

CQ

m

A

ro

1

in

in

II

s:

CO

lU

c

UJ

CO

_o

a

270

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 2. June 2010

TABLE 11. Association between Edible-
North Andaman Island, India.

nest (ENS) and Glossy swiftlets (GS) with microhabitat

use at Chalis-ek,

Microhabitat^

Species

n

Activity (.v ± SD)

x2

p

HaFC

ENS

412

1 1.47 ± 10.09

_

GS

4

6.75 ± 2.06

AFC

ENS

413

11.43 ± 9.38

109.901

<0.001

GS

277

8.32 ± 13.02

IFC

ENS

0

0

GS

174

6.95 ± 7.53

ASC

ENS

10

18.50 ± 13.19

228.748

<0.001

GS

209

24.82 ± 20.74

ISC

ENS

0

0

GS

169

24.07 ± 19.54

BSC

ENS

1

39

0.009

0.923

GS

1

12

HaG

ENS

20

15.05 ± 12.53

GS

3

40.00 ± 26.00

LaG

ENS

7

5.43 ± 5.29

48.672

<0.001

GS

65

18.95 ± 22.84

NG

ENS

3

7.00 ± 5.29

GS

1 1

17.64 ± 8.58

HaFC: >10 m Above Forest Canopy; AFC: 0-10 m Above Forest Canopy: [FC: Inside Forest Canopy; ASC: 0-10 m Above Stream Bank Canopy: ISC: Inside
Stream Bank Canopy; BSC: Below Stream Bank Canopy; HaG: >30 m Above Ground; LaG: 5-30 m Above Ground; NG: 0-5 m Above Ground; — = cannot be
calculated.

lets. Monthly variations in foraging maneuvers
indicate significant seasonal variations in insect
diversity and change in prey preference with
breeding stages (Johnson 1969). Diurnal variation
in insect diversity in the forest is uncertain, although
differing slightly in open lands. This indicates that
most foraging attempts in the microhabitats >10 m
Above Forest Canopy and >30 m Above Ground
may be because of more dispersion of insects at that
height (Johnson 1969).

Habitats and Microhahitats. — Edible-nest
Swiftlets were more active in forested areas,
whereas Glossy Swiftlets were more active in
open paddy lands. The Edible-nest Swiftlet’s
activity is in concordance with previous studies
(Medway 1962, Harrison 1972, Hails and Amir-
rudin 1981, Waugh and Hails 1983, Charles 1987,
Lim and Cranbrook 2002, Nguyen et al. 2002).
The Glossy Swiftlet is the only swiftlet species
known to occur in all habitats. The diet of the
Glossy Swiftlet in Malaysia (Laurie and Tomp-
kins 2000) shows elasticity in food and foraging
habitat selection unlike the Edible-nest Swiftlet.
Eurther validity is given the concept of greater
elasticity by Collins (2000) for Glossy Swiftlets
on Palawan where it has resource partitioning
caused by the presence of Pygmy Swiftlet
(Collocalia trof’lodytes). In this situation, it was
restricted in most of its feeding to <2 m above

ground; a zone more likely to increase vulnera-
bility to land-based predators. That this species
flies so low when “forced” into ecological
partitioning was reported from New Guinea by
Schoddle and Hitchcock (1968), where it was
sharing habitat with the Uniform Swiftlet {Aero-
dramus vanikorensis). It has been shown on that
same island, but at greater altitudes, to share space
with the Mountain Swiftlet {A. hirundinaceus) in a
similar way (Diamond 1972).

Edible-nest Swiftlets were more active in early
morning and late afternoon and were in larger
foraging flocks near caves used for breeding.
Frequency of feeding attempts by both species
was correlated with foraging flock size.

Edible-nest Swiftlets forage high above the
canopy, while Glossy Swiftlets forage at the
canopy level and near the ground. Johnson (1969)
suggested that insects move upwards in air with
an increase in ambient temperature. This may be
the reason for the shift of Edible-nest Swiftlets to
higher altitudes during foraging and Glossy
Swiftlets, being smaller and more agile to
maneuver in confined spaces, to move towards
relatively cooler zones inside the canopy at the
forest edges or stream banks close to open paddy
lands for foraging.

Variations in microhabitat use signify resource
partitioning among Edible-nest and Glossy swift-

Manchi and Sankanm • SWIFTLET FORAGING IN THE ANDAMAN ISLANDS

271

lets at Chalis-ek with Edible-nest Swil'tlets having
affinity for categories, >10 m Above Forest
Canopy, >30 m Above Ground, and Above Forest
Canopy, and Glossy Swiftlets for Inside Forest
Canopy and Inside Stream Bank Canopy in
descending order. Variations in foraging maneu-
vers indicate variations in food preferences by
both species as demonstrated by gut content and
bolus analysis (Medway 1962, Harrison 1972,
Hails and Amiruddin 1981, Waugh and Hails
1983, Charles 1987, Collins 2000, Laurie and
Tompkins 2000). The Edible-nest Swiftlet’s
affinity towards >30 m Above Ground in open
land for foraging was unusual as the species has
been known to prefer forested areas. The plasticity
in food selection by Edible-nest Swiftlets (Charles
1987, Laurie and Tompkins 2000) suggests two
microhabitats, >10 m Above Forest Canopy and
>30 m above ground, may have similar insect
diversity. This demonstrates that forested habitat
and microhabitats above forest canopy level are
important for Edible-nest Swiftlets. Glossy Swift-
lets also had affinity towards both available
habitats.

We infer, in concurrence with previous studies,
that cave-dwelling Edible-nest Swiftlets are
dependent on forests and will be adversely
affected by the present rate of deforestation and
habitat alteration. Changes in land use away from
caves used for breeding sites will also lead to
population declines of Edible-nest Swiftlets be-
cause of its wider foraging ranges. However,
habitat alteration around breeding caves may not
have a marked effect on foraging activities of
Glossy Swiftlets, which have more elasticity in
habitat use.

ACKNOWLEDGMENTS

We acknowledge the Department of Environment and
Forests, Andaman and Nicobar Islands for providing funds
and active collaboration with Salim Ali Centre for
Ornithology and Natural History (SACON), Coimbatore
and for initiating this conservation program and study. We
thank the Swiftlet Protection Team for assistance during the
work. We also thank Dr. Krishnamoorthy Thiyagesan, AVC
College, Mayiladuthurai, for valuable suggestions during
statistical analysis, Archana Waran. Shobha Phaniraj,
Rachna Chandra. Madhuri Ramesh, and Drs, P. A. Azeez,
Padmanabhan Pramod, and Balakrishnan Peroth for com-
ments on previous versions of the manuscript.

LITERATURE CITED

Altmann, J. 1974. Observational study of behaviour;
sampling methods. Behaviour 49:227-265.

Chantler, P. 1999. Family Apodidae (swifts and swift-
lets). Pages .^87-418 in Handbook of the birds of the
world. Volume 5. Barn Owls to hummingbirds (J. del
Hoyo, A. Elliott, and J. Sargatal, Editors). Lynx
Edicions, Barcelona, Spain.

Chantler, P. and G. Drie,s.sens. 2()()0, Swifts - a guide to
the swifts and treeswifts of the World. Pica Press,
Sussex, United Kingdom.

Charles, M. F. 1987. The management of bird’s nest caves
in Sabah, Wildlife Section, Sabah Forest Department,
Sabah, Malaysia.

Collins, C. 2000. Foraging of Glossy and Pygmy swiftlets
in Palawan, Philippines. Forktail 16:53-55.

Diamond, J. M. 1972. Avifauna of the Eastern Highlands of
New Guinea. Publications of the Nuttall Ornitholog-
ical Club Number 12.

Hails, C. J. and A. Amirrudin. 1981. Food samples and
selectivity of White-bellied Swiftlets (Collocalia
esculenta). Ibis 123:328-333.

Harrison, T. 1972. The food of Collocalia Swiftlet (Aves,
Apodidae) at Niah Great Cave in Borneo. Journal of
Bombay Natural History Society 71:376-392.

Johnson, C. G. 1969. Migration and dispersal of insects by
flight. Methuen, London, United Kingdom.

Langham, N. 1980. Breeding biology of the Edible-nest
Swiftlet (Aerodramiis fuciphagus). Ibis 122:447^61.

Lau, A. S. M. AND D. S. Melville. 1994. International
trade in swiftlet nests. Traffic International and World
Wide Fund for Nature, Cambridge, United Kingdom.

Laurie, S. A. and D. M. Tompkins. 2000. The diets of
Malaysian swiftlets. Ibis 142:596-602.

Lim, C. K. and Earl of Cranbrook. 2002. Swiftlets of
Borneo: builders of edible nests. Natural History
Publication, Kota Kinabalu, Borneo.

Manchi, S. 2009. Breeding ecology of the Edible-nest
Swiftlet and the Glossy Swiftlet in the Andaman
Islands. Thesis. Bharathiar University, Coimbatore.
India.

Manchi, S. and R. Sankaran. 2009. Impact of the 2004
earthquake on the limestone caves in North and
Middle Andaman Islands, India. Cuixent Science 97:
1230-1234

Medway, L. 1962. The swiftlets (Collocalia) of Niah Cave.
Sarwak. Part 2. Ecology and regulation of breeding
biology. Ibis 104:228-245.

Medway, L. 1963. The antiquity of trade in edible birds'-
nest. Federal Museum Journal 8:36^7.

Mitchell, K. 2001. Quantilative analysis by the Point-
centered Quarter Method. Department of Mathematics
and Computer Science, Hobart and William Smith
Colleges, Geneva, New York. USA.

Nguyen, Q. P. 1983. The food of Edible-nest Swiftlets
Collocalia fneiphaga germani Oust, in Vietnam. Top
Chi Sinh Vat Hoc 5 (3):5-8.

Nguyen, Q. P., Y. V. Quang, and J. F. Volsin. 2002. The
White-nest Swiftlet and the Black-nest Swiftlet; a
Monograph. Societe Nouvelle Des Editions Boubee.
Paris. France.

Norberg, U. M. 1986. Evolutionary convergence in
foraging niche and flight morphology in insectivorous

272

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2. June 2010

aerial-hawking birds and bats. Omis Scandinavica
17:253-260.

Sankaran, R. 1995. Impact assessment of nest collection
on the Edible-nest Swiftlet (Collocalia fuciphaga) in
the Nicobar Islands. Occasional Report. Salim Ali
Centre for Ornithology and Natural History, Coimba-
tore, India.

Sankaran, R. 1998. The impact of nest collection on the
Edible-nest Swiftlet (Collocalia fuciphaga) in the
Andaman and Nicobar Islands. Report to lUCN. Salim
Ali Centre for Ornithology and Natural History,
Coimbatore, India.

Sankaran, R. 2001. The status and conservation of the
Edible-nest Swiftlet (Collocalia fuciphaga) in the
Andaman and Nicobar Islands. Biological Conserva-
tion 97:283-294.

Sankaran, R. and S. Manchi. 2008. Conservation of the

Edible-nest Swiftlets in the Andaman and Nicobar
Islands. Occasional Report. Salim Ali Centre for
Ornithology and Natural History, Coimbatore, India.

SCHODDE, R. AND W. B. Hitchcock. 1968. Contribution to
Papuasian ornithology 1. Report on the birds of the
Lake Kutubu Area, Territory of Papua and New
Guinea. Paper 13. Division of Wildlife Research and
Technology, CSIRO, Australia.

Spennemann, a. W. 1928. Die ebbaren Vogelnester von
Grissee. Orn. Monatsb. Berlin 1 11-1 12.

SPSS Institute Inc. 1998. SPSS for Windows. Version
10.0. SPSS Institute Inc., Chicago, Illinois, USA.

Tarburton, M. K. 1986. A comparison of the flight
behaviour of the White-rumped Swiftlet and the
Welcome Swallow. Bird Behaviour 6: 72-84.

Waugh. D. R. and C. J. Hails. 1983. Foraging ecology of
a tropical aerial feeding guild. Ibis 125:200-217.

The VMIsoii Journal of Ornithology 122(2):273-278, 2010

GOLDEN- AND BLUE-WINGED WARBLERS: DISTRIBUTION,
NESTING SUCCESS, AND GENETIC DIEFERENCES IN

TWO HABITATS

JOHN L. CONFER,' - KEVIN W. BARNES,' AND ERIN C. ALVEY'

ABSTRACT. — We analyzed phenotypic distribution, nesting success, and genetic purity of Golden-winged Warblers
(Vermivoni clirysoptera) and Blue-winged Warblers {V. pinus) in two ecologically distinct nesting habitats: early-
succession uplands, and swamp forests. The proportion of phenotypically pure Golden-winged Warblers in swamp forests
(94%) differed significantly from uplands (53%) as did the proportion of Golden-winged Warbler pairs (93%) in swamp
forests compared to uplands (48%). Only 1% of the phenotypically pure Golden-winged Warblers in swamp forests paired
with either a hybrid or a Blue-winged Warbler, but 7% of the phenotypically pure Golden-winged Warblers in uplands
formed a hybrid pair. The probability of nesting success for Golden-winged Warbler pairs in swamp forests (65%) was
significantly higher than in uplands (37%). Mitochondrial DNA analyses indicate that all 10 phenotypically pure Golden-
winged Warblers sampled from swamp forests had the ancestral Golden-winged Warbler haplotype, while 10 of 25 Golden-
winged Warblers from uplands had the ancestral Blue-winged Warbler haplotype (P = 0.033). Swamp forests may provide
a source habitat for Golden-winged Warblers with a high phenotypic and genotypic purity even in sympatry with Blue-
winged Warblers. Received 27 August 2009. Accepted 23 December 2009.

The Golden-winged Warbler (Vermivora chry-
soptera) has declined 8.8% per year in U.S. Fish
and Wildlife Region 5 from 1966 to 2007 (Sauer et
al. 2008). This decline occurred in all areas that
surround the site for this study, Sterling Forest
State Park, New York, USA. The Golden-winged
Warbler declined by 53 and 72% in New York
(Confer 2008) and in Pennsylvania (Pennsylvania
Breeding Bird Atlas 2009), respectively, during the
last 20 years. This species has declined from —200
pairs to 15 pairs in New Jersey from 1998 to 2008,
according to statewide surveys by the New Jersey
Division of Fish and Wildlife (S. M. Petzinger,
unpubl. data). The Golden-winged Warbler has
been extirpated in Massachusetts and greatly
declined in Connecticut following intrusion by
the Blue-winged Warbler (V. pinus) despite
14,000 ha of suitable shrubland habitat on utility
rights-of-ways (Confer and Pascoe 2003). Popula-
tion declines of the Golden-winged Warbler
usually follow arrival of the Blue-winged Warbler
with subsequent competition and hybridization
(Confer et al. 2003). However, Golden-winged
Warblers have co-existed with Blue-winged War-
blers for over a century (Eaton 1914) in southern
New York and populations appear to be stable at
— 125 breeding pairs in Sterling Forest State Park
(Confer et al. 1998, this study).

Initial surveys in 1999 and 2000 were conduct-
ed in uplands, where both Golden-winged and

' Department of Biology, Ithaca College, 953 Danby
Road, Ithaca. NY 14850, USA.

^Corresponding author; e-mail: Confer@ithaca.edu

Blue-winged warblers occur frequently. It became
apparent that both species also nested in two
wetland communities described by the New York
State Natural Heritage Program (Reschke 1990).
The shrub swamp community supported about
equal numbers of both species. The red maple
(Acer ruhrum) hardwood forest community, or
swamp forest, attracted primarily Golden-winged
Warblers and our wetland surveys focused on this
habitat.

This study tests the hypothesis that favorable
demographics for Golden-winged Warblers in
swamp forest habitat account for its long-term
stability and its co-existence with Blue-winged
Warblers in Sterling Forest State Park, New
York. Specifically, we hypothesize that Golden-
winged Warblers will have higher nesting
success, lower hybridization rates, and higher
genetic purity in swamp forest habitats than in
uplands.

METHODS

The principal author and a field crew of three to
.seven people studied Golden-winged and Blue-
winged warblers for 5 to 6 days a week for
10 weeks in Sterling Forest State Park. New York
in 2001, 2()()3-2()06, and 2008. Territories of
Golden-winged and Blue-winged warblers often
were contiguous or overlapped and, at times, data
could be acquired by sound from more distant
teiTitories. Each territory was observed for 25-100
person-hrs based on 15-50 person-visits, includ-
ing simultaneous observation of more than one

273

274

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2. June 2010

territory. Male Golden-winged Warblers were
banded as early in the season as possible to assist
delineation of temtorial boundaries and pairing
status. At least 80% of the males were individ-
ually recognizable within 3 weeks of the start of
the season, assisted by a —50% return rate of
banded males and distinctive phenotypes of
occasional hybrids. The intervals between obser-
vations were too long to be used for calculations
of nesting success during 2002 and 2007. We
included observations from 2002 to help estimate
the number of birds and their phenotype in
different habitats.

We made observations each year in 3^ upland
sites separated by 8-14 km, and 5-12 swamp
forests separated by 4—12 km, depending on size
of the field crew. We tried to obtain complete
demographic values and blood samples for every
pair in every patch, which provide an unbiased
estimate of demographic values within each
patch. Sites were selected because they had
typical habitat and provided a large number of
Vermivora territories in a contiguous area, facil-
itating data collection.

Upland sites consisted of a segment of rights-
of-way (ROW) with a managed width of —20 m,
another ROW segment with —80 m of managed
width, a managed successional field, and an
unmanaged area of secondary succession. These
sites provide the total range of upland habitat used
by Vermivora in the park. The two ROW sites
were the only segments of ROW that provided
continuous habitat with eight or more adjacent or
over-lapping Vermivora territories. We estimate
the upland sites we sampled probably supported
more than half of all upland Vermivora territories
in the park based on Geographic Information
System (GlS)-based habitat classification of
satellite imagery, surveys along all ROW in the
park, and our visits to many small patches of
disturbance habitat. Both ROWs have been
managed in a stable vegetative condition for the
last 40 years by a combination of herbicide
application and cutting. Both ROW have about
60% herb cover.

The two upland successional sites used for this
study had —9-12 ha of successional habitat. One
of the upland sties has been used for experimental
habitat management for Golden-winged Warblers
and this management created early secondary
succession habitat. The second successional site
was an abandoned gravel mine with .secondary
succession field/shrub conditions. Most of the

habitat used by Vermivora at this site was flooded
by a large American beaver {Castor canadensis)
pond built in summer 2006. Monitoring at this site
has ended.

All swamp forest patches were between —0.5
and 5 ha and supported 1-3 pairs of Vermivora.
We monitored 12 patches of swamp forest,
summed for all years, which provided a represen-
tative sample of Vermivora demographics for
swamp forests in general.

Nests were observed at 2-5 day intervals. We
visited nests more frequently when nestlings were
expected to be 5-7 days of age and appropriate for
blood sampling and near the time of fledging.
Estimation of nesting success followed Mayfield
(1961). We calculated nesting success starting
with the first day of incubation, assuming
incubation lasted 1 1 days and the nestling stage
lasted 10 days. Many nests discovered during
construction or egg-laying were abandoned. We
attributed most of the high failure rate early in the
nesting cycle to human-induced abandonment and
omitted these nests from calculations of nest
success. We found six second nest attempts of
pairs whose first nest was predated after incuba-
tion started. These re-nests are included in the
sample size for nests.

We used the criteria of Gill (1980) to dis-
tinguish between phenotypically pure and hybrid
birds. We used Chi-square, 2X2, contingency
tests with Yeats correction for continuity with
larger samples to test for differences in the
distribution of pure and hybrid phenotypes in
upland and swamp forests, and Fisher’s exact
probability value for tests where some cells had
expected values <5. We used a /-test to compare
the probability of nesting success between upland
and swamp forest habitats.

Mitochondrial DNA analyses of Golden-
winged and Blue-winged warblers shov/ little
variation within two distinct haplotypes, presum-
ably the ancestral mtDNA of each species. The
two haplotypes differ from each other by 3-5%
(Gill 1997, Shapiro et al. 2004, Dabrowski et al.
2005). Samples from Sterling Forest State Park
collected in 2002 and 2003 showed extensive
introgression with 12 of 28 phenotypically pure
Golden-winged Warblers having the ancestral
Blue-winged Warbler haplolype (Dabrowski et
al. 2005). P. J. Hartman (pers. comm.), continuing
mtDNA analyses of Sterling Forest State Park
samples, found the same haplotype classification
for four replicate samples analyzed by both

Confur ei a/. • WARBLER HABITAT DIFFERENTIATION

275

TABLE 1. Abundance of resident warbler pairs by phenotypes in swamp forests and uplands in Sterling State Park
Forest, New York, 2001-2{X)6 and 2008.

Male and female phenotypes of pairs, males listed first

Habitat

GW" X GW

BR X GW GW X BR

GW X BW

BW X BW

BW X GW

BR X BW

Swamp forests

42

1 0

0

2

0

0

Uplands

46

2 3

1

40

1

2

^ Abbreviations: GW = Golden-winged Warbler, BR = Brewster’s Warbler (hybrid), and BW ^ Blue-winged-Warbler.

laboratories. Pooling results from these two
laboratories, we compare the phenotype-haplo-
type concordance of 35 phenotypically pure
Golden-winged Warbler individuals in upland
and swamp forest habitats in Sterling Forest State
Park.

RESULTS

All upland sites included herbaceous patches of
goldenrod {Solidago spp.) and grasses, and shrub
patches of dogwood {Corniis spp.) and briers
(Riihus spp.). Poison ivy {Toxocodendron radicans)
and Asiatic bittersweet (Celastrus orhicidatus)
were widely distributed and, at times, formed
dense stands. Forests formed at least half of the
boundary for all upland territories. Results from
upland sites were pooled because vegetation varied
as much within one site as among the sites, and
because inspection of our data did not suggest any
difference in population parameters among these
sites. Red maple dominated the canopy in swamp
forests of Sterling Forest State Park, and provided
—60-80% canopy closure. Dead tree trunks
suggest that elms {Ulmus spp.) were once domi-
nant. Spice bush (Lindera benzoin) was the most
widely distributed shrub followed in abundance by
arrowwood (Viburnum recognitum) and wild raisin
(V. cassinoides). Poison sumac (Toxocodendron
vernix) and swamp azalea (Rhododendron visco-
sum) occurred occasionally. Tussock sedge (Carex
stricta) strongly dominated the herb layer and was
the principal substrate for nests in wetlands. Dense
patches of Phragmites occurred in several loca-
tions, although Golden-winged Warblers seldom
perched on this plant and only once used it as a nest
substrate. Data from all swamp forests were pooled
for the same rea.sons applied to the upland sites.

Phenotypically pure Golden-winged Warblers
during 2001-2006, and 2008 comprised 94% (85
of 90) of the resident individuals in swamp forests
and 53% (99 of 186) in uplands (Table 1) (X" =
43.97; df = \, P < 0.001). The proportion of
Golden-winged Warbler individuals in these two

habitats was reflected in pairings among the pure
and hybrid phenotypes. Phenotypically pure
Golden-winged Warblers formed 93% (42 of 45)
of all pairs in swamp forests but only 48% (46 of
95) in uplands (2G = 23.81; df = 1, F < 0.001).
The rate of introgression due to hybrid pairings
was marginally lower in swamp forests. Only 1%
(1 of 85) of the phenotypically pure Golden-
winged Warbler individuals formed a social pair
with either a hybrid or a Blue-winged Warbler in
swamp forests, but 7% (7 of 99) of the
phenotypically pure Golden-winged Warblers
formed a hybrid pair in uplands (Fisher’s exact
one-tailed test; df = 1, F = 0.051).

The probability of nesting success in uplands
varied considerably among years and between
Golden-winged and Blue-winged warblers (Ta-
ble 2). This large variation may be partially due to
differences in predator abundance, although the
variation is certainly partially due to the small
sample of nests per habitat per year. Eastern
chipmunks (Tamias striatus) were at least com-
mon on ROW, and became abundant following
oak (Quercus spp.) mast years when as many as
five to 10 chipmunks were often observed
concun'ently in 2002 and 2004. Black rat snakes
(Elaphe obsoleta) were consistently common on
ROW, probably due to the warming effect of
direct sunshine and the presence of shelter in
exposed rock talus. In contrast, no chipmunks or
black rat snakes were observed in swamp forests
during this study. The abundance of these
predators on most ROW in Sterling Forest State
Park may account for the variable and compara-
tively low nesting success of Golden-winged
Warblers in uplands.

The estimated probability of nesting success for
2001, 2003-2006, and 2008 (Table 2) for Golden-
winged Warblers in swamp forests (65% of 23
nests) was greater than in uplands (37% of 32
nests) (one-tailed r-test; df = 1, F < 0.0485).
Nesting success for Blue-winged Warblers in
uplands was moderately high at 54% based on 28

276

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2, June 2010

TABLE 2. Probability of nesting success for pairs of Golden-wi
State Park Lorest, New York. Six re-nests are includecJ.

nged Warblers and Blue-winged Warblers in Sterling

Year

Golden-winged X Golden-winged Warblers

Blue-winged X Blue-

winged Warblers

Swamp forest

Upland

Swamp forest

Upland

2001

1“ (373'^)

0.49 (4/4)

0

1 (2/2)

2003

0.37 (3/3)

0.7 (7/7)

1 (1/1)

0.34 (3/3)

2004

0.59 (3/3)

0.26 (5/5)

0

0.29 (5/5)

2005

0.57 (6/6)

0.29 (4/4)

0

0.77 (5/4)

2006

0.82 (6/6)

0.23 (9/7)

0.21 (2/1)

0.61 (10/9)

2008

0.44 (2/1)

0.32 (3/3)

0

0.42 (3/3)

Totals

0.65 (23/22)

0.37 (32/30)

0.42 (3/2)

0.54 (28/26)

Probability of nesting success (Mayfield 1961).
Number of nests.

'■ Number of pairs.

nests, but not significantly better than for Golden-
winged Warblers. The low value for nesting
success of Blue-winged Warblers in swamps
(Table 2) is based on only three nests and is not
reliable.

We collected >300 blood samples from adults
and sets of nestlings to ascertain if swamp forests
attracted a population of birds with greater genetic
purity than uplands. Preliminary analyses of
mitochondrial DNA indicate swamp forests sup-
port a higher proportion of Golden-winged
Warbler phenotype-haplotype matches than up-
land habitat. All 10 phenotypically pure Golden-
winged Warblers from swamp forests that have
been analyzed had the ancestral Golden-winged
Warbler haplotype, while 10 of 25 of the
phenotypically pure Golden-winged Warblers
from uplands had the ancestral Blue-winged
Warbler haplotype (Table 3). These preliminary
results strongly suggest swamp forests support a
higher proportion of Golden-winged Warbler
phenotype-haplotype matches than upland habitat.

DISCUSSION

Golden-winged Warblers expanded into the
Northeast during the last 150 years, primarily
into secondary succession shrublands on aban-
doned farmland and most of the populations now
nest in the North Central states, often on clear cut
timberlands (Gill 1980, Huffman 1997). Studies
throughout the Golden-winged Warbler range
de.scribe use of anthropogenic, disturbance eco-
systems on dry sites: e.g., Georgia (Klaus and
Buehler 2001 ), northcentral New York (Confer et
al. 2003), Pennsylvania (Kubel 2005; J. L. Larkin,
unpubl. data), Virginia (Wikson et al. 2007), West
Virginia (Canterbury and Stover 1999), and

Wisconsin (Roth and Lutz 2004). However, other
studies have found nesting by Golden-winged
Warblers in wetlands; Michigan (Berger 1958),
southern New York (this study). North Carolina
(Rossell et al. 2003), and northern New York
according to surveys by the New York State
Department of Environmental Conservation (T. Q.
Rush and T. J. Post, unpubl. data). Some speculate
the historical abundance of beaver meadows and
other wetlands provided the evolutionary crucible
(Hunter et al. 2001 ).

Both wet and dry Golden-winged Warbler
territories are dominated by patches of herbs and
shrubs and, usually, a forested boundary. Upland
sites with these attributes have been described for
central New York (Confer and Knapp 1981),
northcentral New York (Confer et al. 2003),
southern New York (this study), and clearcuts in
Georgia (Klaus and Buehler 2001). GIS mapping
of vegetation has shown Golden-winged Warbler
territories include a diverse mosaic of succession-
al habitat types, especially herbaceous openings,
and are often surrounded by forest in a wetland in
North Carolina (Rossell et al. 2003), and in wet
and dry territories in Sterling Forest State Park (J.
L. Confer and J. L. Larkin, unpubl. data).
Wetlands with many patches of vegetation in
both southern New York (this study) and northern
New York, according to surveys by the New York
State Department of Environmental Con.servation
(T. Q. Rush and T. J. Post, unpubl. data), attracted
a nearly pure Golden-winged Warbler population
despite the presence of Blue-winged Warblers in
nearby uplands.

Our multi-year ob.servations in Sterling Forest
State Park overcome extreme annual variation and
show major differences in population demograph-

Confer ei cil. • WARBLER HABITAT DIFFERENTIATION

277

TABLE 3. Distribution of ancestral Golden-winged or
Blue-winged warbler mitochondrial DNA for birds with
pure Golden-winged Warbler phenotype nesting in
two habitats.

Mitochondrial haplotype

Habitat Golden-winged Warbler Blue-winged Warbler

Swamp Forests 10 0“

Uplands 15 10

“ Fisher's exact probability test; P for two-tailed test = 0.033.

ics between wet and dry habitats. The proportion
of Golden-winged Warblers to Blue-winged
Warblers and hybrid phenotypes is much higher
in swamp forests than in uplands. The absence of
ancestral Blue-winged Warbler mtDNA in the
sample of phenotypically pure Golden-winged
Warblers in swamp forests indicates a history of
infrequent hybridization for Golden-winged War-
blers in swamp forests. We observed a low
frequency of hybrid pairings in swamp forests,
which suggests the current rate of introgression is
also lower in swamp forests than in uplands.
Further, nesting success in swamp forests was
—75% better than nesting success in uplands. Our
results strongly indicate the unusual persistence of
Golden-winged Warblers in the presence of Blue-
winged Warblers in uplands of Sterling Forest
State Park is the result of a source/sink dynamic
between swamp forests and uplands.

Continued stability of genetically pure Golden-
winged Warblers in this region may require that
swamp forests sustain a sufficiently large popu-
lation to buffer stochastic events. We observed
about 55 Golden-winged Warbler pairs during
casual surveys in 2009 with 22 pairs in swamp
forests. We extrapolate, using GIS-based habitat
classification, to an estimate of —125 Golden-
winged Warbler pairs in Sterling Forest State Park
with —50 in swamp forests. The Sterling Forest
State Park population alone may be too small to
perpetuate the Golden-winged Warbler. Fortu-
nately, Sterling Forest State Park is within the
Hudson Highlands, a tri-state parkland of
28,000 ha. Red maple hardwood forests occur
throughout this larger area. Management for
swamp forests throughout the larger Hudson
Highlands could help sustain the Golden-winged
Warbler population in this region. The presence of
Golden-winged Warblers in wetlands with a
mosaic of vegetation in North Carolina and
southern and northern New York with habitat

segregation from Blue-winged Warblers in the
New York studies may provide important exam-
ples for management in a wide area of sympatry.
Further analyses of genetic purity and population
demographics comparing uplands and wetlands
would be valuable.

ACKNOWLEDGMENTS

This work was conducted in collaboration with and
financially supported by the New York State Office of
Parks, Recreation and Historic Preservation, Non-game
Unit of the New York State Department of Environmental
Conservation, New York State Biodiversity Research
Institute, New York State Wildlife Grant program, and
The Sterling Forest Partnership. T. B. Lyons, R. W. Perry,
J. R. Gell, P. G. Novak, and J. A. Hutchinson, all of the
New York State Office of Parks, Recreation and Historic
Preservation, and L. M. Stenzler, D. F. Westneat, M. A. and
J. C. Yrizarri, and P. J. Hamill provided much appreciated
assistance. P. J. Hartman provided a valuable review of
the manuscript. The work of more than 50 field
assistants, principally from Ithaca College, made this study
possible.

LITERATURE CITED

Berger, A. J. 1958. The Golden-winged-Blue-winged
warbler complex in Michigan and the Great Lakes
area. Jack-pine Warbler 36:37-72.

Canterbury, R. A. and S. M. Stover. 1999. The Golden-
winged Warbler: an imperiled migrant songbird of the
southern West Virginia coalfields. Green Lands 29:
44-51.

Confer, J. L. 2008. Golden-winged Warbler (Vermivora
chrysoptera). Pages 468^69 in The second atlas of the
breeding birds of New York State (K. J. McGowan and
K. Corbin, Editors). Cornell University Press, Ithaca,
New York, USA.

Confer, J. L. and K. Knapp. 1981. Golden-winged
Warblers and Blue-winged Warblers: the relative
success of a habitat specialist and a generalist. Auk
98:108-1 14.

Confer, J. L. and S. M. Pascoe. 2003. The avian
community on utility rights-of-ways and other man-
aged shrublands in northeastern United States. Forest
Ecology and Management 185:193-206
Confer, J. L., J. Gebhards, and J. Yrizarry. 1998.
Golden-winged and Blue-winged warblers at Sterling
Forest: a unique circumstance. Kingbird 39:50-55.
Confer, J. L., P. E. Allen, and .1. L. Larkin. 2003. Effects
of vegetation, interspecific competition, and brood
parasitism on Golden-winged Warbler nesting success.
Auk 121: 138-144.

Dabrowski, a., R. Fraser, J. L. Confer, and J. L.
Lovette. 2005. Geographic variability in mitochon-
drial introgression among hybridizing populations of
Golden-winged (Vermivora chrysoptera) and Blue-
winged (V^. pinus) warblers. Conservation Genetics
6:843-853.

278

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

Eaton, E. H. 1914. Birds of New York. Part 2. University
of the State of New York, Albany, USA.

Gill, F. B. 1980. Historical aspects of hybridization
between Blue-winged and Golden-winged warblers.
Auk 97:1-18.

Gill, F. B. 1997. Local cytonuclear extinction of the
Golden-winged Warbler. Evolution 51:519-525.

Huffman, R. D. 1997. Effects of residual overstory on bird
use and aspen regeneration in aspen harvest sites in
Tamarac National Wildlife Refuge, Minnesota. Dis-
sertation. West Virginia University, Morgantown,
USA.

Hunter, W. C., D. A. Buehler, R. A. Canterbury, J. L.
Confer, and P. B. Hamel. 2001. Conservation of
disturbance-dependent birds in eastern North America.
Wildlife Society Bulletin 29:440-455.

Klaus, N. A. and D. A. Buehler. 2001. Golden-winged
Warbler breeding habitat characteristics and nest
success in clearcuts in the southern Appalachian
Mountains. Wilson Bulletin 1 13:297-301.

Kubel, J. 2005. Breeding ecology of Golden-winged
Warblers in managed habitats of central Pennsylvania.
Thesis. Pennsylvania State University, State College,
USA.

Mayfield, H. 1961. Nesting success calculated from
exposure. Wilson Bulletin 73:255-261.

Pennsylvania Breeding Bird Atlas. 2009. http://www.
carnegiemnh.org/atlas/index.htm (accessed 14 July
2009).

Reschke, C. 1990. Ecological communities of New York
State. New York Natural Heritage Program. New York
State Department of Environmental Conservation,
Latham, New York, USA. http://www.dec.ny.gov/
animals/29389. html (accessed 17 July 2009).

Rossell Jr., C. R., S. C. Patch, and S. P. Wilds. 2003.
Attributes ot Golden-winged Warbler territories in a
mountain wetland. Wildlife Society Bulletin 31:1099-
1104.

Roth, A. M. and S. Lutz. 2004. Relationship between
territorial male Golden-Winged Warblers in managed
aspen stands in northern Wisconsin. Forest Science
50:153-161.

Sauer, J. R., J. E. Hines, and J. Fallon. 2008. The
North American Breeding Bird Survey, results and
analysis 1966-2007. Version 5.15.2008. USGS, Pa-
tuxent Wildlife Research Center, Laurel. Maryland,
USA.

Shapiro, L. H., R. A. Canterbury, D. M. Stover, and R.
C. Fleischer. 2004. Reciprocal introgression between
Golden-winged Warblers {Vermivora chrysoptera) and
Blue-winged Warblers (Vc pinus) in eastern North
America. Auk 121: 1019-1030.

Wilson, M. D., B. D. Watts, M. G. Smith, J. P. Bredlau,
AND L. W. Seal. 2007. Status assessment of Golden-
winged Warblers and Bewick’s Wrens in Virginia.
Technical Report Series, CCBTR-07-02. Center for
Conservation Biology. College of William and Mary,
Williamsburg, Virginia, USA.

The Wilson Jounuil of Ornithology 1 22(2):279 -287, 2010

GENETIC AND MORPHOLOGICAL VARIATION OF THE
SOOTY-CAPPED BUSH TANAGER {CHLOROSPINGUS P/LEATUS),
A HIGHLAND ENDEMIC SPECIES FROM COSTA RICA
AND WESTERN PANAMA

TANIA CHAVARRIA-PIZARRO,' - GUSTAVO GUTIERREZ-ESPELETA,'
ERIC J. EUCHS,' AND GILBERT BARRANTES'

ABSTRACT. — We examined the effect of geographic isolation on morphology, genetic structure, and abundance of the
Sooty-capped Bush Tanager {Chlorospingits pHeatus), an endemic species restricted to highlands ot Costa Rica and western
Panama. Abundance and morphology were measured at five study sites and genetic variation was calculated from three
microsatellite loci. We expected geographic discontinuities in this species' distribution to have an effect on its morphology
and genetic structure. Genetic variation was higher within than between populations with no effect of geographic barriers
on population genetic divergence in this species, indicating gene flow is high between populations. Unique alleles were
detected in each population and G, values increased with geographic distance between populations. Some morphological
traits differed between populations, which may be caused by adaptation to different selective pressures in each population.
Molecular data did not differ between the two color morphs that coexist in two isolated populations, which were considered
different species. Received 16 July 2009. Accepted 22 November 2009.

Highland endemic avian species in the moun-
tains of Costa Rica and western Panama cuiTently
show naturally isolated and discontinuous distri-
butions, reflecting their confinement to mountain
peaks and their isolation from other highland
regions in the Neotropics (Stiles 1983; Barrantes
2000, 2009). Morphological and molecular diver-
gence documented for some of these endemic
highland species in Costa Rica correspond to this
natural isolation (Stiles 1983, Barrantes 2000,
Barrantes and Sanchez 2000). For example,
isolated populations of the Volcano Hummingbird
iSelasphorits flammula) and Fiery-throated Hum-
mingbird (Panterpe insignis) differ in coloration
and morphological dimensions (Stiles 1983,
1985). Similarly, geographic bamers have affect-
ed the genetic and morphological divergence
among populations of the Black-and-yellow
Phainoptila {Phainoptila melano.xantha) (Bar-
rantes 2000, Barrantes and Sanchez 2000).

The Sooty-capped Bush Tanager (Chlorospin-
gus pileatiis) is endemic to the highlands of
Costa Rica and western Panama. This species
is naturally restricted to the summit of high
mountains and primarily inhabits forest edges and
open areas in the cloud forest and paramo. Sooty-
capped Bush Tanagers forage in flocks of 5-20
individuals, often accompanied by other small

' Escuela de Biologia, Ciudad Universitaria Rodrigo
Facio, Univer.sidad de Costa Rica, San Jose. Costa Rica.

^Corresponding author;
e-mail: taniachavarria79@yahoo.com

birds (Powell 1985, Stiles and Skutch 1989). A
grayish-green morph of Sooty-capped Bush Tan-
ager (lighter morph) coexists on two high
volcanoes (Irazu and Tumalba) of central Costa
Rica with the darker and more widely distributed
morph. The grayish-green morph was first con-
sidered a different species (Zeledon’s Chloro-
spingus, Chlorospingits zeldoni) by Ridgway
(1905), and Eisenmann (1955) and Slud (1964)
continued C. zelecloni as a different species.
However, Carriker (1910) suggested that C.
zelecloni was a color morph of Sooty-capped Bush
Tanager. Johnson and Brush (1972) analyzed the
structure and pigment composition of feathers of
both taxa and concluded C. zelecloni was a moiph
of Sooty-capped Bush Tanager. However, no
further investigations have been conducted to test
this hypothesis (e.g., using molecular data).

We used molecular data to analyze the effect of
isolation (i.e., distance and barriers between
populations) on genetic structure and moiphology
of the Sooty-capped Bush Tanager. We .specifi-
cally hypothesized that small populations isolated
by effective geographical barriers have a notable
reduction in genetic variation. In addition, we
searched for genetic differences between the two
color morphs on Irazii Volcano. To our knowl-
edge, this is the first fine-.scale genetic analysis,
using microsatellites of an avian species endemic
to the isolated highland regions of Costa Rica and
western Panama. Most species in this hotspot of
diversity and endemism are restricted to a reduced
portion of the highland forests and have low

279

280

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 2, June 2010

FIG. 1. Sampling localities in Costa Rican mountain ranges; Talamanca, Central Volcanic, and Tilaran. Locations of
Monterverde, Irazu, and Cerro de la Muerte are indicated by black arrows.

abundance (Jankowski and Rabenold 2007, Bar-
rantes 2009).

METHODS

Field Data Collection. — We conducted field
work in highland forests of three Costa Rican
mountain ranges: Talamanca Mountain Range,
Central Mountain Range, and Tilaran Mountain
Range (Fig. 1) from July 2003 through December
2005. Censuses and tissue collections were
conducted at five different sites: Monteverde
(10° 19' N, 84° 47' W; 1,550 m asl) in the
Tilaran Mountain Range; Poas Volcano
(10° 1 1' N, 84° 13' W; 2,700 m asl), Irazu Vol-
cano (09° 59' N, 83° 52' W; 3,200 m asl), and
Barva Volcano (10° 08' N, 84° 06' W; 2,800 m
asl) in the Central Mountain Range; and Macizo
Cerro de la Muerte (09° 33' N, 83° 43' W;
3,100 m asl) in the Talamanca Mountain Range.
Mountain ranges and field sites within mountain
ranges are separated by geographic discontinuities

of different magnitude such as watersheds and
mountain passes (Fig. 1 ).

We censused birds 4-7 times in each of the
sites to account for possible fluctuations due to
climatic conditions. Censuses started at 0545 hrs
and all individuals heard and seen within 25 m of
each side of the transect were recorded. Length of
transects varied from 1.9 to 3.0 km among sites.
Thus, abundance was expressed as individuals/km
to allow comparison among sites. We used
abundance categories based on mean values
across censuses for each site. Categories were
defined as very common (>20 individuals
observed/census), common (10-19 individuals/
census), uncommon (5-9 individuals/census), and
rare (<5 individuals/census). We calculated the
area of the potential available habitat for Sooty-
capped Bush Tanagers on each mountain range
using Geographical Information Software (Arc-
View GIS 3.3) on a digital elevation map of
Costa Rica ( 1 :2()(),()()(); Lambert conformal con-

Chaviinia-Pizcirro cf at. • SOOTY-CAPPED BUSH TANAGERS

281

ical projeclion) (Fig. I ). This area was estimated
from the lower limit of the altitudinal distribution
of Sooty-capped Bush Tanager to the summit of
the mountains in each mountain range.

We set 5-8 mist nets (12 m long) during each
visit to a study site. The following measurements
were taken for each individual captured: body
mass, length of the beak from distal end of nares
to the beak tip, width of the beak at front of nares,
length of the culmen from the front of the skull to
the beak tip, tarsus from the tibiotarsus joint to the
distal end of the tarsometatarsus, flattened wing
cord, and tail length. These morphological traits
were chosen as they are expected to respond
adaptively to ecological differences in habitat
(Grant 1986, Price and Boag 1987) and were
taken by the same person (GB). Birds were
banded (plastic bands) to avoid re-measuring the
same individuals. Moiphological data were first
analyzed using a stepwise (forward) discriminant
function analysis (DFA). This option excludes
those variables that have no further effect in
explaining the between-population variance. We
used a MANOVA on those variables included in
the DFA. Statistical analyses were performed in
Statistica Version 6.0 (StatSoft Inc., Tulsa, OK,
USA).

Laboratory Procedures. — ^DNA was extracted
from liver tissue of 56 individuals that had been
obtained in previous years and maintained in the
tissue collection at the Zoology Museum at the
University of Costa Rica. We collected blood
samples from five additional individuals for a
total of 61 individuals (14 from Cerro de la
Muerte, 9 from Irazu Volcano, 18 from Barva
Volcano, 1 1 from Poas Volcano, and 9 from
Monteverde). Tissue samples were preserved in
95% alcohol and stored at —20° C until DNA
extraction. DNA was extracted using a commer-
cial extraction kit for animal tissue and blood
(Promega, Madison, WI, USA) following the
specifications provided by the manufacturer.

Genotypes of the 61 individuals were obtained
using three microsatellite loci (Escpl, Escp4, and
Escp6; Hanotte et al. 1994) isolated from the
related species Emheriza schoeniclus (Yuri and
Mindell 2002). The three microsatellites used
were the most polymorphic for E. schoeniclus.
Microsatellites were amplified using an Eppen-
dorf Mastercycler thermocycler in a volume of
25 pL (12.5 pL of PGR Master Mix 2X
PROMEGA, 2.5 pL [10 pm] of each primer,
5 pL of DNA, and 2.5 pL Fl20) following

protocols for each locus as described in Hanotte
et al. (1994). Genotypes were obtained using an
automatic sequencer (ABl Prism 310, Applied
Biosystems, Palo Alto, CA, USA). Alleles were
sized using program GENOTYPER (3.7NT,
Applied Biosystems, Palo Alto, CA, USA).

We counted the number of alleles per locus and
calculated their frequencies in each population.
We tested deviations of these alleles from Hardy-
Weinberg equilibrium with a Chi-square test
using program GENALEX 6 (Peakall and Smouse
2006); sequential Bonferroni correction was
applied due to performance of multiple simulta-
neous tests (Rice 1989). The observed (Ho) and
expected (He) heterozygosity per locus in each
population were calculated using program ARLE-
QUIN Version 3.1 (Schneider et al. 2000,
Excoffier et al. 2005). An analysis of molecular
variance (AMOVA) was used to assess population
structure with 30,000 permutations to infer
significance of E^i values (Weir and Cockerham
1984). AMOVA calculations were conducted
using ARLEQUIN Version 3.1 (http://cmpg.
unibe.ch/software/arlequin3). We also checked
all loci for the presence of null alleles (Chakra-
borty et al. 1992, Brookfield 1996) using Micro-
Checker Version 2.2.3 (University of Hull, Hull,
UK; http://www.microchecker.hull.ac.uk/; van
Oosterhout et al. 2004). All loci were tested based
on a dinucleotide repeat motif and 1,000 permu-
tations.

We further examined population structure and
admixture using a model-based clustering method
as implemented in the Bayesian clustering
program STRUCTURE (Pritchard et al. 2000,
Falush et al. 2003). We inferred population
structure using a model which allowed for
admixture and correlated allele frequencies. Each
Monte Carlo Markov Chain used to infer
population structure was based on 1,000,000
iterations and a burn-in of 100,000. Multiple runs
were conducted changing the number of putative
populations (K) between one and five. At least
three independent runs were assessed to estimate
the likelihood of the data and the posterior
probabilities for each fixed number of populations
(K).

We calculated genetic distances between pop-
ulations using two parameters: Nei's (Dm) genetic
distance and Es, (Nei 1987, Nei and Kumar 2000).
We correlated genetic distances (E^i values) with
geographic distances (km) using a Mantel test and
examined whether area of habitat available for

282

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2, June 2010

each population con'elated with allele number
and/or genetic distance {Dm) using ARLEQUIN.

RESULTS

Null Alleles. — ^We found no evidence of null
alleles on the first two loci (Escpl and Escp4)
with combined probability for all cases of L* >
0.05. The third locus (Escp6) showed a significant
homozygote excess congruent with the presence
of null alleles (P < 0.001). No evidence of
scoring errors or allele drop-out was found for any
loci. Observed allele frequencies for the third
locus only deviated slightly (at 0.01 level) from
frequencies estimated considering the presence of
null alleles. AMOVA analyses were performed
twice using only Escpl and Escp4 loci, and using
all three loci. Results were similar with small
deviations at the 0.001 level. Thus, we used all
three loci for further analysis.

Genetic Variation and Population Structure. —
Microsatellite loci were highly polymorphic
within each population relative to the number of
genotyped individuals (Table 1 ). Allele number in
any locus ranged from seven to 18 across
populations. The Barva Volcano population had
the highest average number (± SD) of alleles
(9.33 ± 0.66), while Monteverde had the lowest
(5.67 ± 0.66; Table 2). All populations had
unique alleles, although number of these alleles
varied across populations: four in Barva, three in
Cerro de la Muerte and Monteverde, and two in
Irazu and Poas populations (Table 1). Frequency
of the allele Escp6 significantly deviated from
Hardy-Weinberg expectations (lower observed
values than expected) in Cerro de la Muerte {X^
= 64.36, df = 36, P = 0.003), Barva Volcano {X^
= 1 13.56, df = 45, P < 0.0001 ), and Monteverde
{X^ = 48.0, df = 21, P = 0.001).

All populations had high levels of genetic
diversity ba.sed on observed heterozygosity, which
did not deviate significantly from expected
heterozygosity in all populations (Table 2). Ge-
netic variation was higher within populations than
between populations; 98% of the total variance
was distributed within populations and only 2%
among populations (tp^, = 0.0153; P > 0.17).
The.se results are consistent with high levels of
gene flow among populations. Genetic structure
estimates among populations were similar if the
third locus was removed due to the likelihood of
null alleles (tps, = 0.0154; P > 0.18). Similarly,
analyses using STRUCTURE showed no evidence
of significant population structure. The highest

likelihood was obtained when the number of
populations was set to one (K = 1), which
suggests complete admixture.

Pairwise values ranged from 0.04 between
Barva and Poas to 0.09 between Irazu and
Monteverde (Table 3). values and geographic
distances were not significantly correlated (Man-
tel test, r = 0.49, P = 0.09); similar results were
obtained using a simple linear correlation.

We did not detect any genotypic difference
between three individuals of the grayish-green
morph and six individuals of the darker morph in
the Irazu population. Both color morphs shared
some genotypes.

Morphological Variation. — We captured 72
individual Sooty-capped Bush Tanagers; 24 at
Cerro de la Muerte (Estacion Biologica Cerro de
la Muerte), 11 at Irazu, 18 at Barva, 10 at Poas,
and nine at Monteverde. Culmen length, wing
cord, tail length, beak width, and body mass
differed among populations (DFA: f’20.168 = 2.63,
P = 0.0003). The Monteverde population had the
shortest culmen and wing length and the widest
beak; the Irazu population had the longest tail
(Wilks’ lambda = 0.64, df = 16/174, P = 0.04;
Table 4).

Population Size and Habitat Area. — Sooty-
capped Bush Tanagers were very common (>20
individuals/census, corrected by distance) at
Barva and Poas, and common at other sites (10-
19 individuals/census). The highland habitat
available for Sooty-capped Bush Tanagers de-
crea,sed from the Talamanca Mountain Range to
the Tilaran Mountain Range. The Talamanca
Mountain Range has 80.8% of the total natural
highland habitat potentially available for this
species (293,135 ha), while the Central Mountain
Range has 17% and the Tilaran Mountain Range
has 3%. Area of available habitat was not
coiTelated with number of alleles (/• = 0.53, n —
5, P = 0.16) nor with genetic distance (Nei’s) (r
= 0.62, /? = 5, P = 0. 1 1) in Sooty-capped Bush
Tanager populations.

DISCUSSION

All populations of the Sooty-capped Bush
Tanager had high genetic variation as indicated
by genetic diversity indices and number of alleles
per locus. The genetic diversity within popula-
tions of Sooty-capped Bush Tanager is unexpect-
edly high for an endemic species with a small and
naturally fragmented geographical distribution
(Tables I, 2). Within-population genetic variation

TABLE 1. Allele frec|uencies for three microsatellite loci (Esc|.il, Esci.i4, Esc|.t6) in five populations of Sooty-capped Bush Tanager (/; — individuals with amplified
microsatellites per loci in each population) in the highlands of Costa Rica.

Chavanhi-Pizarro d al. • SOOTY-CAPPED BUSH TANAGERS

283

sO

o

o

So

o

O

c

O

c

d

d

d

d

d

•T

u

o

o

o

LU

o

o

o

o

o

d

d

d

d

d

Os

oo

—

os

00

tn

os

o

o

O

(N

cn

—

—

o

d

d

d

d

d

so

>/~)

o

so

o

in

o

o

—

d

d

d

d

d

o

r-

in

os

in

o

CN

o

(n

d

d

d

d

d

r-j

o

in

CN

o

—

o

o

n

d

d

d

d

d

so

in

o

—

O

o

o

o

d

d

d

d

00

•t

o

sO

o

o

o

q

o

o

d

d

d

d

-

o

o

o

o

o

o

o

o

uj

d

d

d

o

sO

o

o

r-

4

o

o

o

—

d

d

d

d

d

ri

-t

in

r^!

o

o

d

d

d

d

00

r»“,

’1’

o

o

o

o

o

o

q

q

d

d

-i-

sO

r-

in

sO

O

q

o

o

o

d

d

d

d

ri

r*-,

r-

o

o

o

o

o

o

o

o

d

d

d

d

os

in

o

os

(U

V

“O

u.

u

V

>

>

C3

>

u

>

V

>

<v

2

'3

N

oo

'C3

c

o

cJ

C3

cS

QQ

O

a-

s

o

O

sO

—

o

o

O

o

o

d

d

d

d

d

00

o

o

o

o

o

o

m

o

so

d

d

d

d

d

'O

in

=i.

u

rJ

o

o

o

o

o

o

o

o

o

d

d

d

d

d

148

s:

—

oo

os

sO

o

o

'sD

o

o

o

o

O

o

o

d

d

d

d

d

00

00

o

sO

o

o

=

o

o

(N

o

d

d

d

d

d

sO

m

so

oo

o

so

o

r-

o

o

o

d

d

d

d

d

■tj-

rn

r-

sO

so

o

o

—

00

o

o

o

o

o

d

d

d

d

d

<N

00

o

so

o

o

00

o

o

O

o

o

so

—

d

d

d

d

d

o

W

r--

1

o

m

o

o

—

o

o

(N

d

d

d

d

d

oo

r-

so

in

<N

o

o

—

o

(N

'Tt

■“

d

d

d

d

d

u

U

sO

r-

CN

(N

<N

o

CN

(n

<N

(N

(N

CN

r4

d

d

d

d

d

<N

o

t^

o

o

(N

o

o

o

d

d

d

d

d

120

CN

r-

o

so

o

os

—

— H

m

CN

00

d

d

d

d

d

—

sO

O

o

o

o

in

o

o

o

o

o

o

so

d

d

d

d

d

3

o

r^

so

o

o

—H

o

o

— ■

d

d

d

d

d

rf

d

162

r-

o

o

o

o

■d

V

3

C

o

d

o

d

o

d

o

d

o

d

o

3

C

(N

'w

C

c

O

o

U

U

<D

— :

LQ

<u

“O

u.

UJ

c

>

O

CQ

(U

OJ

>

>

>

O

c

o

s

CQ

u

<

H

1

U

'3

N

03

d

a

>

CC

CQ

c/5

'C3

O

o,

<

C. Muerte 0.00 0.00 0.27 0.04 0.00 0.04 0.18 0.23 0.04 0.00 0.00 11 0.00 0.00 0.09 0.00

Irazu V. 0.00 0.06 0.07 0.43 0.07 0.00 0.14 0.14 0.07 0.00 0.00 7 0.00 0.00 0.00 0.00

BarvaV. 0.00 0.10 0.00 0.11 0.00 0.29 0.07 0.07 0.04 0.00 0.04 14 0.21 0.04 0.00 0.00

Poas V. 0.00 0.00 0.06 0.12 0.06 0.12 0.06 0.06 0.00 0.00 0.00 8 0.00 0.06 0.12 0.00

Monteverde 0.06 0.17 0.00 0.00 0.00 0.06 0.00 0.00 0.00 0.06 0.00 7 0.00 0.00 0.00 0.12

TABLE 2. Number of individuals in), allele number (Na), observed heterozygosity (Ho), and expected heterozygosity (He) calculated for each locus in Eve populations of
Sooty-capped Bush Tanager in Costa Rica.

284

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122. No. 2. June 2010

r-

to

(N

cn

o

o

O

O

o

d

d

d

d

d

u

X

+ I

+1

+1

+1

+1

to

nC

00

00

00

00

r-

d

d

d

d

d

(O

r-

to

—

—

CN

—

(N

d

d

d

d

d

Ho

+1

+1

+1

-f-l

+1

a

IT)

to

00

o

c/2

'O

r-

NO

+ 1

d

d

d

d

d

C

cd

u

B

m

00

NO

m

NO

V3

m

00

NO

m

NO

•2

d

d

d

d

d

0

H

id

Z

+1

+ 1

+1

+1

+1

r-

m

m

r--

r-

00

m

NO

NO

00

00

os

to

m

to

00

to

r-

r-;

p

—

to

—

(N

—

d

+1

+1

+ 1

+1

+1

p

fO

NO

nO

cq

NO

m

NO

00

iri

ON

00

OJ

nO

NO

m

00

00

00

00

00

d

d

d

d

d

to

o

(N

o

o

X

to

(^1

NO

to

d

d

d

d

d

nO

o

rd

ON

r-

o

00

r-

z

00

r-

X

to

00

ON

00

00

On

d

d

d

d

On

r--

o

On

O

X

r-

00

NO

ON

00

d

d

d

d

d

G

Na

ON

-

o

r-

o

ON

00

ON

X

00

r4

r-

NO

00

00

r-

OO

d

d

d

d

d

r-

oo

r-

c

00

o

X

r-

r-'

00

r-

d

d

d

d

d

a

CJ

u]

Na

00

00

00

00

oo

-t

ON

ty^

o

On

V

3

■o

u.

o

o

>

>

>

>

>

V

s

'O

N

c

Q

cJ

f3

Ui

C3

QQ

O

CL

is expected to be small compared to between-
population variation in species with small ranges
and fragmented distributions, mainly due to
genetic drift or founder effects and small effective
population size (Peterson et al. 1992, Avise 1994).
The absence of genetic structure among popula-
tions of Sooty-capped Bush Tanagers indicates
gene flow may be relatively high among popula-
tions, counteracting the effects of genetic drift,
small effective population size, and local adapta-
tion even in small populations (Avise 1994).
However, time since population isolation may
have also affected the near absence of genetic
structure in this endemic species. Populations are
expected to retain most of their original genetic
composition (Templeton et al. 1995) if isolation is
relatively recent, as it has been hypothesized for
Costa Rican highland endemic species (Stiles
1983, Barrantes and Sanchez 2000, Weir 2006,
Barrantes 2009),

There were some genetic and morphological
differences across populations of the Sooty-
capped Bush Tanager. For example, the presence
of unique alleles in each population indicates that
processes of random drift, inbreeding, and perhaps
local adaptation are causing the genetic differ-
ences among populations. Genetic difference
between populations (Fs, values) tend to increase
with geographic distance (P = 0.09) suggesting
some limitation in dispersal. However, gene flow
among populations likely dilutes the effect of
these processes. Schwartz et al. (1986) indicated
that gene flow rates as low as one migrant per
generation are generally effective to prevent loss
of genetic variation from habitat fragmentation.

Morphology differed among some Sooty-
capped Bush Tanager populations. Moiphological
differences may be caused by different abiotic and
biotic pressures operating on different popula-
tions. For example, the habitat of the Sooty-
capped Bush Tanager in Monteverde is exposed to
relatively intense trade winds year round (Clark et
al. 2000). Birds in this population had shorter
wings (Table 4), which are more efficient at
maneuvering under strong wind conditions (Pen-
nycuick 1968, Thomas 1996, Margjjerie et al.
2007). There is no clear explanation for dilTer-
ences in other morphological traits (e.g., culmen,
tail, and beak width), as fruit species and habitat
structure appear similar across localities (Bar-
rantes and Loiselle 2002). Morphological diver-
gence between populations has been reported for
other endemic birds in the highlands of Costa Rica

Chavania-Pizano ci al. • SOOTY-CAIM^ED lUJSIl T'ANAOERS

285

TABLE 3. Geographic distance (km) (above diagonal)
and values (below the diagonal) between five
populations of Sooty-capped Bush Tanager in the
highlands of Costa Rica.

Site

C. Muerte

Irazii V.

Barva V.

Piiiis V.

Monteverde

C. Muerte

—

47

74

91

142

Irazii V.

0.042

—

34

48

103

Barva V.

0.047

0.063

—

15

71

Poas V.

0.063

0.060

0.040

—

55

Monteverde

0.053

0.090

0.059

0.078

—

(Stiles 1983, 1985; Barrantes and Sanchez 2000),
indicating rapid morphological divergence among
populations, if recent population fragmentation is
assumed (Stiles 1983, 1985; Barrantes and
Sanchez 2000; Barrantes 2009). The lack of
concordance between morphology and genetics
in this study is possibly caused, at least in part, by
the intrinsic characteristics of the microsatellites.
These genetic markers are neutral (or nearly so)
whereas morphological traits are frequently under
strong selective pressures (Zwaitjes 2003, San-
tiago-Alarcon et al. 2006), and changes in traits
presumed to be under selection often do not
correlate with variation of microsatellites (Philli-
more et al. 2008) or mitochondrial DNA markers
(McCormack et al. 2008, Weir et al. 2008).

We found no genetic differences between the
two color morphs in the Irazii population.
Genotypes vary among individuals of each morph
and some genotypes are shared with individuals of
the other morph. Our findings support the
hypothesis of Carriker (1910) and Johnson and
Brush (1972), who suggested that differences in
coloration reflect two morphs of the same species.
We found individuals of the grayish-green moiph
vary greatly in coloration. All grayish-green birds
also had yellow nasal tufts, and one had one
yellow secondary feather in its right wing (GB,
unpubl. data). Evidence that helps explain this

polymorphism is elusive. Birds of both morphs
use the same habitat, forage together, and
apparently mate with each other (Johnson and
Brush 1972; GB, unpubl. data). Johnson and
Brush ( 1972) suggested that brighter coloration of
grayish-green individuals is favored during peri-
ods of high volcanic activity when ash in the
atmosphere reduces visibility. However, informa-
tion on abundance of each morph during high and
low volcanic activity periods is needed to test this
hypothesis. This polymorphism may have evolved
recently by assortative mating with some repro-
ductive barriers, but its detection only would be
possible with detailed genetic and behavior
studies (Yeh 2004, McGlothlin et al. 2005).

The geographic distribution of the Sooty-
capped Bush Tanager on only three mountain
ranges, the number of naturally isolated popula-
tions in = 5), as well as the low number of alleles
analyzed (n = 3) imposed some restrictions on
analyses and limited the statistical power of some
results. For example, the low number of micro-
satellite loci limited use of more detailed analyses
to quantify gene flow. Similarly, lack of signif-
icant correlation between geographic and pairwise
genetic differences was likely caused by the small
number of isolated populations. In addition, we
did not survey Sooty-capped Bush Tanagers in
western Panama where the Fortuna Mountain Pass
may limit bird movement across this geographic
discontinuity, affecting the scope of our findings.

The geographic discontinuities that separate
populations of Sooty-capped Bush Tanagers as
well as the area of available habitat appear to have
had little effect in shaping the genetic structure of
this endemic species. The presence of divergent
morphologies and color moiphs in some popula-
tions and unique alleles in all populations of
Sooty-capped Bush Tanager suggest some isola-
tion and/or local adaptation of populations (Stiles
1983, 1985; Barrantes and Sanchez 2000). This

TABLE 4. Morphological trails (mean ± SD) for five populations of Sooty-capped Bush Tanager in the highlands of
Costa Rica. (Length = mm. body mass = g).

Morphology trails

Site

Culmen length

Wing cord

Tail length

Tarsus length

Beak length

Beak width

Mass

C. Muerte

15.35 ±1.11

71.6

± 2.91

64.0 ± 4.65

23.85

± 0.76

6.38 ± 0.32

6.38 ± 0.32

19.90 ±

1.44

Irazii V.

16.0 ± 0.84

70.3 1

± 2.63

55.0 ± 2.20

24.21

± 0.88

6.41 ± 0.28

6.42 ± 0.28

19.87 ±

0.87

Barva V.

15.22 ± 0.91

69.7

± 2.70

61.66 ± 3.52

23.70

± 0.76

6.40 ± 0.48

6.40 ± 0.48

20.28 ±

1.16

Poas V.

15.88 ± 1.15

69.5

± 2.51

61.93 ± 2.09

23.68

± 0.90

6.36 ± 0.38

6.36 ± 0.38

20.37 ±

2.57

Monteverde

13.58 ± 2.71

68.1

± 4.35

55.57 ± 3.69

23.14

± 1.15

6.43 ± 0.77

6.43 ± 0.77

17.65 ±

1.67

286

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

system of geographically isolated populations
deserves special conservation attention, because
it presents a unique opportunity to further study
incipient causes of population divergence in an
area of high endemism.

ACKNOWLEDGMENTS

We thank Johel Chaves, John McCormack, Jason Weir,
and C. E. Braun tor helpful comments that largely improved
the manuscript. We also thank Cesar Sanchez for helping
with field work and Federico Valverde for logistic support.
Financial support was provided by the Consejo Nacional
para Investigaciones Cientificas y Tecnologicas (CON-
ICIT), the Asociacion Ornitologica de Costa Rica, and the
Vicerrecton'a de Investigacion, Universidad de Costa Rica.

LITERATURE CITED

Avise, J. C. 1994. Molecular markers, natural history and
evolution. Chapman and Hall Press, New York, USA.
Barrantes, G. 2000. Ecology and evolution of Phainoptila
melano.xantha (Bombycilidae, Aves) in the highlands
of Costa Rica and western Panama. Dissertation.
University of Missouri-St. Louis, USA.

Barrantes, G. 2009. The role of historical and local
factors in determining species composition of the
highland avifauna of Costa Rica and western Panama.
Revista Biologi'a Tropical 57 (Supplement l):323-332.
Barrantes, G. and J. Sanchez. 2000. A new subspecies
of Black and Yellow Silky Flycatcher, Phainoptila
melano.xantha (Bombycilidae, Aves), from Costa Rica.
Bulletin of the British Ornithological Club 120:40-46.
Barrantes, G. and B. A. Loiselle. 2002. Reproduction,
habitat use and natural history of the Black-and-
Yellow Silky Flycatcher (Phainoptila melano.xantha),
an endemic bird of western Panama-Costa Rica
highlands. Ornitologi'a Neotropical 13:121-136.
Brookfield, J. F. Y. 1996. A simple new method for
e.stimating null allele frequency from heterozygotc
deficiency. Molecular Ecology 5:453—455.

Carriker Jr., M. A. 1910. An annotated list of birds of
Costa Rica, including Cocos Island. Annals of the
Carnegie Museum 6:314-915.

Chakraborty, R., M. De Andrade, S. P. Daiger, and B.
Budowle. 1992. Apparent heterozygote deficiencies
observed in DNA typing data and their implications in
forensic applications. Annals of Human Genetics
56:45-47.

Clark. K. L., R. O. Lawton, and P. R, Butler. 2000. The
physical environment of Monteverde. Pages 10-25 in
Monteverde: ecology and con.servation of a tropical
cloud fore.st (N. M. Nadkarni and N. T. Wheelwright,
Editors). University of Chicago Press, Chicago,
Illinois, USA.

Eisenmann, E. 1955. The species of Middle American
birds. Transactions of the Linnean Society of New
York 7:1-128.

Excoffier, L., G. Laval, and S. Schneider. 2005.
ARLEQUIN Version 3.0: an integrated software

package for population genetics data analysis. Evolu-
tionary Bioinformatics Online 1 :47-50.

Falush, D., M. Stephens, and J. K. Pritchard. 2003.
Inference of population structure using multilocus
genotype data: linked loci and correlated allele
frequencies. Genetics 164:1567-1587.

Grant, P. 1986. Ecology and evolution of Darwin’s
Finches. Princeton University Press, Princeton, New
Jersey, USA.

Hanotte, O., C. Zanon, A. Pugh, C. Greic, A. Dixon,
AND T. Burke. 1994. Isolation and characterization of
microsatellite loci in a passerine bird: the Reed
Bunting Emheriza schoenielus. Molecular Ecology
3:539-530.

Jankowski, J. E. and K. N. Rabenold. 2007. Endemism
and local rarity in birds of neotropical montane rain
forest. Biological Conservation 138:453—463.

Johnson, N. K. and A. H. Brush. 1972. Analysis of
polymorphism in the Sooty-capped Bush Tanager.
Systematic Zoology 21:245-262.

Marguerie, E., J. B. Mouret, S. Doncieux, and J. A.
Meyer. 2007. Artificial evolution of morphology and
kinetics in a flapping-wing mini-UAV. Bioinspiration
and Biomimetics 2:65-82.

McCormack, J. E., A. T. Peterson, E. Bonaccorso, and
T. B. Smith. 2008. Speciation in the highland of
Mexico: genetic and phenotypic divergence in the
Mexican Jay (Aphelocoma iiltramarina). Molecular
Ecology 12:423-434.

McGlothlin, j. W., P. G. Parker, V. Nolan Jr., and E.
D. Ketterson. 2005. Correlational selection leads to
genetic integration of body size and an attractive
plumage trait in Dark-eyed Juncos. Evolution 59: 658-
671.

Nei, M. 1987. Molecular evolutionary genetics. Columbia
University Press, New York, USA.

Nei, M. and S. Kumar. 2000. Molecular evolution and
phylogenetics. Oxford University Press, New York,
USA.

Peakall, R. and P. E. Smouse. 2006. GENALEX6:
genetic analysis in Excel. Population genetic software
for teaching and research. Molecular Ecology Notes
6:288-295.

Pennycuick, C. j. 1968. A wind-tunnel study of gliding
flight in the pigeon Cohunha livia. Journal of
Experimental Biology 49:509-526.

Peterson, A. T., P. Escalante, and S. A. Navarro.
1992. Genetic variation and differentiation in Mexican
populations of Common Bush-Tanagers and Chestnut-
capped Brush-Finches. Condor 94:244-253.

Phillimore, a. B, 1. P. F. Owens, R. A. Black, J.
CiiiTOCK, T. Burke, and S. M. Clegg. 2008.
Complex patterns of genetic and phenotypic diver-
gence in an island bird and the consequences for
delimiting conservation units. Molecular Ecology
17:2839-2853.

Powell, G. V. N. 1985. Sociobiology and adaptive
significance of interspecific foraging flocks in the
Ncotropics. Ornithological Monographs 36:713-731.

Price, T. D. and P. T. Boag. 1987. Selection in natural
populations of birds. Pages 257-287 in Avian genetics:

Cluivarrki-Pizarro el al. • SOOTY-CAPPED BUSH TANAGERS

287

a population and ecology approach (F. Cooke and
P. A. Buckley, Editors). Academic Press, London,
United Kingdom.

Pritchard, J. K., M. Stephens, and P. Donnelly. 2000.
Inference of population structure using multilocus
genotype data. Genetics 155:945-959.

Rice, W. R. 1989. Analyzing tables of statistical tests.
Evolution 43:223-225.

Ridgway, R. 1905. New genera of Tyrannidae and
Turdidae, and new forms of Tanagridae and Turdidae.
Proceedings of the Biological Society of Washington
17:211-214.

Santiago-Alarcon, D., S. M. Tanksley, and P. G.
Parker. 2006. Morphological variation and genetic
structure of Galapagos Dove (Zenaida galapagoensis)
populations: issues in conservation for the Galapagos
bird fauna. Wilson Journal of Ornithology 118:194-
207.

Schneider, S., D. RossiNt, and L. Excoffier. 2000.
ARLEQUIN Version 2000: a software for population
genetics data analysis. Genetics and Biometry Labo-
ratory, University of Geneva, Switzerland.

Schwartz, O. A., V. C. Bleich, and S. A. Holl. 1986.
Genetics and the conservation of mountain sheep Ovis
canadensis nelsoni. Conservation Biology 37:179-190.

Slud, P. 1964. The birds of Costa Rica. Distribution and
ecology. Bulletin of the American Museum of Natural
History 128:1^30.

Stiles, F. G. 1983. Systematics of the southern forms of
Selasphorus (Trochilidae). Auk 100:311-325.

Stiles, F. G. 1985. Geographic variation in the Fiery-
throated Hummingbird, Panterpe insignis. Ornitholog-
ical Monographs 36:23-30.

Stiles, F. G. and A. F. Skutch. 1989. A guide to the birds
of Costa Rica. Cornell University Press, Ithaca, New
York, USA.

Templeton, A. R., E. Routman, and C. A. Phillips. 1995.
Separating population structure from population histo-
ry: a cladistic analysis of the geographical distribution
of mitochondrial DNA haplotypes in tiger salamander,
Amhystoma trigrinum. Genetics 140:767-782.

Thomas, A. L. R. 1996. The Bight of birds that have wings
and a tail: variable geometry expands the envelope of
flight performance. Journal of Theoretical Biology
183:237-245.

Van Oosterhout, C., W. F. Hutchinson, D. P. M. Wills,
AND P. Shipley. 2004. Micro-Checker: software for
identifying and correcting genotyping errors in micro-
satellite data. Molecular Ecology Notes 4:535-538.

Weir, B. S. and C. C. Cockerham. 1984. Estimating F-
statistics for the analysis of population structure.
Evolution 38:1358-1370.

Weir, J. T. 2006. Divergent timing and patterns of species
accumulation in lowland and highland neotropical
birds. Evolution 60:845-855.

Weir, J. T., E. Birmingham, M. J. Miller, J. Klicka, and
M. A. Gonzalez. 2008. Phylogeography of a
morphologically diverse neotropical montane species,
the Common Bush-Tanager (Chlorospingus ophthal-
micus). Molecular Phylogenetics and Evolution
47:650-664.

Yeh, P. j. 2004. Rapid evolution of a sexually selected trait
following population establishment in a novel habitat.
Evolution 58: 166-174.

Yuri, T. and D. P. Mindell. 2002. Molecular phylogenetic
analysis of Fringillidae, New World nine-primaried
oscines. (Aves: Passeriformes). Molecular Phyloge-
netics and Evolution 23:229-243.

ZwARTJES, P. W. 2003. Genetic variability in migratory and
endemic island songbirds (genus Vireo): a comparative
assessment using molecular and morphological traits.
Conservation Genetics 4:749-758.

The Wilson Journal of Orniihulogy 122(2):288-295, 2010

LONG-TERM CHANGES IN AVIAN COMMUNITY STRUCTURE IN A
SUCCESSIONAL, EORESTED, AND MANAGED PLOT IN A
REEORESTING LANDSCAPE

ELIZABETH W. BROOKS' AND DAVID N. BONTER' --'

ABSTRACT. — We examined a 35-year transition in the breeding bird community at a successional study site in a
reforesting landscape in southwestern New York, USA. Changes in the successional plot were compared with those in two
additional census plots, one in undisturbed forest and the other in a managed tree farm. The territories of 7,429 singing male
songbirds were mapped on the census plots. The most dramatic changes in community structure were in the successional
plot where total number of territories declined between the beginning of the study (.v = 95.8 territories, 1969-1973) and end
of the study (x = 57.2 territories, 1999-2003); grassland/shrub nesting species were nearly extirpated, and the number of
neotropical migrant territories increased from zero in 1969 to 30 in 2003. The average number of neotropical migrant
territories in the undisturbed forest plot declined from the beginning of the study (.v = 54.0, 1975-1979) to the end (.v =
44.8, 2003-2007). The average number of territories increased in the managed tree farm from the beginning of the study (.v
= 53.6, 1983-1987) to the end (.v = 104.0, 2003-2007) largely due to increases in abundance of temperate zone migrants
and resident species. Counts of individual species in the census plots were not highly correlated with counts from regional
Breeding Bird Survey routes. Received 28 March 2009. Accepted 20 October 2009.

Long-term studies are required to better under-
stand the effects of anthropogenic change on
natural ecosystems (Collins 2001), and a need for
detailed long-term studies in ornithology has been
recognized (Wiens 1984). Relatively few studies
follow populations for more than two decades;
many long-term studies have not followed the
same protocol every year (Hall 1984, Leek et al.
1988, McNulty et al. 2008), or have only sampled
avian communities at end points of a long time
period (Ambuel and Temple 1982, Kirk et al.
1997, Haney et al. 2008). Studies examining
changes in community structure following anthro-
pogenic or natural change have tended to focus on
succession following timber harvests or forest
fires (Hagan et al. 1997, Hobson and Schieck
1999, Venier and Pearce 2005).

Recent attention has focused on the inOuence of
forest fragmentation on bird communities (Ste-
phens et al. 2004), but many areas are undergoing
large-scale regeneration rather than forest frag-
mentation. Little attention has been directed at
avian community structure following agricultural
abandonment and secondary succession (but see
Johnston and Odum 1956), a widespread process in
eastern North America. Many areas in New
England support more forested land today than at
any point in the last two centuries as marginal

' Braddock Bay Bird Ob.servatory, P. O. Box 12876,
Rochester, NY 14612, USA.

^Cornell Laboratory of Ornithology, 159 Sapsucker
Woods Road. Ithaca. NY 14850, USA.

’Corresponding author; e-mail: dnb23@cornell.edu

agricultural lands are abandoned and allowed to
undergo secondary succession, resulting in pro-
found changes in bird communities (Hagan 1993).
Grassland bird declines have been attributed, in
part, to abandonment and reforestation of farmland
(Norment 2002); the grassland area in New York
and New England has declined by 60% since the
1930s (Vickery et al. 1994). Secondary succession
and maturation of forests have been cited as
mechanisms driving long-tenn changes in popula-
tions of shaib specialists like the Eastern Towhee
(Pipilo erythrophthalnms) (Hagan 1993) and spe-
cies that prefer eaiiy-successional forest like the
Least Flycatcher {Etnpichnax minimus) (Holmes
and Sherry 1988). Additional studies of avian
community changes during and following agricul-
tural abandonment are warranted, and would
benefit from long-term studies at focal sites.

Intensive, long-term field studies can also be
useful for assessing and validating the utility of
larger-scale monitoring data. For instance, results
from intensive studies have been compared with
regional trends detected by the Breeding Bird
Survey (BBS) (Holmes and Sherry 2001, Ballard
et al. 2003, McNulty et al. 2008). These local,
intensive studies may not be subject to many of
the limitations often associated with BBS data
such as road-bias or changes in observers (Kendall
et al. 1996, Keller and Scallan 1999, Niemuth et
al. 2007). We can have greater confidence in the
results if trend estimates generated from different
methodologies agree.

We document a 35-year transition in the
breeding bird community at a study site that has

288

Brooks cold Bonier • LONG-TERM COMMUNI'I'Y CHANC3HS

289

undergone seeondary succession in a region
experiencing reforestation. In addition, we com-
pare trends on the successional plot with addi-
tional long-term censuses of a forested plot and on
an actively managed tree farm. Our objectives
were to; (1) quantify long-term changes in
breeding bird communities on intensively studied
plots that experienced different management
regimes, and (2) compare changes in the avian
communities on the census plots with regional
trends measured by the Breeding Bird Survey.

METHODS

Intensive Census Plots. — Breeding bird temto-
ries were mapped on three long-term census plots.
Territories were defined by spot mapping (Sven-
son et al. 1970), a reliable method of estimating
populations of breeding birds (James and Warner
1982). At least eight visits were made to each plot
at approximately weekly intervals during the
breeding season from late May to late July
annually. All observations were recorded by a
single observer (EWB) throughout the course of
the study. The census plots were searched during
each visit on foot for ~ 1 .5 hrs, and temtories of
singing males were mapped. All partial territories
(territories extending beyond the boundaries of
the study plots) were considered to be whole
territories in the analysis.

A 9.3-ha “successional” plot transitioned from
an open agricultural field to secondary growth
forest during the study (35 years, 1969-2003).
The “forest” plot was a 16.6-ha area in mixed
deciduous-coniferous forest that was relatively
undisturbed during the course of the study
(33 years, 1975-2007). The “plantation” plot
included 10.7 ha in an actively managed Christ-
mas tree farm; the habitat was maintained at a
relatively early successional stage in the planta-
tion plot, although vegetation height increased
over the course of the study (25 years, 1983-
2007). All three plots were in Allegany County
(42° 14' N, 77° 49' W) in southwestern New
York State.

BBS Trends. — ^The Breeding Bird Survey (BBS)
is an annual survey of bird diversity and
abundance conducted along hundreds ol roadside
routes in North America during the breeding
season each year. A route is composed of a series
of 50 stops distributed across 39.4 km. A skilled
volunteer observer records all individual birds that
are heard or observed at each stop during a 3-min
point count (Robbins et al. 1986). Data from the

two BBS survey routes nearest to the census plots
were analyzed for this study (routes 61045 and
61046). These BBS routes were surveyed by the
same observer as the census plots (EWB) with the
observer being the primary recorder on these
routes beginning in 1991 (route 61045) and 1973
(route 61046).

Data Analysis. — ^Population trends for each
species on each census plot were calculated with
Poisson regression using Proc GENMOD in SAS
(SAS Institute 2003). Poisson models adequately
fit the data in 137 of 141 runs (using the
Deviance Chi-square method; models not fitting
the data showed signs of overdispersion). Species
were included in the analysis if they were
detected in at least 5 years within a plot. Data
from Tree Swallows (Tachycineta hicolor) and
Eastern Bluebirds (Sialia sialis) were eliminated
from analysis as the presence of these species
was dependent upon the availability of artificial
nesting boxes, and nest boxes were not consis-
tently provided throughout the course of the
study.

Birds were grouped according to life history
characteristics (Table 1) and species were classi-
fied by migration strategy (neotropical migrant vs.
temperate migrant/resident) and by preferred
breeding habitat (woodland vs. grassland/shrub).
Classifications were based on life history infor-
mation available through the Patuxent Wildlife
Research Center (Gough et al. 1998). Species
richness was calculated within years as the total
number of species detected on the plot.

Counts for BBS routes 61045 and 61046 were
downloaded from the BBS web site (Sauer et al.
2008). We tested for correlations between BBS
counts and census counts for all years with census
data except for 2006 when the BBS was not
conducted on route 61046. CoiTclations coeffi-
cients (/•) among counts over time for all census
plot/BBS route combinations were calculated
using Proc CORR in SAS.

RESULTS

Widespread changes in the avian community
were detected on the successional plot where
number of total territories declined (;(■ I = 138.8,
P < O.OOl) and grassland-nesting species essen-
tially disappeared. Numerous nesting species were
lost from the plot within the first 10 years
including Clay-colored Sparrow (Spizella pallida).
Vesper Sparrow {Pooecetes gramineus). Savannah
Sparrow {Passerculus sandwichensis), Henslow’s

290

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2. June 2010

TABLE 1. Species, life-history classifications, number of territories recorded, and trend estimates for breeding birds
recorded on long-term studies conducted in successional, forest, and tree farm (plantation) census plots.

Species

Breeding

habitat"

Migration

status^

Successional

n (trend)""

Forest

n (trend)

Plantation

n (trend)

Ruffed Grouse

Bonasa umhellus

w

T

19(+)

6

Wild Turkey

Meleagris gallopavo

w

T

11

Sharp-shinned Hawk

Accipiter striatus

w

T

5

Cooper’s Hawk

A. cooperii

w

T

5

Red-shouldered Hawk

Biiteo lineatus

w

T

10 (+)

Broad-winged Hawk

B. platypterus

w

N

24

Mourning Dove

Zenaida macroura

s

T

18 (+)

6

23 (+)

Black-billed Cuckoo

Coccyzus erythopthalmus

w

N

10 (-)

Barred Owl

Strix varia

w

T

12

Yellow-bellied Sapsucker

Sphyrapicus varius

w

T

9 (+)

Downy Woodpecker

Picoides puhescens

w

T

16

Hairy Woodpecker

P. villas us

w

T

17

Pileated Woodpecker

Dryocopus pileatus

w

T

7

Alder Flycatcher

Empidonax alnorum

s

N

41

Great Crested Flycatcher

Myiarchus crinitus

w

N

10 (+)

Eastern Kingbird

Tyrannus tyrannus

s

N

9

13

Blue-headed Vireo

Vireo solitarius

w

N

11 (+)

81 (+)

Red-eyed Vireo

V. olivaceus

w

N

8

Blue Jay

Cyanocitta cristata

s

T

33 (+)

91

7 (+)

American Crow

Con’us hrachyrhynchos

N/A

T

19(+)

Horned Lark

Eremophila alpestris

s

T

8

Black-capped Chickadee

Poecile atricapillus

w

T

32 (+)

112

Red-breasted Nuthatch

Sitta canadensis

w

T

8 (+)

94

White-breasted Nuthatch

S. carolinensis

w

T

11 (-)

Brown Creeper

Certhia americana

w

T

68

House Wren

Troglodytes aedon

s

N

25

Winter Wren

T. troglodytes

w

T

17 (+)

Golden-crowned Kinglet

Regulus satrapa

w

T

17 (+)

209

Veery

Catharus fuscescens

w

N

22 (-)

Hermit Thrush

C. giittatus

w

T

18 (+)

Wood Thrush

Hylocichla niustelina

w

N

44(-)

American Robin

Turdiis migratorius

s

T

176 (-)

98

102 (-P)

Gray Catbird

Dumetella carolinensis

s

N

30

9

Brown Thrasher

To.xostoma riifum

s

T

I4(-)

17

Cedar Waxwing

Bomhycilia cedrorum

s

T

102 (-r)

20 (+)

136 (-P)

Blue-winged Warbler

Vermivora pinus

s

N

8 (+)

Nashville Warbler

V. ruficapilla

s

N

26

Chestnut-sided Warbler

Dendroica pensylvanica

s

N

17 (+)

26 (+)

Magnolia Warbler

D. magnolia

w

N

78 (+)

308

15 (+)

Yellow-rumped Warbler

D. corona ta

w

T

105 (-H

128 (-I-)

70 (+)

Black-throated Green Warbler

D. virens

w

N

12

337 (-)

Blackburnian Warbler

D. fusca

w

N

10 (+)

323

Prairie Warbler

D. discolor

s

N

7

24

Ovenbird

Seiurus aurocapilla

w

N

18 (+)

77

Mourning Warbler

Oporornis Philadelphia

s

N

19 (+)

Common Yellowthroat

Geothlypis trichas

s

N

195

82

22

Canada Warbler

Wilsonia canadensis

w

N

8 (-)

Scarlet Tanager

Piranga olivacea

w

N

18 (-)

Eastern Towhee

Pipilo erythrophthalmus

s

T

82 (-)

29

Chipping Sparrow

Spizella passerina

s

T

375 (-)

69

430 (-1-)

Field Sparrow

S. pusilla

s

T

129 (-)

102 (-)

Vesper Sparrow

Pooecetes gramineus

s

T

15

Savannah Sparrow

Passerculus sandwichensis

s

T

45 (-)

Grasshopper Sparrow

Ammodranius savannarum

s

N

26 (-)

Song Sparrow

Melo.spiza melodia

s

T

414(-)

39

378 {+)

Brooks and Bonier • LONG-TERM COMMUNITY CHANGES

291

TABLE 1. Continued.

Species

Breeding

liabilaf

Migration

status'’

Successional
n (trend)’’

Forest

n (trend)

Plantation

n (trend)

White-throated Sparrow

Zonotrichia alhicollis

s

T

18 (+)

17

Dark-eyed Junco

Junco hyenialis

w

T

40 (+)

246

Northern Cardinal

Cardinalis cardinalis

s

T

6

9 (-)

Rose-breasted Grosbeak

Pheucticiis ludovicianiis

w

N

9 (-)

Indigo Bunting

Passerine! cyanea

s

N

60 (+)

51

22

Bobolink

Dolichonyx oryzivorus

s

N

31

Red-winged Blackbird

Agelaius phoeniceus

s

T

30

Brown-headed Cowbird

Molothriis ater

s

T

52 (-)

34

19 (+)

Purple Finch

Carpodacus purpureus

w

T

114 (-)

56 (+)

83 (-I-)

House Finch

C. mexicanus

N/A

T

7

American Goldfinch

Spinus tristis

s

T

6

50 (+)

“ Breeding habitat and migration classifications are from Gough et al. (1998), W = woodland nesting species, S = successional species including grassland and
shrubland nesting birds.

T = temperate, including permanent residents and short-distance migrant species. N = neotropical migrants.

Trend sign indicates significant iP < 0.05) population increase (-I-) or decrease (-) over time, and period (.) = insufficient data. No sign indicates the lack of a
significant directional trend.

Sparrow (Ammodramiis henslowii). Bobolink
(Dolichonyx oiyzivorus). Red-winged Blackbird
{Agelaius phoeniceiis), and Common Crackle
(Qiiiscaliis quiscula). The number of territories
of neotropical migrants (/^i = 50.5, P < 0.001)
and woodland-nesting species (x^i = 55.1, P <
0.001) increased despite the decline in total
territories on the successional plot, (Fig. lA).
Positive trends were detected for 17 species (49%,
P < 0.05, Table 1) and declines were detected for
nine species (26%). Total species richness varied
considerably but increased over time (/^i = 23.2,
P < 0.001, Fig. 2).

The avian community on the forest plot
remained relatively stable with significant de-
clines detected for eight species (17%, Table 1)
and increa.ses detected for 12 species (25%). No
trends were detected in total territories {x^\ =0.1,
P = 0.712) or in species richness (x^\ = 0.6, P =
0.430) over time. The number of territories of
neotropical migrants declined (/^i = 9.2, P =
0.003) while territories of shorter-distance mi-
grants and resident species increa.sed (x^i = 12.6,
P < 0.001, Fig. IB).

The total number of territories increased on the
managed tree farm {x'^\ = 101.9, P < 0.001),
where the number of breeding pairs significantly
increased for 11 species (42%, Table 1) and
declined for only three species (12%). Increases in
number of territories were primarily among
resident and temperate-zone migrants {x^\ —
52.1, P < 0.001) with grassland/shrub nesting
species (x^i = 20.8, P < 0.001) and woodland

nesting species (x^i = 55.8, P < 0.001) becoming
more common over time (Fig 1C).

Counts of species on local BBS routes were not
highly correlated with counts of breeding territo-
ries for each species on the census plots.
Comparing counts on BBS route 61045 with the
census plots, the average con-elation coefficients
were greatest for the successional plot {x corre-
lation, r = 0.25), followed by the managed tree
farm {x = 0. 14) and forest plot {x — 0.07). Counts
of individual species on BBS route 61046 were
also best correlated with counts on the succes-
sional plot (.V = 0.23), followed by the forest plot
(T = 0.16) and plantation plot (.v = —0.06). On
average, the counts of individual species on the
two BBS routes were not highly correlated (.v =
0.17).

DISCUSSION

We documented profound changes in commu-
nity structure and abundance of birds in an area
undergoing secondary succession following agri-
cultural abandonment, while a nearby, undis-
turbed forest bird community remained relatively
static. The importance of disturbance and succes-
sion in structuring avian communities has long
been appreciated despite the paucity of long-term
studies documenting the dynamic changes in
communities over time (reviewed by Brawn et
al. 2001). Some form of disturbance is required to
meet the habitat needs of many species, and the
loss of grassland and shrub habitats in North
America has contributed to great declines in birds

292

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 2, June 2010

B. Forest plot

■ Neotropical migrants

■ Woodland

• Temperate/Resident

• Shrub/Grassland

FIG. I. Total territories (// = 7,429) mapped on intensive census plots over 35 years, where temperate migrants and
resident species that nest in early successional habitats demonstrated the strongest trends, declining in the successional
landscape (A), remaining relatively stable in an undisturbed forest plot (B), and becoming more common in the managed
tree farm (plantation, C).

that specialize in these environments (Askins
2000, Brawn et al. 2001),

Clearing of forests in eastern North America
created early-successional habitats that allowed
for geographic spread and numerical increases of
species adapted to these environments. Recent

changes in land use patterns, however, have led to
a dramatic shift in the structure of bird commu-
nities. The area of cultivated cropland east of the
Mississippi River declined by ~22% between
1950 and 2000 (Brown el al. 2005), largely due to
the loss of small farms. These declines in small-

Brooks and Bonier • LONG-TFRIVI COMMUNITY CHANGHS

293

* Successional • Forest o Plantation

FIG. 2. Changes in species richness on census plots in
successional. forested, and managed tree farm (plantation)
habitats.

scale agriculture contributed to similar declines in
populations and distributions of early-succession-
al birds, leading to local and regional extirpation
of some species (Hunter et al. 2001).

Johnston and Odum (1956) documented chang-
es in the richness of species nesting in the
Piedmont of Georgia on plots of various succes-
sional stages with increasing richness recorded
with increasing age of the vegetation. We also
found an increase in species richness during the
early stages of secondary succession. However, in
contrast to the earlier study, density of nesting
birds decreased significantly as our successional
study site transitioned from grassland to forest
over a 35-year period. The rapid decline in
grassland- and shrub-nesting species was not
matched by a similarly rapid increase in wood-
land-nesting birds, many of which are neotropical
migrants.

This lack of replacement by woodland-nesting
species may be a function of a general decline in
the abundance of neotropical migrants in the
region (Robbins et al. 1989). Holmes and Sherry
(2001), in one of the few intensive, long-term
studies of breeding bird communities, recorded a
decline in the abundance of breeding birds in a 30-
year study in relatively undisturbed, contiguous
forest in New Hampshire. Declines were most
pronounced for neotropical migrants. Other stud-
ies have reported long-term declines in popula-
tions of long-distance migrants at sites in West
Virginia (Hall 1984), New Jersey (Leek et al.
1988), and Sweden (Enemar et al. 1994). In

contrast, Askins and Philbrick (1987) detected a
decline and subsequent rebound in abundance ol
many neotropical migrants during a 32-year study
at an isolated 23-ha woodlot in Connecticut. The
amount of forest in the stirrounding landscape
during the Connecticut study initially declined
and then increased, suggesting a link between
abundance of long-distance migrants and regional
forest area.

A general decline in abundance of long-
distance migrants could explain the relatively
shallow increase in neotropical migrant territories
on our successional study plot as the habitat
became more suitable for many of these wood-
land-breeding birds. As in the earlier studies, the
abundance of long-distance migrants declined
over time on the undisturbed forest plot censused
in our study, suggesting that local populations of
neotropical migrants may be declining.

Quantifying changes in the abundance and
distribution of birds remains one of the major
methodological challenges in avian ecology and
conservation biology. Most population and trend
estimates in North America are based on Breeding
Bird Survey data, yet concerns over the adequacy
of the BBS continue to be raised (Keller and
Scallan 1999, Francis et al. 2005) and addressed
through novel analytical techniques (e.g., Royle et
al. 2005). Confidence in trend estimates generated
from BBS data can be strengthened if these trends
agree with data from other monitoring schemes.
For instance, McNulty et al. (2008) compared
long-term trends in an old-growth forest plot with
regional BBS trends in a relatively static land-
scape (the Adirondack Mountains of New York)
and found agreement in the direction of the trend
for 74% of the species monitored. Our compar-
ison of bird counts on local BBS routes with
counts on intensive census plots failed to show
high agreement among the surveys. The greatest
agreement, however, was between the BBS routes
and the successional plot, a logical result in a
region experiencing the abandonment of small-
scale agriculture.

Analyses comparing results of long-term,
intensive research with broad-scale monitoring
programs such as the BBS should provide a more
complete picture of the status of and temporal
changes in bird populations. Intensive studies may
be the only way to monitor trends in abundance
and distribution of .species that are hard to detect,
or species that occur in low densities. For
example. Clay-colored Sparrows were not detect-

294

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

ed on either BBS route in the current study,
although they were recorded nesting within the
successional and managed tree farm plots.

Long-term, intensive studies can provide
unique perspectives on ecological phenomena
that are unattainable in most contemporary studies
that occur over restricted temporal scales. More
long-term research projects should be encouraged
(Wiens 1984). Further, fluctuations in populations
detected in this study and elsewhere (Holmes and
Sherry 2001) highlight the value and necessity of
conducting continuous, long-term studies when
studying population trends.

ACKNOWLEDGMENTS

We especially thank the late Clarence Klingensmith for
help in setting up the study plots, conducting the habitat
surveys in 1969 and 1998, and for advice and encourage-
ment through the years. We thank Cynthia Clements,
Phillips Foster, Tom and Kathy Kent, and the late Alice
Foster and Eddy Foster for permission to conduct the study
on their land. We appreciate the assistance of Gaylord
Rough, Dennis Smith, Rick Walker, and the late Robert
Place. We thank Doris and the late Lou Burton, Peter
Gradoni, and William Howe for a.ssistance in the field.
Research was supported by a grant from the Alfred
University Research Foundation during the 1970 field
season. We thank C. E. Braun and two anonymous
reviewers for comments that improved the manuscript.

LITERATURE CITED

Ambuel, B. and S. A. Temple. 1982. Songbird populations
in southern Wisconsin forests- 1954 and 1979. Journal
of Field Ornithology 53; 149-158.

Askins, R. A. 2000. Restoring North America’s birds:
lessons from landscape ecology. Yale University
Press, New Haven, Connecticut, USA.

ASKIN.S, R. A. AND M. .1. Philbrick. 1987. Effect of
changes in regional forest abundance on the decline
and recovery of a forest bird community. Wilson
Bulletin 99: 7-21.

Ballard, G., G. R. Geupel, N. NtJR, and T. Gardali.
2003. Long-term declines and decadal patterns in
population trends of songbirds in western North
America, 1979-1999. Condor 105: 737-755.

Brawn, J. D., S. K. Robin.son, and F. R. Thomp.son III.
2001. The role of disturbance in the ecology and
conservation of birds. Annual Review of Ecology,
Evolution, and Systematics 32: 251-276.

Brown, D. G., K. M. John.son, T. R. Loveland, and D.
M. Tmeobald. 2005. Rural land-use trends in the
conterminous United States. 1950-2000. Ecological
Applications 15: 1851-1863.

Collins, S. L. 2001. Long-term re.search and the dynamics
of bird populations and communities. Auk I 18; 583-
588.

Enemar, a., B. Cavai.lin, E. Nyholm, I. Rudebeck, and
A. M. Thorner. 1994. Dynamics of a passerine bird

community in a small deciduous wood, S. Sweden,
during 40 years. Ornis Svecica 4; 65-104.

Francis, C. M., J. Bart, E. H. Dunn, K. P. Burnham, and
C. J. Ralph. 2005. Enhancing the value of the
Breeding Bird Survey: reply to Sauer et al. (2005).
Journal of Wildlife Management 69: 1327-1332.

Gough, G. A., J. R. Sauer, and M. Iliff. 1998. Patuxent
bird identification infocenter. Version 97.1. USGS,
Patuxent Wildlife Research Center, Laurel, Maryland,
USA. www.mbr-pwrc.usgs.gov/id/framlst/infocenter.
html (accessed 1 March 2009).

Hagan, J. M. 1993. Decline of the Rufous-sided Towhee in
the eastern United States. Auk 1 10: 863-874.

Hagan, J. M., P. S. McKinley, A. L. Meehan, and S. L.
Grove. 1997. Diversity and abundance of landbirds in
a northeastern industrial forest. Journal of Wildlife
Management 61: 718-735.

Hall, G. A. 1984. Population decline of neotropical
migrants in an Appalachian forest. American Birds
38: 14-18.

Haney, A., S. Apfelbaum, and J. M. Burris. 2008. Thirty
years of post-fire succession in a southern boreal forest
bird community. American Midland Naturalist 159:
421-433.

Hobson, K. A. and J. Schieck. 1999. Changes in bird
communities in boreal mixedwood forest; harvest and
wildfire effects over 30 years. Ecological Applications
9: 849-863.

Holmes, R. T. and T. W. Sherry. 1988. A.ssessing
population trends of New Hampshire forest birds: local
vs. regional patterns. Auk 105: 756-768.

Holmes, R. T. and T. W. Sherry. 2001. Thirty-year bird
population trends in an unfragmented temperate
deciduous forest; importance of habitat change. Auk
118: 589-609.

Hunter, W. C., D. A. Buehler, R. A. Canterbury, J. L.
Confer, and P. B. Hamel. 2001. Conservation of
disturbance-dependent birds in eastern North America.
Wildlife Society Bulletin 29; 440—455.

James, F. C. and N. O. Wamer. 1982. Relationships
between temperate forest bird communities and
vegetation structure. Ecology 63: 159-171.

Johnston, D. W. and E. P. Odum. 1956. Breeding bird
populations in relation to plant succession on the
Piedmont of Georgia. Ecology 37: 50-62.

Keller, C. M. E. and J. T. Scallan. 1999. Potential
roadside biases due to habitat changes along Breeding
Bird Survey routes. Condor 101: 50-57.

Kendall, W. L., B. G. Peterjohn, and .1. R. Sauer. 1996.
First-time observer effects in the North American
Breeding Bird Survey. Auk 1 13; 823-829.

Kirk, D. A., A. W. Diamond, A. R. Smith, G. E. Holland,
AND P. Chytyk. 1997. Population changes in boreal
forest birds in Saskatchewan and Manitoba. Wilson
Bulletin 109: 1-27.

Leck, C. F., B. G. Murray, and J. Swinebroad. 1988.
Long-term changes in the breeding bird populations of
a New-Jersey forest. Biological Con.servation 46: 145-
157.

McNulty, S. A., S. Droege, and R. D. Masters. 2008.
Long-term trends in breeding birds in an old-growth

Brooks and Bonter • LONG-TERM COMMUNITY CHANGES

295

Adirondack forest and the surrounding region. Wilson
Journal of Ornithology 120: 153-158.

Niemuth, N. D., a. L. Dahl, M. E. Estey, and C. R.
Loesch. 2007. Representation of landcover along
Breeding Bird Survey routes in the northern plains.
Journal of Wildlife Management 71: 2258-2265.
Norment, C. 2002. On grassland bird conservation in the
Northeast. Auk 119: 271-279.

Robbins, C. S., D. Bystrak, and P. H. Geissler. 1986.
The Breeding Bird Survey: its first fifteen years, 1965-
1979. Resource Publication 157. USDl, Fish and
Wildlife Service, Washington, D.C., USA.

Robbins, C. S., J. R. Sauer, R. S. Greenberg, and S.
Droege. 1989. Population declines in North- American
birds that migrate to the Neotropics. Proceedings of
the National Academy of Sciences of the USA. 86:
7658-7662.

Royle, J. A., J. D. Nichols, and M. Kery. 2005.
Modelling occurrence and abundance of species when
detection is imperfect. Oikos 1 10:353-359.

SAS Institute. 2003. SAS Version 9.1. Cary, North
Carolina, USA.

Sauer, J. R., J. E. Hines, and J. Fallon. 2008. The North
American Breeding Bird Survey, results and analysis
I966-2(X)7, Version 5.15. USGS, Wildlife Research
Center, Laurel, Maryland, USA. www.mbr-pwrc.usgs.
gov/bbs/bbs.html (accessed 10 February 2009).

Stephens, S. E., D. N. Koons, J. J. Rotella, and D. W.
Willey. 2(K)4. Effects of habitat fragmentation on avian
nesting success: a review of the evidence at multiple
spatial scales. Biological Conservation 115: 101-110.

SvENSSON, S., K. Williamson, C. Ferry, A. H. Joensen, D.
Lea, H. Oelke, and C. S. Robbins. 1970. An
international standard for a mapping method in bird
census work recommended by the International Bird
Census Committee. Audubon Field Notes 24: 722-726.

Venier, L. A. and j. L. Pearce. 2005. Boreal bird
community response to jack pine forest succession.
Forest Ecology and Management 217: 19-36.

Vickery, P. D., A. L. Hunter, and S. M. Melvin. 1994.
Effects of habitat area on the distribution of grassland
birds in Maine. Conservation Biology 8: 1087-1097.

Wiens, J. A. 1984. The place of long-term studies in
ornithology. Auk 101: 202-203.

The Wilson Joiinuil of Ornithology 122(2):296-306, 2010

SCALE-DEPENDENT RESPONSE BY BREEDING SONGBIRDS TO
RESIDENTIAL DEVELOPMENT ALONG LAKE SUPERIOR

MICHELLE T. LORD' AND DAVID J. LLASPOHLER'

ABSTRACT. — We examined the influence of shoreline residential development on breeding bird communities along
forested portions ot Lake Superior and hypothesized that anthropogenic changes related to housing development would
alter bird community structure compared to areas without human development. We used point counts to compare relative
abundance of bird species in relation to residential development at coarse (along 1 km shoreline stretches with and without
housing/cottage development) and fine (developed and undeveloped sides of a shoreline access road) spatial scales during
the 2005 breeding season. More species had development related differences in abundance at the finer-scale analysis than at
the coarse .scale. American Crows (Corvits hrachyrhynchos) and American Robins (Titrclus migratorius) were more
abundant on the developed, shoreline side of shoreline access roads. Red-breasted Nuthatches (Sitta canadensis). Black-
throated Green Warblers (Dendroica virens), and Red-eyed Vireos {Vireo olivaceus) were more abundant on the
undeveloped, inland side of shoreline access roads. Several species were detected exclusively in developed or undeveloped
forest areas. The pattern of development-related differences in relative abundance of bird species depended on the scale at
which data were analyzed, suggesting that many species may respond to habitat differences within the 100 m scale quite
distinct from how they respond to differences at the scale of thousands of meters. Received 27 August 2009. Accepted 23
December 2009.

Rural population growth in many parts of the
United States has increased in the last 15 years
(Long and Nucci 1997, Gustafson et al. 2005).
Growth in parts of the northern Great Lakes
Region has been concentrated around inland lakes
(Radeloff et al. 2001, Gonzalez- Abraham et al.
2006). Two-thirds of previously undeveloped
inland lakes in northern Wisconsin (i.e., lakes
with no residential housing) have become devel-
oped with homes and cottages near the shoreline
since 1965 (Lindsay et al. 2002, Elias and Meyer
2003). Housing development alters shoreline
habitat through changes in habitat structure and
plant species composition, which can create
discontinuities in plant communities resulting in
fragmentation (Brown 2003). These changes have
the potential to influence many taxa including
forest birds (Vale and Vale 1976, Mills et al.
1989, Theobald et al. 1997, Swenson and Franklin
2000).

Many recent studies focusing on the ecological
impacts of nearshore residential development
have examined inland lakes with less attention
on Great Lakes shorelines (Rottenborn 1999,
Lindsay et al. 2002, Elias and Meyer 2003).
Perimeters of both inland lakes and the Great
Lakes provide an interface between aquatic and

' School of Forest Resources and Environmental Science,
Michigan Technological University, Houghton, Ml 49931.
USA.

^Current address: 73 Training Hill, Middletown. CT
06457. USA.

'Corresponding author; e-mail: michelle.ford@gza.com

upland ecosystems but, because of their size and
associated climatological influence, the abiotic
and biotic environment along Great Lakes shore-
lines often differs dramatically from inland lake
riparian zones or adjoining upland habitats
(Eichenlaub 1979, Albert et al. 1986). Thus,
results from studies of inland lakes cannot be
confidently extended to larger lake systems.

North America’s Great Lakes (Huron, Ontario,
Michigan, Erie, and Superior) contain 20% of the
world’s and 95% of North America’s surface
fresh water. The waters of the Great Lakes support
the world’s largest freshwater commercial fishery,
supply drinking water to millions of citizens in the
United States and Canada, and support a multi-
million dollar recreation industry (Great Lakes
Information Network 2005). The amenity values
associated with the 17,542 km of Great Lakes
shoreline have likewise attracted residential
development (Schnaiberg et al. 2002). Shoreline
residential development on the Great Lakes has
accelerated in recent years and now represents one
of the fastest growing segments of rural housing
expansion in the region (Orr 1997).

The Keweenaw Peninsula is in the western
Upper Peninsula of Michigan and is surrounded
on three sides by Lake Superior (Fig. 1). Kewee-
naw County is one of the least populated in
Michigan and much of the shoreline remains
undeveloped; however, residential development in
this area has rapidly increa.sed over the past
decade with population shifts to rural areas (Orr
1997). The Upper Peninsula has not yet felt the

296

Ford am! Flaspohler • BREEDING SONGIBRDS AND DEVELOPMENT

297

effects of rural in-migralion as greatly as other
scenic places in the United States, but land use
changes and population increases are evident (Orr
1997).

We focused on differences between residential-
ly developed and undeveloped shoreline areas of
Lake Superior and measured vegetation habitat
features associated with development known to
influence bird presence and relative abundance.
To our knowledge, this is the first study to
examine the influence of Great Lakes shoreline
residential development on forest breeding bird
communities on any of the Great Lakes. Our
objectives were to; ( 1 ) evaluate differences in
forest vegetation between developed and unde-
veloped areas, (2) examine if relative abundance
of breeding bird species differs between residen-
tially developed and undeveloped shoreline forest
along Lake Superior, and (3) assess the spatial
scale at which development influences bird
presence and abundance.

METHODS

Study Sites. — The study was conducted in two
residentially developed and three undeveloped
shoreline areas along the eastern shore of Lake
Superior in the Keweenaw Peninsula in the
northern-most part of the Michigan’s Upper
Peninsula (Fig. 1). Only east and southeast facing
areas were used to minimize variability in soils,
bedrock geology, site aspect, wind, and forest
type. Developed areas included Hermit’s Cove
(47° 15' N, 88° 07' W) and Rabbit Bay (47° 04' N,
88° 20' W) in Keweenaw and Houghton counties,
respectively. These areas are similarly developed
and are comprised predominately of seasonal
cottages which are bounded on one side by Lake
Superior and on the other by closed-canopy
mixed hardwood and boreal transition forest with
a gravel road along the forest side of the
properties. Cottages at both sites are only on
the lake side of the road and have a mixture of
old (built prior to 1950) and newer homes, as
well as seasonal and permanent residents. Most
cottages are occupied primarily during the
summer months (Manarolla 2005).

Low-elevation and near-shore portions of the
Keweenaw Peninsula are dominated by boreal
transition forests composed primarily of paper
birch (Betiila papyrifera), balsam fir (Abies
halsamea), red maple (Acer riihrum), white spruce
(Picea glaiica), northern white cedar (Thuja
occidentalis), and yellow birch (Betula alleglia-

nieusis). Average precipitation for Keweenaw
County during the breeding sea, son from 1 May
2005 through 31 July 2005 was ~22.8 cm with
a mean temperature of 14° C (Weather Under-
ground Corporation 2005). The soils of the eastern
shore of the Keweenaw Peninsula range from
gravelly and rocky sands to sandy loams and are
poorly drained in lower elevations near the shore
(Albert et al. 1986).

Undeveloped study areas. Smith Fisheries (47°
23' N, 87° 53' W), south Rabbit Bay (47° 02' N,
88° 21' W), and north Hermit’s Cove (47° 15' N,
88° 06' W) were .selected by visiting areas along
the eastern shore in early spring and by using
Geographic Information System (CIS) data and
aerial photographs to identify areas with and
without concentrated residential development
near the shore. Undeveloped study areas were
along sections of shoreline north of Hermit’s
Cove and along sections of shoreline south of
Rabbit Bay, and were placed ^300 m from the
nearest residential stmcture. The Smith Fisheries
study area was —48 km north of Hermit’s Cove,
just north of the town of Lac La Belle, and was
only accessible via logging roads and trails
(Fig. 1). Thirty-six plots were established for
vegetation and bird community sampling, 15 in
undeveloped study areas and 21 in developed
study areas. We collected data at two spatial
scales: (1) broad-scale; developed and undevel-
oped shoreline study areas, and (2) fine-scale:
developed and undeveloped sides of the road
accessing developed shoreline study areas.

Measurement of Habitat Variables. — We mea-
sured habitat characteristics for the broad scale
comparison along transects spaced every 100 m in
both developed and undeveloped shoreline study
areas. Vegetation transect centers along undevel-
oped shoreline study areas were 50 m inland.
Transects along developed shoreline study areas
were centered on the center of the unpaved road
(50-75 m from the shore). We considered roads to
be part of tbe vegetation disturbance associated
with development. A second set of transects was
oriented perpendicular to the shore, and extended
50 m in each direction toward and away from the
shoreline for both developed and undeveloped
shoreline study areas. This method sampled
vegetation in both developed residential property
and the undeveloped forest across the road from
the development along the developed shoreline.
Point-count centers for bird surveys were also
placed in the center of the road so that equal

298

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

O'

SmMti Fisheries Study Area

FIG. 1. Study .sites in Houghton and Keweenaw countie.s, Michigan, USA.

Ford and Flaspoldcr • BREEDING SONGBIRDS AND DEVELOPMENT

299

portions of the point-count area were in devel-
oped/shoreline and undeveloped/inland sides of
the road.

Vegetation sampling methods were adapted
from Noon (1981). Canopy cover and ground
cover were estimated using an ocular tube
approximately every 2 m along each transect.
Coarse woody debris (CWD) was defined as any
piece of downed wood >8 cm in diameter that
crossed a transect. The total pieces of CWD that
crossed a transect were recorded and divided by
the length of the transect (100 m). Diameter at
breast height (DBH) and species were recorded
for all living trees >3 cm in diameter within arms
length, —0.8 m, on either side of the transect.
DBH data were converted to total stem count to
compare the number of stems >3 cm in each
transect. We also calculated deciduous and
coniferous shrub/sapling density along each
transect in developed and undeveloped areas.
Each woody stem ^3 cm in diameter at breast
height was recorded as either coniferous or
deciduous. Average density for each shrub type
was calculated as the total number of shrubs
counted divided by the length of the transect.
Total shrub density was calculated as the sum of
both deciduous and coniferous shrubs divided by
transect length.

A gridded density board was used to measure
understory density and calculate percent cover
from 0 to 3 m above the forest floor (Noon 1981).
Separate density readings were taken facing areas
on each side of the road (i.e., facing the residential
development along the shoreline and facing the
undeveloped inland) by standing on the side of the
road with the density board placed 1 1.3 m from
the road edge. Readings were summed and
converted to percent cover in four height catego-
ries, 0-0.3, >0.3-1, >1-2, and >2 m, and total
understory density 0-3 m.

Bird Surveys. — We used 5 min 50-m fixed-
radius point counts to estimate breeding bird
relative abundance (Bibby et al. 2000, Sutherland
et al. 2004) at developed and undeveloped sites.
Counts started at sunrise, —0600 hrs, and ended
by 1030 hrs between 6 June and 1 July 2005.
Point-count centers were established every 200 m
to minimize double counting individuals (Bibby et
al. 2000). Point-count centers in developed
shoreline study areas were in the center of the
road at the center of vegetation transects, 50-75 m
from the shoreline. Point-count centers within
undeveloped shoreline study areas were 50 m

from the shore to minimize wave noise interfer-
ence. Twenty-one point-count stations were
established in developed shoreline study areas
and 15 in undeveloped shoreline study areas; all
counts were done by M. T. Ford.

The observer waited for 2 min prior to start of
each point-count period (Reynolds et al. 1980).
All birds detected by sight or sound within the 50-
m radius within 5 min were recorded. Birds
detected outside the 50-m radius or seen flying
over were also noted, but were not included in
analyses. All stations were visited two times in
2005, once early and once later in the breeding
season with one of the counts conducted within
30 min of sunrise and the other later in the
morning. Point counts were not conducted in rain
or high winds (> 20 km/hr) (Robbins 1981).

Bird species richness in developed and unde-
veloped areas was estimated from the total
number of species detected in 2005. Relative
abundance for each species was defined as the
maximum number of individuals detected be-
tween the two visits at each station. Birds were
also grouped into nesting/foraging guilds to
investigate guild-level associations with develop-
ment.

Statistical Analysis. — ^We used t- and Wil-
coxon-Mann-Whitney non-parametric tests to
examine differences in bird abundance and
vegetation attributes between developed and
undeveloped areas. The more powerful /-test
was used when substantial deviation from sym-
metry was not detected in boxplots generated
from the raw data (Quinn and Keough 2002). The
Wilcoxon-Mann-Whitney test was used at the
same significance levels to examine nesting/
foraging guild differences between developed
and undeveloped areas. Understory density from
0 to 0.3 m and 1 to 2 m appeared to be
asymmetrical and was analyzed using Wilcoxon-
Mann-Whitney tests. The Shannon-Wiener spe-
cies diversity index (//’) was used to compare
species richness and evenness between developed
and undeveloped areas (Magurran 2004). All data
were tested at a significance level of a = 0.05
unless otherwi.se specified.

RESULTS

Broad Spacial Scale Comparison of Habitat
Characteristics in Developed and Undeveloped
Shoreline Study Areas. — ^Developed shoreline had
higher deciduous and total shrub density than
undeveloped areas. Canopy cover and CWD were

300

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. 2. June 2010

TABLE 1. Vegetation characteristics
Peninsula, Michigan.

in developed and undeveloped

shoreline study areas on

the Keweenaw

Characteristics

Treatment

Mean ± SD

p

Canopy cover'

Developed

76.91 ± 22.33

<0.000

Undeveloped

87.82 ± 12.08

Ground cover

Developed

64.68 ± 16.42

0.312

Undeveloped

61.56 ± 19.16

Coarse woody debris"

Developed

8.38 ± 5.75

<0.000

Undeveloped

16.6 ± 10.16

Number of stems > 8 cm"

Developed

357 ± 205.96

<0.000

Undeveloped

288 ± 166.13

Total shrub density"

Developed

6.3 ± 10.62

0.016

Undeveloped

4.38 ± 9.24

Deciduous shrub density"

Developed

9.38 ± 8.25

<0.000

Undeveloped

4.35 ±6.18

Coniferous shrub density

Developed

3.23 ±4.16

0.194

Undeveloped

4.42 ± 6.05

Total understory density (UD)

Developed

50.03 ± 28.25

0.87

Undeveloped

50.9 ± 33.21

UD 0-0.3 m

Developed

78.19 ± 36.31

0.483

Undeveloped

74 ± 34.17

UD > 0.3-1. 0 m

Developed

51.85 ± 37.2

0.473

Undeveloped

47.04 ± 40.85

UD > 1. 0-2.0 m

Developed

27.83 ± 34.07

0.203

Undeveloped

35.63 ± 37.13

UD > 2.0-3.0 m

Developed

42.24 ± 36.7

0.462

Undeveloped

46.93 ± 37.88

Total diameter at brea.st height (DBH)

Developed

17.77 ± 15.87

0.498

Undeveloped

19.51 ± 59.66

DBH of deciduous trees

Developed

16.65 ± 15.18

0.474

Undeveloped

21.41 ± 96.54

DBH of coniferous trees

Developed

18.92 ± 16.47

0.635

Undeveloped

39.4 ± 13.52

■' Significant (a = 0.05).

greater in undeveloped than developed areas.
Developed and undeveloped shoreline study areas
did not differ in mean DBH for all trees together,
deciduous trees, or coniferous trees. Stem counts
were greater in developed than undeveloped
shoreline study areas (Table 1).

Small Spatial Scale Comparison of Developed
Shoreline and Undeveloped Inland Sides of
Shoreline Road. — ^The residentially developed
shoreline side of the road had lower canopy
cover, ground cover, CWD, stem counts, total
understory density, and understory density from
0-0.3 to >0.3-1 m (Table 2). Mean DBH of
coniferous trees, deciduous trees, and all trees
pooled did not differ between developed and
undeveloped sides of the road (Table 2).

Birds. — We detected 291 individuals of 32
species for all point counts pooled within the
50-m radius point counts during the survey period
(Table 3). Species richness was —34% greater in

developed (29 species) shoreline study areas than
undeveloped (18 species) shoreline study areas.
Ninety percent (29 of 32 species) of all species
detected were in developed shoreline study areas;
only 56% (18 of 32 species) of all species
detected were in undeveloped shoreline study
areas. Fourteen species were found only in
developed shoreline study areas and three species
were only in undeveloped shoreline study areas
(Table 3). The Shannon-Wiener diversity index
(//') was 2.96 in developed and 2.45 in undevel-
oped shoreline study areas. Species evenness was
virtually identical in developed {E = 0.88) and
undeveloped (E = 0.86) shoreline study areas.

The Shannon-Wiener diversity index within
developed shoreline study areas was 2.97 on the
developed side of the road and 2.70 on the inland,
undeveloped side of the road. Species evenness
was similar between shoreline (E = 0.92) and
inland (E = 0.87) sides of the road.

FonI ciihI Flcispohkr • BREEDING SONGBIRDS AND DEVELOPMENT

301

TABLE 2. Vegetation characteristics in developed shoreline and undeveloped inland sides ol shoreline
residentially developed areas on the Keweenaw Peninsula, Michigan.

access roads in

Characierislics

Trealnienl

Mean ± .SD

r

Canopy cover'

Shoreline

69.49 ± 23.31

0.002

Inland

84.34 ± 18.79

Ground cover'

Shoreline

68.16 ± 18.9

0.005

Inland

69.49 ± 12.81

Coarse woody debris"

Shoreline

6.71 ± 6.16

0.008

Inland

10.05 ± 4.83

Number of stems > 8 cm"

Shoreline

138 ± 79.53

<0.001

Inland

219.5 ± 126.5

Total shrub density

Shoreline

5.51 ± 6.99

0.16

Inland

7.1 ± 6.66

Deciduous shrub density

Shoreline

7.71 ± 8.07

0.066

Inland

1 1.05 ± 8.19

Coniferous shrub density

Shoreline

3.32 ± 4.79

0.854

Inland

3.15 ± 3.48

Total understory density"

Shoreline

43.86 ± 39.86

0.047

Inland

56.2 ± 40.33

UD 0-0.3 m"

Shoreline

67.88 ± 40.81

0.01

Inland

88.5 ± 28.06

UD > 0.3- 1.0 m"

Shoreline

41.11 ± 35.9

0.008

Inland

62.59 ± 35.73

UD > 1. 0-2.0 m

Shoreline

25.32 ± 32.19

0.508

Inland

30.34 ± 36.08

UD > 2.0-3.0 m

Shoreline

41.12 ± 35.07

0.784

Inland

43.37 ± 38.67

Total diameter at breast height (DBH)

Shoreline

18.75 ± 16.21

0.194

Inland

17.15 ± 15.64

DBH of deciduous trees

Shoreline

18.13 ± 14.13

0.21

Inland

15.97 ± 15.63

DBH of coniferous trees

Shoreline

19.2 ± 17.56

0.77

Inland

18.68 ± 15.55

^ Significant (y — 0.05).

Broad Spatial Scale Comparison of Bird
Species Relative Abundance. — Only Red-eyed
Vireos (Vireo olivaceus), of 32 total species
detected, were significantly (5 times) more
abundant in developed than undeveloped shore-
line study areas. Ground-nesting birds were
equally abundant in developed and undeveloped
shoreline study areas (P > 0.05). Shrub nesters
were more abundant in undeveloped forest and
canopy nesters were more abundant in developed
shoreline study areas when species were pooled
based on nest site location (P < 0.05) (Table 3).

Fine-scale Comparison of Bird Species Relative
Abundance. — ^Two species, American Crow (Cor-
vus brachyrhynchos) and American Robin (Tur-
dus migratorius) were statistically more abundant
on the developed, shoreline side of the shoreline
access roads. Three species, Black-throated Green
Warbler (Dendroica virens), Red-eyed Vireo, and

Red-breasted Nuthatch (Sitta canadensis) were
more abundant on the undeveloped, inland side of
shoreline access roads (Table 4). Differences in
relative abundance of American Redstarts (Seto-
phaga ruticilla) and Blackburnian Warblers {Den-
droica fusca) approached significance {P = 0.10)
with both detected more frequently on the
undeveloped, inland side of point counts (Ta-
ble 4).

DISCUSSION

Analyses of developed and undeveloped shore-
line areas at the broad scale showed that only one
species was significantly more abundant in
developed shoreline areas and no species were
significantly more abundant in undeveloped
shoreline areas. Fine-scale plot-level analyses in
developed areas suggested that two species were
significantly more abundant on the developed.

302

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2. June 2UI0

TABLE 3. Bird species detected on point counts between May and July 2005
areas on the Keweenaw Peninsula, Michigan.

in developed and undeveloped study

Common name

Species

AOU Code

Number of individual.s detected

Developed Undeveloped

Nesting guild

Ground Shrub

Canopy

American Crow

Corviis

brachyrhynchos

AMCR

5

0

X

American Goldfinch

Spinus tristis

AMGO

1

0

X

American Redstart

Setophaga niticilla

AMRE

14

10

X

American Robin

Tardus migratorius

AMRO

9

0

X

Black-and-white Warbler

Mniotilta varia

BAWW

7

3

X

Black-capped Chickadee

Poecile alricapillus

BCCH

6

0

X

Blackburnian Warbler

Dendroica fusca

BLBW

1

0

X

Blue Jay

Cyanocitta cristata

BLJA

2

1

X

Black-throated Green Warbler

Dendroica virens

BTNW

7

3

X

Chipping Sparrow

Spizella passerina

CHSP

3

0

X

Cape May Warbler

Dendroica tigrina

CMWA

0

1

X

Common Raven

Corvus corax

CORA

0

2

X

Chestnut-sided Warbler

Dendroica

pensylvanica

CSWA

1

0

X

Golden-crowned Kinglet

Regains satrapa

GCKl

6

1

X

Hairy Woodpecker

Picoides villosas

HAWO

2

0

X

Hermit Thrush

Catharus guttafas

HETH

3

10

X

Mallard

Anas platyrhynchos

MALL

2

0

X

Magnolia Warbler

Dendroica magnolia

MAWA

1

1

X

Yellow-rumped Warbler

D. coronata

MYWA

18

4

X

Nashville Warbler

Vermivora ruficapilla

NAWA

23

24

X

Northern Parula

Parula americana

NOPA

24

12

X

Ovenbird

Seiurus aurocapilla

OVEN

9

0

X

Pine Warbler

Dendroica pinus

PIWA

3

0

X

Red-breasted Nuthatch

Sitta canadensis

RBNU

5

3

X

Ruby-crowned Kinglet

Regulus calendula

RCKI

0

1

X

Red-eyed Vireo

Vireo olivaceus

REVI

14

3

X

Ruby-throated Hummingbird

Archilochus coluhris

RTHU

2

0

X

Song Sparrow

Melospiza melodia

SOSP

1

0

X

Swainson’s Thrush

Catharus ustidatus

SWTH

3

8

X

White-throated Sparrow

Zonotrichia alhicoUis

WTSP

7

0

X

White-breasted Nuthatch

Sitta carolinensis

WBNU

1

0

X

Winter Wren

Troglodytes

troglodytes

WIWR

6

13

X

shoreline portion rather than the undeveloped,
inland portion of the shoreline access roads. Three
species were more abundant on the inland,
undeveloped side of the shoreline access roads.
Our results suggest species may be responding to
habitat alterations associated with residential
development at the patch, rather than the larger
stand/landscape scale. We found development-
related differences in natural habitat features
including canopy volume, shrub cover, forest
lloor vegetative cover, and amount of CWD.

Riparian areas in much of the U.S. represent
concentrations of biodiversity and foci for resi-
dential development in rural areas (Naiman et al.
1993, Schnaiberg et al. 2002, Bub et al. 2004,

Gonzalez-Abraham et al. 2006). Shoreline vege-
tation, compared to adjoining inland sites, is
typically denser, more varied in species compo-
sition, and offers greater vertical and horizontal
structural diversity (Riffell et al. 2001). Rotten-
born (1999) proposed that shoreline forests have
unique vegetation characteristics compared to
similar forests distant from the shoreline and
these differences may shape patterns of bird
community composition. Research has shown that
riparian vegetation structure influences breeding
bird communities (Willson 1974, Hostetler and
Holling 2()()()) and altering these habitats through
removal of native vegetation can change bird
community structure. Vegetation structure fre-

Font and Flaspohler • BREEDING SONGBIRDS AND DEVEEOBMENT

303

TABLE 4. Birds detected on undeveloped inland and
developed shoreline sides of shoreline access roads in
residentially developed areas on the Keweenaw Peninsula,
Michigan.

Total

Species

Treatment

count

Mean/ 10 ploCs ± .SD

Zcalc“

AMCR"

Inland

0

0 ± 0

2.03

Shoreline

5

23.8 ± 0.54

AMGO

Inland

0

0 ± 0

0.97

Shoreline

1

4.8 ± 0.22

AMRE

Inland

17

33.3 ± 0.66

1.91

Shoreline

7

28.6 ± 0.56

AMRO

Inland

1

0 ± 0

1.97

Shoreline

8

38.1 ± 0.74

BCCH

Inland

1

4.8 ± 0.22

1.37

Shoreline

4

19 ± 0.4

BLEW

Inland

2

9.5 ± 0.3

1.74

Shoreline

0

0 ± 0

BLJA

Inland

1

4.8 ± 0.22

1.01

Shoreline

2

9.5 ± 0.3

BTNW“

Inland

6

28.6 ± 0.56

2.04

Shoreline

2

9.5 ± 0.3

BAWW

Inland

3

14.3 ± 0.36

0.04

Shoreline

5

23.8 ± 0.44

CHSP

Inland

1

4.8 ± 0.22

0.57

Shoreline

2

9.5 ± 0.3

CSWA

Inland

0

0 ± 0

0.97

Shoreline

1

4.8 ± 0.22

CORA

Inland

0

0 ± 0

1.4

Shoreline

0

5.6 ± 0.23

GCKI

Inland

4

19 ± 0.4

1.17

Shoreline

2

9.5 ± 0.3

HAWO

Inland

1

4.8 ± 0.22

-0.02

Shoreline

1

4.8 ± 0.22

HETH

Inland

0

0 ± 0

0.2

Shoreline

2

9.5 ± 0.3

MALL

Inland

0

0 ± 0

0.97

Shoreline

2

9.5 ± 0.44

MAWA

Inland

1

4.8 ± 0.22

1.4

Shoreline

0

0 ± 0

NAWA"

Inland

14

70 ± 0.66

4.2

Shoreline

8

38.1 ± 0.59

NOPA^

Inland

21

100 ± 0.95

4.02

Shoreline

4

19 ± 0.4

OVEN

Inland

6

28.6 ± 0.64

0.45

Shoreline

3

14.3 ± 0.36

PIWA

Inland

1

4.8 ± 0.22

-0.02

Shoreline

1

4.8 ± 0.22

RBNU'*

Inland

4

19 ± 0.4

2.22

Shoreline

1

4.8 ± 0.22

REVP

Inland

10

47.6 ± 0,51

2.47

Shoreline

4

19 ± 0.40

RCKI

Inland

1

2.8 ± 0.17

0.97

Shoreline

0

0 ± 0

RTHU

Inland

2

9.5 ± 0.30

1.40

Shoreline

0

0 ± 0

SOSP

Inland

0

0 ± 0

1.4

Shoreline

2

9.5 ± 0.3

TABLE 4, Continued.

Total

Species

Treatment

count

Mean/IO plots ± S!)

7x-alc*

SWTH

Inland

2

9.5

-L

0.3

0.7

Shoreline

1

4.8

-+■

0.22

WBNU

Inland

1

4.8

0.22

0.97

Shoreline

0

0

0

WIWR^

Inland

5

24

0.44

3.58

Shoreline

1

5

-h

0.22

WTSP

Inland

3

14

-t-

0.36

0.73

Shoreline

5

24

~h

0.44

MYWA

Inland

10

48

-h

0.68

0.71

Shoreline

9

43

-h

0.51

Mann Whitney Zerit = 1 .96 al a = 0.05 and Zerit — 1 .645 at ot — 0. 1 0. Ho
is rejected when Zcalc > Zerit.

quently influences bird species richness, relative
abundance (Willson 1974, Hostetler and Holling
2000), and foraging behavior (Smith et al. 1998).
We found development-related differences in
natural habitat features including canopy volume,
shrub cover, forest floor vegetative cover, and
amount of CWD. Research has shown that natural
habitat features such as canopy volume, shrub
cover, and forest vegetative cover are influenced
by residential development (Clark et al. 1984,
Christensen et al. 1996, Woodford and Meyer
2003). Many studies (Willson 1974, Burke and
Nol 1998, Sallabanks et al. 2000) have suggested
that changes in the avian community in the
presence of human development are the result of
development-related changes in vegetation rather
than direct human disturbance of birds.

Elias and Meyer (2003) found that undeveloped
shoreline in northern Wisconsin had higher
percent canopy, subcanopy, understory cover,
coarse woody debris, and percentage of shoreline
overhung by trees and sbrubs compared to
developed shoreline stretches. They also found
that plant species richness and diversity were
greater in developed than in undeveloped shore-
line areas. Lindsay et al. (2002) compared bird
community attributes in riparian areas along
developed and undeveloped inland lakes in
northern Wisconsin. They reported that granivo-
rous and omnivorous bird species were more
abundant along developed lakes than undeveloped
lakes. In contrast, insectivorous species were
more abundant along undeveloped than developed
lakeshores (Lindsay et al. 2002). Rottenborn
(1999) reported that shoreline development near
riparian areas in California affected nearby
riparian bird communities and concluded that

304

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2, June 2010

species richness and diversity decreased as native
vegetation was lost to residential development.

Differences in vegetation characteristics be-
tween developed and undeveloped shoreline areas
may be responsible for most of the differences we
observed in bird abundance. Deciduous shrub
density and total shrub density were greater along
developed shoreline, while mean canopy cover
and mean number of CWD pieces were greater
along undeveloped shoreline. The greater number
of all but coniferous shrubs and saplings within
developed shoreline areas is likely the result of
greater light penetration from the more open
forest canopy. Comparison of forest characteris-
tics at the individual plot scale suggested that
within-plot canopy cover, CWD, understory
density from 0 to 0.3 m, >0.3 to 1 m, and overall
understory density were greater on the undevel-
oped side of shoreline access roads. The presence
of development features such as lawns, houses,
outbuildings, walkways and driveways may con-
tribute to these differences.

Christiansen et al. (1996) found that inland lake
shoreline areas in northern Wisconsin with more
cabins had fewer large pieces of dead and down
wood in near shore waters, and speculated that
many property owners remove snags and downed
wood in or near the water’s edge. The same
process may occur around cottages, decreasing the
amount of CWD in these areas. All study areas
have thin soils which, given the close proximity to
winds off Lake Superior, may result in increased
blow-downs in developed shoreline areas, height-
ening the contrast in CWD abundance between
developed and undeveloped shorelines.

The higher abundance of CWD, canopy cover,
understory density from 0 to 0.3 m and from >0.3
to 1.0 m available in the undeveloped inland side
of shoreline access road census locations likely
influenced the relative abundance of several
species. Nashville Warblers (Vemiivora rufica-
pilla) and Winter Wrens {Troglodytes troglodytes)
were both more abundant in the undeveloped
inland portion of shoreline access roads. Nashville
Warblers nest at the ground level and Winter
Wrens often place their nests in the roots of
upturned trees or snags (Ehrlich et. al. 1988). The
higher abundance of CWD and increased under-
story density on the undeveloped, inland side of
the road may be important habitat features for
these two species.

Northern Parula (Panda americana) have been
shown to prefer riparian areas (Moldenhauer and

Regelski 1996) and to be relatively tolerant of
habitat disturbance short of clear cutting (Brooks
1940). Northern Parula, which build nests in the
mid-canopy, have been shown to be positively
correlated with at least 15% canopy cover and,
when breeding, are heavily dependent on an
abundance of epiphytes, particularly lichen (Us-
nea spp.) in their northern range (Collins et al.
1982, Moldenhauer and Regelski 1996). Northern
Parula were more abundant on the inland,
undeveloped side of shoreline access roads than
on the residentially developed side. This may be
the result of lower abundance of epiphytic lichen
related to canopy openings (Esseen and Rehhorn
1998) and/or preference by Northern Parula for
increased canopy cover, which was greater on the
inland side of point-count plots.

Red-breasted Nuthatches depend on the avail-
ability of cavities in dead standing trees or roots of
upturned trees as well as sap from living conifers.
The sap is used to coat the entrance of the cavity
during the incubation period to protect against
nest predators (Ghalambor and Martin 1999). The
increased abundance of CWD in inland areas may
be an indicator of dead, dying trees in the vicinity
and perhaps of standing dead, cavity bearing trees
making it an increasingly suitable habitat for Red-
breasted Nuthatches.

Given that our data were collected over a
single year, we realize there are several limita-
tions associated with our conclusions. Annual
variations in weather, vegetation characteristics
associated with development (i.e., possible col-
lection of CWD by landowners for firewood or
annual maintenance of vegetation within resi-
dential properties), and local bird community all
likely contribute to bird species response within
our study areas. Another recognized limitation is
the potential for birds to naturally nest >50 m
from the shoreline, regardless of the presence of
residential development. Species detection is
another factor which may be identified as a
limitation of our study; it is likely that differ-
ences exist in detection of bird species between
the more open, less vegetated residential shore-
line areas and undeveloped areas with overall
greater sub-canopy density. We believe the
differences found in bird species abundance
between residentially developed and undevel-
oped shoreline areas are legitimate based upon
the known species habitat preferences and the
measured vegetation characteristics of each study
area.

Ford and Flaspo/ilcr • BREEDING SONGBIRDS AND DEVELOPMENT

305

It is essential to clarify the scale at which bird
species respond to changes related to shoreline
development to understand how development-
related habitat alteration interacts with breeding
bird habitat use. Our results suggest birds may
respond more to development-related habitat
changes at the scale of tens of meters, than at
the scale ot the shoreline study areas we measured
(i.e., hundreds of meters).

Ornithologists and ecologists have long recog-
nized the role that scale has in understanding
habitat use in wildlife (Wiens 1989, Levin 1992).
The differing patterns of species composition and
relative abundance we found suggest development
related changes did not elicit a strong response by
most bird species at the scale of the 1 km
shoreline. However, development related changes
likely influence which areas (e.g., developed or
undeveloped side of the road) within a shoreline
area a bird chooses for breeding. Our study was
conducted in a landscape matrix that is >95%
forested, and any forest opening related to
residential development represents a relatively
small landscape disturbance. Landscape condition
has also been shown to be a strong predictor of
bird species presence (Saab 1999, Lee et al.
2002). Our results should be viewed in the context
of the largely unfragmented suiTOunding forest.

ACKNOWLEDGMENTS

This research was supported by the USDA Mclntire-
Stennis Program and the Copper Country Audubon Society.
Robert Froese, Alec Lindsay, Hugh Gorman, and Tom
Drummer were helpful in revising earlier drafts of this
manuscript. Kurt Doran and Kevin Ford assisted with
vegetation data collection.

LITERATURE CITED

Albert, D. A., S. R. Denton, and B. V. Barnes. 1986.
Regional landscape ecosystems of Michigan. School
of Natural Resources, University of Michigan, Ann
Arbor, USA.

Bibby, C. J., N. D. Burgess. D. A. Hill, and S. H.
Mustoe. 2000. Bird census techniques. Academic
Press, London, United Kingdom.

Brooks, M. 1940. The breeding warblers of the central
Allegheny Mountain Region. Wilson Bulletin 52:249-
266.

Brown, D. G. 2003. Land use and forest cover on private
parcels in the upper Midwest USA. 1970 to 1990.
Landscape Ecology 18:777-790.

Bub. B. R., D. J. Flaspohler. and C. J. Huckins. 2004.
Riparian and upland breeding-bird assemblages along
headwater streams in Michigan's Upper Peninsula.
Journal of Wildlife Management 68:383-392.

Burke, D. A. and E. Noi.. 1998. Intluencc of food
abundance, nest-site habitat, and forest fragmentation
on breeding Ovenbirds. Auk I 15:96-104.

Chri,stensen, D. L., B. R. Herwio, D. E. Schindi.er, and S.

R. Carpenter. 1996. Impacts of lakeshore residential
development on coarse woody debris in north temperate
lakes. Ecological Applications 6:1 143-1 149,

Clark, K. L., D. L. Euler, and E. Arm.strong. 1984.
Predicting avian community response to lakeshore
cottage development. Journal of Wildlife Management
48:1239-1247.

Collins, S. L., F. C. James, and P. G, Risser. 1982.
Habitat relation.ships of wood warblers (Parulidae) in
northern central Minnesota. Oikos 39:50-58.

Ehrlich, P. R., D. S. Dobkin, and D. Wheye. 1988. The
birders handbook: a field guide to the natural history of
North American birds. Simon and Schuster Inc., New
York, USA.

Eichenlaub, V. 1979. Weather and climate of the Great
Lakes Region. Notre Dame Press, South Bend, Indiana,
USA.

Elias, J. E. and M. W. Meyer. 2003. Comparisons of
undeveloped and developed shorelands, northern
Wisconsin, and recommendations for restoration.
Wetlands 23:800-816.

Esseen, P. and K. Renhorn. 1998. Edge effects on an
epiphytic lichen in fragmented forests. Conservation
Biology 12:1307-1317.

Ghalambor, C. K. and T. E. Martin. 1999. Red-breasted
Nuthatch (Sitta canadensis). The birds of North
America. Number 459.

Gonzalez- Abraham, C., V. C. Radeloff, R. B. Hammer.
T. J. Hawbaker. S. 1. Stewart, and M. K. Clayton.
2006. Building patterns and landscape fragmentation
in northern Wisconsin, USA. Landscape Ecology 22:
217-230.

Great Lakes Information Network. 2005. Great Lakes
Information Network, <www.great-lakes.net> 12
October 2005. Ann Arbor, Michigan. USA.

Gustafson, E. J., R. B. Hammer, V. C. Radeloff. and R.

S. Potts. 2005. The relationship between environ-
mental amenities and changing human settlement
patterns between 1980 and 2000 in the midwestern
USA. Landscape Ecology 20:773-789.

Hostetler, M. and C. S. Holling. 2000. Detecting the
scales at which birds respond to structure in urban
landscapes. Urban Ecosystems 4:25-54.

Lee, M., L. Fahrig, K. Freemark, and D. J. Currie. 2002.
Importance of patch scale vs. landscape scale on
selected forest birds. Oikos 96:1 10-1 18.

Levin, S. A. 1992. The problem of pattern and scale in
ecology: the Robert H. MacArthur Award Lecture.
Ecology 73:194.3-1967.

Lindsay, A. R.. S. S. Gillum, and M. W. Meyer. 2002.
Intluence of lakeshore development on breeding bird
communities in a mixed northern forest. Biological
Conservation 107:1-11.

Long, L. and A. Nucci. 1997. The 'clean break' revisited:
is U.S. population again deconcentrating? Environ-
mental Planning 29:1355-1366.

306

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 2. June 2010

Magurran, a. E. 2004. Measuring biological diversity.
Blackwell Publishing, Oxford, United Kingdom.

Manarolla, M. T. 2005. Breeding bird community
structure in developed and undeveloped areas along
the Lake Superior shoreline. Thesis. Michigan Tech-
nological University, Houghton, USA.

Mills, S. G., J. B. Dunning, and J. M. Bates. 1989.
Effects of urbanization on breeding bird community
structure in southwestern desert habitats. Condor 91:
416-428.

Moldenhauer, R. R. and D. J. Regelski. 1996. Northern
Parula (Panila americana). The birds of North
America. Number 215.

Naiman, R. j., H. Decamps, and M. Pollock. 1993. The
role of riparian corridors in maintaining regional
biodiversity. Ecological Applications 3:209-212.

Noon, B. R. 1981. Techniques for sampling avian habitats.
USDA, Forest Service, Technical Report RM-87.
Rocky Mountain Forest and Range Experiment
Station, Fort Collins, Colorado, USA.

Orr, B. 1997. Land use change on Michigan’s Lake
Superior shoreline: integrating land tenure and land
cover type data. Journal for Great Lakes Research
23:328-338.

Quinn, G. P. and M. J. Keough. 2002. Experimental
design and data analysis for biologists. Cambridge
University Press, Cambridge, United Kingdom.

Radeloff, V. C., R. B. Hammer, P. R. Voss, A. E. Hagen,
D. R. Field, and D. J. Mladenoff. 2001. Human
demographic trends and landscape level forest man-
agement in the northwest Wi.sconsin pine barrens.
Forest Science 47:229-241.

Reynolds, R. T., J. M. Scott, and R. A. Nussbaum. 1980.
A variable circular-plot method for estimating bird
numbers. Condor 82:309-313.

Riffell, S. K., B. E. Keas, and T. M. Burton. 2001. Area
and habitat relationships of birds in Great Lakes
coastal wet meadows. Wetlands 21:492-507.

Robbins, C. S. 1981. Bird activity levels related to weather.
Studies in Avian Biology 6:301-310.

Rottenborn, S. C. 1999. Predicting the impacts of

urbanization on riparian bird communities. Biological
Conservation 88:289-299.

Saab, V. 1999. Importance of spatial scale to habitat use by
breeding birds in riparian forests: a hierarchical
analysis. Ecological Applications 9:135-151.

Sallabanks, R., j. R. Walters, and J. A. Collazo. 2000.
Breeding bird abundance in bottomland hardwood
forests: habitat, edge, and patch size effects. Condor
102:748-758.

SCHNAIBERG, J., J. RiERA, M. J. TURNER, AND P. R. VOSS.
2002. Explaining human settlement patterns in a
recreational lake district: Vilas County, Wisconsin,
USA. Environmental Management 30:24—34.

Smith, R., M. J. Hamas, and M. E. Dallman. 1998.
Spatial variation in foraging of the Black-throated
Green Warbler along the shoreline of northern Lake
Huron. Condor 100:474-484.

Sutherland, W. J., I. Newton, and R. E. Green. 2004.
Bird ecology and conservation. Oxford University
Press Inc. New York, USA.

Swenson, J. J. and J. Franklin. 2000. The effects of future
urban development on habitat fragmentation in the
Santa Monica Mountains. Landscape Ecology 15:713-
730.

Theobald, D. M., J. R. Miller, and N. T. Hobbs. 1997.
Estimating the cumulative effects of development on
wildlife habitat. Landscape and Urban Planning
39:25-36.

Vale, T. R. and G. R. Vale. 1976. Suburban bird
populations in west-central California. Journal of
Biogeography 3:157-165.

Weather Underground Corporation. 2005. History for
Houghton, Michigan. Weather Underground Corporation.

I October 2005. <www.weathemnderground.com>

Wiens, J. A. 1989. The ecology of bird communities.
Cambridge University Press, Cambridge, United
Kingdom.

Willson, M. F. 1974. Avian community organization and
habitat structure. Ecology 55:1017-1029.

Woodford, J. E. and M. W. Meyer. 2003. Impact of
lakeshore development on green frog abundance.
Biological Conservation 110:277-284.

The Wilson Joiinuil ofOrnilliolof'v 1 22(2):3()7 -3 1 7, 2010

SONGBIRD NEST SURVIVAL IS INVARIANT TO
EARLY-SUCCESSIONAL RESTORATION TREATMENTS IN A

LARGE RIVER ELOODPLAIN

DIRK E. BURHANS,' " BRIAN G. ROOT,- TERRY L. SHAFFER,^ AND DANIEL C. DEY^

ABSTRACT. — We monitored .songbird nest survival in two reforesting, ~50-tia former cropland sites along the Missouri
River in central Missouri from 2001 to 2003. Sites were partitioned into three experimental units, each receiving one of
three tree planting treatments. Nest densities varied among restoration treatments for four of five species, but overall nest
survival rates did not. Nest survival varied with day-of-year and with incidence of brood parasitism by Brown-headed
Cowbirds (Mololhnis ater). Nest survival was higher early and late in the season, and parasitized nests experienced lower
nest survival, despite few complete losses directly attributable to parasitism. Probability of parasitism was inversely related
to distance to the nearest tree, and was much lower than in old field study sites in the same region. High cowbird parasitism
frequencies are usually associated with landscapes low in forest cover, yet these sites in an agriculturally-dominated
bottomland landscape experienced low (~0.8-24%) cowbird parasitism. The assumed negative relationship between
landscape-level forest cover and cowbird parasitism needs further study in habitats other than forest. Received I September
2008. Accepted 26 November 2009.

Studies of nest predation have noted variation
in nesting success among habitat types within the
same species or suites of species (McCoy et al.
1999, Lloyd and Martin 2005), or variation in
nesting success among regions or fragment sizes
(Donovan et al. 1995, Robinson et al. 1995).
These differences could be attributed to predators,
which are known to vary across regions (Thomp-
son et al. 1999, Pietz and Granfors 2000). Predator
identities may vary across habitat types within the
same landscape or region (Chalfoun et al. 2002),
although predator differences may not be reflected
in differences in nesting success (Thompson and
Burhans 2003).

Nest predation is considered the largest cause
of nest loss (Ricklefs 1969, Martin 1992), and
Brown-headed Cowbird (Molothrus ater) brood
parasitism, which also affects nest success, may
also vary by region, landscape, or habitat type
(Robinson et al. 1995, Burhans 1997, Burhans and
Thompson 2006). Much of our present under-
standing about the interactions of habitat and
land.scape features affecting cowbird parasitism
comes from studies in upland forested habitats

' Department of Fisheries and Wildlife Sciences, 202
Natural Resources Building, University of Missouri,
Columbia, MO 6521 1, USA.

^U.S. Fish and Wildlife Service, 6010 Hidden Valley
Road, Suite 101, Carlsbad, CA 9201 1, USA.

■’U.S. Geological Survey, Northern Prairie Wildlife
Research Center, 8711 37th Street Southea.st, Jamestown,
ND 58401, USA.

■•USDA Fore.st Service, Northern Research Station. 202
ABNR Building, Columbia. MO 65211. USA.

^Corresponding author; e-mail: burhansd@gmail.com

(Brittingham and Temple 1983, Robinson et al.
1995, Donovan et al. 2000, Thompson et al.
2000).

We monitored songbird nesting success at two
sites having three contiguous 16.2-ha habitats of
former cropland along the Missouri River. The
habitats varied in densities of planted oaks
{Quercus spp.) and in presence of a homogenous
herbaceous cover crop. The treatments were
sufficiently large to attract reasonable numbers
of species of nesting birds across the different
habitat types. Habitats were physically adjacent to
one another, so that any differences in predation
within the site should have been attributable to
variation in habitat type only, and not to factors
such as region, landscape context, or fragmenta-
tion. Our specific goals were to examine: (1)
whether nesting suceess varied among the planted
habitat types, and (2) whether frequencies of
cowbird parasitism varied among habitat types.

METHODS

Our central Missouri, USA research sites were
managed by the Missouri Department of Conser-
vation and were in predominately agricultural
floodplain landscapes surrounded by agriculture
or early-SLiccessional vegetation originating from
Hoods of the mid-1990s. Plowboy Bend Conser-
vation Area (38° 48' 05" N, 92° 24' 17" W) was a
row-crop agriculture/tloodplain ecosystem west of
the Missouri River’s main channel within a levee-
protected floodplain. The suirounding landscape
within a 5-km radius (Driscoll and Donovan 2004)
centered on Plowboy Bend was ~24% cropland.

307

308

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2. June 2010

26% grassland, 34% forest, and 15% water (based
on circa-2000 TM satellite data interpreted by the
Missouri Resources Assessment partnership;
www.cerc.usgs.gov/morap). Smoky Waters Con-
servation Area (38° 35' 09" N, 91° 58' 03" W) was
72 km southeast of Plowboy Bend between the
main channels of the Missouri and Osage rivers (a
major tributary of the Missouri River). The
surrounding landscape within a 5-km radius was
—24% cropland, 23% grassland, 33% forest, and
18% water. The floodplain at Smoky Waters is
subject to occasional flooding and has not been
protected since a levee was breached in the 1993
and 1995 floods. Smoky Waters was flooded for
3 weeks in June 2001, and Plowboy Bend was
flooded for 1 week during the same time period;
Smoky Waters was also flooded for 1 week in
early May 2002.

The three 16.2-ha restoration treatment units at
each site were cleared of herbaceous vegetation
and disked in autumn 1999. Two of the three units
were planted with swamp white oak (Qiierciis
hicolor) and pin oak {Q. palustris) at a 9 X 9-m
spacing (Dey et al. 2001, Shaw et al. 2003).
Redtop grass {Agrostis gigantea) was planted in
one of the two units, producing a low, dense
ground cover that substantially reduced invasion
of other herbaceous and woody vegetation. We
refer to the combination of planted oaks and
redtop grass as the “redtop” treatment. The “no
redtop” treatment contained the same oak plant-
ings, but without seeded ground cover, and had
diverse mixtures of invasive herbaceous and
woody growth compared to redtop habitats. The
remaining unit at each site served as a “control”
and was not planted to either oak or redtop (Shaw
et al. 2003). Structurally, vegetation in the control
treatments was most similar to that in the no
redtop treatments except for the ab.sence of oaks.
Planted oaks included conventional 1-year old,
not tran.splanted bare root .seedlings, but also
included “RPM®” (Root Production Method)
oaks. These oaks were grown with a special root-
pruning method and attained heights of >1.5 m
within the first year of planting (Grossman et al.
2003, Dey et al. 2004). Both sites received the
same three plantings in the same arrays but, at
Plowboy Bend, the three units adjoined each other
in a “pyramid” fashion, whereas at Smoky
Waters they were adjoined linearly (redtop, no
redtop, control).

Nest Monitoring. — We searched for and moni-
tored nests from late April to early August during

2001-2003. Nests were located by systematically
searching potential nest sites and by observing
behavior of adult birds (Martin and Geupel 1993).
We devoted equal time to searching in each
treatment unit, alternating searches among the
three habitats according to visit and time of day
within each site.

We marked nest locations with plastic flagging
placed >3 m from the nest. We focused on
Dickcissel (Spiza americana). Field Sparrow
{Spizella piisilla). Indigo Bunting (Passerina
cyanea), Red-Winged Blackbird {Agelaius phoe-
niceus), and Song Sparrow {Melospiza melodia)
because they were the most common nesters at the
sites. We monitored nests on average every 3-
4 days except during flooding in 2001 and 2002.
We found nests of the five focal species in
building (22%), laying (20%), incubation (46%),
and nestling (12%) stages.

We visited nests in early morning on the
expected fledging date (based upon observed date
of hatching or estimated age of nestlings) to look
for evidence of fledging, such as fledgling
begging calls, observation of fledglings, parents
carrying food to fledglings, or parents chipping
rapidly nearby. Nests found empty prior to this
date were considered depredated unless we
observed evidence of premature fledging. Nests
were considered “successful” if they fledged at
least one chick; fates of nests where we did not
observe these activities were classified as “un-
known” for the last interval between visits; this
interval was censored to not bias the analysis
(Stanley 2004). We approached nests and viewed
their contents at the maximum distance possible
(—2 to 4 m) to ascertain status, and were careful
not to leave “dead end trails” leading to the nest
(Martin and Geupel 1993). Interval fates other
than “successful” and “depredated” included
“abandoned” and “flooding.” Ne.st failures due
to cowbird brood parasitism were from parental
desertion (eggs remaining in abandoned parasit-
ized nest) or complete brood loss after parasitism
(starvation of host chicks or complete absence of
host chicks or eggs in parasitized nests) and were
classified as “cowbird.” “Disturbance” included
cases where nests were tipped or removed from
the substrate, presumably due to wind' or animal
trampling, but the nest contents remained on the
ground.

Vegetation Measurements. — ^We used vertical
density-board measurements taken between 20
June and 8 July in 2002 from 72 systematically

Burhans el al. • SONGBIRD NEST SURVIVAL IN l-LOODPl.AINS

309

placed locations in each treatment. Vegetation
data were collected in other years, but we chose
2002 data because samples were most complete in
that year, and because 2002 represented a
temporal midpoint in the study. Percent cover of
vertical vegetation was estimated using a nine-
increment density board (2.25 m tall X 0.25 m
wide). Percent cover of living and dead vegetation
was estimated at each 0.25-m increment from a
distance of 15 m. We estimated percent cover in
each increment for forb, grass, and woody
vegetation, combining them to generate an
estimate of mean total percent cover for each
board measurement. Grand means were calculated
for each sample over all of the 0.25-m increments
for each vegetation type of interest (forb, grass,
and total vertical vegetation). We focused on
grass cover, forb cover, and grass height because
visually these differences appeared to distinguish
treatments from each other (despite the presence
of planted oaks, woody cover was negligible
within all treatments because the young seedlings
were small). We added a category for “grass
cover <0.25 m” using only the grass scores from
the lowest increment to quantify low dense grass
cover, which appeared to visually distinguish
redtop treatments from other treatments in the
field. We also added “Mean grass height,” which
was the last-recorded increment having grass
cover on the vertical density board. We randomly
chose 68 of the 72 vegetation samples in each site
X treatment combination due to inconsistent
numbers of samples taken in each treatment.

We measured the distance from ground to the
bottom of the nest cup (±0.01 m; hereafter nest
height) upon finding each nest because nests were,
at times, toppled after predation. Nest height was
measured in 0.0 1-m increments with marked
sticks that were carried daily for nest-searching.
We obtained the spatial coordinates (Universal
Transverse Mercator) after termination of nesting
using a Trimble Geoexplorer Global Positioning
System (GPS) unit with an accuracy of 5-7 m. We
measured the distance between each nest and the
nearest tree (±:3 m in estimated height) using a
Geographic Information System (GIS) to compare
nest locations with locations ol previously
mapped trees and forest patches.

Data Analysis. — We used a general linear
model (Proc MIXED, SAS Version 9.1, 2003)
with an LSMEANS statement to calculate means
and standard errors for each vegetation variable of
interest. We included site as a random effect to

account for differences between the two sites. We
used Likelihood Ratio Tests to assess overall
model significance against a null model that
included only the intercept; we performed multi-
ple comparisons tests for differences in least
squares means among the three treatment types
using a Bonferroni adjustment il the overall model
was significant {P ^ 0.05).

We investigated whether nest densities varied
among restoration treatments by comparing ob-
served numbers of nests in each habitat to
expected numbers under the assumption ot no
difference among restoration treatments. We used
Chi-square goodness-of-fit tests to assess statisti-
cal significance (P < 0.05).

We used the logistic-exposure method (Shaffer
2004) to estimate and analyze nest survival. This
approach uses a generalized linear model with
binomial distribution (interval nest fate = 0 if
failed and 1 if successful) and logit link function
to model daily nest survival in terms of covariates
potentially affecting nest survival (Dinsmore et al.
2002, Shaffer 2004). We treated each interval
between nest checks as an observation, thereby
allowing time-dependent covariates such as nest
stage to change from one interval to the next
during the nesting cycle (Shaffer 2004). We did
not include observation days of building stage
nests in the analysis. Intervals between nest
observations were usually >1 day and varied in
length, and we used the modified logit link
function; g(0) = loge (0''V[1 - 0'^']), where 0 is
the interval survival rate and t is the interval
length in days (Shaffer 2004). The effective
sample size, which follows from the model
likelihood, is the sum of the total number of days
that all nests were known to have survived and the
number of intervals that ended in failure (Rotella
et al. 2004). We used the GENMOD procedure in
SAS (SAS Institute 2000) to estimate parameters
of the logistic-exposure models.

We developed a priori candidate models involv-
ing (1) covariates related to our hypothesis about
effects of the restoration treatment on nest survival,
and (2) additional covariates that we believed
might explain variation in ne.st survival (below).

— Restoration treatment, which was defined as
control, redtop, or no redtop.

— Nest height, which has been positively related
to nest survival for some of the same species
in other studies in the region (Burhans et al.
2002, Burhans and Thompson 2006).

310

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 2. June 2010

— Nesting stage, for which we considered the
stage (laying, incubation, nestling) confirmed
at the terminal visit of the interval to be the
stage tor that interval; e.g., if a nest transi-
tioned from incubation to the nestling stage
during an interval, a value of “nestling” was
assigned to that interval.

— Day-of-year, which may reflect seasonal
variation in predator activity or abundance.
We included both linear and quadratic terms
(day-of-year, [day-of-year]“), the latter term
to account for non-linear nest survival over
the season in models with “day-of-year”
(Grant et al. 2005).

— Cowbird parasitism, which presumably affects
nest predation because of increased begging
calls and nest visits by hosts and/or cowbirds,
which alerts predators to the presence of nests
(Dearborn et al. 1998, Hauber 2000).

— Year, site (Plowboy Bend or Smoky Waters),
bird species, and site were included as
categorical covariates to account for addi-
tional unexplained variation.

We considered models that corresponded to the
covariates and reflected our hypotheses about
relationships of habitat and other factors to nest
survival: (1) day-of-year, (2) cowbird parasitism,
(3) nest height, (4) nesting stage, and (5) habitat.
Each model also included species, year, site, and
year X site to reflect our study design and to
reduce the chance of statistical confounding. We
also considered a model that included only
species, year, site, and year X site, and a global
model that included all covariates.

Nests were considered “successful” for the
relevant interval if at least one young fledged. We
considered complete nest losses due to nest
predation, flooding, unknown weather events, or
apparent trampling by animals, abandonment, or
cowbird parasitism to be “failed” nesting at-
tempts. Only one nest fledged a cowbird but not
host young; it was considered “successful” for
the interval for which it fledged the cowbird.

We evaluated support by comparing multiple
models, fit to the same data, using an information-
theoretic approach (Burnham and Anderson
2002). We considered the model with the lowest
value for Akaike’s Information Criterion (AIC) to
be the best approximating model for the data, and
considered models within two AIC units of the
best model to repre.sent potential best models
(Burnham and Anderson 2002). However, we

report 95% confidence intervals (CIs) for each
parameter estimate based on model-averaging to
account for model-selection uncertainty (Burn-
ham and Anderson 2002). We used model-based
methods (Shaffer and Thompson 2007) to esti-
mate nest survival (including 95% CIs) in relation
to covariates.

We computed species-specific period survival
to show model-based relationships over the entire
nest period in terms of the covariates by raising
daily survival to a power equal to the average
length of the nest cycle for each species (Field
Sparrow = 23 days; Dickcissel = 25.5 days;
Indigo Bunting, Red-winged Blackbird, Song
Sparrow = 27 days; Ehrlich et al. 1988; D. E.
Burhans, unpubl. data). We realize this average
does not correspond to the exact nest cycle for
each species, but it is close for most of the species,
and our goal was to simplify graphic interpreta-
tion of model-based results. We show model-
based relationships between 26-day period sur-
vival rates and covariates other than species by
weighting the daily survival rate estimate for each
species by the proportion of observed nests of that
species. We computed model-averaged survival
estimates for a given covariate by holding
remaining covariates at their mean value. We
calculated period mortality due to a specific cause
(abandonment, predation, parasitism, disturbance)
by multiplying the model-based average period
mortality rate (1 — average period survival) by the
proportion of nests in that category; for example,
if the average period mortality over all nests was
0.75 and the proportion of abandoned nests was
0.10, period mortality due to abandonment would
be estimated as 0.10 X 0.75 = 0.075.

We used logistic regression to model the
probability of cowbird parasitism, including only
nests initiated by the second week of July, as
cowbirds typically do not lay after this time in this
region (Burhans 1997). We included bird species
in each of the models similar to the nest survival
models. Models included: (1) species only, (2)
habitat (redtop, no redtop, control) and species,
(3) site (Plowboy Bend or Smoky Waters) and
species, (4) year and species, and (5) a global
model with all of the above covariates. We also
considered (6) a model incorporating distance to
the nearest tree (>3 m in estimated height; range
= 1-263 m.). Parasitism probability has been
negatively associated with distance to the nearest
tree in other studies (Clotfelter 1998), presumably
because female cowbirds are known to survey for

Biirlums cl al. • SONGBIRD NEST SURVIVAL. IN ELOODPL.AINS

31 1

TABLE I. Distribution of nests across restoration treatments for five songbird species in lower Missouri River
floodplains, 2001-2003. Chi-square tests (df = 2, all tests) were used to assess whether nest number and densities varied
with restoration treatment.

Restoration treatment

Specie.s

Redlop

No redtop

Control

r

r

Dickcissel

22

37

8

18.84

<0.001

Field Sparrow

23

5

22

12.28

0.002

Indigo Bunting

5

31

25

18.23

<0.001

Red-winged Blackbird

70

143

40

66.55

<0.001

Song Sparrow

9

15

8

2.69

0.26

host nests from trees (Hauber and Russo 2000,
Saunders et al. 2003).

RESULTS

We analyzed data from 463 nests representing
an effective sample size of 4,773. This represent-
ed 67 Dickcissel nests (daily survival estimate
assuming constant survival = 0.941; Cl = 0.913-
0.960), 50 Field Sparrow nests (0.958; 0.930-
0.975), 61 Indigo Bunting nests (0.957; 0.934-
0.972), 253 Red-winged Blackbird nests (0.951;
0.936-0.962), and 32 Song Sparrow nests (0.933;
0.888-0.961). Estimated daily survival rate of all
species combined under the assumption of
constant survival was 0.950 (Cl = 0.939-0.959)
and average period survival over the entire nesting
cycle was 0.26. Thus, average period mortality
across the entire nest cycle from all sources was
0.74. Predation (215 losses) accounted for 63%
of period mortality. Flooding (18 losses; 5%);
abandonment, including that attributable to cow-
birds (11 losses, 3%); and unknown (weather or

animal trampling) events (9 losses, 3%) accounted
for additional mortality. Complete nest losses due
to cowbird parasitism occurred only through
abandonment with desertions at one Field Spar-
row nest and three Indigo Bunting nests (1%).
Nests were not equally distributed among resto-
ration treatments for all species except Song
Sparrow (Table 1).

The best approximating model of nest survival
was that with parasitism (Table 2; model-estimat-
ed period survival for parasitized nests: 0.10, Cl
= 0.00-0.40; unparasitized nests: 0.27, 0.19-
0.35). Day-of-year and global models also re-
ceived support (Table 2). We found only weak
support for differences in nest survival attribut-
able to restoration treatment (Table 2, Fig. IB).
The site X year and species covariates were not
related to specific hypotheses, but we include
graphs of the model-averaged values to show the
considerable variation (Fig. 1C, D).

Cowbird Parasitism. — ^We analyzed 426 nests
for probability of cowbird parasitism. Cowbird

TABLE 2. Candidate models of dally nest survival for five songbird species in lower Missouri River floodplains, 2001-
2003. Models with a lower AAIC and a greater Akaike weight (w,) have greater support; K is the number of parameters in
the model.

Model

Variable,s

K

AIC

AAIC

W,

Parasitism

Year X Site, Year, Site, Species, Parasitism"

11

1380.73

0

0.52

Day-of-year

Year X Site, Year, Site, Species. Day-of-year"

12

1382.06

1.34

0.27

Global

Year X Site, Year, Site, Species, Parasitism', Day-of-year",
Nest height. Nesting stage. Habitat

18

1383.98

3.33

0.10

Species

Year X Site, Year, Site, Species

10

1385.29

4.55

0.05

Nest height

Year X Site, Year, Site, Species, Nest height

1 1

1386.74

6.01

0.03

Nesting stage

Year X Site, Year, Site, Species, Ne.sting stage (laying,
incubation, nestling)

12

1387.51

6.79

0.02

Restoration

treatment

Year X Site, Year, Site, Species, Restoration treatment
(redtop, no redtop, control)

12

1388.28

7.56

0.01

Null

1

1420.55

39.76

0.00

“ Unparasitized = 0; parasitized = 1 .
“ Includes day-of-year, (day-of-year)^.

312

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 2, June 2010

FIG. 1. Model-averaged interval survival estimates (±95% confidence intervals) for covariates (A) day-of-year, (B)
restoration treatment, (C) site X year, and (D) species from Missouri River floodplain reforestation sites, 2001-2003.

parasitism frequency was 24.0% for Indigo
Buntings {n = 55 nests); other species were
rarely parasitized (Dickcissel 5.1%, n = 59 nests;
Field Sparrow 4.6%, n = 44 nests; Red-winged
Blackbird 0.8%, n = 239 nests; Song Sparrow
3.5%, n = 29 nests). The best-supported model
included distance to nearest tree and species
(Table 3). Probability of parasitism decreased
with distance to the nearest tree (Fig. 2).

Vegetation. — Mean forb cover varied overall
among treatments (/" = 13.1, df = 2, P < 0.05;
Table 4). Mean forb cover was lower in redtop
versus no redtop treatments (t = 12.64, df = 2,
adj. P = 0.019), but did not differ in pairwise
multiple comparisons between other treatments.
Grass cover <0.25 m varied overall among
treatments (y~ = 8.1, dt = 2, P < 0.05; Table 4)
but did not differ in any multiple-comparisons
tests between pairs of treatments.

DISCUSSION

We found little support for our principal
hypothesis about variability in nest survival due
to differences in floodplain restoration treatments.
Nest densities generally varied among habitat-
treatment types and mean forb cover was lower in
redtop treatments; the model relating nest survival
to restoration treatment had the weakest support
of any model except the null model.

Twedt et al. (2002) working in Mississippi and
Louisiana also evaluated songbird nest survival in
relation to different floodplain restoration treat-
ments. Their study considered plantings with a wider
range of ages (2-10 years) and tree heights (~3-
15 m), and they also found no differences in nest
survival among treatments. Daily nest survival of
Indigo Bunting, Red-winged Blackbird, and Dick-
cissel in their treatments with nest sample sizes >3 1
ranged from 0.919 to 0.948 (Twedt et al. 2002).

Burhaus et ai • SONGBIRD NEST SURVIVAL IN FLOODF^LAINS

313

TABLE 3. Candidate models for iiieidence of cowbird
brood parasitism in lower Missouri River tloodplains,
2001-2003. Models with a lower AAIC and a greater
Akaike weight (« ,) have greater support; K is the number of
parameters in the model.

Model

K

AlC

AAlC

M’,

Distance to nearest tree,
species

6

130.89

0.00

0.95

Global, all variables

1 1

136.88

5.99

0.05

Species only

5

141.96

1 1.07

0.00

Site, species

6

143.91

13.02

0.00

Year, species

7

144.54

13.65

0.00

Restoration treatment, species

7

145.87

14.98

0.00

Null

1

169.36

38.48

0.00

Daily nest survival rates at our sites were
comparable to other studies in central Missouri,
but were both higher and lower than sites in
northern Missouri. Average daily survival for
Field Sparrows in a study at old-field sites <30 km
away (Burhans et al. 2002) was 0.936 ± 0.004
(SE) compared with 0.958 from model-based
estimates in the present study. Daily nest survival
for Indigo Buntings at those same old fields was
0.955 ± 0.003 compared to our model-based
estimate of 0.957. In contrast, daily survival of
Indigo Buntings nesting in riparian forests and
buffer strips in northeastern Missouri was 0.90
(Peak et al. 2004). Our model-based Field
Sparrow (0.330) and Dickcissel (0.203) interval
survival rates (Fig. ID) were lower than Mayfield
success rates for Field Sparrows (0.472 ± 0.6 SE)
and Dickcissels (0.297 ± 0.2) in Conservation
Reserve Program sites in northern Missouri
(McCoy et al. 1999).

A negative relationship between cowbird par-
asitism and nest survival was evident from the
best model (Table 2), although only four complete
nest losses, via nest abandonment, were directly
attributable to cowbird parasitism. Nest abandon-
ment attributed to cowbird parasitism has been
noted frequently for Field Sparrows at more
heavily-parasitized locations (Strausberger and
Burhans 2001), but has rarely been documented
for Indigo Buntings after clutch initiation (re-
viewed in Burhans et al. 2000). We may have
incorrectly attributed bunting abandonments that
were due to death of parents or other disturbances
to cowbird parasitism. That would not affect the
outcome of the logistic-exposure survival analy-
sis, however, because these nests would still be
categorized as parasitized. Parasitized nests could

FIG. 2. Model-based probability of cowbird parasitism
in relation to distance from nearest tree for songbird species
using Missouri River floodplain reforestation sites, 2001-
2003.

have experienced lower survival if predators were
attracted to them because of more frequent and
louder begging calls by cowbird offspring (Dear-
born 1999) or more frequent visits by host parents
(Dearborn et al. 1998). However, we had few
parasitized nests that survived sufficiently long to
contain cowbird nestlings. A final explanation
could be that both nest predators and cowbirds
tended to find the same nests.

Considerable variation in nest survival among
years, sites, and dates was obvious from our most-
supported models of daily survival. Nest survival
in an earlier study of Field Sparrow and Indigo
Bunting in Missouri old-field habitats (Burhans et
al. 2002) was similarly higher in early May and
August and lower in June and early July.
Temporal patterns in nest survival likely reflect
unmeasured variables such as within-season
differences in predator activity or among-year
differences in predator foraging patterns or
abundance. Flooding may have contributed to
important yearly mortality differences, particular-
ly at one site. Sixteen nests were lost to flooding
at Smoky Waters in 2001 and 2002 with 15 of
those lost in 2001, whereas only two nests were
lost to flooding at Plowboy Bend, both in 2001
(Fig. 1 C). We did not detect a strong effect of nest
height on nesting success, although two other
studies from the same region found nest survival
increased with increasing nest height (Burhans et
al. 2002, Burhans and Thompson 2006). The
model-averaged parameter estimate for nest
height (0.03, Cl = —0.10-0.15) suggested a weak
positive effect, but the range of heights in the
present study (0-2.10 m) was lower (0 to 5.5 and

314

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2. June 2010

TABLE 4. Least-squares means (±95% Cl) for vegetation measurements from bottomland restoration sites in lower
Missouri River floodplains, 2002. Vegetation variables (expressed as proportions), except for mean grass height, are mean
cover values across four vertical board measurements taken for each sample. “

Vegetation variable

Restoration treatment

Redlop

No redtop

Control

Mean grass cover

0.21 ± 0.16

0.16 ± 0.16

0.26 ±0.16

Mean forb cover

0.05 ±0.10

0.27 ± 0.10

0.16 ± 0.10

Mean total vegetation

0.27 ± 0.22

0.43 ± 0.22

0.42 ± 0.22

Mean grass height, m

0.80 ± 0.40

0.67 ± 0.40

1 .07 ± 0.40

Grass cover <0.25 m

0.81 ± 0.29

0.23 ± 0.29

0.43 ± 0.29

Sixiy-eighi sample.s w'ere taken for each treatment in each site.

0 to 3.6 m, respectively). The reduced range of
nest heights in our study may not have been
sufficient to reveal an effect, given the positive
influence of extraordinarily tall nests on survival
in Burhans et al. (2002).

Probability of cowbird parasitism decreased
with distance to the nearest tree (Fig. 2). This
finding is in agreement with those of Clotfelter
(1998), Hauber and Russo (2000), and Saunders et
al. (2003). Johnson and Temple (1990) found that
parasitism frequencies in open prairie habitats
were higher near wooded edges, which presum-
ably contained trees for perches, than far from
wooded edges. Twedt et al. (2002) found higher
parasitism frequencies in stands with tall cotton-
wood (Populus spp.) trees compared to younger
stands with shorter trees in Mississippi and
Louisiana floodplain restoration sites.

Our sites experienced low parasitism compared
to more forested (—70%) sites in the region
(Burhans et al. 2000, Burhans and Thompson
2006) despite having low (~33%) landscape-level
forest cover. Field Sparrows experienced parasit-
ism of 11.3% (n = 443 nests) in a study of
cowbird parasitism in old fields within 30 km of
our sites (Burhans and Thompson 2006) compared
to 4.6% in this study; Indigo Buntings experi-
enced parasitism frequencies of 48% (/? = 295
nests) in old fields compared to 23.6% in this
study. Dickcissels were parasitized at 9.6% (/? =
242 nests; Winter 1999) in southwestern Missouri
prairie fragments compared to 5.1% at our sites;
Dickcissels at other sites in the Midwest have
been parasitized at frequencies of 60-85%
(Zimmerman 1983). Landscapes with low forest
cover such as ours are usually associated with
high cowbird abundance and parasitism (Robin-
son et al. 1995, Donovan et al. 2000, Thompson et

al. 2000). We observed Brown-headed Cowbirds
only rarely during 5 years of breeding point
counts at these sites (D. E. Burhans and B. G.
Root, unpubl. data), whereas cowbird detections
were an order of magnitude higher on point counts
at old field sites 30 km distant during the same
period (Burhans and Thompson 2006). Old fields
and large river floodplains are different habitats,
and may be prone to different levels of parasitism
(Robinson and Herkert 1997), possibly because of
differences in vegetation structure. Twedt et al.
(2002) similarly worked at sites having low
landscape-level forest cover (36%), but noted
lower cowbird numbers in stands having short
trees at their restoration sites. Cowbirds could be
abundant in these large river landscapes, but
possibly select habitats within them that have
more or taller trees (Twedt et al. 2002). Studies
have shown that regional and landscape effects
such as forest cover constrain parasitism at more-
local scales (Donovan et al. 2000, Thompson et al.
2000). However, most landscape-level studies
have used forested habitats to document parasit-
ism frequencies and cowbird numbers (e.g.,
Robinson et al. 1995). Several studies have looked
at variation in cowbird numbers and parasitism
among habitats (Hahn and Hatfield 1995, Straus-
berger and Ashley 1997, Robinson et al. 1999),
but the relationship between landscape-level
cowbird abundance and allocation of cowbirds
among habitats deserves further work. Much of
our understanding about landscape determinants
of cowbird parasitism comes from work in the
1980s-2000s in upland forested habitats (Britting-
ham and Temple 1983, Robinson et al. 1995,
Donovan et al. 2000, Thompson et al. 2000).
Birds using forested bottomlands, abandoned
bottomland fields, and grasslands within large

Biirinms cl at. • SONGBIRD NEST SURVIVAL IN I LOODPLAINS

315

Hoodplain agricultural landscapes may differ from
forests in habitat preference by cowbirds, and
require different management considerations.

We do not have data for livestock presence, but
low parasitism in our study could be explained by
the apparent scarcity of cattle operations in
proximity to our sites. Nesting studies have shown
decreases in parasitism with distance from grazed
habitats (Goguen and Matthews 1999, 2000)
despite cowbirds’ ability to commute between
breeding and feeding areas (Thompson 1994).

We found differences in habitat use by the
species in the restoration treatments, but our data
did not show that nest survival varied with
planting treatment. Our findings and those of
Twedt et al. (2002), who also studied agricultural
floodplains low in regional forest cover, indicate
that further study of interactions between land-
scape- and habitat-level patterns in Brown-headed
Cowbird abundance and parasitism frequencies
are warranted.

ACKNOWLEDGMENTS

We thank F. R. Thompson for helpful suggestions about
the data analysis. Many field assistants helped gather
information for this study, including Marjanne Manker,
Kristin Ellis, Charles Sharp. Heather Heitz, Faren McCord,
Sarah Moody, and Tina Foglesong. Maria Furey mapped
site-level geographic data used in the study, and Bill Dijak
provided statistics from those data. Jennifer Grabner helped
with the vegetation data and her crew gathered the
vegetation samples. We thank Soda Popp for generously
allowing use of his access road to Smoky Waters and Terry
Bruns for help when vehicles got stuck. Wes Bailey and
anonymous reviewers provided helpful comments on earlier
drafts. Some equipment, vehicles, and computer support
were courtesy of the USDA Forest Service, North Central
Research Station. This work was funded through the
University of Missouri’s Center for Agroforestry under
cooperative agreements 58-6227-1-004 with the Agricul-
tural Research Service and C R 826704-01-2 with the U.S.
Environmental Protection Agency. The results presented
are the sole responsibility of the principal investigators and/
or University of Missouri and may not represent the policies
or positions of the EPA. Any opinions, findings, conclu-
sions or recommendations expressed in this publication are
those of the author(s) and do not necessarily reflect the view
of the U.S. Department of Agriculture.

LITERATURE CITED

Brittingham, M. C. and S. A. Temple. 1983. Have
cowbirds caused forest songbirds to decline? Biosci-
ence 33:31-35.

Burhans, D. E. 1997. Habitat and microhabitat features
associated with cowbird parasitism in two forest edge
cowbird hosts. Condor 99:866-872.

Buriian.s, D, E. and F. R. Thomp.son III. 2006. Songbird
abundance and parasitism differs between urban and
rural shrublands. Ecological Applications 16:394-
405.

Burhans, D. E., F. R. Thompson III, and .1. Faaborg.
2000. Costs of parasitism incurred by two songbird
species and their quality as cowbird hosts. Condor
102:364-373.

Burhans, D. E., D. Dearborn, F. R. Thomp.son HI. and J.
Faaborg. 2002. Factors affecting predation at song-
bird nests in old fields. Journal of Wildlife Manage-
ment: 66:240-249.

Burnham, K, P. and D. R, Anderson. 2002. Model
selection and multimodel inference: a practical
information-theoretic approach. Springer-Verlag,
New York, USA.

Chalfoun, a. D., M. J. Ratnaswamy, and F. R.
Thompson III. 2002. Songbird nest predators in
forest-pasture edge and forest interior in a fragmented
landscape. Ecological Applications 12:858-867.

Clotfelter, E. D. 1998. What cues do Brown-headed
Cowbirds use to locate Red-winged Blackbird nests?
Animal Behavior 55:1181-1189.

Dearborn, D. C. 1999. Brown-headed Cowbird nestling
vocalizations and risk of nest predation. Auk 116:
448^57.

Dearborn, D. C., A. D. Anders, F. R. Thompson 111, and
J. Faaborg. 1998. Effects of cowbird parasitism on
parental provisioning and nestling food acquisition and
growth. Condor 100:326-334.

Dey, D., j. Kabrick, j. Grabner, and M. Gold. 2001.
Restoring oaks in the Missouri River floodplain. Pages
8-20 in Proceedings of the 29"’ Annual Hardwood
Symposium, 16-19 May 2001. French Lick, Indiana,
USA.

Dey, D. C., W. Lovelace, J. M. Kabrick, and M. A.
Gold. 2004. Production and early field performance of
RPM® seedlings in Missouri floodplains. Pages 59-65
in Proceedings of the 6* Walnut Council Research
Symposium. USDA Forest Service, General Technical
Report NC-243. North Central Research Station, St.
Paul. Minnesota, USA.

Dinsmore, S. j., G. C. White, and F. L. Knopf. 2002.
Advanced techniques for modeling avian nest survival.
Ecology 83:3476-3488.

Donovan, T. M., F. R. Thompson 111. and J. R.
Faaborg. 2000. Cowbird distribution at different
scales of fragmentation: trade-offs between breed-
ing and feeding opportunities. Pages 255-264 in
Ecology and management of cowbirds (J. N. M.
Smith. T. L. Cook, S. 1. Rothstein, S. K. Robinson, and
S. G. Sealy, Editors). University of Texas Press.
Austin, USA.

Donovan, T. M., F, R. Thomp.son III. J. Faaborg, and J.
R. Probst. 1995. Reproductive success of migratory
birds in habitat sources and sinks. Conservation
Biology 9:1380-1395.

Drlscoll. j. L. and T. M. Donovan. 2004. Landscape
context moderates edge effects: nesting success of
Wood Thrushes in central New York. Conservation
Biology 18:1330-1338.

316

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2, June 2010

Ehrlich. P. R.. D. S. Dobkin, and D. Wheye. 1988. The
birder’s handbook. Simon and Schuster Inc., New
York, USA.

Goguen. C. B. and N. E. Matthews. 1999. Review of the
causes and implications of the association between
cowbirds and livestock. Studies in Avian Biology
18:10-17.

Goguen, C. B. and N. E. Matthews. 2000. Local
gradients of cowbird abundance and parasitism
relative to livestock grazing in a Western landscape.
Conservation Biology 14:1862-1869.

Grant, T. A., T. L. Shaffer, E. M. Madden, and P. J.
PlETZ. 2005. Time-specific variation in passerine nest-
survival: new insights into old questions. Auk 122:
661-672.

Grossman, B. C., M. A. Gold, and D. C. Dey. 2003.
Restoration of hard mast species for wildlife in
Missouri using precocious flowering oak in the
Missouri River floodplain, USA. Agroforestry Sys-
tems 59:3-10.

Hahn, D. C. and J. S. Hatfield. 1995. Parasitism at the
landscape scale: cowbirds prefer forests. Conservation
Biology 9:1415-1424.

Hauber. M. E. 2000. Nest predation and cowbird
parasitism in Song Sparrows. Journal of Field
Ornithology 71:389-398.

Hauber, M. E. and S. A. Ru.sso. 2000. Perch proximity
correlates with higher rates of cowbird parasitism of
ground nesting Song Sparrows. Wilson Bulletin 112:
150-153.

Johnson, R. G. and S. A. Temple. 1990. Nest predation
and brood parasitism of tallgrass prairie birds. Journal
of Wildlife Management 54:106-1 1 1.

Lloyd, J. D. and T. E. Martin. 2005. Reproductive
success of Chestnut-collared Longspurs in native and
exotic grassland. Condor 107:363-374.

Martin, T. E. 1992. Breeding productivity considerations:
what are the appropriate features for management?
Pages 455-473 in Ecology and conservation of
neotropical migrant landbirds (J. M. Hagan III and
D. W. Johnston, Editors). Smithsonian Institution
Press, Washington. D.C., USA.

Martin, T. E. and G. R. Geupel. 1993. Protocols for nest
monitoring plots: locating nests, monitoring success,
and measuring vegetation. Journal of Field Ornithol-
ogy 64:507-519.

McCoy, T. D.. M. R. Ryan, E. W. Kurzeje.ski, and L. W.
Burger. 1999. Con.servation Reserve Program: source
or sink habitat for grassland birds in Missouri? Journal
of Wildlife Management 63:530-538.

Peak, R. G., F. R. Thompson 111, and T. L. Shaffer. 2004.
Factors affecting songbird nest survival in riparian
forests in a midwestern agricultural land.scape. Auk
121:726-737.

PiETZ, P. J. and D. a. Granfors. 2000. Identifying
predators and fates of grassland passerine nests using
miniature video cameras. Journal of Wildlife Manage-
ment 64:71-87.

Ricklefs, R. E. 1969. An analysis of nesting mortality
in birds. Smithsonian Contributions to Zoology 9:1-
48.

Robinson, S. K. and J. L. Herkert. 1997. Cowbird
parasitism in different habitats. Illinois Natural History
Reports 348:2-3.

Robinson, S. K., J. D. Brawn, S. F. Morse, and J. R.
Herkert. 1999. Use of different habitats by breed-
ing Brown-headed Cowbirds in fragmented mid-
we,stern landscapes. Studies in Avian Biology 18:52-
61.

Robinson, S. K., F. R. Thompson III, T. M. Donovan, D.
R. Whitehead, and J. Faaborg. 1995. Regional
forest fragmentation and the nesting success of
migratory birds. Science 267:1987-1990.

Rotella, j. J., S. J. Dinsmore, and T. L. Shaffer. 2004.
Modeling nest-survival data: a comparison of recently
developed methods that can be implemented in MARK
and SAS. Animal Biodiversity and Conservation
27:187-205.

SAS Institute Inc. 2000. Version 8. SAS Institute Inc.,
Cary, North Carolina, USA.

Saunders, C. A., P. Arcese, and K. D. O’Connor. 2003.
Nest site characteristics in the Song Sparrow and
parasitism by the Brown-headed Cowbird. Wilson
Bulletin 1 15: 24-28.

Shaffer, T. L. 2004. A unified approach to analyzing nest
success. Auk 121:526-540.

Shaffer, T. L. and F. R. Thompson III. 2007. Making
meaningful estimates of nest survival with model-
based methods. Studies in Avian Biology 34:84-
95.

Shaw, G. W., D. C. Dey, J. Kabrick, J. Grabner, and R.
M. Muzika. 2003. Comparison of site preparation
methods and stock types for artificial regeneration of
oaks in bottomlands. Pages 186-199 in Proceedings of
the 13* Central Hardwood Forest Conference, 1-3
April 2002, Urbana, Illinois (J. W. Van Sambeek, J. O.
Dawson, F. Ponder Jr., E. F. Lowenstein, and J. S.
Fralish, Editors). USDA Forest Service. General
Technical Report NC-234. North Central Research
Station, St. Paul, Minnesota. USA.

Stanley, T. R. 2004. When should Mayfield data be
discarded? Wilson Bulletin 116:267-269.

Strausberger, B. M. and M. V. Ashley. 1997.
Community-wide patterns of parasitism of a host
“generalist” brood-parasitic cowbird. Oecologia
I 12:254-262.

Strausberger, B. M. and D. E. Burhans. 2001. Nest
desertion by Field Sparrows and its possible inlJuence
on the evolution of cowbird behavior. Auk 118:770-
776.

Thompson III, F. R. 1994. Temporal and spatial patterns of
breeding Brown-headed Cowbirds in the midwestern
United States. Auk 1 1 1 :979-990.

THOMP.SON 111, F. R. and D. E. Burhans. 2003. Predation
of songbird nests differs by predator and between field
and forest habitats. Journal of Wildlife Management
67:408-416.

Thomp.son III. F. R.. W. Duak, and D. E. Burhans. 1999.
Video identification of predators at .songbird nests in
old fields. Auk 1 16:259-264.

Thompson 111. F. R.. S. K. Robin.son, T, M. Donovan. J.
R. Faaborg, D. R. Whitehead, and D. R. Larsen.

Bwhcnis et al. • SONGBIRD NEST SURVIVAL IN I’LOODPLAINS

317

2000. Biogeographic. landscape, and local factors
affecting cowbird abundance and host parasitism
levels. Pages 271-270 in Ecology and management
of cowbirds (J. N. M. Smith, T. L. Cook, S. I.
Rothstein, S. K. Robinson, and S. G. Sealy, Editors).
University of Texas Press, Austin, USA.

Twedt, D. J., R. R. Wilson, J. L. Henne-Kerr, and D. A.
Grosshuesch. 2002. Avian response to bottomland

hardwood restoration: the first ten years. Restoration
Ecology 1 0:645-6.^.‘>.

Winter, M. 1999. Nesting biology of Dickcissels and
Henslow’s Sparrows in southwestern Missouri prairie
fragments. Wilson Bulletin I 1 1:315-526.

Zimmerman, J. L. 1983. Cowbird parasitism of Dickcis.sels
in different habitats at different nest densities. Wilson
Bulletin 95:7-22.

The Wilson Journal of Ornithology 122(2):3 18-325, 2010

CAROTENOID-BASED MALE PLUMAGE PREDICTS PARENTAL
INVESTMENT IN THE AMERICAN REDSTART

RYAN R. GERMAIN,'-^ MATTHEW W. REUDINK,'--'

PETER P. MARRA,2 AND LAURENE M. RATCLIEEE'

ABSTRACT. — We examined whether male plumage coloration signals parental quality in the American Redstart
(Seiophaga ruticilla), a highly ornamented, migratory warbler. We measured the relationship between both adult male
arrival date and phenotype (morphology, melanin- and carotenoid-based plumage), and parental care levels of both parents.
Males with brighter flank feathers made more visits to the nest and spent more time at the nest, consistent with the 'good-
parent hypothesis’. Female parental care (number of visits) was negatively correlated with intensity of red of her mate’s tail
feathers and positively associated with her mate’s parental effort. These data indicate offspring of brighter males receive
more care from both parents. Our results suggest carotenoid-based plumage traits of male American Redstarts may have an
important role in intersexual signaling, and add to our understanding of the evolution of multiple ornaments. Received W
July 2009. Accepted 20 January 2010.

Three major models have been developed to
explain patterns of association between male
ornamentation and parental care. The ‘good-
parent’ hypothesis predicts that when paternal
care influences offspring viability directly, males
should signal their ability to provide for offspring
to prospective mates, and more attractive males
should provide more care (Heywood 1989,
Hoelzer 1989, Norris 1990, Hill 1999, Siefferman
and Hill 2003). Some evidence suggests female
parental care does not decrease in response to
increased care by their mate, resulting in a net
benefit to the offspring of ‘good-parent’ males
(Sanz et al. 2000, Schwagmeyer and Mock 2003,
but see Wright and Cuthill 1989, 1990; Markman
et al. 1995). Alternatively, the ‘differential
allocation’ hypothesis predicts females mated to
more attractive males increase their own level of
parental effort to avoid desertion by their mates
and to enhance their fitness through production of
high-quality offspring (Burley 1986, 1988; Studd
and Robertson 1988; Badyaev and Hill 2002). In
this model, attractive males restrict their parental
investment to allocate more energy towards their
own survival, resulting in a negative relationship
between male attractiveness and care (Burley
1986, 1988; Mpller and Thornhill 1998). A third

' Department of Biology. Queen’s University, Kingston,
ON K7L 3N6, Canada.

-.Smithsonian Migratory Bird Center, National Zoolog-
ical Park, 3001 Connecticut Avenue, Northwest, Washing-
ton. D.C. 20008. USA.

'Current address: Trent University, Forensic Science
Department. DNA Building, 2140 East Bank Drive,
Peterborough. ON K9J 7B8, Canada.

'Corresponding author;
e-mail: ryan.germain@qucensu.ca

alternative, the ‘trade-off hypothesis’, similarly
proposes that attractive males provide less care
towards their offspring (to be able to seek extra
mating opportunities), resulting in a negative
relationship between male attractiveness and level
of care (Williams 1966, Magrath and Komdeur
2003, Mitchell et al. 2007). Several studies have
failed to find a relationship between male
plumage and parental care (Rohde et al. 1999,
Smiseth et al. 2001, Cooper and Ritchison 2005).
Some of these studies, however, involve species
with multiple ornaments whose signal function
and context are not well understood.

The American Redstart (Seiophaga ruticilla), a
small (7-8 g) neotropical migratory warbler, is
particularly suited for studies of male plumage
coloration and parental care. American Redstarts
are sexually dichromatic and yearling males
exhibit delayed plumage maturation, suggesting
strong sexual selection on plumage coloration.
Adult males are black with bright orange plumage
patches on their wings, tail, and sides of the breast
(flanks), and a white or black breast, depending on
bib size (Sherry and Holmes 1997). The plumage
of adult male American Redstarts is highly
variable, both in bib size (melanin-based plum-
age) and orange (carotenoid) coloration (Lemon et
al. 1992, Kappes et al. 2009, Reudink et al.
2()()9b). Recent evidence suggests tail feathers of
American Redstarts have an important role in
intrasexLial signaling during the non-breeding
season, where males with brighter tails occupy
higher quality winter territories (Reudink et al.
2009a). Additionally, both bib size (Perreault et
al. 1997) and flank redness have been linked to a
male’s ability to secure paternity at his nest during

318

Gcnnain ci uL • AMERICAN REDSTART I’LUMAGE AND PARENl AL CARE

319

the breeding season, and tail brightness to a
male’s ehances of achieving polygyny (Reudink
et al. 2009b). In particular, flank feathers may
timction as a sexual signal in American Redstarts,
as males will puff out their breast and flank
leathers in a ‘‘flutt display” during courtship
(Sherry and Holmes 1997).

Our objectives were to quantify variation in
plumage ornamentation, morphology, and arrival
behavior of adult male American Redstarts, and
levels of parental care provided by both parents to
offspring to identify which features may provide a
reliable signal to females of male parental effort
in this highly ornamented songbird. We predicted
male plumage coloration would correlate with
levels of parental care, based on recent evidence
suggesting that different plumage regions may be
used to signal information in different contexts
(Reudink et al. 2009a, b; Kappes et al. 2009);
however, we made no a priori predictions as to
which plumage region(s) might be important.

METHODS

Field Data and Feather Collection. — ^We con-
ducted field work at the Queen’s University
Biological Station in southeastern Ontario,
Canada (44° 34' N, 76° 19' W), during two
consecutive breeding seasons (May-Jul 2006-
2007). The 60-ha study site is a mixed deciduous
forest, largely dominated by sugar maple {Acer
saccluiriim) and eastern hop-hornbeam (Ostrya
virginiana). We surveyed transects from 0600 to
1200 hrs EST each day from 1 May to 1 July
across the study site to record the arrival date for
each individual. We designated arrival date as the
number of days after the arrival of the first male
each season (i.e., first date of male an'ival = 0;
male that anJved 5 days later = 5) to standardize
arrival date across seasons. We captured adult
male redstarts on their respective temtories using
mist nests and song playback, usually within
7 days of arrival, and females during the nesting
phase using mist nets near their nest location. All
birds were banded with a Canadian Wildlife
Service aluminum band (permit # I0766C) and a
unique combination of three color bands (exclud-
ing red and orange in males); we recorded
unflattened wing chord length (±1.0 mm) as a
measure of relative body size. We plucked a
single tail feather (third rectrix; R3) and 12-15
orange flank feathers from each male (Quesada
and Senar 2006) (CWS collection permit # CA
0154). We classified adult male bib size using a

1-5 scale ( I = small amount of black plumage, 5
= large amount of black plumage with interme-
diates given half points) following Lemon et al.
(1992).

Parental Care Observations. — We monitored
American Redstart pairs each season (~40 pairs/
.season) daily until their nest was found, and
observed nests every 1-2 days following comple-
tion of nest-building until hatching and recorded
the number of young. We confirmed the number
of nestlings again on day 5 after hatching and
continued to monitor nests until fledging. Fifty
nests were observed: 25 in 2006 and 25 in 2007.
We erected ground-based video cameras (Canon
ZR500) on tripods >5 m from the base of the nest
tree on days 5 and 7 after hatching to record
parental effort. Cameras were focused on the nest
and its immediate (<20 cm) surroundings, and
recording occuiTed between 0600 and 0900 hrs
EST. Each recording lasted 120 min and the first
5 min were discarded to allow parents time to
recover from human disturbance. The remaining
1 15 min were used to calculate provisioning rates
of males and females, and the time each parent
spent at the nest. We recorded and analyzed a total
of 148 hrs of video from 38 nests (/? reduced due
to high predation rates). A trained observer
recorded the measures of parental care through
binoculars at a distance >10 m from the nest
using a stopwatch if there were more nests to be
watched than cameras available (2006 — day 5: n
= 8, day 7: n = 5; 2007 — day 5: n = 2, day 7: n
= 2). There was no difference between video
camera and trained observer nest watches in either
the number of visits to the nest, or total time at the
nest (2-sample t-test assuming unequal variance,
all P values > 0.09). Nests depredated between
days 5 and 7 were excluded from analysis (2006:
n = 7; 2007: n = 5).

We recorded both the total (i.e., sum of days 5
and 7) number of visits to the nest and the total
nest attendance (in sec) for males and females as
measures of parental care. Both variables are
effective measures of parental effort in small
insectivorous birds that make frequent trips to the
nest (Saetre et al. 1995). Total time at the nest for
males may include activities such as vigilance and
removal of fecal sacs, but does not necessarily
represent a direct benefit to rearing of offspring as
it would for females (i.e., brooding); trips to the
nest may serve as a more accurate representation
of care, as males are continually bringing food.
We divided both total number of visits per hour

320

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2. June 2010

and total nest attendance per hour by the number
ot chicks present in the nest to standardize for
brood size and recording time. We compared
standardized parental care variables with raw data
to ensure that parental effort/hr/chick was repre-
sentative of total parental effort (linear regres-
sions, all P < 0.0004). Six males in our study
were polygynous; however, primary and second-
ary nests were temporally separated, and we
included only data from the primary nest (during
the period when the male was caring exclusively
for the first nest). All individuals were identifiable
via color bands, and we were able to ensure that
our data set contained no pair duplications across
both years.

Color Analysis. — ^We mounted feathers from
each male on black paper (Colorline Ebony
#142), and stacked flank feathers as they would
naturally lie on the bird (Siefferman and Hill
2003, Quesada and Senar 2006). Mounting
feathers on a black background is a standard
technique (Siefferman and Hill 2003, Quesada
and Senar 2006), and the paper rrsed had low
(<5%) reflectance, indicating it would have a
minimal impact on our analysis. We omitted
samples from further analysis (n = 5) in
instances where there were too few flank feathers
or the orange area on the tail feather was too
small to obtain reliable readings. We analyzed
male plumage and parental care observations for
both males and females from 16 nests in 2006
and 17 nests in 2007.

We gathered reflectance spectra from flank and
tail feathers of each male using a PX-2 pulsed
xenon light source attached to an Ocean Optics
USB20()0 spectrometer (Dunedin, FL, USA). We
took 25 readings throughout the orange (caroten-
oid-based) region for both flank and tail feathers
following Reudink et al (20()9b). We gathered
raw reflectance data into 10 nm bins from 320
to 700 nm using CLRI.0.3 (Montgomerie 2008),
and averaged across the 25 measurements. We
calculated brightness for tail and flank feathers by
averaging percent reflectance from 320 to 700 nm
(color variable I: ‘brightness’). The carotenoid-
based plumage of American Redstarts consists of
peaks in both the UV and red/orange regions of
the spectrum. We used principal component
analysis (PCA) to collapse the spectrum into a
smaller set of independent variables that describe
the measures of hue and chroma based on shape of
the curve while controlling for variation in
brightness (Cuthill et al. 1999, Montgomerie

2006, Stein and Uy 2006, Reudink et al. 2()09b).
We verified this method by calculating hue and
chroma using the equations: hue=arctan ([R415-510

~ R.S20-4I.s)/R320-700]/|R573-70() ~ ^415-575)/

R320-7(X)J)5 LIV chl'oma = R320-4l.‘s/R320-7(X)> ‘ind
red chroma = R575_7()(/R32o-7(X)-

Flank PC I described 85.5% of the variation in
shape of the reflectance curve, and loaded
positively on short (UV) wavelengths and nega-
tively on the longer (red/orange) wavelength
region of the spectrum. Flank PC 1 associated
negatively with both hue {r = 0.68, n = 32, P <
0.001) and red chroma (r = 0.99, n = 32, P <
0.001), and associated positively with UV chroma
(r = 0.81, n - 32, P < 0.001). Tail feather PC I
explained 48.4% of the variation, and loaded
positively on both the shorter and longer wave-
length regions of the spectrum, and negatively on
intermediate wavelengths. Tail PC I associated
positively with hue (/“ = 0. 16, /? = 33, P = 0.02)
and red chroma (r = 0.61, n = 33, P < 0.001),
but had no association with UV chroma (r <
0.001, n = 33, P = 0.94). The first principal
component from both plumage regions was used
to describe ‘redness’ (color variable 2 = ‘red-
ness’), where birds with greater ‘redness’ have
more negative flank PC I values, and more
positive tail PC I values.

Statistical Analysis. — All statistics were per-
formed using JMP 7.0 (SAS Institute 2007) and R
2.6.1 for Windows (R Development Core Team
2007). We pooled data across both years of study
and, in instances where data for the same
individual were collected across 2 years (/; = 3),
avoided pseudoreplication by randomly selecting
1 year to exclude. We used paired r-tests to
compare the parental care variables across both
observation days (for both males and females), as
well as across gender, and linear regression to
examine the relationship between a pair’s number
of visits to the nest, and total nest attendance. All
plumage variables were tested for co-linearity
using Pearson’s pairwise analysis. We applied
backwards stepwi.se multiple regressions with
parental care variables (male visits/hr/chick,
female visits/hr/chick, male time at the nest/hr/
chick, female time at the nest/hr/chick) as the
response variable to calculate the factors that best
predicted both male and female parental care. We
used measures of male tail brightness, tail redness
(PC I), flank brightness, flank redness (PC 1), bib
score, body size, year, and arrival date as our main
effects in each model.

Germain et al. • AMERIC AN Rlil^S'l AR I' I’LUMACjl-; ANI) PARHN I'AL C'ARB

321

RESULTS

Male American Redstarts did not show a
difference in mean time (sec ± SD) spent al the
nest between days 5 (108.21 ± 112.53) and 7
(77.98 ± 75.66) (paired Mest, ^32 = 1.65, P =
0.1 1); however, mean number of visits for males
on day 7 ( 1 .92 ± 0.73) after hatching were greater
than on day 5 (1.52 ± 0.81) (paired /-test, =
— 2.89, P = 0.007). Females showed no difference
in the mean number of visits to the nest between
days 5 (1.56 ± 0.59) and 7 (1.72 ± 0.68) (paired
Mest, 632 = —1.78, P = 0.09), but spent
significantly more time at the nest on day 5
(778.04 ± 484.18 sec) than day 7 (616.68 ±
523.90 sec) (paired Mest, ^32 = 3.54, P = 0.001).
Females did not visit the nest more or less
frequently than males (paired Mest, — F27,
P = 0.21 ), but did spend a greater amount of time
at the nest (paired Mest, ^34 = —7.91, P <
0.0001 ). There was a significant positive relation-
ship between each member of a breeding pair for
number of visits (;~ = 0.21, F\ -i\ = 8.17, P =
0.008) and nest attendance (r = 0.30, F| 31 =
13.11, F = 0.001).

Pairwise correlations between all male plumage
variables revealed that tail brightness increased
with decreasing tail redness (PC I) (/~ = 0.12,
Fi.31 = 4.30, P = 0.05). No other plumage
variables were significantly coirelated.

Plumage Color and Provisioning. — Results of a
backwards stepwise multiple regression revealed
total number of male visits/hr/chick was signifi-
cantly predicted by the brightness of flank
feathers (Table 1 ). This data set contains one
high value for flank brightness (Fig. 1 ). This point
is not due to a technical eiTor in measurement, as
readings taken before and after this point were
within the normal range, and this point remained
high after repeated measurements. This individual
was also within the normal range for all other
measures. The correlation between male flank
brightness and number of visits/hr/chick remained
significant when this point was excluded (Fig. I ).
Flank brightness was also a significant predictor
of total time spent by males at the nest (Table i ).
Female visits were significantly predicted by year
and mate’s tail redness (Table 1), where lemales
in 2007 made more trips to the nest than those in
2006 (two-sample Mest assuming unequal vari-
ance, t25 = —2.77, P = 0.01). There was a
negative relationship between female visits to the
nest and male tail redness, where females which

made fewer trips to the nest were paired to males
with redder tails (Fig. 2). None of the remaining
variables related to arrival dale, morphology, or
plumage predicted male or female levels ol
parental care.

DISCUSSION

The brightness of flank feathers in adult male
American Redstarts predicted their total number
of visits to the nest, and total time spent at the
nest. Our results are consistent with the ‘good-
parent’ hypothesis and suggest that flank feather
brightness may provide a reliable signal of
parental effort in adult male American Redstarts.
A significant negative correlation was found
between number of female visits to the nest and
redness of her mate’s tail feathers. We found a
strong positive correlation between breeding
partners for visits to the nest, and nest attendance.

The ‘good-parent’ hypothesis predicts that at-
tractive males should signal their ability to provide
increased levels of parental care, compared to less
attractive males (Heywood 1989, Hoelzer 1989).
We show brightness of male carotenoid-based flank
plumage is associated with levels of male parental
care. Plumage brightness has been found to predict
mating success in other species, such as Golden-
collared Manakins (Manacus vitellinus) (Stein and
Uy 2006). In addition, it has been linked to stronger
immune response to novel antigens (Saks et al.
2003a), suggesting it has an important role in
signaling male health status and, potentially, the
ability to provide for offspring.

The pairwise correlation analysis among all
male plumage variables revealed only one signif-
icant association; tail brightness increased with
decreasing tail redness. This .suggests that in
American Redstarts, plumage brightness does not
necessarily coirespond to greater feather caroten-
oid content (Saks et al. 2()()3b. Andersson and
Prager 2006). Instead, plumage brightness may be
influenced by the underlying structural matrix of
feathers (Shawkey and Hill 2005); several studies
have linked structural-based colors with condition/
individual quality (Doiiccl 2002. Doucet and
Montgomerie 2003) as well as the level of male
parental care (Siefferman and Hill 2003). However,
the interactions between structural and pigment-
based components of feather coloration remain
poorly understood and further research is needed to
elucidate patterns of plumage brightness.

Recent work by Kappes et al. (2009) with a
smaller sample of more southerly breeding

322

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122. No. 2. June 2010

TABLE 1. Multiple regression models examining male and female parental care variables in relation to standardized
arrival date, wing length, bib score, flank feather brightness and redness, and tail feather brightness and redness. Initial
models are shown for each model. Final models (i.e., models following removal of nonsignificant effects by stepwise
backw'ard deletion) are also shown where predictor variables were significant. Sample size increased in the final models
where removal ot nonsignificant predictor variables added additional individuals to the model that were missing
information for these variables.

Male visits/hr/chick (initial model, n = 31, = 0.46)

Male visits/hr/chick (final model, /; = 32, r = 0.37)
Male visits/hr/chick (initial model, « = 31, /- = 0.19)

Male visits/hr/chick (final model, n = 32, r =0.1)
Female visits/hr/chick (initial model, /; = 31, r = 0.42)

Female visit.s/hr/chick (final model, n = 32, r = 0.33)
Female visits/hr/chick (initial model, u = 31, r = 0.14)

Female visits/hr/chick (final model, n = 33)

American Redstarts, found males with brighter
flanks provisioned less than duller males, and
those with less red Hanks sired more total young.
Flank brightness and tail brightness in our study,
in agreement with Kappes et al. (2009), were not
related suggesting the two plumage regions may
be u.sed as signals in different contexts. Flank
redness (PC I) is related to genetic paternity

Predictor variables p F P

Year

0.13

0.09

0.77

Standardized aixival

0.01

0.26

0.62

Wing

0.06

0.51

0.48

Bib score

0.06

0.15

0.70

Flank brightness

0.12

3.91

0.06

Flank redness (PC I)

0.03

0.92

0.35

Tail brightness

-0.03

0.12

0.74

Tail redness (PC I)

-0.04

1.37

0.25

Flank brightness

0.11

15.004

0.0006

Year

22.45

0.18

0.68

Standardized arrival

1.34

0.20

0.66

Wing

1.50

0.03

0.88

Bib score

-6.74

0.14

0.71

Flank brightness

4.81

0.52

0.48

Flank redness (PC 1)

3.40

0.84

0.37

Tail brightness

1.31

0.02

0.89

Tail redness (PC 1)

-2.19

0.24

0.63

Flank brightness

6.39

4.34

0.05

Year

0.69

4.15

0.05

Standardized arrival

-0.02

0.69

0.42

Wing

0.02

0.10

0.76

Bib score

0.06

0.30

0.59

Flank brightness

-0.02

0.17

0.69

Flank redness (PC I)

0.004

0.03

0.87

Tail brightness

0.03

0.19

0.67

Tail redness (PC 1)

-0.05

3.24

0.09

Year

0.49

9.30

0.005

Tail redness (PC I)

-0.05

8.16

0.008

Year

357.2

1.14

0.30

Standardized arrival

-9.75

0.27

0.61

Wing

43.61

0.49

0.49

Bib score

64.12

0.3 1

0.58

Flank brightness

-28.60

0.46

0.50

Flank redness (PC 1)

27.75

1.39

0.25

Tail brightness

61.82

1.11

0.30

Tail redness (PC I)

-5.15

0.03

0.86

No significant predictors

(indicating that it may act as a measure of genetic
quality) (Reudink et a). 2()()9b), and our results
suggest that Hank brightness may also act as an
intersexual signal of parental care.

Recent evidence suggests American Redstart
tail feather coloration acts as an intrasexual signal.
Specifically, tail brightness is related to both
territory quality in wintering areas (Reudink et al.

Germain et at. • AMERICAN REDSTART PLUMACE AND PARENTAL CARE

323

FIG. 1. Linear regression of male visits to the nest
(standardized/hr/chick) on flank feather brightness. Gray
regression line (r = 0.34, F\ t,q = 15.29, P = 0.0005)
includes high value (open circle). Black regression line (r
= 0.16, f'|,2y = 5.66, P = 0.02) excludes this individual.

2009a) and polyterritorial polygyny in breeding
areas (Reudink et al. 2009b). We found no
relationship between tail brightness and either
male or female parental care; however, female
visits to the nest were negatively correlated with
the redness of male tail feathers. Plumage redness
is often associated with greater carotenoid depo-
sition into the feather structure; high quality birds
would be assumed to have redder feathers (Saks et
al. 2003b). However, the redness of American
Redstart tail feathers has been found to signifi-
cantly decrease across subsequent breeding sea-
sons (Reudink et al. 2009b). One suggestion is
that female American Redstarts alter their level of
parental care based on the perceived age of their
mates with females providing more care when
paired with older males, although we do not
currently have the data to address this possibility.

Neither male bib size nor arrival date was
correlated with any measure of parental care by
males or females. Consi.stent with previous studies
(Perreault et al. 1997, Reudink et al. 2()09b, but
see Lemon et al. 1992), bib size does not appear to
have a prominent signaling role; however, anival
date has previously been shown to be an important
predictor of reproductive success (Norris et al.
2004, Reudink et al. 2009c). Reudink et al.
(2009c) demonstrated that early arriving males
were more likely to achieve polygyny than those
which arrived at a later date. However, when
males occupy multiple territories and provide care

Male tail feather redness (PC I)

FIG. 2. Linear regression of female visits to the nest
(standardized/hr/chick) on tail feather redness (PC I) (r =
0.15, Ti,3| = 5.60, P = 0.02). X-axis values are from low to
high, where individuals with more negative score (e.g.,
— 10) have tails with lower redness than those with
higher values.

for multiple nests, there may be a limit to the
amount of care one male can provide, as
evidenced by male American Redstarts providing
relatively less care at their second nest, than at
their primary nest (Secunda and Sherry 1991;
RRG, unpubl. data).

The strong relationship between both female
and male visits to the nest, and nest attendance
suggests that offspring of more ornamented males
are receiving the benefit of more care by both
parents. Our results also suggest the potential for
assortative mating, where females that provide
more care pair with males which do the same.
These data conform to models which suggest that
equality of investment in species with biparental
care is an evolutionarily stable strategy (offspring
receive a benefit from both parents providing high
amounts of care, reviewed in Wright and Cuthill
1989). Unfortunately, our data do not allow us to
differentiate the effects of individual feeding
effort from those of territory quality, as both male
and female feeding rates could be affected by
availability of food on their territory. It is also
possible that male coloration is not indicative of
parental care, but rather the association arises
from brighter males obtaining higher quality
teiritories and provisioning chicks more often
with less effort. Male American Redstarts able to
secure high quality territories during the over-
wintering season arrive earlier in breeding areas

324

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2, June 2010

and show higher levels of realized reproductive
success (Reudink et al. 2009c). One possibility is
that socially dominant males arriving earlier from
high quality winter territories also obtain higher
quality territories during the breeding season, in
which case we should observe a relationship
between aixival and provisioning. We did not
observe a relationship between male arrival date
and parental care, but direct measurements of
territory quality are needed to discern how
variations in individual and territory quality
influence provisioning.

The brightness of male American Redstart
flank feathers may have an important role in
intersexual signaling but further research is
needed to clarify female mate choice in this
system. Experimental approaches that measure
female response to changes in male ornamentation
would be most useful. Research on the ornamental
qualities of female plumage to examine if females
with brighter carotenoid-based plumage provide
more for their offspring and pair with brighter
males would also be informative.

ACKNOWLEDGMENTS

We thank the many field assistants who contributed to
this study, and M. M. Osmond and T. D. Barran for
assistance in video analysis. V. L. Friesen, J. J. Nocera, J.
R. Foote. C. F. Richter, and K. E. Delmore provided helpful
comments on earlier versions of this manu.script. We thank
R. D. Montgomerie for use of color analysis equipment, and
F. J. S. Phelan and F. L. Connor for logistical support at
Queen’s University Biological Station. This research was
supported by the Natural Sciences and Engineering
Research Council of Canada. Queen’s University, National
Science Foundation, Smithsonian Institution, Ontario Gov-
ernment (MTCU), American Ornithologi.sts’ Union, Society
of Canadian Ornithologists, Sigma Xi, and the American
Museum of Natural History.

LITERATURE CITED

Ander,sson, S. and M. Prager. 2006. Quantifying colors.
Pages 41-89 in Bird coloration. Volume I, Mechanics
and measurements (G. E. Hill and K. J. McGraw,
Editors). Harvard University Press, Cambridge. Mas-
sachusetts, USA.

Badyaev. a. V. AND G. E. Hill. 2002. Paternal care as a
conditional strategy: distinct reproductive tactics
as.sociated with elaboration of plumage ornamentation
in the House Finch. Behavioral Ecology 13:591-.‘i97.
Burley, N. 1986. Sexual selection for ae.sthetic traits in
species with biparental care. American Naturalist 127:
41.3-445.

BiiRi.HY. N. 1988. The differential allocation hypothesis: an
experimental test. American Naturalist I.I2: 611-628.
CooRER. S. W, AND G. Ri TCiiLSON. 2005. Nestling

provisioning by male and female Yellow-breasted
Chats: no relationships between morphology and
parental care. Journal of Field Ornithology 76: 298-
302.

CuTHiLL, I. C., A. T. D. Bennett, J. C. Partridge, and E.
J. Maier. 1999. Plumage reflectance and the objective
assessment of avian sexual dichromatism. American
Naturalist 153: 183-200.

Doucet, S. M. 2002. Structural plumage coloration, male
body size, and condition in the Blue-black Gras.squit.
Condor 104: 30-38.

Doucet, S. M. and R. Montgomerie. 2003. Multiple
sexual ornaments in Satin Bowerbirds: ultraviolet
plumage and bowers signal different aspects of male
quality. Behavioral Ecology 14: 503-509.

Heywood, J. S. 1989. Sexual selection by the handicap
mechanism. Evolution 43: 1387-1397.

Hill, G. E. 1999. Mate choice, male quality, and
carotenoid-based plumage coloration. Proceedings of
the International Ornithological Congress 22: 1654-
1668.

Hoelzer, G. a. 1989. The good parent process of sexual
selection. Animal Behaviour 38: 1067-1078.

Kappes, P. j., B. j. M. Stutchbury, and B. E.
Woolfenden. 2009. The relationship between carot-
enoid-based coloration and pairing, within- and extra-
pair mating success in the American Redstart. Condor
111: 684-693.

Lemon, R. E., D. M. Weary, and K. J. Norris. 1992. Male
morphology and behavior correlate with reproductive
success in the American Redstart {Setophaga nuicillci).
Behavioral Ecology and Sociobiology 29:399^03.

Magrath, M. J. L. AND J. Komdeur. 2003. Is male care
compromised by additional mating opportunity?
Trends in Ecology and Evolution 18: 424^30.

Markman, S., Y. Yom-Tov, and J. Wright. 1995. Male
parental care in the Orange-tufted Sunbird: behaviour-
al adjustments in provisioning and nest guarding
effort. Animal Behaviour 50: 655-669.

Mitchell, D. P., P. O. Dunn, L. A. Whhtingham, and C.
R. Freeman-Gallant. 2007. Attractive males provide
less parental care in two populations of the Common
Yellowthroat. Animal Behaviour 73:165-170.

M0LLER, A. P. AND R. THORNHILL. 1998. Male parental
care, differential parental investment by females and
sexual .selection. Animal Behaviour 55: 1507-1515.

Montgomerie, R. 2006. Analyzing colors. Pages 90-147 in
Bird coloration. Volume 1. Mechanisms and measure-
ments (G. E. Hill and K. J. McGraw, Editors). Harvard
University Press, Cambridge, Massachusetts, USA.

Montgomerie, R. 2008. CLR, Version 1.03. Queen’s
University, Kingston, Ontario, Canada, (available at
http://post.queensu.cii/~mont/color/).

Norris, D. R., P. P. Marra, T. K, Kyser, T. W. Sherry,
AND L. M. Ratcliffe. 2004. Tropical winter habitat
limits reproductive success on the temperate breeding
grounds in a migratory bird. Proceedings of the Royal
Society of London, Series B 271: 59-64.

Norris, K. .1. 1990. Female choice and the quality of
parental care in the Great Tit Parns major. Behavioral
Ecology and Sociobiology 27: 275-281.

Germain el al. • AMERICAN REDSTART PEIJMACJE AND PARENTAL CARE

325

Pf.rreaui.t, S., R, E. Lemon, and U. Kuhni.e:in. 1997.
Pattern.s and conelales of extrapair paternity in
American Redstarts (Setophafia niiicilla). Behavioral
Ecology 8; 612-621.

Quesada. J. and J. C. Senar. 2006. Comparing plumage
color measurements obtained directly from live birds
and from collected feathers: the case of the Great Tit
Pams major. Journal of Avian Biology 37: 609-616.

R Development Core Team. 2007. R: a language and
environment for statistical computing. R Foundation
for Statistical Computing, Vienna, Austria, http://
www.R-project.org.

Reudink. M. W., P. P. Marra, P. T. Boag, and L. M.
Ratclife. 2009b. Plumage colouration predicts pater-
nity and polygyny in the American Redstart. Animal
Behaviour 77; 495-501.

Reudink. M. W., C. E. Studds, T. K. Kyser, P. P. Marra,
AND L. M. Ratcliffe. 2009a. Plumage brightness
predicts non-breeding season temtory quality in a
long-distance migratory songbird. Journal of Avian
Biology 40:34-41.

Reudink, M. W., P. P. Marra, T. K. Kyser, P. T. Boag,
K. M. Langin, and L. M. Ratcliffe. 2009c. Non-
breeding season events influence polygyny and extra-
pair paternity in a long-distance migratory bird.
Proceedings of the Royal Society of London, Series
B 276:1619-1626.

Rohde, P. A., A. Johnsen, and J. T. Lifjeld. 1999.
Parental care and sexual selection in the Bluethroat,
Luscina s. svecica: a field-experimental test of the
differential allocation hypothesis. Ethology 105: 651-
663.

Saetre, G. P., T. Fossnes, and T. Slagsvold. 1995. Food
provisioning in the Pied Flycatcher; do females gain
direct benefits from choosing bright-colored males?
Journal of Animal Ecology 64: 21-30.

Saks, L., O. Indrek, and P. Horak. 2003a. Carotenoid-
based plumage coloration of male Greenfinches
reflects health and immunocompetence. Oecologia
134: 301-307.

Saks, L., K. McGraw, and P. FIorak. 2003b. How feather
colour reflects its carotenoid content. Functional
Ecology 17: 555-561.

Sanz, j. j., S. Kranenharg, and J. M. Tinbf.rgen. 2000.
Differential response by males and females to
manipulation ol partner contribution in the Great Tit
{Pams major). Journal of Animal Ecology 69: 74-84.

SAS Institute. 2007. .IMP® Version 7.0.2. Duxbury,
Pacific Grove, California, USA.

SCHWAGMEYER, P. L. AND D. W. MOCK. 2003. HoW
consistently are good parents good parents? Repeat-
ability of parental care in the House Sparrow, Passer
domesticHS. Ethology 109: 303-313.

Secunda, R. C. and T. W. Sherry. 1991. Polyterritorial
polygyny in the American Redstan. Wilson Bulletin
103:190-203.

Shawkey, M. D. and G. E. Hill. 2005. Carotenoids need
stnictural colours to shine. Biology Letters 1: 121-125.

Sherry, T. W. and R. T. Holmes. 1997. American
Redstart (Setophaga ruticilla). The birds of North
America. Number 277

Siefferman, L. and G. E. Hill. 2003. Structural and
melanin coloration indicate parental effort and repro-
ductive success in male Eastern Bluebirds. Behavioral
Ecology 14: 855-861.

Smiseth, P. T., j. Ornborg, S. Andersson, and T.
Amundsen. 2001. Is male plumage reflectance
correlated with paternal care in Bluethroats? Behav-
ioral Ecology 12: 164-170.

Stein, A. C. and J. A. C. Uy. 2006. Plumage brightness
predicts male mating success in the lekking Golden-
collared Manakin (Manaciis vitellimis). Behavioral
Ecology 17: 41-47.

Studd, M. V. and R. j. Robertson. 1988. Differential
allocation of reproductive effort to territorial estab-
lishment and maintenance by male Yellow Warblers
(Dendroica petechia). Behavioral Ecology and Socio-
biology 23: 199-210.

Williams, G. C. 1966. Adaptation and natural selection.
Princeton University Press, Princeton, New Jersey. USA.

Wright, J. and 1. Cuthill. 1989. Manipulation of .sex
differences in parental care. Behavioral Ecology and
Sociobiology 25: 171-181.

Wright, J. and I. Cuthill. 1990. Manipulation of sex
differences in parental care: the effect of brood size.
Animal Behaviour 40: 462—47 1 .

The iVilso?! Journal oj Ornithology 1 22(2):326-333, 2010

VARIATION IN PLUMAGE COLORATION OF NORTHERN
CARDINALS IN URBANIZING LANDSCAPES

TODD M. JONES,' AMANDA D. RODEWALD,'-" AND DANIEL P. SHUSTACK' -

ABSTRACT. — Biologists know relatively little of how one of the most important avian phenotypic signals, feather
coloration, may be affected by anthropogenic changes resulting from urbanization. We examined the relationship between
urbanization and carotenoid-based plumage color of Northern Cardinals (Cardinalis cardinalis) in 13 riparian forests
distributed across a rural-to-urban landscape gradient in central Ohio, USA. Feathers and morphometric measurements were
collected from breeding territorial males (n = 129) and females (/? = 145) during March-August 2006-2008. Plumage
brightness of males, but not temales, increased with body condition (i.e., size-adjusted mass) and declined with amount of
urbanization surrounding forests in which cardinals bred. The extent plumage coloration reflected condition was partially
mediated by landscape composition. Specifically, the relationship between brightness and body condition was most
pronounced in the most rural land.scapes. The interdependency of male coloration and body condition may be more relaxed
in urban than rural landscapes if carotenoid-rich foods from anthropogenic and/or invasive sources are more available, or
are accessible to birds across a wide range of condition. Received 11 May 2009. Accepted 18 January 2010.

Most studies of birds in human-dominated
landscapes have described striking and relatively
predictable patterns in response of species diver-
sity and abundance to intensity of urban develop-
ment (Blair 1996, Anderies et al. 2007, Devictor
et al. 2007, Tratalos et al. 2007). Subtle
consequences of urbanization, such as changes
in behavior or coloration are less commonly
examined. Biologists continue to have a poor
understanding of how urbanization might influ-
ence plumage coloration, despite a century of
interest in causes and consequences of this
important phenotypic signal in birds (Anders.son
1994, Bortolotti 2006). Most studies of avian
coloration in urban systems have focused on
detrimental effects of anthropogenic pollution
upon levels of carotenoids in plant and animal
matter (Sillanpaa et al. 2008) as well as plumage
coloration, as in Great Tits (Pants major) (Eeva et
al. 1998, Isaksson et al. 2005, Isaksson and
Andersson 2007), The presence of certain exotic
plants, some of which are positively associated
with urban development (Borgmann and Rode-
wald 2005), also can affect coloration by altering
access to diet-derived pigments. For example,
Mulvihill et al. (1992) showed that access to
exotic honeysuckle fruits (Lonicera spp.) contain-
ing rhodoxanthin, a carotenoid-group (xantho-

' The Ohio State University, School of Environment and
Natural Resources, 2021 Coffey Road, Columbus, OH
43210, USA.

^Current address: Massachusetts College of the Liberal
Arts. Biology Department, 375 Church Street. North
Adam.s. MA 01247. USA.

’Corresponding author; e-mail; rodcwald.l@osu.edu

phylls) pigment, changed retrix-tip colors of
Cedar Wax wings (Bomhydlla cedrorum) from
yellow to orange. However, the consequences of
urbanization affecting avian coloration remain
poorly understood despite these studies.

One of the primary ways in which urban
development might affect plumage coloration is
by altering food resources. Cities generally
contain a variety of novel food resources,
including bird seed at feeders, fruit from exotic
plants, and refuse that can change diets. These
changes in food resources can directly affect
coloration through increased availability of pig-
ment-containing food sources. The link between
diet and plumage coloration should be particularly
strong for species that have carotenoid-based
plumages, which are highly dependent on carot-
enoid pigments obtained from food sources
(Mpller et al. 2000, McGraw et al. 2001, Hill et
al. 2002). However, access to pigments is not the
sole determinant of coloration because individuals
may differ in their ability to use ingested pigments
(McGraw and Hill 2001). A variety of urban-
associated factors, such as abundant human-
commensal predators, oxidative stressors (e.g.,
pollution), and high avian densities, may constrain
the ability of individuals to acquire or physiolog-
ically sequester carotenoids for color production
(Mpller et al. 2000, Isaksson and Andersson
2007). Consequently, plumage quality .is known
to reflect general health or immunocompetence
(Andersson 1994, Saks et al. 2003, Jawor and
Breitwisch 2004, Dawson and Bortolotti 2006,
Maney et al. 2008), individual quality (Hill 1991,
Kristiansen et al. 2006), and quality of the
environment in which birds live (Arriero and

326

Jones et al. • NORTHERN CARDINALS IN URBANIZING LANDSCAPES

327

TABLE 1. Land
central Ohio, USA

cover characteristics within a

l-kni radius of 1 3 riparian-forest study sites in

urbanizing landscapes in

Study site

Urban index

I5uiidings‘

Fore.si width
(m)

Proportion of l-kni-radius area covered

Agriculture

Lawn

Pavenienl

Road

Public Hunting

-1.15

210

1 94

0.32

0.08

0.01

0.01

Prairie Oak.s

-1.12

58

1 48

0.47

0.12

0.03

0.02

Three Creek.s

-0.71

92

1 33

O.IO

0.10

0.04

0.02

South Galena

-0.57

185

1 63

0.I4

0.30

0.02

0.01

Galena

-0.48

360

277

0.I5

0.22

0.04

0.02

Elk Run

-0.16

812

1 67

0.3 1

0.27

0.06

0.05

Woodside Green

0.32

1,227

1 04

O.l I

0.40

0.07

0.05

Rush Run

0.75

1,61 1

1 50

0.00

0.41

0.09

0.06

Cherrybottoni

0.76

997

1 65

0.02

0.36

0.16

0.07

Kenny

0.89

1,733

1 26

0.00

0.34

0.17

0.06

Casto

1.25

1,776

202

0.00

0.42

0.20

0.08

Lou Berliner

1.26

2,272

1 56

0.00

0.28

0.23

0.08

Tuttle

1.61

2,285

1 60

0.00

0.34

0.30

0.09

“ Number of buildings per landscape.

Fargallo 2006). Thus, coloration might be a useful
indicator of cryptic detrimental or beneficial
effects of urbanization upon an ecosystem.

We used the Northern Cardinal {Cardinalis
cardinalis) as a model species to examine the
extent to which coloration varied with individual
body condition and the amount of urbanization
surrounding forested parks in central Ohio, USA.
The Northern Cardinal is a socially monogamous,
dichromatic, year-round resident that exhibits a
single annual molt immediately after the breeding
season (Halkin and Linville 1999). Cardinals
display a brilliant red carotenoid-based plumage
that is highly dependent upon their diet (Linville
and Breitwisch 1997, McGraw et al. 2001). The
species is known to respond positively to
urbanization, likely as a consequence of more
abundant nesting substrates, fruit resources, bird
feeders, and warmer winter temperatures than in
comparable rural landscapes (Leston and Rode-
wald 2006, Rodewald and Bakermans 2006).
Previous studies indicate that forests within urban
landscapes contain nearly three times the amount
of fruit and nearby bird feeders than in rural areas
(Atchison and Rodewald 2006, Leston and
Rodewald 2006). Increased fruit re.sources are
primarily from the presence of exotic and invasive
species, such as Amur honeysuckle (Lonicera
maackii) and multiflora ro.se (Rosa midtiflova).
These increased food resources represent one of
many ways that urbanization may inOuence the
condition and coloration of cardinals.

Specifically, we asked (1) does plumage

coloration vary along a rural-to-urban landscape
gradient, and (2) does plumage coloration reflect
differences in condition among individuals?

METHODS

Study Sites. — Cardinals were studied at 13
riparian forest sites within the Scioto River
Watershed of central Ohio, USA (Table 1,
modified from Rodewald and Shustack 2008b).
Study sites consisted of mature forest corridors
>250 m long and >100 m wide with all study
sites separated by >2 km. Riparian forests across
the study area were generally similar in both
shape and landscape configuration (i.e., they were
narrow, linear, highly-connected forests in frag-
mented landscapes).

We described the amount of urbanization
suiTounding each study site using an urban index
developed for a larger concurrent study of avian
communities across 35 sites (Rodewald and
Shustack 2()08b). Landscape composition within
a 1-km radius of each site was calculated ba.sed on
digital orthophotographs (2002-2004) and exist-
ing data from county auditors. An urban index
was calculated for each site to represent the
amount of urbanization surrounding that particu-
lar forest. Positive values of the urban index
indicate increasingly urban forests, which were
generally surrounded by >800 buildings within
the 1-km radius, and bordered by residential or
commercial areas. Negative urban index values
represent more rural forests, which were generally
surrounded by <360 buildings and bordered by

328

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 2, June 2010

agricultural fields or pastures (described in
Rodewald and Shustack 2008b).

Feather Sampling. — ^Feather samples were col-
lected from 281 Northern Cardinals captured in
mist nets during the breeding season from early
March to mid-August 2006—2008. We measured
right tarsus length (nearest 0. 1 mm), mass (g), and
length of the right unflattened right wing chord
(nearest 0.5 mm) for all individuals captured. Six
to 10 feathers were collected from the breast of
each male while 3-6 feathers were collected from
the underwing coverts of both wings of females.
These areas of feather collection were chosen
because they are known to be important areas for
cardinal coloration and have been used in
previous studies (Linville and Breitwisch 1997;
Wolfenbarger 1999a, c; Jawor et al. 2003, 2004).
Collected feathers were labeled, stored in manila
coin envelopes, and placed in a freezer until
measured.

Color Measurements. — Feathers from each
individual were superimpo.sed in layers of three
on 4.5 X 7.5 cm white note cards to imitate the
plumage surface of a bird (Bennett et al. 1997,
Quesada and Senar 2006). Superimposed feathers
were photographed at a distance of 2 cm using
a sLipra-macro feature of a Fujifilm FinePix
S800()fd digital camera. Feathers were photo-
graphed inside an 20 X 20 X 20 cm Digital
Concepts Lighting Studio (with 2 20-watt lights
directed at the feathers) to standardize lighting
conditions. Images were imported into Adobe
Photoshop CS 2™, a program which quantified
hue, saturation, and brightness. Hue was defined
as a certain point on the color spectrum and was
measured on a circular scale in degrees with 0°
(360°) being the purest red color (i.e., red with
few blue or yellow tones). The program quantified
saturation as a percentage of density with i)%
being least den.se and 100% being most den.se
(i.e., least amount of gray tones). Brightness was
quantified on a percentage grayscale with 0%
being black and 100% being white. The color-
picker tool (3X3 eye-dropper sample) was used
to randomly measure and quantify plumage
coloration at 10 points within the largest overlap-
ping area of the superimposed feathers. We
examined the precision of our color measurements
by repeating them 10 times and calculating the
coefficient of variation for measurements for each
individual. Within-individual measurements were
preci.se for brightness (CV range = 0.5^. 4; mean
± SF = 1.93 ± 0.04) and saturation (CV range =

0.5-5. 7; mean ± SE = 1.82 ± 0.05). Hue
measurements showed only small amounts of
variation (CV range = 1.8-34.9; mean ± SE =
7.26 ± 0.25). Consequently, the mean of the 10
repeat measurements was used to represent
plumage coloration for each individual. Our
approach to color quantification is widely used
(Kilner 1997, Kilner and Davies 1998, Dale 2000,
Gerald et al. 2001, Eitze and Richner 2002,
Bortolotti et al. 2003, Bezzerides et al. 2007,
Surmaki and Nowakowski 2007), but it does not
necessarily reflect how plumage is perceived by
other cardinals. Also, this method is not sensitive
to UV light as perceived by birds (Cuthill 2006),
although previous studies involving carotenoid
plumages reported positive correlations among
yellow-red spectrum (500-700 nm) and UV
reflectance peaks (Senar and Quesada 2006).

Data Analysis. — ^We reduced the number of
color variables by applying a principal compo-
nents analysis to hue, saturation, and brightness
measures separately for males and females. The
first two factors for males explained 47%
(eigenvalue = 1 .40) and 34% (eigenvalue =
1.01) of the variation in plumage color. The first
factor (hereafter called “brightness”) was strong-
ly correlated with brightness (—0.75) and satura-
tion (0.84). whereas the second factor (hereafter
referred to as “hue”) was strongly correlated with
hue (0.89). Only the first factor was useful for
females explaining 55% (eigenvalue = 1.67) of
the variation in plumage color. This first factor
(hereafter called “hue”) loaded positively for hue
(0.86) and negatively for saturation (—0.85).

We first developed a condition index to
examine relationships between coloration and
body condition using principal component analy-
sis (PC A) on wing and tarsus length for males and
females separately (SAS Institute 2002). The first
factor explained 60% of the variation in body size
in males (eigenvalue = 1.19) and was positively
correlated with both wing and tarsus (0.77 for
each). The first factor for females explained 62%
of the variation in body size (eigenvalue = 1.25)
and was positively correlated with both wing and
tarsus (0.79 for each). We then regressed mass
against this body-frame-size factor for. 129 males
and 145 females. Mass was positively associated
with body size for males (F| 127 = 1 1.92, P <
0.001), and residuals were used as a body
condition index, which indicated if individuals
were heavier (positive value) or lighter (negative
value) than expected for a given body (frame)

Jones et al. • NORTHliRN CARDINALS IN URBANIZING LANDSCARLS

329

size. Variation in female mass was not a function
of body size (F1.143 = 20.72, F = 0.399), and
we used mass alone to examine condition of
females.

We used mixed effects models to test how (I)
condition was related to the urban index, year, and
urban by year interaction (urban and year treated
as fixed effects; site treated as random effect), and
(2) how plumage coloration was influenced by
condition, the urban index, and a condition by
urban interaction (urban and condition treated as
fixed effects; site and year treated as random
effects). These models were used separately for
males and females, and for each of the two
principal components describing plumage color.
We also considered how plumage variation
among individuals might be related to urbaniza-
tion for males and females, separately, by
examining coefficient of variations for color
metrics at each site, by male/female, and year.
We used a mixed effect model with year as a
random effect and the urban index as a fixed
effect.

RESULTS

Our derived principal component (lower values
indicate brighter plumage) indicated male plum-
age was brighter as condition improved (condition
(3 = -0.12 ± 0039 SE; Ei,,,, = 10.09, P =
0.002, Fig. I A), but declined with amount of
urban development suiTounding forest patches
(urban p = 0.17 ± 0.079 SE; = 4.92, P =
0.029, Fig. IB). We detected a significant condi-
tion by urbanization interaction (P = 0.10 ±
0.036 SE; Fi.m = 7.77, P = 0.006, Fig. 2),
indicating the relationship between brightness and
condition was stronger in rural than urban
landscapes. Condition varied among years (pa-
rameter estimates were —1.99 ± 0.479 SE for
2006, -1.44 ± 0.479 SE for 2007 with 2008 as
the reference year; F\,\\ \ = 9.54, P < O.OOl), but
was not directly related to the urban index (Ci 1 1 1
= 1.76, P = 0.1869) nor via an urbanization by
year interaction (C| m = 0.22, P = 0.8025). Hue
was not significantly related to either condition
(^i.iii = 9.49, P = 0.3484) or urbanization
(F| III = 0.06, P = ().80()), nor was there a
condition by urbanization interaction (E|,iii =
0.79, P = 0.376). We found no evidence that
among-individual variation of male coloration at
sites changed along the rural-urban gradient for
either brightness (Fi.is = 0.01, P = 0.917) or hue
28 = 0.54, P = o'^^467).

We found no evidence that plumage coloration
of females was related to mass (/” 1.127 “ 0.50, P
— 0.480), urbanization (A|,i27 = 0.28, P —
0.597), or a mass by urbanization interaction
(/’I.I27 = 0.22, P = 0.638). Variation in
coloration among females was not significantly
related to urbanization (^1.29 = 1.19, /^ = 0.284).
We found some evidence that mass varied with
urbanization in some years (urban by year
interaction estimates: —1.47 ± 0.75 SE for
2006; 0.60 ± 0.808 SE for 2007 with 2008 as
the reference year; Ei 127 = 3.60, P = 0.0301),
but mass was not directly related to either
urbanization (E| 127 = 2.19, E = 0.1416) or year
(E| 127 = 1.04, P — 0.3568). Female mass
declined with urbanization in 2006, but showed
no pattern in 2007 and 2008.

DISCUSSION

Brightness of male plumage increased with
body condition, but declined as the amount of
urbanization surrounding forests in which cardi-
nals bred increased. The relationship between
brightness and body condition was stronger in
more rural than urban landscapes. The interde-
pendency of coloration and body condition may
be relaxed in urban landscapes if birds in
comparatively poor condition have access to
carotenoid-rich foods due to supplemental feeding
or fruits of exotic shrubs. We know in our study
system that urban development promotes invasion
by exotic Amur honeysuckle (Borgmann and
Rodewald 2005), which also is the largest source
of fruit at our sites (Leston and Rodewald 2006).
Mulvihill et al. (1992) reported that access to
exotic honeysuckle fruits containing rhodo-
xanthin, a carotenoid-group (xanthophylls) pig-
ment was responsible for orange, rather than
yellow, tipped retrices of Cedar Waxwings. The
weaker interdependency of condition and colora-
tion in urban landscapes may explain why we
failed to find a direct relationship between male
condition and urbanization, despite the reduced
brightness of urban males.

Others have identified links between plumage
and individual quality (Linville et al. 1998.
Wolfenbarger 1999b, Jawor and Breitwisch
2004), but we found the extent plumage coloration
reflected condition was partially mediated by
landscape composition. Our findings suggest
plumage coloration may prove to be a less useful
indicator of male quality in urban landscapes. The
potential decoupling of coloration and quality

330

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2. June 2010

FIG. 1 . Relationship of plumage brightness as described by the first principal component to (A) body condition and (B)
urban index of 129 male Northern Cardinals in riparian forests of central Ohio, 2006-2008.

may have important con.sequences given that
coloration can serve as a usetiil cue in mate
selection (e.g., sexual selection; Darwin 1871,
Andersson 1994) and, in the case of cardinals, is
known to promote assortative mating (Jawor et al.
2003).

Our finding that among-individual variation in
plumage was not related to urbanization provides
insight into the assertion that synanthropic birds
may overmatch resources in urban environments
(Shochat 2004). Studies have shown that when
individuals have more access or are less limited in
.sequestering carotenoid resources, they converge
on a similar plumage di.splay, resulting in less
variation among individuals (Hill 1992). The

comparatively dull plumage of urban birds
suggests that densities may exceed resource
levels. However, in the case of overmatching,
one also would expect to find high variation
among individuals in coloration, as only a few
birds would have access to the best resources.
Carotenoids are generally considered to be a
limiting resource among birds displaying carot-
enoid-based plumages (Mpller et al. 2000, Dale
2006), specifically for Northern Cardinals (Lin-
ville and Breitwisch 1997). One would expect
increased intraspecific competition would con-
strain individual access to carotenoids in areas
where carotenoids are especially limited due
either to environmental gradients or density-

Jones et al. • NORTHERN CARDINALS IN URBANIZING LANDSCAPES

331

Rural — — Urban

PIG. 2. The relationship between body condition and plumage brightness (i.e., principal component I) was stronger for
rural (n = 42) than urban (h = 86) male cardinals breeding in riparian forests of central Ohio, 2006-2008.

dependence, thereby increasing variation in plum-
age coloration among individuals. High variation
in carotenoid-based coloration among individuals
might be an indicator of the availability of
reduced carotenoids in a particular environment.
We found no evidence of greater variation in
access to carotenoids than individuals in more
rural forests despite greater densities of breeding
pairs in urban landscapes (Leston and Rodewald
2006, Rodewald and Bakermans 2006). Supple-
mental food resources associated with residential
land, such as bird feeders and fruiting plants
(Atchison and Rodewald 2006, Leston and
Rodewald 2006) may permit high densities of
cardinals to use similar or greater per capita food
resources than birds living in more rural environ-
ments. This scenario of apparent resource-match-
ing is consistent with patterns reported by
Rodewald and Shustack (2008a), which showed
that survival, reproductive productivity, and
energetic condition of cardinals were similar
across the rural-to-urban gradient.

The presence of a relationship between color-
ation and urbanization suggests development
surrounding forests inlluences the ability of
cardinals to either acquire or sequester carotenoid
pigments. However, there are a number of
important caveats to our findings. First, given
that coloration of cardinals may improve with age
(Wolfenbarger 1996), we cannot exclude the
possibility that different age structure in urban
and rural populations contributed to the observed

patterns. Second, there is a possibility that
cardinals molted and grew feathers at different
locations than where they were captured, which
may reflect different food conditions. However,
cardinals in the study system: (1) established
breeding territories beginning in March, (2)
generally molted in August while still on teiritory,
and (3) had high site fidelity across years. Third,
we focused only on plumage coloration, and other
ornaments (e.g., mask, crown, and beak colora-
tion) of cardinals also may be important indicators
of quality (Jawor and Breitwisch 2004). Our study
provides some evidence that urban disturbances
can influence plumages of Northern Cardinals in
our system, but continued study is needed to
identify and understand the underlying mecha-
nisms of plumage differences.

ACKNOWLEDGMENTS

This research was generously supported by a Schwab
Associate Scholarship grant from The Ohio State Univer-
sity. College of Biological Sciences, and a Research
Experience for Undergraduates (REU) supplement from
the National Science Foundation (DEB-0639429 to A. D.
Rodewald). Field components of this research were
partially supported by the National Science Foundation
( DEB-034()879 and DEB- 0639429 to A. D. Rodewald) and
the Ohio Division of Wildlife. This work would not have
been possible without dedicated efforts of L. J. Kearns, I. J.
Ausprey, and J. R. Sinith-Castro. We appreciate helpful
comments from W. M. Masters and P. G. Rodewald that
improved this manuscript. We thank the following for their
work in the field; B. T. Adams. Elizabeth Ames. Christine
Austin, S. C. Buescher, L. M. Koerner, S. E. Lehnen, W. W.

332

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

Li, L. L. MacArthur, Angela Petersen, J. E. Price, E. C.

Rogers, Kaitlin Uppstrom. Benjamin Van Allen, and R. M.

Zajac. We are grateful to Franklin County Metro Parks,

Columbus Parks and Recreation, Ohio Division of Wildlife,

The Nature Conservancy, City of Bexley, Gahanna Parks

and Recreation, and private landowners for their coopera-
tion and access to sites.

LITERATURE CITED

Anderies, j. M., K. Madhusudan, and E. Shochat. 2007.
Living in the city: resource availability, predation, and
bird population dynamics in urban areas. Journal of
Theoretical Biology 247:36-49.

Andersson, M. B. 1994. Sexual selection. Princeton
University Press, Princeton, New Jersey, USA.

Arriero, E. and j. a. Fargallo. 2006. Flabitat structure is
associated with the expression of carotenoid-based
coloration in nestling Blue Tits (Panis caeruleus).
Naturwissenschaften 93: 1 73-180.

Atchison, K. A. and A. D. Rodewald. 2006. The value of
urban forests to wintering birds. Natural Areas Journal
26: 280-288.

Bennett, A. T. D., I. C. Cuthill, J. C. Partridge, and K.
Lunau. 1997. Ultraviolet plumage colors predict mate
preferences in starlings. Proceedings of the National
Academy of Sciences of the USA 94: 8618-8621.

Bezzerides, L. a., K. j. McGraw, R. S. Parker, and J.
Husseini. 2007. Elytra color as a signal of chemical
defense in the Asian Ladybird Beetle Harmonia
a.xyridis. Behavioral Ecology and Sociobiology 61:
1401-1408.

Blair, R. B. 1996. Land use and avian species diversity
along an urban gradient. Ecological Applications
6:506-519.

Borgmann, K. L. and a. D. Rodewald. 2005. Forest
restoration in urbanizing landscapes: interactions
between land uses and an exotic shrub. Restoration
Ecology 13:334-340.

Bortoloiti, G. R. 2006. Natural selection and coloration:
protection, concealment, advertisement, or deception?
Pages 3^0 in Bird coloration: evolution and function
(G. E. Hill and K. J. McGraw, Editors). Volume 2.
Harvard University Press, Cambridge. Massachusetts,
USA.

Bortolotti. G. R., K. j. Fernie, and J. E. Smits. 2003.
Carotenoid concentration and coloration of American
Kestrels {Falco sparveriiis) disrupted by experimental
exposure to PCBs. Functional Ecology 17: 651-
657.

Cuthill, 1. C. 2006. Color perception. Pages 3-35 in Bird
coloration: mechanisms and measurements (G. E. Hill
and K. J. McGraw, Editors). Volume 1. Harvard
University Press, Cambridge. Massachu.setts, USA.

Dale. J. 2000. Ornamental plumage does not signal male
quality in Red-billed Queleas. Proceedings of the
Royal Society of London, Series B 267: 2143-2149.

Dali:. .1. 2006. Intraspecific variation in coloration. Pages
36-86 in Bird coloration: function and evolution (G. E.
Hill and K. .1. McGraw, Editors). Volume 2. Harvard
University Press, Cambridge, Mas.sachu.setts, USA.

Darwin, C. [1871] 1981. The descent of man, and selection
in relation to sex. Princeton University Press, Prince-
ton, New Jersey, USA.

Dawson, R. D. and G. R. Bortoloiti. 2006. Carotenoid-
dependent coloration of American Kestrels predicts
ability to reduce parasitic infections. Naturwis-
senschaften 93: 597-602.

Devictor, V., R. Julliard, D. Couvet, A. Lee, and F.
Jiguet. 2007. Functional homogenization effect of
urbanization on bird communities. Conservation
Biology 21: 741-751.

Eeva, T., E. Lehikoinen. and M. Ronka. 1998. Air
pollution fades the plumage of the Great Tit.
Functional Ecology 12: 607-612.

Fitze, P. S. and H. Richner. 2002. Differential effects of a
parasite on ornamental structures based on melanins
and carotenoids. Behavioral Ecology 13:401-407.

Gerald, M. S., J. Bernstein, R. Hinkson, and R. A. E.
Fosbury. 2001. Formal method for objective assess-
ment of primate color. Journal of Primatology 53: 79-
85.

Halkin, S. L. and S. U. Linville. 1999. Northern Cardinal
(Cardinalis cardinalis). The birds of North America.
Number 440.

Hill, G. E. 1991. Plumage coloration is a sexually selected
indicator of male quality. Nature 350: 337-339.

Hill, G. E. 1992. Proximate basis of variation in carotenoid
pigmentation in male House Finches. Auk 109: 1-12.

Hill, G. E., C. Y. Inouye, and R. Montgomerie. 2002.
Dietary carotenoid predict plumage coloration in wild
House Finches. Proceedings of the Royal Society of
London, Series B 269: 1119-1124.

ISAKSSON, C. AND S. Andersson. 2007. Carotenoid diet
and nestling provisioning in urban and rural Great Tits
Panes major. Journal of Avian Biology 38: 564-572.

ISAKSSON, C., J. Ornborg, E. Stephensen, and S.
Andersson. 2005. Plasma glutathione and carotenoid
coloration as potential biomarkers of environmental
stress in Great Tits. EcoHealth 2: 138-146.

Jawor, j. M. and R. Breitwisch. 2004. Multiple
ornaments in male Northern Cardinals, Cardinalis
cardinalis, as indicators of condition. Ethology 110:

1 13-126.

Jawor, J. M., S. U. Linville, S. M. Beall, and R.
Breitwisch. 2003. Assortative mating by multiple
ornaments in Northern Cardinals (Cardinalis cardina-
lis). Behavioral Ecology 14: 515-520

Jawor, J. M., N. Gray, S. M. Beall, and R. Breitwisch.
2004. Multiple ornaments correlate with aspects ol
condition and behavior in female Northern Cardinals,
Cardinalis cardinalis. Animal Behaviour 67: 875-882.

Kilner, R. 1997. Mouth colour is a reliable signal of need
in begging Canary nestlings. Proceedings of the Royal
Society of London, Series B 264: 963-968.

Kilner, R. and N. B. Davies. 1998. Nestling mouth
colour: ecological correlates of a begging signal.
Animal Behaviour 56: 705-712.

Kristiansen, K. O., J. O. Bustnes, I. Folstad, and M.
Helberg. 2006. Carotenoid coloration in Great Black-
backed Gull Lariis marinits reflects individual quality.
Journal of Avian Biology 37: 6-12.

Jones ct III. • NORTHERN CARDINALS IN URBANIZING LANDSCAPES

333

Lkston, L. F. V. AND A, D, Rodkwald. 2()()6. Are urban
forests ecological traps for unticrstory birds? An
examination using Northern Cardinals. Biological
Conservation 131: 566-574.

Linville, S. U. and R. Breitwisch. 1997. Carotenoid
availability and plumage coloration in a wild popula-
tion of Northern Cardinals. Auk 1 14: 796-800.

Linville, S. U., R. Breitwisch, and A. J. Schilling.
1998. Plumage brightness as an indicator of parental
care in Northern Cardinals. Animal Behaviour 55:
119-127.

Maney, D. L., a. K. Davis, C. T. Goode, A. Reid, and C.
Showalter. 2008. Carotenoid-based plumage color-
ation predicts leukocyte parameters during the breed-
ing season in Northern Cardinals (Cardiiuilis cardina-
lis). Ethology 1 14: 369-380.

McGraw, K. J. and G. E. Hill. 2001. Carotenoid access
and intraspecific variation in plumage pigmentation in
male American Goldfinches {Carduelis tristis) and
Northern Cardinals (Ccirdinalis cardinalis). Functional
Ecology 15:732-739.

McGraw, K. J., G. E. Hill, R. Stradi, and R. S. Parker.
2001. The influence of carotenoid acquisition and
utilization on the maintenance of species-typical
plumage pigmentation in male American Goldfinches
(Carduelis rristis) and Northern Cardinals (Cardinalis
cardinalis). Physiological and Biochemical Zoology
74: 843-852.

M0LLER, A. P., C. Biard, j. D. Blount, D. C. Houston, P.
Ninni, N. Saino, and P. F. Sural 2000. Carotenoid
dependent signals: indicators of foraging efficiency,
immunocompetence or detoxification ability? Avian
Poultry Science Review 1 1:137-159.

Mulvihill, R. S., K. C. Parkes, R. C. Leberman, and D.
S. Wood. 1992. Evidence supporting a dietary basis
for orange-tipped rectrices in the Cedar Waxwing.
Journal of Field Ornithology 63: 212-216.

Quesada, j. and j. C. Senar. 2006. Comparing plumage
colour measurements obtained directly from live birds
and from collected feathers: the case of the Great Tit
Pams major. Journal of Avian Biology 37: 609-616.

Rodewald, a. D. and M. H. Bakermans. 2006. What is
the appropriate paradigm for riparian forest conserva-
tion? Biological Conservation 128: 193-200.

Rodewald, A. D. and D. P. Shustack. 2008a. Consumer
resource matching in urbanizing landscapes: are

synanlhropic species over-matching? Ecology 89:
515-521.

Rodewald, A. D. and D. P. Shu.stack. 2008b. Urban
flight: understanding individual and population-level
responses of Nearctic-neotropical migratory birds to
urbanization. Journal of Animal Ecology 77: 83-91.

Saks, L., 1. Ots, and P. Horak. 2003. Carotenoid-based
plumage coloration of male Greenfinches reflects
health and immunocompetence. Oecologica 134:
301-307.

SAS Institute. 2002. SAS/STAT, Software Release 9.1.
SAS Institute, Cary, North Carolina, USA.

Senar, J. C. and J. Quesada. 2006. Absolute and relative
signals: a comparison of melanin- and carotenoid-
based patches. Behaviour 143: 589-595.

Shochat, E. 2004. Credit or debit? Resource input changes
population dynamics of city-slicker birds. Oikos
106:622-626.

Sillanpaa, S., j. P. Salminen, E. Lehikoinen, E.
Toivonen, and T. Eeva. 2008. Carotenoids in a food
chain along a pollution gradient. Science of the Total
Environment 406:247-255.

SuRMACKl, A. AND J. K. NowAKOWSKi. 2007. Soil and
preen waxes influence expression of carotenoid-based
plumage coloration. Naturwissenschaften 94: 829-
835.

Tratalos, j., R. a. Fuller, K. L. Evans, R. G. Davies,
S. E. Newson, j. j. D. Greenwood, and K. J.
Gaston. 2007. Bird densities are associated with
household densities. Global Change Biology 13:
1685-1695.

Wolfenbarger, L. L. 1996. Fitness effects associated with
red coloration of male Northern Cardinals (Cardinalis
cardinalis). Dissertation. Cornell University, Ithaca,
New York, USA.

Wolfenbarger, L. L. 1999a. Female mate choice in
Northern Cardinals: is there a preference for redder
males? Wilson Bulletin 111: 76-83.

Wolfenbarger, L. L. 1999b. Is red coloration of male
Northern Cardinals beneficial during the non-breeding
season?: a test of status signaling. Condor 111: 655-
663.

Wolfenbarger, L. L. 1999c. Red coloration of male
Northern Cardinals correlates with mate quality and
territory quality. Behavioral Ecology 10: 80-90.

The Wilson Journal oj Ornithology 122(2);334-339, 2010

ANNUAL PRECIPITATION AFFECTS REPRODUCTION OF THE
SOUTHERN GREY SHRIKE (LANIUS MERIDIONALIS)

ODED KEYNAN' AND REUVEN YOSEP--^

ABSTRACT. — We studied the breeding ecology of the Southern Grey Shrike (Lanins mericlionalis) at the Shezaf Nature
Reserve, Arava Valley, Israel, an extremely arid desert with mean annual rainfall of 30 mm. We color-banded 128 shrikes
during 2007-2009. The breeding season lasted from late February until late June. We found 34 nests and a correlation
between years with amount of precipitation and number of breeding pairs. Second broods after successful broods were
found only in 2007 and were all unsuccessful. Most nests were in Acac ia trees; pairs that nested in dry trees, lacking foliage,
were more likely to fail. The average clutch size was 3.44; average number of nestlings was 1.95; and average number of
nestlings that fledged was 1.24. These results are lower than other studies of the species, probably a result of the scarcity of
food in the arid environment. Nest survival rate was higher during incubation than during rearing of nestlings (0.71 vs. 0.62),
and total nest survival was 44.5%. The arid Arava Valley influences breeding success of Southern Grey Shrikes, and the
present .severe drought could negatively influence persistence of the shrike population in the Arava Valley. Received 6
September 2009. Accepted 23 November 2009.

Rainfall is known to have profound effects on
the life history of many bird species and different
populations within species (Lloyd 1999, Morrison
and Bolger 2002, Mondajem and Bamford 2009).
Water is a strong limiting factor in arid ecosys-
tems, where rainfall is generally low, erratic, and
unpredictable (Noy-Meir 1973). This leads to low
primary productivity and low food availability
when combined with high temperatures (Tieleman
et al. 2004). Birds that reproduce in arid
ecosystems respond to these limiting factors by
reducing clutch size (Lloyd 1999, Tieleman et al.
2004) or breeding earlier in the season or by not
breeding at all in regions where conditions are
particularly unfavorable (Immelmann 1973).

Arid ecosystems are characterized by high
between-year variations in both amount and
duration of productivity due to the unpredictabil-
ity of rains (Illera and Diaz 2006). Thus,
individual birds that live in these environments
must have a high phenotypic flexibility to survive
and reproduce (Ricklefs and Wikelski 2002).
Lloyd (1999) showed that >50% of the species
in South Africa’s arid zones had inter-seasonal
and inter-annual variations in clutch size due to
changes in food availability. Tieleman et al.
(2004), in their re,search on life history variation
along an aridity gradient in larks, showed that nest
predation increa.sed while both growth rate and
clutch size decreased with increasing aridity.

' Department of Zoology, George S. Wise Faculty of Life
•Sciences, Tel Aviv University. Israel.

^ International Birding and Research Center in Eilat, P. O.
Box 774, Eilat 88000. Israel.

•’Corresponding author; e-mail: ryoseftieilatcity. co.il

The Southern Grey Shrike {Lanins meridiona-
lis) is a passerine (Laniidae). Birds in this family
are known for their similarity to diurnal raptors in
morphology, especially the strong and sharp bill,
and their predatory behavior (Yosef 2008).
Shrikes do not posses strong talons, in contrast
to raptors, and impaling behavior evolved in the
family as a way to dismember prey (Smith 1973).

The Southern Grey Shrike was recently sepa-
rated from the Great Grey Shrike (L. excnhitor)
(Lefranc and Worfolk 1997). It is widely
distributed in the Middle East, North Africa, and
central Asia (Lefranc and Worfolk 1997); in
Israel, it is resident along the Dead Sea Ritt
Valley (Shirihai 1996). There are few studies of
its reproductive biology in extremely arid ecosys-
tems (e.g., Yosef 1992, Budden and Wright 2000);
more research is needed to understand the species’
response to extreme aridity and a fluctuating,
scarce food base.

Our objectives were to: ( 1 ) characterize the
breeding ecology of the Southern Grey Shrike in
an extremely arid environment, (2) compare it to
previous studies on populations of the same
species in other and similar environments, and
(3) characterize the changes in the reproductive
biology within our population following changes
in precipitation between years. We hypothesized
that, similar to Tieleman et al. (2004), we would
find an effect of aridity on reproductive effort
and success of the Southern Grey Shrike. We
predicted that, because of the reduced prey base in
desert areas, the number of clutches of a pair of
shrikes in the season would be low due to
exposure and lack of appropriate breeding sites.

334

Keymin and Yosef - BRFEDING OJ- SOUTHERN C3REY SHRIKES

335

METHODS

Study Area. — ^The study was conducted at the
Shezaf Nature Reserve during April 2006 to July
2009 in the southern part of Israel, —30 km south
ot the Dead Sea in the Arava Rift Valley between
the settlements of Hazeva (30° 46' N, 35° 16' E)
and Ein Yahav (30° 45' N, 35° 15' E). The Arava
Valley is an extremely arid zone (UNESCO 1977)
with a mean annual winter rainfall of 35 mm that
occurs in a range of 6-9 days with large annual
variations in total rainfall and temporal and spatial
distribution (Anava et al. 2000). The mean
temperature during summer is 38° C with daily
temperatures that can rise to 49° C (Goldreich and
Kami 2001). The flora of the Shezaf Nature
Re.serve is dominated by Acacia trees {Acacia
tortilis and A. raddiana) and scattered shrubs
(ZiUa spiuosa, Lyciiim shawii, and Haloxylon
persicum) that occur only in the dry river beds
(wadis).

Field Procedures. — We color-banded 128
Shrikes for individual recognition. Eorty-nine
(38.3%) adults were captured in a treadle trap
(Yosef and Lohrer 1992) and 79 (61.7%) nestlings
were banded at the nest 10-11 days post-hatching.
All shrikes were accustomed to presence of
human observers. Territories were mapped by
watching the shrikes’ movements in their home
ranges and by watching aggressive behavior
towards other shrikes; temitory size was calculat-
ed as the minimum polygon bounded by defended
points of the habitat (Yosef and Grubb 1992).

Nests were located during the breeding season
and on each visit were checked to ascertain stage
of the reproductive cycle (egg laying, incubation,
nestling, fledging). We gathered data for 34 nests
(23 in 2007, 1 1 in 2008) and calculated their
probability of survival from incubation to fledging
using Mayfield’s method (Mayfield 1961, 1975).
The average duration of incubation and nestling
days were from Yosef (1992). We recorded the
height of the nest above ground, the species of
tree in which the nest was built, and its condition
(dry with no foliage vs. green). Means are
expressed (± SD), but groups were compared
using Mann Whitney-f/ or y- tests. We chose
P < 0.05 as the minimum acceptable level of
significance.

RESULTS

Breeding Season and Nesting Attempts. — ^The
breeding season lasted from late Eebruary until

late June with peak breeding activity in March
(Fig. 1.). Three pairs had second attempts after
successful first broods in 2007. None of the.se
attempts fledged any young. None of the pairs in
2008 had second nesting attempts even after
having successfully fledged young from the first
brood (Table I). Three nesting attempts were
made following nest failure, and one female
nested four times with three different males, but
only the fourth attempt was successful (Table 1 ).

We found 34 nests during the two breeding
seasons. There was a significant decline in nesting
attempts between 2007 and 2008 (23 vs. 1 1 ; =

4.24, df = \, P = 0.04; Table 1). We found a
significant correlation between number of breed-
ing pairs and amount of precipitation between
2007 and 2009 (Spearman correlation /y = 1.00,
P < 0.0001, Fig. 2).

Nest Construction. — ^Thirty-three (97%) nests
were in Acacia trees {Acacia raddiana and A.
tortilis) and one (2.9%) was in a Commicarpus
plumhagineus bush. Twenty-six (76.5%) nests
were in green trees and eight (23.5%) in dry trees
{y~ = 9.529, df = 1, = 0.002, n = 34). Nests in

dry trees may have been more likely to fail than
those in trees with green foliage (Mann Whitney-
U z = —1.445, P — 0.09, n — 34). Average nest
height was 1.86 ± 0.64 m (range 0.47-3.37). Nest
height did not affect breeding success (Spearman
correlation P = 0.49, n = 34 for clutch size, 0.43
for hatching success, 0.47 for fledging success)

Clutch Size and Incubation. — The average
clutch size was 3.44 ± 0.89 eggs per nest (2-5)
for the two breeding seasons and did not differ
between years (Mann Whitney-U z = —0.59, P =
0.59, n = 28; Table 1 ). Only females brooded
clutches while males foraged and brought food to
the female at or near the nest. Males also guarded
the area sim'ounding the nest tree.

Hatching Success and Nestlings. — ^The average
number of eggs that hatched per nest was 2.14 ±
1.28 (0-4) for all nests and did not differ
significantly between years. Nestling success per
nest, for pairs that fledged at least one young per
nesting attempt, was 2.6 ± 1 .03 for the two
breeding seasons and did not differ between years
(Mann Whitney-U z = — 1.2, P = 0.23, n = 29)
(Table 1). At least one nestling fledged from
20 (58.8%) nests during the entire study. The
average number of fledglings per nest was 1.53
± 1.48 (0-4) and did not differ between years
(Mann Whitney-U z = —0.16, P = 0.89, /? = 17
nests).

336

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

FIG. 1. Average reproductive activity (nests/month) of Southern Grey Shrikes for 2007-2009 in the Shezat Nature
Reserve, Arava Valley, Israel.

Fledglings. — Young remained in the nest tree
for 2-5 days post-fledging and both parents cared
for them. The young remained with the parents for
up to 6 weeks after fledging. Only five (6.3%) of
79 banded nestlings remained in the study area
after leaving their parents, and only one (1.3%)
fledgling from 2007 bred in the study area in
2008. Both parents cared for the fledglings in the
first 3 days after they started to follow the parents
in the territory but subsequently the parents
separated. In three ca.ses (17.6%), females stopped
caring for fledglings and started incubating a
second clutch. In six other cases (35.4%), females
left the breeding area and left the males to care tor
the fledglings. In five cases (29.4%) both male
and female cared for the fledglings for the entire
post-fledging period and, in three cases (17.6%),

all fledglings were lost after a few days and the
parents separated.

Nest Survival Rate. — Six of 34 nests watched
during the incubation period failed while being
exposed for 298.5 days for a daily mortality rate
of 2% per nest day. Daily nest survival rate was
98%. The survival rate for the entire period was
71%. Seventy-three (74.5%) of 98 eggs laid in 28
nests that succeeded in hatching at least one egg,
hatched. Eleven (39.2%) of 28 nests watched
during the nestling period failed while being
exposed for 408.5 days for a daily mortality rate
of 2.3% per nest day. Daily nest survival rate was
97.3%. The survival rate for the entire period was
62.5%. Probability of survival of a nest from start
of incubation until the young fledged was 0.71 X
0.745 X 0.625 = 0.329 (i.e., the probability that

TABLE 1. Between season variances (.v ± SD) in the nesting ecology ot the Southern Grey Shrike in Shezaf Nature

Reserve, Israel.

Breeding parameter

Breeding pairs
Second brood due to failure
Second brood after success
Breeding attempts
Average clutch size
Average nestlings per nest
Fledging success

2007

16

4

3

23

3.52 ± 0.89
2..50 ± 0.78
1.17 ± 1.30

2008

Statistic.s

9
2
0
1 1

3.27 ± 0.90
2.8 ± 1.03
1.36 ± 1.39

f „ = 34, f = 1,96. df = \. P = 0.16
Insufficient data
Insufficient data

,j = 34, f = 4.24, df = \, P = 0.04
Mann Whitney-(7 n = 28, z = -0.59, P = 0.59
Mann Whitney-U /; = 29 z = — 1 .2, P = 0.23
Mann Whitney-U n = M,?.. = —0.16, P = 0.89

Kcynaii am/ YoscJ • BRCHDING ()!■ SOUTHERN GREY SHRIKES

337

FIG. 2. Number of breeding pairs of Southern Grey Shrikes in Shezaf Nature Reserve, Arava Valley, Israel as a
function of annual precipitation (line) between 2007 and 2009,

any nest would fledge at least one young was
32.9%).

DISCUSSION

Our study evaluated the hypothesis that we
would find an effect of aridity on reproductive
effort and success of the Southern Grey Shrike.
Our results are consistent with our predictions
that, because of the reduced prey base in deserts,
the number of clutches laid by a pair in the season
would be low because of exposure and lack of
appropriate breeding sites. However, a longer and
more detailed study is required to understand the
ecological constraints that influence Southern
Grey Shrike breeding capability in the desert
and their ability to persist in this harsh environ-
ment.

The results of the 2007, 2008, and 2009
breeding seasons reflect the combined impact of
arid habitats and multi-year drought on the
breeding biology of the Southern Grey Shrike at
the Shezaf Nature Reserve. Nest survival rate was
higher in our study during the incubation period
than during the nestling period. This supports our
initial prediction that harsh conditions in the
Arava affect the breeding success of shrikes.
Differences between the incubation and brooding
periods can be explained by the need for more
food during rearing of the nestlings, a need that

was harder to achieve the closer the season was to
the summer when prey availability decreases
greatly in the desert.

We found no between-year variations in start or
length of the breeding season. These findings
suggest that, unlike many other avian species.
Southern Grey Shrikes do not rely on rain as a
stimulus for the onset or end of the breeding
season. Differences among regional studies in
length of the breeding period (Table 2) also
cannot be explained by rainfall. This conclusion
does not agree with Illera and Diaz (2006) and
Mondajem and Bamford (2009), who suggested
that rainfall was the cue that stimulated repro-
ductive activities in birds living in unpredictable
environments. Lloyd (1999), however, found rain
as a stimulus for breeding activity for only some
of the birds of the South African arid zones. We
believe Southern Grey Shrikes do not use rainfall
as a stimulus for timing and duration of the
breeding season because of the wide range of
habitats they occupy. Southern Grey Shrikes
occur in many habitats that differ in their
predictability. Thus, rainfall cannot be a reliable
cue for the onset of the breeding season.

The main between-year differences in our study
were the number of breeding pairs at the study site
and their breeding attempts. Some of the pairs in
2007 attempted second broods after success but

338

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122. No. 2, June 2010

TABLE 2. Life history variances (.v ± SD) in the current study in comparison to other reproductive studies of Southern
Grey Shrike.

Current study

Budden and Wrislit
2000

Inbar 1995

Yosef 1992

Campos el al.
2000

Lepley et al. 2000

Location

Hazeva. Israel

Hazeva. Lsrael

Ml. Hernion. Israel

Negev Desert, Israel

Northern Spain

Caru-Seche. France

Avg. rainfall (mm)

30

30

1,500

90

325

>500

Breeding season

Feb-Jun

Feb-Jun

Unavailable

Jan-Aug

Mar-Jul

Mar-Jun

Max. # of broods

2

2

Unavailable

4

1

1

Avg. clutch size

3.44 ± 0.89

3.9 ± 0.57

4.6 ± 0.5

5.8 ± 0.6

5.7 ± 0.4

5.2 ± 1.1

Avg. hatchlings/nest

2.14 ± 1.28

1.09

Unavailable

4.7 ± 2.1

2.1 ± 2.4

3.1

Avg. fledglings/nest

1.53 ± 1.48

Unavailable

Unavailable

3.6 ± 1.4

Unavailable

1.54 ± 1.9

all attempts failed. In contrast, none of the pairs in
2008 and 2009 had any signs of trying a second
brood. This difference indicates that number of
breeding attempts was most probably related to
food abundance. Bird species that Hedge more
than one brood decrease in breeding success as the
breeding season advances (Verhulst and Nilsson
2008). This decrease in breeding success occurs
mainly because of climatic factors that affect food
availability (Takagi 2001, Verhulst and Nilsson
2008, Yosef and Zduniak 2008, Mondajem and
Bamford 2009). It is possible that food was so
scarce during 2008 that the pairs could invest in
only one brood. Southern Grey Shrikes can fledge
up to four broods in less arid environments, where
food is more abundant and the duration of the
breeding season is longer (Table 2).

Far fewer pairs live and breed in the Arava
Valley than in the Sede Boker area (Shirihai
1996). We suggest that conditions in the Arava
are marginal for the Southern Grey Shrike and are
possibly a limiting factor to its distribution. The
severe droughts in the past few years have
affected the numbers of breeding pairs in the
Arava (Fig. 2). The number of pairs declined
sharply between years (Table 1 ), reaching a low
of three in 2009 (Fig. 2), and it appears shrikes
a.ssess habitat quality before they initiate breeding
(Yosef 2008). It is possible that some relocate to
other breeding habitats if food availability is low.
It is also possible that some individuals that nested
during 2007 did not survive the harsh summer or
were unable to continue to defend a good territory
due to drought.

George et al. (1992) reported that drought
.severely affected total bird density and breeding
success. They showed that some species recov-
ered numbers in the year alter the drought but
others did not. Bolger et al. (2005) found that an
extreme drought dramatically reduced both the

number of pairs breeding and breeding success in
four species of passerine birds in southern
California.

The Southern Grey Shrike population in Shezaf
Nature Reserve has declined drastically in the past
3 years, due to a severe drought. It will be of
interest to ascertain whether the population can
persist and will recover in the future if there are
good rains.

ACKNOWLEDGMENTS

We thank Amotz Zahavi, Ido Izhaki. Moshe Inbar, Ronni
Ostreicher, and the staff of Hazeva Field Study Center; and
Harel Ben-Shachar of the Israel Nature Re.serves and Parks
Authority. We thank Tom Cade and an anonymous
reviewer for improving an earlier draft of the manuscript.

LITERATURE CITED

Anava, a.. M. Kam, a. Shkolnik, and A. A. Degen.
2000. Seasonal field metabolic rate and dietary
intake in Arabian Babblers (Twdoides squamiceps)
inhabiting extreme desert. Functional Ecology 14:
607-613.

Bolger. D. T., M. A. Patten, and D. C. Bo.stock. 2005.
Avian reproductive failure in response to an extreme
climatic event. Oecologia 142: 398^06.

Budden, a. and J. Wright. 2000. Nestling diet, chick
growth and breeding success in the Southern Grey
Shrike {Lanius nicridionalis). The Ring 22: 165-172.
Campos, F., F. Guiterrez-Corchero, and M. A. Her-
nandez. 2007. Fenologia reproductora y xito repro-
ductor del alcaud: n real, Lunins meridiomdis, en zonas
agricolas del Norte de Espania. Ecologia 21 : 167-174.
George, T. L., A. C. Fowler. R. L. Knight, and L. C.
McEwen. 1992. Impact of severe drought on grassland
birds in western North Dakota. Ecological Applica-
tions 2: 275-284

CiOi.DREiCH. Y. AND O. Karni. 2001. Climate' and
precipitation regime in the Arava Valley, Israel. Israel
Journal of Earth Sciences 50: 53-59.

Ii.lera, J. C. and M. Diaz. 2006. Reproduction in an
endemic bird of a semi arid island: a food mediated
process. Journal of Avian Biology 37: 447—456.

Keyiniii and Yosef - BREEDING OE SOUTHERN GREY SHRIKES

339

Immelmann, K, 1973. Role of the environment in
reproduction as source ol "predictive” information.
Pages 121-147 in Breeding biology of birds (D. S.
Earner, Editor). National Academy of Sciences,
Washington, D.C., USA.

Inbar, R. 199.3. Shrike nesting in Mount Hermon, Israel.
Proceedings Western Eoundation of Vertebrate Zool-
ogy 6: 215-217.

Lefranc, N. and T. Worfolk. 1997. Southern Grey
Shrike. Pages 140-144 in A guide to the shrikes of the
world. Yale University Press, New Haven, Connecti-
cut, USA.

LEPLEY, M., C. P. GUtLLAUME, A. NEWTON, AND M.
TttEVENOT. 2000. Breeding biology of Southern Grey
Shrike {Lanins meridionalis) in Crau-Seche (Bouches
du Rhone, Prance). Alauda 68: 35—43.

Lloyd, P. 1999. Rainfall as a breeding stimulus and clutch
size determinant in South African arid-zone birds. Ibis
141: 637-643.

MAYFtELD, H. 1961. Nesting success calculated from
exposure. Wilson Bulletin 73: 255-261.

Mayfield, H. 1975. Suggestions for calculating nest
success. Wilson Bulletin 87: 456^66.

Mondajem, a. and a. Bamford. 2009. Influence of
rainfall on timing and success of reproduction in
Marabu Storks Lentoplilos crumeniferus. Ibis 151:
344-351.

Morrison, S. A. and D. T. Bolger. 2002. Variation in a
sparrow’s reproductive success with rainfall: food and
predator-mediated processes. Oecologia 133: 315-324.

Noy-Meir. 1. 1973. Desert ecosystems: environment and
producers. Annual Review of Ecology and Systematics
4: 25-5 1 .

Ricklefs, R. E. and M. Wikelski. 2002. The physiology/
life history nexus. Trends in Ecology and Evolution
17: 462^68.

Shirihai, H. 1996. The birds of ksrael. Academic Press,
London, United Kingdom.

Smi th, S. M. 1973. Eood manipulalion by young passerines
and the possible evolutionary history of impaling by
shrikes. Wilson Bulletin 85: 318-322.

Takagi, M, 2001. Some effect of inclement weather
conditions on the survival and condition of Bull-
headed Shrike nestlings. Ecological Research 16: 55-
63.

TIELEMAN, B. I., J. B. WtLLIAMS, AND G. H. Vl.SSER. 2004.
Energy and water budgets of larks in a life history
perspective: parental effort varies with aridity. Ecol-
ogy 85: 1399-1410.

UNESCO. 1977. Map of world distribution of arid
lands. MAB Technician Note 7. UNESCO, Paris,
Prance.

Verhulst, S. AND J.-A. Nilsson. 2008. The timing of
bird’s breeding seasons: a review of experiments that
manipulated timing of breeding. Philosophical Trans-
actions of the Royal Society London, Series B 363:
399-410.

Yosef, R. 1992. Prom nest building to fledging of young in
Great Grey Shrikes (Lanins excubitor) at Sede Boqer,
Israel. Journal of Ornithology 133: 279-285.

Yosef, R. 2008. Laniidae. Pages 732-796 in Handbook of
the birds of the world. Volume 13 (J. del Hoyo, A.
Elliott, and D. A. Christie, Editors). Lynx Edicions,
Barcelona, Spain.

Yosef, R. and T. C. Grubb Jr, 1992. Territory size
influences nutritional condition in nonbreeding Log-
gerhead Shrikes {Lanins Indovicianns) a ptilochronol-
ogy approach. Conservation Biology 6: 447-449.

Yosef, R. and P. E. Lohrer. 1992. A composite treadle/
Bal-Chatri trap for Loggerhead Shrikes. Wildlife
Society Bulletin 20: 116-118.

Yosef, R. and P. Zduniak. 2008. Variation in clutch size,
egg size variability and reproductive output in the
Desert Pinch (Rhodospiza obsolete). Journal of Arid
Environments 72: 1631-1635.

The Wilson Journal of Ornithology 122(2):340-345, 2010

HOME RANGE SIZES AND HABITAT USE OE NELSON’S AND

SALTMARSH SPARROWS

W. GREGORY SHRIVERT^' THOMAS P. HODGMAN,-
JAMES P. GIBBS,' AND PETER D. VICKERY'

ABSTRACT. — Nelson’s (Ammodramus nelsoni) and Saltmarsh (A. caudacutus) sparrows are sympatric breeders in tidal
marshes of the southern Gulf of Maine. These sparrows hybridize, have different mating strategies, and males do not defend
territories or provide parental care. We estimated and compared core area sizes, home range sizes, and habitat u.se between
species and between males and females. We radio-marked 140 sparrows (63 Nelson’s and 77 Saltmarsh sparrows) during
three breeding seasons (1999-2001) at Scarborough Marsh, Maine, USA. Home ranges of male A. nelsoni were 2.3 times
larger (± SE) (1 19.68 ± 19.43 ha) than those of male A. caiidacutiis (52.85 ± 8.68 ha). Home range sizes of female
Nelson’s and female Saltmarsh sparrows did not differ from each other (female Nelson’s home range = 43.58 ± 13.10 ha;
female Saltmarsh home range = 27.8 1 ± 6.3 ha). More than 40% of male and 1 8% of female home ranges had two discrete
core areas and, in most instances, each core area corresponded to a separate lunar cycle. We suggest that differences in
mating strategies, densities, and adaptation to nesting in tidal marshes explain the larger home range estimates for male
Nelson’s Sparrows. Female and male Nelson’s Sparrows' home ranges had more Spartina alterniflora cover and female
Saltmarsh Sparrows’ home ranges had greater Juncus gerardii cover than random locations. Home ranges of female
Saltmarsh Sparrows had less Spartina alterniflora cover and more Juncus gerardii cover than female Nelson’s Sparrows.
We did not detect any differences in vegetation variables between male Saltmarsh and male Nelson's span'ow home ranges.
Received 1 1 September 2009. Accepted 10 January 2010.

Home range and resource use can occur at
multiple spatial and temporal scales that may
change during the annual cycle for most bird
species. Movement patterns and space require-
ments are important parameters to consider when
attempting to understand populations and how
they might respond to different management
actions or habitat change scenarios (Barbour and
Litvaitis 1993). Home range studies can provide
valuable information about habitat selection, use
of space, and in making predictions about how
environmental change can alter a species’ avail-
able habitat (White and Garrott 1990).

Nelson’s {Antmoclrcimns nelsoni previously
known as Nelson’s Sharp-tailed Sparrow [AOU
1998, Chesser et al. 2009] ) and Saltmarsh (A.
caiulacutus previously known as Saltmarsh Sharp-
tailed Sparrow [AOU 1998, Chesser et al. 2009])
sparrows are listed as conservation priorities due
to our limited understanding of their basic

' .State University of New York. College of Environmen-
tal Science and Forestry, I Forestry Drive, Syracuse, NY
13210. USA.

^Bird Group. Maine Department of Inland Fisheries and
Wildlife, 650 State Street. Bangor, ME 04401, USA.

'Center for Ecological Research, P. O. Box 127,
Richmond. ME 04357, USA.

■‘Current address: Department of Entomology and
Wildlife Ecology. University of Delaware, Newark, DE
19717, USA.

^Corresponding author; e-mail: gshriver@udel.edu

demographic parameters and habitat requirements
(Rich et al. 2004). Previous re.search into the
breeding ecology of these sparrows has described
responses to landscape features (Benoit and
Askins 1999, Shriver et al. 2004), dispersal
patterns and survival rates (DiQuinzio et al.
2001), and second (Nocera et al. 2007) and third
(Woolfenden 1956, Greenlaw and Rising 1994)
order habitat selection (Johnson 1980). These
studies have increased our understanding of the
breeding ecology for these taxa from Long Island,
New York, USA (Woolfenden 1956, Greenlaw
and Rising 1994) to the Canadian Maritimes
(Nocera et al. 2007), supporting the species split
(AOU 1995). These aid in making predictions
about how changes to available habitat may affect
the persistence of populations of these .species.
Our objectives were to; (1) use radiotelemetry to
estimate species- and male/female-specific home
range size, and (2) resource use to further our
understanding of how these non-territorial and
promi.scuous (Greenlaw and Rising 1994) species
use their breeding habitat.

METHODS

We used mist nests to capture sparrows at
Scarborough Marsh, Maine, USA (43° 34' N, 70°
22' W) during three breeding seasons (20 May-15
Aug) from 1999 to 2001. All sparrows were
individually marked with a U.S. Geological
Survey band and 3 color bands. We conducted

340

S/irivcr el a/. • NELSON'S AND SALTMARSIl SPARROW HOME RANGE SIZES

341

our study on a lOO-ha portion of this l,5()()-ha salt
marsh. Spurt ina patens, S. alterniflora. Distich I is
spicata, Jiinciis gerartlii, and salt pans (depres-
sions with little or no vegetation and generally
high salinity levels) dominated this area of
Scarborough Marsh.

We used radiotelemetry to relocate individuals
in the marsh by homing during each of three
breeding seasons. We attached 1-g radio trans-
mitters {<5% body mass, Holohil Systems Ltd.,
Caip, ON, Canada) to adult males and females of
both species using the thigh-harness method
(Rappole and Tipton 1991). We assessed the fit
of transmitters using a 30 X 25 X 25 cm, slotted,
clear-plastic cage to view span'ows after trans-
mitter attachment; if a harness and transmitter did
not appear to fit properly, we removed it and
released the sparrow. We attempted to locate
radio-marked birds each day by homing using
a receiver (Communication Specialists Inc.,
Orange, CA, USA) and 4-element, hand-held
antenna. We recaptured as many birds as possible
at the end of the season to remove transmitters.

The surface complexity and zonation of the
marsh vegetation in our study area facilitated
accurate plotting of sparrow locations. We used a
high resolution, color-infrared, aerial photograph
to record locations of individual spaiTows each
field day until batteries in transmitters failed or
transmitters were removed. This geo-referenced
photograph allowed us to plot locations of
sparrows to within 3 m. We overlaid a transpar-
ency on the photograph each field day, and
marked date, sparrow locations, transmitter fre-
quency, band number, species, and male or female
on the overlay. We examined the dominant
vegetation within a 1-m-radius circle of each
sparrow location and categorized vegetation as:
(1) Spartina alterniflora, (2) S. patens, (3) S.
patens! S. alterniflora, (4) Jnncns gerarclii, (5)
Distichlis spicatalS. patens, (6) un-vegetated salt
pan, or (7) vegetated salt pan. We used the
proportion of each vegetation type across all
locations for a specific sparrow in the habitat use
analysis. We also sampled 174 randomly located
1-m-radius circular plots to estimate the avail-
ability of each vegetation type within the study
area. We used the geo-referenced photograph as
the base layer to digitize these data using
ArcView 3.3 and used the animal movement
extension (Hooge and Eichenlaub 1997) to
estimate 50% (hereafter, core area) and 95%
(hereafter, home range) fixed kernel home ranges.

We used only those individuals with >20
locations for all analyses. We used univariate
ANOVA to test for differences in core area and
home range sizes between species and males/
females (Zar 1999). We calculated the number of
sparrows that had two di.screte core areas (i.e., two
50% kernel home ranges) and measured the
distance between the centroids for each core area.
We used univariate ANOVA with Tukey HSD
post-hoc tests to explore differences in vegetative
cover between randomly located points, and
locations of males and females for each species,
and between the two species. We used a
Kolmogorov-Smirnov goodness of fit test to
examine if each variable fit a normal distribution
(Zar 1999). All statistical tests were conducted in
SPSS (Version 16.0) with alpha = 0.05.

RESULTS

We estimated core area and home range size for
63 radio-marked Nelson’s and 77 radio-marked
Saltmarsh spaiTows at Scarborough Marsh, Maine
in 1 999-200 1 . Core area and home range sizes of
males were larger than for females, both for
Nelson’s (core area: F|_62 = 8.5, P = 0.005; home
range: Fi,e2 — ll-C P = 0.001) and Saltmarsh
sparrows (core area: F\jf, = 4.2, P = 0.045; home
range: F] 75 = 5.7, P = 0.019) (Table 1). Both
core area and home range sizes of male Nelson’s
Sparrows were larger than those of male Salt-
marsh Sparrows (core area: F| 51 = 4.3, P =
0.043; home range: F1.51 = 10.6, P = 0.002)
(Table 1). We found no difference in core area or
home range sizes of female Nelson’s and female
Saltmarsh sparrows (core area: F| 77 = 0.18, P =
0.674; home range: F\J■^ = 1.1, P — 0.289)
(Table 1). Distances between core areas, for those
individuals that maintained two core areas, were
greater for male Nelson’s than for male Saltmarsh
spaiTows (F| 54 = 4.5, P = 0.038), and were
greater for female Nelson’s compared to female
Saltmarsh sparrows (F| 34 = 8.0, P = 0.008)
(Table 1).

Female and male Nelson’s Sparrows' home
ranges had more Spartina alterniflora cover than
random locations (F| 237 = 16.2, P < 0.001) but
did not differ between each other (F = 0.142).
The amount of S. patens cover differed among
Nelson’s Sparrows and random locations (F| 237
= 6.7, P = 0.002) with less S. patens in the home
ranges of male Nelson’s Sparrows (F < 0.001,
Table 2). Spartina patenslS. alterniflora cover
differed among Nelson's Span'ows and random

342

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

TABLE 1 . Characteristics (.v ± SE) of radio-marked Nelson’s and Saltmarsh sparrows fixed kernel 50% core areas and
95% home ranges at Scarborough Marsh, Maine, USA, 1999 to 2001.

Nelson’s Sparrow

Saltmarsh Sparrow

Variable

Females Ui = 34)

Males (« = 29)

Females (n = 44)

Males (H = 33)

Core area, ha

Home range, ha

Home ranges with two cores

Core area distance, m

6.15 ± 1.66
43.58 ± 13.10

9 (27%)

161.40 ± 38.85

18.42 ± 3.91

119.68 ± 19.4

13 (45%)
204.28 ± 25.61

5.25 ± 1.38
27.81 ± 6.30

8 (18%)

77.56 ±13.16

9.64 ± 1.67
52.85 ± 8.68

14 (42%)
158.99 ± 17.55

locations {F^ 2^l — 5.2, P = 0.006) with less
Spartina patens! S. alterniflora in male Nelson’s
Sparrow home ranges (P = 0.018, Table 2). Both
male and female Nelson’s Sparrow home ranges
had more vegetated pan cover than random
locations (^1.237 = 38.1, P < 0.001, Table 2).

Spartina alterniflora cover differed among
Saltmarsh Sparrow home ranges and random
locations {F\,2S2 — 8.4, P < 0.001, Table 2) with
greater cover in male Saltmarsh Sparrow home
ranges than in random locations {P < 0.001).
Spartina patens cover differed among Saltmarsh
Sparrow home ranges and random locations
(F\.252 — 6.6, P = 0.002, Table 2) with less
cover in home ranges of male Saltmarsh Sparrows
than in random locations (P = 0.004). Spartina
patensIS. alterniflora cover differed among Salt-
marsh Sparrow home ranges and random locations
(^1.252 = 5.5, P — 0.005, Table 2) with less cover
in home ranges of male (P = 0.039) and female
{P = 0.004) Saltmarsh Sparrows than in random
locations (Table 2). Juncus gerardii cover dif-
fered among Saltmarsh Sparrow home ranges and
random locations (^1252 = 7.1, P = 0.001,
Table 2) with more cover in female Saltmarsh
Sparrow home ranges than in random locations [P
= 0.001, Table 2). Both male and female Salt-
marsh Sparrow home ranges had more vegetated

pan cover than random locations (^1,237 = 48.2, P
< 0.001, Table 2). Female Saltmarsh Sparrow
home ranges had more Spartina alterniflora cover
(7^1,79 = 4.5, P = 0.038) and more Juncus
gerardii cover (F] 79 = 4.6, P = 0.035) than
female Nelson’s Span'ow home ranges (Table 2).
We did not detect any differences in vegetation
variables between home ranges of male Saltmarsh
and male Nelson’s span'ows (Table 2).

DISCUSSION

Home range sizes of both species were 2-10
times larger than those reported for other
emberizid span'ows (Patterson and Petrinovich
1978, Reed 1985, Delany et al. 1995, Bechtoldt
and Stouffer 2005, Jones and Bock 2005, Cox and
Jones 2007). Male/female-specific home range
size was approximately twice as large for
Nelson’s than for Saltmarsh sparrows. Male
Nelson’s Sparrow home range sizes were three
times larger than those of females. All home
ranges for both species and males and females
overlapped within the 100-ha study area. Home
range sizes for male Saltmarsh Spairows were
twice those of females, but both were smaller than
home ranges of male Nelson’s SpaiTOws. Wool-
fenden (1956) estimated male Saltmarsh Sparrow
home ranges to be 3^ times larger ( 1 .2-1.6 ha for

TABLE 2. Proportion ± SE of .seven vegetation types within Nelson’s and Saltmarsh sparrow home ranges and at
random points at Scarborough Marsh. Maine, 1999-2001.

Nelson's Sparrow

Saltmarsh Sparrow

Vegetation type

Females

Males

Females

Males

Random

Spartina alterniflora
S. patens

S. patensIS. alterniflora
.innens gerardii
Disticlilis spicata
Un-vegetated Pan
Vegetated Pan

0.188 ± 0.021
0.276 ± 0.023
0.104 ± 0.02!
0. 1 86 ± 0.025
0.012 ± 0.007
0.005 ± 0.002
0.188 ± 0.021

0.288 ± 0.041
0.178 ± 0.021
0.044 ± 0.010
0.214 ± 0.030
0.006 ± 0.005
0.013 ± 0.005
0.203 ± 0.032

0.132 ± 0.017
0.288 ± 0.029
0.087 ± 0.013
0.283 ± 0.035
0.006 ± 0.003
0.004 ± 0.001
0.155 ± 0.021

0.216 ± 0.032
0.201 ± 0.025
0.081 ± 0.029
0.216 ± 0.042
0.001 ± 0.001
0.015 ± 0.006
0.235 ± 0.030

0.069 ± 0.017
0.435 ± 0.031
0.239 ± 0.030
0.1 15 ± 0.022
0.008 ± 0.006
0.041 ± 0.014
0.019 ± 0.009

S/irivcr et al. • NELSON'S AND SALTMARSII SPARROW HOME RANGE SIZES 343

males vs. 0.4 ha for females) than those of
females. Greenlaw and Rising (1994) also report-
ed a 4-fold difference in male (4.3 ha) versus
female (1.1 ha) home range size from studies of
Saltmarsh Spanows. It is difficult to compare our
findings with other home range estimates given
our use of radiotelemetry. The non-territorial
mating system of these spaiTows (Woolfenden
1956, MuiTay 1969, Post and Greenlaw 1982,
Greenlaw 1993, Shriver et al. 2007) generally
results in more time spent by males searching for
receptive females or females soliciting copulatory
chases (Greenlaw and Rising 1994, Shriver et al.
2007). This, coupled with a lack of territorial
boundaries, partially explains the large home
ranges observed for both taxa in our study.

Male Nelson’s SpaiTows had 33% larger home
ranges than male Saltmarsh Sparrows which may
be explained by differences in mating systems and
breeding densities between the two .species. Male
Nelson’s Sparrows spend a substantial amount of
time guarding females (up to 43 hrs with the same
female, Shriver et al. 2007) while male Saltmarsh
Sparrows practice “scramble competition polyg-
yny’’ (Greenlaw and Rising 1994) where small
groups (3-6) of males chase receptive females,
copulate, and then usually perch to search for
another receptive female (W. G. Shriver et al.,
unpubl. data). Male Saltmarsh Sparrows have not
been observed guarding females (Shriver et al.
2007). Twenty-seven percent of female Nelson’s
Sparrows maintained two distinct core areas and
the distance between these core areas was two
times greater than for female Saltmarsh Sparrows.
This may cause male Nelson’s Sparrows to travel
farther to encounter females compared to male
Saltmarsh Sparrows. We estimated there were
three times more female Saltmarsh Sparrows than
female Nelson’s Sparrows (1998 to 2001; W. G.
Shriver et al, unpubl. data) in our study area. We
suggest, based on our banding data from this site,
greater dispersion of female Nelson’s Sparrows
(especially in lower density areas), greater
distances between female Nelson’s Sparrows'
core areas, and the mate guarding behavior of
male Nelson’s Sparrows, at least partially explain
differences in home range sizes for these species.

Other aspects of sharp-tailed sparrow breeding
ecology appear to be synchronous with new moon
tidal events (Shriver et al. 2007) and we reviewed
the timing of the maintenance of each core area.
More than half of the sparrows with two discrete
core areas did not use them simultaneously, but

instead core area use was timed with separate
lunar cycles. Most nests typically flood during
spring tides (Gjerdrum et al. 2005, Shriver et al.
2007), and it is apparent that re-nesting female
sparrows moved within the study area to construct
a second nest (during the subsequent lunar event).
The large ranges (and multiple core areas) of
males are likely a consequence of spatial shifts
exhibited by females. Female Nelson’s Spairows
do not re-nest immediately after a Hooding event
(Shriver et al. 2007), and appear to move further
to re-nest than female Saltmarsh Sparrows. Home
ranges of male Nelson’s Sparrows, therefore, are
larger and distance between cores further as they
attempt to maintain close proximity to females
during re-nesting attempts.

Patterns in habitat use for these sparrows
differed from studies of nest site selection
(DiQuinzio et al. 2002, Gjerdrum et al. 2005,
Shriver et al. 2007). We found spaiTows more
frequently in Spartina alteniiflora and J uncus
gerardii cover and less often in S. patens cover
than was available in the study area. Vegetated
pan cover was also more common within sparrow
home ranges than was available in the study area.
Saltmarsh Sparrow female home ranges had the
greatest proportion of Juncus gerardii compared
to random locations or other sparrow home
ranges. Gjerdrum et al. (2008) also found Juncus
gerardii was the key predictor variable in
explaining variation in number of nests and
fledgling production.

Tidal marshes are limited globally in extent to
45,000 km" with 15,500 km" (34%) occumng
along the Atlantic and Gulf coasts of North
America (Bertness 1999, Mendelssohn and Mckee
2000, Greenberg et al. 2006). The coastlines of
North America support the greatest number of
endemic tidal marsh vertebrate species, including
Saltmarsh Sparrows (Greenberg et al. 2006),
increasing the global significance of these eco-
systems. Resource professionals engaged in tidal
marsh conservation have reason to be concerned
about persistence of tidal marsh vertebrate .species
given their geographically limited habitat and
vulnerability to habitat loss and fragmentation
resulting from sea level rise. Even moderate
estimates (30 cm rise) of .sea level rise had
negative effects on the estimated persistence of
Seaside Sparrows (Ainniodranius maritimus) in
Connecticut (Shriver and Gibbs 2004). Our study
was conducted in one of the larger marshes in the
Gulf of Maine where these species co-occur, near

344

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 2. June 2010

the northern limit of Saltmarsh Sparrows and the
southern limit of Nelson’s Sparrows. These
conditions certainly influence the spatial use
patterns of these sparrows and our estimates of
home range size for these taxa may not apply to
smaller sites where these sparrows are relatively
common and allopatric. Future home range
studies at other sites using similar techniques
would be valuable to examine the variation in
these parameters given different marsh sizes and
location within the range of each species. Space
requirements and habitat use information similar
to those pre.sented here and elsewhere (Gjerdrum
et al. 2008) could then be used to model the
effects of climate change on these species which
would aid in conservation planning efforts.

ACKNOWLEDGMENTS

We thank J. J. Bacon, M. J. Baumflek, G. C. Hall, C. N.
Jacques, B. J. MacCulloch, A. B. Sacerdote, and M. J.
McElroy for assistance in the field. Jan Taylor (USFWS,
Region 5) provided early and continued support for this
project. This research was funded in part by The Maine
Outdoor Heritage Fund, Maine Department of Inland
Fisheries and Wildlife Federal Aid in Sport Fish and
Wildlife Restoration - Project W82R. U.S. Fish and
Wildlife Service, and a National Fstuarine Research
Reserve Graduate Fellowship awarded to WGS. J. S,
Greenlaw, C. S. Flphick, J. J. Nocera, and P. W. C. Paton
provided valuable comments on early versions of this
manu-script and are warmly acknowledged.

LITERATURE CITED

American Ornithologists’ Union (AOU). 1995. Fortieth
supplement to the American Ornithologists' Union
checklist of North American birds. Auk 1 12:819-8.50.
American Ornithologi.sts’ Union (AOU). 1998. Check-
list of North American birds. Seventh Edition.
American Ornithologi.sts’ Union, Washington, D.C.,
USA.

Barbour. M. S. and J. A. Litvaitis. 1993. Niche
dimensions of New England cottontails in relation to
habitat patch size. Oecologia 95: 321-327.
Bechtoldt, C. L. and P. C. Stouffer. 2005. Home-range
size, response to fire, and habitat preferences of
wintering Henslow’s Sparrows. Wilson Bulletin 117:
21 1-225.

Benoit, L. K. and R. A. Askins. 1999. Impact of the
spread of Phra^mitcs on the distribution of birds in
Connecticut tidal marshes. Wetlands 19: 194-208.
BERTNE.SS. M. D. 1999. The ecology of Atlantic shorelines.
Sinauer Associates Inc., Sunderland, Massachu.setts,
USA.

Chf;,s.ser. R. T., R. C. Bank.s, F. K. Barker, C. Cicero. J.
L. Dunn. A. W. Kratter, I. J. Lovette. P. C.
Ra.smu.s.sen. j. V. Rem.sen Jr., J. D. Rlsing, D. F.
Stotz, and K. Winki;r. 2009. Fiftieth Supplement to

the American Ornithologists’ Union checklist of North
American birds. Auk 126:705-714.

Cox, J. A. AND C. D. Jones. 2007. Home range and survival
characteristics of male Bachman’s Sparrows in an old-
growth forest managed with breeding season burns.
Journal of Field Ornithology 78:263-269.

Delany, M. F., C. T. Moore, and J. D. R. Progulske.
1995. Territory size and movement of Florida
Grasshopper Sparrows. Journal of Field Ornithology
66: 305-309.

DiQuinzio, D. A., P. W. C. Paton, and W. R. Eddleman.

2001. Site fidelity, philopatry, and survival of
promiscuous Saltmarsh Sharp-tailed Sparrows in
Rhode Island. Auk 1 18: 888-899.

DiQuinzio, D. A., P. W. C. Paton, and W. R. Eddleman.

2002. Nesting ecology of Saltmarsh Sharp-tailed
Sparrows in a tidally restricted .salt marsh. Wetlands
22: 179-185.

Gjerdrum, C., C. S. Flphick, and M. Rubega. 2005. Nest
site selection and nesting success in saltmarsh
breeding sparrows: the importance of nest habitat,
timing, and study site differences. Condor 107: 849-
862.

Gjerdrum, C., C. S. Flphick, and M. A. Rubega. 2008.
How well can we model numbers and productivity of
Saltmarsh Sharp-tailed Sparrows (Ammodramus caii-
dacutus) using habitat features? Auk 125: 608-617.

Greenberg, R., J. E. Maldonado, S. Droege, and M. V.
McDonald. 2006. Tidal marshes: a global perspective
on the evolution and con.servation of their terrestrial
vertebrates. BioScience 56: 675-685.

Greenlaw, J. S. 1993. Behavioral and moiphological
diversification in Sharp-tailed Sparrows (Ammodramus
caiidaciilus) of the Atlantic Coast. Auk 1 10: 286-303.

Greenlaw, J. S. and J. D. Rising. 1994. Sharp-tailed
Sparrow (Ammodramus caiidaciitus). The birds of
North America. Number 1 12.

Hooge, P. N. and B. Eichenlaub. 1997. Animal
movement extension to Areview. Version 2.0. USGS,
Ala,ska Science Center-Biological Science Office,
Anchorage, Alaska, USA.

JOHN-SON, D. H. 1980. The comparison of usage and
availability measurements for evaluating resource
preference. Ecology 6 1 : 65-7 1 .

Jones, Z. F. and C. E. Bock. 2005. The Botteri’s Sparrow
and exotic Arizona grasslands: an ecological trap or
habitat regained? Condor 107: 731-741.

Mendelssohn, 1. A. and K. L. Mckee. 2000. Saltmarshes
and mangroves. Pages 501-536 in North American
terrestrial vegetation (M. Barbour and W. D. Billings,
Editors). Cambridge University Press, . Cambridge,
United Kingdom.

Murray Jr., B. G. 1969. A comparative study of Le
Conte’s and Sharp-tailed .sparrows. Auk 86: 199-231.

Nocera, J. J., T. M. Fitzgerald, A. R. Hanson, and G. R.
Milton. 2007. Differential habitat u.se by Acadian
Nelson’s Sharp-tailed Sparrows: implications for
regional con.servation. Journal of Field Ornithology
78: 50-55.

PA iTitR.soN, T. L. AND L. Petrinovici I. 1978. Territory size
in the White-crowned Sparrow (Zonotrichia leu-

Shrivcr el al. • NELSON'S AND SALTMARSII SPARROW HOME RANGE SIZES

345

cophrys): measurement and stability. Condor 80: 97-
98.

PcxsT, W. AND J. S. Greenlaw. 1982. Comparative costs of
promiscuity and monogamy: a test of reproductive
effort theory. Behavioral Ecology and Sociobiology
10: 101-107.

Rappole, J. H. and A. R. Tipton. 1991. New harness
design for attachment of radio transmitters to small
passerines. Journal of Field Ornithology 62: 335-337.
Reed, J. M. 1985. A comparison of the “flush” and spot-
map methods for estimating the size of Vesper
Spanow teiTitories. Journal of Field Ornithology 56:
131-137.

Rich, T. D., C. J. Beardmore, FI. Berlanga, P. J.
Blancher, M. S. W. Bradstreet, G. S. Butcher, D.
W. Demarest, E. H. Dunn, W. C. FIunter, E. E.
Inigo-Elias, j. A. Kennedy, A. M. Martell, A. O.
Panjabi, D. N. Pashley, K. V. Rosenberg, C. M.
Rustay, j. S. Wendt, and T. C. Will. 2004. Partners
in Flight North American Landbird Conservation Plan.
Cornell Laboratory of Ornithology, Ithaca, New York,
USA.

Shriver, W. G. and j. P. Gibbs. 2004. Projected effects of
sea-level rise on the population viability of Seaside

Sparrows (Ammodramus niciritimiis). Pages 397-410 in
Species conservation and management: case studies
(fj. R. Ak9akaya, M. Burgman, O. Kindvall, C. C.
Wood, P. SJogren-Gulve, J. S. fJatfield, and M. A.
McCarthy, Editors). Oxford University Press, Oxford,
United Kingdom.

Shriver, W. G., T. P. Hodgman, J. P. Gibbs, and P. D.
Vickery. 2004. Landscape context influences salt
marsh bird diversity and area requirements in New
England. Biological Conservation 119: 545-553.

Shriver, W. G., P. D. Vickery, T. P. FIodgman, and J. P.
Gibbs. 2007. Flood tides affect breeding ecology of
two sympatric sharp-tailed sparrows. Auk 124: 552-
560.

White, G. C. and R. M. Garrott. 1990. Analysis of
wildlife radio-tracking data. Academic Press, New
York, USA.

Woolfenden, G. E. 1956. Comparative breeding behavior
of Ammospiza caudacuta and A. maritima. University
of Kansas, Publications of the Museum of Natural
History 10: 45-75.

Zar, j. H. 1999. Biostatistical analysis. Fourth Edition.
Prentice Hall, Upper Saddle River, New Jersey,
USA.

The Wiiso)! Journal of Ornithology 122(2):346-353, 2010

ASSESSING GENERALIZED EGG MIMICRY: A QUANTITATIVE
COMPARISON OF EGGS OF BROWN-HEADED COWBIRDS AND

GRASSLAND PASSERINES

DWIGHT R. KLIPPENSTINE' - AND SPENCER G. SEALY'"'

ABSTRACT. — Perceived similarity in appearance of eggs of the Brown-headed Cowbird (Molothrus ater) and several
grassland passerine hosts has been suggested to represent a form of generalized mimicry evolved to impede egg
discrimination. We tested this hypothesis by quantitatively comparing four parameters (ground color, maculation color,
distribution of maculation, and maculation density) among cowbird eggs and those of six grassland passerines known to
preferentially accept cowbird eggs over non-mimetic blue eggs. Cowbird eggs did not significantly differ in three of four
parameters (mean ground color, and color and density of maculation) from the average grassland passerine egg measured
for this community, in southern Saskatchewan, when all grassland passerine eggs were pooled. Cowbird eggs sufficiently
overlapped grassland pas.serine eggs in these three parameters that 88% of cowbird eggs measured were statistically
indistinguishable from among all grassland pas.serine eggs. The frequency at which cowbird eggs were misclassitied as eggs
of each grassland passerine ranged from 8 to 48%, which suggests that a single cowbird egg is capable of mimicking the
eggs of more than one species. These results support the hypothesis that cowbirds have evolved a generalized egg
appearance that mimics the eggs of multiple grassland passerine hosts within a single community. Received 31 October
2008. Accepted 23 October 2009.

Evolution of egg mimicry by obligate brood
parasites is regarded as one of the most common
responses to development of egg discrimination
and ejection by intolerant hosts (Underwood and
Sealy 2002, Kilner 2006). Egg mimicry evolves
because ejection of unlike eggs by hosts provides
a selective advantage to parasitic eggs that match
the clutch in which they are laid (Davies and
Brooke 1991, Soler et al. 2003). The extent of
similarity gradually escalates through a co-
evolutionary race as hosts refine their egg-
discrimination abilities over subsequent genera-
tions (Moksnes et al. 1990, Davies and Brooke
1991). The evolutionary equilibrium hypothesis
posits that whenever the level of mimicry exceeds
a species’ ability to discriminate between eggs,
the host accepts the parasitic egg (Davies and
Brooke 1988, Lotem and Nakamura 1998).
Presumably, acceptance occurs becau.se of the
greater cost of mistakenly ejecting the wrong egg
(Davies and Brooke 1988, Lotem and Nakamura
1998).

Egg mimicry has been documented or suspect-
ed in several parasitic systems (Underwood and
Sealy 2002). Most notably different groups of
female Common Cuckoos (Ciiciiliis canonts) have
evolved eggs that closely mimic those of a

' Department of Biological .Scienees, University of
Manitoba, Winnipeg, MB R3T 2N2. Canada.

^Current address: Faculty of Medicine. University of
Manitoba. Winnipeg, MB R3E 3P.3, Canada.

’ Corresponding author: e-mail: sgsealy@cc.umanitoba.ca

preferred host (Davies and Brooke 1988, March-
etti et al. 1998, Gibbs et al. 2000). The Brown-
headed Cowbird {Molothrus ater, hereafter cow-
bird) has been largely exempt from these
speculations (Sealy et al. 2002, Mermoz and
Ornelas 2004) because individual female cow-
birds are known to parasitize multiple species
(Fleischer 1985, Alderson et al. 1999). Rothstein
(2001) suggested cowbird egg mimicry could not
evolve due to the immense variability in egg
appearance each female cowbird would encounter
when parasitizing several species. He also argued
that because cowbirds are known to parasitize
hundreds of species it would be more beneficial to
avoid hosts that discriminate among eggs than to
enter a costly co-evolutionary race. A recent
comparison of tolerant and intolerant hosts of the
cowbird also found that egg appearance was not
correlated with acceptance of cowbird eggs (Peer
and Sealy 2004).

The close resemblance in color and maculation
pattern between cowbird eggs and those of many
grassland passerines, despite skepticism, has been
suggested to be a form of generalized egg
mimicry (Peer et al. 2000, Davis et al. 2002,
Frontispiece in Klippenstine and Sealy 2008).
Generalized cowbird egg mimicry is the idea that
cowbird eggs have been adapted to match the
appearance of eggs laid by multiple hosts. The
eggs of a parasite in traditional mimicry evolve to
resemble those of a single, intolerant host species
(Underwood and Sealy 2002). Generalized egg
mimicry, however, incorporates the selective

346

K/ippcnsfilic and Scaly • COMPARISON OF COWBIRD AND GRASSLAND PASSERINE EGGS 347

torces provided by all hosts exhibiting egg
discrimination. This causes the parasite’s eggs to
assume the average color and maculation pattern of
eggs from all intolerant hosts in proportion to the
selective pressure provided by each host. A single
parasite egg can be laid randomly in any host nest
in true generalized egg mimici^ and still have a
high probability of matching the host’s eggs.

Indirect evidence of generalized cowbird egg
mimicry has come from experiments that showed
several grassland passerines accept cowbird eggs
despite possessing an ability to eject less mimetic
foreign eggs {Peer et al. 2000, Davis et al. 2002,
Klippenstine and Sealy 2008). Therefore, the
tolerance of parasitism observed in grassland
passerines has, in part, been attributed to their
inability to identify cowbird eggs among their
clutch, i.e., evolutionary equilibrium hypothesis
(Peer et al. 2000, Davis et al. 2002, Klippenstine
and Sealy 2008). Generalized cowbird egg mim-
icry is plausible because most grassland passerine
eggs are similar in color and maculation pattern
(Peer and Sealy 2004, Klippenstine and Sealy
2008). Grassland passerines also are historic hosts
of cowbirds and, presumably, have been exposed to
parasitism sufficiently long for the required co-
evolutionary race to have occurred (Mayfield 1965,
Rothstein 1994, Igl and Johnson 2007).

We quantitatively compared color and macula-
tion pattern of cowbird eggs to eggs of a
community of grassland passerines known to
accept cowbird eggs, but eject non-mimetic eggs,
to examine the possibility of generalized cowbird
egg mimicry (Klippenstine and Sealy 2008);
Sprague’s Pipits (Anthus spragueii). Vesper Spar-
rows (Pooecetes gramineus). Savannah Span'ows
(Passerculiis sandwichensis), Baird’s Sparrows
(Ammodramus hairdii). Chestnut-collared Long-
spurs (Calcariiis ornatus), and Western Meadow-
larks (Stwnella neglecta). We predicted if gener-
alized cowbird egg mimicry has evolved that: (I)
cowbird eggs would match the average color and
maculation pattern present among the eggs of all
six grassland passerines; (2) cowbirds would have
incorporated enough of the community-wide
variation that a proportion of their eggs would
match those of each grassland passerine; and (3)
an individual cowbird egg would be capable of
mimicking the eggs of more than one species.

METHODS

Fieldwork was conducted in southcentral Sas-
katchewan, Canada from 7 May to I July 2001

and 9 May to 15 July 2002. Sites were 60 km
south of Regina within a lOO-knri area centered
about the abandoned hamlet of Dummer (49° 50'
N, 104° 49' W). Nests were found in cultivated
and uncultivated land using predominantly a
dragging rope as described by Klippen.stine and
Sealy (2008).

Photographs of Eggs. — ^Appearance of cowbird
eggs and grassland passerine eggs was quantified
using photographs taken with a 35-mm Pentax
automatic focus camera and 100-speed color
Kodak film for increased resolution (Fleischer
and Smith 1992). Photographs of complete
clutches were desired unless this was not possible
because of partial predation, cowbird parasitism,
or if eggs broke when handled. Cowbird eggs
were photographed whenever a nest was naturally
parasitized and were included in the photograph
of the host’s eggs.

All eggs were temporarily placed in an egg box,
which measured —20 by 50 cm and had a foam
mat on the bottom with six indentations that held
the eggs in place during each photograph. Three
color chips (red, green, and blue), each with three
shades of their corresponding color (light, medi-
um, and dark), were placed on the mat and
included in the photograph (Fleischer and Smith
1992, Villafuerte and Negro 1998). The egg box
was placed in a portable hood to exclude outside
lighting and a camera stand within the hood was
used to ensure consistent photography. Films were
developed as 4 X 6 color prints with a matte finish
to decrease glare. Photographs in 2001 were
developed throughout the field season, but all
photographs in 2002 were developed on the same
day to reduce variation in color, contrast, and
brightness caused by developing at different
times. Prints were scanned with a flatbed scanner
at 300 dpi for digital analysis.

Measurements of Egg Parameters. — Four egg
parameters were measured: ground color, macu-
lation color, distribution of maculation, and total
density of maculation. Eggs were isolated from
their respective photographs using the magic
wand tool of Adobe Photoshop 4.0 to enhance
the accuracy of measurements. Maculation was
measured using inverted black and white images
of each egg. Markings were highlighted as bright
white spots, leaving the ground color of the egg
black. The proportion of maculation on the blunt
and pointed ends of each egg was measured by
centering a 20- X 30-pixel rectangular, selection
box one-quarter the distance from the pointed end

348

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2, June 2010

and one-quarter from the blunt end of each egg,
and measuring the proportion of area covered in
white compared to black using the Northern
Eclipse Program (modified from Fleischer
1985). Density of maculation was calculated as
the percent total area covered in maculation for
both blunt and pointed ends of each egg on a scale
of 0 to 100, i.e., percent X 100. The distribution of
maculation was calculated as the percent of the
total density of maculation occupying only the
blunt end, i.e., blunt end maculation divided by
total maculation multiplied by 100. Eggs with
evenly distributed maculation had values close to
50 and values increased as the proportion of
maculation on the blunt end of the eggs increased.

Ground and maculation colors were measured
from the color images of each egg using the RGB
program (Villafuerte and Negro 1998). Three
samples of ground color were captured using
Adobe Photoshop 4.0, i.e., three square 10 X 10
pixels, along the length of the egg, and avoiding
the area of brightness caused by glare of the
camera’s flash. Three to six spots or scrawls were
sampled in the case of maculation. All selected
areas were transferred into the RGB program,
which calculated the red, green, and blue value of
each pixel on a scale of 0 to 256 (Villafuerte and
Negro 1998). The average red, green, and blue
values were assigned to the background color and
color of maculation for each species’ egg.
Preliminary analy.ses revealed that for eggs of
all species the proportion of red, green, and blue
was highly correlated to each other in both ground
color and maculation color, respectively. The
amounts of red, green, and blue for both
maculation and ground color were summed
because they were not independent; the new value
approximated the darkness or lightness of the
color on a scale of 0 to 768 with an egg’s color
becoming darker as this number decreased. This
was possible because ground color and macula-
tion color of each species’ eggs examined tended
to be brown, differing only in intensity of color,
which would be seen as lightening or darkening.
The correlation between red, green, and blue
would have been lower or absent had different
colors, i.e., blue, pink, etc., been abundant on the
eggs.

The extent to which the shade o( each color
varied within a photograph due to minor differ-
ences in illumination, development, and film was
calculated by measuring the discrepancy between
observed and theoretical values for the three

shades of the color chips photographed with the
eggs. A correction factor was computed that
represented how much red, green, and blue was
added or subtracted from the photograph during
its exposure and development (Villafuerte and
Negro 1998). This correction factor was calculat-
ed for each photograph by averaging the differ-
ences between theoretical and observed values for
all three shades of each color.

Statistical Analyses. — Eggs of all six grassland
passerines were pooled to calculate the average
parameters for this community. The total number
of eggs for each species was reduced to 15 by
randomly eliminating additional eggs to ensure
that all species were represented equally when
pooled. All reductions were repeated twice and
the results were compared to ensure that no bias
occurred during the initial reduction. Only 14
Sprague’s Pipit eggs were available for compar-
ison, and the average of all 14 eggs was inserted
to make up the missing value (Zar 1996). We
opted to pool the eggs of each species in equal
proportions rather than in proportion to their
frequency of parasitism, relative abundance or
frequency of cowbird egg ejection, i.e., as a
measure of selective pressure, for two reasons: ( 1 )
these parameters do not vary significantly among
species within this community (Davis 2003,
Klippenstine and Sealy 2008); and (2) currently,
levels of parasitism and relative abundance of
each species varies between years and habitat
structure (Davis 2003, Klippenstine and Sealy
2008). A single measure of these parameters may
not accurately represent the selective pressure
provided by each species currently or throughout
their co-evolution with cowbirds.

We used Discriminate Analysis (DA) to
examine the extent of discrimination possible
between cowbird eggs and the average grassland
passerine egg using the categories of maculation
(density and distribution of maculation) and color
(ground and maculation color). An ANOVA was
used to identify which parameter or parameters
created the difference if discrimination was
observed in any of these categories (Manly
1986). All four parameters were recombined,
based on the latter analyses, to acquire the correct
collection of parameters that minimized discrim-
ination between cowbird eggs and the average
grassland passerine egg. The predicted group
membership function following a DA was used
to assess overlap between cowbird eggs and
grassland passerine eggs using minimum discrim-

Klippciislinc and Sca/y • COMPARISON OF COWBIRD AND GRASSLAND PASSERINE EG(}S 349

TABLE 1. Mean parameter values for the average grassland passerine egg and Brown-headed Cowbird eggs.

Species («) Density of maculation, %

Disiribuiion of maculalion. %

Ground color

Maculation color

Average grassland passerine (90) 23.5 ± 2.0

Brown-headed Cowbird (50) 28.7 ± 2.2

75.8 ± 2.2*

64.3 ± 1.9

472. 1 ± 9.4
453.6 ± 7.2

372.0 ± 7,1

362.1 ± 6.4

* Differs (P = 0.05) from eowbird eggs.

ination parameters. This function works by
attempting to re-assign each egg blindly to its
correct species, listing the portion of eggs for each
species that was misclassified as being the eggs of
all other species (Manly 1986). The portion of
misclassification between two species’ eggs
represented the extent of overlap or “mimicry”
between their eggs.

The predicted group membership analysis was
conducted on two scales to fully assess mimicry:

( 1 ) the frequency of misclassification that oc-
curred when eggs of all six grassland species were
compared simultaneously with eowbird eggs; and

(2) the frequencies of misclassification that
occurred when eowbird eggs were compared
separately to eggs of each grassland species. We
examined whether the sum of misclassification
frequencies between eowbird eggs and eggs of
each species was greater than the proportion of
eowbird eggs misclassified when compared to all
species simultaneously to learn whether individ-
ual eowbird eggs mimic those of multiple species.
If, however, each eowbird egg only matches the
eggs of one species, the sum of misclassification
frequencies involving each species should be
equal to the proportion misclassified when all
eggs were compared together. Differences in the
frequency of misclassification between species
were as,sessed using a Chi-square test. There was a
tendency for greater discrimination because the
same data used in the initial DA were used in the
predicted group membership (Manly 1986). This
analysis was conservative, as it was more likely to
separate species than place them together. Statis-
tical significance was set at alpha = 0.05 for all
analyses, and all average values are given as mean
± SE.

RESULTS

Cowhird Eggs versus Average Grassland
Passerine Egg. — Cowbird eggs were indistin-
guishable in color (ground and maculation
combined) from the average grassland passerine
egg for this community (DA, yj2 = 1.8, E = 0.40;

Table 1, Eig. 1), but were readily discernable in
maculation (density and distribution of macula-
tion combined; DA, x'2 = 13.5, P = 0.001).
Dissimilarities in overall maculation resulted
exclusively from cowbird eggs being far more
evenly maculated than the average grassland
passerine egg {P\ = 12.3, P = 0.001), as no
difference in the total density of maculation was
observed between the two groups (P\ = 2.5, P —
0.17, Table 1). Thus, minimum discrimination
between cowbird eggs and the average grassland
passerine egg was achieved by excluding distri-
bution of maculation as a parameter (DA, xS =
2.8, P = 0.43 vs. xS = 13.5, P = 0.009 for all
four parameters).

Overlap BetM'een Cowbird Eggs and Grassland
Passerine Eggs. — ^The predicted group member-
ship analysis was significantly more likely to
misclassify cowbird eggs on the basis of mini-
mum discrimination parameters, i.e., color and
total density of maculation, than eggs of all other
grassland passerines (x‘i > 4.4, P < 0.05 for all
species; Table 2). Overall, 88% of cowbird eggs
were misclassified compared to 4 to 70% of grass-
land passerine eggs (Table 2). Between 4 and
67% of grassland passerine eggs were misclassi-
fied as being the eggs of another grassland
passerine, an indication of the close resemblance
between grassland passerine eggs (Table 2).

Between 8 to 48% of cowbird eggs were
misclassified, depending on the species, when
cowbird eggs were compared to eggs of each
species separately using minimum discrimination
parameters (Table 3). The sum of all cowbird egg
misclassifications for all six species greatly
exceeded the proportion of cowbird eggs misclas-
sified when all eggs were compared simulta-
neously (186 vs. 88%; Tables 2, 3). The propor-
tion of cowbird eggs misclassified did not differ
(28 to 49%; X'4 = 6.0, P = 0.20) among eggs of
Sprague’s Pipit, Vesper, Savannah, and Baird’s
sparrows, and Chestnut-collared Longspurs (Ta-
ble 3). A substantially lower proportion of
cowbird eggs were misclassified as Western

350

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

Host species

Host egg Cowbird egg

Sprague’s Pipit

Vesper Sparrow

Savannah Sparrow

Baird’s Sparrow

Chestnut-collared

Longspur

Western Meadowlark

FIG. 1. Example.s of remarkable similarities between gra.ssland pas.serine eggs and cowbird eggs in naturally
parasitized nests.

Meadowlark eggs (8%) than the other five species
(X^i > ^ ^ species; Table 3).

DISCUSSION

Our results support the hypothesis that cowbird
eggs have evolved to generally mimic the appear-
ance of eggs from multiple grassland passerines.
Cowbird eggs were indistinguishable from the
ground and maculation colors, and total density of
maculation as measured for the average grassland

passerine eggs from this community in southern
Saskatchewan (Table I ). All six grassland species
examined have been experimentally demonstrated
to eject non-mimetic eggs (Klippenstine and Sealy
2008). Thus, our results support the idea that
ejection of discordant cowbird eggs has forced
them to assume the average appearance of eggs
from all intolerant hosts.

Cowbird eggs so closely approximated the
color and maculation density of grassland passer-

KUppeusthw and Sealy • COMPARISON OF COWBIRD AND GRASSLAND PASSFZRINE EGGS 351

TABLE 2. Proportion of correct identification and
misclassification observed for each species when
compared simultaneously using minimum discrimination
parameters (ground color, maculation color, and density
of maculation).

Eggs (%) mi.sclassiried £

IS . . .

Species {n)

Eggs correctly
identified
(%)

Other
grassland
passerine eggs

Brown-headed

Cowbird

eggs

Other

species

overall

Sprague’.s Pipit
(14)

71

22

7

29

Vesper Sparrow
(21)

43

57

0

57

Savannah

Sparrow (34)

41

56

3

59

Baird's SpaiTow
(37)

54

30

16

46

Chestnut-collared
Longspur (40)

30

67

3

70

Western

Meadowlark

(24)

96

4

0

4

Brown-headed
Cowbird (50)

12

88

*

00

00

* Differs (P = 0.05) from all other grassland passerine eggs.

ine eggs that 88% ( 1 of every 1 . 1 eggs) of cowbird
eggs were misclassified as belonging to another
species (Table 2). Between 8 and 48% of cowbird
eggs were misclassified depending on the grass-
land hosts, when compared to eggs of each species
separately (Table 3). Thus, any cowbird egg laid
randomly in a Sprague’s Pipit, Vesper Sparrow,
Savannah Sparrow, Baird’s Sparrow, or Chestnut-
collared Longspur nest has between a “one-in-
three” and a “one-in-two” chance of matching
the hosts eggs; the chance of matching for
Western Meadowlarks is significantly lower at
approximately “one-in-ten” (Table 3). Effective
mimicry requires most cowbird eggs to match the
eggs of at least one host. This is especially true
considering female cowbirds are thought to
exhibit little to no host selection (Fleischer
1985, Alderson et al. 1999). “Generalized”
mimicry necessitates that color and maculation
density of cowbird eggs overlap to a significant
extent with the eggs of all six species.

Another important component of generalized
mimicry is the idea that a single cowbird egg may
mimic the eggs of more than one species. This
contrasts with the elaborate mimetic system
evolved by the Common Cuckoo. Different
subgroups or “gentes” of female cuckoos in that

TABLE 3. Proportion of nii.scla.ssificulion observed
using minimum discrimination parameters (ground color,
maculation color, and density of maculation) between
cowbirds eggs and eggs of individual species.

Cowbird

Species (H)

eggs misclassified as eggs of each
grassland passerine i%)

Sprague's Pipit ( 14)

28

Vesper Sparrow (21 )

38

Savannah Sparrow (34)

36

Baird's Sparrow (37)

28

Chestnut-collared

Longspur (40)

48

Western Meadowlark (24)

8*

Total

186

* Differs (P = 0.05) from all other grassland eggs.

system mimic the eggs of a preferred host (Davies
and Brooke 1988, Marchetti et al. 1998, Gibbs et
al. 2000). The benefit of gentes is that it allows
cuckoos, as a species, to parasitize several species
that exhibit egg discrimination while lowering
competition for nests (Davies and Brooke 1988).
Cowbird gentes are supported (Fig. 1), illustrating
some of the remarkable similarities we discovered
between cowbird eggs and the egg of each
grassland passerine. However, the high frequency
at which cowbird eggs were misclassified when
compared to the eggs of each species (8 to 48%,
sum total 186%; Table 3) is only possible if
individual cowbird eggs are capable of matching
the eggs of more than one species. We conclude
that eggs of individual female cowbirds have not
evolved to mimic those of a prefeixed species.

Cowbird eggs were more evenly maculated
than the average grassland passerine egg despite
the strong resemblance in three of four parameters
(Table 1 ). This discrepancy may ari.se because an
unforeseen cost exists for cowbird eggs that are
more unevenly maculated, e.g., increased rates' of
depredation (Underwood and Sealy 2002, Kilner
2006). Cowbird eggs have similarly been shown
to be rounder than host eggs to increase shell
strength and reduce the potential for damage
during laying or by intolerant hosts (Rohwer and
Spaw 1988, Pieman 1989). Alternatively, diver-
gences in maculation distribution might reflect the
parameters used by grassland passerines to
discriminate cowbird eggs. Rothstein (1982)
found that Gray Catbirds (Dumctella carolinensis)
were more responsive to changes in color than
maculation. and American Robins (Turdus mi-
gratoriiis) were more sensitive to color than size
(also Underwood and Sealy 2006). Differences in

352

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

host discrimination alter the selective pressure
placed on cowbird eggs by each host and modify
which parameters are mimicked (Rothstein 1982,
Underwood and Sealy 2006). Consequently, the
distribution of maculation may not be imitated by
cowbirds because grassland passerines discrimi-
nate foreign eggs solely on the basis of color and
density of maculation.

Our results show that cowbird eggs were most
readily distinguished from eggs of Western
Meadowlarks on the basis of color and maculation
density (Tables 2, 3). This corresponds with the
level of intolerance observed for this species
compared to the other five grassland passerines
(Klippenstine and Sealy 2008). Western Mead-
owlarks ejected 67% of cowbird eggs and 91% of
the non-mimetic eggs added to their nests. The
five remaining grassland passerines, in compari-
son, accepted all or nearly all cowbird eggs and
ejected about 20% of non-mimetic eggs (Klip-
penstine and Sealy 2008). Western Meadowlarks
apparently eject cowbird eggs more readily
because of the lower extent of matching between
the eggs of the two species. Peer and Sealy (2004)
rated Western Meadowlarks eggs similar in color
and maculation to cowbird eggs. They also
suggested egg appearance was not correlated with
this host’s intolerance of parasitism, but was
related to their larger beaks. We suspect the
difference arises because they compared the color
and maculation of eggs using human observers. It
is extremely difficult, however, for a human
observer to assess the density of maculation
independent of its distribution, which in this study
had a profound affect. Consequently, discrepan-
cies observed between the two studies emphasize
the benefits of assessing egg appearance quanti-
tatively.

Generalized cowbird egg mimicry is not the
only possible explanation of the close resem-
blance between cowbird eggs and grassland
passerine eggs. Concealment of eggs in the
nesting habitat from depredation is a primary
force directing the evolution of egg appearance in
most species (Underwood and Sealy 2002, Kilner
2006). Therefore, the same forces promoting the
evolution ot color and maculation density in
grassland passerine eggs may have .selected the
same pattern in cowbird eggs (Underwood and
Sealy 2002, Kilner 2006). We propo.se two criteria
to distinguish between resemblance due to nest
concealment and cowbird egg mimicry. The first
is that cowbird egg mimicry requires host species

that exhibit egg discrimination and ejection. The
presence of egg discrimination in all six species
strongly supports the idea of cowbird egg mimicry
(Klippenstine and Sealy 2008). A worthwhile test
of this criterion would be to examine the extent of
matching between cowbird eggs and eggs of
grassland passerines that have not evolved egg
recognition. The second criterion is that nest
concealment would not necessarily make cowbird
eggs better at matching grassland passerine eggs
than eggs of any other grassland passerine. That
cowbird eggs were significantly more likely to be
misclassified than all other grassland passerine
eggs (88% vs. 4-70%) supports generalized
cowbird egg mimicry (Table 2).

The benefits of mimicking grassland passerine
eggs are obvious for cowbirds. Despite having
evolved egg discrimination, all six grassland
species fully or partially accept cowbird eggs
(Davis et al. 2002, Klippenstine and Sealy 2008).
This tolerance has translated into reasonable
breeding success for cowbirds within this com-
munity (Davis et al. 2002, Davis 2003, Klippen-
stine and Sealy 2008).). However, we believe the
value of generalized cowbird mimicry is mo.st
likely limited to only a few ho.st assemblages.
Communities where host eggs are highly diver-
gent in appearance are probably not conducive to
generalized mimicry (Rothstein 2001). General-
ized cowbird egg mimicry, even for grassland
passerines examined in this study, is feasible only
because each host exhibits a relatively crude level
of egg recognition (Klippenstine and Sealy 2008).
Greater selective pressure will be placed on
cowbirds to evolve more species-specific egg
mimicry as continued parasitism forces grassland
passerines to develop their ability to discriminate
foreign eggs (Moksnes et al. 1990, Davies and
Brooke 1991 ). Cowbird egg mimicry of grassland
passerine eggs may eventually resemble the type
of mimicry exhibited by the Common Cuckoo.
Alternatively, as Rothstein (2001) suggested,
cowbirds may in the future choose to avoid such
troublesome hosts.

ACKNOWLEDGMENTS

We are indebted to the landowners who permitted us to
conduct our research on their land. We thank M. A. Talbot
and N. L. Haalboom for their hard work in the field;
Cornelius Klippenstine for use of his farmhouse; C. S.
Bjornsson and Erwin Huebner for technical assistance; N.
C. Kenkel, M. A. Klippen.stine, D. M. Gillis, .1. F. Hare, and
M. V. Abrahams for statistical advice; M. D. Kujath, L. C.
Graham for commenting on earlier drafts of the manuseript;

Klippeusliiw ami Scaly • COMPARISON OF COWBIRD AND GRASSLAND l>ASSERINE EGGS 353

and two anonymous reviewers for eonstruelivc comments
that improved this manuscript. This study was funded by a
Discovery Grant from the Natural Sciences and Engineer-
ing Research Council of Canada to S. G. Scaly and through
in-kind services provided by S. K. Davis and the
Saskatchewan Watershed Authority (formerly Saskatche-
wan Wetlands Conservation Corporation). This paper is
dedicated to the memory of B. A. Klippenstine.

LITERATURE CITED

Alderson, G. W., H. L. Gibbs, and S. G. Sealy. 1999.
Determining the reproductive behaviour of individual
Brown-headed Cowbirds using microsatellite DNA
markers. Animal Behaviour 58:895-905.

Davies, N. B. and M. de L. Brooke. 1988. Cuckoos
versus Reed Warblers: adaptations and counter
adaptations. Animal Behaviour 36:262-284.

Davies. N. B. and M. de L. Brooke. 1991. Coevolution of
the cuckoo and its hosts. Scientific American 264:92-98.
Davis, S. K. 2003. Nesting ecology of mixed-grass prairie
songbirds in southern Saskatchewan. Wilson Bulletin
115:1 19-130.

Davis, S. K., D. R. Klippenstine, and R. M. Brigham.
2002. Does egg recognition account for the low
incidence of cowbird parasitism in Chestnut-collared
Longspurs (Calcarius ornaliis)! Auk 1 19:556-560.
Fleischer, R. C. 1985. A new technique to identify and
assess the dispersion of eggs of individual brood
parasites. Behavioral Ecology and Sociobiology
17:91-99.

Flei.scher, R. C. and N. G. Smith. 1992. Giant Cowbird
eggs in the nests of two icterid hosts: the use of
morphology and electrophoretic variants to identify
individuals and species. Condor 94:572-578.

Gibbs, H. L., M. D. Sorenson, K. Marchetti, M. de L.
Brooke. N. B. Davies, and H. Nakamura. 2000.
Genetic evidence for female host-specific races of the
Common Cuckoo. Nature 407:183-186.

IGL, L. D. AND D. H. JOHN.SON. 2007. Brown-headed
Cowbird, Molothrus ater, parasitism and abundance in
the Northern Great Plains. Canadian Field-Naturalist
121:239-255.

Kilner. R. M. 2006. The evolution of egg colour and
patterning in birds. Biological Reviews 81:383^06.
Klippenstine, D. R. and S. G. Sealy. 2008. Differential
ejection of cowbird eggs and non-mimetic eggs by
grassland passerines. Wilson Journal of Ornithology
120:667-673.

Lotem, A. and H. Nakamura. 1998. Evolutionary
equilibria in avian brood parasitism: an alternative to
the “arms race-evolutionary lag” concept. Pages 22.3-
235 ill Parasitic birds and their hosts: studies in
coevolution (S. I. Rothstein and S. K. Robinson,
Editors). Oxford University Press, Oxford. United
Kingdom.

Manly, B. F. J. 1986. Multivariate statistical methods: a
primer. Chapman and Hall. New York, USA.
Marchetti, K., H. Nakamura, and H. L. Gibbs. 1998.

Host-race formation in the Common Cuckoo. Science
282:471-472.

Mayfield, H. 1965. The Brown-headed Cowbird, with old
and new hosts. Living Bird 4:13-28.

Mermoz, M. E. and J. F. Ornelas. 2004. Phylogenetic
analysis of life-history adaptations in parasitic cow-
birds. Behavioral Ecology 15:109-119.

MOKSNES, A., E. R0SKAFT, A. T. Braa, L. Korsnes, H. M.
Lampe, and H. C. Pedersen. 1990. Behavioral
responses of potential hosts towards artificial cuckoo
eggs and dummies. Behaviour 1 16:64-89.

Peer, B. D. and S. G. Sealy. 2004. Correlates of egg
rejection in hosts of the Brown-headed Cowbird.
Condor 106:580-599.

Peer, B. D., S. K. Robinson, and J. R. Herkert. 2000.
Egg rejection by cowbird hosts in grasslands. Auk
1 17:892-901.

PiCMAN, J. 1989. Mechanism of increased puncture
resistance of eggs of Brown-headed Cowbirds. Auk
106:577-583.

Rohwer, S. and C. D. Spaw. 1988. Evolutionary lag versus
bill-size constraints: a comparative study of the
acceptance of cowbird egg by old hosts. Evolutionary
Ecology 2:27-36.

Rothstein, S. 1. 1982. Mechanisms of avian egg recogni-
tion: which egg parameters elicit responses by rejecter
species? Behavioral Ecology and Sociobiology
11:229-239.

Rothstein, S. I. 1994. The cowbird’s invasion of the far
west: history, causes and consequences experienced by
host species. Studies in Avian Biology 15:301-315.

Rothstein, S. 1. 2001. Relic behaviours, coevolution and
the retention versus loss of host defences after
episodes of avian brood parasitism. Animal Behaviour
61:95-107.

Sealy, S. G., D. G. McMaster, and B. D. Peer. 2002.
Tactics of obligate brood parasites to secure suitable
incubators. Pages 254-269 in Avian incubation:
behaviour, environment, and evolution (D. C. Deem-
ing, Editor). Oxford University Press, Oxford, United
Kingdom.

SoLER, J. J., J. M. Aviles, M. Soler, and A. P. Mgller.
2003. Evolution of host egg mimicry in a brood
parasite, the Great Spotted Cuckoo. Biological Journal
of the Linnean Society 79:551-563.

Underwood, T. J. and S. G. Sealy. 2002. Adaptive
significance of egg coloration. Pages 280-298 in
Avian incubation: behaviour, environment, and evo-
lution (D. C. Deeming. Editor). Oxford University
Press, Oxford. United Kingdom.

Underwood. T. J. and S. G. Sealy. 2006. Parameters of
Brown-headed Cowbird Malotlinis ater egg discrim-
ination in Warbling Vircos Vireo yilviis. Journal of
Avian Biology 37:457^66.

Vii.LAFUERTE. R. AND J. J. Negro. 1998. Digital imaging
for colour measurement in ecological research.
Ecology Letters 1:151-154.

Zar. j. 1996. Biostalistical analysis. Third Edition. Prentice
Hall, Upper Saddle River. New Jersey. USA.

The Wilson Journal of Ornithology 122(2);354-360, 2010

FLIGHT FEATHER MOLT OF TURKEY VULTURES

ROBERT M. CHANDLER, PETER PYLE,--' MAUREEN E. ELANNERY,'
DOUGLAS J. LONG,''^ AND STEVE N. G. HOWELL''

ABSTRACT. — We document the molt sequence of flight feathers in Turkey Vultures {Cathartes aura) based on studies
of captive and wild birds, and examination of museum specimens. We found an unusual pattern of primary replacement,
which appears to be a modified form of Stajfelmauser, or stepwise wing molt. A Stajfehnauser-Wke strategy for replacement
of the secondaries is also described. These patterns of feather replacement appear to be adaptations to maintain flying
performance while replacing all primaries and most secondaries during each molt. To what extent molt patterns in Turkey
Vultures reflect convergent adaptation for flight, rather than ancestral characters useful for phylogenetic studies, remains
unknown. Received 15 June 2009. Accepted 12 January 2010.

Maintaining optimal flight performance is
essential for Turkey Vultures (Cathartes aura), a
species adapted for long-term soaring (Rosser and
George 1986, Tucker 1987). This species locates
its primary diet of carrion by detecting odor
plumes while aloft, and spends most daylight
hours searching for food (Owre and Northington
1961, Stager 1964). It is a migratory species that
can travel hundreds or thousands of kilometers
seasonally between southern wintering and north-
ern breeding areas (Stewart 1977, Kirk and
Houston 1995, Bildstein and Zalles 2001).

Little is known about the molt of Turkey
Vultures despite its widespread distribution and
abundance (Kirk and Mossman 1998). Identifying
and understanding molt processes in birds can be
accomplished in several ways. Miller (1941),
Stresemann (1963), and Pyle (2005) used data
from museum study skins of raptors to make their
basic molt sequence hypotheses. Analyzing molt
patterns from study skins can be difficult and may
not represent the progression of a single individ-
ual, and sample sizes of large birds are typically
too low to undertake comprehensive analyses.
Houston (1975) observed flight feather replace-
ment in both captive and wild individuals of Old
World vultures (Gyps), but molt in captive birds

' Department of Biological and Environmental Sciences,
Georgia College and State University. Milledgeville. GA
31061. USA.

^Institute for Bird Populations, P. O. Box 1346, Point
Reyes Station. CA 94956, USA.

' Department of Ornithology and Mammalogy, California
Academy of Sciences, 55 Music Concourse Drive, San
Francisco. CA 941 18. USA.

■* Department of Natural Sciences, Oakland Museum of
California, 1000 Oak Street. Oakland. CA 94607, USA.

'PRBO Conservation .Science, 3820 Cypress Drive #1 1,
Petaluma, CA 94954. USA.

'’Corresponding author; e-mail: bob.chandler@gcsu.edu

may not reflect that of wild birds (cf. Pyle 2005).
Snyder et al. (1987) based their molt study of
California Condors (Gymnogyps californiamis) on
photographs of wild birds in flight. Bloom and
Clark (2001) used banding records with some
recaptures to study the molt of the Golden Eagle
(Acitiila chrysaetos), but sample sizes of known
individual birds are low in these studies. Ideally, a
complete understanding of molt strategies would
be based on the study of wild birds, captive
individuals, and specimens; we present such an
analysis of molt in Turkey Vultures.

METHODS

A. M. Rea began collecting molt sequence data
on Turkey Vultures in 1972 as part of a larger
study on the phylogenetic relationship of New
World vultures (Rea 1983). Chandler joined the
project in 1978, collecting feathers from three
adult individuals held captive at the San Diego
Natural History Museum (SDNHM) in San Diego,
California, USA and studying molt year-round on
several others. Flannery and Long tabulated molt
data on five vultures (4 adults and 1 first year bird)
between 2002 and 2004, which were held captive
by the California Academy of Sciences (CAS) in
San Francisco, California. The extended study
period of captive individuals allows calculation ot
feather replacement rates both within and between
individuals to establish ranges of variation. Pyle
and Howell visually examined hundreds of
vultures in the field, and Pyle and Flannery
examined molt in over 20 museum specimens.

We classified ages of Turkey Vultures using
head and beak coloration (Henckel 1981), and
known history for captive individuals. Age coding
of birds follows Pyle (1997); HY (hatching or first
calendar year), SY (second calendar year), TY
(third calendar year), and ASY (at least third

354

Chandler el al. • TURKEY VULTURE MOLT

355

calendar year). Chandler collected data from two
captive (SDNHM) ASYs during 1972-1980 and a
third ASY during 1979-1980. Flannery and Long
collected data from four captive ASYs (3 males
and 1 female known to be ages 7+ to 25+ years)
and a female received as an FlY, housed at CAS
during 2002-2004. The ASYs were from Okla-
homa, Pennsylvania, and Texas; the HY, received
with down still present on the head and the
majority of the beak still dark, was from
Tennessee and amved in December 2001. All
eight individuals were kept in outdoor pens,
ensuring they were acclimated to the natural local
photoperiod and climate.

Primaries, secondaries, and rectrices for the
vultures studied by Chandler (SDNHM) were
painted with a dot and slash code to identify the
specific feather (a dot = 1 and a slash = 5), and
with a year-specific color for each individual for
each year of study. The aviary was checked
several times daily for dropped feathers, which
were labeled and dated. Wings were examined
visually for the individuals at CAS, and dropped
feathers were examined to tabulate wing molt
data. Primaries and secondaries on each healthy
wing of each individual were scored as old,
missing, in pin, growing, or new. The percentage
of feather length grown was recorded for each
growing feather. Data were collected weekly
during the first year of the study and biweekly or
monthly in subsequent years. Drop dates for each
primary on each bird were estimated by calcu-
lating primary-specific growth rates and obtain-
ing mean, back-calculated dates for each feather.
Feathers not dropped that year were designated as
being retained in both studies.

Pyle and Howell recorded data from 1998 to
2006 on wing molt from Turkey Vultures in the
field. Binoculars were used to score primaries as
missing, growing, new, or old. Molt was
assumed if the same primary was missing or
growing on both wings simultaneously; individ-
uals with a missing or growing primary on only
one wing were not included in the data .set. Data
were collected year-round and, although most
data were obtained from central California, at
least 10 birds each from Maryland, Indiana,
Mexico, and El Salvador were scored. Vultures
observed with dark heads and beaks, and no
primary molt in June-December were classified
as HYs and not included in the data set. Analysis
was based on date of observation and progres-
sion of molt, including either no molt, or a score

representing the number of the primary that was
growing on the observation date. Individuals
with two primaries being replaced simultaneous-
ly from different parts of the wing (e.g., an inner
and an outer primary) were scored twice in the
data set. Secondary molt patterns afso were
scored when visible. We analyzed the data using
a two-tailed Utest with JMP Version 5.1
statistical software (SAS Institute 2002) to
examine whether or not there was a difference
in mean dates of molt between captive and wild
birds.

More than 20 specimens at CAS, Museum of
Vertebrate Zoology (MVZ), SDNHM, and the
National Museum of Natural History (USNM)
were examined. Primaries and secondaries were
scored in the same manner as captive birds, and
replacement sequences were inferred based on
wear dines reflecting protracted molts (Pyle
2005).

RESULTS

Primaries. — ^All data indicate the 10 primaries
of a Turkey Vulture molt sequentially from the
innermost primary (PI) to the outermost primary
(PIO). This process started for the SY at CAS in
late January with PI and was completed in late
July with PIO (Pig. 1). This SY then replaced PI-
PS for a second time within the year in
September-November (Pig. I ). The following
year, as a TY, this individual continued the
sequence where it had suspended the year before,
replacing P4-P10 in order between early March
and early August, and replacing PI -PS in August-
September (Pig. 1). The captive ASYs at both
locations replaced P5-P10 in March-September
and replaced P1-P2 in July-August (Pig. 1). This
pattern averaged later than the drop dates recorded
for the younger bird as an SY and a TY.
Replacement of PS and P4 in ASYs was more
variable (Pig. 1).

Based on the above results, data on P1-P4 from
birds in the field were partitioned as either
molting in August-November or in January-
March (Pig. 2). The remaining primaries. P5-
PIO had sequential mean dates of molt, beginning
in April-May (P5) and finishing in September-
October (PIO). This pattern was similar to that
recorded for captive birds of all ages (Pig. 2).
Mean dates of molt for P1-P4 and P7-P8 were not
significantly different between captive and wild
birds. However, mean drop dates were signifi-
cantly earlier (two-tailed Me.st, P < 0.01) in

356

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122. No. 2, June 2010

Captive data

10 -

A ♦ □

9 -

A ♦ □

8 -

El

7 -

♦ A

o

E5

CC

6 -

A □

5

□ ♦ A

4 -

□

♦

►

□

3 -

A

t>

A

2 -

A

«□

1 -

ib.

□ ♦ A

J F M A M

J J A S

0

N

D

Mouth

♦ TY

□ AHYavc

FIG. I. Calculated feather drop date by month for Turkey Vultures. Data collected from Turkey Vultures held captive
at the California Academy of Sciences. SY and TY data collected from same individual during 2002 and 2003. Open
triangle represents assumed drop date for PI in a SY calculated using percentage feather grown; actual drop date occurred
prior to acquisition of the bird.

captive than in wild populations for P5, P6, P9,
and PIO. This difference is most likely attribut-
able to the higher energy requirements for
breeding and food acquisition observed in wild

populations that could suspend, slow down, or
otherwise delay molt.

Three vultures observed in the field were
replacing PI during 20-29 January. All three

Field obseivatious

3

T

J

29

I »

26

I ■ <

97

I ■ 1

42

I ■ 1

19

I -■ 1

9

I •— I

10

t— ■ — I

1 1 « 1 1

F M A M J J

56

I ■-

26

I ■ 1

H

30

I ■ (

28

I ■ 1

19

I ■ I

T 1 1 1

A S O N

Month

D

FIG. 2. Mean (± SE) feather drop date for Turkey Vultures by month based on field observations.

Chaiullcr et al. • TURKEY VULTURE MOLT

357

were classified as SYs by head plumage and
evenly worn primaries (P2-P10). Most of 19
vultures observed replacing P2-P3 in February
and March (Fig. 2) were classified as SYs. In
contrast, many of the 77 vultures observed
replacing P1-P3 in August-November (Fig. 2)
were ASYs. Vultures replacing P4 in February-
April (/; = 19) included three SYs and seven
ASYs. Two vultures that could not be partitioned
into our defined categories were replacing P3 on 4
January and P4 on 7 December (Fig. 2). Both of
these individuals were observed in central Cali-
fornia. ASY vultures showing no primary molt (/?
= 181) were observed between 6 September and
17 March. The mean (± SD) date for individuals
showing no molt was 14 January ±49 days.

Secondaries. — ^There were three consistent foci
for initiating molt of secondaries based on data
from the SDNHM captive birds, specimens, and
field observations: the tertials and secondaries 1
and 5 (SI and S5, numbered proximally from
outermost). Molt proceeded proximally from SI
and S5 and distally from the tertials, and it may
become irregular in the area from S8 to SI 1. Six
secondaries were retained from 1973 to 1978 on
captive ASYs (SDNHM), two for one bird and
four for the other individual. Combined locations
of retained secondaries for the two birds were: S4,
S7 (twice), S8 (twice), and SIO. SI dropped
regularly in April followed by S2 and S5 in May
and June, respectively. S3 and S6 usually dropped
in June, S4 and S7 in July, S8 in August or
September (but at times as early as mid-July), and
S9 and SIO dropped between June and late
September; S9 always dropped before SIO. There
was a tendency for molt to terminate at SI 1 and
S12 with drop dates between May and June, and
S12 following Sll. Smaller secondaries (S1-S4)
dropped together in April and May and as late as
August. The tertiary feathers, as with the smaller
secondaries, had a tendency to drop together
between May and August with no di.scernable
sequence.

There appeared to be .some interannual varia-
tion in the number of secondaries retained. For
example, 1979 was an atypical year of extraordi-
narily high numbers of retained feathers. One
vulture retained five secondaries and two rectri-
ces, another had nine .secondaries and one rectrix
retained that year. A possible rea.son is that the
birds suffered from lead poisoning. The following
year, after the birds had recovered from the
apparent lead poisoning, no feathers were re-

tained. Retained feathers dropped in May or June
of the next year.

Rectrices. — ^Tail feathers generally had a con-
sistent pattern for the six feather pairs (SDNHM).
On one side, numbering rectrices from the central
feather R1 to the outermost feather R6, the
sequence was R1-R6-R2-R4-R3-R5. R1 dropped
between late April and May, R6 dropped regularly
in May. The largest rectrix, R2, dropped in late
May or June, followed by R4 in late June or July,
R3 in July or August, and R5 in mid-August to
early September. Tail feathers retained from the
previous year dropped in May or early June of the
next year. Three rectrices were retained during the
first 6 years (1973-1978), a pair of R4s for one
vulture and a single R5 for another.

Feather Synchrony and Growth. — ^Time elapsed
between dropping of paired (right and left)
feathers was calculated for P5 in captive birds at
SDNHM. The average elapsed time was 7 1/2 days
after the first of the pair had dropped (range 1-
41 days). PI and P6 through PIO held to this
average or had an even shorter elapsed time, at
times less than 24 hrs. P2 through P4 had the
greatest range with an average of less than 10 days
for P2 to an average of 1 month for P3 and P4.
The maximum time for an individual feather was
41 days between drop dates for P3.

DISCUSSION

Basic concepts of flight feather molt in diurnal
soaring hawks, eagles, and Old World vultures
have been presented by Miller ( 1941 ), Stresemann
(1963, 1966), and Stresemann and Stresemann
( 1966). Later studies on wing molt by Brown and
Amadon (1968), Houston (1975), Brown (1976),
Edelstam (1984), Clark (2004), and Pyle (2005)
have led to hypotheses about the role of flight
feather replacement schemes that may be used to
infer possible differences in flight strategies and
aerial performance. Birds molt for specific
reasons during certain times of the year and this
molt is not random feather loss, but has a definite
sequence that has an adaptive advantage in
optimizing aerodynamics with respect to feather
maintenance (Hedenstrom and Sunada 1999.
Rohwer 1999, Pyle 2006).

Large birds exhibit several remigial replace-
ment .strategies detectable within a year and
between years by wear patterns, contrasting
feather appearance, and retained feathers (Pyle
2005, 2006, 2008). The most common strategy of
molt among large birds that need to maintain

358

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 2, June 2010

flight, involves a Stajfelmauser pattern meaning
“staggered molt" (Stresemann and Stresemann
1966) and also known as “stepwise molt”.
Falconids typically replace all primaries and
secondaries during each prebasic molt using
multiple waves originating at P4 and S4, whereas
larger accipitrids can retain primaries and sec-
ondaries from a previous season as part of a
Stajfelmauser sequence (Pyle 2005). The physio-
logical demands of producing new feathers and
the time required to complete a full molt are
correlated to body mass and efficiency for flight
requirements of larger birds (Shugart and Rohwer
1996, Pyle 2005). Sequential replacement of
feathers in waves without having large gaps,
increases the efficiency of wings and still allows
larger birds to fly. A Stajfelmauser pattern occurs
in a wide range of larger birds, e.g., pelicans,
herons, and condors (Pyle 2006, 2008).

As widespread and common as Turkey Vultures
are throughout the Western Hemisphere, it is
surprising that no comprehensive review has been
published about their molt (but note comments
and references by Kirk and Mossman 1998). For
example, Stresemann and Stresemann (1966)
mentioned that primaries are replaced irregularly,
but Jackson (1988) correctly reported that, hairing
some aberrant irregular patterns, the molt takes
place annually and within a single year, being
replaced in a serial inner-to-outer sequence. Rea
(1983) included findings by Chandler on second-
aries and rectrices.

Our data produce a clearer picture of the flight
feather molt of Turkey Vultures, which show
similarities to other New World vultures, and the
California Condor (Snyder et al. 1987). However,
the molt of Turkey Vultures is more regular, more
symmetrical between wings, and in a 1 -year rather
than the condor’s 2-year cycle. There have been
no comprehensive studies published on other
cathartid vultures for additional comparisons or
differences in molt in the Cathartidae.

Further comparisons of flight feather molt with
other species of birds can be made in two ways,
either by comparing molt in avian tamilies related
to the Cathartidae, or by comparing molt adapta-
tions in other avian families that show morpho-
logical convergence with soaring flight. Unfortu-
nately, the proximate affinity of cathartid vultures
to other avian groups (e.g., Falconitormes or
Ciconii formes) remains unresolved (Ligon 1967,
Rea 1983, Emsiie 1 988, Sibley and Ahlquist 1990,
Avi.se et al. 1994, Helbig and Seibold 1995, Wink

1995, Livezey and Zusi 2007). The wing molt of
Turkey Vultures, depending on phylogenetic
relationships, may reflect an ancestral pattern.
Molt patterns similar to those in Turkey Vultures
are shared by some species of Ciconii formes
(Bloesch et al. 1977, Shugart and Rohwer 1996), a
proposed ancestral group to Cathartidae. Howev-
er, similarities in molt patterns are also seen in
several other taxonomic groups, and are consid-
ered to be a widespread convergent molt adapta-
tion in many taxa of larger soaring birds (Pyle
2006). Further data are needed to address this
nature versus nurture issue.

Molt Terminology. — ^Turkey Vultures have an
unusual pattern of primary replacement (Figs. 1,
2) that could be interpreted in several ways. One
inteipretation would be that the second molt cycle
of SY birds involves replacement of P1-P3
(occasionally P1-P4) twice. This would necessi-
tate labeling one of the replacements, either that
of January-March or that of August-November,
as a prealternate molt. Each successive prebasic
molt typically would then begin at P4 (or P5) in
the spring and finish with P1-P3 (or P1-P4) in the
fall. However, prealternate molts are not known to
occur among families thought to be close relatives
of New World vultures.

A second interpretation would be that replace-
ment of P1-P3 in SYs in the fall represents an
advanced start to the third prebasic molt, concur-
rent with completion of the second prebasic molt.
Each subsequent prebasic molt would then begin
with PI in the fall, usually (but not always)
suspend for the winter, and resume with P4 in the
spring (Pyle 2008). This is essentially a modifi-
cation of Staffelmauser, in that all primaries are
replaced in 1 year via two waves. Incomplete-to-
complete replacement of the secondaries also
appears to follow a Staffelmauser-\\]s.Q strategy
found in other large birds (Pyle 2006). However,
molt of primaries differs from text-book Staffel-
mauser in that replacement of PI in the second
wave is delayed until after the preceding wave has
replaced the middle primaries; the molt assumes
an annual regularity not found in other species
exhibiting Staffelmauser. The advancement of
some or all primary molt in selected species of a
family, to occur before another life-history event
(migration, breeding, chick-feeding), has been
documented among loons, puffins, gulls, and
diurnal raptors (Howell 2001, Howell and Pyle
2005, Pyle 2005). However, in these groups the
second prebasic wing molt typically occurs or

Cluiiicller el al. • TURKEY VULTURE MOLT

359

starts lafcr in SYs than in subsequent ages, whieh
is not the case in Turkey Vultures.

A third interpretation would be to treat the first
wing molt of Turkey Vultures as a preformative
molt and the molt of P1-P3 of SYs in the fall as
the start of the second prebasic molt. The idea of a
preformative molt being limited to flight feathers
has not previously been recognized, but Howell
(in press) has argued that Staffelmaiiser can
develop in two ways. One is the conventional
view that the first wing molt pertains to the second
prebasic molt (a normal schedule), the other is
that the process of Staffelmaiiser is kick-started by
insertion of a preformative wing molt (an
accelerated schedule). Species exhibiting acceler-
ated schedules include Osprey (Pandion haliae-
tiis), and most if not all Pelecaniformes (Howell,
in press). The first wing molt in these species
starts earlier than subsequent wing molts, which is
also true of Turkey Vultures. To what extent
normal and accelerated schedules reflect ancestral
traits or environmental responses remains un-
known.

ACKNOWLEDGMENTS

R. M. Chandler especially acknowledges A. M. Rea for
initially inviting him to participate in a molt sequence
study. We thank W. S. Clark for his constructive review and
L. D. Chandler for her help. Many museums and their
curators and staff have allowed us to study specimens in
their care. For this we thank; Philip Unitt at the San Diego
Natural History Museum; S. L. Olson, H. F. James, and R.
B. Clapp at the National Museum of Natural History,
Washington. D.C.; J. P. Dumbacher at the California
Academy of Sciences, San Francisco; and Carla Cicero and
the late N. K. Johnson at the Museum of Vertebrate
Zoology, University of California, Berkeley. We also thank
the SDNHM which supported this research while A. M. Rea
was Curator and R. M. Chandler was Collection Manager;
as well as the California Academy of Sciences for the
support of D. J. Long and M. E. Flannery.

LITERATURE CITED

Avise, j. C., W. S. Nelson, and C. G. Sibley. 1994. DNA
sequence support for a close phylogenetic relationship
between some storks and New World vultures.
Proceedings of the National Academy of Science of
the USA 91:5173-5177.

Bildstein, K. L. and j. Zalles. 2001. Raptor migration
along the Mesoamerican land corridor. Pages 119-141
in Hawkwatching in the Americas. (K. L. Bildstein and
D. Klem Jr., Editors), Hawk Migration Association of
North America, North Wales. Pennsylvania, USA.
Bloesch. M., M. Dizerens, and E. Sutter. 1977. Die
mauser der schwungfedern beim Weisslork Ciconia
ciconia. Ornithological Beob. 74:161-188.

Bloom. P. H. and W. S. Clark. 2001. Molt and sequence
of plumages of Golden Eagles and a technique for in-
hand ageing. North American Bird Bander 26:97-1 16.

Brown, L. 1976. Classification and distribution. Birds of
prey, their biology and ecology. Hamlyn Publishing
Group Limited, London. Great Britain.

Brown, L. and D. Amadon. 1968. Eagles, hawks, and
falcons of the world. Country Life Books, Hamlyn
Publishing Group Limited, London, Great Britain.

Clark, W. S. 2004. Wave moult of the primaries in
acciptrid raptors, and its use in ageing immatures.
Pages 795-804 in Raptors worldwide. World Working
Group on Birds of Prey (R. D. Chancellor and B. U.
Meyersburg, Editors). Budapest, Hungary.

Edelstam, C. 1984. Patterns of moult in large birds of
prey. Acta Zoologica Fennica 21:271-276.

Emslie, S. D. 1988. The fossil history and phylogenetic
relationships of condors (Ciconiiformes: Vulturidae) in
the New World. Journal of Vertebrate Paleontology
8:212-228.

Hedenstrom, a. and S. Sunada. 1999. On the aerody-
namics of moult gaps in birds. Journal of Experimental
Biology 202:67-76

Helbig, a. and I. Seibold. 1995. Are storks and New
World vultures paraphyletic? Molecular Phylogenetics
and Evolution 4:315-319.

Henckel, R. E. 1981. Ageing the Turkey Vulture HY to
ASY. North American Bird Bander 6:106-107.

Houston, D. C. 1975. The moult of the White-backed and
Ruppell’s Griffon vultures. Gyps africaniis and G.
ruppellii. Ibis 1 17:474^88.

Howell, S. N. G. 2001. Molt of the Ivory Gull. Waterbirds
24:438-442.

Howell, S. N. G. In press. Molt in North American birds.
Peterson Reference Guides. Houghton Mifflin Co.,
Boston, Massachusetts, USA.

Howell, S. N. G. and P. Pyle. 2005. Molt, age
determination, and identification of puffins. Birding
37:412-418.

Jackson, J. A. 1988. Turkey Vulture, Cathartes aura.
Pages 25^2 in Handbook of North American birds.
Volume 4. Diurnal Raptors (Part 1). (R. S. Palmer,
Editor). Yale University Press, New Haven. Connecti-
cut, USA.

Kirk, D. A. and D. C. Houston. 1995. Social dominance
in migrant and resident vultures at carcasses: evidence
for a despotic distribution? Behavioral Ecology and
Sociobiology 36:232-233.

Kirk, D. A. and M. J. Mossman. 1998. Turkey Vulture
(Cathartes aura). The birds of North America.
Number 339.

Ligon, j. D. 1967. Relationships of the catharlid vultures.
University of Michigan, Museum of Zoology. Occa-
sional Paper 651:1-26.

Livezey, B. C. and R. L. Zusi. 2007. Higher-order
phylogeny of modern birds (Theropoda. Aves: Neor-
nithes) based on comparative anatomy. 11. Analysis
and discussion. Zoological Journal of the Linnean
Society 149:1-95.

Miller, A. H. 1941. The significance of molt centers

360

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2. June 2010

among the secondary remiges in Falconiformes.
Condor 43:1 13-1 15.

OwRE, O. T. AND P. O. Northington. 1961. Indication of
the sense of smell in the Turkey Vulture, Cathartes
aura (Linnaeus), from feeding tests. American Mid-
land Naturalist 66:200-205.

Pyle. P. 1997. Identification guide to North American birds.
Part I. Slate Creek Press. Bolinas, California, USA.

Pyle, P. 2005. Remigial molt patterns in North American
Falconiformes as related to age, sex, breeding status,
and life-history strategies. Condor 107:823-834.

Pyle, P. 2006. Stajfelmauser and other adaptive strategies
for wing molt in large birds. Western Birds 37:179-
185.

Pyle, P. 2008. Identification guide to North American
birds. Part II. Slate Creek Press, Point Reyes Station,
California, USA.

Rea. a. M. 1983. Cathartid affinities: a brief overview.
Pages 26-54 in Vulture biology and management (S.
R. Wilbur and J. A. Jackson, Editors). University of
California Press, Berkeley, USA.

Rohwer. S. 1999. Time constraints and moult-breeding
tradeoffs in large birds. Proceedings of the Interna-
tional Ornithological Congress 22:568-581.

Rosser, B. W. C. and J. C. George. 1986. Slow muscle
fibres in the pectoralis of the Turkey Vulture
(Cathartes aura)', an adaptation for soaring flight.
Zoologischer Anzeiger 3^:252-258.

SAS 1N.ST1TUTE Inc. 2002. JMP Version 5.1. SAS Institute
Inc., Cary. North Carolina, USA.

Shugart, G. W. and S. Rohwer. 1996. Serial descendent
primary molt or Staffelmauser in Black- crowned
Night Herons. Condor 98:222-233.

Sibley, C. G. and J. E. Ahlquist. 1990. Phylogeny and
classification of birds: a study in molecular evolution.
Yale University Press, New Haven. Connecticut, USA.

Snyder, N. E. R., E. V. Johnson, and D. A. Clendenen.
1987. Primary molt of California Condors. Condor
89:468^85.

Stager, K. 1964. The role of olfaction in food location by
the Turkey Vulture (Cathartes aura). Los Angeles
County Museum Contributions in Science 81:1-63.

Stewart, P. A. 1977. Migratory movements and mortality
rate of Turkey Vultures. Bird-Banding 48:122-124.

Stresemann, E. 1963. Taxonomic significance of wing
molt. Proceedings of the International Ornithological
Congress 13: 171-175.

Stresemann, E. 1966. Inheritance and adaptation in moult.
Proceedings of the International Ornithological Con-
gress 14:75-80.

Stresemann. E. and V. Stresemann. 1966. Die Mauser
der Vogel. Journal Eur Ornithologie 107:1-53.

Tucker, V. A. 1987. Gliding birds: the effect of variable
wing span. Journal of Experimental Biology 133:33-
58.

Wink, M. 1995. Phylogeny of Old World and New World
vultures (Aves: Accipitridae and Cathartidae) inferred
from nucleotide sequences of the mitochondrial
cytochrome b gene. Zeitschrift fur Naturforschung
50:868-882.

Short Communications

The Wilson Joiinuil of Ornitholoi’y l22(2):36l-365, 2010

The First Reported Case of Cooperative Polyandry in the Red-footed Booby:

Trio Relationships and Benefits

Lei Cao,' “ Guixia Zhao,' Shan Tang,' and Hanzhao Guo'

ABSTRACT. — We report the first observations of
cooperative polyandry in the Red-footed Booby (Siila
siila), a trio that remained stable over at least 4
consecutive years. Both males were heterosexual;
neither was dominant. One male was more sexually
active, but mounting activity suggested shared paternity
over the years. Trio males, compared to monogamous
males, invested more in pair bond maintenance activity
which possibly assisted in maintaining a stable breeding
unit. Trio members, like monogamous adults, shared
breeding duties evenly, reducing the total investment for
each trio member relative to that of monogamous
adults. The extra foraging effort of the trio enabled
similar provisioning rates in good and bad years. We
suggest trio formation improved female lifetime repro-
ductive success due to enhanced nest, egg, and chick
care; maintained stable food provisioning rates despite
highly heterogeneous foraging conditions; and in-
creased male survivorship due to sharing of breeding
duties. Red-footed Booby trio formation is probably an
“opportunistic deviation” from normal monogamy
which is favored by differential costs and benefits to
males and females. Received 15 July 2009. Accepted 10
January 2010.

Most seabirds are monogamous (Lack 1967),
i.e., cooperative mating systems are rarely ob-
served and there is little information on parental
investment or cost and benefits to the individuals
involved (Stacey and Koenig 1990, Phillips 2002,
Ludwigs 2004). Brown Skuas (Stercorariiis ant-
arcticiis), forming polygynous and polyandrous
trios, were no more reproductively successful than
monogamous pairs (Phillips 2002). A Common
Tern (Sterna hiriinclo) trio raised more chicks than
comparable monogamous pairs through higher
brooding and provisioning rates (Ludwigs 2004).

The Red-footed Booby (Siila sula) is socially
monogamous showing reversed sexual size di-
morphism (Nelson 1978); both adults share
parental duties equally (Weimerskirch et al.

' School of Life Sciences, University of Science and
Technology of China, Hefei 230027, Anhui Province,
China.

"Corresponding author; e-mail: caolei@ustc.edu.cn

2006). We report the first description of cooper-
ative polyandry in the Red-footed Booby, a trio
that nested five times in the same tree during
2004-2007. We studied trio male bond mainte-
nance activities and breeding behavior to test
whether trio members shared parental duties as in
monogamous pairs.

METHODS

Study Site and Field Work. — ^Dong Island (16°
40' 03" N, 1 12° 43' 52" E) is in the northern South
China Sea and supports ~35,500 pairs of Red-
footed Boobies (Cao 2005). A trio nested five
times during 2004-2007: in March 2004, January
2005, November 2005, December 2006 (when all
nests in the colony were destroyed by a typhoon),
and March 2007. Trio male bond maintenance
behavior during the nest-building period was
observed for 115 hrs in 2006 and 124 hrs in
2007, standardized to 100 hrs for analysis. The
trio and four adjacent monogamous nests (pair
numbers: N9, N1 1, N12, and N13) were studied in
2007 from nest building to chick fledging to
compare adult behavior. Nest construction started
between 6 and 19 March and chicks fledged from
17 to 23 August. Observation periods/breeding
durations of the five nests were: trio = 893 hrs/
1 69 days; N9 = 873 hrs/ 1 64 days; N 1 1 = 874 hrs/
167 days; N12 = 880 hrs/ 166 days; and N13 =
803 hrs/ 158 days; activities were standardized to
1,000 hrs for analysis (heterosexual mounting was
standardized to 100 hrs). Age of birds was
classified based on plumage (Nelson 1978) and
males and females were classified using color of
bare facial parts (James 2001) and voice (Lormee
et al. 2003). Trio individuals were uniquely color
marked. Measurements were taken of the trio and
paired adults following Lowe (1989).

Studies of breeding behavior focused on bond
maintenance and division of labor. Bond mainte-
nance was measured as frequency of sexual
activities (sky-pointing and mounting) and mate
preening per unit of time. Division of labor
included nest attendance (% time spent by each

361

362

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2, June 2010

TABLE 1. Trio male bond maintenance activities during nest building in 2006 and 2007. Values rounded to the nearest
whole number for clarity of presentation; NS = not significant {P > 0.05).

.Activity (n/100 hrs)

2006

2007

2006 vs. 2007

CTA

o-B

CTA VS. CfB

ctA

CTB

OCA vs. o-B

CfA

o-B

Sky-pointing

4

17

P < 0.01

2

31

P < 0.001

NS

P < 0.05

Heterosexual mounting

23

43

P < 0.05

23

16

NS

NS

P < 0.001

Homosexual mounting

3

10

P < 0.05

2

5

NS

NS

NS

Heterosexual preening

30

10

P < 0.001

32

13

P < 0.01

NS

NS

Homosexual preening

17

4

P < 0.01

24

9

P < 0.01

NS

NS

adult), territorial defense (frequency of flight-
circuiting, fighting, wing-flailing, and forward head
waving; Nelson 1978), nest construction (frequency
of delivery of nesting material by each bird),
incubation (hrs spent by each adult; data were
standardized to 300 hrs for analysis), provisioning
and foraging; provisioning was measured by the
percentage of days that: ( 1 ) each adult fed the chick,
and (2) foraging trips were undertaken by each
adult during the observation period. Provisioning
and foraging data collection occurred during the
late chick-rearing period after brooding ceased.

Chicks were weighed at 0700, 1600, and
2000 hrs daily from 60 to 88 days after hatching
(peak body mass period before fledging; Cao
2005) to learn whether they had been fed and the
amount of food delivered (calculated as the
difference between successive weighings).

Breeding success (fledglings produced divided
by the number of eggs laid) was calculated for the
trio and the general population by checking nests
every 5 or 10 days. Data affected by the super
typhoon in December 2006 were omitted from the
calculation, as a typhoon of this strength, which
destroyed all nests in the colony, is a rare
occurrence (about once every 20 years).

Statistical Analyses. — ^The binomial distribution
was used to examine differences between activity
levels of trio individuals and between those of trio
and monogamous individuals (Tables 1, 2) using
the equation: : = \p - 0.5l/y{ Var(y;)) based on the
null hypothesis that a p-value of 0.5 is expected if
there is no difference between activity levels (p =
proportion of the combined activities attributable
to one individual and Var(/7) — p(\ — p)/u \u =
combined activities of the 2 individuals]).

We calculated the number of standard deviation
units (-) that the activity level of the trio member
(.Y) was from the mean activity level of the
monogamous sample (.v) using the equation z =
(x - a:)/s, where s = standard deviation for the

monogamous sample, to examine differences
between division of labor activity levels of trio
and monogamous pair individuals (Table 2).

We used the r-score for sample sizes greater
than 30, and the f-distribution for smaller sample
sizes (Fowler and Cohen 1996); all tests were
two-tailed. The r-test was used to examine
differences between adult measurements (mass,
wing length, etc) and between food deliveries to
chicks. Data are presented as mean ± SD.

RESULTS

Measurements. — ^The trio consisted of adult
males o*A and crB, and an adult female (9A).
Body masses of o*A (948 g) and 9A (961 g) were
similar, ~ 10% heavier than CB (856 g), but body
masses of trio members were not different (P >
0.05) from those of paired adults (male = 852 ±
47.1 g. n = 17; female = 956 ± 77.9 g, n = 14);
differences were not significant (P < 0.05)
between culmen lengths (crA = 84.0 mm; crB
= 85.5 mm; 9A = 85.0 mm; paired adults: male
= 82.7 ± 2.64 mm, n = 17; female = 84.7 ±
2.94 mm, n - 14) or wing lengths {CA =
386 mm; o*B = 373 mm; 9A = 377 mm; paired
adults: male = 377 ± 8.1 mm, n = 17; female =
382 ± 5.9 mm, n = 14).

Breeding Behavior. — ^Trio males were hetero-
sexual (Table 1); no agonistic behavior was
observed, even when both males were present
during copulation and sky-pointing. Male A mate-
preened significantly more and o*B sky-pointed
significantly more in both years, but ctB did
significantly more mounting than cA in 1 year.
Male A exhibited similar bond maintenance
behavior in both years, but cB sky-pointed
significantly more and did less heterosexual
mounting in 2007. The trio indulged in far more
bond maintenance activity than monogamous
pairs, but frequency of heterosexual mounting
was not significantly different (Table 2).

SHORT COMMUNICATIONS

363

TABLE 2. Breeding behavior of trio and monogamous pair members during nest building to chick fledging in 2007.
Values rounded to the nearest whole number for clarity of presentation; trio male duplicate data are for CfA and CB,
respectively; NS = not significant {P > 0.05).

Trio

N9

Nil

N12

NI3

p

Sky-pointing (;;/ 1,000 hrs)

Male(s)

5/48

0

10

8

0

NS/P < 0.01

Female

0

1

0

0

0

Total

53

1

10

8

0

P < 0.01

Heterosexual mounting (/t/100 hrs)

Male(s)

23/16

23

34

30

14

NS/NS

Total

39

23

34

30

14

NS

Mate preening (between sexes)

Male(s)

139/36

2

17

0

25

P < 0.01/NS

(n/ 1.000 hrs)

Female

3/10

12

13

0

44

NS/NS

Total

188

14

30

0

68

P < 0.05

Nest attendance (hrs/ 1,000 hrs)

Male(s)

384/486

397

311

372

408

NS/NS

Female

411

404

488

371

467

NS

Total

1,281

801

800

743

875

P < 0.01

cr/Q ratio

30:38:32

50:50

39:61

50:50

47:53

NS

Territorial defense (/7/l,000 hrs)

Male(s)

184/258

111

114

119

93

P < 0.01/P < 0.01

Female

no

136

135

98

103

NS

Total

551

247

249

217

197

P < 0.01

c/g ratio

33:47:20

45:55

46:54

55:45

47:53

NS

Nest construction

Nest material supply (/;/ 1,000 hrs)

Male(s)

82/78

235

237

256

54

NS/NS

Female

1

6

I

2

3

NS

Total

161

241

238

258

56

NS

Nest building («/l,000 hrs)

Male(s)

24/27

13

5

8

5

P < 0.05/P < 0.05

Female

108

192

212

240

44

NS

Total

159

205

217

248

49

NS

Totals

cf/g ratio

33:33:34

56:44

53:47

52:48

56:44

NS

Incubation (hrs/300 hrs)

Male(s)

69/89

125

118

151

137

P < 0.05/NS

Female

142

175

182

149

163

NS

<y/g ratio

23:30:47

42:58

39:61

50:50

46:54

NS

Provisioning

Days chick fed by each adult (%)

Male(s)

89/21

62

77

56

69

NS/P < 0.05

Female

39

86

59

67

46

NS

cr/g ratio

60:14:26

42:58

57:43

46:54

60:40

NS

Foraging

Days foraging trips made (%)

Male(s)

100/86

95

100

100

100

NS/P < 0.05

Female

96

100

95

94

92

NS

<y/g ratio

35:31:34

49:51

51:49

52:48

52:48

NS

Trio members shared nest attendance, total nest
construction, and foraging activities evenly,
except crA did more provisioning and less
incubation, crR conducted more temtorial defense
and less provisioning and foraging, and both
males did more nest building than monogamous
pairs. Trio members shared parental duties
equally, which reduced individual contributions
to nest construction, incubation, and provisioning
while enhancing nest attendance, territorial de-
fense, and foraging effort through the presence of
an additional adult.

Food Delivery and Breeding Success. — Signif-
icantly more food was delivered to chicks of
monogamous parents in 2007 compared to 2004
(2007; 3,744 ± 371 g, n = 6; 2004: 2,837 ± 95 g,
n = 3', P < 0.05) because adults delivered food on
a greater percentage of days (2007; 97 ± 2.2%, n
= 6; 2004: 75 ± 3.5%, n = 3. P < 0.05). The trio
delivered significantly more food in 2004 than the
average monogamous pair (3,485 g vs. 2,837 ±
95 g, n = 3, P < 0.05) as they fed their chick on a
significantly higher percentage of days (97% vs.
75 ± 3.5%, n = 3,P < 0.05).

364

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

Monogamous pairs in 2007 delivered more
tood than in 2004 because foraging conditions
were better. The trio overcame poor feeding
conditions in 2004 due to their combined foraging
effort which maintained food delivery at a similar
level in both years (2004: 97%; 2007: 96%).
Excluding the typhoon year, the trio successfully
fledged four chicks, i.e., one chick/year, compared
with 0.49 (/; = 35) chicks/year in the general
population.

DISCUSSION

Trio Relationships. — ^Measurements and body
mass suggested trio members were not the result
of associations between poor quality individuals.
Male A adopted the female position more often
during homosexual mounting in one of the years,
but its overall .sexual behavior was similai' to that of
monogamous males. Trio members, like monoga-
mous adults (Weimerskirch et al. 2006), shai'ed
breeding duties equally, but males indulged in more
bond maintenance than monogamous males (espe-
cially crA), which may have contributed to
maintaining trio stability over A+ years, as suggest-
ed by Gonzalez et al. (2006) for the Spanish
Imperial Eagle (Aquila adalherti). Neither trio male
seemed dominant, consistent with “type one”
cooperative polyandi'y of Heinsohn et al. (2007)
where males share territorial defense and are non-
aggressive towards each other.

Benefits of Being a Trio. — ^Breeding success for
the trio female was double that of monogamous
females, similar to observations of Common Terns
(Ludwigs 2004), and Galapagos (Buteo gaiapa-
goensis) and Harris’ hawks (Parahiiteo iinicinctns)
(Eaaborg and Bednarz 1990), but male productivity
did not differ. Suggested reasons for increased trio
individual lifetime reproductive success (Phillips
2002, Ludwigs 2004, Gonzalez et al. 2(X)6) also
apply to the Red-footed Booby trio:

(1) enhanced nest, egg, and chick care from
three adults;

(2) stable food delivery in an heterogeneous
environment (Weimerskirch et al. 2005): the
extra foraging adult maintaining similar food
provisioning rates in good and bad seasons
may be adaptive in dynamic environments
(Rubenstein and Lovette 2007);

(3) enhanced survival: sharing of breeding
duties between trio members particularly
benefits males which lo.se body condition

late in the chick-rearing period (Lormee et
al. 2003, Weimerskirch et al. 2006).

Why Aren't There More Trios? — Assuming a
random sample, observations of 54 Dong Island
nests suggested that cooperative polyandry was
rare, occurring at <2% of Red-footed Booby
nests. This is surprising as Red-footed boobies
have a long reproductive period and the benefits
to be gained from kinship selection may be
expected to lead to more common trio formation.

Reversal in heterosexual mounting activity of
crB over 2 years suggested potentially shared
paternity; both males contributed to parental care
because neither knew which was the father
(Jamieson et al. 1994, DeLay et al. 1996). Paternal
uncertainty could explain trio rarity among Red-
footed Boobies, given the lack of benefit to male
reproductive success of being a trio member. A
future priority should be to conduct genetic
analysis to ascertain chick parentage.

Cooperation may arise as a result of a skewed sex
ratio (Young et al. 2008). We have no direct
information on the sex ratio on Dong Island but, of
~ 2,000 nests inspected, only two cases of adult
males breeding with females in immature plumage
suggests the male: female ratio does not differ
greatly from unity. We believe it is most likely
that Red-footed Booby trio formation is an
“opportunistic deviation’’ from the norm, as
proposed for Common Terns (Ludwigs et al. 2004).

ACKNOWLEDGMENTS

We thank C. S. Li, Y. L. Pan, and L. H. Peng for
as.sistance with field work, and M. A. Barter, A. D. Fox.
Ken Rogers, and two anonymous referees for improving the
paper. This study was supported by the National Natural
Science Foundation of China (Grant Numbers 30570253
and 30940010) and a Chinese Academy of Scienees
Innovation Project (Grant Number KZCX2-YW-ON503).

LITERATURE CITED

Cao, L. 2005. Population ecology of the Red-footed Booby
on the Xisha Archipelago. Dissertation. Lanzhou
University. Lanzhou. China.

Delay, L. S., J. FaaborCi, .1. Naranjo, S. M. Paz, T.
DeVries, and P. G. Parker. 1996. Paternal care ip
the cooperatively polyandrous Galapagos Flawk.
Condor 98:300-311.

Faaborg, j. and .1. C. Bednarz. 1990. Galapagos and
Harris" hawks: divergent causes of soeiality in two
raptors. Pages 359-383 in Cooperative breeding in
birds: long-term studies of ecology and behavior (P. B.
Stacey and W. D. Koenig, Editors). Cambridge
University Press, Cambridge, United Kingdom.

SHORT COMMIJNICA FIONS

365

Fowler, J. and L. Cohen. 1996. Staiistic.s for ornilholo-
gisls. Second Edition. British Trust for Ornithology,
Norfolk, United Kingdom.

Gonzalez, L. M„ A. Margalida, R. Sanchez, and J.
Oria. 2006. Cooperative breeding in the Spanish
Imperial Eagle Aqiiila culalherti: a case of polyandry
with male reversed sexual behaviour? Ibis 148:159-
163.

Heinsohn, R., D. Ebert, S. Legge, and R. Peakall. 2007.
Genetic evidence for cooperative polyandry in reverse
dichromatic Ectectus parrots. Animal Behaviour
74:1047-1054.

James, D. J. 2001. Sexual dimorphism in the colouration of
facial bare parts in Red-footed Boobies Sula sulci
ruhripes in the Coral Sea. Pages 99-104 in Herald
Cays Scientific Study Report, Geography Monograph
Series 6 (L. Comben, Editor). Royal Geographical
Society of Queensland, Brisbane, Australia.

Jamieson, 1. G., J. S. Quinn, P. A. Rose, and B. N. White.
1994. Shared paternity among non-relatives is a result
of an egalitarian mating system in a communally
breeding bird, the Pukeko. Proceedings of the Royal
Society of London, Series B 257:271-277.

L.ack, D. 1967. Interrelationships in breeding adaptations
as shown by marine birds. Proceedings of the
International Ornithological Congress 14:3^2.

Lormee, H., P. Jouventin, C. Trouve, and O. Chastel.
2003. Sex-specific pattern in baseline corticosterone
and body condition changes in breeding Red-footed
Boobies Sula sula. Ibis 145:212-219.

Lowe, K. M. 1989. The Australian bird bander’s manual.
Australian National Parks and Wildlife Service,
Canberra.

Ludwigs, J. D. 2004. A case of cooperative polyandry in
the Common Tern. Waterbirds 27:31-34.

Nelson, J. B. 1978. The Sulidae. Oxford University Press,
Oxford, United Kingdom.

Pllii.LlPS. R. A. 2002. Trios of Brown Skuas at Bird Island.
South Georgia: incidence and composition. Condor
104:694-697.

Rubenstein, D. R. and 1. J. Lovette. 2007. Temporal
environmental variability drives the evolution of
cooperative breeding in birds. Current Biology
17:1414-1419.

Stacey, P. B. and W. D. Koenig. 1990. Cooperative
breeding in birds. Long-term studies of ecology and
behaviour. Cambridge University Press, Cambridge,
United Kingdom.

Weimerskirch, H., M. Le Corre, S. Jaquemet, and F.
Marsac. 2005. Foraging strategy of a tropical seabird,
the Red-footed Booby, in a dynamic marine environ-
ment. Marine Ecology Progress Series 288:251-261.

Weimerskirch, H., M. Le Corre, Y. Ropert-Coudert, A.
Kato, and F. Marsac. 2006. Sex-specific foraging
behaviour in a seabird with reversed sexual dimor-
phism: the Red-footed Booby. Oecologia 146:681-
691.

Young, L. C.. B. J. Zaun, and E. A. VanderWere. 2008.
Successful same-sex pairing in Laysan Albatross.
Biology Letters 4:323-325.

The Wilson Journal of Ornithology 122(2):365-369, 2010

Further Evidence of Breeding by Shiny Cowbirds in North America

Matthew J. Reetz,' "^'^ Jacob M. Musser,^ and Andrew W. Kratter^

ABSTRACT. — We report evidence of egg-laying in
northern Florida by the Shiny Cowbird {Molothrus
honariensis). Examination of the ovary and oviduct of a
female trapped in Alachua County, Florida in late June
2006 revealed ovulation and recent passing of eggs.
Two other Shiny Cowbird females captured in late May

'Department of V/ildlife Ecology and Conservation,
University of Florida. P. O. Box 1 10430, Gainesville, FL
32611, USA.

^Department of Ecology, Evolution, and Behavior,
University of Minnesota, 100 Ecology. 1987 Upper Buford
Circle, St. Paul. MN 55108, USA.

■’Florida Museum of Natural History, Box 117800,
University of Florida, Gainesville, FL 3261 1, USA.

■‘Current address: Department of Forest and Wildlife
Ecology, University of Wisconsin. 1630 Linden Drive,
A230A Russell Laboratories, Madison, W1 53706, USA.

■'’Corresponding author; e-mail: mreetz@wi.sc.edu

and early June had convoluted oviducts but showed no
other signs of recent egg-laying. These three specimens
add to the few female Shiny Cowbirds captured in
North America. Our observations represent funher
evidence that Shiny Cowbirds likely breed in North
America and additional evidence of breeding outside of
southern Florida. Received 10 August 2009. Accepted JO
December 2009.

The Shiny Cowbird (Molothrus honariensis) is
a generalist obligate brood-parasite that has been
slowly expanding its breeding range northward
over the last eenttiry from South America through
the Caribbean to North America (Post et al. 1993,
Cruz et al. 2000). Range expansion by the Shiny
Cowbird has caused concern due to contact with
novel hosts without recent exposure to brood

366

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

parasitism (Cruz et al. 1985). For example, arrival
of Shiny Cowbirds in the West Indies threatened
several endemic songbird populations (Wiley
1985, Cruz et al. 1989, Post et al. 1990). Shiny
Cowbirds were first recorded in North America in
southern Florida in 1985 (Smith and Sprunt 1987).
By the mid 1990s, they were observed during the
breeding season (Mar-Aug) in numerous locali-
ties (primarily coastal) throughout Florida (Cruz
et al. 2000) and have been recorded in small
numbers west along the Gulf Coast to Oklahoma
and Texas, and north along the Atlantic Coast
(Post et al. 1993) as far as New Brunswick,
Canada (Benoit 1995).

Both male and female Shiny Cowbirds have
been observed often in the southeastern United
States (Post et al. 1993, Lowther and Post 1999),
but the vast majority of specimens collected
during the breeding season are males (ORNIS
data base http://olla.berkeley.edu/ornisnet, ac-
ces.sed 6 July 2009). We report the collection of
three female Shiny Cowbirds in Florida that
contribute to only five other known females
collected in North America (Sykes and Post
2001; W. Post, pers. comm.). Despite documented
expansion of Shiny Cowbirds in the United States
and a general increase in their abundance (Low-
ther and Post 1999), only a handful of records
indicate breeding, most coming from southern
Florida (Smith and Sprunt 1987, Kale 1989,
Stevenson and Anderson 1994) and one from
coastal Georgia (Sykes and Post 2001). We
present further evidence of breeding by Shiny
Cowbirds in North America through females
collected in northcentral Florida.

METHODS

We captured birds in Alachua County, Florida
(29° 39' N, 82° 19' W) from April to July 2006
and 2007 in drop-down traps (2.4 X 2.4 X 1.8 m)
baited with food and live cowbird decoys as part
of a study on reproduction of female Brown-
headed Cowbirds (Molothrus ater). Traps were
checked daily from ()9()() to IKK) or 1400 to
1600 hrs EST. Trapping was conducted at
cowbird feeding sites, including the University
of Elorida Beef Research, Hor.se Teaching, and
Santa Fe Beef Research units.

Initial specimen identifications were performed
at the Florida Museum of Natural History
(Gainesville, FL) by AWK, using criteria in Pyle
(1997). Specimens were prepared as round skins,
and two tissue samples (muscle, liver, and heart)

were saved from each. Few diagnostic features
distinguish female Shiny from Brown-headed
cowbirds, and genetic analyses of tissues were
conducted at the Bell Museum of Natural History
(St. Paul, MN) to confirm species identity. Whole
genomic DNA was extracted from ethanol-
preserved tissue samples using the QIAGEN
Blood and Tissue Kit (Qiagen, Valencia, CA,
USA). We amplified a 356-base pair region of the
mitochondrial gene Cytochrome h and a 535-base
pair intron from the Z-linked gene MUSK
(muscle, skeletal, receptor tyrosine kinase) using
a PGR protocol similar to Groth and BaiTOw-
cloLigh (1999), but with an initial annealing
temperature of 58° C. Primers used in PGR
amplification and sequencing were El 5656 (Bark-
er 2004) and HI 61 37 (Sorenson et al. 1999) for
cyt b and MUSF-I3F, MUSF-13R, and MUSK-
I3P2 (Kimball et al. 2009) for MUSK intron 3.
Successful amplification was confirmed on aga-
rose gels and products were prepared for sequenc-
ing by enzymatic digestion (Werle et al. 1994).
Sequencing was performed using BigDye Termi-
nator technology on an ABI 3730 Sequencer and
alignment was conducted by eye in the program
SEQUENCHER (Genecodes, Ann Arbor, MI,
USA). We compared the three unknown specimens
to known M. ater and M. honariensis sequences in
Genbank (cyt b: AE089041, AE089043, and
EE529951), and to new sequences obtained from
vouchered tissue samples EMNH 334794 (cyt b;
GU25654I; MUSK: GU256545) and EMNH
350707 (MUSK: GU256546) to verify the identity
of samples. Reference specimens of M. ater were
collected in Illinois and Minnesota whereas those of
M. honariensis were from Puerto Rico.

Ovaries and oviducts were excised and weighed
using a Mettler Toledo Analytical Balance (model
AB54-S). Ovaries and oviducts were examined
with a dissection probe and Bausch and Lomb
Stereo Zoom 4 (7-3()X magnification) dissecting
microscope. Oviduct width was measured to the
nearest millimeter using a standard metric ruler.
Breeding condition of each female was assessed
using pre-ovulatory and post-ovulatory follicles
(Scott and Ankney 1983). Pre-ovulatory follicles
represent developing ova that are subsequently
released into the abdominal cavity and taken into
the oviduct for egg production. Post-ovulatory
follicles remain attached to the ovary alter
ovulation, and are reabsorbed over time. We
recorded the number and size of any pre-ovula-
tory follicles (those that could potentially produce

SNORT COMMUNICATIONS

367

TABLE

June 2007.

1. Reproductive measurements of female Shiny

Cowbirds trapped in Alachua County,

Florida, April-

Specimen #

Collection

date

(

Oviduct condition

3viduct width
(mm)

Oviduct mass
(g)

Ovary mass
(g)

Width of
pre-ovulatory
follicles (mm)

Area of
post-ovulatory
follicles (mnr)

UF 45834

29 May

Minute, convoluted

2

<0.20

<0.05

None

None

UF 46076

8 Jun

Minute, convoluted

3

<0.20

<0.05

None

None

UF 46077

20 Jun

Enlarged, convoluted

8

0.9861

0.6055

9.41

6.17

6.32

7.75 (atretic)

3.56

the yolk of an egg) in each specimen. Widths of pre-
ovulatory follicles were measured to the nearest
0.01mm using a Mitutoyo Absolute Digimatic
caliper (model CD6: CS). Round, orange-yellow
pre-ovulatory follicles were distinguished from
degenerating or atretic follicles which are irregular
in shape and cream-colored. We also recorded the
number and size of all recognizable post-ovulatory
follicles, which are flat and oval-shaped and have
an open slit (stigma) through which ova are released
into the abdominal cavity during ovulation. Only
follicles with ruptured stigma were considered post-
ovulatory follicles and were distinguished from
burst atretic follicles (Pearson and Rohwer 1998),
which are ruptured but typically contain a cream-
colored residue that is lacking in post-ovulatory
follicles. The area of each flattened post-ovulatory
follicle was estimated as the product of width and
maximum lengths measured to the nearest 0.01mm.

RESULTS

Three (UF #’s 45834, 46076, 46077) of 147
female cowbird specimens collected were thought
to be Shiny Cowbirds based on morphological
characteristics, especially the longer, less deep
bill, shorter wing, and longer tail of the Shiny
versus Brown-headed cowbird (Pyle 1997). Com-
parison of known M. uter and M. honariensis
sequences revealed differences at nine sites at cyt
b (P = 0.03 uncorrected distance) and three sites
at MUSK (P = 0.006 uncorrected proportional
distance). This difference in genetic distance
among loci is expected as mitochondrial DNA is
known to have a relatively rapid rate of evolution.
All three unknown specimens had identical
sequences at both the cyt b (Genbank Accession
#’s GU256544, GU256542, GU256543) and
MUSK (GU256549, GU256547, GU256548) loci.
These sequences were a perfect match with M.
honariensis references at both loci. Our limited
number of known references prevents us from

addressing the extent to which these differences
are polymorphic sites rather than fixed substitu-
tions. The extent of differentiation and agreement
of both mitochondrial and nuclear DNA provide
compelling evidence in identifying the three
specimens as M. honariensis.

Two females collected (UF #’s 45834, 46076)
showed no recent signs of egg-laying because
they had small ovaries, minute oviducts, and no
pre- or post-ovulatory follicles (Table 1 ). The
oviduct of each of these females was convoluted
indicating they had passed eggs during the current
breeding season. The third female (UF # 46077)
had numerous signs of recent egg-laying. The
ovary had three orange-yellow pre-ovulatory
follicles that were in good condition and followed
a size sequence consistent with actively laying
Brown-headed Cowbirds in Florida (MJR, unpubl.
data). The size of the largest pre-ovulatory follicle
was similar to follicles in Brown-headed Cow-
birds that would likely have ovulated on the day
following capture (MJR, unpubl. data). The ovary
had one ruptured follicle that was likely post-
ovulatory because the rupture occun'ed along the
stigma and contained no residue that would be
consistent with atresia. The size of this post-
ovulatory follicle was similar to those in Brown-
headed Cowbird daily laying sequences that were
estimated to have ovulated 4 days before collec-
tion (MJR, unpubl. data). The ovary also included
one atretic follicle that was flaccid, lacked a
clearly ruptured stigma, and contained cream-
colored lluid. The oviduct of the specimen was
large (~ Ig) and convoluted indicating it passed an
egg during the cuiTent breeding season. Mass
measurements of the ovaries (range = 0.1224-
0.9172 g, .V = 0.4335 g, SD = 0.1853 g) and
oviducts (range = 0.5802-1.851 1 g, .v = 1.2569 g,
SD = 0.2902 g) of 46 female Brown-headed
Cowbirds in breeding condition (MJR, unpubl.
data) were similar to those for this specimen.

368

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2. June 2010

Collection dates for all three specimens were within
the period when the majority of female Brown-
headed Cowbirds in Florida are in breeding
condition (26 Apr-24 Jun) (MJR, unpubl. data).

DISCUSSION

Shiny Cowbirds have been reported in the
United States, especially Florida, during the
breeding .season for nearly 25 years. To date,
however, there has been scant evidence of
breeding. Seven of eight male Shiny Cowbirds
captured in the southeastern U.S. in 1992 had
enlarged testes (Post et al. 1993), and male
cowbirds have been observed courting females
(Post et al. 1993, FFWCC 2003). However, it is
not clear in these cases whether copulation and
subsequent egg-laying occurred. There are four
observations of fledgling Shiny Cowbirds from
southern Florida, three of which were being fed by
adults of other species (Smith and Sprunt 1987,
Kale 1989, Stevenson and Anderson 1994). These
records are not considered definitive (Lowther
and Post 1999) because of difficulties distinguish-
ing between Juvenile Shiny and Brown-headed
cowbirds. The latter species has also recently
expanded into the southeastern U.S., occurs in
much higher numbers, and has been confirmed
breeding throughout Florida (FFWCC 2003).
However, Sykes and Post (2001) captured a
female Shiny Cowbird on the coast of Georgia
in early July 2000 that had an enlarged ovary and
three developing ova in its oviduct. Until now, no
further substantial evidence of breeding by Shiny
Cowbirds in North America has been reported.

We note our records also cannot be considered
definitive. The two females with convoluted
oviducts may have passed eggs earlier in the
breeding season, but no other signs of recent
reproduction were evident. Shiny Cowbirds may
be capable of substantial long-range movements
(Post et al. 1993), and breeding could have
occurred outside of North America. It is also
possible the female that was in reproductive
condition may have ovulated and laid unlertilized
eggs. However, the appearance of the ovary and
oviduct indicate the female would have produced
a series of unfertilized eggs. We regard this as
unlikely. Furthermore, we do not know if eggs
produced by this female were laid in an
appropriate location (i.e., a potential host nest).
Our observation of a female Shiny Cowbird in
reproductive condition that likely recently passed

an egg lends support to breeding by Shiny
Cowbirds in the United States.

The initial colonization of Florida and the
southeastern United States by Shiny Cowbirds
provoked alarm about negative impacts on native
breeding species (Atherton and Atherton 1988,
Paul 1988). Extremely limited documentation of
breeding by Shiny Cowbirds and near absence of
collected females after nearly 25 years in the
southeastern United States indicate that either the
species is having little success colonizing main-
land North America, or that difficulties in
documenting breeding and/or identifying females
or Juveniles resulted in unden'eporting of breeding
by this species. Focused collection studies of
cowbirds may be helpful in identifying the extent
of breeding by Shiny Cowbirds in these initial
decades of colonization of mainland North
America. Collections will likely improve Shiny
Cowbird abundance estimates by avoiding diffi-
culties in differentiating females of similar
cowbird species.

ACKNOWLEDGMENTS

We thank D. W. Armstrong. J. D. Driver, and J. A.
McQuagge (University of Florida) for access to field sites.
M. L. Avery, J. S. Humphrey, K. L. Keacher, M. P.
Milleson, and E. A. Tillman (USDA, National Wildlife
Research Center) provided logistical support in trapping
cowbirds. We thank D. W. Steadman (Florida Museum of
Natural History) for support in reproductive analyses. P. W.
Sykes (USGS, Patuxent Wildlife Research Center) provided
information on Shiny Cowbird museum specimens. We
thank K. E. Sieving (University of Florida), and S. M.
Lanyon and F. K. Barker (Bell Museum) for laboratory and
logistical support. Invaluable field assistance was provided
by A. W. Cook. This work was supported by the Florida
Fish and Wildlife Conservation Commission’s program,
Florida's Wildlife Legacy Initiative, and the U.S. Fish and
Wildlife Service's State Wildlife Grants program (SWG()4-
031). Protocols were approved under University of Florida
lACUC # D954.

LITERATURE CITED

Atherton, L. S. and B. H. Atherton. 1988. Florida
region. American Birds 42:60-63.

Barker, F. K. 2004. Monophyly and relationships of wrens
(Aves: Troglodytidae); a congruence analysis .of
heterogeneous mitochondrial and nuclear DNA se-
quence data. Molecular Phylogenetics and Evolution
3l:486-.304.

Benou, G. 1995. A day to remember: first Shiny Cowbird
in Canada. Birders Journal 4:89-90.

Cruz. A., T. Manolis, and J. W. Wii.ey. 1985. The Shiny
Cowbird: a brood parasite expanding its range in the
Caribbean region. Ornithological Monographs 36.

SHORT COMMUNICATIONS

369

Cruz, A., J. W. Wiley, T. K. Nakamura, and W. Post.
1989. The Shiny Cowbird Molothrus honariensis in the
West Indian region: hiogeographieal and ecologieal
implieations. Pages 5I9-.‘S40 in Biogeography of the
West Indies: past, present, and future (C. A. Woods,
Editor). Sandhill Crane Press, Gainesville, Florida, USA.
Cruz, A., .1. W. Prather, W. Post, and J. W. Wiley.
2000. The spread of Shiny and Brown-headed
cowbirds into the Florida region. Pages 47-57 in
Ecology and management of cowbirds and their hosts:
studies in the conservation of North American
passerine birds (J. N. M. Smith, T. L. Cook, S. I.
Rothstein, S. K. Robinson, and S. G. Sealy, Editors).
University of Texas Press, Austin, USA.

Florida Fish and Wildlife Conservation Commission
(FFWCC). 2003. Florida’s breeding bird atlas: a
collaborative study of Florida’s birdlife. http://www.
wildflorida.org/bba (accessed 16 June 2009).

Groth. j. G. and G. F. Barrowclough. 1999. Basal
divergences in birds and the phylogenetic utility of the
nuclear RAG-1 gene. Molecular Phylogenetics and
Evolution 12:115-123.

Kale II, H. W. 1989. Florida birds. Florida Naturalist
62:14.

Kimball, R. T., E. L. Braun, F. K. Barker, R. C. K.
Bowie, M. J. Braun, J. L. Chojnowski, S. J.
Hackett, K. L. Han, J. Harshman, V. Heimer-
Torres, W. Holznagel, C. j. Huddleston, B. D.
Marks, K. J. Miglia, W. S. Moore, S. Reddy, F. H.
Sheldon, J. V. Smith, C. C. Witt, and T. Yuri. 2009.
A well-tested set of primers to amplify regions spread
across the avian genome. Molecular Phylogenetics and
Evolution 50:654-660.

Lowther, P. and W. Post. 1999. Shiny Cowbird (Molothrus
honariensis). The birds of North America. Number 399.
Paul, R. T. 1988. Florida region. American Birds 42:1278-
1281.

Pearson, S. F. and S. Roiiwer. 1998. Determining clutch
size and laying dates using ovarian follicles. Journal ol
Field Ornithology 69:587-594.

Po.sT, W., T. K. Nakamura, and A. Cruz. 1990. Patterns
of Shiny Cowbird parasitism in St. Lucia and
southwestern Puerto Rico. Condor 92:461^69.

Post, W., A. Cruz, and D. B. McNair. 1993. The North
American invasion pattern of the Shiny Cowbird.
Journal of Field Ornithology 64:32^1.

Pyle, P. 1997. Identification guide to North American
birds. Part 1. Columbidae to Ploceidae. Slate Creek
Press, Bolinas, California, USA.

Scott, D. M. and C. D. Ankney. 1983. The laying cycle of
Brown-headed Cowbirds: passerine chickens? Auk
100:583-592.

Smith, P. W. and A. Sprunt IV. 1987. The Shiny Cowbird
reaches the United States: will the scourge of the
Caribbean impact Florida’s avifauna too? American
Birds 41:370-371.

Sorenson, M. D., J. C. Ast, D. E. Dimcheff, T. Yuri, and
D. P. Mindell. 1999. Primers for a PCR-based
approach to mitochondrial genome sequencing in
birds and other vertebrates. Molecular Phylogenetics
and Evolution 12:105-114.

Stevenson, H. M. and B. H. Anderson. 1994. The birdlife
of Florida. University Press of Florida, Gainesville,
USA.

Sykes Jr., P. W. and W. Post. 2001. First specimen and
evidence of breeding by the Shiny Cowbird in
Georgia. Oriole 66:45-51.

Werle, E., C. Schneider, M. Renner, M. Volker, and
W. Fiehn. 1994. Convenient single-step, one tube
purification of PCR products for direct sequencing.
Nucleic Acids Research 22:4354-4355.

Wiley, J. W. 1985. Shiny Cowbird parasitism in two avian
communities in Puerto Rico. Condor 87:165-176.

The Wilson Journal of Ornithology 122(2):369-374, 2010

Circumstantial Eviidence for Infanticicie of Chicks of the Communal Smooth-
billed Ani {Crotophaga ani)

J. S. Quinn, A. Samuelsen,' M. Barclay,' G. Schmaltz,'"^ and H. Kahn''^

ABSTRACT. — The Smooth-billed Ani (Crotophaga
ani) is a plural-breeding, joint-nesting species that
engages in both cooperation and reproductive compe-

' Biology Department, McMaster University, Hamilton,
ON L8S 4K1, Canada.

'Biology Department, University of the Fraser Valley,
Abbotsford, BC V2S 7M8, Canada.

■’Current address: Department of Zoology and Fisheries,
University of Agriculture, Faisalabad, Pakistan.
■'Corresponding author; e-mail: quinn@mcmaster.ca

tition. We report circumstantial evidence for infanticide
of chicks by adults that joined a group after completion
of the communal clutch. We provide evidence that some
of the group adults captured after the infanticide were
unrelated to any of the chicks and may have acted to
increase chances of renesting. Received 30 Jiilv 2009.
Accepted 3 December 2009.

Infanticide of young by adults is an infrequent,
yet widespread behavior among many taxa

370

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2. June 2010

(Bertram 1975, Hrdy 1979, Sherman 1981, Hrdy
and Hausfater 1984) and, in some species, it
represents a considerable proportion of juvenile
mortality (Sherman 1981, Packer and Pusey
1983). It is typically considered a reproductive
strategy associated with reducing wasted resourc-
es or parental care directed towards unrelated
young by potential foster parents (Pierroti 1991,
Quinn et al. 1994, Eggert and Muller 2000) or
step-parents (Rohwer et al. 1999).

Resources and care in cooperatively breeding
species are often directed towards non-filial
young. A number of features, including genetic
relatedness among adult group members within
cooperatively breeding groups, affects which
individuals breed and how much. Infanticide has
been proposed as a mechanism to adjust a group’s
reproductive skew, the extent to which individuals
monopolize reproduction (Clutton-brock et al.
1998, Hager and Johnstone 2004, Macedo et al.
2005). Reproductive skew may be adjusted to
support group stability (Vehrencamp 1977, 1978;
Vehrencamp and Quinn 2004), but immigrants
arriving soon after fertilization are likely to be
compromised by delayed reproduction if offspring
care is costly.

Power struggles and resultant take-overs have
been described in social animals including lions
(Panthera leo) (Bertram 1975, Packer and Pusey
1983) and Acorn Woodpeckers (Melanerpes
formicivorus) (Koenig 1981, Hannon et al.
1985). The immigrant individual or group, in
these cases, displaces existing same sex group
members and subsequently may kill existing
offspring. Infanticide has been reported in com-
munal breeding animals, although the evidence is
often indirect as observing infanticide directly is
unlikely (Stacey and Edwards 1983). Research to
establish the relatedness of victims and perpetra-
tors is needed, particularly in plural-breeding
species, species with breeding groups in which
more than one male and female are breeders (Trail
et al. 1981, Macedo et al. 2001).

The Smooth-billed Ani (Crotophaga ani) is a
plural-breeding, joint-nesting neotropical cuckoo
of the subfamily Crotophaginae. Anis form
socially monogamous pairs that cooperate as
a communal group to rai.se mixed broods of
young in a single nest (Davis 1940, Quinn and
Startek-Foote 2000). Ani group size ranges from
two to 17 individuals with up to five females,
typically unrelated to each other (Quinn and
Startek-Foote 2000; J. S. Quinn, unpubl. data).

Genetic data confirm that extra-pair fertilizations
occur in this species (J. S. Quinn, unpubl. data).
Smooth-billed Anis engage commonly in egg
tossing and egg burial, resulting in failure of an
average of 56% of eggs laid by multi-female
groups (Schmaltz et al. 2008). These behaviors
may lead to reduced competition for young
hatching from eggs that are typically laid after
tossing or burial episodes by their parents.
Skewing reproduction may explain this behavior,
but it is possible that ovicidal adults are increasing
hatching synchrony to limit size asymmetry that
might lead to death of their young. Our objectives
are to: (1) describe circumstantial evidence for
chick infanticide, and (2) show that some adults in
the group after the apparent infanticide were
unrelated to the nestlings and may have been new
group members.

METHODS

Adults in groups were counted once a month as
they departed roosts during the morning and
numbers were recorded opportunistically during
nest checks. Nests were checked daily during
laying until incubation began and then were
resumed 2 days prior to expected first hatching.
Adults were captured using mist nets and
playback in early morning at the roost site. Blood
samples were taken from captured adults and
chicks to allow microsatellite genotyping and
parentage exclusion. We genotyped with six
microsatellite loci (45()B2, C5, C14, SNX07,
SNXI4, and SNX17; Blanchard and Quinn
2001, Gregory and Quinn 2005). Relatedness
(Queller and Goodnight 1989) was examined
using GenAlEx6.2 (Peakall and Smouse 2006).

OBSERVATIONS

We observed dead chicks or their disappear-
ance in 23 of 90 broods monitored over 10 years
of study, generally September through December
or early January inclusive from 1998 until 2007 at
Cabo Rojo National Wildlife Refuge, Puerto Rico
(17° 59' N, 67° 10' W). We inferred the loss of 10
broods (43.5%) to have been cau.sed by brown rats
(Rattiis ratOis), Red-tailed Hawks (Buteo jamai-
censis), or red fire ants (Solcnopsis invicta) based
on direct observations or signs at the nest
(Table 1). The collected evidence from one case
does not appear to be the result of the.se predators
and is more consistent with infanticide. We
observed chicks that had been recently killed
with skin stripped off the skull in 2006 in the

SHORT COMMUNICATIONS

371

TABLE 1.

with evidence.

Numbers of failed Smooth-billed Ani broods monitored between 1998 and 2007 (inclusive) and likely cause

Number of

broods

Suspected cause

Evidence/sign of predation at nest

4

Red-tailed Hawk

Small bits of llesh left in the nest; eggs were not predated; one observed
on time lapse videotape.

2

Rat

Bone fragments remained in nest; small sharp tooth indentations in plastic
color bands in one case; only body left (head eaten) in other case.
Video observations indicate that rats completely eat newly hatched
nestlings starting with the head.

4

Red fire ant

Small pustules on skin of dead chicks; ants seen on chicks or in nest.

12

Unknown

Chicks too young to leave nest; nothing remaining. Most consistent with
rat or Red-tailed Hawk predation.

1

Infanticide

Scalp split, skull showing; aggressive vocalizations and movement of
adults near nest; post-laying period increase in group size.

“Gully” group and describe known circumstanc-
es along with microsatellite genotype data for
some captured adults and the nestlings (Table 2).

A small group of up to four adults (2 banded
and 2 not banded) occupied the teiritory, bred, and
tended eggs in a nest discovered by us after clutch
completion. Genetic analysis suggested the pres-
ence of an additional breeding male and female
(likely unbanded and hence less detectable, or
brood parasitism by a non-member of the group).
A color-banded bird that had not previously been
observed on territory was seen shortly after the
last young hatched and ob.servations showed that
at least four additional adults had recently joined
the group. We observed at least seven adults in the

nest tree on 18 October engaging in aggressive
vocalizations and interactions that were obscured
by foliage. We observed no chick-predators in the
area during this time. We found two dead chicks
in the bottom of the nest following these
observations. Four living nestlings observed on
1 8 October were not seen again and are presumed
to have died. We subsequently captured three
adults beside the nest tree using mist nets and
playback. One month after the nestling deaths we
observed one of the adults, which had been
banded on 22 October, nest building on another
teiTitory (Table 2).

We genotyped all six Gully-group nestlings and
all identified color banded adults. Dyads of

TABLE 2. Timing of event.s in 2006 at the Smooth-billed Ani nest on the “Gully" territory.

Date

Observations

16-28 September

28-29 September
2 October
12 October

14 October

15 October

16 October
18 October

21 October

22 October
5 November

16 November

16 December

Up to 4 birds including color-banded RG/A (male), WY/YAblt (female), and unbanded birds
observed in territory.

Two birds candying nesting material.

Nest found with 6 eggs (many tossed eggs beneath).

Eggs 1, 2, and 3 hatched.

Six chicks - 3 new from eggs 4, 5, and 6.

GWYAb (first and only time observed in territory)".

Eour adults observed at nest; 1 with food.

Six or 7 adults in nest tree; aggressive vocalizations heard several times; copulation observed
between unbanded male and female with Ag on right leg; 2 dead nestlings in nest; 1 chick
scrambled away; 3 chicks banded.

Seven adults observed at nest tree; no further record of live chicks on territory.

Three adults captured and banded: BB/AbY (female), RG/AbR (female), WR/AbW (male).

Up to 10 adults observed together on territory interacting without aggression; 6 unbanded. 3 or 4
banded including: BB/AbY (female). RG/AbR (female), and WY/YAblt (female).

BB/AbY (female) and unbanded bird observed nest building in another territory east northeast at
least two territories distant (~ I km).

RG/AbR (female) and WR/AbW (observed in “Gully" territory).

Band combination misread.

372

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2. June 2010

TABLE 3. Exclusion analysis of sampled adults and young from the 2006 “Gully” territory based on six ani
microsatellites. “Included” individuals are those with genotypes consistent with parentage of particular chicks.
Included pairs have genotypes consistent with parentage by the included pair. “Excluded” adults are not shown.

Chick

Included females

Included males

Included pairs

A

WY/YAblt, RG/AbR

RG/A

RG/A X WY/YAblt

B

WY/YAblt

C

RG/AbR

D

RG/AbR

RG/A

RG/A X RG/AbR

E

RG/A, RW/AbW

F

RG/AbR

sampled adults in the group showed mean (± SE)
relatedness values of —0.083 ± 0.105 (range
—0.28 to +0.30). These values suggest the adults
sampled were not close relatives, consistent with
analyses of other Smooth-billed Ani groups (J. S.
Quinn, unpubl. data). Exclusion analysis suggest-
ed the two previously color banded adults
associated with the territory during egg laying
were parents of some young as was one of the
adults captured subsequently (Table 3).

Observations (Table 2) and genotyping (Ta-
ble 3) are consistent with an immigration event
and subsequent infanticide as group size increased
after clutch completion. Some group members
were not related to any of the young, providing a
likely motive for infanticide (to reduce the time to
the next breeding attempt; Macedo and Melo
1999). Aggressive behavior occui+ed in the nest
tree on the day the two dead offspring were
di.scovered. We did not see the remaining four
offspring sub.sequently and presumed they had
died. Banded individuals that had been seen on
the territory prior to clutch completion and one
that was caught subsequently were likely genetic
parents of some of the young. The remaining
individuals captured after clutch completion were
not closely related to any young or sampled adults
in the group.

Elimination of offspring at the chick stage is
not the only way to initiate renesting. We have
tracked egg rejection for several years. Most of
those cases were limited to the egg-laying phase
and did not lead to renesting, but we observed two
cases, both in 2003, where eggs were apparently
pecked, leaving a large hole in each. Thirty eggs
were laid, three of those were ejected from the
nest, nine were buried in the nest structure, and 18
eggs were pecked leading to nest abandonment in
one group of 12 birds (“North House"). Thirty
eggs were laid by another group of seven adults
(“South Eence"); two of those were ejected, 26

were buried in the nest structure, and the
remaining two in the nest bowl were pecked
leading to nest abandonment. Both of these groups
were multi-female breeder groups. We have not
observed other avian species that are egg
predators near nests of Smooth-billed Anis and
suspect these cases were egg rejection. The
abandonment presumably increased chances for
renesting. The North House group did renest
successfully soon after the clutch destruction.
Eggs were destroyed between 19 and 22 Novem-
ber 2003 and the new nest was found on 8
December with 15 eggs in the nest cup, two tossed
eggs beneath, and five buried eggs. Twenty-two
eggs were laid over a maximum of 16 to 19 days;
therefore, the new nest was initiated soon after the
eggs in the previous nest had been destroyed.

DISCUSSION

Egg rejection is a reproductive tactic that we
previously described as common in Smooth-billed
Ani groups (Schmaltz et al. 2008). Egg rejection
is limited to groups with more than one breeding
female and may relate to reproductive competition
(Schmaltz et al. 2008). Destruction of complete
clutches led to nest desertion and subsequent
renesting in one of the two cases described here.
Macedo et al. (2005) argue that individuals failing
to lay or fertilize eggs in the group nest may
engage in infanticide (or ovicide) to initiate a
renesting in which they may improve their
reproductive success. The ovicide described here
is consistent with either intra-group activity or
with a joining event following clutch completion.

Infanticide of Smooth-billed Ani chicks has not
been directly observed, but we report indirect
supporting evidence. Alternative hypotheses are
possible, but our circumstantial evidence suggests
the chicks were killed by new group members,
unrelated to the offspring, presumably to hasten a
new reproductive bout (Macedo and Melo 1999).

SHORT COMMUNICATIONS

373

These observations are consistent with other
competitive behaviors seen in this communal
nesting species.

Our observations on infanticide support group
joining over a territory take-over by a rival group.
Habitat did not appear to be in shortage. We have
observed vacant territories that are used some
years and not others during ongoing research
suggesting that Smooth-billed Ani habitat was not
saturated in our study area (J. S. Quinn, unpubl.
data). Observed aggression associated with other
temtory intrusion events included apparent tres-
passing by neighbors, or group joining events by
one to several individuals (J. S. Quinn, unpubl.
data). We have not observed a temtory take-over.
The group membership following the “immigra-
tion event” reported here included both original
members and new birds. Subsequent observations
did not include within group aggression, although
one of the presumed immigrants was later seen
nest building on another temtory suggesting the
group was not completely stable following the
immigration.

ACKNOWLEDGMENTS

We thank the U. S. Fish and Wildlife Service staff at
Cabo Rojo National Wildlife Refuge and Fred Schaffner for
logistical support. Funding for this work was provided by
NSERC (to JSQ). FIK was supported by the Higher
Education Commission, Islamabad, Pakistan.

LITERATURE CITED

Bertram, B. C. R. 1975. Social factors inlTuencing
reproduction in lions. Journal of Zoology, London
177:463^82.

Blanchard, L. and J. S. Quinn. 2001. The characteriza-
tion of microsatellite loci in the communally breeding
Smooth-billed Ani (Crotopha^a ani). Molecular Ecol-
ogy Notes 1:152-154.

Clutton-Brock, T. H., P. N. M. Brotherton, R. Smith,
G. M. McIlrath, R. Kansky, D. Gaynor, M. J.
O'Riain, and j. D. Skinner. 1998. Infanticide and
expulsion of females in a cooperative mammal.
Proceedings of the Royal Society of London, Series
B 265:2291-2295.

Davis, D. E. 1940. Social nesting habits in the Smooth-
billed Ani. Auk 57:179-218.

Eggert, A.-K. and j. K. Muller. 2000. Timing of
oviposition and reproductive skew in cobreeding
female burying beetles [Nicrophorus vespilloicJes).
Behavioral Ecology 11:357-366.

Gregory, S. M. and J. S. Quinn. 2005. Microsatelite
isolation from four avian species comparing two
isolation techniques. Molecular Ecology Notes 6:87-89.
Hager, R. and R. A. Johnstone. 2004. Infanticide and

control of reproduction in cooperative and communal
breeders. Animal Behaviour 67:941-949.

Hannon, S. J., R. L. Mumme, W. D. Koenig, and F. A.
Pitelka. 1985. Replacement of breeder and within
group conflict in the cooperatively breeding Acorn
Woodpecker. Behavioral Ecology and Sociobiology
17:303-312.

Hrdy, S. B. 1979. Infanticide among animals: a review,
classification, and examination of the implications for
the reproductive strategies of females. Ethnology and
Sociobiology 1:13^0.

Hrdy, S. B. and G. Hausfater. 1984. Comparative and
evolutionary perspectives on infanticide: introduction
and overview. Pages 13-35 in Infanticide: comparative
and evolutionary perspectives (G. Hausfater and S. B.
Hrdy, Editors). Aldine Press, New York, USA.

Koenig, W. D. 1981. Space competition in the Acorn
Woodpecker: power struggles in a cooperative breed-
er. Animal Behaviour 29:396-409.

Macedo, R. H. F. and C. Melo. 1999. Confirmation of
infanticide in the communally breeding Guira Cuckoo.
Auk 116:847-851.

Macedo, R. H. F., M. Cariello, and L. Muniz. 2001.
Context and frequency of infanticide in communally
breeding Guira Cuckoos. Condor 103:170-175.

Macedo, R. H. F., J. S. Quinn, and M. Lima. 2005.
Reproductive skew and individual strategies: infanti-
cide or cooperation? Acta Ethologia 8:92-102.

Packer, C. and A. E. Pusey. 1983. Adaptations of female
lions to infanticide by incoming males. American
Naturalist 121:716-728.

Peakall, R. and P. E. Smouse. 2006. Genalex 6: genetic
analysis in Excel. Population genetic software for
teaching and research. Molecular Ecology Notes
6:288-295.

PiEROTTi, R. 1991. Infanticide versus adoption: an inter-
generational conflict. American Naturalist 138:1140-
1 158.

Queller, D. C. and K. F. Goodnight. 1989. Estimating
relatedness using genetic markers. Evolution 43:258-
275.

Quinn, J. S. and J. M. Startek-Foote. 2000. Smooth-
billed Ani (Crotophaga ani). The birds of North
America. Number 539.

Quinn, J. S., L. A. Whittingham, and R. D. Morris.
1994. Infanticide in skimmers and terns: generalised
territorial attacks or intergenerational conflict? Animal
Behaviour 47:363-367.

Rowher. S., j. C. Herron, and M. Daly. 1999. Step-
parental behaviour as mating effort in birds and other
animals. Evolution of Human Behaviour 20:367-390.

Schmaltz, G., C. Lentz, and J. S. Quinn. 2008.
Competition and waste in the communally breeding
Smooth-billed Ani: effects of group size on egg-laying
behaviour. Animal Behaviour 76:153-162.

Sherman, P. W. 1981. Reproductive competition and
infanticide in Belding’s ground squirrels and other
animals. Pages 31 1-331 in Natural selection and .social
behavior (R. D. Alexander and D. W, Tinkle, Editors).
Chiron Press, New York, USA.

Stacey, P. B. and T. C. Edwards. 1983. Possible cases of

374

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2. June 2010

infanticide by immigrant females in a group-breeding
bird. Auk 100:731-733.

Trail, P. W., S. D. Strahl, and J. L. Brown. 1981.
Infanticide in relation to individual and flock histories in
a communally breeding bird, the Mexican Jay {Aphelo-
coma iiltramarina). American Naturalist 1 18:72-82.

Vehrencamp, S. L. 1977. Relative fecundity and parental
effort in communally nesting anis, Crotophaga
siilcirostris. Science 197:403-405.

Vehrencamp, S. L. 1978. The adaptive significance of
communal nesting in Groove-billed Anis {Crotophaga
siilcirostris). Behavioral Ecology and Sociobiology
4:1-33.

Vehrencamp, S. L. and J. S. Quinn. 2004. Joint laying
systems. Pages 177-196 in Ecology and evolution of
cooperative breeding in birds (W. D. Koenig and J. L.
Dickinson, Editors). Cambridge University Press,
Boston, Massachusetts, USA.

The Wilson Journal of Ornithology 122(2):374-378, 2010

Flight Speeds of Migrating Red-necked and Horned Grebes

Laurence C. Binford'-'* and Joseph A. Youngman"

ABSTRACT. — Air speed of Red-necked Grebes
{Podiceps grisegena), migrating in “no wind” condi-
tions along a 2.64-km path along Lake Superior,
Michigan in fall averaged 61.5 km/hr. Speed increased
significantly using tail winds adjusted for wind vectors
as the migration season advanced: 68 km/hr in August,
70.1 km/hr in September, and 78.7 km/hr in October.
Horned Grebes {P. auritus) averaged 55.5 km/hr with
no winds and, with tail winds, increased from 63.5 km/
hr in September to 71.5 km/hr in October. Whether the
increases indicate that late fall migrants are time
constrained to complete their movement to wintering
areas rapidly, or are a consequence of having to tJy
faster because of larger body mass at that season,
remain to be fully explored. We also present Bight
speeds of 10 additional species of waterbirds migrating
along the same path. Received 22 April 2008. Accepted
28 October 2009.

The north shore of Keweenaw County, Michi-
gan, at the tip of the Keweenaw Peninsula, is a
leading line for waterbirds migrating from west to
east over Lake Superior (Binford 2006). Fall
migrants fly due east, clo.se to and parallel to the
shoreline, and rarely vary their altitude, direction,
or speed making this an ideal location for studying
flight speed. Red-necked Grebes (Podiceps grise-
gerut) and Horned Grebes (P. aiiritii.s) are among
the common migrants. We took advantage of the
exceptional observation conditions to measure
their flight speed under differing conditions at

' Foster Hall I 19. Museum of Natural Science, Louisiana
Slate University, Baton Rouge, LA 70803, USA.

^3631 1 U.S. Highway 41, Chassell, Ml 49916, USA.

’ Deceased.

several phases of the fall migration. We also
documented speeds for other passage waterbirds.

METHODS

Study Area and Field Methods. — ^The south
shore of Lake Superior in Keeweenaw County,
Michigan trends west-east. We selected two
locations, “Dan’s Point,” 6.4 km west of Copper
Harbor, and “Kirkish Point,” 3.8 km west of
town for observations. These peninsulas project
farther into the lake than others in Keweenaw
County. We chose a spot at each locality within
sight of the other, about 5 m above lake surface,
and on a west-east line paralleling the coast.
Surveyor R. S. Haataja measured the distance at
2640.35 m using Ashteck GPS (Global Position-
ing System) receivers and double-checking the
data with a Topcon GPT 1003 Total Station using
infrared beam and reflector. We used 2,640 m in
our calculations. I. D. Wijayratne located a spot
on the bedrock 14 m due north of each point and
about 5 m above lake level using an optical
transit. We marked these with orange paint and
positioned over each an orange pole 4 cm in
diameter and 3 m tall. We directed a 20X spotting
scope at the pole at each point. We timed birds
using identical, continuously running .stop watch-
es, synchronized at the start of each .session, and
u.sed citizen-band radios to communicate between
stations.

We gathered data on mornings between 13
August and 21 October 1999-2003. Binford
manned the west point, and Youngman (replaced
by Jake Musser in 2003) was at the east point.
We used binoculars to detect grebe migrants and

SHORT COMMUNICATIONS

375

switched to spotting scopes for counting and
timing. Binford chose which individuals or
Bocks to time, spacing his choices to give
Youngman sufficient time to complete the
previous observation. Binford detected subjects
arriving from the west. When the lead bird’s
head was aligned with his pole, he used a
stopwatch to record the time to the nearest
second. He then radioed Youngman and gave
him an unique observation number, total flock
size, and the identity of included species,
including the lead bird, and the approximate
distance offshore (“close,” “moderate,” or
“far”). Youngman used the last datum to locate
flocks, which he was usually able to detect
quickly, giving ample time for verification and
recording time of passage, along with wind and
temperature data. Each member of a flock was
considered to have the same time as the lead bird
if the distance between the leader and the
following birds remained the same (data from
shifting flocks were eliminated).

Youngman used an anemometer (Skywatch
Meteos, JDC Electronics, Yverdon-les-Bains,
Switzerland) to measure wind velocity and
direction. He recorded high and low wind speeds
during a 15-sec period immediately after a flock
passed his pole and used the means to calculate air
speed. Winds <3 km/hr, or too variable to
ascertain direction, were considered 0 speed. Only
observations with west or north-northwest, or east
to east-northeast winds were used. Elight speeds
with southerly components were excluded be-
cause the land mass sheltered the anemometer,
although not birds over the lake.

We minimized problems inherent in timing
birds over a measured course (Schnell 1965).
Clicking a stopwatch at exactly the same instant a
bird reached a pole was difficult. However, both
observers had the same problem and responded at
about the same time in relation to bird-pole
position, the distance actually measured was
nearly the same as pole to pole. We doubt this
difference exceeded 1 sec, which is negligible
considering the time (2-1- min) for most birds to
complete the course. Most recorded birds flew
<30 m above the water, sufficiently low that wind
speeds probably differed little from those at shore
observation points.

We ignored birds flying westward (possibly
slowing to search for foraging areas) and excluded
flocks in which birds changed direction, shifted
internal position, merged with another flock, or

obviously changed speed. We deleted the obser-
vations if there were discrepancies between the
two sightings (e.g., number of birds, identity ol
lead bird, distance offshore).

Statistical Analysis. — We calculated air speed
as recorded ground speed minus vectored tail
wind (W, NW) velocity or plus vectored head
wind (E, ENE) velocity. Vectoring assumed that
due west and east winds were 100% effective,
east-northeast 75%, and northwest 50%. We used
one-way analysis of variance (ANOVA, Minitab
Version 13 for Windows) to test for significance
of difference in air speed between months, and
between tail, head, and zero winds.

RESULTS

All 70 Red-necked Grebe flocks for which we
obtained useful data were migrating in W (/? =
25,) NW (/? = 26), ENE (/? = 6), E (n = 6), or
zero {n — 7) winds. Thirteen were of single
individuals. 32 contained 2-14 individuals, and 6
contained 17-35 birds; 13 flocks were mixed with
other species. There was no significant difference
in flight speed among these groups. Elight speeds
averaged ~ 60 km/hr in “no wind” conditions
throughout the fall migration, but tail wind speeds
were progressively greater as the season advanced
(Table 1): August — 61.8 km/hr; September —
70.0 km/hr; October — 78.7 km/hr. The differenc-
es were statistically significant (F = 41.89; df =
2, P = 0.001). Air speeds were unrelated to wind
conditions, at least in August and September, as
indicated by similar tail wind velocities in August
(10.8 km/hr) and September (10.7 km/hr).

All 17 Horned Grebe flocks were moving in
NW (/? =14), E (/? =1), or zero (u = 2) winds.
Twelve contained 3-8 individuals, and one
contained 14; four flocks consisted of mixed
species, total individuals ranging from two to
nine. This species, perhaps because of its slow
speed and unique behavior of flying low. at times
below wave crests, rarely joined other species,
and did not lead a mixed-species flock; no other
species joined a flock composed predominately of
Horned Grebes. “No wind” speed averaged
55.5 km/hr in October. Tail wind speed increased
from 63.5 km/hr in September to 71.5 km/hr in
October (F = 21.14; df = 1, F = 0.001), even
though similar wind speeds prevailed in both
months (9.3 km/hr and 9.0 km/hr, respectively).

Data for other species of waterbirds were
obtained during the same periods (Table 1).

376

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 2. June 2010

•3

O

o

o.

r--

so

in

p

■ — ■

'5

r-i

in

p

d

in

m

p

*3

+ 1

vC

d

—

+1

sd

mi

d

Q

sO

in

' — '

sO

m

m

"3 C/5 ^

C w

-3

<N

m

p

00

5

— i

00

00

d

!7 S ^

G

o

in

so

m

m

-3

U

O

—

c.

■J-.

o

r--

o

n

V-

"2

1

ci

m

.—v

—

ON

cn

'.c

-3

in

ON

—

(n

ni

d

p

p

■.A

+ 1

sd

+ 1

sd

+1

d

d

U.

sO

so

m,

vO

<

—

oo

r^i

ni

oo

E

d

d

d

sd

d

c

o

in

so

in

in

m

U

!>

<N

E

*3 G — r

nj

—

00

ON

CN

c C/5 s:

O

+1 'll

>

ni

sC

d

p

d

00

p

o

■3 c

-3

+ 1

T

+1

d

+ 1

t

00

—

ll ^

cn

p

p

p

■3

in

—

sd

ml

d

L.

o

1

CJ

C-

r?

-3

—

d

cn

3

U

oo

p

cn

u

D.

(N

— '

rd

—

r-i

r-

—

-3

+ 1

d

+ 1

d

+ 1

sd

J

3

in

m,

in

m

' — '

E

oc

—

p

m,

o

>

G

in

d

d

m

in

in

m

mi

mi

T3

u.

r3

—

m

—

oo

(N

i>

“3

o

m

•—

00

3

"t

in

00

p

p

ri

p

Olj

•>2

oc

rd

o

sd

in

p

mi

p

d

m

. zz

+1

+ 1

+ !

d

+ 1

d

+ 1

mi

+ 1

d

+1

d

c5

<

NO

00

00

r'

Uh

0^

r-

O'!

00

nj

—

r-

4;

ON

si

m

1

p

r*"

oc

d

— !

d

d

d

00

d

sd

00

d

00

—

sC

m,

NO

mj

r-

m,

sO

m>

sO

in

'J'.

■73

15

u.

w

■3

Q _

E

00

m

s.

•^C/5 5

o

in

ON

—

—

00

^ +1 u
_ - Wl

3

ni

in

p

o

r|

p

'o

d

00

ni

OO

ni

p

00

00

d

—

ri

—

■« « 3

3

+ 1

d

in

+ 1

d

+ 1

d

+1

nO

+ 1

sd

L.

f-^ Cti

U

W

c.

nj

+ 1

T

T

n

J.

1

oc

00

p

r|

00

p

ri

00

rr

00

3

C

d

d

d

sd

d

00

d

sd

d

sd

iT.

5

-u

-u

O-

y:

E

in

—

00

— >

V

.£Z

*3

Q.

m

o

ri

n

p

E

p

p

p

r--

00

E

ir

w

in

d

p

00

d

ml

O

mi

p

d

p

3

o

+1

ml

+ 1

+ 1

d

+ 1

3

+ 1

+ 1

oo

On

00

00

—

G

1

in

o

m.

r-

1

O

p

—

ON

p

ci

LU

3

d

ni

d

d

sd

d

ri

sd

sd

ri

d

oc

r-

sO

00

ON

00

sO

r^

NO

CQ

<

H

o

JC

V5

u

1)

L.

u

a

00

fU

a.

c/5

"3

0/

3

<

a.

(i>

cn

o

o

O)

u.

O

O

cn

r-

o

ri

r'

-i*

TD

1

<U

1

T

r^<

1

1

F

ri

“O

D

—

ON

n

Ui

O

—

d::

X

ri

o

^

(N

m

p

mj

d

ml

p

P ^

p

P

p ^

mi

in

1

+ 1

Nd

00 —

d

—

00 —

m,

1

sO

nO

r-

1

m

00

(N

00

p

d

d

00

m

NO

m

--'N

O (N

0

(10

0

(i:

0

(2:

o 3

d

d

S'

r',

p

p

p ^

p

sd

d *”

d

— .

d ”

ri

+ 1

Nd

NO ^

r-

r--

1

in ^

1

r-

4

1

p

rvi

1

p

p

00

00

d

mi

d

so

so

sD

so

m

rT

nO

(N

ON

p

p p-

p

rj*

00 ^

, '

d

00 ^

d

NO

1

d ^

1

+ 1

1

rr

cr 00

m

p

d

— d

00

— •

d

NO

00

— , ,

p

o

P ^

p

sd

in

P

in

sd

d

—

+ !

NO

in ^

so

so

[

m ^

so

1

NO

1

m

ON

oo

ON

p

mi

sd

d

sd

d

so

in

in

m,

lo

d

n

so

r-

p

—

mi

d

p

p

d

00

d

p

00

d

m

+ 1

d

d

d

so

ml

+ 1

d

NO

+ 1

d

r-

r-

r-

r-

00

m

4

1

00

4

4

m

1

p

-4

00

sd

d

00

ri

00

00

r-

so

so

nO

NO

so

r-

NO

S'

d

d

S

00

mi

m

d

—

rj

rj

p

d

p

in

d

o

+ !

d

nO

sd

d

d

+ 1

d

4-1

d

1

n

1

1

1

00 '

(N

1

o

1

p

1

p

1

p

p

1

00

p

1

p

d

d

sd

00

oc

NO

z;

d

nO

so

r^

ri

p

p

r-i

r^i

p ^

o

p

ri

p

r|

p

d

oo

d

d

d

sd

d

d

00

d

d ”■

o

4-1

00

00

ON

r-

oo

+ 1

r-

4-1

m

1

si

1

m

r-

1

m.

oo

5

Nd

3

O

sd

3

d

sd

00

00

r--

r-

r-

00

r-

p

CJ

c

o

S

C

O

OJ

OIJ

5

a

d

•o

D-

•5

I

G

o

o

cn

"O

o

I

<u

OX)

3

C3

5

u.

CJ

OD

c

Co

O

<L)

c

1^3

CJ

O

c

o

E

c

a

CJ

'G

s

1

c

G

G

cn

Urn

o

O

O

on

G

4

G

Co

■'d

o

5

o

U

a

OJ

E

<

P

d)

OJ

o

P

3

(U

u.

o

p

d

3

CO

§

5

TABLE I. Conlinued.

SHORT COMMUNICATIONS

377

Q

•a CO

o § =

•o Q
c 00

'i Tl

S s

2S ^

ON

in

ON

in

00

(N

<N

ro

O

r-«

r-

r--

n

+ 1

00

+ 1

(N

NO

NO

O NO
vO NO

IN

r-

+1

ON
I

cn ON

NO

in

ON

+ 1 ^
in On

+ 1 S

I

ON

d

I

NO ON
ON 00

NO
+ 1

r-

no ri
oc 00

<u

C3

0

>>

CJ

•0

Co

CJj

u.

U

5

Q

c

"O

(U

0

-a

0

-5

0

"2

0

■y‘.

>3

(J

Sc

c

"Tv

C3

CO

*C3

0

r-“

ID

u

u

§

CJJ

c

c

G

C

P

0

tS

X)

•o

ID

§

CO

U

OC

DISCUSSION

Measured flight speeds for grebes are almost
non-existent, and no data for either the Red-
necked or Horned grebe are presented in recent
Birds of North America compilations (Stout
and Nuechterlein 1999, Stedman 2000). We
found flight speeds in windless conditions
averaged 61.2 km/hr for Red-necked Grebes
and 55.5 km/hr for Horned Grebes (vs. 60 km/
hr for Eared Grebes [P. nigricollis]; Cullen et
al. 1999).

Air speeds for both species increased through
the fall migration period, even when tail winds
were considered. This increase may have
relevance to the question of whether migrants
fly at minimum power velocity (V^p), maxi-
mum distance velocity (V,na), or somewhere in
between (e.g., Alerstam 1990, 2003). Tail wind
data, for example, suggest that Red-necked
Grebes fly slower in August, just as they are
arriving at the Great Lakes (where many stop
and molt), and that speeds increase progressive-
ly in Septeir,ber and October, as the grebes
prepare to leave for wintering areas on the
Atlantic Coast. The Horned Grebe had a similar
pattern for September-October. This pattern
might indicate that grebes in early fall empha-
size energy conservation, whereas in late fall
migration, when time to reach wintering areas is
constrained, the ability to fly faster and cover
long distances rapidly is more important.
However, larger birds necessarily fly faster than
smaller birds of the same species, because flight
speed is proportional to mass (Rayner 1988).
Data on weight changes in migrating Red-
necked and Horned grebes are not available.
Eared Grebes lay on large fat stores while at fall
staging areas, so that body mass is 50% greater
on departure than on arrival (Jehl 1988, 1997;
Jehl et al. 2003; Jehl and Henry 2010). If Red-
necked and Horned grebes show a similar
pattern, the expected increase in air speed
between arriving and departing birds would
approximate 5%, which is much less than we
found in early versus late fall samples. Addi-
tional data on mass and body composition of
these species at different seasons are needed to
move the discussion forward.

Regardless of whether the increased speed is
due to behavioral adjustment, increased body
mass, or both, seasonal differences in flight speed
can be expected. Detailed interspecific or intra-

378

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2, June 2010

specific comparisons must use data from the same
time period or from birds in the same physiolog-
ical condition. For this reason, the data in Table 1
should be used cautiously.

ACKNOWLEDGMENTS

We thank I. D. Wijayratne and R. S. Haataja for surveying;
Jaw-chuan Lin for statistical analyses; J. V. Remsen. S. J.
Stedman. and J. R. Jehl Jr. for reviews; Amy Slagle, Don
Slagle, and M. J. Hester for logistical support; and the Copper
Country Audubon Club (Houghton, Michigan, USA) for
funding. Jehl and Stedman revised the final manuscript after
the unexpected death of the senior author (LCB). Original
data on flight speed and conditions are filed at the Museum of
Natural History, Louisiana State University, Baton Rouge.

LITERATURE CITED

Alerstam, T. 1990. Bird migration. Cambridge University
Press, Cambridge, United Kingdom.

Alerstam, T. 2003. Bird migration speed. Pages 253-267
in Avian migration (P. Berthold, E. Gwinner, and E.
Sonnenschein, Editors). Springer-Verlag, Berlin, Ger-
many.

Binford, L. C. 2006. Birds of the Keweenaw Peninsula,
Michigan. Miscellaneous Publication Number 195.
Museum of Zoology, University of Michigan, Ann
Arbor, USA.

Cullen, S. A., J. R. Jehl Jr., and G. L. Nuechterlein.
1999. Eared Grebe {Podiceps nigricollis). The birds of
North America. Number 433

Jehl Jr., J. R. 1988. Biology of the Eared Grebe and
Wilson’s Phalarope in the nonbreeding season: a study
of adaptations to saline lakes. Studies in Avian
Biology 12:1-74.

Jehl Jr., J. R. 1997. Cyclical changes in body composition
in the annual cycle and migration of the Eared Grebe
{Podiceps nigricollis). Journal of Avian Biology
28:132-142. ^

Jehl Jr., J. R. and A. E. Henry. 2010. The postbreeding
migration of Eared Grebes. Wilson Journal of
Ornithology 122:217-227.

Jehl Jr., J. R., A. E. Henry, and H. I. Ellis. 2003.
Optimizing migration in a reluctant and inefficient
flier: the Eared Grebe. Pages 199-209 in Avian
migration (P. Berthold, E. Gwinner, and E. Son-
nenschein, Editors). Springer-Verlag, Berlin, Ger-
many.

Rayner, j. V. M. 1988. Form and function in avian flight.
Current Ornithology 5:1-66.

SCHNELL, G. D. 1965. Recording flight-speed of birds by
Doppler radar. Living Bird 4:79-87.

Stedman, S. J. 2000. Horned Grebe (Podiceps cmritus). The
birds of North America. Number 505.

Stout, B. E. and G. L. Nuechterlein. 1999. Red-necked
Grebe (Podiceps grisegena). The birds of North
America. Number 465.

The Wilson Journal of Ornithology 122(2):378-380, 2010

Thermoregulatory Behavior in Migratory European Bee-eaters

{M crops apiaster)

Reuven Yosef

ABSTRACT. — Birds in hyper-arid environments
have acute problems of energy and water balance, and
thermoregulate both physiologically and behaviorally. 1
report on European Bee-eaters (Merops apiaster)
engaged in a previously unreported thermoregulatory
behavior of diving into the sea and in salt ponds with
high levels of salinity. This behavior may also explain
the previously reported, but unexplained, finding of bee-
eaters inside a tiger shark (Galeocerdo ciivier) in the
Red Sea. The.se observations should instigate future
experiments on the subject of .selective use of salt water
for evaporative cooling and thermoregulatory behavior
by desert birds. Received 27 August 2009. Accepted 9
December 2009.

' International Birding and Research Centre in Eilat, P. O.
Box 774, Eilat 88000. Israel; e-mail: ryo.sef@eilatcity.co.il

Animals in deserts use a variety of mechanisms
to maintain their body temperature within phys-
iologically acceptable limits and minimize water
loss (Louw and Seely 1982). Heat stressed birds
engage in panting or gular fluttering to affect
evaporative cooling (Tieleman 2002, pers. obs.).
Other physiological adaptations in hot environ-
ments include greater tolerance of tissues to high
temperatures and dehydration, heterothermy (fluc-
tuation of body temperature in response .to
environmental stress), reducing insulation, and
appropriate coloration (Louw and Seely 1982,
Maclean 1996, Tieleman 2002).

Behavioral thermoregulation is a suite of
behavioral activities that minimize the exposure
of the organism to heat stress and reduce water
loss at high temperatures (Wunder 1979). It is

SHORT COMMUNICATIONS

379

common to see birds exploit cooler terrestrial
micro-climates and bathe in water to remove body
heat by evaporative cooling (Dawson and Barthol-
omew 1968, Cook 1997).

The International Birding and Research Centre
in Eilat, Israel has reclaimed a local garbage dump
as a Bird Sanctuary to help migratory bird
populations (Cherrington 2001, Tal 2002, Ro-
senzweig 2003). The site has commercial salt
ponds and a fresh water lake that are less than
50 m apart. Eilat is at the southernmost tip of
Israel within the Sahara-Arabian desert belt
characterized by extreme temperatures (mean =
9.5 to 39.4° C; Israel Meteorological Service,
www.ims.gov.il), very low precipitation (mean
annual rainfall = 17 mm), and high solar radiation
(mean = 6.32 kWh m7day, Eaiman et al. 2003).

I report observations of migrant European Bee-
eaters (Merops apiaster) diving into the sea and
salt ponds with high levels of salinity (>15%).

OBSERVATIONS

European Bee-eaters migrate through the Eilat
bottleneck in both seasons in large numbers. They
are among the last migrants in spring (May-Jun)
and their passage at Eilat coincides with the peak
heat waves of late spring-early summer when
ambient temperatures can range from 1 8 to 48° C.
Temperatures peak on many occasions during
periods of khamsins, dry, hot, and dusty local
wind that carries great quantities of sand and dust
from the deserts with speeds to 160 km/hr, and a
rise of temperatures as much as 20° C in 2 hrs.

I observed 57 flocks of bee-eaters diving into
the shallow salt ponds in the Bird Sanctuary or
off-shore in the Red Sea in spring 2005-2009 for
a total of 36 khamsin days with temperatures
above 43° C. Similar observations were also made
from the shores of the Red Sea. However, the
outcome of this behavior far out at sea was too
distant to describe properly. Bee-eaters were also
present on days when temperatures were under
42° C, but were not ob.served diving into the salt
ponds. All observations were conducted with
Swarovski 15 X 56 binoculars.

Bee-eaters perched in the shade of Prosopis
alba and Acacia radciiana trees surrounding the
salt pond at the southern extremities of the Bird
Sanctuary. I considered them heat stressed
because of their open beaks, tongues thrust out
and gular fluttering. Every 15-20 min small
groups of bee-eaters, 8-39 birds/flock, would
take off and dive head first into the shallow salt

pond. On most occasions the whole flock would
take off in a circling Bight, calling, and diving
into the salt pond. Rarely did single birds remain
behind when the flock was engaged in diving.

The bee-eaters flew into the shadiest parts of
the trees and perched there for the next 30-45 min
after flying from the water, or paddling to the
shoreline. The bee-eaters spread their wings while
wet and allowed the breeze and the low humidity
of the area (15-18%) to dry the feathers during
the period immediately after they perched. I
believe the birds achieved evaporative cooling
since bee-eaters did not engage in gular fluttering
for the first 15-20 min after diving.

The maximum depth of the salt pond was
200 mm. I observed 576 bee-eaters diving into the
ponds with 389 flying from the water surface and
1 1 1 unable to do so and had to paddle ashore
where they lay on the ground for several minutes
with their wings stretched out before flying to a
perch. The remaining 76 bee-eaters were dazed
for many minutes, possibly from hitting the
bottom of the pool. They surfaced from the dive
but did not move, and had trouble reaching the
shoreline. These birds were collected, rehabilitat-
ed (allowed to rest in the shade and fed sugar
water), and returned to the wild a day or two later.
On days when I was not present at the Bird
Sanctuary, 117 carcasses washed up on shore
suggesting this is a common phenomenon.

I saw no bee-eaters diving into the nearby fresh
water lake which is much deeper (~ 4 m), and
which has fish.

DISCUSSION

Ery (1984, 2001) reviewed bathing (a.k.a.
splashing) by .several species of bee-eaters. The
major reasoning forwarded by most published
reports was that bee-eaters were feeding on
insects or fish. Ery (2001:313) states that cooling,
dislodging parasites, drinking or cleansing are
alternatives to what seems “to be essentially some
.sort of social comfort activity" in the gregarious
bee-eaters. My observations of bee-eaters diving
into salt ponds, devoid of fish or other possible
food sources, during the hottest days of the season
strengthens the argument that in the desert
“splashing" has a thermoregulatory function.

The described behavior may explain the finding
of bee-eaters inside a tiger shark (Galeocerdo
cuvier) in the Red Sea (Yosef et al. 2002). Perhaps
the Bee-eaters involved had dived into the Red
Sea for cooling and were unable to take off from

380

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2. June 2010

the surface and were eaten by the shark, or had
drowned and collected from the body of water.
The latter appears to be the more realistic option
because tiger sharks are considered to be noctur-
nal foragers (Compagno and Niem 1998).

The ability of some birds to discriminate fresh
and saline waters is debated (Maclean 1996).
However, most studies have investigated the
tolerance of birds to saline water - drinking
tolerance, body mass loss, serum and urine
osmolarity, and fecal water loss (Maclean 1996).
It has also been suggested that a possible added
value could be the removal of ectoparasites in the
salt water. However, this has yet to be proved. I
was unable to find literature on how birds use
water sources and modify their behavior to
thermoregulate in extreme temperatures. King
and Chastant (2008) report on cormorants (Pha-
lacrocorax spp.) and darters (Auhinga spp.) that
regurgitate water into their chicks open mouths
and over their bodies on warm and windless days
during the hottest hours to help them maintain
their body temperatures.

We still lack information on the ability of
seawater to have a cooling effect on heat-stressed
birds. My observations should instigate future
experiments on the subject of evaporative water
loss and thermoregulatory behavior in desert
birds.

ACKNOWLEDGMENTS

I thank Berry Pin.show for di.scu.ssing the .subject. Susan
Craig, Keith Bildstein, and an anonymous reviewer
improved an earlier version of the paper.

LITERATURE CITED

Cherrington, M. 2001. To save a watering hole. Pages 8-
14 in The best American science and nature writing
2001 (E. O. Wilson, Editor). Houghton Mifflin
Company, New York, USA.

Compagno, L. J. V. and V. H. Niem. 1998. Carcharhinidae.
Requiem sharks. In FAO identification guide for

fishery purposes. The living marine resources of the
western central Pacific. (K. E. Carpenter and V. H.
Niem, Editors). FAO, Rome, Italy.

Cook, W. E. 1997. Avian desert predators. Springer-
Verlag, Heidelberg, Germany.

Dawson, W. R. and G. A. Bartholomew. 1968.
Temperature regulation and water economy of desert
birds. Pages 357-394 in Desert biology. Volume 1 (G.
W. Brown Jr., Editor). Academic Press, New York,
USA.

Faiman, D., D. Feuermann, P. Ibbetson, A. Zemel, A.
Ianetz, V. Liubansky, and I. Setter. 2003. Data
processing for the Negev radiation survey: Ninth year.
TMY-V4. Israel Ministry of Energy and Infrastructure,
Jerusalem.

Fry, C. H. 1984. The bee-eaters. T & AD Poyser,
Staffordshire, United Kingdom.

Fry, C. H. 2001. Family Meropidae (Bee-eaters). Pages
286-341 in Handbook of the birds of the world.
Volume 6, Mousebirds to Hornbills (J. del Hoyo, A.
Elliott, and J. Sargatal, Editors). Lynx Edicions,
Barcelona. Spain.

King, D. T. and J. E. Chastant. 2008. Water regurgitation
by adult Double-breasted Cormorants: a possible
mechanism to assist in chick thermoregulation.
Waterbirds 31:283-284.

Louw, G. N. AND M. K. Seely. 1982. Ecology of desert
organisms. Longman House, Essex, United Kingdom.

Maclean, G. L. 1996. Ecophysiology of desert birds.
Springer-Verlag, Heidelberg, Germany.

Rosenzweig, M. L. 2003. Win-win ecology-how the
earth’s species can survive in the midst of human
enterprise. Oxford University Press, New York, USA.

Tal. a. 2002. Pollution in a promised land-an environ-
mental history of Israel. University of California Press,
Los Angeles, USA.

Tieleman, B. I. 2002. Avian adaptation along an aridity
gradient-physiology, behavior and life history. Dis-
sertation. Rijks Universiteit, Groningen, The Nether-
lands.

Wunder, B. a. 1979. Evaporative water loss from birds:
effects of artificial radiation. Comparative Biochem-
istry and Physiology Part A: Physiology 63:493-494.

Yosef, R., D. Zakai, M. Rydberg-Heden, and R.
Nikolajsen. 2002. An unusual record of a European
Bee-eater Merops apiaster from Eilat-inside a tiger
shark Galeocerdo ciivier. Sandgrouse 24:140-142.

SHORT COMMUNICATIONS

381

The Wilson Joiinial of Ornithology 122(2):38l-384, 2010

Population Status of Chuck- will’ s-widow (Caprinmlgus carolinensis) in

the Bahamas

William K. Hayes, Elwood D. Bracey,“ Melissa R. Price,' Valerie Robinette,'

Eric Gren,' and Caroline Stahala-^

ABSTRACT. — The Chuck-wiH’s-widow {Caprinuil-
giis carolinensis) in the Bahama Islands has been
regarded as a rare to uncommon winter visitor. We
conducted breeding season surveys on the three largest
northern islands (North Andros, Grand Bahama, and
Great Abaco) to examine the status of this species. We
encountered singing birds on most survey routes on all
three islands, suggesting that sizeable breeding popula-
tions are widespread in the northern Bahamas with an
aggregate estimate of 500-1.000 pairs. Our density
estimates were somewhat less than those from the
primary range in the United States, suggesting either a
lower carrying capacity in the Bahama Islands or
recently established populations that have yet to reach
carrying capacity. Received 24 August 2009. Accepted 3
January 2010.

Much remains to be learned about Chuck-
wilTs-widow (Caprimitlgus carolinensis) and
other nightjars (Caprimulgiformes), largely be-
cause of their nocturnal and secretive habits. The
species breeds across much of the eastern United
States and winters in southern Florida, Central
America, the West Indies, and northern South
America. Gaps remain in our knowledge of the
species’ distribution, abundance, and habitat use
despite its distinctive nocturnal song, which
greatly facilitates detection (Straight and Cooper
2000). The species has gradually expanded its
range in North America in the past century
(Straight and Cooper 2000).

The status of Chuck-will’s-widow in the
Bahama Islands remains unclear. The species is
considered a rare to uncommon winter visitor
(Brudenell-Bruce 1975, Emien 1977, Buden 1987,
White 1998, Straight and Cooper 2000). Northrop
(1891) collected a male with enlarged testes

' Department of Earth and Biological Scicncc.s, Loma
Linda University, Loma Linda, CA 92350, USA.

-2305 Royal Palm Way, Treasure Cay, Abaco. Bahamas.
-’Department of Biological Science, Florida State Uni-
versity, Tallahassee. FL 32306, USA.

“'Corresponding author, e-mail: whayes@llu.edu

during the breeding season (15 May) on Andros,
but was uncertain of nesting. Paterson (1972)
observed a bird incubating a single egg on
Andros, but Bond (1973:3, 1984:20) rejected the
record as a misidentified Antillean Nighthawk
{Chordeiles gundlachii). However, Buden (1992)
and Buden and Sprunt (1993) heard calling birds
and flushed a pair during visits to several Exumas
cays in April through June, suggesting possible
breeding. Calling birds subsequently heard from
Eebruary to July on several islands, documented
in North American Birds seasonal reports, led to
increased speculation of nesting. Breeding was
eventually confirmed for Grand Bahama (Norton
1999; Norton and White 2001, 2002) and Abaco
(Norton 2000). However, several questions re-
main unanswered. Were these breeding records
extralimitai, or indicative of larger breeding
populations? If the latter, how large might these
populations be, and how widespread are they?

The objective of our study was to clarify the
population status of Chuck-will’s-widow in the
Bahama Islands. Our results, when compared to
similar surveys from the species’ continental
range in the United States, provide inferences
about the bird’s abundance and breeding status.

METHODS

We conducted breeding season surveys on the
three largest northern islands: North Andros,
Grand Bahama, and Great Abaco. We followed
generally practiced protocols for surveys of
nightjars (e.g.. Cooper 1981, Wilson 2008).
Surveys were conducted along 8-20 km routes
that included 9-12 brief stops (usually 10) of 2-
5 min duration at 1 km or greater intervals.
Surveys, with a few exceptions, were completed
during >50% moon illumination. Surveys were
conducted during either the hour before sunrise or
within 1-2 hrs following sunset, depending on
when the moon was above the horizon. Weather
varied from clear to overcast but with relatively
low wind and absence of rain. We recorded the

382

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2, June 2010

GPS (Global Positioning System) coordinates,
time, and number of calling birds for each stop.
We broadcast a recording of the song by speaker
or whistled a cruder imitation of the song by
mouth for most stops. We did not differentiate
between spontaneous and solicited calls, as both
were frequently detected. Calling birds could at
times be heard at >500 m; we tried not to double-
count individuals on consecutive stops, but this
may have resulted in slight underestimation bias.

We (MRP, VR, and EG) conducted seven
surveys on North Andros during 2009. These
included: 10 and 12 May (mornings) south of
Nichol’s Town; 17 May (morning) between
Mastic Point and San Andros Airport; 2 June
(evening) west of Stafford Creek; and 3, 4, and 5
June (evenings) between Stafford Creek and Love
Hill. These routes traversed primarily Caribbean
pine (Piiuis carihaea) forest habitat with scattered
openings, but coppice, marsh, and residential
areas were present at some stops.

WKH conducted three surveys during 2007 on
Grand Bahama. These included: 1 July (morning)
from Bassett Cove to the oil storage facility east
of Riding Point; 1 July (evening, 12 stops, no
moonlight) near the water towers at Lucayan
Estates; and 2 July (morning, 9 stops) in Lucaya
along West Beach and Midshipman roads. All
routes were in primarily pine forest with scattered
openings; however, the first included occasional
coppice and residential areas, and the third
transected a rural residential area.

EDB conducted repeated surveys along two
routes (9 stops each) between Treasure Cay and
Leisure Lee on Great Abaco during 2009. These
included four evening surveys on 5 May, 10 May
(no moonlight), I June, and 5 June. Additional
evening surveys repeated in July and August
yielded no calling birds and were excluded from
analysis. Routes included pine forest with .some
coppice. CS conducted two surveys in 2009
further south on Great Abaco. These were 14
July (morning) along Great Abaco Highway
between Crossing Rocks and Abaco National
Park, and 15 July (morning) along Mid Road
towards Hole-in-the-Wall. The routes passed
through pine forest with variable understory, and
several stops included coppice.

We derived the following variables for each
route: total number of calling birds; mean birds
per stop along route; proportion of stops with bird
detections; number of birds calling per occupied
stop (with birds calling); and density (pains/knr).

Numbers were adjusted for routes of 9 or 1 2 rather
than 1 0 stops to reflect the standard 1 0-stop route
(i.e.. birds/stop multiplied by 10). We assumed
that all calling individuals were detected within a
500-m radius to estimate population density
(calling males = pairs/kni"). Thus, the number
of calling birds detected at each stop corresponded
to a minimum density of pairs per 0.785 km- (area
of circle with 500 m radius). This assumption is
subject to .several factors influencing detection
rate, including some birds being heard at greater
distances and many individuals not calling during
the brief stops.

RESULTS

We encountered 42 calling birds along the 14
survey routes (Table 1 ). The number of calling
birds per 10-stop route varied from zero to 12.2.
Calling birds were detected on four of the seven
(57%) Andros routes and on 100% of the three
Grand Bahama and four Abaco routes. Grand
Bahama appeared to host the highest density of
individuals. One nine-stop route through a rural
residential area yielded 1 1 calling birds. The
maximum number of calling birds recorded per
stop was two, and this was consistent for all three
islands. Three birds were heard at several stops on
Grand Bahama but, in each case, one was judged
to be a bird heard at the previous stop and was not
counted.

Repeated surveys along two routes on Great
Abaco illustrated nightly and .seasonal variation.
Calling birds after sun.set numbered four (5 May),
five (10 May), six (1 Jun), and seven (5 Jun) for
the 18 stops of both routes combined. No birds
were detected on three additional surveys (22 and
30 Jul, 5 Aug), and these surveys were excluded
due to lateness of season. Our mean estimate of
calling birds for the first four surveys (5.5) was
21% below the maximum count (7), suggesting
detection of 79% or less for individual surveys.

DISCUSSION

Standardized surveys provide the best under-
standing of the relative abundance of Chuck-
will's-widow, and only males are known to give
the primary song (Straight and Cooper 2000). A
number of factors can affect calling rates and
influence detection rates and density estimates.
Male calling peaks in late spring in continental
populations and may decrease markedly in July
(e.g.. Cooper 1981, Straight and Cooper 2000).
Both calling rale and duration of calling through-

TABLE I. Chuck-will's-widows on North Andros (7 survey routes). Grand Bahama (3 routes), and Great Abaco (4 routes), Bahama Islands. Values include .v ± SD and rangi
within parentheses for birds per 10-stop route and associated variables.

SI lORT COMMUNICATIONS

383

in

NO

'e

00

f^i

+ 1

?

in

CO

ri

'E

d

d

c.

00

o

.3

—

—

c

_ — *

Q

+1

+1

+1

00

00

—

CO

o

d

d

c.

•y;

<N

CJ

-o

—

1

1

’5.

3

U

o

CO

in

U

o

o

d

d

c.

+1

+ 1

+ 1

-o

o

CO

y:

oo

3

o

r>

n

O

m

1

CO

O

r>

1

a

—

o

CN

33

;§

3}-

$

ro

mi

d.

+ 1

+ 1

+ 1

V‘.

ON

CO

NO

#

oi

GO

CO

n

oT

m,

rn

—

d

CL

d

1

1

ri

!

c

o

d

y;

C

00

00

CL

in

o

d

d

d

'El

+1

+ 1

+1

-t

o

o

—

in

d

r\

r i

1

1

mj

1

£

1

O

1

n

1

C,

00

CO

y.

-o

—

ini

d

5

+ 1

+ 1

+ 1

vO

o

mi

d

E

o

sC

sO

"E

H

■X.

CL

c

y

o

—

OO

r-

CO

O

h-

ir.

Q

d

U.

CJ

£

a

-C

C

3

<

<

•o

C3

1 .

U

O

QD

a

P

z

a

6

u

>

•C

'U

-3

O

-a

OJ

rr.

3

-D

a.

V

3

c

c

f-

Gill the night increase significantly with greater
moonlight (Cooper 1981, Mills 1986, Wilson and
Watts 2006, Woods and Brigham 2008).

Our surveys suggest a substantia) breeding
population of Chuck-will’s-widows exists on the
large northern islands in the Bahamas. Abundance
varied locally, but calling males were widespread
on each of the large islands surveyed (North
Andros, Grand Bahama, and Great Abaco). Birds
calling in summer on New Providence (e.g.,
Norton and White 2002) and the Exumas (Buden
1992, Buden and Sprunt 1993) suggest breeding
populations on these islands as well.

Relative densities from our surveys (,v = 0.14-
0.56 birds/stop among the 3 islands) were
generally lower than surveys within the primary
range of continental populations. Cooper (1981),
from 20 surveys in northern Georgia, reported 0-
1.75 calling individuals per stop {x = 1.37 for 6
nights with >0.5 lunar illumination). James and
Neal (1986) listed counts of 1.77 and 2.32/stop in
Searcy and Clay counties, Arkansas, respectively.
Baumgartner and Baumgartner (1992) reported
counts of 1.67 and 1.98/stop in northeast and
southeast Oklahoma, respectively. More recently,
Wilson (2008) reported on coordinated surveys
from seven states having two or more surveys
conducted in 2007. These included: Florida with
1.54 birds/stop, Alabama with 1.50, South Car-
olina with 1.30, Georgia with 0.98, Mississippi
with 0.90, Virginia with 0.44, and North Carolina
with 0.42. Our surveys differed somewhat from
tho.se of Wilson (2008) in that stops were spaced
more clo.sely and listening time was briefer. We
also used tape playback or vocal imitation to help
stimulate calling.

We do not know whether the current breeding
population in the Bahama Islands represents a
recent colonization or has been overlooked. The
relatively low densities compared to continental
populations suggest either a lower carrying
capacity in the Bahama Islands or recently
established populations that have yet to reach
carrying capacity. Population densities at the
expanding northern periphery of the continental
range are generally low (Hunt 2008). Future
surveys could improve our knowledge on these
possibilities. Two birds collected from Cuba
during the breeding .season with enlarged gonads
suggest possible breeding there as well (Straight
and Cooper 2000).

Our density estimates provide for a rough
approximation of the total population in the

384

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2. June 2010

Bahamas. The estimates are limited by three
major factors (see also Wilson 2008): (1) some
birds were detected at distances >500 m, resulting
in overestimation; (2) not all males present were
calling, resulting in underestimation; and (3)
survey routes may not have been representative
of the habitats available on each island. We
suspect our numbers are underestimates. We
multiplied our estimates by the quantity of
suitable habitat on each island based on detailed
maps. As many as 214 pairs exist on Andros (0. 18
pairs/km“; 5,957 km" total area; 20% of total area
assumed to be suitable habitat), 390 pairs on Grand
Bahama (0.7 1 pairs/km"; 1 ,373 km" total area; 40%
suitable habitat), and 192 pairs on Abaco (0.38
pairs/km"; 1,681 km" total area; 30% suitable
habitat). New Providence (207 km") presumably
hosts a breeding population, and smaller popula-
tions may exi.st on other islands. We conclude the
Bahama Islands supports an aggregate population in
the vicinity of 500-1,000 pairs.

Clearly, the recent scattered reports of singing
and nesting birds reflect sizeable breeding popu-
lations rather than extralimital records. Efforts
should be undertaken to locate additional popula-
tions on other islands and to examine whether
populations are gradually increasing in size.

ACKNOWLEDGMENTS

We thank the following for their .support: Forfar Field
Station and Insular Species Conservation Society (Andros
research); Grand Bahama Power Company (Grand Ba-
hama): and Bahamas National Trust and Friends of the
Environment (Abaco).

LITERATURE CITED

Baumgartner, F. M. and A. M. Baumgartner. 1992.
Oklahoma bird life. University of Oklahoma Press,
Norman, USA.

Bond, J. 1973. Supplement 18 to the Check-list ot birds of
the West Indies (1956). Academy of Natural Sciences
of Philadelphia. Pennsylvania. USA.

Bond, J. 1984. Supplement 25 to the Check-list ot birds of
the We.st Indies (1956). Academy of Natural Sciences
of Philadelphia. Penn.sylvania. USA.
Brudenele-Bruce. P. G. C. 1975. The birds of New
Providence and the Bahama Islands. Collins, London,
United Kingdom.

Buden. D. W. 1987. The birds of the southern Bahamas.

British Ornithologists’ Union, London, United King-
dom.

Buden, D. W. 1992. The birds of the Exumas, Bahama
Islands. Wilson Bulletin 104:674-698.

Buden, D. W. and A. Sprunt IV. 1993. Additional
observations on the birds of the Exumas, Bahama
Islands. Wilson Bulletin 105:514—518.

Cooper, R. J. 1981. Relative abundance of Georgia
caprimulgids based on call-counts. Wilson Bulletin
93:363-371.

Emlen, j. T. 1977. Landbird communities of Grand
Bahama Island: the structure and dynamics of an
avifauna. Ornithological Monographs 24.

Hunt, P. 2008. Northeast nightjar survey: 2008 results.
Report to the New Hampshire Audubon Society, www.
vtecostudies.org/PDF/Nightjar08.pdf (accessed 4 Au-
gust 2009).

James, D. A. and J. C. Neal. 1986. Arkansas birds: their
distribution and abundance. University of Arkansas
Press, Fayetteville, USA.

Mills, A. M. 1986. The influence of moonlight on the
behavior of goatsuckers (Caprimulgidae). Auk
103:370-378.

Northrop, J. I. 1891. The birds of Andros Island,
Bahamas. Auk 8:64-80.

Norton, R. L. 1999. West Indies Region. North American
Birds 53:436-437.

Norton, R. L. 2000. West Indies. North American Birds
54:426-427.

Norton, R. L. and A. White. 2001. West Indies. North
American Birds 55:370-372.

Norton, R. L. and A. White. 2002. West Indies. North
American Birds 56:496-M97.

Paterson. A. 1972. Nesting of Chuck-will’s-widow on
Andros Island. Bahamas. Auk 89:676-677.

Straight, C. A. and R. J. Cooper. 2000. Chuck-will’s-
widow (Caprimiilgiis ccirolincn.'iis). The birds of North
America. Number 499.

White, A. W. 1998. A birder's guide to the Bahama Islands
(including Turks and Caicos). American Binding
Association, Colorado Springs, Colorado, USA.

Wilson, M. D. 2008. The Nightjar Survey Network:
program construction and 2007 Southeastern Nightjar
Survey results. Center for Conservation Biology
Technical Report Series, CCBTR-08-001 . College of
William and Mary, Williamsburg, Virginia, USA.

W1L.SON, M. D. AND B. D. Watts. 2006. The effect of
moonlight on detection of Whip-poor-wills: implica-
tions for long-term monitoring strategies. Journal of
Field Ornithology 77:207-21 1.

Woods, C. P. and R. M. Brigham. 2008. Common
Poorwill activity and calling behavior in relation to
moonlight and predation. Wilson Journal of Ornithol-
ogy 120:505-512.

SHORT COMMUNICATIONS

385

The Wilson Journal of Ornithology 122(2):385-387, 2010

Yellow Rails Wintering in Oklahoma

Christopher J. Butler,' " Lisa H. Pham,' Jill N. Stinedurf,' Christopher L. Roy,'
Eric L. Judd,' Nathanael J. Burgess,' and Gloria M. Caddell'

ABSTRACT. — Yellow Rails (Cotnrnicops novebor-
acensis) were recently discovered to migrate through
southeastern Oklahoma in small numbers during fall
with a few records into December and a single record in
January. We made seven trips (3 in Nov 2008 and 1/
month from Dec 2008 through Mar 2009) to Red Slough
Wildlife Management Area in McCurtain County
(Oklahoma) to catch and band Yellow Rails. Twenty-
five Yellow Rails were banded and birds were observed
during each month. Rails were encountered in areas
dominated by Sporohohts spp. averaging 44 cm in
height in areas with 4 cm or less of standing water.
Yellow Rails appear to overwinter in small numbers in
McCurtain County, Oklahoma —300 km north of the
Gulf Coast. Received 22 June 2009. Accepted 15
October 2009.

Yellow Rails (Coturnicops noveboracensis) are
secretive, nocturnal birds that breed in the
northern United States and Canada with a disjunct
population in southcentral Oregon (Bookhout
1995). The total population of Yellow Rails is
estimated to be only 17,500 individuals (Butcher
et al. 2007). The species is on the Audubon’s
Society’s Red WatchList (Butcher et al. 2007) and
is a species of special concern in Canada
(Environment Canada 2006), and most of the
states where breeding occurs (Grace et al. 2005).
Yellow Rails generally leave breeding areas
between mid-August and November (Stenzel
1982, Bookhout 1995, Popper and Stern 2000,
Goldade et al. 2002) and winter along the coast,
from Texas to North Carolina (Bookhout 1995).
Several studies of Yellow Rails have been
conducted in breeding areas but little is known
about their migratory or wintering ecology
(Bookhout 1995).

The first Yellow Rail recorded in Oklahoma
was collected on 7 March 1842 in what is now
Delaware County (Tomer 1959). More than a
century passed before Yellow Rails were again
reported in Oklahoma. In 1954, one was reported

' Department of Biology, University of Central Okla-
homa. Edmond, OK 73034, USA.

^Corresponding author; e-mail: cbutlerl l@uco.edu

killed by a mowing machine (Heck and Arbour

2008) . These authors summarized published dates
of occuiTence for the 20th century and reported
one record from the 1 960s, three records from the
1970s, no records during the 1980s, and four
records from the 1990s. W. D. Arbour began
conducting regular rope drags to flush Yellow
Rails beginning in 2001 at Red Slough Wildlife
Management Area (WMA) in McCurtain County,
Oklahoma. Yellow Rails were observed annually
at this location from 2001 to 2008 (Heck and
Arbour 2008). They are now considered to be
regular autumn migrants from 15 October through
26 November in southeastern Oklahoma (OBRC
2004).

Yellow Rails were also recorded during De-
cember in 2 years (2004 and 2005) at Red Slough
WMA and a single record exists from January
(2004) at this location (Heck and Arbour 2008).
We hypothesized that Yellow Rails may over-
winter in small numbers in southeastern Okla-
homa given the species has been recorded as late
as mid-January in McCurtain County.

METHODS

Red Slough Wildlife Management Area (33°
44' 05" N, 94° 38 13" W; Fig. 1) is in McCurtain
County, Oklahoma, and consists of —1,000 ha of
moist soil management areas. 1.160 ha of
(reforesting) bottomland hardwoods, and 165.6 ha
of re.servoirs for a total of 2,325.6 ha (USDA

2009) . Red Slough WMA is cooperatively man-
aged by the U. S. Forest Service, Natural
Resources Conservation Service, Oklahoma De-
partment of Wildlife Conservation, and Ducks
Unlimited (USDA 2009).

We traveled to Red Slough WMA seven times
between November 2008 and March 2009 to
capture and band Yellow Rails (3 times in Nov
and once monthly from Dec through Mar).
Vegetation was sampled at nine locations where
birds were flushed during November 2008. A 1 X
I ni" frame was placed at each location where we
flushed Yellow Rails and vegetation height of

386

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

FIG. 1. Red Slough Wildlife Management Area (black star) in McCurtain County, Oklahoma.

each species was measured as well as percent
cover following Daubenmire (1959). Yellow Rails
may have run before flushing (which could
potentially bias the results), but the habitat
appeared to be homogeneous and any bias was
expected to be minimal.

We began capturing Yellow Rails —30 min
after sunset and continued our efforts for 3 hrs
each night. We used a 12-m nylon rope weighted
with bottles (filled with rocks) which two people
dragged through damp, grassy areas where rails
had previously been encountered. Rope-dragging
caused rails to flush and we were able to shine a
handheld spotlight on the area where the bird
landed and catch them in a small handheld net.
Each bird was banded with a USGS aluminum
band, classified to age following Pyle (2008), and
released.

RESULTS

Twenty-five Yellow Rails were banded from
November 2008 through March 2009 (Table 1).
One bird banded in November 2008 was recap-
tured in December 2008 and one bird banded in

TABLE 1. Trapping effort and capture.s of Yellow
Rails, McCurtain County, Oklahoma, 2008-2009.

# hours

Month of sampling New birds Recaptures

November 2008

18

13

2 (both banded
earlier in Nov)

December 2008

6

9

1 (banded in Nov)

January 2009

6

1

1 (banded in Dec)

February 2009

6

0

0

March 2009

6

2

0

December 2008 was recaptured in January 2009.
A single Yellow Rail was flushed in February
2009 but we were unable to catch it. During
November, most birds captured (9 of 13) were
classified as adults (i.e., AHY) following Pyle
(2008). In contrast, from December 2008 through
March 2009, most birds captured (8 of 12) were
classified as young (i.e., HY/SY).

Fourteen genera in eight plant families were
identified in the nine vegetation plots. Seven plots
were dominated by grasses (60-100% cover),
mainly Sporoholus spp. (dropseed), some identi-
fied as Sporoholus compositus. Other grasses on
these seven plots included Steinchisma hians
(gaping grass) and Panicum spp. (panicgrass).
One plot was dominated by flatsedge (Cyperus
spp.) with 85% cover, and another by an
unidentified Polygonaceae with 97.5% cover.
Other plants encountered included Ambrosia spp.
(ragweed), Ammannia spp. (toothcup), Andropo-
gon virginiciis (broomsedge bluestem), Boltonia
asteroides (white doll’s daisy), Eleocharis spp.
(spike-rush), Hypericum drummondii (St. John’s
wort), Juncus spp. (rush), O.xalis spp. (wood
sorrel), and Tridens soictus (long-spike tridens).

The dominant vegetation averaged (± SE)
44.1 ± 4.6 cm in height in areas where rails were
captured. Ten Yellow Rails were captured in areas
with no standing water and fifteen were captured
in shallow water <4 cm deep.

DISCUSSION

Yellow Rails appear to winter in small numbers
at Red Slough WMA in McCurtain County,
Oklahoma. This is considerably farther north than
previously reported wintering areas, because Red

SHORT COMMUNICATIONS

387

Slough WMA is ~ 300 km north of the Gulf
Coast. It is unknown whether this is a recent
wintering range expansion or whether Yellow
Rails have wintered here but not been previously
detected. Most birds banded during winter (Dec-
Mar) were HY/SYs, which suggests this species
may undergo differential migration and young
winter farther north than adults.

Yellow Rails are reported to use the drier areas
of Spartina spp. marshes along the coast (Book-
hout 1995). Radiotelemetry in coastal Texas
suggested that Yellow Rails prefer areas domi-
nated by saltgrass {Distichlis spicata) and gulf
cordgrass {Spartina spartinae) (Grace et al. 2005).
Wayne (1905: 396-397) collected wintering
Yellow Rails near Charleston, South Carolina
away from the immediate coast in a “low, wet
piece of open land with a dense growth of short,
dead grass”. He collected 58 Yellow Rails
between 1903 and 1918 at this freshwater locality
(Post 2008). Heck and Arbour (2008) suggested
Yellow Rails in Oklahoma occur primarily in fall
panicum (Panicum dichotomiflorum) and long-
leafed rushgrass {Sporobolus asper) during fall
migration. We generally found Yellow Rails in
areas dominated by Sporobolus spp. rather than
Panicum spp., although the number of higher taxa
(families and genera) at our sites was fairly high.
It is possible that Yellow Rails choose wintering
habitat based on structural components (e.g., areas
dominated by grassy vegetation <50 cm in height
with little or no standing water) rather than by
dominant plant genera, as suggested for other rail
species (Conway 1995, Melvin and Gibbs 1996).

Yellow Rails are difficult to detect and may
winter farther inland than previously reported.
Given that they overwinter in small numbers in
southeastern Oklahoma, observers elsewhere in
the southeastern United States should be alert for
the possibility of Yellow Rails overwintering in
damp grassy areas.

ACKNOWLEDGMENTS

We thank W. D. Arbour, R. A. Bastarache, and the U. S.
Forest Service for assistance in locating potential sites tor
Yellow Rails. We thank E. J. Davis, M. E. Stone, B. M.
Wilcox, B. D. Rising, J. A. Young, S. M. Still, D. R. Wood,
and his ornithology class for assi.stance capturing Yellow
Rails. We thank the Oklahoma State University Fore.st
Resources Center for providing accommodations in
McCurtain County. This research was supported by the
Office of Research and Grants at the University of Central
Oklahoma. T. A. Bookhout and an anonymous reviewer
provided helpful comments on the manuscript.

LITERATURE CITED

Bookhout, T. A. 1995. Yellow Rail (Coiiiniicops nove-
horacensis). The birds of North America. Number 139.
Butcher, G. S., D. K. Niven, A. O. Panjabi, D. N.
Pashley, and K. V. Rosenberg. 2007. WatchList:
The 2007 WatchList for United States birds. American
Birds 61:18-25.

Conway, C. J. 1995. Virginia Rail (Rallus limicola). The
birds of North America. Number 173.

Daubenmire, R. 1959. A canopy-coverage method of
vegetation analysis. Northwest Science 33:43-64.
Environment Canada. 2006. Species of special concern:
Yellow Rail. Quebec Region, Canadian Wildlife Service,
Environment Canada, Quebec City, Quebec, Canada.
http://www.qc.ec.gc.ca/faune/oiseaux_menaces/html/rale_

jaune_e.html (accessed 4 Februaiy 2009).

Goldade, C. M., j. A. Dechant, D. H. Johnson, A. L.
Zimmerman, B. E. Jamison, J. O. Church, and B. R.
Euliss. 2002. Effects of management practices on
wetland birds: Yellow Rail. USDI, Geological Survey,
Northern Prairie Wildlife Research Center, James-
town, North Dakota, USA.

Grace, J. B., L. K. AuLAtN, H. Q. Baldwin, A. G.
Billock, W. R. Eddleman, A. M. Given, C. W.
Jeske, and R. Moss. 2005. Effects of prescribed fire
in the Coastal Prairies of Texas. Open File Report
2005-1287. USDI, Geological Survey, Lafayette,
Louisiana, USA. http://www.nwrc.usgs.gov/factshts/
2005-1287-fire-in-coastal-texas-report.pdf (accessed 8
October 2009).

Heck, B. A. and W. D. Arbour. 2008. The Yellow Rail in
Oklahoma. Bulletin of the Oklahoma Ornithological
Society 41:13-15.

Melvin, S. M. and J. P. Gibbs. 1995. Sora (Porzana
Carolina). The birds of North America. Number 250.
Oklahoma Bird Records Committee (OBRC). 2004.
Date guide to the occurrences of birds in Oklahoma.
Fourth Edition. Oklahoma Ornithological Society,
Tulsa, USA.

Popper, K. J. and M. A. Stern. 2000. Nesting ecology of
Yellow Rails in southcentral Oregon. Journal of Field
Ornithology 7 1 :460^66.

Post, W. 2008. Wintering ecology of Yellow Rails based
on South Carolina specimens. Wilson Journal of
Ornithology 120:606-610.

Pyle. P. 2008. Identification guide to North American
birds. Part 11. Anatidae to Alcidae. Slate Creek Press,
Point Reyes, California, USA.

Stenzel, j. R. 1982. Ecology of breeding Yellow Rails at
Seney National Wildlife Refuge. Thesis. Ohio State
University, Columbus, USA.

Tomer, J. S. 1959. An Oklahoma record of the Yellow Rail.
Auk 76:94-95.

U.S. Department of Agriculture (USDA). 2009. Red
Slough Wildlife Management Area. USDA, Forest
Service, Hot Springs, Arkansas, USA. http://www.fs.
fed.us/r8/ouachita/natural-resources/redslough/info.
shtml (accessed 26 May 2009).

Wayne, A. T. 1905. Notes on certain birds taken or seen
near Charleston, South Carolina. Auk 22:395^00.

388

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2, June 2010

The Wilson Journal of Ornithology 122(2):388-39L 2010

Breeding of the Giant Laughingthriish (Garrulax maximus) at Lianhuashan,

Southern Gansu, China

Jie Wang,* " Chen-Xi Jia,' Song-Hua Tang,* Yun Fang,* and Yue-Hua Sun*’^

ABSTRACT. — We describe the nest sites, nests,
eggs, and incubation and provisioning behavior of the
endemic Giant Laughingthrush (Garrula.x maximus) in
a coniferous forest (2,850-2,950 m asl) at Lianhua-
shan, southern Gansu, central China. We found seven
shallow bowl-shaped nests in Picea-Ahies trees, 4.0 ±
1.5 m (x ± SD, n = 1) above the ground during May
and June 2003, 2007, and 2008. Clutch size was 2.2 ±
0.4 unspotted blue eggs (2-3, n = 6) of which 1.4 ±
0.5 nestlings hatched (1-2, n = 7), and 1.0 ± 1.0
young fledged (0-2, n = 7). Three nests failed,
possibly due to predation or abandonment during
prolonged rainfall. Both males and females incubated
clutches; nest attentiveness during the day decreased
from 92.6 ± 0.9% before hatching to 59.4 ± 1.5%
during days 3-7 of the nestling period. Both parents
fed the nestlings ( 1 .0 ± 1.0 times/hr) and consumed the
feces (0.3 ± 0.5 times/hr) during the 7-15 days after
hatching Received 28 March 2009. Accepted 26
October 2009.

The endemic Giant Laughingthrush {Garrulax
maximus) is distributed from southern Gansu and
southeastern Qinghai to southern and southeastern
Tibet and northern Yunnan in southern and central
China (Zheng 2005, Thompson 2007). Habitats
occupied include open broadleaf and mixed
broadleaf-coniferoLis forests, bamboo {Phyllosta-
cliys spp.) scrub in broadleaf evergreen forest, oak
(Quercus spp.) forest with scattered conifers, and
relatively open broadleaf evergreen forest with
shrub understory at 2,135^,115 m elevation
(Thompson 2007).

The Giant Laughingthrush is the largest laugh-
ingthrush (Cheng et al. 1987, Rasmussen and
Anderton 2005) and was previously treated as a
subspecies of Spotted Laughingthrush (G. ocella-
tus) (Ali and Ripley 1996). Little is known about
its natural history and breeding except one

' Key Laboratory of Animal Ecology and Conservation
Biology. Institute of Zoology. Chinese Academy of
.Sciences. Beijing 1 00 1 01. China.

^Graduate School of Chinese Academy of Sciences,
Beijing 100049, China.

^Corresponding author; e-mail: sunyh@ioz.ac.cn

description based on a single nest in a bamboo
clump, which contained two eggs (Ludlow and
Kinnear 1944). We describe the nests, eggs,
nestlings, and nesting behavior of this species in
an alpine conifer-dominated forest in southern
Gansu, China.

METHODS

Study Area. — Our study area was in the
Lianhuashan Nature Reserve (34° 57' N, 103°
46' E) in southern Gansu Province, China. The
forested area occurs on north, northeast, and
northwest-facing slopes; only shrubs and grasses
grow on south-facing slopes. Coniferous forest is
the most prevalent cover type and is dominated by
Dragon spruce {Picea asperata) and Earges fir
{Abies fargesii). The other cover types are; (1)
coniferous-deciduous forest, including spruce, fir,
Himalayan birch {Benda utilis), and willow {Salix
spp.); and (2) shrublands including willow, sea
buckthorn {Hippophae rhamnoides), and barbeiry
{Berheris spp.). The study area has been described
by Sun et al. (2003). The mean annual tempera-
ture is 5. 1-6.0° C, with a maximum of 34.0° C
and minimum of —27.1° C. The climate is
semiarid, and the annual precipitation is about
65 cm.

Field Procedures. — We located seven nests (6
during incubation and 1 during the nestling stage)
by following nesting-related activities of adults or
by systematically checking individual trees in the
conifer-dominated forest between 2,850 and
3,100 m elevation during three breeding seasons
(Apr-Jul in 2003, 2007, and 2008). We inspected
nests every 3-5 days to identify hatching and
fledging dates and, if possible, cause of nest
failure.

Incubation or brood care was documented using
data loggers (Tinytag Plus 2 Gemini Data
Loggers, West Sussex, UK) at two nests, one of
which was monitored for 194 hrs during 1 1 days.
The probe was placed among the eggs to take a
temperature reading every 15 sec. Length of on-
and off-nest bouts, and behaviors of provisioning

SHORT COMMUNICATIONS

389

TABLE 1. Giant Laughingthrush nests observed in the Lianhuashan Nature
during May and June 2003, 2007, and 2008. DBH = diameter at breast height.

Reserve, southern Gansu,

central China

Nest #

Date found

Stage

Nest tree

Nest

Clutch size

Nest fate

Species

DBH

(cm)

Height above
ground (m)

Distance to
the trunk (m)

Eggs

Nestlings

Fledglings

1

25 May 2003

Incubation

Spruce

12

7.0

0.40

3

2

2

Successful

2

27 May 2003

Incubation

Spruce

10

3.0

0.04

2

2

2

Successful

3

9 June 2003

Nestling

Spruce

14

4.5

0.10

2

2

Successful

4

24 May 2007

Incubation

Fir

32

3.8

2.00

2

1

1

Successful

5

9 June 2007

Incubation

Spruce

13

4.0

0.05

2

1

0

Abandoned

6

26 May 2008

Incubation

Fir

45

2.4

2.10

2

1

0

Depredated

7

3 June 2008

Incubation

Spruce

26

3.5

2.20

2

1

0

Depredated

young and removing feces were recorded at one
nest by observation (33.7 hrs during days 6-10,
12-13, and 15 of the nestling period) and at
another nest by video recorder (1.9 hrs on the 15th
day of the nestling period). All observers and
equipment were camouflaged using branches and
leaves.

Nest characteristics were measured after termi-
nation of nesting and included height of the nest,
distance of the nest to trunk, and height and

diameter at breast height (DBH) of the nest tree.
All data are expressed as A ± SD.

RESULTS

Nests were placed in Dragon spruce or Farges
fir with three in branches of large conifers (DBH
= 34 ± 10 cm) at a height of 2. 4-3. 8 m above
ground and 2. 0-2.2 m from the tmnk. Four nests
were in small spruce (DBH = 12 ± 2 cm) nearly
touching the trunk at a height of 3. 0-7.0 m

FIG. 1 . Nest with two eggs of the Giant Laughingthrush in a Dragon spruce (Picea asperata) at Lianhuashan, southem
Gansu, central China. Photograph by Chenxi Jia.

390

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

On-nest bout Off-nest bout - )IG Nest attentiveness

0

O)

1-^

Q)

0

13

<

0

13

0

0

0

Day

FIG. 2. Length of on- and off-nest bouts and nest attentiveness of the Giant Laughingthrush at one nest (found in 2007)
during the incubation and nestling periods. Day 0 = batching day. The numbers on the asterisks are the time (hrs)
monitored by data loggers whereas the numbers on/under the bars are the sample sizes of on/off-nest bouts.

(Table 1). Nests were shallow and bowl-shaped
(Fig. 1) with an inside diameter of 1 1.6 ± 0.7 cm
( 1 0.8- 1 2.2 cm, n = 4), outside diameter of 1 9.3 ±
0.9 cm (18.5-20.0 cm), inside depth of 5.1 ±
1.3 cm (4. 0-6. 5 cm), and outside height of 9.4 ±
1 . 1 cm (8.0-10.5 cm). Two nests were 78 and 80 g
in wet weight, respectively. The inside bowl of the
nest was lined with thin strands of bamboo,
madder {Ruhia spp.), and spiraea (Spiraea spp.)
stems, whereas the outer bowl and foundation
were mainly of twigs, 25 cm (10-38 cm) in
length, mostly of honeysuckle (Lonicera spp.)
with a few from spruce, fir, and birch.

Clutch size was 2.2 ± 0.4 eggs (2-3, n = 6)
with 1.4 ± 0.5 nestlings hatched (// = 7) and 1.0
± 1 .0 young fledged (n = 7). Ten unspotted blue
eggs were 33.5 ± 1.5 mm (31.6-36.2 mm) in
length, 22.9 ± 0.7 mm (21.6-24.4 mm) in
diameter, and 8.8 ± 0.8 g (7.8-10.0 g) in weight.
One nest was abandoned in the nestling stage,
possibly due to prolonged rainfall (15-22 Jun
2007). One nest was possibly predated as one egg
had a small hole in the eggshell during the

incubation stage. The other nest was predated at
2344 hrs (Beijing Time, revealed by data logger)
during the nestling stage with the nest overturned
and feathers of the adult scattered on the ground.

Mean length of on-nest bouts decreased from
59.4 ± 39.1 min (1.5-136.0 min) before hatching
to 27.6 ± 18.0 min (3.0-72.5 min) during the
nestling period, whereas mean off-nest bouts
increased from 4.6 ± 2.8 min (0.5-10.5 min) to
15.6 ± 10.1 min (2.5-45.0 min) (Fig. 2). Nest
attentiveness during the day decreased from 92.6
± 0.9% before hatching to 59.4 ± 1.5% during
days 3-7 of the nestling period. The female began
diurnal activities at 0613 hrs (±28 min, n = 9)
and began nocturnal brooding at 1931 hrs
(±28 min, n = 9).

Both parents provisioned nestlings and re-
moved (or consumed) feces at a frequency of
1.2 ± 1.5 (0-6) and 0.5 ± 0.9 (0-3) times/hr,
respectively, based on observation (32 hrs) and
video recording (1.9 hrs) during days 7-15 of the
nestling period. The nestling period appeared to
be 17-18 days (2 nests).

SHORT COMMUNICATIONS

391

DISCUSSION

Nests in conifers were higher than the nest in
the bamboo clump (1.2 m) reported by Ludlow
and Kinnear (1994) and the nest of the Spotted
Laughingthrush (<2 m) (Ali and Ripley 1996).
Nest materials used by the Giant Laughingthrush
were similar to those used by the closely related
Snowy-cheeked Laughingthrush (G. sukatschewi)
(Bi et al. 2003) and the Spotted Laughingthrush
(Ali and Ripley 1996), but did not include moss,
birch leaves, or strips of honeysuckle bark, which
are frequently used by the Elliot’s Laughingthrush
(G. elliotii) (Jiang et al. 2007). We did not find
nests in bamboo clumps, which were quite short
(<1.8 m) and scattered under the conifer trees or
in small patches within the forest. The bamboo in
this area had been nearly clear-cut by the local
people 5 years earlier.

The unspotted blue color of the eggs was
similar to the Snowy-cheeked Laughingthrush (Bi
et al. 2003) but possibly differed from the Spotted
Laughingthrush, which has eggs that are spotless
or with a few chocolate-brown specks near the
broad end (Ali and Ripley 1996). Mean clutch
size was similar to that of the Spotted Laugh-
ingthrush (normal = 2 eggs, Ali and Ripley
1996), but less than the high-altitude Brown-
cheeked Laughingthrush (G. henrici) (2.6 eggs,
Lu et al. 2008), Snowy-cheeked Laughingthrush
(3.5 eggs; Jie Wang, unpubl. data), and Elliot’s
Laughingthrush (3.4 eggs, Jiang et al. 2007).

Both parents incubated eggs and brooded
nestlings, similar to most other laughingthrushes
(Thompson 2007). Frequency of provisioning
nestlings was lower than for the Snowy-cheeked
Laughingthrush (1.2 vs. 7.9 times/hr; Bi et al.
2003). One reason may be that adult Giant
Laughingthrushes seemed to be vigilant and prone
to decrease their activities during the nestling
stage, as the frequency of provisioning young
estimated from video recordings (7 times in
1.9 hrs) was higher than from observation (31
times in 32 hrs).

ACKNOWLEDGMENTS

We are grateful to Dan Strickland for help with English
and J,-L. Li, Y.-X. Jiang, Sh. -Q. Zhao, and the .staff of
Lianhuashan Nature Reserve tor field assistance. We
appreciate the constructive comments of K. E. Miller, C.
E. Braun, and one anonymous reviewer on earlier drafts of
the manuscript. The work was supported by National
Natural Science Foundation of China (Grant 306201301 10).

LITERATURE CITED

Ali, S. and S. D. Ripley. 1996. Pages 29-31 in Handbook
of the birds of India and Pakistan. Second Edition.
Volume 7. Oxford University Press, New Delhi, India.
Bi, Z.-L., Y. Gu, C.-X. Jia, Y.-X. Jiang, and Y.-H. Sun.
2003. Nests, eggs, and nestling behavior of the Snowy-
cheeked Laughingthrush {Garrulax sukatschewi) at
Lianhuashan Natural Reserve, Gansu, China. Wilson
Bulletin 115:474^77.

Cheng, T. H., Z. Y. Long, and B. L. Zheng. 1987. Pages
113-115 in Fauna Sinica (Aves. Volume 11. Passer-
iformes, Muscicapidae II, Timaliinae). Science Press,
Beijing, China.

Jiang, Y.-X., Y.-Z. Zhu, and Y.-H. Sun. 2007. Notes on
reproductive biology of Elliot’s Laughingthrush at
Zhuoni, Gansu. Sichuan Journal of Zoology 26:555-
556.

Lu, X., G.-H. Gong, and X.-H. Zeng. 2008. Reproductive
ecology of Brown-cheeked Laughingthrush (Garrulax
henrici) in Tibet. Journal of Field Ornithology 79:152-
158.

Ludlow, F. and N.-B. Kinnear. 1944. The birds of
southeastern Tibet. Ibis 86:43-86, 176-208, 348-389.
Rasmussen, P. C. and J. C. Anderton. 2005. Page 413 in
Birds of south Asia: the Ripley Guide. Volume 2.
Smithsonian Institution, Washington, D.C., USA, and
Lynx Edicions, Barcelona, Spain.

Sun, Y.-H, J. E. Swenson, Y. Fang, S. Klaus, and W.
SCHERZINGER. 2003. Population ecology of the Chi-
nese Grouse, Bonasa sewerzowi, in a fragmented
landscape. Biological Conservation 110:177-184.
Thompson, H. S. S. 2007. Page 253 in Handbook of the
birds of the world. Volume 12. Picathartes to tits and
chickadees (J. del Hoyo, A. Elliott, and D. A. Christie.
Editors). Lynx Edicions, Barcelona, Spain.

Zheng, G.-M. 2005. Page 253 in A checklist on the
classification and distribution of the birds of China.
Science Press, Beijing, China.

392

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 2, June 2010

The Wilson Journal of Ornithology 1 22(2):392-395, 2010

First Description of the Nest of Jocotoco Antpitta {Grallaria ridgelyi)

Harold F. Greeney'-'''^ and Mery E. Juina J' -

ABSTRACT. — The Jocotoco Antpitta (.Grallaria
ridgelyi) is an endangered and poorly studied inhabitant
ot montane bamboo (Chusc/uea spp.) thickets of
extreme southeastern Ecuador. There is nothing known
of the breeding biology of Jocotoco Antpitta apart from
a single record of a dependent juvenile, and we describe
a nest of this species for the first time. The nest was a
bulky cup composed primarily of dead plant materials
and tirmly supported by a large clump of epiphytes on
the side of a dead trunk. The single nestling was
provisioned at a rate of 1.96 feedings/hr during the final
5 days prior to Hedging. Two adults brought food to the
nestling, often delivering prey from a nearby worm-
feeding station created by the Jocotoco Foundation. The
pair we studied may breed twice a year, but this may
have been facilitated by their proximity to the artificial
feeder. Received I Ma\ 2009. Accepted 17 September
2009.

The Jocotoco Antpitta (Grallaria ridgelyi) is
one of the rarest species of this poorly known
genus that encompasses 31 species in the family
Grallariidae (Remsen et al. 2009). Despite its
large size and distinctive plumage, the Jocotoco
Antpitta was only described 10 years ago (Krabbe
et al. 1999), having been overlooked due to its
propensity to forage in the thick undergrowth of
Chu.sc/uea bamboo-covered slopes (Krabbe et al.
1999, Ridgely and Greenfield 2001, Heinz et al.
2005, Freile et al. 2010). The Jocotoco Antpitta is
apparently confined to southeastern Ecuador and
is considered Endangered by BirdLife Interna-
tional (2009). The breeding biology of the
Jocotoco Antpitta remains undocumented with
only a single description of a juvenile accompa-
nying two adults (Greeney and Gelis 2()()5b). We
de.scribe the first nest of Jocotoco Antpitta from

' Yanayacu Biological Station & Center for Creative
Studies. Cosanga, Napo Province, Ecuador, c/o 721 Loch y
Amazonas, Quito. Ecuador.

^ Fimdacion de Conservacion Jocotoco, Avenida Los
Shyris N37-146 y El Comcrcio. Quito, Ecuador.

'Current address; University of Nevada, Department of
Biology 0314, 1664 North Virginia Street. Reno, NV
89.‘i.‘)7. USA.

■‘Corresponding author; e-mail: revmmo.ss@yahoo.com

the Tapichalaca Biological Reserve in southeast-
ern Ecuador.

METHODS

We made all observations at the Tapichalaca
Biological Reserve (04° 30' S, 79° 10' W) near
Valladolid in extreme southeastern Ecuador. The
teiTain is typically steep for this elevation in the
Andes (~ 2,500 m) with vegetation characterized
by high epiphyte density, a canopy height of 15-
30 m, and a dense understory comprised largely of
CIntsquea bamboo (Krabbe et al. 1999). We made
nest measurements to the nearest 0.5 cm and
behavioral observations at the nest using a tripod-
mounted video camera placed ~ 5 m from the
nest. We recorded 16.3 hrs of video from 1030 to
1800 hrs on 10 November 2008 (5 days prior to
fledging) and from 0700 to 1715 hrs the day
before fledging.

RESULTS

We observed two adult Jocotoco Antpittas on
the morning of 5 November 2008 gathering food
items at the worm feeder and black-light trap
adjacent to the Tapichalaca Lodge (~ 2,500 m
asl). Adults carried multiple invertebrates from
this feeding area, leaving in the direction of the
subsequently discovered nest. We again ob.served
adults carrying prey in the same direction on 6
November.

On 8 November 2008 at 1000 hrs MEJ found
the nest, 30 m from the feeders, which contained a
single large nestling. The nestling was well
feathered at the time of discovery, cinnamon on
the back, chestnut on the crown and upper breast,
and finely barred with black on all visible areas
(Fig. 1). Most of the ocular area was still bare
with just a faint hint of the white malar pattern of
adults. The upper mandible was dark with a pale
orange tip while the lower mandible was mostly
dull orange. The gape was orange but we did not
examine the mouth lining or remove the nestling
for closer inspection of plumage patterns. The nest
was 3.6 m above the ground on the side of a
vertical, rotting tree trunk that was 4.1 m tall

SHORT COMMUNICATIONS

393

FIG. 1. Nestling Jocotoco Antpitta (G;'a//«n<7
in southeastern Ecuador, 3 days prior to fledging (Photo-
graph by M. E. Juifia).

(56 cm dbh) (Fig. 2). The nest was placed flush
against the trunk and was supported primarily by a
cluster of Tillansia bromeliads and secondarily by
a 4-cm diameter, horizontal branch crossing under
the nest. The nest was a large cup comprised
predominantly of dead plant materials, mostly
dicot and bromeliad leaves. The cup was lined
internally with a thick layer of fine black rootlets
and bare fern rachi.ses.

Internal measurements of the nest cup were
15.5 X 14.0 cm in width (measured at perpen-
dicular angles) and 12.5 (back) to 17.0 cm (front)
deep. Externally, the nest was 23.5 by 26 cm in
diameter and 18 cm tall. The area around the nest
contained a dense understory and, along with the
dense ~ 20 m canopy, provided nearly 90%
shading of the nest. The nest was directly shaded,
predominantly by Chusquea bamboo, as well as
epiphytic ferns and orchids. The nest was on the
northeast side of the trunk on a northeast facing
slope where it received most of its sun in the
mornings.

FIG. 2. Nesting site of Jocotoco Antpitta {Grallaria
ridgelyi) in southeastern Ecuador. The white arrow points to
the nest itself (Photograph by M. E. Juina).

Transcriptions of video at the nest revealed the
nestling was brooded for only 7% of daylight
hours. We were unable to ascertain if more than
one adult brooded, but we confirmed that two
adults brought food. Adult feeding visits to the
nest were brief (mean ± SD = 2.2 ± 7.5 min),
during which time they frequently probed sharply
or rapidly into the nest lining. The nestling was
provisioned at a rate of 1.96 feedings/hr with
absences from the nest lasting 29.7 ± 24.4 min.
The nestling often followed the adults to the rim
of the nest as they left after delivering food during
the day prior to leaving the nest. In the absence of
adults, the nestling frequently stood and stretched
its wings or shuffled about inside the nest,
occasionally pecking at small invertebrates mov-
ing around the nest.

DISCUSSION

The nest of Jocotoco Antpitta is a deep, bulky
cup similar in form to the described nests of other
Grallaria (Greeney et al. 2008). The nest was well
supported by bromeliads, against the trunk and
branch of a dead tree, most similar to the nest
situations of Rufous Antpitta (G. rufula) (Greeney

394

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2, June 2010

and Gelis 2005a), White-bellied Antpitta (G.
hypoleiica) (Price 2003), Yellow-breasted An-
tpitta (G. flavotincta) (Greeney et al. 2009),
Variegated Antpitta (G. varia) (Protomastro
2000), Pale-billed Antpitta (G. carrikeri) (Wie-
denfeld 1982), Great Antpitta (G. excelsa)
{Koefed and Auer 2004), and some Scaled
Antpitta (G. giiatimalensis) (Dobbs et al. 2001,
2003). The nest was composed predominantly of
dead plant material, most similar to nests of
Chestnut-naped Antpitta (G. nuchalis) (Juina et al.
2009), Chestnut-crowned Antpitta (G. ritficapilla)
(Martin and Greeney 2006), Plain-backed Antpitta
(G. haplonota) (Greeney et al. 2006), Watkins’s
Antpitta (G. watkiusi) (Martin and Dobbs 2004),
Variegated Antpitta (Quintella 1987, Protomastro
2000), Pale-billed Antpitta (Wiedenfeld 1982),
and White-bellied Antpitta (Price 2003), as well
as some nests of Moustached Antpitta (G. alleni)
(Freile and Renjifo 2003, Londono et al. 2004,
Greeney and Gelis 2006), and Scaled Antpitta
(Miller 1963; Rowley 1966; Dobbs et al. 2001,
2003).

Previous observations of an older, dependent
fledgling Jocotoco Antpitta (Greeney and Gelis
2005b) were made on 30 November 2003 (HFG,
Linpubl. data). We also observed young fledglings
accompanying adults to the worm teeder on
multiple days in July and November 2007 and in
April 2008. Observations of juveniles, with the
exception of the previously published report of
Greeney and Gelis (2005b), were all near the
worm feeder maintained by the Jocotoco Foun-
dation which is presumably visited by only one
pair of Jocotoco Antpittas (MEJ, pers. obs.).
These records suggest at least two broods a year.
The effects of these feeders are unstudied despite
the many birding lodges in Ecuador that now use
worm feeders for Gi'cillcivio antpittas. We believe
it is possible the year-round surplus of food
provided by the feeders in our study area may
have encouraged this pair of Jocotoco Antpittas to
reproduce throughout the year, a pattern which
may not be common for pairs without dietary
supplementation. The importance of this possibil-
ity and the long-term repercussions on adult
longevity and health deserve further study at this
site and at others where worm feeders are used.

Clutch size remains undocumented for Jocotoco
Antpitta, but repeated sightings of single fledg-
lings (Greeney and Gelis 2()05b, this study)
combined with our observations of a single
nestling, suggest this species may lay a smaller

clutch than most other GraUaria (Greeney et al.
2008). Alternatively, single fledglings may reflect
brood or clutch reduction from the typical two-
egg clutches of the genus (Greeney et al. 2008).

ACKNOWLEDGMENTS

Field work was supported by the Jocotoco Foundation.
We thank the staff and park guards of the Tapichalaca
Biological Reserve, most especially Franco Mendoza and
Timo Brockmeyer. Preparation of this manuscript was
supported by National Geographic Grant W38-08. The
PBNHS continues to aid our natural history work in
Ecuador. This is publication Number 194 of the Yanayacu
Natural History Research Group.

LITERATURE CITED

BirdLife International. 2009. Jocotoco Antpitta - Bird-
Life species factsheet. http;//www.birdlife.org/datazone/
species/index.html (accessed 1 August 2009).

Dobbs, R. C., P. R. Martin, and M. J. Kuehn. 2001. On
the nest, eggs, nestlings, and parental care of the
Scaled Antpitta (GraUaria giiatimalensis). Ornitologia
Neotropical 12:225-233.

Dobbs, R. C., P. R. Martin, C. Batista, H. Montag, and
H. F. Greeney. 2003. Notes on egg laying, incubation
and nestling care in Scaled Antpitta GraUaria
guatimalensis. Cotinga 19:65-70.

Freile, J. F. and L. M. Renjifo. 2003. First nesting records
of the endangered Moustached Antpitta, GraUaria
alleni. Wilson Bulletin 115:11-15.

Freile, J. F., J. L. Parra, and C. H. Graham. 2010.
Distribution and conservation of GraUaria and Gral-
lariciUa antpittas (Grallariidae) in Ecuador. Bird
Conservation International 21: In press.

Greeney, H. E. and R. A. Gelis. 2005a. A nest of the
Rufous Antpitta GraUaria ritfiUa depredated by a
Turqoise Jay Cyanolyca turcosa. Cotinga 24:1 10-1 1 1.
Greeney, H. E. and R. A. Gelis. 2005b. Juvenile plumage
and vocalization of the Jocotoco Antpitta GraUaria
ridgelyi. Cotinga 23:79-81.

Greeney, H. E. and R. A. Gelis. 2006. Observations on
parental care of the Moustached Antpitta (GraUaria
alleni) in northwestern Ecuador. Ornitologia Neotrop-
ical 17:313-316.

Greeney. H. F., R. C. Dobbs, P. R. Martin, and R. A.
Gelis. 2008. The breeding biology of GraUaria and
Grallariciila antpittas. Journal of Field Ornithology
79:113-129.

Greeney, H. F., M. E. Juina J., S. M. 1. Lliquin, and J. A.
Lyons. 2009. First nest description of the Yellow-
breasted Antpitta GraUaria flavotincta in northwest
Ecuador. Bulletin of the British Ornithological Club
129:256-258.

Greeney, H. E., R. A. Gelis, C. Dingle, F. J. Vaca B., N.
Krabbe, and M. Tidwell. 2006. The ne.st and eggs of
the Plain-backed Antpitta (GraUaria haplonota) in
eastern Ecuador. Ornitologia Neotropical 17:601-604.
Heinz, M., V. Schmidt, and M. Schaefer. 2005. New

SHORT COMMUNICATIONS

395

distributional record for the Jocotoco Antpitta Gral-
larici ridi’clyi in south Ecuador. Colinga 23:24-26.

JuiN.A .1, M. E., J. B. C. Harris, and H. F. Greeney. 2(X)9.
Description of the nest and parental care of the Chestnut-
naped Antpitta (Grcillarici michalis) from southern
Ecuador. Ornitologia Neotropical 20:305-310.

Kofoed, E. M. and S. K. Auer. 2004. First description of
the nest, eggs, young, and breeding behavior of the
Great Antpitta (Gnillaria excelsa). Wilson Bulletin
1 16:105-108.

Krabbe, N. and T. S. Schulenberg. 2003. Family Formicar-
iidae (ground-antbirds). Pages 682-731 in Handbook of
the birds of the world. Volume 8. Broadbills to tapaculos.
(J. A. del Hoyo, A. Elliott, and D. A. Christie, Editors).
Lynx Edicions, Barcelona, Spain.

Krabbe, N., D. J. Agro, N. H. Rice, M. Jacome, L.
Navarrete, and F. Sornoza. 1999. A new species of
antpitta (Formicariidae: Gi allaria) from the southern
Ecuadorian Andes. Auk 116:882-890.

Londono, G. a., C. a. Saavedra-R., D. Osorio, and J.
MartInez. 2004. Notas sobre la anidacion del Tororoi
Bigotudo {Grallaria alleni) en la Cordillera Central de
Colombia. Ornitologia Colombiana 2:19-24.

Martin, P. R. and R. C. Dobbs. 2004. Description of the
nest, egg and nestling of Watkin’s Antpitta Grallaria
watkinsi. Cotinga 21:35-37.

Martin, P. R. and H. F. Greeney. 2006. Description of the
nest, eggs and nestling period of the Chestnut-crowned
Antpitta Grallaria ruficapilla from the eastern Ecua-
dorian Andes. Cotinga 25:47-49.

Miller, A. H. 1963. Seasonal activity and ecology of the
avifauna of an American equatorial cloud forest.
University of California Publications in Zoology
66:1-78.

Price, E. R. 2003. First description of the nest, eggs,
hatchlings, and incubation behavior of the White-
bellied Antpitta (Grallaria hypoleuca). Ornitologia
Neotropical 14:535-539.

Protomastro, j. j. 2000. Notes on nesting of Variegated
Antpitta Grallaria varia. Cotinga 14:39^1.

Quintela, C. E. 1987. First report of the nest and young of
the Variegated Antpitta (Grallaria varia). Wilson
Bulletin 99:499-500.

Remsen Jr., j. V., C. D. Cadena, A. Jaramillo, M. Nores,
J. F. Pacheco, M. B. Robbins, T. S. Schulenberg, F.
G. Stiles, D. F. Stotz, and K. J. Zimmer. 2009. A
classification of the bird species of South America.
American Ornithologists’ Union. http://www. museum.
lsu.edu/~Remsen/SACCBaseline.html (accessed 1
August 2009).

Ridgely, R. S. and P. j. Greenfield. 2001. Birds of
Ecuador. Cornell University Press, Ithaca, New York,
USA.

Rowley, J. S. 1966. Breeding records of birds of the Sierra
Madre del Sur, Oaxaca, Mexico. Proceedings of the
Western Foundation of Vertebrate Zoology 1:107-
204.

Wiedenfeld, D. a. 1982. A nest of the Pale-billed Antpitta
(Grallaria carrikeri) with comparative remarks on
antpitta nests. Wilson Bulletin 94:580-582.

The Wilson Journal of Ornithology 122(2):395-398, 2010

Nesting Ecology of the Grey-backed Shrike {Lanius tephronotus) in

South Tibet

Xin Lu,'-^ Chen Wang,' and Tonglei Yu'

ABSTRACT. — We studied the breeding ecology of
the Grey-backed Shrike (Lanius tephronotus) in alpine
shrub habitats in south Tibet. Shrikes nested between
late May and early July at elevations of 4,010-4,540 m
with a delay in nesting time with increa.sed elevation.
Nests were built mostly in bushes (83%), 0.7-2. 1 m
above ground. Clutch size averaged (± SD) 4.12 ±
0.67, range = 3-5 eggs and was smaller for pairs that
nested later at higher elevations. Incuhation by females
began before clutch competition and lasted 15-18 days.
Nestlings were cared for by both parents tor 14—15 days.
Breeding success, considered as the number of nest
attempts that fledged at least one young, was 45.8%.
Shrikes nesting in alpine habitats, compared with their

'Department of Zoology, College of Life Sciences,
Wuhan University, Wuhan, 430072, China.

-Corresponding author; e-mail: luxinwh@gmail.com

lowland congeners, experienced shorter breeding sea-
sons, produced fewer broods, smaller clutches, and
larger eggs. They followed a life history strategy that
allowed them to compensate for reduced annual
fecundity under harsh conditions. Received 29 June
2009. Accepted I November 2009.

Lanins shrikes occur throughout the world
except for South America and Australia (Lefranc
and Worfolk 1997). Among the world’s Lanins
species, the Grey-backed Shrike (L. tephronotus)
breeds at the highest elevations in the Himalayas
on the Tibetan plateau, and peripheral zones at
elevations of 2,700 to 4,500 m. This species is a
partial migrant and most shrikes winter at low
elevations between 300 and 2,500 m within their

396

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. 2. June 2010

breeding range while a few migrate to lowland
Southeast Asia (Lefranc and Worfolk 1997).
However, little information pertaining to its
reproduction has been published (Tsering et al.
2002). We studied the breeding ecology of the
Grey-backed Shrike in a shrub-covered alpine
valley in south Tibet. Our objectives are to report
on the adaptation of life history traits of shrikes to
high-elevation environments, and provide base-
line data for species conservation.

METHODS

Study Area. — Field data were gathered in the
Xiongse Valley (29° 27' N, 91° 40' E) near Lhasa,
Tibet during 1996 and 1999-2004. Shrub vegeta-
tion characterizes the study area, including roses
{Rosa sericea) and barberries (Berber is hem-
leyana) at 3,980^,550 m, and Wilson Junipers
(Sabina pingii) at 4,550-4,980 m on south-facing
slopes; spiraeas (Spiraea japonica) at 4,200-
4,400 m, rhododendrons (Rhododendron nivale),
and alpine willows (Salix sclerophylla) at 4,400-
5,100 m on north-facing slopes.

Field Procedures. — Our study site was a 400-ha
plot that covered the entire elevational range of
the shrub vegetation. All Grey-backed Shrike
nests located were described, and nest contents
measured. Subsequent visits were made to
examine nesting success. Several nests were
checked infrequently because of their remoteness.
We pooled the data of all years for analysis
because of the small number of nests found during
the study period. We chose P < 0.05 as the
minimum acceptable level of statistical signifi-
cance. Means are expressed ± SD.

RESULTS

Gray-backed Shrikes are the only Lanins
species in Lhasa area. They arrived in the valley
at 4,400 m in early May (2 to 12 May), singly
(40% of 15 sightings) or in pairs (60%). Most
departed by early October, but a few individuals
were occasionally seen above 4,200 m until mid-
December.

Nesting elevations ranged between 4,010 and
4,540 m with 69% (25 of 36 nests) below 4,300 m.
Shrikes tended to nest near streams (<5() m; 61%
of 36 nests). A 1 82-ha plot on south-facing slopes
was systemically searched and seven shrike nests
were found, producing a breeding density of 3.8
pairs/ 100 ha.

Shrikes nested in several plant species, of
which most (61%) were in the two most common

species, roses and barberries. Shrike nests con-
structed in bushes were nearer the ground ( 1 .5 ±
0.4 m, range = 0.7-2.!, n = 30) than those in
trees (2.8 ± 0.2 m, range = 2. 5-3.0, n = 6; t =
7.26, P < 0.001). Individual bushes selected for
nesting had dense foliage. Mean vegetation cover
around shrike nests was 34 ± 18% (range = 10-
85, n = 30), similar to the overall coverage (36%)
in the study area. However, a few shrike nests
were built in solitary tall bushes. Nests were cup-
shaped and materials in the exterior layer of the
nest cup consisted of thin twigs and roots of
bushes. The interior layer consisted of feathers,
sheep hair, and, at times, moss. Dimensions of
seven nests were; exterior diameter 14.6 ± 0.9 cm
(range = 13-16), interior diameter 9.0 ± 0.8 cm
(range = 8-10), inside depth 6.8 ± 1.4 cm (range
= 5-8.6), and outside height 15.3 ± 2.1 cm
(range = 12-19).

Females did not lay eggs immediately after they
finished building the nest. The duration between
completion of nest construction and laying of the
first egg was 6 to 10 days (n = 5 nests). Egg-
laying began in late May and ended by early July
with a peak in early June (Fig. 1 ). Shrikes nesting
at higher elevations initiated clutches later than
those nesting at lower elevations (r = 0.41, =

32, P = 0.02).

One egg was laid per day (n = 13 daily
monitored eggs from 6 nests). Eggs were whitish
with light-brown spots, especially on the large
end. Fresh mass of 11 eggs was 5.0 ± 0.16 g
(range = 4.7-5. 2). Measurements of 46 eggs from
12 nests averaged 25.7 ± 1.17 mm (range =
23.8-28.2) in length and 18.6 ± 0.48 mm (range
= 17.8-20.0) in breadth. Clutch size varied from
three to five eggs (4.12 ± 0.67, n = 25) with four
being most common (56%). We detected a strong
decline in clutch size over the season (/• = —0.43,
n = 25, P = 0.03). However, it was not
significant once we controlled for elevation
(partial coiTelation coefficient, r = -0.28, n =
22, P - 0.18). Clutch size decreased significantly
as elevations increased (r = —0.46, P — 0.02),
but did not if we considered the egg-laying time
(partial correlation coefficient, r — -0.34, P —
0.1 1 ). The decrease of clutch size was affected by
both breeding date and nesting elevation.

Incubation, considered as the period from the
laying of the last egg to hatching of the first egg,
started before clutch completion (n = 6 nests, 3
began with the first egg, 2 with the second egg, and 1
with the third egg). Only females incubated;

SHORT COMMUNICATIONS

397

FIG. 1. Temporal distribution (all data combined) of date of laying of the first egg by Grey-backed Shrikes m the
Xiongse Valley, south Tibet. Clutches initiated are arranged in 5-day periods.

incubation lasted 15 days at 4,050 m (n = 1) and 17-
18 days at >4,350 m (/? = 3). Males defended the
nest during incubation and, at times, fed incubating
females. Nestlings were provisioned by both female
and male parents. Fledging occurred at 14-15 days
(14.3 ± 0.6, n = 3) of age and fledglings weighed
34.3 ± 0.6 g (range = 32.2-36.9, n = 6), which was
7 1 % of the adult female weight (48 . 1 g, average of 4
birds measured in the field).

Fifty of 58 eggs (86.2%) from 14 clutches
hatched successfully, resulting in a brood size of
3.71 ± 1.00 (/? = 14). Thirty of 33 hatchlings
(90.0%) from nine clutches survived to fledging.
Brood size at fledging was 3.36 ± 0.81 {n =11).
Overall breeding success, measured as the per-
centage of clutches that produced at least one
fledged young, was 45.8% (11 of 24 known-fate
attempts). The probability of a nest being
successful calculated using Mayfield’s (1975)
method was 0.455 (i.e., 45.5%). Two (15.4%) of
13 failed attempts were due to nest abandonment
because of inclement weather (nestlings died in
the nest after continuous rainfall), and 1 1 (84.6%)
because of predation (damaged nest structure)
likely by mountain weasels (Mustela ahaica), the
most common carnivore in alpine areas.

DISCUSSION

Grey-backed Shrikes, compared to their Lhasa
counterpart and other lowland congeners (Ta-
ble 1), had a short breeding season (~ 50 days).
They produced fewer and smaller clutches but
larger eggs than their lowland relatives. Alpine

shrikes had average clutch size similar to
conspecifics in lowland Lhasa, but their maximum
clutch size was one egg less. Shrike clutches
within the alpine valley became smaller with
increasing elevations. Invertebrates in the Tibetan
plateau are scarce in terms of species diversity and
population abundance decreases with increased
elevations (Wang et al. 1992). Poor food avail-
ability could prevent shrikes nesting at high
elevation from producing larger clutches. Selec-
tion should cause high-elevation shrikes to
allocate more energy for each individual offspring
(fewer but larger eggs) to improve their survival
in harsh unpredictable alpine environments, which
may compensate for reduced fecundity resulting
from both fewer and smaller clutches. The shift in
reproductive investment from offspring number
toward offspring quality follows the general trend
in avian life history shifts along elevational
gradients (Badyaev 1997, Badyaev and Ghalam-
bor 2001, Lu 2005, Bears et al. 2009). Badyaev
and Ghalambor (2001) suggested that high-
elevation birds provide greater parental care to
their offspring in terms of longer incubation and
nestling periods than their low-elevation counter-
parts. Grey-backed Shrikes nesting in alpine
habitats versus lowland Lhasa had longer incuba-
tion and nestling periods (Table 1). Relative to
other species of shrikes at the same latitude, Grey-
backed Shrikes appeared to not be typically longer
in these two parameters of parental care. This
differentiation might be related to interspecific
difference in body size or other traits.

398

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2. June 2010

TABLE 1. Life history parameters of

some Lanins shrikes nesting near

30° N.

L. tephronotns

L. tephronolus

L. excubitor

L. ludovicianus

L. schach

Locality

Lhasa, China

Lhasa, China

Sede Boqer,
Israel

Florida, USA

Yangzhou, China

Latitude, N

29° 27'

29° 32'

30° 52'

27° 05'

32° 23'

Elevation, m

4,300

3,650

475

11

10

Breeding season

50

65

120

160

100

length, days

Broods per year

1

1

1^

1-3

1-2

Clutch size (range)

4.1 (3-5)

4.0 (3-6)

5.8 (5-7)

4.3 (2-6)

5.9 (4-8)

Egg volume, mm^

8,916

10,153

8,689

8,484

Egg volume/female

196

161

183

149

body mass

Incubation period.

16.7 (15-18)

15.3

16.8 (14-18)

17

14.0 (12-15)

days (range)

Nestling period.

14.3 (14-15)

12.9

17.2 (16-19)

15

12.5

days (range)

Source

This study

Tsering et al.
2002

Yosef 1992

Yosef 2001

Yan and Ma 1991,

Hu et al. 2007

Grey-backed Shrike nests were intermediate in
probability to be successful (45.5%), compared
with those of other species of shrikes (Red-backed
Shrike [L. coUurio]: 35.3%, Horvath et al. 2000;
Woodchat Shrike [L. senator]: 63.8%, Talagi
2001; Loggerhead Shrike [L. ludovicianus]:
60.0%, summarized by Yosef 2001). High-
elevation shrikes, despite presumably higher
juvenile survival, had lower annual productivity
as a result of fewer and smaller clutches, and
relatively low nesting success. This suggests that
this species is more at risk if their habitats are
altered by changes in land management activities
or climate.

ACKNOWLEDGMENTS

We thank the Buddhists in the Xiongse Nunnery for
logistic assistance and Guohong Gong. Liying Zhang,
Xianhai Zeng, and Xiaoyan Ma for field assistance. We also
thank Reuven Yosef and an anonymous reviewer for their
comments on the manuscript. This work was conducted at
the Field Research Station for Tibetan Wildlife, which is
jointly administered by Wuhan University and Tibet
University.

LITERATURE CITED

Badyaev, a. V. 1997. Avian life history variation along
altitudinal gradients: an example with cardueline
finches. Oecologia 1 1 1:365-374.

Badyaev, A. V. andC. K. Ghalambor. 2001. Evolutional
gradients: trade-off between parental care and fecun-
dity. Ecology 82:2948-2960.

Bear.s. H., K. Martin, and G. C. White. 2009. Breeding
in high-elevation habitat results in shift to slower life-

history strategy within a single species. Journal of
Animal Ecology 78:365-375.

Horvath R., R. Farkas, and R. Yosef. 2000. Nesting
ecology of the Red-backed Shrike (Lanins coUnrio) in
northeastern Hungary. Ring 22:127-132.

Hu, J., T. P. Guan, C. Q. Zhou, and J. C. Hu. 2007.
Growth of Rufous-backed Shrike (Lanins scliach).
Sichuan Journal of Zoology 26:152-154.

Lefranc, N. and T. Worfolk. 1997. Shrikes: a guide to
the shrikes of the world. Pica Press, Sussex, United
Kingdom.

Lu, X. 2005. Reproductive ecology of Blackbirds (Tnrdns
niernla ma.xiinns) in a high-altitude location, Tibet.
Journal of Ornithology 146:72-78.

Mayfield, H. F. 1975. Suggestions for calculating nest
success. Wilson Bulletin 87:456^66.

Talagi, M. 2001. Some effects of inclement weather
conditions on the survival and condition of Bull-
headed Shrike nestlings. Ecological Research 16:55-
63.

TSERING, R.. L. E. JOHANNESSEN, and T. S0LHOY. 2002.
Breeding biology of the Grey-backed Shrike (Lanins
tephronotns). Proceedings of the International Orni-
thological Congress 23:289.

Wang B. H.. W. H. Yuan, C. M. Wang, F. S. Huang, S.
H. Tang, and D. W. Lin. 1992. The Xizang insect
fauna and its evolution. Henan Science and Technol-
ogy Publishing House, Zhengzhou, China.-

Yan, a. H. and j. S. Ma. 1991. Studies on ecology of the
Brown-backed Shrikes (Lanins schacli). Chinese
Journal of Zoology 26:30-32.

Yo.SEF, R. 1992. From nest building to fledging of young in
Great Grey Shrikes (Lanins twcnhilor) at Sede Boqer,
Israel. Journal of Ornithology 133:279-285.

Yosef, R. 2001. Nesting ecology of resident Loggerhead
Shrikes in southcentral Florida. Wilson Bulletin 113:
279-284.

SHORT COMMUNICATIONS

399

The Wilson Journal of Ornithology 122(2):399-402, 2010

The First Reported Hybridization of Abert’s and Canyon Towhees {Pipilo spp.)

R. Roy Johnson'"^ and Steven L. Hopp^

ABSTRACT. — Two mixed pairs of towhees were
found in irrigated desert yards near Tucson, Arizona. The
first known towhee Fi hybrids, from a female PipUo
aherti (Abert’s Towbee) and male P. fiiscns (Canyon
Towhee) were studied from winter of 1998-1999 through
summer 2000. This mixed pair raised at least eight young
in three broods during the two breeding seasons. Young
were so similar to P. fuscus that, if seen outside this
context, they would probably not be identified as hybrids.
A second mixed pair, also near Tucson, suggests that
hybridization between P. aherti and P. fuscus may be
more common than originally thought. Lack of previously
reported hybridization between P. aherti and P. fuscus
may be either due to internal or external isolating
mechanisms, limited survival and longevity, or human
failure to recognize hybrids. Received 2 September 2009.
Accepted 10 January 2010.

Hybridization among birds is relatively common,
occurring in an estimated 10% of non-marine species
(Grant and Grant 1992). The tendency to cross can be
strong for some species pairs. For example, hybrid-
ization is common in three of the four species of ‘ ‘red-
eyed” towhees (Pipilo spp.) of the United States and
Mexico (Sibley 1954; Sibley and West 1958, 1959;
Sibley and Sibley 1964; Greenlaw 1996). Hybridiza-
tion has not previously been reported in the complex
of four species of towhees collectively known as
“brown towhees.” The California Towhee (P. criss-
alis) of the Pacific West does not occur sympatrically
with the other three species of “brown towhees.”
However, P. fuscus (Canyon Towhee) occurs
throughout a portion of its range with both P.
aherti in western Arizona and extreme northwestern
Mexico, and with P. alhicollis (White-throated
Towhee) of central Mexico (Davis 1951, Johnson
and Haight 1996, AOU 1998). P. aherti and P .
fuscus occur together throughout much of the range
of P. aherti in southern and central Arizona, but P.
aherti occurs in more mesic riparian bottoms (Tweit
and Finch 1994) while P. fuscus generally inhabits

'Johnson & Haight Environmental Consultants, 3755
South Hunters Run, Tucson, AZ 85730, USA.

^Environmental Studies, Emory and Henry College,
Emory, VA 24327, USA.

•’'Corresponding author; e-mail: rroylois@aol.com

adjacent riparianlands or drier uplands (Johnson
and Haight 1996).

OBSERVATIONS

SLH discovered a mixed pair of “brown
towhees” nesting on Tanque Verde Creek during
February and March 1999, —24 km east of
Tucson. The immediate area represents an eco-
tone between the riparian mesquite (Prosopis
velutina) bosque (woodland) along Tanque Verde
Creek, typical P. aherti habitat, and upland
Sonoran desertscrub, habitat often occupied by
P. fuscus. This mixed pair of towhees raised at
least three broods of F| hybrids during the next
two breeding seasons (Appendix).

The Abert’s Towhee was observed on 16 April
1999 carrying nesting material into a 3.4 m tall
ornamental juniper (Juniperus spp.). The Abert’s
Towhee was followed closely by the Canyon
Towhee, suggesting the Abeit’s was the female
because the male of a mated pair of Canyon Towhees
has been reported to fly with the female as she carries
nesting material (Johnson and Haight 1996; J. T.
Marshall, pers. comm.). This was subsequently
verified by extracting DNA from the feathers and
ascertaining the young possessed the mtDNA of
Abert’s Towhee (R. M. Zink, pers. comm.) The pair
frequently engaged in sc/ueal-duets when greeting
each other, a trait of both species (Marshall 1964).

The nest was 1.9 m above ground and measured
7.6 cm tall with 12 cm outside, and 7.6 cm inside
diameters. The nest contained three eggs on 27 April
and on 8 May the nest contained three nestlings,
e.stimated to be 4-5 days of age. The nestlings
weighed 18.2, 25.5, and 26.5 g on 1 1 May; the
smallest did not have its eyes open. We banded all 3
nestlings.

We observed 22 provisioning trips to the nest on
12 May, 12 by the Abert’s and 10 by the Canyon
Towhee. The Canyon Towhee flew south into
desertscrub during forays from the nest, while the
Abert’s Towhee flew north into the mesquite bosque
for foraging. This difference in habitat preference by
each adult was noted on numerous occasions.

The young fledged on 18 or 19 May, and the
adults immediately moved them into a wooded

400

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2. June 2010

region too thick for close observations. They led the
fledglings back into open areas within ~1 week
making them easier to observe. We observed parents
feeding the young between 27 and 31 May.
Particular young were generally favored by individ-
ual parents, suggesting brood splitting (Marshall and
Johnson 1968, Johnson and Haight 1996). However,
on two occasions we observed a single fledgling
being fed by both parents. One fledgling was
observed bathing in a birdbath on 30 May.

The female spent more time in the mesquite
bosque during this period, and the male was more
attentive to the young, leading us to suspect the
pair was engaged in building a second nest. We
observed the pair copulating on 29 May with the
P. fiisciis in the superior position.

The P. fusciis adult was observed alone in
mid-June with one of the banded fledglings.
The central breast spot of the young bird was
evident at this stage but the throat patch was
less conspicuous than that of the adult. The adult
and young engaged in squeal -duet vocalization
when meeting, a behavioral trait previously unre-
ported between adult and young “brown towhees.”
None of the banded young was observed following
this initial post-fledgling period.

We discovered both adult towhees feeding
unbanded fledglings in late August 1999 and
recorded their appearance (Appendix). We later
saw only the adult mixed pair as they frequently
visited the bird-feeding area together throughout
the winter months. The parents were observed
feeding three fledglings on 15 May 2000. A
Cooper’s Hawk (Accipter cooperii) flew overhead
as all five towhees fed on cracked corn in the open
yard on 17 May. Both parents uttered a series of
short, sharp, shrill peer or pink calls. These calls
sounded identical in both parents and caused the
fledglings to “freeze.” The young were observed
sporadically (and described. Appendix) over the
next few weeks and the adults remained on
teiritory, but were not seen after November 2000.

DISCUSSION

Intrageneric Hybridization. — Several factors are
important regarding this Tucson “brown towhee”
hybridization. The.se include: (1) hybridization has
not been reported earlier between P. aherti and P.
fuscus in the Tucson region, although studies ot
“brown towhees” have been conducted in this
vicinity for more than 100 years (Bendire 1890;
Marshall 1960, 1964; Marshall and John.son 1968;
Johnson and Haight 1996); (2) the pair bond

between congeneric species that are not sisters
(Zink 1988, Zink and Dittmann 1991) was
sufficiently strong to last 2 years and allow raising
of at least three broods; (3) the female P. aberti
selected a male P. fuscus at a locality where other
P. aberti are not uncommon; and (4) the problem of
detecting hybrids is exacerbated by the fact the Fj
hybrids appeared so similar to P. fuscus. However,
no interbreeding was found among several hundred
color-banded birds in a multi-year study in an
environment inhabited by both species, near
Tucson, Arizona (Marshall 1960, pers. comm.).

The residential yard near Tanque Verde Creek
was in an ecotone which provided feeding sites.
Both towhee species prefer feeding in open areas
immediately adjacent to brush into which they
may escape if disturbed (Marshall 1960, Dawson
1968, Marshall and Johnson 1968). Thus, the open
yard between the brushy hillside and mesquite
bosque provided an ideal situation for hybridiza-
tion of P. fuscus and P. aberti.

A second mixed pair of “brown towhees”
occupied a territory during 2007-2009 in the
foothills of the Tucson Mountains —12 km west
of downtown Tucson (S. M. Russell, pers. comm.;
RRJ, pers. obs.). The immediate surrounding
landscape is a mixture of native upland Sonoran
desertscrub with artificial plantings maintained by
drip irrigation, forming habitat more mesic than
native Sonoran desertscrub. This pair, similar to
the Tanque Verde mixed pair, was first noted as
they commonly appeared at and left a feeding
station together. The pair would disappear into the
adjacent desertscrub on leaving the feeding
station. Male/female identity of this pair of P.
fuscus and P. aberti remain unknown along with
their breeding history. The pair built a nest in
2008 that failed to fledge young (S. M. Russell,
pers. comm.). However, Russell noted an imma-
ture “brown towhee” during the 2009 breeding
season that he judged to have at least some P.
aberti characteristics, e.g., black markings on the
face. Russell noted only the single mixed towhee
pair in the yard during this period with no pairs of
either P. aberti or P. fuscus. During October and
November Rus.sell noted two P. aberti (presum^
ably a pair) in this yard but no P. fuscus.

Habitats are newly created through landscape
disturbance that provide new niches. Processes are
telescoped in time in human-disturbed habitats,
compared to natural processes such as ecological
succession. Both sets of mixed P. aberti and P.
fuscus pairs in our study were in irrigated residential

SHORT COMMUNICATIONS

401

yards. Disturbed habitat has had a profound
influence on avian hybridization on the Great
Plains in numerous passerine species. The entire
Great Plains may now be considered a large hybrid
zone between several related eastern and western
species that nest in woody vegetation (Gill 1998).

Shifting Habitat Selection by Abert’s Towhee. —
A shift in habitat selection and local distribution
by P. aberti has occuired recently in Tucson when
arid desertscrub ecosystems were replaced by
relatively mesic irrigated yards and parks. P .
aberti was not listed as a resident of housing
developments in Tucson’s suburban uplands in
studies as recently as the 1980s (Tweit and Tweit
1986, Mills et al. 1989). P. aberti has now been
recorded in suburban yards as far as 7 km from its
original riparian habitat along the Santa Cruz
River, increasingly encroaching on areas that had
been occupied by P. fuscns (Tucson Audubon
Society 2001-2009; RRJ, pers. obs.). These
changes have increased the probability of P. aberti
and P.fiiscHS occurring together and the possibility
of hybridization. Whether P. aberti will displace P .
fiiscus in much of suburban Tucson or the two
species will coexist is unknown.

A similar movement occurred by P. aberti in
the Phoenix region during the 1900s. The species
moved from the now dry Salt and Gila rivers into
suburban yards and agricultural areas (Johnson et
al. 1987, Rosenberg et al. 1987). However, in
Phoenix, P. aberti does not come into contact with
P. fuscus, which occurs only in more remote arid
upland areas (Rea 1983, Johnson et al 1987).

It is probable that disturbed native habitat in the
Tucson area, as well as ecotonal factors, facilitat-
ed the Tanque Verde towhee hybridization and
Tucson Mountain mixed pairing. It is possible that
hybridization is a recent phenomenon caused by
newly created habitats stemming from human
landscape alteration. It is also possible the
difficulty in distinguishing hybrid offspring has
led to an underestimation of the frequency of
hybridization. Thus, a search should be undertak-
en in avian collections for hybrid towhees.

ACKNOWLEDGMENTS

This paper benefited from discussions with J. T.
Marshall, leading expert on “brown towhees,” and from
discussions with L.T, Haight. Mitochondrial DNA analysis
verifying the P. aberti was the female of the pair was
conducted by R. M. Zink. S. M. and R. O. Russell assisted
in banding young hybrids, and S. M. Russell provided
information about the second mixed pair in the Tucson

Mountains. We thank C. E. Braun, J. B. Dunning, and an

unknown reviewer for criticjues of an early manuscript.

LITERATURE CITED

American Ornithologists’ Union (AOU). 1998. Check-
list of North American birds. Seventh Edition. American
Ornithologists’ Union, Washington, D.C., USA.

Bendire, C. E. 1 890. Notes on Pipilo fuscus mesoleucus and
Pipilo aberti, their habits, nests and eggs. Auk 7:22-29.

Davis, J. 1951. Distribution and variation of the brown
towhees. University of California Publications in
Zoology 52: 1-120.

Dawson, W. R. 1968. Pipilo aberti. Abert’s Towhee. Pages
632-638 in Life histories of North American cardinals,
grosbeaks, buntings, towhees, finches, sparrows, and
allies (O. L. Austin jR., Editor). U.S. National Museum
Bulletin Number 237, Part 2.

Gill, F. B. 1998. Hybridization in birds. Auk 1 15:281-283.

Grant, P. R. and B. R. Grant. 1992. Hybridization of bird
species. Science 256:193-197.

Greenlaw, J. S. 1996. Spotted Towhee {Pipilo maculatus).
The birds of North America. Number 263.

Johnson, R. R. and L. T. Haight. 1996. Canyon Towhee
{Pipilo fuscus). The birds of North America. Number 264.

Johnson, R. R., L. T. Haight, and J. M. Simpson. 1987.
Endangered habitats versus endangered species: a
management challenge. Western Birds 18:89-96.

Marshall Jr., J. T. 1960. Interrelations of Abert and
Brown towhees. Condor 62:49-64.

Marshall Jr., J. T. 1964. Voice in communication and
relationships among brown towhees. Condor 66:345-
356.

Marshall Jr., J. T. and R. R. Johnson. 1968. Pipilo
fuscus mesoleucus. Canyon Brown Towhee. Pages
622-630 in Life histories of North American cardinals,
grosbeaks, buntings, towhees, finches, sparrows, and
allies (O. L. Austin Jr., Editor). U.S. National Museum
Bulletin Number 237, Part 2.

Mills, G. S., J. B. Dunning Jr., and J. M. Bates. 1989.
Effects of urbanization on breeding bird community
structure in southwestern desert habitats. Condor
91:416^28.

Rea, a. M. 1983. Once a river: bird life and habitat changes on
the middle Gila. University of Arizona Press. Tucson.

Rosenberg, K. V., S. B. Terrill, and G. H. Rosenberg.
1987. Value of suburban habitats to desert riparian
birds. Wilson Bulletin 99: 642-654.

Sibley, C. G. 1954. Hybridization in the red-eyed towhees
of Mexico. Evolution 8:252-290.

Sibley, C. G. and F. C. Sibley. 1964. Hybridization in the
red-eyed towhees of Mexico: the populations of the
southeastern plateau region. Auk 81:479-504.

Sibley, C. G. and D. A. West. 1958. Hybridization in the
red-eyed towhees of Mexico: the eastern plateau
populations. Condor 60:85-104.

Sibley, C. G. and D. A. West. 1959. Hybridization in the
Rufous-sided Towhees of the Great Plains. Auk
76:326-338.

Tucson Audubon Society. 2001-2009. Tucson (Arizona)

402

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 2. June 2010

bird count, www.tucsonbirds.org (accessed 19 August
2009).

Tweit. R. C. .and D. M. Finch. 1994. Abeil’s Towhee (Pipilo
aberti). The birds of Noith America. Number 111.
Tweit. R. C. and J. C. Tweit. 1986. Urban development
effects on the abundance of some common resident
birds of the Tucson area of Arizona. American Birds
40:430-436.

Zink, R. M. 1988. Evolution of brown towhees: allozymes,
morphometries and species limits. Condor 90:72-82.
Zink. R. M. and D. L. Dittmann. 1991. Evolution of
brown towhees: mitochondrial DNA evidence. Condor
93:98-105.

APPENDIX. — Characteristics of eight young
Fi hybrids in three broods from Pipilo aberti X P.
fusciis cross. Colors and characteristics were
ascertained through field and photographic com-
parisons with the parents, by examination of
specimens in the University of Arizona bird
collection, and from Birds of North America
accounts (Tweit and Finch 1994, Johnson and
Haight 1996). All characteristics are most like P.
fuscus unless otherwise noted.

General body color of all eight was intermediate
between that of the two parents but most other
characteristics were similar to adult P. fuscus. All
showed throat patches varying from light to darker
and striping around the patch varied from much to
none. No young had the dark facial mask or
uniform pinkish-brown of breast, flanks, and belly
of female P. aberti. The only obvious P. aberti

characteristic for any of the young was the dark
rusty crissum (under-tail coverts) of five F,
hybrids, in contrast to the lighter, buffy crissum
of P. fiscus. Photographs available by using the
second author’s complete name in a search engine
or at http://zeeman.ehc.edu/envs/Hopp/index.html.
Brood # 1 (May 1999; 3 Fi Hybrids
Two with hint of rufous crown; third with
uniform gray-brown crown. All with central
breast spot. Two with more extensive, bright,
rusty crissum, most like P. aberti.

Brood # 2 (August 1999; 2 F) Hybrids)

Both with slight but detectable rufous crown.
Throat patch for both notably darker and less
pronounced than in adult P. fuscus. One with
slight breast spot. Tails of both longer than adult
P. fuscus. Both with more extensive, bright rusty
crissum, most like P. aberti.

Brood # 3 (May 2000; 3 F, Hybrids)

One with evident and two with slight rufous
crown. One with central breast spot smaller and
less conspicuous than adult P. fuscus, second with
heavy streaking approaching a spot, and third
without a spot. Tails of both similar in length to
adult P. fuscus, slightly shorter than P. aberti.
Bright rusty crissum in one, less evident in
second, and third with no color difference
between belly and crissum, appearing similar to
front view of female Brown-headed Cowbird
(Molothriis a ter).

The Wilson Journal of Ornithology 1 22(2):402^05, 2010

Yellow-billed Cuckoo Hatched and Fed by a Red-winged Blackbird

Ken Yasukawa'

ABSTRACT. — I observed a Red-winged Blackbird
(Agelaius phoeniceus) nest in southcentral Wisconsin,
USA that received a Yellow-billed Cuckoo (Coccyziis
ainericanus) egg. The cuckoo egg was laid after the
blackbird clutch was complete and the female had
begun incubation. The parasitic egg was considerably
different in size and color, but was accepted by the
female Red-winged Blackbird. The female, but not the
male, afso cared for the nestling, which was strikingly
different in plumage, gape markings, and begging
behavior, and which grew despite its 3^ day hatching

' Beloit College. Department of Biology. Beloit, Wl
5351 I. USA: e-mail: yasukawa@bcloit.edu

delay relative to its nest mates. Rapid development and
exaggerated begging behavior of the Yellow-billed
Cuckoo nestling may enable a parasitic cuckoo nestling
to compete successfully with host nestlings for food.
Received 24 August 2009. Accepted II December 2009.

Yellow-billed (Coccyzus ainericanus) and Black-
billed (C. erythropthahnus) cuckoos are the only
altricial birds that lay in the nests of other birds as
apparent facultative, as opposed to obligate, inter-
specific brood parasites (Hughes 1999). Wyllie
(1981) noted intraspecific parasitism also occurs.

SHORT COMMUNICATIONS

403

FIG. 1. Nestling Yellow-billed Cuckoo (right) begging in a brood of Red-winged Blackbirds.

and Fleischer et al. (1985) used egg protein analysis
to verify that two female Yellow-billed Cuckoos
contributed to a large clutch. The majority of records
of interspecific parasitism are between Yellow-
billed and Black-billed cuckoos (Nolan and Thomp-
son 1975), but 1 1 other species are known to have
been parasitized by Yellow-billed Cuckoos. Many
of the unusual aspects of North American cuckoo
reproductive ecology, including the tendency to
vary clutch size in relation to food supply and rapid
nestling development, may have facilitated evolu-
tion of brood parasitism in cuckoos (Nolan and
Thompson 1975). I report an unusual case of brood
parasitism of a Red-winged Blackbird {Agelaiiis
phoeniceiis) by a Yellow-billed Cuckoo and discuss
an aspect of cuckoo nestling behavior that may have
facilitated the evolution of parasitism.

OBSERVATIONS

On 21 May 2009 I found a Red-winged
Blackbird nest containing four incubated eggs at
Newark Road Prairie in southcentral Rock
County, Wisconsin (42° 32' N, 89° 08' W). I
checked this nest daily at ~ 0530 hrs CST. On 24
May I found an egg of a Yellow-billed Cuckoo in
addition to the four blackbird eggs. The cuckoo
egg was considerably larger than the blackbird
eggs (Hughes [1999] gives a mean egg mass of

9.1 g for eastern populations of Yellow-billed
Cuckoos, whereas Yasukawa and Searcy [1995]
give mean masses of 4.02-4.07 g for Red-winged
Blackbirds). The egg was pale bluish-green and
unmarked, unlike the blackbird eggs, which were
pale blue with dark brown blotches. All five eggs
were warm to the touch, indicating the female was
incubating them. On 28 May I observed one
newly hatched blackbird nestling and by the next
morning, there were three blackbird nestlings
along with one blackbird egg and the cuckoo egg.
When I checked the nest on the morning of 1
June, the Yellow-billed Cuckoo egg had hatched.
The cuckoo nestling was thus 4 days younger than
the first-hatched blackbird nestling and 3 days
younger than the other two blackbird nestlings.
The fourth blackbird egg did not hatch and
showed no signs of development when 1 examined
its contents on the morning of 3 June.

The Yellow-billed Cuckoo nestling was strik-
ingly distinct from the Red-winged Blackbird
nestlings in appearance (Fig. I ) and behavior.
Unlike its blackbird nest mates, when begging the
cuckoo nestling rapidly vibrated its head, pro-
duced a sustained hissing begging call while
showing its distinctive gape with white spots.
Happed its wings rapidly, and attempted to gain a
position above the blackbird nestlings by standing

404

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2, June 2010

on them. In contrast, the blackbird nestlings
moved their heads only slightly as they stretched
their necks and gaped. The blackbird gape lacked
the white spots of the cuckoo, and the blackbird
nestlings did not flap their wings or jostle for
position. The blackbird begging call was a series
of discrete chirps rather than a sustained hiss. I
made a 1-hr video recording of the nestlings on 4
June to document the differences in nestling
begging behavior between the two species.
Unfortunately, a predator, probably a common
raccoon (Procyon lotor), took the nestlings that
night, and I could not continue my observations of
this unusual case of brood parasitism to learn
whether the fledgling would have received care.

DISCUSSION

Unusual reproductive behavior in Yellow-billed
Cuckoos is associated with abnormal food abun-
dance (Nolan and Thompson 1975) but, to the best
of my knowledge, 2009 was not an outbreak year
for eastern tent caterpillars (Malacosoma ameri-
cana) or periodic cicadas {Magicicada spp.), and
the numbers of annual cicadas (Tihicen spp.) did
not seem unusual. The most recent outbreak of
periodic cicadas in Rock County, Wisconsin was
2007 (Hamilton 2007).

The egg and nestling of the Yellow-billed
Cuckoo were quite different from those of the
host Red-winged Blackbird, but the egg was
accepted (incubated) and the female blackbird
fed the nestling. I did not observe the male
blackbird feeding nestlings. Nickell (1954) re-
ported a similar case in Oakland County,
Michigan. Furthermore, the 3^ day difference
in hatching did not appear to disadvantage the
cuckoo nestling, whereas a blackbird nestling
experiencing the same hatching asynchrony
would normally die of starvation (Holcomb and
Twiest 1971, Strehl 1978).

Red-winged Blackbirds accept Brown-headed
Cowbird {Molothrii.s ater) eggs and care for
cowbird nestlings and fledglings, and Brown-
headed Cowbird parasitism is common on my
study area (Clotfelter and Yasukawa 1999).
Yellow-billed Cuckoos are observed on my study
area in most breeding seasons as well, although
they are not common. Given the number of
studies of Red-winged Blackbird reproductive
biology, extensive overlap in Red-winged Black-
bird and Yellow-billed Cuckoo breeding ranges,
and dearth of similar records in the literature.

Yellow-billed Cuckoo parasitism of Red-winged
Blackbirds is clearly rare.

Nolan and Thompson (1975) suggested that
North American cuckoos have traits that facilitate
the evolution of brood parasitism (see also
Hughes 1999). These traits include nomadic
movements in response to food abundance, large
egg size, variable clutch size in response to food
availability, intraspecific brood parasitism, occa-
sional use of abandoned nests of other species,
short breeding cycle (17 days from egg laying to
fledging is among the shortest of any species of
bird), rapid nestling development (nestlings can
raise their heads above the nest rim and are fully
feathered within 2 hrs after hatching), extreme
hatching asynchrony, and an unusual diet of food
items that can be difficult to find. Yellow-billed
Cuckoos occasionally lay eggs in nests of other
species (called “incipient parasitism” by Nolan
and Thompson 1975), and a necessary behavior is
in place to allow evolution of brood parasitism if
parasitism was to become more advantageous
than parental care (Hamilton and Orians 1967).
North American cuckoos have not evolved
obligate brood parasitism, however, perhaps
because species recognition, song learning, and
ecological constraints act to maintain the advan-
tage of parental behavior.

Acceptance of a Yellow-billed Cuckoo egg and
nestling by a Red-winged Blackbird may be
another indication of the plastic nature of parental
care in North American songbirds (Shy 1982).
However, the exaggerated begging behavior of the
nestling and its ability to overcome hatching 3^-
days later may indicate they facilitated the
Yellow-billed Cuckoo’s ability to secure food
from the female Red-winged Blackbird despite
the obvious differences in plumages, begging
calls, and begging behavior of the two species.
The ability to compete for food in novel and/or
more exaggerated manners than host nestlings
would also facilitate the evolution of obligate
brood parasitism.

ACKNOWLEDGMENTS

1 dedicate this .short communication to my mentor, Val
Nolan Jr. I thank C. S. Anderson for field assistance and
Beloit College for permission to work at Newark Road
Prairie and financial support. S. G. Sealy and .1. M. Hughes
provided background information on North American
cuckoo parasitism, and E. D. Ketter.son, E. D. Clotfelter,
and two anonymous reviewers made suggestions to improve
earlier versions of the manuscript.

SHORT COMMUNICATIONS

405

LITERATURE CITED

Cl.OTFELTER, E. D. AND K. Yasukawa. 1999. Impact ot
brood parasitism by Brown-headed Cowbirds on Red-
winged Blackbird reproductive success. Condor
101:105-114.

Fleischer, R. C., M. T. Murphy, and L. E. Hunt. 1985.
Clutch size increase and intraspecific brood parasitism in
the Yellow-billed Cuckoo. Wilson Bulletin 97: 125-127.

Hamilton, K. 2007. Wisconsin Pest Bulletin 52 (10), 15
June 2007. Trade and Consumer Protection. Wisconsin
Department of Agriculture, Madison, USA. pestbulle-
tin.wi.gov/pests.jsp?categoryid=32&issueid = 84 (ac-
cessed 1 December 2009).

Hamilton 111, W. J. and G. H. Orians. 1967. Evolution of
brood parasitism in altricial birds. Condor 67:361-382.

Holcomb, L. C. and G. Twiest. 1971. Growth and
calculation of age for Red-winged Blackbird nestlings.

Bird-Banding 42:1-17.

Hughes, J. M. 1999. Yellow-billed Cuckoo (Coccyzus
onicf'icofius). The birds ol North Ameiica. Number
418.

Nickell, W. P. 1954. Red-wings hatch and rai.se a Yellow-
billed Cuckoo. Wilson Bulletin 66:137-138.

Nolan Jr., V. and C. F. Thompson. 1975. The occurrence
and significance of anomalous reproductive activities
in two North American non-parasitic cuckoos Coccy-
zus spp. Ibis 1 17:496-503.

Shy, M. M. 1982. Interspecific feeding among birds: a
review. Journal of Field Ornithology 53:370—393.

Strehl, C. 1978. Asynchrony of hatching in Red-winged
Blackbirds and survival of late and early hatched birds.
Wilson Bulletin 90:653-655.

Wyllie, I. 1981. The cuckoo. Universe Books, New York,
USA.

Yasukawa, K. and W. a. Searcy. 1995. Red-winged
Blackbird {Agelaius phoeniceits). The birds of North
America. Number 184.

The H'ilsoii Journal of Ornithology 122(2):406-415. 2010

Ornithological Literature

Robert B. Payne, Review Editor

FIELD GUIDE TO THE SONGBIRDS OE
SOUTH AMERICA; THE PASSERINES. By
Robert S. Ridgely and Guy Tudor. University of
Texas Press, Austin, Texas, USA. 2009: 750
pages and 122 color plates with facing-page maps.
ISBN: 978-0-292-71979-8. $49.95 (paper).—
There is much to celebrate with publication of
this revamp and update of the first two volumes of
The Birds of South America {BOSA: Oscine
passerines, 1989; Suboscine passerines 1994) in
a more compact format under a single cover.
Undoubtedly the most anxiously awaited aspect of
Songbirds is the tremendous contribution of Guy
Tudor’s new art; illustrations of 406 species
(some as only busts) have been added to the

— 1,220 previously depicted ones, a whopping
33% increase. There remain not illustrated about
347 native species of South American passerines,
roughly 18% of the authors’ total number of

— 1,974 (excluding a few vagrants). Now repre-
sented are all but one (I think) standing genus
(Stymphalornis antbirds, endemic to southeast
Brazil).

The back cover calls attention to the need for
“a more compact guide that birders can take into
the field.” Given that they intend to show us how
to identify 1,981 species of passerine birds, some
of which are hard to see well, all while traipsing
through the rainforests and de.serts on the Bird
Continent, they may have set themselves up for
disappointment. On some levels, they have
succeeded, it’s just hard to say whether they have
tackled the monster or the monster has tackled
them. Weighing in at 1.28 kg, the book can be
carried in the field by most birders, but not very
comfortably by anyone. Perhaps most tellingly,
almost no one has opted to tote it on recent South
American birding walks where most folks would
like to be carrying a field guide. I’ve not yet seen
anyone with the plates cut out and bound
.separately, all too often the fate of BOSA, but
the specter has already come up. The ideal bulk
may lie about halfway between Songbirds and the
much smaller Birds of South America: Non-
passerines (Erize el al. 2006, Princeton University
Press).

Tudor’s illustrations are wonderful; it’s hard to

imagine ever seeing paintings of many of these
birds that are more “spot-on.” In large measure
because he is an avid and highly observant birder
who understands the importance of accurate
shapes and postures, Tudor is able to encapsulate
what we need for field identification in the strict
confines of afield guide. As is inevitably the case
when a large number of species is treated, a few
illustrations leave much to be desired (most of
which are holdovers from BOSA). White-bearded
Antshrike {Biatas nigropectus, Plate 20, #3) never
appears uncrested (except in a museum tray) — the
new, undersized male figure and a female bust, at
appropriate scale, could have admirably replaced
the poor BOSA ones; Plain-crested Elaenia
{Elaenia cristata, Plate 44, #7), unrecognizable
as is, shows a shaip, spiky crest virtually all the
time; White-winged Cotinga (Xipholena atropur-
purea, Plate 69, #7) is a much darker “blackish-
purple” as described in the text; Black-collared
Swallow (Atticora melanoleuca, Plate 76, #10)
invariably looks black-and-white (not navy blue)
in the field even at close range, and the posture is
atypical; and all of the pipits (Anthus spp., Plate
84, perhaps with the exception of Correndera
Pipit, A. correndera) remain unidentifiable in and
of themselves.

Other closely similar groups of birds, such as
the dreaded “tyrannulets,” are well done al-
though there are regrettable problems with a few
species that occur in syntopy (together in the same
habitat) through much of their ranges, particularly
Rough-legged Tyrannulet (Phyllomyias burmeis-
teri, Plate 41, #16); Greenish Tyrannulet (P.
vire.scens, Plate 41, #17); and Mottle-cheeked
Tyrannulet [Phylloscartes ventralis, Plate 46, #4).
The most similar-looking small tyrannids, scat-
tered over .several plates (mostly 41, 42, 46),
could have been grouped in a much more useful
manner for field identification, for starters by
separating exclusively Andean from non-Andean
species on different plates, then bunching the
members of each broad group that actually
overlap geographically and occur in the same
habitats. Because the small South American
tyrannids, too many of which are not illustrated,
present the most complex field identification

406

ORNITHOLOGICAL LITERATURE

407

problems (female Sporophila seedeaters aside), I
had hoped they would be given high priority in
Songhii'cls.

The choice was made to insert many ot the new
illustrations at varying scales different from the
rest of the (existing) figures on the plate. Some of
the most problematic instances are Plate 52, #8,
11; Plate 53, #3, 4, 11, 13 relative to the
themselves and the rest of that plate; Plate 61,
#5, 7; Plate 7 1, #4; Plate 101, #2, 3; and Plate 1 14,
#12 (Dickcissel [Spiza americaiia] should be
significantly larger than Greenish Yellowfinch
[Sicalis olivascens]'.). It is disheartening to see the
stunning Araripe Manakin {Autilophia hoker-
manui, Plate 64, #17) relegated to a corner-
squeeze at about half-size, below its sister,
Helmeted Manakin (A. galeata). Its illustration
points to another unfortunate feature of all plates
having new illustrations: something happened in
the production stage so all of the new figures look
faded relative to those in BOSA. Thus, the red and
black on the Araripe Manakin are significantly
paler or less saturated than these same colors on
the Helmeted Manakin. A number of figures of
females also have been newly illustrated, an
important advancement, but a field guide really
must treat all distinctive female plumages (of the
species illustrated at all). Almost no juvenal
plumages are shown or described. One of the
few, Andean Laniisoma {Laniisoma huckleyi,
Plate 66, #8), is mislabeled the adult female. I
love the new illustrations of the mockingbirds
(Plate 83) in flight; it would have been great to see
much more of this kind of field-oriented illustra-
tion.

On the plates, each species illustration is
numbered and some also have letters A, B, C,
indicating the figure represents a distinctive
subspecies. To interpret this information, one
needs to go to the text. This cumbersome system
could have been made much less so by simply
putting the letters on the species’ map to inform
users of where that taxon (or group of taxa) is
expected.

The South American distribution of each
species is shown on a correspondingly numbered
map on pages opposite or near its illustration,
sometimes requiring a little hunting although I got
used to it. I really like these maps. Several levels
of geographic zoom are used to show most of the
highly restricted ranges more clearly. The contrast
between colors chosen for ranges, including
migrants, and also rivers and political boundaries

for most countries (to the state/department level in
the tightest zooms), also works well. Without
having looked at most of the maps critically, I
think it’s fair to say the level of accuracy is
admirably high. 1 happened to notice that the
maps for Black Antbird {Cercomacra serva) and
Blackish Antbird (C. iiigrescens) were switched
(their names are bad enough!). Similarly, numbers
labeling the maps and figures for Green Honey-
creeper (Chlorophanes spiza) and Red-legged
Honeycreeper (Cyaiierpcs caeruleus) are mixed
up (numbers are correct on the maps and text, so
switch the species’ numbers on the plate of your
copy to match). Problems on the level of not
showing Saffron-crested Tyrant-Manakin {Neo-
pehna chrysocephahim) in northern Peru; of
showing Long-winged Antwren {Mynnothenila
longipennis) extensively between the Rio Napo in
Peru and the Negro in Brazil where it is largely or
completely absent; of missing the interface
between Squamate Antbird (Myrmeciza squa-
mosa, shown extensively north into Rio de
Janeiro) and White-bibbed Antbird (M. loricata,
shown far too extensively through interior Bahia,
and ranging well into northern coastal Sao Paulo);
of White-breasted Antbird (Rhegmatorhina hojf-
mannsi) crossing (and extensively so) the Rio
Juruena into northem Mato Grosso; of Seira do
Mar Tyrannulet (Phylloscartes dijficilis) occumng
all along the coast of southeast Brazil where
known from only a few scattered points high in
the mountains; and of showing the threatened
Black-and-white Monjita {Xolmis dominicauus)
over much too wide an area in Rio Grande do Sul
(it was never known to be so amply distributed,
even historically) — are somewhat more frequent.
Maps for a few very poorly known birds are
annoyingly problematic. For example, an undoc-
umented, single-observer sight-record of the ultra-
rare Kinglet Calyptura {Calyptura cristata. Plate
66, #6) in coastal northern Sao Paulo is mapped
and cited in the text as if it were fact! This map
should show only the single point of known
occurrence (from late Oct 1996) just below
Teresopolis, Rio de Janeiro with an aiTOW
pointing to a red dot and at most a question mark
anywhere else; the maximum zoom should have
been provided here.

In a field guide, the species accounts should
foster transformation of the painting on the plate
to a living bird by infusing vital information on
habitat and behavior while enabling even the most
difficult field identifications. My overall impres-

408

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 2, June 2010

sion is that over 90% of the species accounts in
Songbirds are good to excellent; space limitations
in the Journal precluded my elaboration of
examples. Transcriptions of vocalizations of most
species are offered, which is usually better than
not having anything but not nearly as helpful as
actually hearing the sounds. I highly recommend
obtaining recordings (primary vocalizations of
most species have been published or are available
on the internet) ahead of travel to South America.

The taxonomy used in Songbirds closely
parallels that of the standing South American
Checklist Committee (SACC) of the American
Ornithologists’ Union. The SACC was obviously
heavily referenced although it was not cited
anywhere in a 16-page section toward the back
of the book titled “Notes on Taxonomy and
English Names.’’ However, the authors “express-
ly note’’ (page 2) that they “adhere to the
decisions on English names and taxonomy’’ that
were recently published on behalf of the Interna-
tional Ornithological Congress (IOC; Gill et al.
2006-2009, Birds of the World, Recommended
English Names, Princeton University Press). The
first bird in the book, Campo Miner (Geositfa
poeciloptera), provides a poignant example of the
taxonomic swirl that underlies Songbirds. It boils
down to; (1) the IOC classification, which
incorporated new information up to the end of
2004, maintaining the long-standing Geobates
contrary to a 2003 publication suggesting it be
merged into Geositta-, (2) the SACC invoking
primarily genetic data published in 2005 to merge
Geobates into Geositta; and finally (3) Songbirds
incontrovertibly following the SACC, not the
IOC.

Ornithologists and birders frequently express
their opinions that some taxa or populations of
birds could or should be considered distinctive
enough to merit species status. The unilateral
splitting or lumping of species by authors of such
works as Songbirds and other guide books in
which no data or analyses of the case(s) are
presented for peer-review or independent evalua-
tion (now or a hundred years from now) reside in
a bubble drifting around the perimeter of science,
a “gray” subspecies of taxonomy. In Songbirds,
Ridgely and Tudor are “gray’’ on —25 taxa
traditionally considered subspecies. All are inter-
esting cases worthy of focu.sed study, and the
SACC and other taxonomic committees will soon
consider not only all of them but also all of the
equally or more interesting cases that, for one

reason or another (we have no data to judge), were
not mentioned in Songbirds. For one example, the
Stigmatiira wagtail-tyrants go from two to four
species absent any analysis at all but the well-
supported (and accepted by the SACC) elevation
to species status of Schistocichla lencostigma
satiirata (Roraiman Antbird, in the Spot-winged
Antbird complex) published in 2005 was not
recognized.

English names are not bound by scientific
principles, and their u.sage for South American
birds became particularly unstable after the mid-
1980s, when BOS A introduced changes to names
of long standing. In Songbirds, the 2006 IOC
classification is followed, even in cases where a
novel name of Ridgely’ s from BOSA (e.g.. Noble
Antthrush for Chamaeza nobilis) was rejected in
favor of the traditional name (Striated Antthrush).
But this has set up some conflicts. For example, a
nomenclatural note on page 691 reads, “We
rename Mitrephanes phaeocercus the Northern
(rather than the Common) Tufted Flycatcher,
prefeiTing to restrict the use of the name
“Common’’ to species that truly are so.’’ This
flycatcher was described from Mexico and most
of its range lies in Middle America. Ridgely and
Gwynne (1989, A Guide to the Birds of Panama,
Princeton University Press) introduced the mod-
ifier “Common,’’ and described its status in
Panama as “Common.’’ Pointless self-reversal
like this does not inspire confidence and promotes
instability of names. True to form, Ridgely has not
resisted unveiling a few English names that
almost seem designed to shake things up. My
vote for the standout is ... “Shrike-like Tanager’’
(in place of the entrenched White-banded Tana-
ger, Neothraiipis fasciata), not to be confused with
Lanio Shrike-Tanagers. In sum, there is a great
deal of arbitrariness displayed throughout the
taxonomy and English naming used in Songbirds.

The new Ridgely and Tudor Field Guide to the
Songbirds of South America should become a
standard in the libraries of all birders, ornitholo-
gists, and con.servationists interested in neotrop-
ical birds, and it is sure to make birds more
accessible to more people, which is at the
grassroots level of effective bird conservation. If
you ever dream of birding in South America, buy
and thoroughly enjoy this book. Take it to South
America with you and use it to ably identify the
vast majority of passerine birds you will see. —
BRET WHITNEY, Louisiana State Univer-
sity Museum of Natural Science, 119 Foster

ORNITHOLOGICAL LITERATURE

409

Hall, Baton Rouge, LA 70803, USA; e-mail;
ictinia@earthlink.nel

PALEOGENE BIRDS. By Gerald Mayr.
Springer-Verlag, Berlin Heidelberg, Germany.
2009: 262 pages and 64 figures. ISBN: 978-3-
540-89627-2 ($189.00 USD, cloth; ~$25 per
chapter in electronic form). — The Paleogene
Period, from —65 to 23 million years ago, witnessed
profound shifts in the earth system including the
origin of permanent ice sheets in Antarctica. Global
cooling from peak “greenhouse Earth” conditions
is linked to a retreat of Eocene paratropical forests
previously extending to the poles and the develop-
ment of the “icehouse” conditions of the present.
Across the same time period, shifts in the ecology,
diversity, and distribution of nearly every part of the
biota are inferred. New insights into every part of
these Paleogene shifts have been rapidly accruing in
recent years. The record of sea-surface temperatures
and global climates has been significantly refined,
and vertebrate paleontologists continue to further
inform our knowledge of changing faunas.

The rapid pace of new discoveries of fossil bird
species in recent years belies the general idea of
their impossibly fragmentary record. The early
history of extant crown clade avian diversity,
species of which are parts of the major bird
lineages alive today, has been partially tracked in
several short review articles eager to summarize
the rush of new data recovered in recent years.
Because relevant paleo-ornithological discoveries
are moving forward at an arguably unprecedented
pace, any summaries must be proximate, abbre-
viate, and soon out of date. However, the last
book-length review focusing on an overlapping
time period and set of taxa is the 1999 Second
Edition of A. Feduccia’s “The Origin and
Evolution of Birds’’ (Yale University Press). With
a few exceptions, including updating several
chapters presenting the author’s views on the
dinosaurian affinities of birds, most ol its content
is from the 1996 First Edition. It is into this
context that Gerald Mayr, Curator of Ornithology
at the Senckenberg Institute, offers a significant
contribution, a detailed new guide to fossil avian
diversity from the first part of the Cenozoic. In
—250 dense small-format pages, “Paleogene
Birds" neatly presents insights into —40 million
years of the early evolution of extant bird
lineages. The book is largely organized taxonom-
ically but also includes a brief overview of

principal fossil localities, a review of debates
concerning the relationships among major clades,
and conclusions regarding the historical patterns
of avian diversity in the Northern and Southern
hemispheres.

Mayr, who recognizes in his preface that any
review in the present intellectual context will
quickly need updating as new discoveries are
made, quite thoroughly summarizes the earlier
literature but focuses on key recent discoveries, a
great many of which he has made. The commu-
nity would profit from this book even if it were
merely a synthesis of Mayr’ s own work. A prolific
author, he has published over 150 papers since
1995, more than any other paleo-ornithologist
over the same time period.

Ornithologists of all kinds will be intrigued
by the bizarre turns that avian evolution took in
the first part of the Cenozoic. The book presents
exceptional fossils that record novel ecologies
and distributions for well known groups; e.g..
Old World hummingbird relatives, giant flight-
less divers, and sea birds with bony pseudo-
teeth. However, at -$189 USD for a new copy
of this slim volume, many non-specialists may
elect to visit their local libraries or hope for a
lower-price soft cover edition. The intended
primary readership of the book is the paleo-
ornithological community. However, it will also
serve as an important reference for other
ornithologists and paleontologists seeking ac-
cess to specific parts of the literature. The book
seems organized for, and best approached by,
skipping from section to section, rather than
reading from cover to cover. Illustration print
quality is excellent, although the specialist
might wish for an oversized volume with color
illustrations and larger photographs.

One is well served by the detailed presentation
of Mayr’s perspective as that perspective is often
thought-provoking. Much of the data presented is
fascinating, and comments made as small asides
could be expanded into journal articles of
considerable interest. Some of the more contro-
versial ideas forwarded by Mayr and summarized
in the book concern phylogenetic relationships
among major lineages of birds; e.g., regarding the
affinities of penguins with parts of former
“Pelecaniformes.” While not all readers will
agree with these hypotheses, all are supported by
clearly presented evidence that is well positioned
to inform new research questions.

We still have much to learn about patterns and

410

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 2. June 2010

potential drivers of avian evolution. Mayr clearly
stands with the rest of us in wondering what new
bizarre forms we may discover tomorrow. It is
striking how few avian taxa are generally shown in
the skies of Paleogene ecosystem reconstructions.
It is hard not to feel as if we have been missing
significant patterns in vertebrate evolution across
the Cenozoic greenhouse to icehouse transition.
This book offers an important opportunity to renew
our focus on the diversity and distributions of birds
as significant parts of Paleogene environments. —
JULIA A. CLARKE, Associate Professor,
University of Texas, Department of Geological
Sciences, 1 University Station Cl 100, Austin,
TX 78712, USA; e-mail: Julia_Clarke@jsg.
utexas.edu

BIRDS OF BORNEO. By Susan Myers.
Princeton University Press, Princeton, New Jer-
sey, USA. 2009: 272 pages, 1,592 color illustra-
tions, 630 maps. ISBN: 978-0-691-14350-7.
$29.95 (paper). — At last, ornithologists and bird-
watchers in Borneo have a complete and accurate
guide to birds that they can fit in their pockets. For
30 years, folks have worked to this end, and it is
extremely gratifying to have reached it. Starting in
the late 1970s, Bertram Smythies’ magnificent
handbook. The Birds of Borneo ( 1960, 1968), was
out of print and the first modern field guide to the
birds of southeast Asia (King et al. 1975) did not
cover Borneo. Thus, there was a great need for a
handy book on Bornean birds. The Sabah Society
helped alleviate the problem by publishing an
inexpensive version of Smythies (1981), but it
was too big for field use. Charles Francis (1984)
also helped by producing a pocket guide consist-
ing of Smythies’ plates, but the illustrations
without text were not adequate for identification
of many species. For the last 1 5 years, Bornean
bird enthusiasts have done pretty well with
MacKinnon and Phillipps (1993), the first field
guide to birds of the Indo-Malayan Archipelago. It
has excellent illustrations by Karen Phillipps, but
is now out of print. Also, because it includes all
the birds of Sumatra, Java, and Borneo, it is too
big and bulky for comfortable use in the field.
Becairse it was published before knowledge of
Bornean bird distribution and ecology had quite
gelled, it also contains lots of small mistakes.

Those problems are now history. The tremen-
dous recent growth in .sophisticated birdwatching
in Borneo, combined with comprehensive updates

in the scientific literature (Smythies 1999, Shel-
don et al. 2001), set the stage for a good field
guide. The guide’s author, Susan Myers, has led
Victor Emmanuel Nature Tours in eastern Asia
for more than 15 years. Thus, she has been an
integral part of the ecotourism phenomenon. She
combines great experience in finding and identi-
fying birds in Borneo with a grand perspective of
Asian birds as a whole. The result of these
congruent forces is a truly outstanding field guide.
It is small, has extensive color illustrations of all
the birds, a complete set of maps, and all the
information needed on plumage, distribution,
habitat, elevation, and behavior to sort out the
species. It really is excellent.

The quality of any field guide depends on
insightful, pithy descriptions, and high quality
illustrations. This book does well in both areas. A
good test of its description quality is how well the
author deals with difficult groups. Such a group in
Borneo is the Cyornis flycatchers. These blue and
brown birds of the forest understory look much
the same, have nondescript songs, and often
confuse newcomers (Smythies resorted to a key
to deal with them). Myers, with simple, skillful
descriptions of plumage, similar species, micro-
habitats, and behavior makes it easy to sort them
out. As for illustrations, each species is depicted
individually next to its description and distribu-
tion map, instead of in plates of similar taxa.
Sexually dimorphic plumages, age variants, and
occasionally alternative subspecies are provided.
Most of the figures have been reproduced
individually from the plates in Robson (2000),
but 295 (of 180 species) were prepared specifi-
cally for this book. This helps explain the large
number of contributing artists (16) and the
variation in illustration quality. Despite variation,
only a couple of paintings are sub-par (e.g.,
Malaysian Rail-babbler [Enpetes macrocerus] and
crows). All the illustrations are more than
adequate for identification, and most are really
good. 1 particularly liked the trogons, hornbills,
bulbuls, babblers, and muscicapine flycatchers.
My favorites are the wren-babblers.

By virtue of its simplicity and thoroughness, this
is an outstanding field guide. 1 really cannot wait to
take it into the fore.st and give it a field test. No
doubt it will hold up well. — FREDERICK H.
SHELDON, Louisiana State University Muse-
um of Natural Science and Department of
Biological Sciences, Baton Rouge, LA, 70803,
USA; e-mail: fsheld(®lsu.edu

ORNITHOLOGICAL LITERATURE

LITERATURE CITED

Francis, C. M. 1984. Pocket guide to the birds of
Borneo. Sabah Society and World Wildlife Fund
Malaysia, Kuala Lumpur, Malaysia.

King, B. F., E. C. Dickinson, and M. W. Woodcock.
1975. A field guide to the birds of south-east Asia.
Collins, London, England.

MacKinnon, J. and K. Phillipps. 1993. A field
guide to the birds of Borneo, Sumatra, Java, and
Bali. Oxford University Press, Oxford, United
Kindom.

Robson, C. 2000. A guide to the birds of southeast
Asia. Princeton University Press, Princeton, New
Jersey, USA.

Sheldon, P. FI., R. G. Moyle, and J. Kennard. 2001.
Ornithology of Sabah: history, gazetteer, annotated
checklist, and bibliography. Ornithological Mono-
graphs 52:1-285.

Smythies, B. E. 1960. The birds of Borneo. Oliver and
Boyd, London, England.

Smythies, B. E. 1968. The birds of Borneo. Second
Edition. Oliver and Boyd, London, England.
Smythies, B. E. 1981. The birds of Borneo. Third
Edition. Sabah Society and Malayan Nature Soci-
ety, Kuala Lumpur, Malaysia.

Smythies, B. E. 1999. The birds of Borneo. Eourth
Edibon. Natural History Publications, Kota Kina-
balu, Malaysia.

BIRDS AND BATS OF PALAU. By H.
Douglas Pratt and Mandy T. Etpison. Mutual
Publishing L.L.C., Honolulu, Hawaii, USA. 2008;
290 pages. ISBN: 978-1-56647-871-7. $29.95
(paper). — Palau lies at the western edge of
Micronesia. As a result of its relative proximity
to the rich avifaunas of Asia, the Philippines, and
New Guinea, Palau sustains a larger set of both
resident and migrant birds than any other
Micronesian island group. Most of Palau also is
blessed with a rugged, raised limestone topogra-
phy that has stifled deforestation and human
habitation. It’s a great place for birdwatching.

When I visited Palau in 1995 and 1997, I
carried two field guides. The first was the rather
small Field Guide to the Birds of Palau by John
Engbring (1988, Conservation Office, Koror,
Palau), which covered mainly resident species.
The second was the more comprehensive A Field
Guide to the Birds of Hawaii and the Tropical
Pacific by H. Douglas Pratt, P. L. Bruner, and D.
G. Berrett (1987, Princeton University Press,
Princeton, NJ), which covered all island groups
in Micronesia and Polynesia. H. Douglas Pratt and
Mandy T. Etpison now have produced a nice, up-

41 1

to-date field guide for the birds (and bats) of
Palau, whether resident or migrant. The text of
this book has an informal style of writing, which
may be good for some readers and not so good for
others. There are no subheadings in the species
accounts, and one must look through paragraphs
of text to glean information on field marks,
vocalizations, etc. The scattering of typos, un-
italicized scientific names, grammatical errors,
etc. is not enough to distract the average reader.
The only maps are aerial photographs or satellite
imagery; while attractive with their intense shades
of blue and green, the lack of a clearly labeled,
line-drawing map hinders one’s understanding of
Palauan geography.

The overall design of the book is lively and
attractive. The species accounts feature magnifi-
cent photographs, mainly by Etpison. Because
nearly all of the photographs are unlabeled,
however, you must make your own judgment
about the age and gender of the bird depicted (if
those attributes are determinable), not to mention
what it was doing and where the photograph was
taken. The single color plate of endemic birds,
painted by Pratt, is excellent although it is placed,
without warning and essentially unreferenced, in
the introductory material. The 19 species of
scolopacid sandpipers depicted on pages 199 to
207 are erroneously placed in the plover family
(Charadriidae).

Disagreement about common names of birds
abounds in the tropical Pacific, so it is little
surprise that some of these issues surface in Birds
and Bats of Palau. Eor example, following the
lead of F. Gill and M. Wright (2006, Birds of the
World: Recommended English Names. Princeton
University Press), Pratt and Etpison call Sterna
hergii (page 224) the “Swift Tern’’ rather than
the more conventional, non-apodid name Great
Crested Tern. On pages 28 and 228-231, I
congratulate the authors for not adopting Gill
and Wright’s name “Angel Tern’’ for Gygis alha
(= Gygis Candida), although they do use this
simply heavenly name in the u.seful checklist of
species on pages 278-282, which presents the
distribution of each species on each major island.

I’m not a big fan of proposing new common
names in field guides, even though this practice
has become widespread in these days of over-
splitting to increase perceived endemism. The
account of the kingfisher Todiramphus (often
placed in the genus Halcyon) cinnamominus
pelewensis helps explain my reservations. Usually

412

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 2. June 2010

known as the Micronesian Kingfisher with
distinctive subspecies on Guam (T. c. cinnamo-
miniis), Palau (7. c. pelewensis), and Pohnpei (7.
c. reichenbachii), Pratt and Etpison now call it the
Rusty-capped Kingfisher and regard it as endemic
to Palau at the species level, even though they list
it as 7. c. pelewensis in the species account,
checklist, and index. Here I quote the first and last
paragraphs of the species account (pages 88-90)
as they lay out their thinking about this bird.

“In most previous books this bird has been
called Micronesian Kingfisher because it was
regarded as one of 3 subspecies that make up that
species. However, HOP has accumulated evidence
(see below) that shows that those 3 supposed
subspecies could not possibly interbreed success-
fully and that there is really no such thing as the
‘Micronesian Kingfisher.’ Until the details are
properly published, however, we continue to use
the traditional taxonomy, but re-christen this
endemic Palau soon-to-be species with a distinc-
tive English name.’’

“Differences among the so-called ‘Micronesian
Kingfishers’ include voice, adult and juvenile
plumage coloration, and nesting habits. Those on
Pohnpei are larger, and look very much like
Rusty-caps as adults, but the Juveniles resemble
the strikingly different Guam kingfisher. In that
one, the white parts of the Rusty-cap’s plumage
are a rich rusty chestnut, and. unlike the other 2,
adult females are distinctive with a white belly,
but Juveniles are like adults. These visual
differences are what scientists call an ‘isolating
mechanism.’ This is any behavioral, ecological, or
physical characteristic that prevents two species
from interbreeding, during the mating process or
after pairs have formed. For example, Pohnpei
kingfishers nest in cavities in termite nests, while
the Rusty-capped uses tree cavities, since the
termite nests in Palau are used by Collared
Kingfishers. So even if, hypothetically, a Pohnpei
bird and Rusty-cap would mate, they would not be
able to decide where to nest! And we suspect that
the 3 Micronesian forms might not recognize each
other visually either.’’

Aside from the pitfalls of trying to mix popular
writing with supposedly .scholarly thinking, and
issues of biological versus phylogenetic species
concepts, they mention nothing about whether
there is molecular evidence for differentiation
among these kingfishers. Changes in nomencla-
ture are compromised until they are properly
documented in a peer-reviewed publication.

In the drawings that show the external parts of
birds on page 29, the bill is divided into the
“upper mandible’’ (= culmen or rostrum) and
“lower mandible’’ (= mandible). For decades,
ornithologists have puzzled other vertebrate
biologists by using the word “mandible” (=
lower Jaw) in describing the upper part of the bill.
Also on page 29, the tarsal Joint (ankle) is called
the knee, which of cour.se is actually one Joint
higher, where the femur Joins the tibiotarsus and
fibula. Similarly, they incoiTectly use “ankle” for
the Joint were the toes articulate with the
metatarsals.

In spite of my criticisms, nobody interested in
birds should visit Palau without this fine book. I wish
that it had been available during my two trips there,
and I hope that it becomes widely used by residents
and visitors to this fascinating island group. —
DAVID W. STEADMAN, Florida Museum of
Natural History, University of Florida, Gaines-
ville, FL 32611, USA; e-mail: dws@flmnh.ufl.
edu

THE ECOLOGY AND CONSERVATION OF
ASIAN HORNBILLS: FARMERS OF THE
FOREST. By Margaret F. Kinnaird and Timothy
G. O’Brien. University of Chicago Press, Chi-
cago, Illinois, USA, and London, UK. 2007: 315
pages, 8 pages of color photos, numerous figures
throughout. ISBN: 978-0-226-43712-5. $45.00
(cloth). — Hornbills are easily the most conspicu-
ous of Asian forest birds. Their huge size and
massive, sculptured bills project a striking image
that is advertised by the hornbill’s loud, demand-
ing voice and whooshing wing beats. Dedicated
fruit-eaters, hornbills have a dominant role in the
seed-dispersal of trees and other plants. Sadly,
both their large size and food habits predispose
them to extirpation by hunting and forest loss, and
hornbills are among the most threatened of Asian
birds.

This book offers an impressive compilation of
field research and literature review on the
evolution, ecology, and con.servation of these
wonderful birds. Hornbills, despite their boister-
ous nature, are for the most part hard to get to
know. Imagine the challenges of studying a bird
that spends its life out of reach high in the forest
canopy, that roams over vast areas of forest
tracking hundreds of species of fruits, and that
nests deep within the tree-hollow of a forest giant.
A few teams of researchers have met the.se

ORNITHOLOGICAL LITERATURE

413

challenges over the past three decades. Kinnaird
and O’Brien foremost among them and, as a
result, the mysteries of the complex life ot the
rainforest hornbills are revealed in this book.

The authors begin with two chapters that
introduce the hornbills, their phylogeny and
biogeography. I was surprised to learn the
radiation of 3 1 species and nine genera of Asian
hombills, primarily rainforest dwellers, is be-
lieved to have been derived from savannah and
woodland inhabiting African hornbills. Speciation
within the geographically complex southeastern
Asian land mass and myriad islands has created a
hot-spot of hornbill diversity centered on the
Malaysian Peninsula and the islands ot Sumatra,
Borneo, and Java, all once connected. Several
unique features of hornbills are also discussed.
Naturally, this includes a section on “Casque
form and function’’ that weighs the evidence for
the horny casque’s role in social signaling
(species recognition, sex, and age), structural
support for the long bill, and as a battering ram
in one amazing species, the Helmeted Hornbill
(Rhinoplax vigil), the males of which engage in
aerial jousting with head-on collisions!

The next two chapters characterize the forest
habitat and diet of these hornbills. Habitats range
from highly seasonal deciduous subtropical for-
ests to evergreen tropical forests with only the
most subtle seasonal variation. An important
climatic influence is the occurrence once every
few years of an El Nino Southern Oscillation
drought that causes the synchronized fruiting of
many species of trees which, at other times,
withhold bearing fruit. This boom-or-bust food
availability affects which years hornbills nest.
Although capable and eager to hunt small
vertebrates and insects with that deadly bill, forest
hornbills achieve at best a dismal success rate and
instead must rely on fruits to fill their dietary
requirements. Dietary studies document 497
species of fruits consumed, with oily fruits
favored. With some success, the authors tackle
the seemingly impossible task of characterizing
the temporal and spatial availability of fruit and
how that availability affects hornbill abundance,
community structure, and reproduction.

Once they’ve established an ecological context
for forest hornbills, the authors delve into a range
of topics on hornbill reproduction and social
systems, specifically monogamy, territoriality,
and cooperative breeding. It’s worth highlighting
one topic. A behavior for which hornbills are

famous is that the female seals herself into the
nesting cavity during incubation and nestling care.
The brick-hard seal is made from a mortar of
regurgitated fruit, feces, and other materials
delivered by her mate and pasted to the rim of
cavity entrance. Apart from fortifying the nest
against predators, the .seal could serve other
functions, such as creating a more suitable
microclimate within the nest. However, the
authors argue for a social explanation, that the
female is forcing the male to be the sole provider
for the family.

Much of the field research on hornbills has
focused on the evolutionary and ecological
function of hornbills in seed dispersal-their role
as farmers of the forest, as the book’s title
suggests. The authors exhaustively review this
research and then attempt to model populations of
hornbills, not just for the sake of the hornbills
(minimal viable populations) but for the forest
they serve (ecological functional populations).
This topic leads into the final chapters on the
threats to hornbill survival and the outlook for the
future. And threats there are aplenty. Asian
hornbills and the forests that support them are in
a disastrous situation through large-scale defores-
tation and hunting driven by a dense and growing
human population. The rate of forest loss in places
like Sumatra is mind-boggling and depressing to
any sensibility for biodiversity and wilderness.
The authors map out a full range of realistic
options for the conservation of hornbills and their
forests, including parks and protected area, grass-
roots activities, and species management.

1 was impressed by how much published
research there is on hornbills, given the difficulty
in studying canopy birds in tropical forests and the
modest effort worldwide devoted to this field. 1
was also impressed by the command that Kinnaird
and O’Brien have of their subject material.
Throughout the book, the authors take care to
bring together available data on each subject and
to explore it in depth. The text is laden with
numerical summaries and statistical tests that
greatly bolster their explanations. This quantita-
tive approach is enhanced by numerous tables and
figures. The data are nevertheless limited, and the
authors over-reach in their inteipretation in
places, such as dismissing for lack of evidence
the role of predation in the evolution of nest-
sealing (see review by M. Whitmer, 2008, Auk
125: 996-997). Kinnaird and O’Brien are certain-
ly entitled to their views by the wealth of

414

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 2. June 2010

irifoiTnation assembled here, but it may take
decades more research by other scientists before
some of these ideas are sorted out. There are eight
pages of color photographs of hornbills by Tim
Laman. The text is written in an easily read
narrative style, although it is information-packed
and therefore rather slow going. I didn’t notice any
errors. This book will be a classic on hornbill
biology, and on the ecology and evolution of a
group of fruit-eating and seed-dispersing birds. —
THANE K. PRATT, P. O. Box 420, Volcano, HI
96785, USA; e-mail: thane-linda@earthlink.net

THE SECOND ATLAS OF BREEDING
BIRDS IN NEW YORK STATE. Edited by
Kevin J. McGowan and Kimberley Corwin.
Cornell University Press, Ithaca, New York,
USA, 2008: 712 pages, 25 two-page color
paintings, 245 halftone drawings, 164 charts and
graphs, and 512 color maps. ISBN: 978-0-8014-
4716-7. $59.95 (cloth). — This important work is
the result of the first statewide re-survey of the
distribution of breeding birds. The first survey,
The Atlas of Breeding Birds of New York State,
edited by Robert F. Andrie and Janet R. CaiTolI
(1988), summarized the survey data collected
between 1980 and 1985. For the Second Atlas,
1,207 volunteers surveyed essentially the same
5,279 blocks from 2000 to 2005, exactly 20 years
later, compiling over half a million records
representing 248 species and three hybrids. The
result is an invaluable record, not only of the
current distribution of breeding bird populations,
but also the change in distribution over the past
two decades. Change is occurring at a surprising
rate; the Second Atlas documents changes, both
increa.ses and decrea.ses, experienced by more
than half of the state's species.

The book opens with six introductory chapters.
In the first, Kimberley Corwin introduces the
project and the methodology. As in the original
atlas, the spatial units were based on lO-km X 10-
km ( 100 knV) squares, each divided into four 25-
knr survey “blocks.” Thus, the entire state was
delineated into a total ol 5,333 blocks. The
operational definitions of possible breeding,
probable breeding, and confirmed breeding are
given, and there is a discussion of how the data
were processed, including limitations and possible
biases. The chapter concludes with 15 color maps
that illustrate in great detail characteristics of the
state, such as cities and major roads, river

systems, and selected lakes (e.g., useful as
landmarks); climate data such as precipitation
and temperature; and habitat categories based
on land use, forest types, and ecoregions. The
latter is particularly useful for interpreting the
distribution maps presented in the individual
species accounts.

Chapter 2, by Kevin McGowan and Benjamin
Zuckerberg, is a thorough comparison of the first
and second surveys. Survey effort is quantified
not only project wide, but also by region and by
ecozone in relation to species richness. The heart
of this chapter is the amount of change for each
species, based on the number of blocks in which
they were detected. This information is presented
in two tables, the first aixanged taxonomically for
ease of finding a particular species, and the
second in rank order of change. Additional tables
are limited to those species in which the changes
were statistically significant. An analysis of
change with respect to breeding habitat revealed
that grassland species by far had the greatest
decline in distribution.

In Chapter 3, Gregory Endinger and Timothy
Howard introduce the reader to the different
habitats of the state, categorized by ecoregions,
ecozones, bird conservation regions, and natural
and cultural ecological communities. This chapter
is followed by an examination of land-use
changes, presented by Charles Smith and Peter
Marks. Both chapters should be of interest to New
York residents and nonresidents alike, and they
help place the distribution changes into perspec-
tive. Chapter 5, by John Peterson, is a discussion
of professional and amateur ornithology in New
York and, in Chapter 6, Kenneth Rosenberg and
Michael Burger identify conservation priorities
regarding the state’s breeding birds.

The main part of the book is the collection of
species accounts, contributed by 41 authors. The
text in each account consists of a brief introduc-
tion to the natural history of the species, its status
from the first atlas, the changes that have occurred
during the past 20 years, and conservation
implications. There are two maps: one showing
the 2()()()-2()()5 distribution, and the other the 20-
year change. There also is a table that summarizes
the number of blocks for which breeding was
possible, probable, and confirmed in 1980-85 and
2000-05, and a graph that shows the USGS
Breeding Bird Survey trend in New York from
1965 through 2005.

The layout is excellent. Each species account

ORNITHOLOGICAL LITERATURE

415

covers two pages with the text and an attractive
halftone portrait of the species on the left, and the
maps, table, and graph on the right. Thus, the
reader is presented each entire account without
having to turn additional pages.

This book is a remarkable accomplishment. It
belongs on the shelf of anyone interested in the
changing distribution of birds in the northeastern
United States. However, because the volume is so

beautifully done, especially the 25 color paintings
of New York habitats, it will be equally at home
on the coffee table. Finally, at a time when college
textbooks command over $100 lor the paperback
editions, the $59.95 list price for The Second Atlas
represents a lot of value. — JOHN A. SMALL-
WOOD, Montclair State University, Mont-
clair, NJ 07043, USA; e-mail: smallwoodj@
montclair.edu

Presidents of The Wilson Ornithological Society

J. B. Richards

1888-1889

Lawrence H. Walkinshaw

1958-1960

Lynds Jones

1890-1893

Harold E. Mayfield

1960-1962

Willard N. Clute

1894

Phillips B. Street

1962-1964

Reuben M. Strong

1894-1901

Roger Tory Peterson

1964-1966

Lynds Jones

1902-1908

Aaron M. Bagg

1966-1968

Frank L. Burns

1909-1911

H. Lewis Batts Jr.

1968-1969

W. E. Saunders

1912-1913

William W. H. Gunn

1969-1971

T. C. Stephens

1914-1916

Pershing B. Hofslund

1971-1973

W. F. Henninger

1917

Kenneth C. Parkes

1973-1975

Myron H. Swenk

1918-1919

Andrew J. Berger

1975-1977

Reuben M. Strong

1920-1921

Douglas A. James

1977-1979

Thomas L. Hankinson

1922-1923

George A. Hall

1979-1981

Albert F. Ganier

1924-1926

Abbot S. Gaunt

1981-1983

Lynds Jones

1927-1929

Jerome A. Jackson

1983-1985

J. W. Stack

1930-1931

Clait E. Braun

1985-1987

J. M. Shaver

1932-1934

Mary H. Clench

1987-1989

Josselyn Van Tyne

1935-1937

Jon C. Barlow

1989-1991

Margaret Morse Nice

1938-1939

Richard C. Banks

1991-1993

Lawrence E. Hicks

1940-1941

Richard N. Conner

1993-1995

George Miksch Sutton

1942-1943

Keith L. Bildstein

1995-1997

S. Charles Kendeigh

1943-1945

Edward H. Burtt Jr.

1997-1999

George Miksch Sutton

1946-1947

John C. Kricher

1999-2001

Olin Sewall Pettingill

1948-1950

William E. Davis Jr.

2001-2003

Maurice Brooks

1950-1952

Charles R. Blem

2003-2005

Walter J. Breckenridge

1952-1954

Doris J. Watt

2005-2007

Burt L. Monroe Sr.

1954-1956

James D. Rising

2007-2009

John T. Emlen Jr.

1956-1958

E. Dale Kennedy

2009-

THE WILSON JOURNAL OL ORNITHOLOGY

Editor CLAIT E. BRAUN

Editorial Board RICHARD C. BANKS

5572 North Ventana Vista Road
Tucson, AZ 85750-7204
E-mail: TWILSONJO@comcast.net

JACK CLINTON EITNIEAR
SARA J. OYLER-McCANCE
LESLIE A. ROBB

Editorial NANCY J. K. BRAUN
Assistant

Review Editor ROBERT B. PAYNE

1306 Granger Avenue
Ann Arbor, MI 48104, USA
E-mail: rbpayne@umich.edu

GUIDELINES FOR AUTHORS

Please consult the detailed “Guidelines for Authors’’ found on the Wilson Ornithological Society web site
(http://www.wiLsonsociety.org). All manuscript submissions and revisions should be sent to Clait E. Braun,
Editor, The Wilson Journal of Ornithology, 5572 North Ventana Vista Road, Tucson, AZ 85750-7204, USA.
The Wilson Journal of Ornithology office and fax telephone number is (520) 529-0365. The e-mail address is
TWilsonJO@comcast.net

NOTICE OF CHANGE OF ADDRESS

Notify the Society immediately if your address changes. Send your complete new address to Ornithological
Societies of North America, 5400 Bosque Boulevard, Suite 680, Waco, TX 76710.

The permanent mailing address of the Wilson Ornithological Society is: %The Museum of Zoology, The
University of Michigan, Ann Arbor, MI 48109, USA. Persons having business with any of the officers may
address them at their various addresses given on the inside of the front cover, and all matters pertaining to the
journal should be sent directly to the Editor.

MEMBERSHIP INQUIRIES

Membership inquiries should be sent to Timothy J. O Connell, Department of Natural Resources Ecology and
Management, Oklahoma State University, 240 AG Hall, Stillwater, OK 74078; e-mail: oconnet@ okstate.edu

THE JOSSELYN VAN TYNE MEMORIAL LIBRARY

The Josselyn Van Tyne Memorial Library of the Wilson Ornithological Society, housed in the University of
Michigan Museum of Zoology, was established in concurrence with the University of Michigan in 1930. Until
1947 the Library was maintained entirely by gifts and bequests of books, reprints, and ornithological magazines
from members and friends of the Society. Two members have generously established a fund for the purchase
of new books; members and friends are invited to maintain the fund by regular contribution. The fund is
administered hy the Library Committee. Jerome A. Jackson, Florida Gulf Coast Univeristy, is Chairman of the
Committee. The Library currently receives over 200 periodicals as gifts and in exchange for The Wilson Journal
of Ornithology. For information on the Library and our holdings, see the Society’s web page at http;//
www.wilsonsociety.org. With the usual exception of rare books, any item in the Library may be borrowed by
members of the Society and will be sent prepaid (by the University of Michigan) to any address in the United
States, its possessions, or Canada. Return postage is paid by the borrower. Inquiries and requests by borrowers,
as well as gifts of books, pamphlets, reprints, and magazines, should be addressed to: Josselyn Van Tyne
Memorial Library, Museum of Zoology, The University ol Michigan, 1 109 Geddes Avenue, Ann Arboi, Ml
48109-1079, USA. Contributions to the New Book Fund should be sent to the Treasurer.

This issue of The Wilson Journal of Ornithology was published on 18 May 2010.

Continued from outside back cover

346 Assessing generalized egg mimicry: a quantitative comparison of eggs of Brown-headed Cowbirds
and grassland passerines
Dwight R. KUppenstirie and Spencer G. Sealy

354 Flight feather molt of Turkey Vultures

Robert M. Chandler, Peter Pyle, Maureen E. Flannery, Douglas J. Long, and Steve N. G. Howell
Short Communications

361 The first reported case of cooperative polyandry in the Red-footed Booby: trio relationships and
benefits

Lei Gao, Guixia Zhao, Shan Tang, and Hanzhao Guo

365 Further evidence of breeding by Shiny Cowbirds in North America
Matthew J. Reetz, Jacob M. Musser, and Andrew W. Kratter

369 Circumstantial evidence for infanticide of chicks of the communal Smooth-billed Ani {Crotophaga
ant)

J. S. Quinn, A. Samuelsen, M. Barclay, G. Schmaltz, and H. Kahn

374 Flight speeds of migrating Red-necked and Fiorned grebes
Laurence C. Binford and Joseph A. Youngman

378 Thermoregulatory behavior in migratory European Bee-eaters {Merops apiaster)

Reuven Yosef

381 Population status of Chuck-will’s-widow {Caprimulgus carolinensis) in the Bahamas

William K. Hayes, ElwoodD. Bracey, Melissa R. Price, Valerie Robinette, Eric Gren, and Caroline Stahala

385 Yellow Rails wintering in Oklahoma

Christopher J. Butler, Lisa H. Pham, Jill N. Stinedurf, Christopher L. Roy, Eric L. Judd, Nathanael J.
Burgess, and Gloria M. Caddell

388 Breeding of the Giant Laughingthrush {Garrulax maximus) at Lianhuashan, southern Gansu, Chint
Jie Wang, Chen-Xi Jia, Song-Hua Tang, Yun Fang, and Yue-Hua Sun

392 First description of the nest of Jocotoco Antpitta {Grallaria ridgelyi)

Harold E Greeney and Mery E. Juifia J

395 Nesting ecology of the Grey-backed Shrike {Lanius tephronotus) in South Tibet
Xin Lu, Chen Wang, and Tonglei Yu

399 The first reported hybridization of Abert’s and Canyon towhees {Pipilo spp.)

R. Roy Johnson and Steven L. Hopp

402 Yellow-billed Cuckoo hatched and fed by a Red-winged Blackbird
Ken Yasukawa

406 Ornithological Literature

Robert B. Payne, Book Review Editor

The Wilson Journal of Ornithology

(formerly The Wilson Bulletin)

Volume 122, Number 2 CONTENTS June 2010

Major Articles

207 Not the Nice sparrow (The 2007 Margaret Morse Nice Lecture) '

Douglas W. Mock and P. L. Schwagmeyer

217 The postbreeding migration of Eared Grebes
Joseph R. Jehl Jr. and Annette E. Henry

228 Postfledging and natal dispersal of Crested Ibis in the Qinling Mountains, China

Yu Xiao-Ping, Xi Yong-Mei, Lu Bao-Zhong, LiXia, Gong Ming-Hao, Shi Liang, and Dong Rong

236 Morphological and genetic variation between migratory and non-migratory Tropical Kingbirds
during spring migration in central South America

Alex E. Jahn, Douglas J. Levey, Izeni Pires Parias, Ana Maria Mamani, Julian Quillen Vidoz, and
Ben Ereeman

244 Red-cockaded Woodpecker male/female foraging differences in young forest stands
Kathleen E. Eranzreb

255_ Foraging habits and habitat use by Edible-nest and Glossy swiftlets in the Andaman Islands, India
Shirish S. Manchi and Ravi Sankaran

273 Golden- and Blue-winged warblers: distribution, nesting success, and genetic differences in two
habitats

John L. Confer, Kevin W. Barnes, and Erin C. Alvey

279 Genetic and morphological variation of the Sooty-capped Bush Tanager {Chlorospingus pileatus), a
highland endemic species from Costa Rica and western Panama
Tania Chavarria-Pizarro, Gustavo Gutierrez-Espeleta, Eric J. Euchs, and Gilbert Barrantes

288 Long-term changes in avian community structure in a successional, forested, and managed plot in a
reforesting landscape
Elizabeth W. Brooks and David N. Bonter

296 Scale-dependent response by breeding songbirds to residential development along Lake Superior
Michelle T. Eord and David J. Elaspohler

307 Songbird nest survival is invariant to early-successional restoration treatments in a large river
floodplain

Dirk E. Burhans, Brian G. Root, Terry L. Shajfer, and Daniel C. Dey

318 Carotenoid-based male plumage predicts parental investment in the American Redstart
Ryan R. Germain, Matthew W Reudink, Peter P Marra, and Laurene M. Ratclijfe

326 Variation in plumage coloration of Northern Cardinals in urbanizing landscapes
Todd M. Jones, Amanda D. Rodewald, and Daniel P Shustack

334 Annual precipitation affects reproduction of the Southern Grey Shrike {Lanius meridionalis)

Oded Keynan and Reuven Yosef

340 Home range sizes and habitat use of Nelson’s and Saltmarsh sparrows

W. Gregory Shriver, Thomas P. Hodgman, James P Gibbs, and Peter D. Vickery

Continued on inside back cover

IU\U

<b

Wilson Journal

of Urnithology

(^)IU

SMlH.h

Volume 122, Number 3, September 2010

MCZ

LIBRARY

SlP 1 5 2010

HAFVVARD

ur-jiVERSITY

Published by the
Wilson Ornithological Society

41312!

THE WILSON ORNITHOLOGICAL SOCIETY
FOUNDED 3 DECEMBER 1888

Named after ALEXANDER WILSON, the first American ornithologist.

President — E. Dale Kennedy, Biology Department, Albion College, Albion, MI 49224, USA; e-mail:
dkennedy@albion.edu

First Vice-President— Robert C. Beason, P. O. Box 737, Sandusky, OH 44871, USA; e-mail:
Robert.C.Beason@gmail.com

Second Vice-President — Robert L. Curry, Department of Biology, Villanova University, 800 Lancaster
Avenue, Villanova, PA 19085, USA; e-mail: robert.curry@villanova.edu

Editor — Clait E. Braun, 5572 North Ventana Vista Road, Tucson, AZ 85750, USA; e-mail: TWILSONJO@
comcast.net

Secretary — John A. Smallwood, Department of Biology and Molecular Biology, Montclair State University,
Montclair, NJ 07043, USA; e-mail: smallwoodj@montclair.edu

Treasurer — Melinda M. Clark, 52684 Highland Drive, South Bend, IN 46635, USA; e-mail: MClark@tcservices.biz

Elected Council Members — Jameson F. Chace, Sara R. Morris, and Margaret A. Voss (terms expire
2011); Mary Bomberger Brown, Mary Garvin, and Mark S. Woodrey (terms expire 2012); Mark
Deutschlander, Paul G. Rodewald, and Rebecca J. Saffan (terms expire 2013).

Living Past-Presidents — Pershing B. Hofslund, Douglas A. James, Jerome A. Jackson, Clait E. Braun,
Mary H. Clench, Richard C. Banks, Richard N. Conner, Keith L. Bildstein, Edward H. Burtt Jr., John
C. Kricher, William E. Davis Jr., Charles R. Blem, Doris J. Watt, and James D. Rising.

Membership dues per calendar year are: Regular, $40.00; Student, $20.00; Family, $50.00; Sustaining,
$100.00; Life memberships, $1,000.00 (payable in four installments).

The Wilson Journal of Ornithology is sent to all members not in arrears for dues.

THE WILSON JOURNAL OF ORNITHOLOGY
(formerly The Wilson Bulletin)

THE WILSON JOURNAL OF ORNITHOLOGY (ISSN 1559-4491) is published quarterly in March, June,
September, and December by the Wilson Ornithological Society, 810 East 10th Street, Lawrence, KS 66044-8897. The
subscription price, both in the United States and elsewhere, is $40.00 per year. Periodicals postage paid at Lawrence, KS.
POSTMASTER: Send address changes to OSNA, 5400 Bosque Boulevard, Suite 680, Waco, TX 76710.

All articles and communications for publication should be addressed to the Editor. Exchanges should be addressed
to The Josselyn Van Tyne Memorial Library, Museum of Zoology, Ann Arbor, MI 48109, USA.

Subscriptions, changes of address, and claims for undelivered copies should be sent to OSNA, 5400 Bosque
Boulevard, Suite 680, Waco, TX 76710, USA. Phone; (254) 399-9636; e-mail: business@osnabirds.org. Back issues or
single copies are available for $12.00 each. Most back issues of the Journal are available and may be ordered from OSNA.
Special prices will be quoted for quantity orders. All issues of the journal published before 2000 are accessible on a free
web site at the University of New Mexico library (http://elibrary.unm.edu/sora/). The site is fully searchable, and full-text
reproductions of all papers (including illustrations) are available as either PDF or DjVu files.

© Copyright 2010 by the Wilson Ornithological Society
Printed by Allen Press Inc., Lawrence, KS 66044, USA.

COVER: Wilson’s Phalaropes (Phalaropus tricolor). Illustration by Robin Corcoran.

@ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).

MC2

LIBRARY

SEP 1 5 2010

HARVARD

UNIVERSITY

FRONTISPIECE. The Baywing (Agelaioides badiiis), an endemic species of southern South America,
is the primary host of the most specialized of the parasitic cowbirds, the Screaming Cowbird (Molothrus
nifoaxilUiris). It is also a secondary host of the generalist Shiny Cowbird (M. honariensis), which results
in a rare, triangular host-parasite system. Photograph of an adult from the Province of Buenos Aires,
Argentina by Alec Earnshaw.

Wilson Journal

of Ornithology

Piihlisbed by tbe Wilson Ornitbological Society

VOL. 122, NO. 3 September 2010 PAGES 417-634

The Wilson Journal of Ornithology 122(3):417^31, 2010

REPRODUCTIVE SUCCESS AND NESTLING GROWTH OF THE
BAYWING PARASITIZED BY SCREAMING AND SHINY COWBIRDS

MARIA C. DE MARSICO,''2 BETTINA MAHLER,' AND JUAN C. REBOREDA'

ABSTRACT. — We studied the breeding biology of the Baywing (Agelaioides badius), a shared host of Screaming
(Molothnis rufoaxillaris) and Shiny (M. bonariensis) cowbirds. We monitored 193 nests from December 2002 to March
2007 in the Province of Buenos Aires, Argentina. Baywings used a wide variety of nesting sites, mainly old nests of
furnarids. Their breeding season lasted from late November to February and was closely matched by that of Screaming
Cowbirds. The breeding season for Shiny Cowbirds started in late September but overlapped that of Baywings. Frequency
and intensity of Screaming Cowbird parasitism were 93% and 5 eggs per parasitized nest, while for Shiny Cowbirds they
were 16% and 1.4 eggs. Host clutch size was 4.0 ± 0.1 eggs and did not vary with time of breeding. Weight at hatching and
age of maximum growth were similar for host and Screaming Cowbird nestlings. Shiny Cowbird nestlings had higher
weight at hatching and lower age of maximum growth than the other two species. Screaming and Shiny cowbird nestlings
had higher growth rates and asymptotic weights than host nestlings. Sex-specific growth curves of Screaming Cowbirds
indicated males had higher growth rate and asymptotic weight than females. Only 19% of the nests produced fledglings.
Host egg survival, hatching success, and nestling survival were 0.92, 0.88, and 0.94, respectively. Excluding nest failures,
hosts fledged 0.78 chicks per egg laid. Baywings were smaller than Screaming and Shiny cowbirds, and experienced a high
frequency and intensity of parasitism. However, the effect of parasitism on host hatching success and chick survival was
low and comparable to that observed in larger hosts. Received I September 2009. Accepted H March 2010.

Obligate brood parasites lay their eggs in nests
of individuals of other species (hosts), which
provide all parental care for eggs and young.
Brood parasitism in birds has evolved indepen-
dently at least seven times, one of them in the
genus Molothnis, within the New World icterine
blackbirds (Lanyon 1992, Davies 2000, Sorenson

' Departamento de Ecologi'a, Genetica y Evolucidn.
Facultad de Ciencias Exactas y Naturales, Universidad de
Buenos Aires, Pabelldn II Ciudad Universitaria,
C1428EHA Buenos Aires, Argentina.

^Corresponding author;
e-mail: de_mar.sico@ege.fcen.uba.ar

and Payne 2002). This genus encompasses five
parasitic species that show great variation in
number of hosts they use (Ortega 1998). The basal
species of the clade, the Screaming Cowbird (M.
rufoaxillaris), is an extremely specialist parasite
with only three known host species: Baywing
(Agelaioides hadius) (Hudson 1874, Friedmann
1929), Chopi Blackbird (Gnorimpsar chopi) (Sick
1985, Fraga 1996, Mahler et al. 2009), and the
Brown-and-yellow Marshbird (Pseudolei.stes vir-
e.scens) (Mermoz and Reboreda 1996, Mermoz
and Fernandez 2003). The Bay wing is largely the
primary host among these species with frequen-
cies of parasitism often exceeding 80% of the

417

418

THE WILSON JOURNAL OL ORNITHOLOGY • Voi 122, No. 3. September 2010

nests (Mason 1980, Fraga 1998). In contrast, the
Shiny Cowbird {M. bonariensis) is one of the
most derived species (Lanyon 1992) and an
extremely generalist parasite with more than 230
known hosts (Ortega 1998, Lowther and Post
1999). Shiny Cowbirds also parasitize Bay wing
nests but at a relatively low frequency ( 1 6-20% of
the nests; Mason 1980, Fraga 1998). Screaming
and Shiny cowbirds occasionally converge in
Baywing nests resulting in an unusual three-sided
host-parasite system (Fraga 1998).

There are several reports of Screaming and
Shiny cowbird parasitism in nests of Baywings
(Hudson 1874, Friedmann 1929, Mason 1980,
Fraga 1998), but there are relatively few quanti-
tative data on the breeding biology of this host and
its interactions with both parasites. Previous
studies indicate Baywings seldom build their
nests, but exploit a wide variety of covered
nesting sites, including old nests built by other
species and holes in trees (Hoy and Ottow 1964,
Fraga 1988). This nesting behavior has been
proposed as an explanation for the late breeding of
this species in Argentina (Friedmann 1929),
where nesting attempts of Baywings typically
occur from November to March (Fraga 1998).
Baywings lay clutches of three to five eggs, but
host eggs are often outnumbered by eggs of
parasites (Hoy and Ottow 1964, Mason 1980,
Fraga 1998, De Marsico and Reboreda 2008). The
intensity of Screaming Cowbird parasitism in
nests of Baywings is among the highest reported
for a cowbird host (Mason 1986, Fraga 1998,
Reboreda et al. 2003, Ellison et al. 2006, but see
Kattan 1997, Trine 2000).

Multiple parasitism is likely to be costly for
Baywings because Screaming and Shiny cowbirds
often puncture host eggs during nest visits, which
may reduce host clutch size even in unparasitized
nests (Hoy and Ottow 1964, Fraga 1998, Massoni
and Reboreda 2002, Astie and Reboreda 2006).
Screaming and Shiny cowbirds, in addition, have
a shorter incubation period than Baywings (Fraga
1998), which provides parasite nestlings a head
start when parasitism is properly .synchronized
with host laying (Fraga 1998, Dure Ruiz et al.
2008, Fiorini et al. 2009). This increases the risk
of host nestlings being outcompeted by larger
cowbird nestlings (Payne and Payne 1998, Hoover
2003, Dure Ruiz et al. 2008).

Fraga ( 1986) estimated the effect of Screaming
Cowbird parasitism on egg and nestling survival
of Baywings. His estimates, however, may have

overestimated the costs of parasitism because
other sources of breeding failure such as nest
predation were rare during his study (Fraga 1986).
In addition, data of host and parasite nestling
growth in Bay wing nests are .scarce. Fraga (1998)
estimated growth parameters for Baywing and
Screaming Cowbird nestlings, but his results did
not account for non-independence among host
nestlings from the same brood. Screaming and
Shiny cowbirds also have a marked sexual size
dimorphism (Mason 1987), but sex-specific
growth curves of Screaming and Shiny cowbirds
are lacking. Male and female parasite nestlings
may have different growth patterns, and may
differentially compete for food with host nestlings
and experience different costs of competition
(Weatherhead 1989, Tonra et al. 2008). Growth
patterns of nestlings of both parasite species in
Baywing nests have not been compared. Scream-
ing and Shiny cowbird nestlings in nests of the
Brown-and-yellow Marshbird, another shared
host, did not differ in growth constant, asymptotic
weight, and age of maximum growth (Mermoz
and Fernandez 2003); it remains unknown if the
same is true for the primary host of the Screaming
Cowbird.

Our objective was to use a large data set of 193
nests from five consecutive breeding seasons to
examine: ( 1 ) quantitative data on several aspects
of the nesting biology of the Bayw'ing, (2)
baseline estimates of host reproductive success
and nest productivity, and (3) novel data on
nestling growth in Baywing nests, providing the
first report of sex-specific growth curves of
Screaming and Shiny cowbirds.

METHODS

Study Area. — Our study was conducted from
December 2002 to March 2007 at the private
reserve El Destino near Magdalena (35° 08' S, 57°
23' W) in Buenos Aires Province, Argentina. The
study site is a flat area of 320 ha within the
Biosphere Reserve Parque Costero del Sur (MAB-
UNESCO). Vegetation comprises woodland
patches dominated by Celtis tala and Scutia
hii.xifolia surrounded by grasslands and marshes
(Cagnoni et al. 1996). Baywings are year-round
residents in this area.

Potential nesting sites for Baywings within our
study area included old nests of other species and
cavities in trees. We placed 50 wood nest boxes
between 2002 and 2003 to facilitate data collec-
tion. Boxes were 30 X 20 X 16 cm (height, width.

Dc Marsico et uL • NESTING ECOLOGY OE BAYWINCJS

419

and depth), and had a 58-mm diameter circular
entrance hole and a removable roof to allow nest
inspection. All boxes were attached to trees at the
edge or in the interior of woodland patches at
heights of 1.8-2. 5 m. Nest boxes were separated
by at least 50 m. Previous observations indicated
that Baywings did not use empty boxes, and we
partially filled them with artificial nests of wood
sticks and plant fibers. One hundred additional
nest boxes, similar yet smaller than those for
Baywings (25 X 17 X 13 cm in height, width, and
depth, entrance hole = 45 mm), were placed in
the area (also attached to trees at a height of 1.2-
1.8 m) starting in 2003 and were also available to
Baywings. Most of these boxes had inactive
House Wren (Troglodytes aedon) nests by the
time Baywings started breeding. All nest boxes
were cleaned prior to the beginning of each
breeding season, and the large ones were re-filled
with new artificial nests.

Data Collection. — We looked for Baywing
nests throughout each breeding season (mid
Nov-late Feb). Most nests were found along
edges of woodland patches or in isolated trees at
heights of 1.3-10 m. We limited our study to
those nests at heights < 5 m, which could be
inspected using a portable ladder. Nest boxes were
checked twice each week to detect any new
nesting attempt. We recorded whether each nest
was in a natural site or a nest box. We classified
natural sites by identity of the nest builder when it
was known, or the nest architecture.

Active nests (i.e., nests with a defending pair of
Baywings) were checked every 1-3 days until
they fledged chicks or failed. We assigned each
nest an initiation date, corresponding to laying of
the first Baywing egg. This date was ascertained
directly when the nest was found before or during
host laying, or indirectly through backdating from
hatching dates (considering an incubation period
of 13 days since the laying of the penultimate egg;
Fraga 1998), when the nest was found during
incubation. We examined the nest contents during
each visit taking all eggs and nestlings from the
nest. Individual eggs were assigned to the host or
each parasite species on the basis of background
color, spotting pattern, and shape (Friedmann
1929; Fraga 1983, 1986). Maximum width and
length of host and parasite eggs were measured to
the nearest 0.1 mm using a vernier caliper.
Hatchlings of each species were identified using
skin and bill coloration as diagnostic cues
(Friedmann 1929; Fraga 1979, 1986). We as-

signed the laying date of Screaming and Shiny
cowbird eggs either directly or through backdat-
ing from hatching dates (incubation period:
12 days for both species; Fraga 1998) whenever
possible. We assumed that laying in half of the
cases occurred the same day the egg was found,
and in the other half the eggs were laid the
previous day when nests were inspected every
other day and the parasite egg appeared between
two successive visits. All eggs and nestlings were
marked with waterproof ink for identification.
Nestlings were weighed daily or every other day
to the nearest 0.1 or 0.5 g using Pesola spring
scales of 10- and 60-g capacity, respectively. We
visited nests at approximately the same hour of
the day whenever possible to record weights at
regular daily intervals. All nestlings were banded
at 10-1 1 days of age with a unique combination of
color leg bands and a numbered aluminum band.

We took a small amount (15-30 pL) of blood
from a subsample of parasite nestlings at 8 to
1 1 days of age through brachial vein puncture with
a 31G needle. Blood was collected with an 80 pL
heparinized capillary tube, immediately mixed
with 0.5 mL of lysis buffer (lOOmM Tris pH 8,
10 mM NaCl, lOOmM EDTA, 2% SDS) and stored
at room temperature until analysis. Individuals
were identified as male or female genetically
following Ellegren (1996). DNA was extracted
from blood samples using a standard salting-out
protocol (Miller et al. 1988) and diagnostic sex-
linked alleles were amplified using the P8/P2
primer set (Griffiths et al. 1998). Amplifications
were performed in 10 pL reaction volumes using
50-100 ng of DNA template, 0.5 pM forward and
reverse primers, 0.25 pM dNTPs, 2.5 mM MgCE,
and 0.25 u Taq-Polymerase. Annealing tempera-
tures were set at 50° C and repeated for 30 cycles.
PGR products were separated in 3% agarose gels
stained with ethidium bromide. The presence of
one band indicated males (ZZ), whereas two bands
indicated females (ZW).

We stopped visiting nests once nestlings were
banded and blood-sampled to avoid inducing
premature fledging. However, we continued
observing the nest daily or every other day from
a distance of 5-10 m to identify the date of
fledging. Host and parasite fledglings typically
remain in the natal territory for at least 3 weeks
(Fraga 1991), and we could confidently identify
whether young fledged successfully or not.

We artificially parasitized 56 Baywing nests
with fresh Shiny Cowbird eggs (// = 5 1 ) or newly

420

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 3, September 2010

hatched nestlings {n = 5) obtained from parasit-
ized Chalk-browed Mockingbird (Mimus saturni-
nus) and House Wren nests. Shiny Cowbird eggs
laid in Chalk-browed Mockingbird and House
Wren nests did not differ in width from those in
Baywing nests (ANOVA: Fip?, = 2.3, P = 0.10;
D. T. Tuero, unpubl. data). However, eggs from
Chalk-browed Mockingbird nests were longer
(ANOVA: F2,98 = 8.7, P < 0.001) and had
greater volume (ANOVA: F2p^ = 5.8, P < 0.004)
than those in nests of the other two host species
(Tukey post-hoc test: P < 0.05; D. T. Tuero,
unpubl. data). Artificially parasitized nests were
checked as described above, and data on eggs and
nestlings were collected in the same way as in
naturally parasitized nests.

Data Analysis. — Frequency of parasitism was
calculated as number of nests with parasite eggs
or chicks over total number of nests found and
intensity of parasitism as number of parasitic eggs
per parasitized nest. Overall intensity of parasit-
ism was estimated using a subset of nests found
before or during host laying to avoid underesti-
mating the number of cowbird eggs that were laid
asynchronously and later removed from the nest
cup by Bay wings (Hoy and Ottow 1964, Fraga
1998, De Marsico and Reboreda 2008). We
compared the intensity of parasitism among years
for each cowbird species using the Kruskal-Wallis
test. Data from 2002 were excluded for the
analysis corresponding to Shiny Cowbirds be-
cause only one of the parasitized nests was found
early in the nesting cycle. We used the Kolmo-
gorov-Smimov test for two independent samples
(Siegel and Castellan 1989) to compare the
frequency distribution of Baywing nesting at-
tempts and frequency of parasitic events of
Screaming Cowbirds during the breeding sea.son
(divided into weekly intervals). We estimated the
extent of overlap between the breeding seasons of
Baywings and Shiny Cowbirds using laying dates
of parasite eggs in nests of Baywings (n = 23) and
of two primary hosts in the study area. Chalk-
browed Mockingbird (n = 335 eggs) and House
Wrens (n = 176 eggs). Data on laying dates for
these species were collected during 2003, 2006,
and 2007 (V. D. Fiorini and D. T. Tuero, unpubl.
data).

We estimated clutch size of Baywings using a
subsample of 32 nests found before laying, visited
daily during laying, and which survived at least
until laying ended. We used the.se criteria to avoid
missing egg losses becau.se hosts often remove

punctured eggs (Massoni and Reboreda 2002,
Astie and Reboreda 2006). We estimated total
clutch size from nests parasitized after the host
began to lay that survived until hatching. We used
two estimates; initial clutch size, defined as the
number of host and parasite eggs that remained in
the nest by the end of host laying, and final clutch
size, defined as the number of host and parasite
eggs by the end of incubation. Host incubation
period was estimated as the number of days
elapsed since laying of the last host egg until
hatching of the last host chick (Nice 1954). The
incubation period of Screaming and Shiny cow-
bird eggs was estimated as the number of days
elapsed since the start of incubation until the day
of hatching of the parasite chick (Briskie and
Sealy 1990). We calculated the number of parasite
chicks at hatching from nests that were parasitized
during host laying that survived until the nestling
stage.

We calculated egg volume (cm^) for host, and
Screaming and Shiny cowbirds as 0.515 / w’,
where / and w are egg length- and width (in cm);
0.515 is a species specific constant (Nolan and
Thompson 1978). Measures of Bay wing eggs
were averaged over host clutch for statistical
analyses. The sample of Shiny Cowbird eggs
included those in Baywing nests (n = 1 1 ) as well
as those in nests of Chalk-browed Mockingbird (/?
= 61) and House Wrens (n = 29) from the same
study area (D. T. Tuero, unpubl. data).

Growth curves were built using data from
nestlings that survived at least 10 days and were
weighed daily or every other day. Nestlings that
were depredated or died in the nest were excluded
from the analyses. We used non-linear regressions
to adjust daily weights of each nestling following
the logistic function for weight: W = A/(l +
e\p(—K*(T — To))), where A is the asymptotic
weight, K is the growth constant and Tq is the age
of maximum growth (Ricklefs 1967). Regressions
used a non-linear model and least-squares esti-
mation minimized with the Quasi-Newton option
in SYSTAT 9.0 (SPSS, Chicago, IL, USA).
Parameter estimates and weight at hatching were
obtained for each individual nestling from the
adjusted curves. Maximum growth rate (gmax) was
calculated using the expression: grnax = K * A/4
(Richner 1991). Host nestlings from the same
brood were not independent, and growth param-
eter estimates of individual nestlings from the
same nest were averaged to generate a growth
curve for each brood. We used data from 69

Dc- Marsico et al. • NESTING ECOLOGY OE BAYWINGS

421

Baywing nestlings in 25 nests (5 broods with I, 4
with 2, 9 with 3, 6 with 4, and I with 5 nestlings)
to build growth curves. Five broods were
unparasitized, 12 were parasitized by Screaming
Cowbirds only (9 with 1, and 3 with 2 parasitic
chicks), three were parasitized by Shiny Cowbirds
only (each with a single parasitic chick), and five
had one chick of each parasite species. Mean
(± SE) brood size was 4.0 ± 0.3 chicks per brood,
including host and parasite chicks. We obtained
growth data for 18 Screaming Cowbird nestlings
from 13 broods (10 with 1, 3 with 2, and 1 with 5
Screaming Cowbird chicks). These broods had, on
average, 2.9 ± 0.4 host chicks (range = 0^), and
three also had one Shiny Cowbird chick. Mean
brood size was 4.7 ± 0.4 chicks per brood (range
= 1-8). Data from Screaming Cowbird chicks
from the same brood were averaged. We also
obtained growth data for 10 Shiny Cowbird
nestlings from 10 different broods. These broods
had, on average, 3.4 ± 0.7 host chicks (range =
2-5), and four were also parasitized by Screaming
Cowbirds (3 with 1, and 1 with 2 Screaming
Cowbird chicks). Mean brood size was 4.9 ± 0.3
(range = 3-6). We ascertained the gender of 10
Screaming (6 females and 4 males) and seven
Shiny (5 females and 2 males) cowbird nestlings.
We used this subsample to obtain sex-specific
growth parameters of each parasite species, but
comparisons of growth parameters among Bay-
wing, and Screaming and Shiny cowbird nestlings
used the full data set.

The date of fledging was established directly in
a few cases where we observed the fledglings
leaving the nest. Chicks in most nests fledged
between two successive visits, and we assumed
that in half of the cases they did it the same day
we observed them out of the nest, and in the other
half, that they fledged 1 day earlier.

We estimated nest survival as the proportion of
nests that produced at least one host or parasite
fledgling. Host and parasite reproductive success
was estimated using three parameters: (1) egg
survival, calculated as the proportion of host or
parasite eggs that survived until the end of
incubation in nests that survived until the nestling
stage; (2) hatching success, calculated as the
proportion of host or parasite eggs that hatched
from those that survived; and (3) chick survival,
calculated as the proportion of host or parasite
chicks that fledged from those that hatched in
nests that survived the entire nesting cycle.
Survival of parasite eggs was estimated from

nests found before or during host laying that
survived until hatching. We considered as para-
sitized those clutches that received cowbird eggs
during laying or incubation to analyze the effect
of parasitism on egg survival and hatching
success. We considered as parasitized those nests
with parasite chicks to analyze the effect of
parasitism on nestling survival. Counts included
parasitic eggs or chicks artificially placed in
Baywing nests. Overall host productivity in
successful nests was estimated as the number of
host chicks fledged divided by host clutch size.
Productivity of Screaming and Shiny cowbirds
was estimated as the number of parasite chicks
fledged divided by the total number of parasite
eggs laid in nests that were found before or during
host laying and survived until fledging, including
those that were laid before the beginning of host
laying. Only naturally parasitized nests were
considered to estimate productivity.

Statistical tests followed GenStat Discovery
Edition 3 (GenStat 2007) and results are presented
as T ± SE. All tests are two tailed and significance
was accepted at P < 0.05.

RESULTS

Nest Site Use. — We found 193 Bay wing nests.
Eighty-nine (46%) were in old, closed nests made
of sticks built by species of the Fumariidae (e.g.,
Anwnbius annumbi, Phacellodoinus spp., Synal-
laxis spp.), 40 (21%) in large nest boxes, 16 (8%)
in old Rufous Hornero {Furnarius rufus) nests, 16
(8%) in old Great Kiskadee (Pitangus sulphur-
atus) nests, 14 (7%) in cavities in trees, 10 (5%)
in small nest boxes previously used by House
Wrens, five (3%) were open cup-nests built by
Baywings, and three (2%) were in unusual nesting
sites (an old Chalk-browed Mockingbird nest, a
broken tree branch, and an old nest of the paper
wasp [Polybia sciiteUaris]). There were no
differences among years in frequency of nests
found in natural or artificial nesting sites (x‘4 =
6.6, P = 0.16).

Considering nests with known clutch initiation
date (/? = 116), the median date of clutch
initiation differed among nest types, indicating
that use of nest sites varied during the breeding
season (Kioiskal-Wallis test: 7/5 = 13.9, P =
0.017, 77 = 12 Great Kiskadee nests, 27 nest
boxes, 9 natural cavities, 53 old furnarid nests
made of sticks, 7 Rufous Hornero nests, and 8
nests of other types). Post-hoc comparisons
indicated that old Great Kiskadee nests were

422

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 3. September 2010

TABLE 1. Clutch and brood composition at Baywing nests in the Reserve El Destino (2002-2006). Bw =
unparasitized nests. Bw -r Sc = nests parasitized by Screaming Cowbirds only, Bw + Sh = nests parasitized by Shiny
Cowbirds only, and Bw + Sc + Sh = nests parasitized by both species. Only naturally parasitized nests with known clutch
size were considered for egg and chick counts. Values represent number of eggs or nestlings of each species as x ± SE and
the corresponding sample sizes (number of nests).

Bw

Bw + Sc Bw + Sh

Bw + Sc + Sh

Eggs laid

Bw

4.1 ± 0.3 (12)

3.7 ± 0.1 (142) 0(1)

3.8 ± 0.2 (21)

Sc

4.9 ± 0.3 (142)

4.6 ± 0.5 (21)

Sh

1.0(1)

1.2 ± O.I (21)

Eggs incubated

Bw

4.6 ± 0.2 (7)

3.5 ± 0.1 (80)

3.6 ± 0.3 (11)

Sc

2.4 ± 0.2 (49)

1.8 ± 0.4 (21)

Sh

0.6 ± 0.2 (21)

Young at hatch

Bw

4.4 ± 0.2 (5)

3.2 ± 0.2 (49)

2.8 ± 0.5 (6)

Sc

1.0 ± 0.1 (49)

1.0 ± 0.4 (6)

Sh

0.7 ± 0.3 (6)

Young Oedged

Bw

4.7 ± 0.3 (3)

2.9 ± 0.2 (28)

2.7 ± 0.9 (3)

Sc

1.2 ± 0.2 (28)

0.3 ± 0.3 (3)

Sh

0.3 ± 0.3 (3)

occupied significantly later than small nest boxes
{P < 0.05). Baywings used non-active nests in
most cases, but in two cases we observed
usurpation of active nests ( 1 nest of Firewood-
gatherer [Amimbius anmimhi] and I nest box
occupied by House Wrens). Only five nests in our
sample were built by Baywings; all were open cup
nests with a lax structure of grasses and a ‘roof
provided by surrounding foliage.

Nest reuse occurred both within and among
breeding seasons. We observed nest reuse in 51/
51 ca.ses within breeding sea.sons when the nest
failed due to clutch ejection by hosts in response
to multiple parasitism, but we only observed nest
reuse in 1/46 cases when the nest failed due to
predation. We ob.served nest reuse in 1 1 cases
among breeding seasons (nest sites that were used
in 2 or more consecutive breeding seasons),
although we could not establish if the same
breeding pair was involved. Reuse occurred in
five of these cases in nests that had been
successful in the previous breeding season
whereas in the others, the nest failed in the
previous breeding season.

Incidence of Cowhird Parasitism. — Ninety-four
percent of the nests (181/193) were parasitized,
155 by Screaming Cowbirds only, one by Shiny
Cowbirds only, and 25 by Screaming and Shiny

cowbirds. Frequency of parasitism did not differ
among years (Screaming Cowbird: yf4 = 0.38, P
= 0.98; Shiny Cowbird: = 7.19, P = 0.13).

There were no differences in frequency of
parasitism combining data from all years between
natural nests (/; = 153) and nest boxes {n = 40)
for Screaming (x'l = 1.44, P = 0.76) and Shiny
(X'l = 0.52, P = 0.47) cowbirds.

Average intensity of parasitism was 5.0 ± 0.3
eggs for Screaming Cowbirds {n = 140 nests) and
1.4 ± 0.1 eggs for Shiny Cowbirds {n = 19 nests;
Table 1). We did not detect differences among
years in intensity of parasitism by Screaming
(Kruskal-Wallis test: H4 = 5.9, P = 0.21) or
Shiny (Kruskal-Wallis test: H4 — 3.3, P = 0.34)
cowbirds; the later result should be inteipreted
cautiously given the small sample size. Pooling
data from all years indicated the intensity of
Screaming Cowbird parasitism did not differ
between natural nests {n = 98) and nest boxes
(// = 42; Mann-Whitney C-test: Z = —0.31, P =
0.76). Similarly, the intensity of Shiny Cowbird
parasitism did not differ between natural nests (/;
= II) and nest boxes (/? = 8; Mann-Whitney
C-test: Z = -0.61, P = 0.54).

Screaming Cowbirds parasitized Baywing nests
either before, during, and after host laying. On
average, a parasitized nest received 1 .3 ± 0.3

De Marsico el al. • NESTING ECOLOGY OE BAYWINGS

423

Screaming Cowbird eggs before host laying, 1.2
± 0.2 during host laying, 1.1 ±0.2 eggs during
the incubation stage, and 0.4 ± 0.1 eggs after
nestlings had hatched (/; = 39 parasitized nests
found before or during host laying that survived
until the nestling stage). Eggs laid before host
laying were removed from the nest by Baywings
before they started to lay. Eggs laid after the host
began laying were accepted and incubated, but
only those laid during host laying typically
hatched. Shiny Cowbirds also parasitized Bay-
wing nests either before or during host laying.
Nests received, on average, 0.8 ± 0.6 and 0.5 ±
0.6 Shiny Cowbird eggs before and during host
laying, respectively (n = 18 parasitized nests).

Breeding Seasons. — The earliest and latest
clutch initiation dates for Baywings were 27
November and 7 Eebruary, respectively. The first
Screaming Cowbird parasitism was recorded on
26 November (in a nest at pre-laying stage) and
the latest on 11 February (Fig. 1). The frequency
of Screaming Cowbird parasitism matched the
frequency distribution of Baywing nesting at-
tempts in all breeding seasons (Kolmogorov-
Smimov test: P > 0.10 for all comparisons;
Fig. 2).

The earliest Shiny Cowbird parasitism event
was recorded on 29 September in a Chalk-browed
Mockingbird nest and the latest, on 23 January in
a Baywing nest. Even though Shiny Cowbirds
started laying earlier than Baywings, they over-
lapped considerably in breeding seasons (Fig. 3).
Shiny Cowbird parasitism mostly (44%) occuiTed
in December and January. Nearly 90% of all
Baywing clutches were initiated during the same
interval. Only 10% of Bay wing nesting attempts
began too late to be parasitized by Shiny
Cowbirds.

Clutch Size, Incubation, and Nestling Periods. —
Baywings laid, on average, 4.0 ± 0.1 eggs per
nest (range = 2-5, mode = 4, « = 32 nests), and
host clutch size did not vary with clutch initiation
date (Spearman rank correlation: r = 0.22, Z =
1.23, P = 0.23, n = 32 nests). Initial clutch size in
nests parasitized solely by Screaming Cowbirds
was 5.3 ± 0.2 eggs, and final clutch size was 6.4
± 0.2 eggs (n = 49 nests). Only two nests
naturally parasitized by Shiny Cowbirds survived
until hatching, and they were also parasitized by
Screaming Cowbirds. Initial clutch size in these
nests was 7.0 ± 1.0 eggs, and final clutch size was
8.0 ± 1.0 eggs (Table 1).

Host incubation period was 13.0 ± 0.1 days

■^T-ooincNia)“2f^®^cNa)

(MOOt-cNCNjO'^'^C^OO

FIG. 1. Breeding sea.son of (A) Baywings and (B)
Screaming Cowbirds in Reserve El Destino between 2002
and 2006. Bars represent the number of nesting attempts
initiated (;? = 160) or the total number of Screaming
Cowbird eggs laid (n = 686) during the breeding season
divided into weekly intervals.

with a range of 12-14 days and a mode of 13 (n =
42 clutches). Screaming and Shiny cowbirds had an
incubation period of 12.0 ± 0.1 days (range = 1 1-
13 days, n = 40 and 14 eggs of Screaming and
Shiny cowbirds, respectively). Average number of
Baywing chicks at hatching was 3.2 ± 0.1 (range
= 0—5, n = 60). Nests parasitized by Screaming
Cowbirds only, had 1 .3 ± 0.2 parasite chicks at
hatching (n = 37). Nests naturally parasitized by
Shiny Cowbirds (n = 2) had one Shiny Cowbird
and one Screaming Cowbird chick each. Combin-
ing host and parasitic chicks, Baywings reared, on
average, 4.2 ± 0.2 chicks per nest (range = 1-8, n
= 60 nests; Table 1). The mean nestling period
was 14 days (range = 12-16; n = 20 nests).

424

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122. No. 3. September 2010

2003

2004

(U

>

2005

<M

ZQQCiQQ->-5-3^

<N

ooLncgcn'C'^cncDcNjoj

Ot-CNJCsjO'^'^'^OO

FIG. 2. Baywing and Screaming Cowbird breeding seasons. Curves represent the cumulative relative frequency of
Baywing nesting attempts (white) and parasite eggs laid (black) along the breeding season. Sample sizes are: 2003 = 26
nests and 78 eggs, 2004 = 29 nests and 130 eggs, 2005 = 35 nests and 139 eggs, and 2006 = 60 nests and 312 eggs. Year
2002 was excluded due to small sample sizes.

Egg Size and Nestling Growth. — Egg size and
volume differed between host and parasites
(Table 2). Shiny Cowbird eggs were wider than
those of Baywings and Screaming Cowbirds

(ANOVA: F2.552 = V6.8, P < 0.001) and had
greater volume (ANOVA: F2.552 = 52.2, P <
0.001; Tukey post-hoc test: P < 0.05). Egg length
did not differ among species (ANOVA: F2.552 =

■g

25 80

</)

O)

O)

(D

$

o

U

c

in

C/5

M—

o

0

JD

E

13

70
60
50
40 ^
30
20
10
0

Q.

-5--

U

0

0

tj

>

>

>

>

0

0

0

0

0

c

c

C

C

0)

0

0

0

0

0

0)

0)

<D

0)

03

CO

ra

(0

ro

C/)

0

0

0

z

Z

z

z

□

Q

Q

Q

Q

—3

—>

~3

“3

G>

CD

CO

0

CO

0

1^

00

in

CM

03

ID

CM

C33

CD

CNJ

0

CN

CM

0

CM

0

0

■<—

CM

CM

0

CM

FIG. 3. Seasonal distribution of Shiny Cowbird parasitism in Reserve El De.stino during the 2002-2006 breeding
seasons. Bars represent the number of parasite eggs laid within each weekly interval in ne.sts of Baywings, Chalk-browed
Mockingbirds, and House Wrens (n = 534). The dashed line indicates clutch initiation date of the earliest Baywing nesting
attempt during this study (27 Nov).

De Marsico cl al. • NESTING ECOLOGY OE' BAYWINCjS

425

TABLE 2. Length, width, and volume (.v ± SE) of Baywing, and Screaming and Shiny cowbird eggs. Values
coiTesponding to host eggs represent average egg size and volume of all eggs in 98 clutches. Some Bay wing (Bw) nests in
our sample were artificially parasitized with Shiny Cowbird eggs collected from Chalk-browed Mockingbird (CbM) and
House Wren (HW) nests, and we present data on size and volume of Shiny Cowbird eggs in the three host species. Data
from Chalk-browed Mockingbird and House Wren nests were provided by V. D. Fiorini and D. T. Tuero (unpubl. data).

Specie.s

Host

n

Length (cm)

Width (cm)

Volume (cm')

Baywing

346

2.36 ± 0.01

1.78 ± 0.00

3.87 ± 0.02

Screaming Cowbird

Bw

357

2.37 ± 0.01

1 .79 ± 0.00

3.94 ± 0.02

Shiny Cowbird

Bw

1 1

2.30 ± 0.03

1.86 ± 0.00

4.11 ± 0.16

CbM

61

2.43 ± 0.02

1.89 ± 0.01

4.49 ± 0.05

HW

29

2.34 ± 0.00

1.86 ± 0.00

4.20 ±0.10

1.6, P = 0.20). Differences in egg volume were
not attributable to larger Shiny Cowbird eggs
coming from Chalk-browed Mockingbird nests
because the results did not change when these
eggs were removed from the data set (ANOVA;
P2.491 = 8.9, P < 0.001).

We found differences in growth parameters
between host and parasite chicks, as well as
between Screaming and Shiny cowbird chicks
(Table 3, Fig. 4). Overall, both Screaming and
Shiny cowbirds had higher growth constant
(ANOVA: p2,45 = 6.5, P = 0.003), maximum
instantaneous growth rate (ANOVA: P2.45 — 45.8,
P < 0.001 ), and asymptotic weight (F2,45 — 37.3,
P < 0.001) than Bay wing nestlings (Tukey post-
hoc test, P < 0.05). Shiny Cowbirds had a higher
weight at hatching (ANOVA: F2A5 — 6.7, P =
0.003) and lower age of maximum growth (^2,45
= 8.8, P < 0.001; Tukey post-hoc test, P <
0.001) than both Screaming Cowbird and host
nestlings.

Sex-specific growth curves for Screaming and
Shiny cowbirds varied (Table 3, Fig. 4). Male
Screaming Cowbird nestlings (77 = 4) did not
differ from females (/? = 6) in weight at hatching
(/-test: / = — 1 . 15, df = 8, P = 0.28), growth

constant (/-test: / = —0.09, df = 8, P = 0.93), and
age of maximum growth (/-test: / = —0.46, df =
8, P = 0.66; Table 3). However, male nestlings
had a higher maximum instantaneous growth rate
(/-test: / = —3.84, df = 8, P = 0.005) and reached
a higher asymptotic weight than females (/-test: /
= -5.53, df = 8, P < 0.001; Table 3). Sample
sizes for Shiny Cowbirds were too low for
statistical analyses, but growth curves suggest
that differences in males and females similar to
those of Screaming Cowbirds might occur in this
species (Table 3, Fig. 4).

Baywing Reproductive Success. — Only 35
(19%) of all nests fledged chicks and all but one
produced host fledglings. The exception was a
parasitized nest where all host eggs failed to hatch
and only two Screaming Cowbird chicks fledged.
The number of Baywing chicks fledged per nest
was 3.0 ± 0.2 (77 = 34 nests, range = 1-5).
Overall, host productivity in nests that survived
the entire nesting cycle was 0.78 fledglings per
egg laid (77 = 25 nests with known clutch size).
Losses were mostly due to egg punctures by
Screaming and Shiny cowbirds and hatching
failures. On average, survival of host eggs was
0.92 (77 = 54 nests). Egg survival was higher in

TABLE 3. Growth parameters (x ± .SE) of Bay wing, and Screaming and Shiny cowbird nestlings obtained from
adjusted growth curves. Data for host nestlings were averaged over each brood. Data for parasite nestlings were based on
individual nestlings of each gender. Sample sizes = 69 Baywings in 25 broods, 6 female and 4 male Screaming Cowbirds,
and 5 female and 2 male Shiny Cowbirds.

Screaming Shiny

Parameter

Baywing

Male

Female

Male

Female

Hatch weight, g

3.6 ± 0.1

3.8 ± 0.4

3.3 ± 0.2

4.9 ± 0.3

4.4 ± 0.2

Growth constant/day

0.47 ± 0.01

0.50 ± 0.02

0.49 ± 0.02

0.5 1 ± 0.03

0.5 1 ± 0.03

Maximum growth rate, g/day

4.25 ± 0.07

6.37 ± 0.14

5.23 ± 0.22

6.10 ± 0.14

5.31 ± 0.32

Age of maximum growth, days

5.8 ± 0.1

6.2 ± 0.2

6.0 ± 0.2

5.3 ± 0.2

5.2 ± 0.1

Asymptotic weight, g

36.7 ± 0.4

51.6 ± 1.9

42.5 ± 0.6

48.3 ± 1.8

41.4 ± 1.0

426

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 3, September 2010

Screaming Cowbird

Shiny Cowbird

O)

sz

D)

0

$

cn

c

C/)

0

Age (days)

FIG. 4. Growth curves of Screaming and Shiny cowbird nestlings reared by Baywings. Points correspond to daily
weights (x ± SE) as a function of age for male (black circles) and female (white circles) nestlings of each species, estimated
from the corresponding adjusted logistic functions. The dotted line in each graph represents the growth curve of the host
with adjusted daily weights (.v ± SE) averaged across all host nestlings within the same brood (/? = 25 broods with 1-5 host
nestlings) Sample sizes were four and six female Screaming Cowbirds, and two and five female Shiny
Cowbirds, respectively.

nests that were not parasitized during the egg
stage than in parasitized nests (Mann-Whitney
(/-test: Z = -2.03, P = 0.042, n = 10
unparasitized and 44 parasitized nests). On
average, hatching success was 0.88 {n = 54
nests). Hatching success did not differ between
parasitized and unparasitized nests (Mann-Whit-
ney (/-test: Z = 0.00, P > 0.99). There were no
statistical differences in survival of Baywing
chicks between parasitized {n = 20) and unpar-
asitized nests (n = 8; Mann-Whitney (/-test: Z =
— 1.52, P = 0.13). Nestling survival was, on
average, 0.94 {n = 28 nests). Brood reduction
occurred in six parasitized nests where one (n =
5) or two (/? = 1) host nestlings died due to
starvation or nest crowding.

Screaming Cowbird Reproductive Success. —
About 89% (615/694) of the Screaming Cowbird
eggs in Baywing nests failed as a result of nest
predation or desertion (/? = 146 naturally

parasitized nests found before or during host
laying). Survival of Screaming Cowbird eggs in
nests that survived until hatching was 0.93 {n =
48 nests). Hatching success was 0.62 (/? = 47
nests), and hatching failures were mostly due to
parasitic females laying eggs too late for success-
ful incubation. Hatching success of Screaming
Cowbird eggs laid during host laying was 0.92 (//

= 40 nests). Overall nestling survival was 0.77 (/?
= 20 nests). Mortality of parasitic chicks occun'ed
at six nests as a result of late hatching (/? = 4
nests) and partial nest predation (n = 2 nests).
Screaming Cowbird productivity was 0.14 fledg-
lings per egg laid (n = 20 nests).

Shiny Cowbird Reproductive Success. — Nearly
91% of Shiny Cowbird eggs naturally laid (30/33,
n = 26 nests) and 82% of those artificially placed
in Baywing nests (40/49, n = 49 nests) failed due
to nest predation or desertion. Combining natural
and artificial parasitism. Shiny Cowbird egg
survival was 0.95 (/? = 19 nests) and hatching
success was 0.94 (/? = 18 nests). Nestling survival
was 0.91 (n = 11 nests) and only one Shiny
Cowbird chick died, presumably due to partial
nest predation. Sample size was too small to
estimate productivity. Only one of three Shiny
Cowbird eggs naturally laid produced a fledgling
{n — 3 nests). The others were laid before host
laying and therefore rejected by hosts.

DISCUSSION

Nest Site Use. — Baywings used a wide variety
of nesting sites, but the more commonly used
were old, clo.sed nests built by several furnarid
species. This ob.servation contrasts with a previ-
ous study in a nearby area, which reported that

De Mursico el al. • NESTING ECOLOGY Ol- FiAYWINGS

427

nest boxes and secondary cavities were preferred
nesting sites (Fraga 1988, 1998). These differenc-
es, however, could be explained by local variation
in abundance of suitable nesting sites at the time
Bay wings start to breed (Hoy and Ottow 1964).
Some authors have observed Baywings fighting
with owners to occupy their nests (Friedmann
1929, Jaramillo and Burke 1999), but we seldom
observed nest piracy, suggesting that nesting sites
were not limiting during our study (see also Hoy
and Ottow 1964, Fraga 1988). It is not clear why
in most cases Baywings use old nests of other
species instead of building their own nests, but it
is possible this behavior allows breeding pairs to
save time and energy (Hauber 2002, Wiebe et al.
2007). Additionally, closed nests might offer
better protection against predators than open cup
nests (Martin and Li 1992, Auer et al. 2007).

Breeding Seasons. — Our results indicate the
breeding season of Baywings typically starts in
early December and may extend to early March,
considering nestling and postfledgling stages of
late nesting attempts. Fraga (1998), in the
Province of Buenos Aires, observed laying as
early as October, but we did not observe any sign
of nesting activity before late November. The
breeding season of Screaming Cowbirds over-
lapped extensively that of Baywings. However,
female Screaming Cowbirds started parasitism in
advance of host laying indicating that nest
building and defense activities by the host may
stimulate laying behavior of the parasite. It is also
possible the parasite’s breeding season began
earlier if Screaming Cowbirds in our study area
also parasitize Brown-and-yellow Marshbirds,
which breed from October to December (Mermoz
and Reboreda 1996, Mermoz and Fernandez
2003). Shiny Cowbirds began to breed much
earlier in our study area than Baywings, parasit-
izing other hosts such as Chalk-browed Mocking-
birds and House Wrens starting in late September
(Fiorini and Reboreda 2006, Tuero et al. 2007).

Friedmann (1929) suggested differences in time
of breeding would be the cause of the low
frequency of Shiny Cowbird parasitism in nests
of Baywings. However, our data indicate only
10% of the nesting attempts of Baywings began
after the Shiny Cowbird breeding season ended.
Breeding seasons of Baywings and Shiny Cow-
birds overlapped more extensively than previously
estimated (90 vs. 65-80%; Fraga 1998), and
female Shiny Cowbirds would have broad oppor-
tunities for parasitizing this host. An alternative

explanation for the low frequency of Shiny
Cowbird parasitism of this host is that the parasite
would have low reproductive success with Bay-
wings (Fraga 1998), but this hypothesis remains
untested.

Incidence of Cowbird Parasitism. — Nearly all
Baywing nests were parasitized by Screaming
Cowbirds, and regularly more than once (Hoy and
Ottow 1964, Mason 1980, Fraga 1998). These
high levels of parasitism did not seem to be an
artifact of nest boxes (Rattan 1997), as the
frequency and intensity of parasitism by Scream-
ing and Shiny cowbirds did not differ between
nest boxes and natural sites. Our results did not
appear to be influenced by the shortage of host
nests in a particular year, as frequency and
intensity of parasitism were consistent among
consecutive breeding seasons. Females of other
brood parasites parasitize nests within nearly
exclusive breeding areas, which may allow them
to reduce the costs associated with intraspecific
competition (Langmore et al. 2007). Several
observational and genetic studies indicate that
cowbirds, however, lack territorial behavior
(Rattan 1997, Hahn et al. 1999, Mermoz and
Reboreda 1999, McLaren et al. 2003, Strausberger
and Ashley 2005, Ellison et al. 2006). Thus,
multiple parasitism can arise as a consequence of
a limitation of suitable nests, high fecundity rates
of female cowbirds, or high population density of
the parasite relative to the number of host
breeding pairs (Martinez et al. 1998, Strausberger
1998, McLaren et al. 2003, Ellison et al. 2006).
Multiple parasitism by Screaming Cowbirds,
which are constrained to locate and parasitize
near a single host species, could also be the result
of different females parasitizing the same nest and
individual females parasitizing a nest more than
once as a result of low nest availability.

Baywing Reproductive Success. — Cowbird par-
asitism reduces host fitness in several ways
(Massoni and Reboreda 1998, Lorenzana and
Sealy 1999, Zanette et al. 2005, Astie and
Reboreda 2006). We found parasitized nests lost
more host eggs than unparasitized nests, but
hatching success and nestling survival did not
differ between parasitized and unparasitized nests.
Our results indicate survival of Baywing eggs and
nestlings was higher than in previous reports
(Eraga 1986, 1998), possibly because, during our
study, most host egg and nestlings were lost due to
nest desertion and predation rather than to egg
punctures or competition with parasite nestlings.

428

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 3. September 2010

We rarely observed brood reduction in parasitized
nests and nestling mortality due to parasitic mites,
as reported previously (Fraga 1986). Regular
cleaning of nest boxes may have contributed to
decrease the incidence of ectoparasite infestations
in our study population. However, given that most
nests occurred in natural sites, we have confidence
that box cleaning had little influence on our
estimates of nestling survival. Costs of parasitism
for Baywings are similar to those observed in
other host species of the Shiny Cowbird that are
larger than the parasite (Mermoz and Reboreda
2003, Astie and Reboreda 2006). Productivity is
often lowered in hosts larger than the parasite by
puncture or removal of host eggs by female
cowbirds rather than by nestling mortality, as host
young are seldom outcompeted by parasitic
nestlings (Eckerle and Breitwisch 1997, Lichten-
stein 1998, Lorenzana and Sealy 1999, Astie and
Reboreda 2006), unless a marked hatching
asynchrony exists (Soler et al. 1996, Dure Ruiz
et al. 2008). Baywings are smaller than Screaming
and Shiny cowbirds; thus, it is expected that host
nestlings compete strongly for food with the
parasite young. However, Baywings are cooper-
ative breeders (Fraga 1991) and it is possible that
helpers at the nest contribute to ameliorate the
costs of parasitism by increasing nest provisioning
rates (Fraga 1991).

Nestling Growth. — We did not find differences
in growth patterns between Screaming and Shiny
cowbird nestlings in nests of Baywings. Overall,
Screaming and Shiny cowbird nestlings did not
differ in asymptotic weight, growth constant, and
maximum instantaneous growth rate. This is
consistent with similar comparisons between
parasite nestlings of both species reared in
Brown-and-yellow Marshbird nests (Mermoz
and Fernandez 2003). However, parasite nestlings
differed in weight at hatching and age of
maximum growth, which could be the result of
differences in egg size between both species
(Nolan and Thompson 1978). A higher weight at
hatching combined with a shorter incubation
period could provide Shiny Cowbird nestlings a
head start to overcome competition, which can be
particularly important in nests of host species
larger than the parasite (Astie and Reboreda 2006,
Fiorini et al. 2009). However, evidence for a
positive effect of egg size on nestling performance
is equivocal (e.g., Schifferli 1973, Magrath 1992,
Reed et al. 1999).

Screaming Cowbird eggs and hatchlings were

similar in mass to those of Baywings, although
parasitic juveniles and adults are quite larger than
the host (Mason 1987, Fraga 1998). Females in
other parasite species have evolved eggs similar in
size to those of the host, presumably in response
to host discrimination against larger or smaller
eggs (Mason and Rothstein 1986, Marchetti
2000). No experiments have been conducted on
Baywings to test whether they are able to
discriminate eggs by size. However, that they
readily accepted larger Shiny Cowbird eggs
artificially placed in their nests (this study)
suggests Baywings do not discriminate foreign
eggs on the basis of size. Screaming Cowbird eggs
in Baywing nests usually have low chances of
success as a consequence of poor synchronization
of parasitism and high nest failure rates (Hoy and
Ottow 1964, Fraga 1998, De Marsico and
Reboreda 2008). Thus, a selective pressure may
exist on female Screaming Cowbirds to increase
the total number of eggs laid at the expense of
laying smaller eggs (Fontaine and Martin 2006,
Martin et al. 2006).

Screaming Cowbird nestlings were sexually
dimorphic in size and presented sex-specific
growth curves. Males grew faster and were larger
than females over the entire nestling period,
consistent with a recent study of Brown-headed
Cowbirds (Molothrus oter) (Tonra et al. 2008, but
see Weatherhead 1989). Our sample sizes were
too small to compare growth parameters between
male and female Shiny Cowbirds, but our data
indicate that sex-specific growth patterns similar
to that observed in Screaming Cowbirds may
occur in this species. The occun’ence of sex-
specific growth curves among parasitic cowbirds
implies that gender should be considered when
analyzing the costs of competition and factors
affecting nestling growth in host and parasite
species.

ACKNOWLEDGMENTS

We thank Fundacion Elsa Shaw de Pearson for allowing
us to conduct this study at Reserve El Destino. D. T. Tuero
and V. D. Fiorini generously provided data for laying dates
and egg size of Shiny Cowbirds. We are grateful to
reviewers Rosendo Fraga and Peter Lowther for comments
and suggestions that greatly helped to improve the
manuscript. MCDM was supported by a fellowship from
the Con.sejo Nacional de Investigaciones Cientificas y
Tecnicas (CONICET). BM and JCR are re.search fellows of
CONICET. This work was supported by research grants of
Agencia Nacional de Promocion Cienti'fica y Tecnologica.

Dc Marsico cl al. • NESTING ECOLOGY OE BAYWINGS

429

University of Buenos Aires, and the Student Grant Program

of the Neotropieal Grassland Conservaney.

LITERATURE CITED

Astie, a. a. and J. C. Reboreda. 2006. Costs of egg
punctures and parasitism by Shiny Cowbirds (Molo-
thnis bonariensis) at Creamy-bellied Thrush (Tiirdits
amaurochalinus) nests. Auk 123:23-32.

Auer, S. K., R. D. Bassar, J. J. Fontaine, and T. E.
Martin. 2007. Breeding biology of passerines in a
subtropical montane forest in northwestern Argentina.
Condor 109:321-333.

Briskie, j. V. and S. G. Sealy. 1990. Evolution of short
incubation periods in the parasitic cowbirds, Molo-
rhnis spp. Auk 107:789-793.

C.agnoni, M., a. M. Faggi, and A. Ribichich. 1996. La
vegetacion de la Reserva El Destino (Partido de
Magdalena, Provincia de Buenos Aires). Parodiana
9:25-44.

Davies, N. B. 2000. Cuckoos, cowbirds and other cheats. T.
& A. D. Poyser, London, United Kingdom.

De MArsico, M. C. and J. C. Reboreda. 2008. Egg-laying
behavior in Screaming Cowbirds: why does a
specialist brood parasite waste so many eggs? Condor
110:143-153.

Dure Ruiz, N. M., M. E. Mermoz, and G. J. Fernandez.
2008. Effect of cowbird parasitism on brood reduction
in the Brown-and-yellow Marshbird. Condor 1 10:507-
513.

Eckerle, K. P. and R. Breitwisch. 1997. Reproductive
success of the Northern Cardinal, a large host of
Brown-headed Cowbirds. Condor 99:169-178.

Ellegren, H. 1996. First gene on the avian W chromosome
(CHD) provides a tag for universal sexing of non-ratite
birds. Proceedings of the Royal Society of London,
Series B 263:1635-1641.

Ellison, K., S. G. Sealy, and H. L. Gibbs. 2006. Genetic
elucidation of host use by individual sympatric
Bronzed Cowbirds (Molothrus aeneiis) and Brown-
headed Cowbirds (M. ater). Canadian Journal of
Zoology 84:1269-1280.

Fiorinl V. D. and j. C. Reboreda. 2006. Cues used by
Shiny Cowbirds {Molothrus bonariensis) to locate and
parasitize Chalk-browed Mockingbird (Miinus satitr-
niniis) nests. Behavioral Ecology and Sociobiology
60:379-385.

Fiorini, V. D., D. T. Tuero, and J. C. Reboreda. 2009.
Shiny Cowbirds synchronize parasitism with host
laying and puncture host eggs according to host
characteristics. Animal Behaviour 77:561-568.

Fontaine, J. J. and T. E. Martin. 2006. Parent birds assess
nest predation risk and adjust their reproductive
strategies. Ecology Letters 9:428^34.

Fraga, R. M. 1979. Differences between nestlings and
fledglings of Screaming and Bay-winged cowbirds.
Wilson Bulletin 91:151-154.

Fraga, R. M. 1983. The eggs of the parasitic Screaming
Cowbird {Molothrus rufoaxillaris) and its host, the
Bay-winged Cowbird {M. badius): is there evidence
for mimicry? Journal fur Ornitologie 124:187-193.

Fraga, R. M, 1986. The Bay-winged Cowbird {Molothrus
badius) and its brood parasites: interactions, coevolu-
tion and comparative efficiency. Thesis. University of
California, Santa Barbara, USA.

Fraga, R. M. 1988. Nest sites and breeding success of Bay-
winged Cowbirds {Molothrus badius). Journal fiir
Ornithologie 129:175-183.

Fraga, R. M. 1991. The social system of a communal
breeder, the Bay-winged Cowbird Molothrus badius.
Ethology 89:195-210.

Fraga, R. M. 1996. Further evidence of parasitism of
Chopi Blackbirds {Gnorinwpsar chopi) by the spe-
cialized Screaming Cowbird {Molothrus rufoaxillaris).
Condor 98:866-867.

Fraga, R. M. 1998. Interactions of the parasitic Screaming
and Shiny cowbirds {Molothrus rufoaxillaris and M.
bonariensi.s) with a shared host, the Bay-winged
Cowbird {Molothrus badius). Pages 173-193 in
Parasitic cowbirds and their hosts. Studies in coevo-
lution (S. I. Rothstein and S. K. Robinson. Editors).
Oxford University Press, New York, USA.

Friedmann, H. 1929. The cowbirds. A study in the biology
of social parasitism. C. C. Thomas, Springfield,
Illinois, USA.

GenStat. 2007. GenStat Discovery Edition 3. Lawes
Agricultural Trust, Rothamsted Experimental Station,
Hertfordshire, United Kingdom.

Griffiths, R., M. C. Double, K. Orr, and R. J. G.
Dawson. 1998. A DNA test to sex most birds.
Molecular Ecology 7:1071-1075.

Hahn, D. C., J. A. Sedgwick, I. S. Painter, and N. J.
Casna. 1999. A spatial and genetic analysis of
cowbird host selection. Studies in Avian Biology
18:204-217.

Hauber, M. E. 2002. Is reduced clutch size a cost of
parental care in Eastern Phoebes {Sayornis phoebe)!
Behavioral Ecology and Sociobiology 5 1 :503-509.

Hoover, J. P. 2003. Multiple effects of brood parasitism
reduce the reproductive success of Prothonotary
Warblers. Protonotaria citrea. Animal Behaviour
65:923-934.

Hoy, G. and J. Ottow. 1964. Biological and oological
studies of the molothrine cowbirds (Icteridae) of
Argentina. Auk 81:186-203.

Hudson, W. H. 1874. Notes on the procreant instincts of
the three species of Molothrus found in Buenos Aires.
Proceedings of the Zoological Society of London
1 1:153-174.

.lARAMlLLO, J. and P. Burke. 1999. New World blackbirds:
the icterids. Princeton University Press. Princeton.
New Jersey. USA.

Kattan, G. H. 1997. Shiny Cowbirds follow the 'shotgun'
strategy of brood parasitism. Animal Behaviour
53:647-654.

Langmore. N. E., G. j. Adcock, and R. M. Kilner. 2007.
The spatial organization and mating system of the
Horsfield's Bronzed-cuckoos Chalcites basalis. Ani-
mal Behaviour 74:403-412.

Lanyon, S. M. 1992. Interspecific brood parasitism in
blackbirds (Icterinae): a phylogenetic perspective.
Science 255:77-79.

430

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 3. September 2010

Lichtenstein. G. 1998. Parasitism by Shiny Cowbirds of
Rufous-bellied Thrushes. Condor 100:680-687.

Lorenzana, J. C. and S. G. Sealy. 1999. A meta-analysis
of the impact of parasitism by Brown-headed Cow-
birds on its hosts. Studies in Avian Biology 18:241-
253.

Lowther. P. E. and W. Post. 1999. Shiny Cowbird
(Molothriis bonariensis).'Yhe birds of North America.
Number 399.

Magrath, R. D. 1992. The effect of egg mass on the
growth and survival of blackbirds: a field experiment.
Journal of Zoology 227:639-654.

Mahler, B., Y. Sarquis Adamson, A. G. Di Giacomo, V.
A. Confalonieri, and J. C. Reboreda. 2009.
Utilization of a new host in the host-specialist brood
parasite Molothnis rufoaxillaris'. host switch or host
acquisition? Behavioral Ecology and Sociobiology
63:1603-1608.

Marchetti, K. 2000. Egg rejection in a passerine
species: size does matter. Animal Behaviour 59:877-
883.

Martin, T. E. and P. Li. 1992. Life-history traits of open-
vs. cavity-ne.sting birds. Ecology 73:579-592.

Martin, T. E., R. D. Bassar, S. K. Bassar, J. J. Fontaine,
P. Lloyd, H. A. Mathewson, A. M. Niklison, and
A. Chalfoun. 2006. Life-history and ecological
correlates of geographic variation in egg and clutch
mass among passerine species. Evolution 60:390-
398.

Martinez, J. G., J. J. Soler, M. Soler, and T. Burke.
1998. Spatial patterns of egg laying and multiple
parasitism in a brood parasite: a non-territorial system
in the Great Spotted Cuckoo (Clamator glamdarius).
Oecologia 117:286-294.

Mason, P. 1980. Ecological and evolutionary aspects of
host selection in cowbirds. Thesis. University of
Texas. Austin, USA.

Mason, P. 1986. Brood parasitism in a host generalist, the
Shiny Cowbird. II. Host selection. Auk 103:61-69.

Mason, P. 1987. Pair formation in cowbirds: evidence
found for Screaming but not Shiny cowbirds. Condor
89:349-356.

Ma.son, P. and S. I. Rothstein. 1986. Coevolution and
avian brood parasitism: cowbird eggs show evolution-
ary response to host discrimination. Evolution
40:1207-1214.

Massonl V. AND J. C. Reboreda. 1998. Costs of brood
parasitism and the lack of defenses on the Yellow-
winged Blackbird-Shiny Cowbird system. Behavioral
Ecology and Sociobiology 42:273-280.

Mas.soni, V. AND J. C. Reboreda. 2002. A neglected cost
of brood parasitism: egg punctures by Shiny Cowbirds
during inspection of potential host nests. Condor
1()4;407-412.

McLaren, C. M.. T. J. Underwood, and S. G. Sealy.
2003. Genetic and temporal patterns of multiple
parasitism by Brown-headed Cowbirds (Molothm.s
aler) on Song Sparrows (Melospiz.a melodia). Canadi-
an Journal of Zoology 81:281-286.

Mfrmoz, M. E. and G. j. Fernandez. 2003. Breeding
success of a specialist brood parasite, the Screaming

Cowbird, parasitizing an alternative host. Condor
105:63-72.

Mermoz, M. E. and j. C. Reboreda. 1996. New host for a
specialized brood parasite, the Screaming Cowbird.
Condor 98:630-632.

Mermoz, M. E. and J. C. Reboreda. 1999. Egg-laying
behaviour by Shiny Cowbirds parasitizing Brown-and-
yellow Marshbirds. Animal Behaviour 58:873-882.

Mermoz, M. E. and J. C. Reboreda. 2003. Reproductive
success of Shiny Cowbirds (Molothnis hoiuiriensis)
parasitizing the larger Brown-and-yellow Marshbird
(Pseudoleistes virescens) in Argentina. Auk 120:1 128-
1139.

Miller, S. A., D. D. Dykes, and H. F. Polesky. 1988. A
simple salting out procedure for extracting DNA from
human nucleated cells. Nucleic Acids Research
16:1215.

Nice, M. M. 1954. Problems of incubation periods in North
American birds. Condor 56:173-197.

Nolan, V. and C. F. Thompson. 1978. Egg volume as a
predictor of hatchling weight in the Brown-headed
Cowbird. Wilson Bulletin 90:353-358.

Ortega, C. P. 1998. Cowbirds and other brood parasites.
University of Arizona Press, Tucson, USA.

Payne, R. B. and L. L. Payne. 1998. Brood parasitism by
cowbirds: risks and effects on reproductive success
and survival in Indigobirds. Behavioral Ecology 9:
64-73.

Reboreda, J. C., M. E. Mermoz, V. Massonl A. A. Astie,
and F. L. Rabuffetti. 2003. Impacto del parasitismo
de cria del Tordo Renegrido (Molothnis hoiuiriensis)
sobre el exito reproductive de sus hospedadores.
Hornero 18:77-88.

Reed, W. L., A. M. Turner, and P. R. Sotherland. 1999.
Consequences of egg-size variation in the Red-winged
Blackbird. Auk 116:549-552.

Richner, H. 1991. The growth dynamics of sexually
dimorphic birds and Fisher’s sex ratio theory: does
sex-specific growth contribute to balanced sex ratios?
Functional Ecology 5:19-28.

Ricklefs, R. E. 1967. A graphical method of fitting
equations to growth curves. Ecology 48:978-983.

Schifferli, L. 1973. The effect of egg weight on the
subsequent growth of nestling Great Tits Pams major.
Ibis 115:549-558.

Sick, H. 1985. Ornitologia brasileira: uma introducgao.
University of Brasilia, Brazil.

Siegel, S. and N. J. Castellan. 1988. Nonparametric
statistics for the behavioral sciences. Second Edition.
McGraw-Hill, New York, USA.

Soler, M., J. G. Martinez, and J. J. Soler. 1996. Effects
of brood parasitism by the Great Spotted, Cuckoo on
the breeding success of the magpie host: an experi-
mental study. Ardeola 43:87-96.

Soren.son, M. D. and R. B. Payne. 2002. Molecular
genetic perspectives on avian brood parasitism.
Integrative and Comparative Biology 42:388^00.

Strausberger. B. M. 1998. Temporal patterns of host
availability. Brown-headed Cowbird brood parasitism,
and parasite egg mass. Oecologia 1 16:267-274.

Strau.sberger. B. M. and V. M. A.shley. 2005. Host use

Dc Marsico et at. • NESTING ECOLOGY OF BAYWINGS

431

strategies of individual female Brown-headed Cow-
birds Molothrus ater in a diverse avian community.
Journal of Avian Biology 36:313-321.

Tonra, C. M., M. E. Hauber, S. K. Heath, and M. D.
Johnson. 2008. Ecological correlates and sex differ-
ences in early development of a generalist brood
parasite. Auk 125:205-213.

Trine, C. L. 2000. Effects of multiple parasitism on
cowbird and Wood Thrush nesting success. Pages
135-144 in Ecology and management of cowbirds and
their hosts (J. N. M. Smith, T. L. Cook, S. I. Rothstein,
S. K. Robinson, and S. G. Sealy, Editors). University
of Texas Press, Austin, USA.

Tuero, D. T., V. D. Fiorini, and J. C. Reboreda. 2007.

Effects of Shiny Cowbird Molothrus honariensis
parasitism on different components of House Wren
T’/'og/or/vte.y wcr/w? reproductive success. Ibis 149:521-
529.

Weatherhead, P. j. 1989. Sex-ratios, host-specific repro-
ductive success, and impact of Brown-headed Cow-
birds. Auk 106:358-366.

WiEBE, K. L., W. D. Koenig, and K. Martin. 2007. Cost
and benefits of nest reuse versus excavation in
cavity-nesting birds. Annales Zoologici Fennici
44:209-217.

Zanette, L., E. Macdougall-Shakleton, M. Clinchy,
AND J. N. M. Smith. 2005. Brown-headed Cowbirds
skew offspring sex ratios. Ecology 86:815-820.

The Wilson Journal of Ornithology 122(3):432-438, 2010

BOBOLINK EGG MASS VARIABILITY AND NESTLING

GROWTH PATTERNS

BARBARA FREI,' ' DAVID M. BIRD,' AND RODGER D. TITMAN'

ABSTRACT. — The Bobolink (Dolichonyx oryzivorus) is an obligate grassland species that is declining throughout its
range in North America. There are few data available on Bobolink eggs and nestlings; this information is necessary for
conservation planning efforts. Egg mass was recorded for 175 eggs from 37 nests in Quebec and eastern Ontario in 2006-
2007. Hatching asynchrony was evident with high between-clutch variation in egg mass. Egg mass did not differ with
clutch size. Nest initiation date was positively correlated with smalle.st egg size and negatively correlated with within-clutch
egg mass deviation. Nestling wing length, tarsus length, and mass were measured for 166 nestlings ranging from 2 to
10 days of age. Bobolink nestlings fledged below adult size and mass, achieving 87.7 ± 2.3%, 67.6 ± 1.5%, and 55.1 ±
0.4% of breeding adult tarsus length, mass, and wing length (± SE), respectively. Received 13 December 2008. Accepted 4
February 2010.

The Bobolink (Dolichonyx oryzivorus) is an
obligate grassland species that has recently
received attention concerning population declines,
mainly due to habitat loss and agricultural
intensification (Bollinger et al. 1990, Herkert
1997). Studies have examined aspects of breeding
ecology (e.g., Bollinger and Gavin 1992, Winter
et al. 2004), but little information is available on
eggs and nestlings, including linear nestling
measurements. In addition, reported values are
based on small sample sizes (e.g., mean egg mass,
n — 5; Martin and Gavin 1995).

Nestling growth rates and patterns in passerines
are useful measures to examine the influence of
environmental, parental, and social parameters to
which the nestlings are subject. Knowledge of a
time-frame of critical growth for body compo-
nents may provide insight into vulnerable periods
during the nesting cycle. Nestling growth may be
affected by pre-hatching (e.g., egg mass and
incubation) and post-hatching factors. Egg mass
variability may result from physiological differ-
ences among females or environmental variables
such as geographic location, weather, and food
availability (Christians 2002). As first hypothe-
sized by Lack (1967), a trade-off exists between
clutch size and egg size; females have a limited
amount of energy for egg production that must be
shared between the number of eggs and size of the
eggs. In addition, some species exhibit seasonal
change in egg size (Magrath 1992) or seasonal

' Avian .Science and Con.servation Centre, Department of
Natural Resource Sciences, McGill University, Macdonald
Campus, Ste-Anne-de-Bellevuc, QC H9X 3V9, Canada.

^Corresponding author;
e-mail; barbara.frei@mail.mcgill.ca

effects of within-clutch egg size variability
(Takagi 2003).

Nestling growth patterns may be described using
several different parameters: ( 1 ) growth rate, (2)
percent of adult weight attained by dependent
young in the nest, and (3) shape of the growth curve
(Ricklefs 1968a). Standard growth rate curves
show mass increasing throughout the nestling
period and reaching near-adult weights at fledging.
There are two alternate curves where (a) nestlings
attain a peak mass above adult weight and decrease
to adult weight just before or after fledging, and (b)
growth levels remain below adult values at
fledging (Ricklefs 1968a). Growth rates may vary
between individual broods of the same species, but
growth patterns are species-specific (Ricklefs
1968a). For example, nestlings that fledge above
adult weight are primarily oceanic species and
aerial insectivores with long nestling periods, and
are independent feeders post-fledging (Ricklefs
1968b). Conversely, ground-feeding species fledge
below adult weight and rely on well-developed legs
for escaping predators (Ricklefs 1968b).

The goal of our study was to identify patterns
and variability of egg mass and nestling growth in
a population of Bobolinks by comparing: (1) egg
mass within- and between clutches; (2) the
relationship between (a) clutch size and egg mass
and (b) nest initiation date and egg mass; (3)
nestling growth between years of .5tudy, study
sites and between broods; and (4) Bobolink
nestling growth patterns by percentage of adult
size and mass attained at fledging.

METHODS

Stiuly Area. — This study was conducted in
2006-2007 at three sites in southwestern Quebec

432

Frei Cl al. • BOBOLINK EGG MASS AND NESTLING GROWTH

433

and eastern Ontario with established Bobolink
populations. The sites consisted of a private hay
farm near Hemmingford, Quebec (45° 05' N, 73°
36' W), an agricultural park (Bois-de-la-Roche) in
Senneville, Quebec (45° 26' N, 73° 56' W), and a
wildlife conservation area (Atocas Bay) in
Lefaivre, Ontario (45° 36' N, 74° 51' W). All
sites consisted of multiple hayfields dominated by
timothy (Phleiiiii pratense) and smooth brome
(Bromiis ramosiis ssp. racemosiis) at Hemming-
tord, and a broader mixture of forbs and grasses at
Atocas and Bois-de-la-Roche.

Measurements. — We measured and numbered
eggs on day of discovery or as they were laid, and
weighed them to the nearest 0. 1 g with a portable
electronic balance. We took nestling measure-
ments twice during consecutive visits 3^ days
apart between 0700 and 1500 hrs EST. We
marked individual nestlings on the initial visit
by coloring their toenails using non-toxic perma-
nent markers. Nestlings were separately classified
by age during each visit, as asynchronous
hatching resulted in different ages of nestlings
within the same nest. Nestling age was classified
using a combination of external physical traits;
amount of down, feather tracts, feather eruption
from sheaths, and opening of eyes. Photos were
taken at each visit and compared to known-age
nestlings to confirm classification. We measured
length of the unflattened wing (wing length) to the
nearest 0.5 mm using a ruler with a wing stop, and
tarsometatarsus (tarsus) length to the nearest
0.1 mm with dial calipers. We weighed nestlings
to the nearest 0. 1 g. The same person performed
all measurements to reduce observer bias. We
banded nestlings at > 7 days of age using
aluminum bands. We monitored all nests until
completion, and nests that fledged at least one
young were recorded as successful.

Statistical Analysis. — We checked egg and
nestling data for normality using Shapiro-Wilks’
test (Zar 1999). All measurements were normally
distributed, allowing use of untransformed data in
subsequent analyses. We compared egg mass
between years using Student’s r-test and between
sites using one-way ANOVA (Zar 1999). We
examined the variance of egg mass between and
within clutches using one-way ANOVA. We
examined if mean egg mass per clutch varied
with changes in nest initiation date using linear
regression. We also examined the variance of
mean egg mass with clutch size using one-way
ANOVA. We examined the effect of initiation

date on mass of the smallest egg in each clutch
using linear regression. We u.sed the standard
deviation of mean egg mass per clutch as a
measure of within-clutch egg mass variation. We
regressed standard deviations with Julian initia-
tion date to examine if within-clutch variation
changed with initiation date of the nest.

We measured a subset (/; = 102) of nestlings
twice over the span of 10 days, obtaining a mixed
longitudinal sample of nestling growth (Ricklefs
1983). We measured the remaining nestlings once
[n = 64). We calculated mean age measurements
for nestling wing length, tarsus length, and mass
from 2 to 10 days of age, and examined variance
across all ages using ANOVA. We used nested
ANOVA in which year of study, study site, and
brood were nested within nestling age. Missing
values for day 8 at Bois-de-la-Roche, and day 10
for both Bois-de-la-Roche and Atocas were
replaced with mean values for those dates. We
calculated the percent of adult size and mass that
was reached each day of nestling growth.
Individual nestling measurements were divided
by adult measurements, giving a percentage of
adult growth. These were averaged to acquire
mean (± SE) percentage of adult growth for
nestlings of 2-10 days of age.

Male and female Bobolinks are dimoiphic, but
cannot be assigned to gender without genetic
testing as nestlings. Thus, we averaged the
reported adult male and female measurements
from the breeding season to use as general adult
measurements (Martin and Gavin 1995). We
reported all mean values ± SE and used SPSS
16.0 (SPSS Inc., Chicago, IL, USA) and JMP
8.0.1 (SAS Institute Inc., Cary, NC, USA) for all
analyses.

RESULTS

Hatching Phenology and Egg Measurements. —
Three of 25 nests for which nestling measure-
ments were obtained hatched synchronously
(12%); 20 nests had one marginal nestling
(80%), and two nests had two marginal nestlings
(8%). Nestlings in three of four re-nests were each
I day apart in age.

Measurements were taken for 175 eggs from 37
nests. Sixteen of the 53 nests monitored were
excluded from egg calculations as they were
found either during the nestling stage or depre-
dated prior to measurements being taken. Mean
egg mass over both years and all sites (/; = 175)
was 2.69 ± 0.05 g (1.6-3. 5 g). We found no

434

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 3, September 2010

FIG. 1. Linear regression with 95% Cl of smallest egg mass in Bobolink clutch compared to initiation date at Atocas
Bay, Ontario, Bois-de-la-Roche, Quebec, and Hemmingford, Quebec during 2006 and 2007.

differences between Bobolink egg mass between
years of study (7175 = —1.87, P = 0.062) and
among study sites {Fx■^^ = 1.45, P = 0.24).

We found a between-clutch effect of egg mass
across Bobolink females (F37 = 13.089, P <
0.001). The model sum of squares accounted for
between-clutch variation only (12.773) and the
total sum of squares accounted for both between-
and within-clutch variation (16.824). Thus, be-
tween-clutch variation accounted for 15.9% of the
total variation.

We found no differences in egg mass between
clutch sizes (F37 = 0.830, P = 0.487) and mean egg
ma.ss was not related with initiation date (r = 0.088;
F37 = 1 .89, P = 0. 10). Egg mass of the smallest egg
in the clutch increa,sed with initiation date (r = 0.13;
/r3^ = 4.390, P = 0.045, [i = 0.021 ± 0.01 SE;
Fig. 1 ). Within-clutch egg ma.ss variation was not
related to clutch size (F37 = 0.246, P = 0.863), but
decrea,sed with greater date of nest initiation (r =
0.277; F37 = 1 1 .5(X), P = 0.002; Fig. 2).

Nestling Measurements and Growth. — We re-
port mean ± SE nestling measurements for
nestlings at ages 2-10 days (Table 1). Bobolink
nestlings had ~ 5-10% larger tarsi in 2007 than in
2006 (F268 = ^-87 1, P < 0.001) and ~ 10-15%
more mass in 2007 than in 2006 (F268 ~ 3.620, P
< 0.001). Wing length did not differ between
years (F268 = 1-382, P - 0.204). Year of study

had an impact on two of the three variables tested;
we tested among study sites within each year to
prevent possible year effects. None of the
variables for 2006 differed among study sites:
nestling wing length (F268 = 1.221, F = 0.279),
tarsus length (F268 = 0.518, F = 0.898), and mass
(^268 0.530, F = 0.890). However, all variables

differed among sites in 2007: nestling wing length
(^268 “ 1.859, F = 0.033), tarsus length (F268 =
1.838 F = 0.036), and mass (F268 = 2.330, F =
0.005). Among site-differences were most evident
between younger nestlings with Atocas > Bois-
de-la-Roche > Hemmingford nestlings; there
were no differences among sites at nestling age
8-10 days. All variables in all years differed
between individual broods.

Nestling growth as a percentage of adult size
and mass increased predictably across nestling
age (Fig. 3). Nestlings for all variables were
below adult size or mass at day 10, just prior to or
at Hedging. Nestling measurements as percentage
of adult ± SE at day 10 were: wing length = 55.1
± 0.4%, tarsus length = 87.7 ± 2.3%, and mass
= 67.6 ± 1 .5%.

DISCUSSION

Asynchronous Hatching. — Hatching asynchro-
ny was prevalent in the Bobolink population we
studied. Most nests (80%) had a single late-

Frei cl al. • BOBOLINK EGG MASS AND NESTLING GROWTH

435

FIG. 2. Linear regression showing within-ckitch deviation of Bobolink egg mass compared to initiation date at Atocas
Bay, Ontario, Bois-de-la-Roche, Quebec, and Hemmingford, Quebec during 2006 and 2007.

hatched nestling, resulting from incubation begin-
ning with the penultimate egg, as previously
described for Bobolinks (Martin 1974). Few nests
had two late-hatched nestlings (8%), far fewer
than in the closely related Red-winged Blackbird
(Agelaius phoeniceus) where 8.1% of nests
hatched synchronously, 41.9% had one late-
hatched nestling, and 43.3% had two late-hatched
nestlings (Forbes and Glassey 2000). Three of
four re-nests in our study had nestlings staggered
in age, a result of incubation beginning with the
first egg laid. This, paired with smaller clutch
size, suggests females shorten the re-nesting cycle
as Bobolinks rarely re-nest (Martin and Gavin
1995).

Egg Mass Variation Within- and Between
Clutches. — Between-clutch variation of egg mass
far exceeded within-clutch variation. This pattern
is common for most bird species with an average
of 70% of variation in egg size due to variation
between- rather than from within-clutches (Chris-
tians 2002). Egg size appears to reflect differences
among individual females, and often increases
with age and experience (Christians 2002).
Overall, consistency of egg and clutch size is a
consequence of selective pressures and energy
expenditure limitations (Lack 1947).

Egg Mass Variation with Clutch Size and Nest
Initiation Date. — Our data did not support Lack’s
(1967) trade-off hypothesis that greater energy

TABLE 1. Mean (± SE) wing, tarsus, and mass measurements for Bobolink nestlings (/;
Ontario, Bois-de-la-Roche, Quebec, and Hemmingford. Quebec during 2006 and 2007.

= 268) at Atocas Bay,

Nestling age
(days)

n

Wing length
(cm)

Tarsus length
(cm)

Mas,s

(g)

2

6

1 .02 ± 0.03

0.95 ± 0.03

6.5 ± 0.6

3

24

1 .28 ± 0.04

1.08 ± 0.04

7.4 ± 0.3

4

47

1.78 ± 0.04

1.31 ± 0.04

10.0 ± 0.3

5

39

2.38 ± 0.05

1.67 ± 0.04

14.2 ± 0.3

6

39

3.00 ± 0.05

1.86 ± 0.05

16.4 ± 0.3

7

33

3.34 ± 0.08

2.07 ± 0.09

17.3 ± 0.8

8

46

4.22 ± 0.06

2.22 ± 0.03

19.9 ± 0.9

9

27

4.72 ± 0.04

2.33 ± 0.04

21.5 ± 0.4

10

7

5.10 ± 0.04

2.25 ± 0.06

21.6 ± 0.5

436

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 3. September 2010

FIG. 3. Nestling growth represented as a percent of breeding adult Bobolink measurements ± SE for wing length,
tarsus length, and mass for Bobolink nestlings pooled for sites and years.

expenditure during laying results in larger clutch-
es of smaller eggs. This hypothesis has been
empirically supported in several studies (Black-
burn 1991, Olsen et al. 1994), but others have
failed to find a negative relationship between
clutch size and egg size (Rohwer and Eisenhauer
1989). The opposite has been demonstrated to be
true, in some cases, where smaller clutches
contain smaller eggs (Galbraith 1988).

Female Bobolinks appear to have laid larger
eggs more consistently as the nesting season
progressed, as nest initiation date was positively
correlated with egg mass of the smallest egg and
negatively conelated with within-clutch variation.
In temperate regions, such as eastern Canada,
ambient temperature increases throughout the
breeding season resulting in decreased daily
energy expenditure for homeotherms and in-
creased food availability for insectivores (Avery
and Krebs 1984, Eeva et al. 2000). In turn, this
relaxes energy expenditure constraints related to
egg-laying activities and may result in larger eggs
later in the season. Mean egg mass of Common
Blackbirds [Tunhis menila) was correlated with
air temperature during the period when the species
was undergoing rapid follicular growth (Magrath
1992), and Great Tits (Pams major) roosting in
heated nest boxes laid larger eggs than those in
cooled nest boxes (Nager and Zandt 1994).

Few studies have investigated within-clutch
.seasonal variation of egg size. Takagi (2003) in

northern Japan during a study of Bull-headed
Shrikes (Lanins bucephalus) found a seasonal
increase in intra-clutch variation of egg volume
due to a sea.sonal size decline in the smallest egg
within the clutch. The inverse was true in the
Bobolinks we studied; within-clutch variation
decreased as the mass of the smallest egg
increased over the season. Bobolinks in our study
area began laying from mid- to late May. It is
often cool and rainy during this period resulting in
high energetic costs for laying females, leading to
smaller smallest eggs and higher within-clutch
variation earlier in the breeding season. Common
Swifts (Apus apus) similarly lay lighter eggs
during unfavorable weather (O’Conner 1979).

Nestling Growth. — Bobolink young at fledging
were well below adult size and mass. Thus, the
Bobolink growth curve is a variant of the
‘standard’ growth curve in which growth levels
off below adult weight (Ricklefs i968a). Bobolink
young are often described as poor flyers for
several days post-lledging (Martin and Gavin
1995), a result of leaving the nest prior to full
feather emergence. Eastern Meadowlarks (Stnr-
nella magna) (Lanyon 1995) and Savannah
Sparrows (Passercnlns sandwichensis) (Wheel-
wright and Rising 2008) are also ground-nesting
grassland birds that fledge at below adult mass.
This may be adaptive for ground-nesting birds,
whose young are especially vulnerable to preda-
tors and disturbance. Nestlings of species with

Frei el al. • liOBOLlNK EGG MASS AND NESTLING GROWTH

437

well-protected nests generally remain in nests
longer than young in exposed nests (Ricklefs
1968a).

Growth ol wing, tarsus, and mass was greatest
between nestling days 3 and 6. The inflection
point occurred at about days 8-9, when growth
leveled off below adult size and mass. Studies in
New York and Wisconsin indicated the Bobolink
nestling mass gain inflection point occuiTed 8 days
after hatching (Martin and Gavin 1995). Tarsus
length was the closest to adult size at fledging of
the three variables measured. This accelerated
growth of a specific nestling characteristic may be
a reflection of the selective pressure to enable
young to run from the nest by day 7 if disturbed
(Martin and Gavin 1995). Eastern Meadowlarks
and Savannah Sparrows also attain near-adult
tarsus length at fledging and are known to escape
from nests by running prior to fledging (Lanyon
1995, Wheelwright and Rising 2008). Incomplete
growth prior to fledging is associated with
provisioning by parents after the young leave the
nest, when they complete their growth to adult
size (Martin and Gavin 1995).

This is the first report of linear nestling
measurements, as well as the first comparison of
egg size and within-clutch variation with clutch
size and initiation date for Bobolinks. This study
provides baseline data for future studies, as well
as information on timing of critical growth for
nestlings, which should prove useful for conser-
vation and management planning for this declin-
ing grassland species.

ACKNOWLEDGMENTS

Field assistance for this project was ably provided by
Carine Lecoeur. Kate Robinson, Rachel Theoret-Gosselin,
and Rachel Verkade. Funding for this project was provided
by the Natural Sciences and Engineering Research Council
(NSERC), Fondation de la faune du Quebec (FFQ),
Canadian Wildlife Service (CWS). Bird Protection Quebec
(BPQ), and McGill University. We thank Marie-Anne
Hudson, Shawn Craik, and two anonymous reviewers who
provided thoughtful criticisms to the manuscript. We thank
Andrew Stairs, the City of Montreal, and Ducks Unlimited
for use of their hayfields for the study. We are grateful to
the McGill Bird Qbservatory (MBO) for training in banding
and equipment use, and the Avian Science and Conserva-
tion Centre (ASCC) for support.

LITERATURE CITED

Avery, M. 1. and J. R. Krebs. 1984. Temperature and
foraging success of Great Tits Pams major hunting for
spiders. Ibis 126:33-38.

Blackburn, T. M. 1991. An interspecific relationship

between egg size and clutch size in birds. Auk
108:973-977.

Bollinger, E. K. and T. A. Gavin. 1992. Easiern
Bobolink populations: ecology and conservation in
an agricultural landscape. Pages 497-506 in Ecology
and conservation of neotropical migrants (J. M. Hagan
and D. W. Johnston, Editors). Smith.sonian Institute
Press, Washington, D.C., USA.

Bollinger, E. K., P. B. Bollinger, and T, A. Gavin.
1990. Effects of hay-cropping on eastern populations
of the Bobolink. Wildlife Society Bulletin 18:142-
150.

Christians, J. K. 2002. Avian egg size: variation within
species and inflexibility within individuals. Biological
Reviews 77:1-26.

Eeva, T., S. Veistola, and E. Lehikoinen. 2000. Timing
of breeding in subarctic passerines in relation to food
availability. Canadian Journal of Zoology 78:67-78.

Forbes, S. and B. Glassey. 2000. Asymmetric sibling
rivalry and nesting growth in Red-winged Blackbirds
{Agelaius phoeniceus). Behavioural Ecology and
Sociobiology 48:413^17.

Galbraith, H. 1988. Effects of egg size and composition
on the size, quality and survival of Lapwing Vanellus
vanellus chicks. Journal of Zoology 214:383-398.

Herkert, j. R. 1997. Bobolink (Dolichonyx oryzivoms)
population decline in agricultural landscapes in the
midwestem USA. Biological Conservation 80:107-
112.

Lack, D. 1947. The significance of clutch size. Ibis
89:302-352.

Lack, D. 1967. The significance of clutch size in
waterfowl. Wildfowl 9:67-69.

Lanyon, W. E. 1995. Eastern Meadowlark (Sturnella
magna). The birds of North America. Number 160.

Magrath, R. D. 1992. Seasonal changes in egg-mass
within and among clutches of birds: general explana-
tions and a field study of the Blackbird Tiirdns menda.
Ibis 134:171-179.

Martin, S. G. 1974. Adaptations for polygynous breeding
in the Bobolink {Dolichonyx oiyzivorons). American
Zoologist 14:109-119.

Martin, S. G. and T. A. Gavin. 1995. Bobolink
(Dolichonyx oiyzivorons). The birds of North Amer-
ica. Number 176.

Nager. R. G. and H. S. Zandt. 1994. Variation in egg size
in Great Tits. Ardea 82:315-328.

Q’Conner. R. j. 1979. Egg weights and brood reduction in
the European Swift (Apns apns). Condor 81:133-145.

Olsen, P. D., R. B. Cunningham, and C. F. Donnelly.
1994. Is there a trade-off between egg size and clutch
size in altricial and precocial nonpasserines? A test of
a model of the relationship between egg and clutch
size. Australian Journal of Zoology 42:323-328.

Ricklefs, R. E. 1968a. Patterns of growth in birds. Ibis
1 10:419^51.

Ricklefs, R. E. 1968b. Weight recession in nestling birds.
Auk 84:30-35.

Ricklefs, R. E. 1983. Avian postnatal development. Pages
1-83 in Avian biology (D. S. Earner and J. R. King.
Editors). Academic Press. New York, USA.

438

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3, September 2010

Rohwer, F. C. and D. I. Eisenhauer. 1989. Egg mass and
clutch size relationships in geese, eiders, and swans.
Ornis Scandinavica 20:43^8.

Takagi, M. 2003. Seasonal change in egg-volume variation
within clutch in the Bull-headed Shrike, Lanius
bucephalus. Canadian Journal of Zoology 81:287-
293.

Wheelwright, N. T. and J. D. Rising. 2008. Savannah

Sparrow {Passerculus sandwichensis). The birds of
North America. Number 45.

Winter, M., D. H. Johnson, J. A. Shaffer, and W. D.
SvEDARSKY. 2004. Nesting biology of three grassland
passerines in the northern tallgrass prairie. Wilson
Bulletin 116:211-223.

Zar, j. H. 1999. Biostatistical analysis. Third Edition.
Prentice Hall, Upper Saddle River, New Jersey, USA.

The IVilson Journal of Orni/liology 1 22(3):439-446, 2010

BREEDING PHENOLOGY AND NESTING SUCCESS OF THE
YUCATAN WREN IN THE YUCATAN PENINSULA, MEXICO

JESUS VARGAS-SORIANO,'^ JAVIER SALGADO ORTIZ,- AND
GRISELDA ESCALONA SEGURA^

ABSTRACT. — The Yucatan Wren (Campylorhynchus yucatanicus) is a highly restricted endemic species inhabiting the
coastal scrub at the northern portion of the Yucatan Peninsula. We describe the breeding phenology and nesting success of
this endangered species from April to September 2007 for a population at Ria Celestun Biosphere Reserve. We found 232
nests of which only 1 10 (47%) were active at either incubation or nestling stages. Yucatan wrens initiated nest building in
late April, but clutch initiation occurred in early June and lasted until the end of July, resulting in a breeding season of
2 months. Nests were exclusively in coastal scrub and the transition between coastal scrub and black mangrove (Avicennia
germinans) forest. Eleven species of trees were used as nesting substrate, but three included 75% of all nests found. Clutch
size (x ± SD) was 3 ± 1.5 eggs with incubation and nestling periods averaging 16 ± 1.0 days, and 16.5 ± 1.9 days,
respectively. Mayfield estimates of daily survival rate for incubation and nestling periods were 0.968 ± 0.005 and 0.975 ±
0.005, respectively with nesting success of 46%. The average number of fledglings per successful nest was 2.5 ± 1.3.
Predation was the main cause of nest mortality accounting for 54% of the active nests. Parental care was provided by both
parents, but participation of a third individual feeding nestlings was recorded at three nests, providing evidence for
occasional cooperative breeding. Increasing human development in the coastal region of the Yucatan Peninsula may
represent a serious threat to conservation of the Yucatan Wren due to habitat restriction and high dependency on three
species of trees as nesting substrate. Received 3 December 2008. Accepted 4 Fehruar}’ 2010.

Information on life history traits and social
systems is scarce or lacking for many species of
birds, especially those at tropical latitudes (Martin
1996, 2004). Estimation of productivity is one of
the most important demographic parameters
necessary to describe numerical fluctuations in
space and time (Donovan et al. 1995, Jehle et al.
2004, Powell and Knutson 2006). Three approach-
es have been suggested to measure bird produc-
tivity: (1) percentage of nests constructed by
females per year (Ricklefs and Bloom 1977,
Anders and Marshall 2005), (2) nesting success
(based on number of nestlings from individual
broods that survive until the end of the reproduc-
tive period as fledglings), and (3) estimation of
reproductive success (based on number of indi-
viduals that survive and reproduce in the follow-
ing reproductive cycle). Estimation of reproduc-

' Laboratorio de Vida Silvestre y Colecciones Cientificas,
Centro de Estudios de Desarrollo Sustentable y Aprove-
chamiento de la Vida Silvestre, Universidad Autonoma de
Campeche, Agustin Melgar s/n entre Juan de la Barrera y
Calle 20, Colonia Buenavista, 24039, San Francisco de
Campeche, Campeche, Mexico.

^Laboratorio de Ornitologi'a, Editlcio B4, Facultad de
Biologia, Universidad Michoacana de San Nicolas de
Hidalgo, Ciudad Universitaria, Morelia, Michoacan, Mex-
ico.

^EI Colegio de la Frontera Sur Unidad Campeche, Calle
10 No. 264 Colonia Centro, 24000, San Francisco de
Campeche, Campeche. Mexico.

"Corresponding author; e-mail: abucefalo@hotmail.com

live success provides information on population
dynamics, and the probability of species extinc-
tion or survival. Reproductive success and
proportion of nests constructed by females have
been difficult to estimate due to logistical
limitations of following females during one or
more reproductive seasons, or as a result of not
knowing the final fate of individuals in time and
space, particularly once they become independent
of their parents and disperse to new areas
(Ricklefs and Bloom 1977). However, nests are
relatively easy to monitor and provide robust data
on productivity. Thus, nesting success has been
frequently used to evaluate population changes in
bird communities (Ricklefs and Bloom 1977.
Johnson and Shaffer 1990, Knadle et al. 2001,
Jehle et al. 2004, Anders and Marshall 2005.
Powell and Knutson 2006).

Researchers can monitor nests to estimate
nesting success and factors that influence repro-
ductive success including vegetation structure,
depredation, nest parasitism, and climatic condi-
tions (Hazier 2004, Martin et al. 2006). This
information is essential in understanding the
ecology of species, which can suggest strategies
for habitat conservation and management (Peter-
son et al. 1993).

The Yucatan Wren {Campylorhynchus yucata-
nicus) occurs exclusively in the coastal portion of
the Yucatan Peninsula in the states of Campeche
and Yucatan, where it inhabits the thin line of

439

440

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 3. September 2010

FIG. I. Study Area showing transects at Ria Celestun Biosphere Reserve, Yucatan, Mexico.

coastal scrub (Peterson and Chalif 1989, Howell
and Webb 1995, Brewer 2001). It is endemic to
the Yucatan Peninsula and considered by the
Mexican Government as threatened and is includ-
ed on the lUCN red list as near threatened (Diario
Oficial de la Federacion 2002, BirdLife Interna-
tional 2008).

Information on the biology of this species has
been limited to general descriptions of distribution
and habitat use, and is largely anecdotal based on
observations by Zimmerman (1957), who provid-
ed a brief description of the nest and behavior of
individuals from a population at Sisal, Yucatan.
Our objectives were to obtain information on (1)
breeding seasonality, (2) length of incubation and
brooding periods, (3) clutch size, (4) parental care,
and (5) nesting success.

METHODS

Study Arec/.— The study was conducted during
April to September 2007 at Ria Celestun Bio-
sphere Re.serve at the northwestern portion of the
Yucatan Peninsula, Mexico (20° 46' and 21° 06'
N, 90° I r and 90° 25' W) on the boundaries of

the states of Campeche and Yucatan (Fig. I). The
climate of the area is warm semi-dry with rain
occurring mainly during summer. Annual precip-
itation averages 750 mm with a dry season
occurring from December to May and a wet
season from June to November. The mean annual
temperature is 26.5° C with May being the
warmest month (average 29° C) (CONANP 2000).

The most abundant types of vegetation at Ria
Celestun Biosphere Reserve are coastal scrub, and
mangrove and flooded lowland forests (CONANP
2000). Coastal scrub has two phases of develop-
ment with the first adjacent to the beach and
integrated by clumps of shrubs where the main
species are bay cedar (Suriana inaritima) and sea
rosemary {Toumefortia gnapluilodes). The second
corresponds to scrub towards the interior of the
dunes in a zone where the sand is less altered and
canopy height ranges from 3 to 5 m. The most
common plant species in this zone are white tree
(Bnivaisia herlonderiana), wood luck seed (Tlie-
vetia gaumeri), sea grape (Coccoloha uvifera),
geiger tree (Cordia sehestena), black poison wood
(Metopium brownie), peccary tree {Sideroxylon

Wniras-Soriano cl al. • YUCATAN WREN NESTING SUCCESS

441

americiiniaii), joewood {Jcicquinia aiirantiaca),
peccary wood (Caesalpinia vesicaria), blackhead
(Pithecellobiiun keyeiise), wild sage {Liintcuui
involiicrata), blacktorcli {Erithalis friiticosa),
mexican cotton {Gossypiitm hirsiitiim), and cen-
tury plant {A^cive angustifolia).

Mangrove forest is abundant along the estuary
and is represented by red (Rhizophoni mangle),
black {Avicennia germinans), and white mangrove
{Laguncitlaria racemosa) with tree heights rang-
ing from 12 to 14 m. Other species such as
pickleweed (Baris maritima) and sea purslane
(Sesiivinni portiilacastnun) together with numer-
ous species of grasses (Cyperaceae and Grami-
neae) also occur along drainage channels forming
patches, characterized by short trees ranging from
1.5 to 3 m in height, known as logwood patches.

The flooded lowland forest merges between the
coastal scrub and mangrove forest and is repre-
sented by logwood tree (Haematoxyhim campe-
chianum), milk tree (Cameraria latifolia), black
poison wood, chicle tree (Manilkara zapota),
gambo limbo (Biirsera simaruba), nance (Byrso-
nima crassifolia), black olive tree (Bucida bu-
ceras), ceiba tree (Ceiba aesculifolia), buttercup
tree (Cochlospermum vitifolium), and button
mangrove (Conocarpus erectus).

Nest Searching and Monitoring. — We estab-
lished line transects ranging from 4 to 6 km in
length in each of the three main vegetation types
within the study area. We visited each transect
starting in April 2007 to locate wrens and nesting
sites based on presence of old nests. Other wren
species are known to maintain year-round nests,
apparently used as dorms (Robinson et al. 2000,
Yaber and Rabenold 2002). Yucatan Wrens are no
exception, as old nests were found along each
transect; old nests were used as evidence of wren
presence and potential nesting sites. Each site was
checked for nesting activity every third day with
observations throughout morning (0630-1200 hrs
EDT) and afternoon (1600-1930 hrs), following
Martin and Geupel (1993). We found new nests,
primarily by scanning trees along transects
(Yucatan Wrens make nests of grasses and the
yellow material contrasts with the dark back-
ground of trees used as nesting substrate). We also
used parental behavioral cues including observa-
tion of birds carrying nesting material, alarm calls
of wrens during both incubation and brooding,
and parents delivering food to nestlings (Ralph et
al. 1996). Few nests were found by accidentally
flushing adults from the nest while walking. We

recorded the number of nests found in each
vegetation type and identified the species of trees
used as nesting substrate.

We checked nests every 2-4 days until failure
or success to estimate clutch initiation, length of
incubation and brooding periods, and final nest
fate. We monitored nests daily when a transition,
hatching or fledging, was anticipated. Yucatan
Wren nests are dome shaped and bulky with a
tubular entrance; we used a fiberscope (Provision
model 960, Chicago Miniature Inc., Hackensack,
NJ, USA.) to examine nest contents and avoid
damage to nests and contents. We considered a
nest failed if all eggs or nestlings disappeared and
successful if one or more young fledged.

We used three video cameras (SONY DCR-
HC30/HC40) to record parental care and to
document cooperative breeding as this behavior
has been reported for other species of wrens
(Rabenold 1984), Cameras were used at 45 of 1 10
active nests for a total of 200 hrs of video
recording. Video cameras were used only at nests
with nestlings. Each video was later reviewed
with the aid of a TV monitor to calculate parental
visitation rate, feeding, and presence of individ-
uals other than the parental pair.

Statistical Analyses. — We report means ± SD
for all variables unless otherwise specified.
Nesting success and daily nest survival probabil-
ities were calculated following Mayfield (1961,
1975).

RESULTS

Timing of Breeding. — We found 232 nests
throughout the 2007 breeding season of which
110 (47%) were active: laying (59), incubation
(40), and nestling (11). The remaining 122 (53%)
nests were abandoned during the building stage.
Yucatan Wrens started nest building in early
April, but most nests were built in June (50%)
with fewer in May (35%), July (13.5%), and April
(1.3%). Nests were build by both parents and the
average time to build a nest was 6 ± 1.5 days
(range 4-8, n = 149). Clutch initiation started in
June despite early nest building. The earliest
clutch recorded was on 1 1 June and the latest on
28 July; the breeding season extended for
2 months (Jun-Jul). There was a time lag of 19
± 8.5 days (range 5-33, n = 24) between the
end of nest building and laying of the first eggs.
Old nests within temtories were not used for
breeding.

442

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. 3. September 2010

Cocos nucifera

Maytenus phyllanthoides

Coccoloba uvifera

o Jacquinia aurantiaca

o

Avicennia gerrninans
^ Caesalpinia vesicaria

Sunana mantima
Sideroxylon celastrina
Pithecellobium keyense
Sideroxylon americanum
Conocarpus erectus

0 10 20 30 40 50 60 70 80

Number of nests

FIG. 2. Tree species used by Yucatan Wrens as nesting substrate at Ria Celestun Biosphere Reserve, Yucatan. Mexico.

Clutch Size, Length of Incubation and
Nestling Periods. — Clutch size averaged 3 ± 1.5
eggs (range 1-5; n = 99 nests), but half of the
nests (49.5%) had four eggs. The proportion of
nests with one egg was 10.1%, two eggs =
18.2%, three eggs = 21.2%, and five eggs = 1%.
Yucatan Wrens typically laid one egg per day and
females initiated incubation with laying of the last
egg. The incubation period averaged 16 ±
1.0 days (range 15-17, n ~ 47). The earliest
hatching date was 28 June and the latest was 1 1
August. Hatching was synchronous as all eggs
within individual broods hatched within 24 hrs
after the first egg hatched.

Yucatan Wren nestlings fledged on average
16.5 ± 1.9 days after hatching (range 14-19; n =
5 1 ). The average number of young fledged per
successful nest was 2.5 ± 1.3 (1^ young; n —
51 ). The earliest date of fledging was 14 July and
the latest was 23 August.

Parental Care. — Incubation feeding by Yuca-
tan Wrens was not observed. Both parents fed
nestlings at all nests observed (/? = 51) for
parental care. The average number of provisioning
visits/hr by both parents combined was 7 ± 1.6
(range 5-9 visits). We recorded the presence of a
third adult feeding young at three nests, confirm-
ing occasional cooperative breeding in this
species. Fledglings remained clo.se to the nesting
site for 2-3 days after fledging and both parents
continued providing care to fledglings, indepen-
dently of number.

Nesting 5//cxr.v,v.— The Mayfield estimate of
daily survival for the incubation period was 0.968
± ().()()5 and 0.975 ± 0.005 for the nestling

period. Estimated nesting success was 46%. Nest
predation was the only cause of nesting failure,
accounting for loss of 54% of all active nests. We
found no evidence of brood parasitism by
cowbirds of Yucatan Wren nests. Ectoparasites
(Diptera: Muscidae) were observed on 20 juve-
niles in seven nests, but no mortality was related
to parasitism.

Habitat Use and Nesting Substrate. — One
hundred and fifty of the 232 nests (65%) were
in coastal scrub, while the remaining 82 (35%)
were in flooded lowland forest (between the
coastal scrub and mangrove forest). Yucatan
Wrens used 1 1 species of trees as nesting
substrate (Eig. 2). However, only three species,
button mangrove, peccary tree, and blackhead
supported 75% of all nests found (Eig. 2).

DISCUSSION

The breeding season of Yucatan Wrens at
Celestun was highly seasonal, occuiring over a
period of ~ 3 months (May-Jul). Nest building
started during the dry season in late April, but
clutch initiation was not recorded until June
suggesting initiation of breeding is closely related
to seasonal occurrence of rainfall. Previous
studies on avian breeding phenology have sug-
gested that onset of breeding of tropical birds is
closely connected with rainy seasons' (Ahumada
2001, Tarraoux and McNeil 2003). It is possible
that rainfall may also be a primary stimulus for
reproductive activity for Yucatan Wrens. We did
not measure the rainfall regime for the study
period, but Salgado-Ortiz et al. (2008) reported
that 75% of the rain occurs from June to

Vurgas-Soriano cl at. • YUCA I AN WREN NESTING SUCCESS

443

September in Celestun with a monthly maximum
in September. Our observations of a time lag that
averaged 20 days from the end of nest building to
laying of the first egg in June supports the
possibility that Yucatan Wrens may be using rain
as an environmental cue to adjust timing of
breeding.

Clutch initiation of Yucatan Wrens extended
over 2 months (Jun-Jul) with no evidence of
second nesting attempts after either failure or
success. The clutch initiation period is similar to
that recorded for Cactus Wrens {Campylo-
rhynchiis bninneicapilhis) in Arizona (Simons
and Martin 1990), but shorter than that of
Stripe-backed Wrens (C. nuchalis) in Venezuela
(Piper 1994). The difference in duration
(2 months) of the breeding season between
Yucatan and Striped-backed wrens may be
explained by seasonal effects occurring in the
flooded savanna where the latter species occurs.
The savanna remains relatively wet weeks after
arrival of the first rains, presumably resulting in
food being available for a longer period, which in
turn ensures the possibility of more nesting
attempts.

The average clutch size of Yucatan Wrens was
three eggs, similar to that of Rufous-naped Wrens
(C. rufinucha) in Costa Rica (Skutch 1985), and
slightly smaller than that of Cactus Wrens (4
eggs) in Arizona (Simons and Simons 1990). The
smaller clutch size of the Yucatan Wren appears
to fit the latitudinal trend in clutch size from
temperate to tropical latitudes. However, the
evidence is inconclusive because ~ 50% of
Yucatan Wren nests had four eggs suggesting
differences in clutch size may not be significant,
as has been found for open cup nesting species
(Martin 1996).

The incubation period averaged 16 days, sim-
ilar to that of the Cactus Wren; however,
incubation started with the laying of the last egg
by Yucatan Wrens in contrast to the first egg for
Cactus Wrens (Farley and Stuart 1994). It is
possible Yucatan Wrens begin incubation with the
last egg to synchronize hatching. We observed
that hatching of all eggs occurred within 24 hrs of
hatching of the first egg, but were unable to obtain
detailed description of hatching synchrony. The
difference in initiation of incubation versus that of
Cactus Wrens could be attributed to the large
change of temperatures throughout the day
occurring in the arid desert where Cactus Wrens
occur (Mezquida 2004). Cactus Wrens starts

breeding in March (Simons and Martin 1990),
coinciding with frequent cold fronts from the
north resulting in extreme weather conditions.
Initiation of incubation with laying of the first egg
could be a local adaptation to protect eggs from
unstable weather conditions (Hansell 2000).

Parental care, as confirmed by video cameras,
was provided by both parents and an occasional
adult helper at three nests. We were not able to
capture and color-band individuals, but videotap-
ing provided evidence of individuals in addition to
the parents visiting active nests and providing
food to nestlings. We suspect that cooperative
breeding its more widespread in Yucatan Wrens,
but confirmation requires study with color-banded
individuals. Austad and Rabenold (1986) and
Rabenold (1990) reported cooperative breeding
has been documented in nine of the 13 species of
Campy lorhyncluis, not including the Yucatan
Wren. Our evidence is scarce, but the highly
restricted coastal arid scrub where the Yucatan
Wren occurs and dramatic seasonal changes have
been suggested as environmental or causative
factors for occurrence of cooperative breeding in
other species (Rabenold 1990).

Predation was the main cause of failure of
Yucatan Wren nests. Stutchbury and Morton
(2001) report nest depredation rates for tropical
birds as a group are higher (65-80%) compared to
those of temperate zone birds. The nest loss (54%)
for Yucatan Wrens at Celestun is less than
reported as typical for tropical birds. Nest
depredation has been reported to generally be
higher for open cup nests that for closed nests
(Collias and Collias 1984, Martin 1995, Auer et
al. 2007). Lower nest depredation of Yucatan
Wrens could be a result of the dome-shaped nest.
Clumping of nests has been reported to reduce the
probability of depredation, increasing reproduc-
tive success (Bowman and Hams 1980, Martin
1993). Yucatan Wrens built .several nests in the
same tree; we did not study the direct effect of
clumped nests on nesting success, but this
behavior could explain the lower depredation rate
of Yucatan Wren nests compared to Mangrove
Warblers {Dendroica petechia Ivyanti) nesting in
the same area (67-88%) (Salgado-Ortiz et al.
2008). The Mayfield estimate of daily survival
rate of Yucatan Wren nests (0.978) was higher
than reported for Mangrove Warblers (0.950) in
the same area (Salgado-Ortiz et al. 2008), but was
similar to that reported for Cactus Wrens (0.99)
(Ricklefs 1968). Daily survival rates of Yucatan

444

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3. September 2010

and Cactus wrens provide evidence to support the
hypothesis that dome-shaped nests survive better
than open cup nests (Martin 1996).

Construction of several nests in clumps appears
to be widespread behavior among the Troglody-
tidae (Marquez-Valdelamar 1998). Nests built, but
not intended for breeding are used as “dummy”
nests, a strategy suggested to reduce depredation
and increase chances of adult survival (Marquez-
Valdelamar 1998, Robinson et al. 2000, Puga-
Vazquez 2008). Robinson et al. (2000) demon-
strated that only half of the nests constructed by
Song Wrens (Cyphorhinus phaeocepholiis) in
Panama were used for reproduction, while the
others were constructed to confuse nest predators.
Yucatan Wrens used 47% of the total new nests
found over the breeding season. Some nests in our
study were destroyed during the building stage by
other individuals of the same species or uniden-
tified predators, while others were finished, but
remained unused for reproduction, supporting the
“dummy” anti-predator strategy.

Yucatan Wrens nested exclusively within
coastal scrub and the transition to mangrove
forest. This vegetation zone extends < 1 km
inland. Yucatan Wrens mostly used three species
of trees with button mangrove being used as the
main nesting substrate. This indicates Yucatan
Wrens are highly dependent on this tree species as
nesting substrate. Previous studies have shown
that other species of Troglodytidae construct nests
mainly in trees with thorns, particularly of the
genus Acacia. It has been hypothesized that use of
thorny plants as nesting substrate is for protection
from predators (Young et al. 1990, Marquez-
Valdelamar 1998, Skutch 2001). This also has
been suggested for Cactus Wren, which uses
cactus (Opuntia spp.) as nesting substrate (Face-
mire et al. 1990). Yucatan Wrens at Celesttin u.sed
species lacking thorns, despite several thorny
species occurring in coastal scrub. We believe the
high use of button mangrove as nesting substrate
reflects the greater abundance of this tree species
within the preferred nesting habitat. Yucatan
Wrens were not locally rare in our study area,
but are highly habitat restricted and show marked
preference in nesting substrate. Button mangrove
is widely irsed as firewood by local people who
are reducing the availability of this species. This
use may have negative implications on future
nesting success of Yucatan Wrens. The current
human population increa.se and increasing land
development resulting from the touri.st industry in

coastal areas are becoming serious threats to
future conservation of Yucatan Wrens and their
habitat.

ACKNOWLEDGMENTS

We are grateful to Victor Nah Chin for help in the field
and Rodolfo Noriega Trejo, R. E. Gongora Chi'n, and
Rodolfo ColK Dfaz for help with plant identification.
DUMAC, A.C., research field station, and personal from
Celestiin Biosphere Reserve provided logistic support. The
study was supported financially by El Colegio de la
Frontera Sur (ECOSUR) and the National Council of
Science (CONACyT Scholarship 225107 and SNl 21467).
The study was conducted under research permit SGPA/
DGVS/03757/06.

LITERATURE CITED

Ahumada, J. a. 2001. Comparison of the reproductive
biology of two neotropical wrens in an unpredictable
environment in northeastern Colombia. Auk 1 18:191-
210.

Anders, A. D. and D. M. R. Marshall. 2005. Increasing
the accuracy of productivity and survival estimates in
assessing landbird population status. Conservation
Biology 19:66-74.

Auer, S. K., R. D. Bassar, J. J. Fontaine, and T. E.
Martin. 2007. Breeding biology of passerines in a
subtropical montane forest in northwestern Argentina.
Condor 109:321-333.

Austad, S. N. and K. N. Rabenold. 1986. Demography
and the evolution of cooperative breeding in the
Bicolored Wren, Cainpylorhynchits griseu.s. Behaviour
97: 308-324.

BirdLife International. 2008. Campy lorhynchus yiica-
taniciis. In lUCN 2009. lUCN Red List of Threatened
Species. Version 2009.2. <www.iucnredlist.org>.
Bowman, G. B. and L. D. Harris. 1980. Effect of spatial
heterogeneity on ground-nest predation. Journal of
Wildlife Management 44:806-813.

Brewer, D. 2001. Wrens, dippers, and thrashers: a guide to
the wrens, dippers, and thrashers of the world. Yale
University Press, New Haven, Connecticut, USA.
CoLLlAS, N. E. and E. C. Collias. 1984. Nest building and
bird behavior. Princeton University Press, Princeton,
New Jersey, USA.

CoMisibN Nacional de Areas Naturales Protegidas
(CONANP). 2000. Programa de Manejo de la Reserva
de la Biosfera Ria Celestiin. Mexico, D.F.

Diario Oficial de la FederaciOn. 2002. Norma Oficial
Mexicana NOM-059-ECOL-200I . Proteccion ambien-
tal-Especies nativas de Mexico de flora y fauna
silvestres-Categon'as de riesgo y espccificaciones para
su inclusion, exclusion o cambio-Lista de especies en
ric.sgo. Sccretan'a de Medio Ambiente y Recursos
Naturales. Segunda Seccion del miercoles 6 de marzo
dc 2002. Mexico, D.F.

Donovan, T. M., F. R. Thompson 111. J. Faaborg, and
J. R. Proust. 1995. Reproductive success of migratoiy

Vargas-Soriano cl al. • YUCATAN WREN NES I ING SUCCESS

445

birds in habitat sources and sinks. Conservation
Biology 9:1380-1395.

Facemire, C. F., M. E. Fac'emire, and M. C. Facemire.
1990. Wind as a factor in the orientation of entrances
of Cactus Wren nests. Condor 92:1073-1075.

Farley, G. and J. Stuart. 1994. Atypical nesting sites of
the Cactus Wren. Texas Journal of Science 46: 193-
195.

FIansell, M. 2000. Bird nest and construction behaviour.
Cambridge University Press, Cambridge, New York,
USA.

Hazler, K. R. 2004. Mayfield logistic regression: a
practical approach for analysis of nest survival. Auk
121:707-716.

Howell, S. and S. Webb. 1995. A field guide to the birds
of Mexico and northern Central America. Oxford
University Press, New York, USA.

Jehle, G., a. a. Yackel Adams, J. A. Savidge, and S. K.
Skagen. 2004. Nest survival estimation: a review of
alternatives to the Mayfield estimator. Condor 106:
472^84.

Johnson, D. H. and T. L. Shaffer. 1990. Estimating nest
success: when Mayfield wins. Auk 107:595-600.

Knadle, G. E., D. L. Brubaker, and K. S. Brubaker.
2001. Nest success is not an adequate comparative
estimate of avian reproduction. Journal of Field
Ornithology 72:527-536.

Marquez-Valdelamar, L. 1998. Monografias de las
especies mexicanas de la familia Troglodytidae (aves).
Thesis. Facultad de Ciencias, Universidad Autonoma
de Mexico. Mexico, D.F.

Martin, T. E. 1993. Nest predation and nest sites: new
perspectives on old patterns. Bioscience 43:523-
532.

Martin, T. E. 1995. Avian life history evolution in relation
to nest sites, nest predation, and food. Ecological
Monographs 65:101-127.

Martin, T. E. 1996. Life history evolution in tropical and
south temperate birds: what do we really know?
Journal of Avian Biology 27:263-272.

Martin, T. E. 2004. Avian life-history evolution has an
eminent past: does it have a bright future? Auk
121:289-301.

Martin, T. E. and G. R. Geupel. 1993. Protocols for nest
monitoring plots: locating nests, monitoring success,
and measuring vegetation. Journal of Field Ornithol-
ogy 64:507-5 19.

Martin, T. E., R. D. Bassar, S. K. Bassar, J. J. Fontaine,
P. Lloyd, H. A. Mathewson, A. M. Niklison, and
A. Chalfoun. 2006. Life history and ecological
correlates of geographic variation in egg and clutch
mass among passerine species. Evolution 60:390-
398.

Mayfield, H. F. 1961. Nesting success calculated from
exposure. Wilson Bulletin 73:255-261.

Mayfield, H. F. 1975. Suggestions for calculating nest
success. Wilson Bulletin 87:456-466.

Mezquida, E. T. 2004. Patrones de orientacion de los nidos
de Passeriformes en una zona arida del Centro-Oeste
de Argentina. Omitologfa Neotropical 15:145-153.

Peterson, A. T., O. A. Flores-Villela, L. S. Leon-

Paniagua, j. E. Llorhnte-Bousquets, M. a. Luis
Martines, a. G. Navarro-Siguenza, M. G. Torres-
ChAvez, and 1. Varga.s-Fernandez. 1993. Con.ser-
vation priorities in Mexico: moving in the world.
Biodiversity Letters 1:33-38.

Peterson, R. T. and E. L. Chalif. 1989. Aves de Mexico,
Guta de Campo. Diana, Mexico D.F.

Piper. W. H. 1994. Courtship, population, nesting behavior,
and brood parasitism in the Venezuelan Stripe-backed
Wren. Condor 96:654-671.

Powell, L. A. and M. G. Knutson. 2006. A productivity
model for parasitized multibrooded song birds. Condor
108:292-300.

Puga-Vazquez, I. L. 2008. Relacion de la anidacion de
Uropsila leucogastra con factores ambientales en
Calakmul, Campeche, Mexico. Thesis. Facultad de
Ciencias Quhnico Biologicas, Universidad Autonoma
de Campeche. San Francisco de Campeche, Cam-
peche, Mexico.

Rabenold, K. N. 1984. Cooperative enhancement of
reproductive success in tropical wren societies.
Ecology 65: 871-885.

Rabenold, K. N. 1990. Helping and communal breeding in
birds. Wilson Bulletin 102: 563-565.

Ralph, C. J., G. R. Geupel, P. Pyle, T. E. Martin, D. F.
Desante, and B. MilA. 1996. Manual de Metodos de
Campo para el Monitoreo de Aves Terrestres. USDA,
Forest Service, General Technical Report PSW-GTR-
159. Southwest Research Station, Albany, California.
USA.

Ricklefs, R. E. 1968. The survival rate of juvenile Cactus
Wrens. Condor 70:388-389.

Ricklefs, R. and G. Bloom. 1977. Components of avian
breeding productivity. Auk 94:86-96.

Robinson, T. R., W. D. Robinson, and E. C. Edwards.
2000. Breeding ecology and nest-site selection of Song
Wrens in central Panama. Auk 1 17:345-354.

Salgado-Ortiz, j., P. P. Marra, T. S. Sillett, and R. J.
Robertson. 2008. Breeding ecology of the Mangrove
Warbler {Dendroica petechia biyanti) and compara-
tive life history of the Yellow Warbler subspecies
complex. Auk 125:402-410.

Simons, L. S. and T. E. Martin. 1990. Food limitation of
avian reproduction: an experiment with the Cactus
Wren. Ecology 71:869-876.

Simons, L, S. and L. H. Simons. 1990. Experimental
studies of nest-destroying behavior by Cactus Wrens.
Condor 92:855-860.

Skutch, a. F. 1985. Clutch size, nesting success, and
predation on nests of neotropical birds, reviewed.
Ornithological Monographs 36:575-594.

Skutch, A. F. 2001. Life history of the Riverside Wren.
Journal of Field Ornithology 72: 1-1 1 .

Stutchbury, B. j. M. and E. S. Morton. 2001.
Behavioral ecology of tropical birds. Academic Press.
London, United Kingdom.

Tarraoux, a. and R. McNeil. 2003. Influence of rain on
the breeding and molting phenology of birds in a thorn
woodland of northeastern Venezuela. Neotropical
Ornithology 14:371-380.

YAber, M. C. and K. N. Rabenold. 2002. Effects of

446

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3. September 2010

sociality on short-distance, female-biased dispersal in
tropical wrens. Journal of Animal Ecology 71:1042-
1055.

Young, B. E., M. Kaspari, and T. E. Martin. 1990.

Species-specific nest selection by birds in ant-acacia
trees. Biotropica 22:310-315.

Zimmerman, D. A. 1957. Some remarks on the behavior of
the Yucatan Cactus Wren. Condor 59: 53-58.

The Wilson Journal of Ornithology 1 22(3):447^54, 2010

BREEDING BIOLOGY AND NATURAL HISTORY OF THE
SLATE-THROATED WHITESTART IN VENEZUELA

ROMAN a. RUGGERA' -^ and THOMAS E. MARTIN'

ABSTRACT. — We provide details on the breeding biology of the Slate-throated Whitestart (Myiobonts miniatus) from
126 nests found during seven breeding seasons, 2002-2008, at Yaeambii National Park, Venezuela. Nesting activity peaked
in late April and May. Only the female built the nest and incubated the eggs. Males rarely visited the nest during these
stages. Mean clutch size (2.1 ± 0.04 eggs, n = 93) was the smallest recorded for the Slate-throated Whitestart. Incubation
and nestling period lengths were 15.3 ± 0.3 1 {n = 21 ) and 10.8 ± 0.24 (n = 7) days, respectively. Attentiveness (% of time
on the nest) during incubation (59 ± 1.6%, n = 52) was similar to other tropical warblers and much lower than northern
relatives. This caused a relatively low egg temperature (34.40 ± 0.33° C, n = 6 nests, 20 days) compared with north
temperate birds. Both parents fed nestlings and increased their provisioning rates with nestling age. Growth rate based on
nestling mass (k = 0.521 ± 0.015) was faster than for other tropical passerines but slower than northern relatives. Predation
was the main cause of nesting failure and rate of predation increased with age of the nest. An estimated 15% of nests were
successful based on an overall Mayfield daily predation rate of 0.053 ± 0.007. This study confirms a strong latitudinal
variation in life history traits of warblers. Received 17 September 2009. Accepted 12 March 2010.

New World wood warblers (Parulidae) are a
small-sized and mostly insectivorous group of
birds. Much work has been done on north
temperate species, but resident tropical warblers
are relatively poorly studied (but see Skutch 1945,
1954; Greeney et al. 2008; Cox and Martin 2009;
Morales-Rozo et al. 2009). One of the most
broadly distributed genera of this family is
Myiobonts, which consists of 12 species (Sibley
and Monroe 1990, Curson et al. 1994, Ridgely and
Tudor 1994). The most widespread species of the
genus is the Slate-throated Whitestart (M. minia-
lus) which is a resident of mountain ranges
between 700 and 2,500 m elevation in humid
and wet forests (Ridgely and Tudor 1994). It
occurs from Mexico to southern Bolivia, and
unconfirmed data extend its distribution to
northern Argentina (Di Giacomo 1995). It also
is vagrant in southern Arizona, New Mexico, and
west Texas (Dunn and Garrett 1997). However,
the most northern breeding record is in the Sierra
Madre Oriental, Mexico (McCormack et al.
2005).

Slate-throated Whitestart populations are not
threatened anywhere within their range and
tolerate nesting in disturbed areas (Collins and
Ryan 1994, Curson et al. 1994, Mumme 2010).

' uses Montana Cooperative Wildlife Re.search Unit.
University of Montana, Missoula, MT 59812, USA.

^Current Address: CONICET-Instituto de Ecologi'a
Regional (lER), Facultad de Cs. Naturales e Instituto M.
Lillo. Universidad Nacional de Tucuman, C.C. 34, 4107
Yerba Buena, Tucuman, Argentina.

^Corresponding author; e-mail: raruggera@yahoo.com.ar

For these reasons, the species is evaluated as one
of Least Concern (lUCN 2009).

The wide distribution of this species, which is
almost equal to that of the entire genus, makes the
Slate-throated Whitestart an interesting species
for examining geographical variation. The Slate-
throated Whitestart has great plumage and genetic
variability along its range (Perez-Eman 2005).
The 12 subspecies recognized show a dine in
belly colors from red to orange and yellow from
North to Central and South America (Curson et al.
1994). Yet, intraspecific and interspecific com-
parisons in life-history traits of Myiohorus are
scarce (Skutch 1945, 1954; Collins and Ryan
1994; Mumme 2010).

M. m. ballux is the most common race in the
Andes of Venezuela, Colombia, and part of
Ecuador. Genetic information for our study site
(Yacambu National Park, Venezuela) is lacking,
but geographical proximity to Cubiro as well as
the occurrence of some slight orange in the breast
suggests M. m. ballux is also the race occurring in
Yacambu (Phelps and Phelps 1950; J. L. Perez-
Eman, pers. comm.).

Our objective is to provide new information on
the breeding biology of the Slate-throated White-
start in a montane forest in northwest Venezuela.
We compare our results with life-history traits
recorded for other warblers, especially other
Myiobonts species.

METHODS

Study Area. — This study was conducted during
seven consecutive breeding seasons, from early
March to early July in 2002 to 2008 at Yacambu

447

448

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3. September 2010

National Park (09° 42' N, 69° 42' W) in Lara
State, Venezuela. This is a montane forest
consisting mainly of primary forest in the northern
part of the Andes. The park ranges from 500 to
2,200 m, but our study plots were restricted to
1,350-2,000 m elevation (Fierro-Calderon and
Martin 2007). The rainy season spans from mid-
April to mid-July with a peak from May to June.

Field Procedures. — Slate-throated Whitestart
nests were found almost solely by observing
parental behavior, especially during nest building
when it is easiest to find them (Martin and Geupel
1993). Nests were checked every other day,
except at stage-changing events (hatching or
fledging), when they were checked every day or
twice per day. Nest size was measured using a
ruler with an accuracy of 0.01 cm for outer
diameter (from edge to edge of nest) and height
(exterior bottom-to-top of nest), and inner diam-
eter (from edge to edge of cup) and height
(bottom-to-top of cup). Eggs and nestlings were
weighed with a portable electronic scale (ACCU-
LAB, Elk Grove, IL, USA) with an accuracy of
±0.001 g. Mean egg mass was calculated only for
eggs weighed between days 0 (last egg laid) and 2
of incubation. We measured the tarsus of nestlings
using digital calipers (Mitutoyo, Kingsport, TN,
USA) with an accuracy of 0.01 mm.

We videotaped nests during incubation and
nestling periods in 6-8 hr sessions, starting within
30 min of sunrise (Martin and Ghalambor 1999,
Fierro-Calderon and Martin 2007). This allowed
us to calculate parental attentiveness (percentage
of time on nest for incubating or brooding), as
well as on- and off-bout duration and parental
provision rates. We sought to videotape incuba-
tion in three periods to examine age-related
changes in behaviors: early (2-3 days after the
last egg was laid), middle (days 6-10) and late (2-
3 days before average hatching date) incubation.
We recorded parental behavior in two stages
during the nestling period: early (day 2-3 after
hatch day) and pin break (day in which the eighth
pin feather broke its sheath). The latter is possible
because pin feather tips turn white the day before
they break. Activity recorded in videotapes as
well as observation of predation events during
nest checks allowed us to identify some of the
predators.

We recorded egg temperatures by inserting
thermisters on day 1 or 2 of the incubation period
into one egg in a nest and the hole was sealed with
glue (Weathers and Sullivan 1989). The wire was

passed through the back of the nest and connected
to a HOBO Stowaway XTI datalogger (Onset
Corp., Bourne, MA, USA) that recorded temper-
atures every 12-24 sec for 5-7 days per nest,
unless the nest was depredated earlier (Martin et
al. 2007).

Statistical Analyses. — All means are reported
with their standard errors (SE). Sample sizes are
number of nests in all cases unless otherwise
noted. Mean initiation date was calculated only
for nests in which the exact egg laying date was
known. Nesting season length was estimated as
the number of days for the middle 90% of nests
initiated (excluding the earliest and latest 5% of
initiations) following Martin (2007). We calcu-
lated the incubation period as the number of days
between the last egg laid and the last egg hatched
(Briskie and Sealy 1990, Martin 2002). Incubation
periods were only calculated for nests in which
the exact day the last egg was laid and exact hatch
day were known. We quantified the nestling
period as the number of days between the last egg
hatching and the last nestling fledging, again
based only on exact observations. We used
analyses of variance (ANOVA) with significance
at a = 0.05 to test changes in parental behavior
during incubation and nestling stages, and used
LSD post-hoc tests among periods. Nestling
growth curves were calculated following Remes
and Martin (2002). We included measurements of
mass and tarsus length for nestlings in which the
exact age was known for these analyses. Mortality
and survival analyses of nests were calculated
considering the elapsed time of the observations
to allow inclusion of nests found in different
stages following Mayfield (1975). Nests that
failed because of experimental activities were
excluded from estimation of daily survival or
mortality.

RESULTS

Nest Description. — The highest nest located
was at an elevation of 1,674 ± 5 m asl; we did not
see or hear any Slate-throated Whitestarts above
this elevation even though we worked to 2,000 m.
Nests were in small nooks on banks or slopes and
often built in open or disturbed areas- including
road banks or human-made trails inside the forest.
However, nests were also found on slopes within
undisturbed forest. Nests were highly cryptic
because they were usually hidden by surrounding
shrubs, ferns, or grasses. Only the female was
observed going to and from the nest (mainly in

Riiggcni and Martin • BREEDING BIOL()C}Y Ol- A TROPICAL WARBLER

449

T3

B

ro

c

B

00

CD

c

o

c/3

CD

JD

E

ID

FIG. 1. Numbers of nests initiated (i.e., first egg laid)
each week beginning 1 March for Slate-throated White-
starts at Yacambii National Park, Venezuela during 2002-
2008, Only nests for which initiation date was observed are
included (/; = 106).

morning hrs) to build the nest. The male was
nearby in the surrounding trees, usually singing
and hunting insects. The first nests of the season
were often built over 2 to 4 weeks in a rhythm that
started with several nest-building visits per day
that slowly increased over time to much more
active building. Nests of later attempts were built
over 3 to 4 days in much more active behavior.

The nest was a dome-shaped construction
entirely built with dry material including pieces
of grass, several types of dry leaves, roots, and
small twigs. The lining consisted of fine and soft
dry fibers. The entrance was circular to elliptical
(wider than higher). Nest orientation at 73 nests
spanned from 1 to 334° with no orientation
preferences. The internal dimensions of the cup
averaged 49.1 ± 4.7 mm in diameter and 31.4 ±
9.5 mm in height (n = 66). The outer dimensions
of the nest averaged 103.9 ±21.2 mm and 64.8 ±
25.0 mm for diameter and height, respectively.

Nesting Chronology. — The nesting period of
Slate-throated Whitestarts in Yacambu spanned
from late March to late June based on 126 nests
over seven breeding seasons. The earliest and
latest nests across all years of study were initiated
on 26 March and 23 June, respectively. Both nests
were found before the first egg was laid, yielding
precise documentation of these extremes, but even
back-calculating nests that were already initiated
when found did not exceed the.se extremes. The
mean initiation date was 7 May (± 2.2 days, n =
106) and highest nesting activity occuired at the
end of April and through May (Fig. 1 ).

Re-nesting activities were observed within 6-
8 days after a nest failed and within a 15-m radius

of the previous nest (// = 10). Banded individuals
allowed confirmation that it was a re-nesting
attempt by a particular pair of individuals. The
nesting season of the Slate-throated Whitestart in
Yacambii was 77 days, but this species appears to
be single brooded; we did not observe any
breeding activity of the same pair in the same
season after a successful brood.

Clutch Size and Eggs. — Mean clutch size for
the Slate-throated Whitestart was 2.1 ± 0.04 eggs
(/? = 93). We recorded one instance of a single
egg clutch (1%), 15 nests had three eggs (16%),
and the rest had two eggs (83%). Three-egg
clutches did not exceed 25% of the clutches in any
year, but three-egg clutches were observed in all
years except 2004 (n = 14 in 2004).

Eggs were laid on consecutive days and
generally early in the morning (i.e., before 0800
hrs). The eggs were elliptical to sub-elliptical and
the shell was dull white with iiTegularly distrib-
uted dark red to brown spots. Mean egg mass was
1.61 ± 0.01 g (n = 152 eggs) and ranged from
1.20 to 1.94 g. The mean egg mass represented
19% of adult female body mass (10.1 ± 0. 15 g, /?
= 20 females).

Incubation Period. — Incubation began when
the last egg was laid, but nest attentiveness was
highly variable on that day. Eggs hatched
synchronously in all cases. Only the female
incubated, and the male rarely visited the nest to
feed the female. The male did not visit during 6-
8 hrs of video observations at 31 of 52 nests
sampled during incubation. Overall, the male
averaged 0. 1 1 ± 0.02 trips/hr to the nest (n = 52)
during incubation.

The incubation period of the Slate-throated
Whitestart averaged 15.3 ± 0.31 days (n = 21)
varying from 14 to 18 days. Overall, nest
attentiveness (percent of time spent on the nest
incubating) during this stage averaged 59 ± 1.6%
in = 52). Nest attentiveness increased (E2.49 =
15.8, P < 0.001) from early incubation to mid-
incubation but did not continue to increase in late
incubation (Eig. 2A). The increase in nest atten-
tiveness was caused by reduction in duration of
off-bouts with increasing age of the embryo
(Fig. 2B; ^249 = 5.4, P = 0.008). Duration of
on-bouts did not change (^2,49 = 1.49, P = 0.17)
with age of the embryo (Fig. 2B). On- and off-
bouts averaged 33.4 ± 1.63 min and 22.8 ±
2.65 min (// = 52 in both cases), respectively.
Temperature of Slate-throated Whitestart eggs
averaged 34.40 ± 0.33° C over 24-hr periods (/? =

450

THE WILSON JOURNAL OL ORNITHOLOGY • Voi 122, No. 3, September 2010

Incubation period

FIG. 2. Mean ± SE for (A) nest attentiveness, and (B)
on- and off-bout durations across three periods of
incubation: early (days 1-5), middle (days 6-10), and late
(days 11-17) for Slate-throated Whitestarts at Yacambti
National Park, Venezuela. Sample sizes reflect numbers of
nests. Different letters among periods indicate significant
differences (a = 0.05) based on LSD tests of each bout type
(on- vs. off-bouts) separately.

6 ne.sts, 20 days of sampling) based on an overall
mean taken from the means of each nest.

Nestling Period. — Hatchlings had sparse natal
down with pink-orange skin, and were blind. The
eighth primary feather broke its sheath on days 6
or 7. Brooding attentiveness was by females alone
and averaged 63 ± 2.9% {n = 10) at the
beginning of the nestling period (days 0-3 of
age) and decreased dramatically (/^u.s == 92.6, P
< 0.001 ) to 1 1 ± 4.9% (/? = 7) at pin break. Both
females and males provisioned the nestlings at an
average rate of 3.77 ± 0.44 trips/hr (// = 10) early
in the nestling period (days 0-3), but provisioning
rate markedly increased (F\,\5 — 25.7, P < 0.001 )
to 9.41 ± 1.18 trips/hr (// = 7) at pin break.

Nestlings fledged in 10.8 ± 0.24 days in = 1)
after hatching, ranging from 10 to 11.5 days.
Growth rate ba.sed on nestling mass (Fig. 3A) was
greater than when based on tarsus length
(Fig. 3B). Nestling weights on the last 2 days in

12 ■

A

10 ■

O)

C/)

c/)

cc

8 -
6 -

4 •

2 -
0 ■■

I

I

I

/c= 0.521 +0.015

T 1 I

18

E

E 15

cn 12 ■

c

o

c/) 9

3

CO

,C0 6

B

.1

!i

M'

I

I

/c= 0.334 + 0.016

0 2 4 6 8 10 12

Nestling age (day)

FIG. 3. (A) Mass and (B) tarsus length measurements

of Slate-throated Whitestart nestlings at Yacambu National
Park, Venezuela plotted against age. Growth rate constant
(k) is indicated in each graph.

the nest (days 10 and 1 1 ) averaged 10.78 ± 0.22 g
(// = 5 fledglings) compared with mean adult
mass (10.1 ± 0. 15 g for the female [n = 20] and
10.1 ± 0.1 1 g for the male [/? = 36]).

Nest Survival. — Eighteen of 126 nests were
excluded from nest survival analyses because of
disruption by experimental activities or because
they were abandoned before egg-laying. Overall
daily mortality rate for the remaining nests was
0.065 ± 0.008 in = 108 nests; 1,084.5 exposure
days) and only 15% of the nests were successful
(ba,sed on a total nesting period of 30.2 days).
Overall daily nest predation rate was 0.053 ±
().()()7. Daily nest predation rates increased with
age of the nest from 0.037 ± 0.021 (/? = 81.5
exposure days) during egg-laying to 0.052 ±
().()()8 in = 736 exposure days) during incubation
to 0.060 ± 0.015 in = 267.0 exposure days)
during the nestling stage. Predators included
mostly birds (e.g., small raptors, jays, and
woodwrens) but also small mammals (e.g.,
squirrels and tayra [Eira harhara]).

Rtiggcra amt Martin • BREEDING BIOIX3GY OE A TROPICAL. WARBLER

451

DISCUSSION

The nesting season for Slate-throated White-
start in Yacambil (late Mar to late Jiin) was almost
identical to reports elsewhere in Venezuela
(Collins and Ryan 1994), and Costa Rica (Skutch
1954, Mumme 2010). In general, the highly
seasonal breeding activity began at the end of
the dry season and continued through the rainy
season.

Behavior during nest building was consistent
with previous observations reported by Skutch
(1945, 1954) and Mumme (2010) in Central
America, and for other Myioborus species (e.g.,
M. pictiis [Marshall and Baida 1974] and M.
flaviverte.x [Morales-Rozo 2009]). Characteristics
of Slate-throated Whitestart nests appear to be
highly conservative along the broad range of the
species. Slate-throated Whitestart nests in Ya-
cambu were in the same types of places reported
from Central America (Skutch 1954, McCormack
et al. 2005, Mumme 2010) and Venezuela (Ewert
1975, Collins and Ryan 1994). Slate-throated
Whitestart nests were quite similar to some other
members of the genus, including M. melanoce-
phalus (Greeney et al. 2008), M. torquatus
(Skutch 1954), M. bnmniceps (Auer et al. 2007),
and M. flavivertex (Morales-Rozo et al. 2009).
However, some tropical Myioborus species, such
as M. ornatiis, apparently build open cup nests
(Curson et al. 1994; O. Cortes, pers. comm.).
Painted Redstart (M. pictiis), the most northern
species of the group, only occasionally builds an
enclosed nest, and more often makes an open cup
with the roof substituted by natural overhangs
such as rocks, logs, and pine needles (Marshall
and Baida 1974). Open-cup nests are common for
warblers outside of the tropics. The placement of
nests in small holes in the ground is also typical of
some North American parulids, such as ground-
nesting Vennivora, which do not build a roof
(Martin 1993).

Clutch size for Slate-throated Whitestart at our
study site was relatively constant with 83% of
nests containing two eggs. The mean clutch size
in Yacambti (2.1 eggs) was relatively small
compared with that of M. m. pallidiventris only
250 km from our .study site, which averaged 2.7
eggs based on six nests (Collins and Ryan 1994).
This difference may reflect sample size issues, but
it is notable that four of their six nests (67%) had
three eggs, while only 15 of 93 nests (16%) had
three eggs in our study. Moreover, we did not find

more than 25% of nests with three eggs in any
individual year. Other Myioborus species near the
equator also had two-egg clutches. M. melanoce-
phalus in Ecuador had two eggs in all of .seven
nests (Greeney et al. 2008), and M. flavivertex in
Colombia had two eggs in the only nest ob.served
(Morales-Rozo et al. 2009). Basileuterus tristria-
tus in Yacambu had a mean clutch size of 1.96
eggs and only one of 96 clutches had three eggs
(Cox and Martin 2009). Thus, two eggs near the
equator may be more typical of warblers, but the
frequency of occurrence of three eggs needs
further study in Myioborus.

Variation in clutch size among latitudes is
clear. M. m. aurantiacus in Costa Rica usually
laid three eggs with mean clutch sizes of 2.84
eggs from 13 nests (Skutch 1954) and 2.89 from
82 nests (Mumme 2010). The only nest observed
for M. m. hellmayri in Guatemala had three eggs
(Skutch 1954), M. bnmniceps in Argentina had a
mean clutch size of 2.6 eggs (Auer et al. 2007),
while M. pictus in Arizona had a mean clutch size
of 3.7 eggs (Marshall and Baida 1974). In
contrast, warblers in the north temperate region
typically have mean clutch sizes of 4-5 eggs
(Martin 1988). Thus, tropical warblers have a
strong reduction in clutch size compared to north
temperate warblers, but variation within the
tropics and subtropics both in Slate-throated
Whitestart races and among congeners also
appears evident.

Incubation behavior and duration (15.3 days) of
the Slate-throated Whitestart in Yacambu seem to
be typical tor tropical and subtropical warblers
(see Cox and Martin 2009) and differ substantially
from northern relatives. The mean incubation
period was about a day longer than for other
.SLibspecies (Skutch 1954, Collins and Ryan 1994.
Mumme 2010). However, other whitestart species
and most tropical warblers have incubation
periods ranging from 14 to 16 days (Martin et
al. 2007, Cox and Martin 2009), and our results fit
in the middle of this range. Longer incubation
periods in the tropics and subtropics, compared to
northern species, have been attributed to lower
nest attentiveness which causes colder incubation
temperatures (Martin 2002, Martin et al. 2007.
Martin and Schwabl 2008, Londono 2009). Nest
attentiveness of Slate-throated Whitestarts at our
study site was relatively low hut similar to other
tropical and southern warblers (Cox and Martin
2009). Skutch (1954) studied one nest of Slate-
throated Whitestart in Guatemala during middle

452

THE WILSON JOURNAL OF ORNITHOLOGY • Voi 122, No. 3. September 2010

incubation, and both nest attentiveness (67.4%)
and duration of bouts (on-bouts = 37.6 min; off-
bouts = 18.2 min) were similar to our observa-
tions for mid-incubation (Fig. 2).

Length of the nestling period (10-1 1 days) had
low variation within Yacambu and was similar to
that observed for Slate-throated Whitestarts in
other localities (Skutch 1954, Collins and Ryan
1994, Mumme 2010). Nestling periods in other
species of Myioboriis are similar to Slate-throated
Whitestarts (Skutch 1954, Marshall and Baida
1974, Auer et al. 2007, Greeney et al. 2008). Most
warblers have a nestling period within a range of
9-13 days (Cox and Martin 2009) even when
including north temperate and tropical species
(Elliott 1969, Martin 1995, Remes and Martin
2002, Martinez et al. 2004, Di Giacomo 2005,
Auer et al. 2007, Cox and Martin 2009). The
duration of the nestling period appears to be a
more conservative life-history trait within Paruli-
dae than incubation period.

Slate-throated Whitestarts in Costa Rica made
20.3 provisioning trips/hr, on average, during
nestling ages 5-9 days (Mumme 2010), which is
twice the rate that we observed (9.41 trips/hr) for
this species in Yacambu on days 6-7. Despite the
much higher provisioning rate, nestlings averaged
the same weight on days 7-8 in both places: 9.5 g
in Costa Rica (Mumme 2010) and 9.5 ± 0.17g {n
= 24) in Venezuela (this study). The larger clutch
size in Costa Rica (Mumme 2010) may account
for the higher provisioning rate and similar
nestling size compared with our site. The larger
clutch size and similarly sized nestlings with
higher parental provisioning rates in Costa Rica
(Mumme 2010) compared with our site may
suggest that food is more limiting at our site for
this species.

Growth rate estimates are scarce for tropical
warblers. One estimate based on limited data for
the Slate-throated Whitestart in Venezuela (Col-
lins and Ryan 1994) found that nestlings grew at a
rate (k = 0.522) based on mass almost identical to
our estimate from Yacambu (k = 0.521, Fig. 3).
Generally, tropical passerines grow more slowly
than species in temperate regions (Ricklefs 1976,
Cox and Martin 2009). Our estimated growth rate
(Fig. 3) was slower than estimates for north
temperate warblers (i.e., Remes and Martin
2002, Cox and Martin 2009), but faster than for
many other tropical passerines (Ricklefs 1976).
This relatively fast growth of nestlings given slow
embryonic development (i.e., long incubation

periods) demonstrates that developmental rates
in embryonic and post-natal stages are not
constrained by each other.

Nest design is thought to be an important factor
influencing predation rate and failure from
weather (Collias and Collias 1984, Auer et al.
2007). The last may be especially important in
tropical and subtropical montane forests where
rains during the breeding season are quite intense.
Enclosed nests in these environments may afford
nest contents protection from rain (Collias and
Collias 1984) as well as predation (Auer et al.
2007). The daily predation rate of Slate-throated
Whitestarts in Yacambu was higher than the
average for all non-cavity-nesting species studied
at this site (0.041 ± 0.003, n = 38 spp.; TEM,
unpubl. data). This average was remarkably
similar to the average for 10 non-cavity-nesting
species studied in Panama (Robinson et al. 2000).
Thus, the domed ground nests of this warbler in
Yacambu are readily found by predators.

The Slate-throated Whitestart in Yacambu had
a relatively high daily predation rate, small clutch
size, long incubation period, and low nest
attentiveness during incubation, typical of many
tropical passerines (Skutch 1954; Martin et al.
2000, 2007; Cox and Martin 2009). Warblers
appear to have strong divergences among relatives
across geographic space, making them an impor-
tant group for studying latitudinal changes in life-
history strategies. Some traits, such as clutch size
seem to have geographic variation even within the
tropics within this species and genus. Other
features of Slate-throated Whitestart nesting
behavior, including nest construction and place-
ment, nestling period length, and growth rate are
relatively conservative. Focusing effort on par-
ulids in tropical latitudes in particular could
strengthen many aspects of life history theory.

ACKNOWLEDGMENTS

We thank Carlo.s Bosque for substantial logistical support
in aiding this work. We also thank many assistants for
valuable help in the field. We are grateful to Amy Stokes
for help with data summary and T. E. Martin’s and Pedro
Blendinger’s laboratories for helpful comments on the
manuscript. We are also grateful to J. L. Perez-Eman, J. G.
Blake, and two anonymous reviewers for valuable com-
ments and improvements to the manuscript. This study was
made possible in part by support under NSF grants DEB-
9981527, DEB-0543178, and DEB-0841764 to T. E.
Martin. Permit numbers are DM/0000237 from EONACIT,
PA-lNP-005-2004 from INPARQUES, and 01-03-03-1 147
from Ministerio del Ambiente. Any use of trade names is to

Riiggeni aiul Mcirtiii • BREEDING BIOLOGY OF A TROPICAL WARBLER

453

aid method descriptions only and does not imply endorse-
ment by the U.S. Government.

LITERATURE CITED

Auer, S. K., R. D. B.'Vssar, J. J. Fontaine, and T. E.
Martin. 2007. Breeding biology of passerines in a
subtropical montane forest in northwestern Argentina.
Condor 109:321-333.

Briskie, j. and S. G. Sealy. 1990. Evolution of short
incubation periods in the parasitic cowbirds, Molo-
thnis spp. Auk 107:789-793.

CoLLiAS, N. E. and E. Collias. 1984. Nest building and
bird behavior. Princeton University Press, Princeton,
New Jersey, USA.

Collins, C. T. and T. P. Ryan. 1994. Notes on breeding
biology of the Slate-throated Redstart {Myioboms
miniatiis) in Venezuela. Ornitologt'a Neotropical
5:125-128.

Cox, W. A. and T. E. Martin. 2009. Breeding biology of
the Three-Striped Warbler in Venezuela. Wilson
Journal of Ornithology 121:667-678.

CuRSON, J., D. Quinn, and D. Beadle. 1994. Warblers of
the Americas: an identification guide. Houghton-
Mifflin, New York, USA.

Di Giacomo, A. G. 1995. Dos especies nuevas para la
avifauna argentina. Hornero 14:77-78.

Di Giacomo, A. G. 2005. Aves de la Reserva El Bagual.
Pages 201^66 in Historia natural y paisaje de la
Reserva El Bagual, Provincia de Formosa, Argentina.
Inventario de la fauna de vertebrados y de la flora
vascular de un area protegida del Chaco Humedo (A.
G. Di Giacomo and S. F. Krapovickas, Editores).
Temas de Naturaleza y Conservacion 4. Aves
Argentinas/Asociacion Omitologica del Plata, Buenos
Aires, Argentina.

Dunn, J. L. and K. L. Garrett. 1997. A field guide to the
warblers of America. Houghton-Mifflin, Boston,
Massachusetts, USA.

Elliott, B. G. 1969. Life history of the Red Warbler.
Wilson Bulletin 81:184-195.

Ewert, D. 1975. Notes on nests of four avian species from
the Coastal Cordillera of Venezuela. Wilson Bulletin
87:106.

Fierro-CalderOn, K. and T. E. Martin. 2007. Repro-
ductive biology of the Violet-chested Hummingbird in
Venezuela and comparisons with other tropical and
temperate hummingbirds. Condor 109:680-685.

Greeney, H. F., P. R. Martin, R. C. Dobbs, R. A. Gelis,
A. BOcker, and H. Montag. 2008. Nesting ecology
of the Spectacled Whitestart in Ecuador. Ornitologt'a
Neotropical 19:335-344.

lUCN. 2009. lUCN Red List of threatened species. Version
2009.1. www.iucnredlist.org.

Londono, G. a. 2009. Eggs, nests, and incubation behavior
of the Moustached Wren (Thryothorus geniharhis) in
Manu National Park, Peru. Wilson Journal of Orni-
thology 121: 623-627.

Marshall, J. and R. P. Balda. 1974. The breeding
ecology of the Painted Redstart. Condor 76:89-101.

Martin, T. E. 1988. Nest placement: implications for

selected life history traits, with special reference to
clutch size. American Naturalist 132:900-910.

Martin, T. E. 1993. Nest predation and nest sites: new
perspectives on old patterns. BioScience 43:523-532.

Martin, T. E. 1995. Avian life history evolution in relation
to nest sites, nest predation and food. Ecological
Monographs 65:101-127.

Martin, T. E. 2002. A new view of avian life-history
evolution tested on an incubation paradox. Proceed-
ings of the Royal Society of London, Series
B 269:309-316.

Martin, T. E. 2007. Climate correlates of 20 years of
trophic changes in a high-elevation riparian system.
Ecology 88:367-380.

Martin, T. E. and G. R. Geupel. 1993. Nest-monitoring
plots-methods for locating nests and monitoring
success. Journal of Field Ornithology 64:507-519.

Martin, T. E. and C. K. Ghalambor. 1999. Males helping
females during incubation. I. Required by microcli-
mate or constrained by nest predation? American
Naturalist 153:131-139.

Martin, T. E. and H. Schwabl. 2008. Variation in mater-
nal effects and embryonic development rates among
passerine species. Philosophical Transactions of the
Royal Society of London, Series B 363:1663-1674.

Martin, T. E., P. R. Martin, C. R. Olson, B. J.
Heidinger, and j. j. Fontaine. 2000. Parental care
and clutch sizes in North and South American birds.
Science 287:1482-1485.

Martin, T. E., S. K. Auer, R. D. Bassar, A. M. Niklison,
AND P. Lloyd. 2007. Geographic variation in avian
incubation periods and parental influences on embry-
onic temperature. Evolution 61:2558-2569.

Martinez, W. E., V. D. Piaskowski, and M. Teul. 2004.
Reproductive biology of the Gray-crowned Yellow-
throat (Geothlypis poliocephala palpebralis) in central
Belize. Ornitologia Neotropical 15:155-162.

Mayfield, H. F. 1975. Suggestions for calculating nest
success. Wilson Bulletin 87:456-466.

McCormack, J. E., G. Castaneda Guayas.amin, B. Mila,
and F. Heredia Pineda. 2005. Slate-throated Red-
starts (Myioboms miniatus) breeding in Maderas del
Carmen, Coahuila, Mexico. Southwestern Naturalist
50:501-503.

Morales-Rozo, a., E. Rodriguez-Ortiz, B. Freeman,
C. A. Olaciregui, and C. D. Cadena. 2009. Notas
sobre el nido y los pichones del Abanico Colombiano
(Myioboms flavivertex: Parulidae). Ornitologia Neo-
tropical 20: 113-119.

Mumme, R. L. 2010. Breeding biology and nesting success
of the Slate-throated Whitestart (Myioboms miniatus)
in Monteverde, Costa Rica. Wilson Journal of
Ornithology 122: 29-38.

P£rez-Eman, j. L. 2005. Molecular phylogenetics and
biogeography of the neotropical redstarts (Myioboms-,
Aves, Parulinae). Molecular Phylogenetics and Evo-
lution 37:51 1-528.

Phelps, W, H. and W. H. Phelps Jr. 1950. Lista de las
aves de Venezuela con su distribucion. Parte 2.
Passeriformes. Boletin de la Sociedad Venezolana de
Ciencias Naturales 12:1^27.

454

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 3. September 2010

RemeS, V. AND T. E. Martin. 2002. Environmental
influences on the evolution of growth and develop-
mental rates in passerines. Evolution 56: 2505-2518.

Ricklefs, R. E. 1976. Growth rates of birds in the humid
New World tropics. Ibis 118:179-207.

Ridgely, R. S. and G. Tudor. 1994. The birds of South
America. Volume 2. University of Texas Press, Austin,
USA.

Robinson, W. D., T. R. Robinson, S. K. Robinson, and J.
D. Brawn. 2000. Nesting success of understory forest
birds in central Panama. Journal of Avian Biology 31:
151-164.

Sibley, C. G. and B. L. Monroe Jr. 1990. Distribution and
taxonomy of birds of the world. Yale University Press,
New Haven, Connecticut, USA.

Skutch, a. E. 1945. Studies of Central American redstarts.
Wilson Bulletin 57:216-234.

Skutch, A. F. 1954. Life histories of Central American
birds: Families Fringillidae, Thraupidae, Icteridae,
Parulidae and Coerebidae. Cooper Ornithological
Society, Berkeley, California, USA.

Weathers, W. W. and K. A. Sullivan. 1989. Nest
attentiveness and egg temperature in the Yellow-eyed
Junco. Condor 91: 628-633.

The IVilsun Journal oj Ornilholof'v 1 22(3):455-464, 2010

REPRODUCTIVE BIOLOGY OF A GRASSLAND SONGBIRD
COMMUNITY IN NORTHCENTRAL MONTANA

STEPHANIE L. JONES, J. SCOTT DIENI,' AND PAULA J. GOUSE'"

ABSTRACT. — Successful conservation of grassland bird populations requires basic information on their breeding
biology; in particular, information from undisturbed native prairie over an extended period of time. We pre.sent data
collected at Bowdoin National Wildlife Refuge in northcentral Montana on the reproductive biology of six grassland bird
species that breed in mixed-grass prairie: Sprague’s Pipit (Anthiis spragneii). Savannah (Passercuhts sandwichensis).
Grasshopper (Ammod ramus savannariim), and Baird’s (A. hairdii] sparrows, Chestnut-collared Longspur (Calcarius
ornatus), and Western Meadowlark (Stiirnella neglecta). Basic measures of reproductive biology are presented, including
estimates of nest density, phenology, fecundity, parasitism rates, and nest success, and how these and other characteristics
varied across years. Nests {n = 1,494) of the six focal species accounted for 98% of all passerine nests found during 1997-
2007; Chestnut-collared Longspurs (51%) were the dominant breeding species. Total nest density across years ranged from
20 to 41 nests per 40 ha (CV = 26%) on unburned sites. Mean clutch initiation date and clutch size varied little across
years; however, clutch size tended to decrease over the course of a season, regardless of bird species. Daily nest survival
rates did not differ markedly among bird species, but did vary substantially among years, suggesting that year-dependent
factors were affecting nest success among all species similarly. Received 20 December 2008. Accepted 28 February 2010.

Declines in grassland bird populations have
been consistently greater and geographically more
widespread than in any other avifaunal group in
North America (Knopf 1996, Askins et al. 2007,
Sauer et al. 2008), which has been particularly
evident for breeding endemics of the northern
mixed-grass prairie (Knopf 1996). This has been
attributed to many factors, including habitat
conversion and fragmentation (Askins et al.
2007), woody vegetation encroachment (Houston
and Schmutz 1999), removal of native grazers
(Knopf 1994, 1996) and fire (Higgins 1984), and
increased predation and parasitism (Basore et al.
1986, Davis and Sealy 2000).

Conservation of grassland bird populations
requires basic information on their breeding
biology, especially data from undisturbed native
prairie over an extended period of time. Large
expansive areas of shortgrass and mixed-grass
prairies still exist in the western portion of the
Great Plains, including northcentral Montana,
where these grasslands are relatively intact. In
addition, northern prairie grasslands are funda-

' U.S. Fish and Wildlife Service, P. O. Box 25486 DFC,
Denver, CO 80225, USA.

“Redstart Consulting, 403 Deer Road, Evergreen, CO
80439, USA.

^U.S. Fish and Wildlife Service, Bowdoin National
Wildlife Refuge, Malta, MT 59538, USA.

■‘Current address: U.S. Fish and Wildlife Service,
Charles M. Russell National Wildlife Refuge, P. O. Box
1 10, Lewiston, MT 59457, USA.

’Corresponding author;
e-mail: stephanie_jones(§>fws.gov

mentally dynamic. Local temporal variability in
grassland conditions are largely a function of
weather, which can influence plant community
structure and composition, both within- and
between seasons (Igl and Johnson 1999, Winter
et al. 2005, Askins et al. 2007). Fluctuations in
habitat conditions result in notable shifts in local
bird population densities (Igl and Johnson 1999,
Winter et al. 2005) as many native grassland bird
species adapt to shifting habitat conditions with
nomadic behavior (Clark and Shutler 1999,
Ahlering and Johnson 2006, Jones et al. 2007).
Consequently, long-term studies focusing on the
same population can be instrumental to under-
standing how reproductive characteristics vary
over time.

We studied six species, two of which, Sprague’s
Pipit {Anthus spragueii) and Baird’s Sparrow
(Ammodramus hairdii) are breeding endemics to
the northern mixed-grass prairie, and a third.
Chestnut-collared Longspur (Calcarius ornatus),
breeds only in the short to mixed-grass prairie of
the western and northern Great Plains. These
species have been the subject of recent studies
on their reproductive biology (Davis and Sealy
1998, Winter 1999, Davis 2003), and on habitat
and management effects (Madden et al. 1999.
Davis 2005). However, few studies have occuired
in Montana (Hill and Gould 1997, Robbins and
Dale 1999, Green et al. 2002), or over an extended
period of years. Savannah (Passercuhts .sandwi-
chensis), and Grasshopper (Annnodramus .savan-
na rum) sparrows, and Western Meadowlarks
(Sturnella neglecta) have been well-studied

455

456

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3, September 2010

throughout much of their range, but relatively
few data for these species are available from the
northern mixed-grass prairie (Wheelwright and
Rising 1993, Lanyon 1994, Vickery 1996).

We present 1 1 years of data on the reproductive
biology of six grassland bird species that breed in
mixed-grass prairie of northcentral Montana:
Sprague’s Pipit; Savannah, Grasshopper, and
Baird’s sparrows; Chestnut-collared Longspur;
and Western Meadowlark. Our objectives were to
characterize: ( 1 ) songbird community structure, (2)
nesting phenology, (3) fecundity, (4) brood para-
sitism rates and nest success, and (5) to describe
how these characteristics varied across years.

METHODS

Study Area. — Our study was conducted during
1997-2007 at Bowdoin National Wildlife Refuge
(NWR) in Phillips County, near the town of Malta
in northcentral Montana (48° 24' N, 107° 39' W;
—750 m asl). The study area consisted of five
permanent sites (26-59 ha) totaling 218 ha of flat
to gently rolling native mixed-grass prairie
(Fig. 1). Data for this analysis were collected at
four sites beginning in 1997. One of these sites
partially (3 ha) burned in a wildfire in 1994, and
was then prescribed burned during spring 2004; a
second site was prescribed burned in spring 2007
and was not monitored during 2004. A fifth site,
which had been prescribed burned during spring

2000, was added in 2001 . No other burning events
have occurred on the study sites since refuge
documentation began in 1936. Grazing by cattle
was gradually phased out during 1973-1977.

The climate is continental and semiarid,
characterized by strong winds and high evapora-
tion rates. Mean long-term annual and seasonal
(May-Jul) precipitation totals are 34 and 18 cm,
respectively (National Climatic Data Center
2007). Mild drought conditions prevailed during
the study period; mean annual and seasonal
precipitation totals were 32 and 14 cm, respec-
tively (National Climatic Data Center 2007). The
plant community was dominated by western
wheatgrass {Pascopynim smithii), needle-and-
thread (Stipa coinata), blue grama (Bouteloua
gracilis), and clubmoss (Selaginella densa).
Invasive plant species were isolated and sparse.
Shrubs (Sarcobatus vermicidatus, Artemisia cana,
Ceratoides lanata) were sparse, and trees absent,
except sporadically outside the periphery of two
study sites.

Photographic records at Sprague’s Pipit nests (u
= 5) documented Northern Harrier {Circus
cycmeus), garter snake (Tliamnopliis spp.). West-
ern Meadowlark, and deer mice (Peromyscus
spp.) preying upon nestlings during this study
(S. L. Jones, unpubl. data). Other potential nest
predators have been observed on, or within, the
immediate vicinity of our study sites, including:

Jones el at. • NESTING OE GRASSLAND SONGBIRDS

457

American badger (Taxidea taxiis), long-tailed
weasel (Mnstela fremita), red fox (Vidpes vulpes),
coyote {Canis latnms), mice and voles (Zapus,
Reithrodontomys, and Microtiis), Richardson’s
ground squirrel (Spermophihis ricluirdsonii), bull-
snake (Pituopliis melanoleucus), prairie rattle-
snake (Crotcdus viridis), gulls {Lcinis spp.), Short-
eared Owl (Asia flanimeiis). Loggerhead Shrike
{Lanins ludovicianns), and Black-billed Magpie
{Pica luidsonia).

Nest Searching. — Sites were searched for nests
3-5 times per week from early May through late
July in an attempt to locate all active nests each
year. Search techniques (% successful) included
behavioral observations (3%) (Martin and Geupel
1993), opportunistic foot flushes (36%), and rope
dragging (61%) (Davis 2003, Dieni and Jones
2003). Nests were marked for relocation by
placing a discrete strip of plastic flagging on the
ground ~2.5 m on either side of the nest. Nests
were monitored every 2-A days until the nesting
outcome was ascertained. Nesting stage at initial
discovery was assigned by candling eggs (Loke-
moen and Koford 1996) or estimating age of
nestlings (Jongsomjit et al. 2007). Nesting out-
comes were: (1) successfully fledged at least one
young of the parental species, (2) depredated, (3)
abandoned (eggs or nestlings left permanently
unattended), or (4) unknown. Observations of
fledglings within 3 days of expected fledging, the
presence of feces and feather scales in the nest,
fledglings near the nest, and adults uttering alarm
calls nearby or feeding new fledglings within 50 m
of the nest were treated as evidence of reproduc-
tive success. Predation was assumed when the
nest, eggs or nestlings too young to fledge
disappeared or were destroyed.

Statistical Analysis. — Number of nests located
for each focal bird species was summarized only
for the three sites which had not been disturbed by
prescribed fire to quantify annual variation of
nesting density. Site effects were ignored given
the relatively close spatial proximity of the study
sites.

The nest was the primary sample unit for the
analysis of reproductive parameters and nest
success. Clutch initiation dates (CID) were
estimated by calculating the date of clutch
initiation from egg or nestling age or from known
dates of major events (e.g., hatching). Reproduc-
tive parameters are reported using standard
measures of central tendency. Data variation was
summarized using the coefficient of variation (CV

= SD/mean; Zar 1999); annual variation for many
parameters was summarized only for the domi-
nant species with adequate sample sizes. Trend
lines (across years) were fit using the method of
least squares; strength of trends was characterized
using simple beta {h) and correlation (Pearson r)
coefficients.

We estimated nest daily survival rates (DSR)
using the survival model in program MARK
(White 2008) for the six focal bird species, and to
examine how DSR varied temporally across years
(1997-2007). Program MARK uses a generalized
linear approach to modeling DSR and maximum
likelihood estimation to derive model coefficients
and sampling variances (Dinsmore et al. 2002,
Rotella et al. 2004). Regression models were
constructed using the logistic transformation
(logit) as the link function and natural logs
[ln(DSR/l-DSR)]. Bird species and year were
treated as nominal-scale variables (6 and 11
levels, respectively) with each level introduced
into the regression model as an artificial explan-
atory variable and 0 or 1 coding; second-order
interactions between species and year were added
with the appropriate cross-product terms. One
candidate model set was evaluated within an
information-theoretic framework (Burnham and
Anderson 2002). Program MARK was used to
calculate Akaike’s Information Criterion coirect-
ed for small sample sizes (AIC,.), ranking the fit of
each model in ascending order of AIC^. values.
Estimates for each species and year were
calculated using the model-averaging approach
(Burnham and Anderson 2002) weighted accord-
ing to that model’s likelihood in the set (normal-
ized Akaike weight). The intercept-only model
(where the regression coefficient was set to zero)
was included in a set with the covariate of interest
(species or year), which served to reduce model-
selection bias of the estimate (Burnham and
Anderson 2002). A point estimate of nest success
was calculated as DSR raised to the mean number
of days of the nesting period (incubation and
nestling stages combined) for each species
(Rotella 2007).

RESULTS

Nine species of passerines were actively
breeding on our study area. Nests (/; = 1,494)
of the six focal species accounted for 98% of all
passerine nests found during 1997-2007. Only a
small number of nests of Horned Lark {Eremo-
phila alpestris-, n = 5), Lark Bunting {Calamos-

458

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3, September 2010

TABLE 1. Annual variation in nesting totals for six songbird species, Bowdoin National Wildlife Refuge, Montana.
Numbers of nests discovered in the three undisturbed sites were summarized across years (1997-2006). Total nests were
summarized across all songbird species encountered. Values are adjusted per 40 ha. Linear trends in nest totals across years
are summarized using Pearson (/•) and beta (b) coefficients.

Statistic

Sprague's Pipit

Savannah Sparrow

Grasshopper Sparrow

Baird's Sparrow

Chestnut-collared Longspur

Western Meadowlark

Total nests

Mean

2.7

4.6

2.7

1.5

15.4

2.6

30.4

CV

0.49

0.30

0.33

1.05

0.31

0.46

0.25

Min

1.3

2.8

1.5

0

10.7

0.8

20.6

Max

5.3

7.4

4.1

4.8

22.4

4.1

41.2

n

94

164

97

52

547

92

1,075

r

0.16

-0.64

0.14

-0.76

-0.53

-0.61

-0.56

h

0.1

-0.3

0.0

-0.4

-0.8

-0.2

-1.3

pizci melcmocorys\ n — 7), and Vesper Sparrow
(Pooecetes gramineus\ n = 18) were found.
Chestnut-collared Longspurs were the dominant
breeding species accounting for over 51% of all
passerine nests discovered over all years, followed
by Savannah Sparrow (17%), Sprague’s Pipit
(8.4%), Grasshopper Sparrow (8.1%), Western
Meadowlark (7.8%), and Baird’s Sparrow (6%).
Mean rank in nest abundance across all years
followed a similar order: Chestnut-collared Long-
spur (1.0, CV = 0.0), Savannah Sparrow (2.2, CV
= 0.20), Western Meadowlark (3.9, CV = 0.27),
Grasshopper Sparrow (4.1, CV = 0.19), Sprague’s
Pipit (4.2, CV = 0.33), and Baird’s Sparrow (5.6,
CV = 0.18).

Annual variation in total nests was summarized
collectively for three unburned study sites (Ta-
ble 1). Annual variation (CV) in nest density
ranged from 1.05 for Baird’s Sparrow to 0.30 for

Savannah Span'ow among focal species. Variabil-
ity in total songbird nest density was moderate
(CV = 0.25) ranging from 41 nests/40 ha in 1997
to 21 nests/40 ha in 2001. A modest declining
trend for total songbird nests/40 ha was evident {b
= — 1.3, r = —0.56) across all years. This trend
was pronounced for Baird’s SpaiTow (r = —0.76),
but also evident for Savannah Sparrow (r =
—0.64), WesteiTi Meadowlarks (r = —0.61), and
Chestnut-collared Longspurs (r = —0.53). Nests
were primarily discovered during incubation
(67%) with frequency of nests discovered gener-
ally decreasing thereafter. Relatively few nests
were discovered during the nest-building phase
(Fig. 2).

Reproductive Parameters. — The CID frequency
distribution for Chestnut-collared Longspurs was
multimodal across all years, exhibiting three
peaks. Bimodal distributions were evident for

Build Laying-

Incubation —

Nestling

509

(Z)

c 400
300

^ 200
I 100
0 ■

15

222

266

110 122

168

68

14

0 1-3 4-6 7-9 10-12 13-15 16-18 19-21 22-29

Days in nesting cycle

FIG. 2. Frequency distribution of nest age-intervals at time of discovery, Bowdoin National Wildlife Refuge, Montana,
1997-2007 Oi = 1.494). Nests were primarily located during incubation with frequency of nests generally decreasing
thereafter. Relatively few nests were located during the nest-building phase.

Jones el al. • NESTING OF GRASSLANO SONGBIRDS

459

TABLE 2. Frequency distributions For clutch initiation dates (CID) for six songbird species at Bowdoin National
Wildlife Refuge, Montana, 1997-2()()7. Values include peaks (modes), mean, and median nesting dates. Nesting period
includes the range of activities observed across years for each species from the earliest C!D date to last nest fledged.
Frequency distributions were multimodal for three species and generally positively skewed towards early season breeding
for all species.

Species

Nest peaks

Mean

Median

Earliest

Latest

Potential nesting period

Sprague’s Pipit

23 May

5 Jun

25 May

7 May

31 Jul

07 May-25 Aug

Savannah Sparrow

24 May, 27 Jun

6 Jun

2 Jun

11 May

19 Jul

1 1 May-09 Aug

Grasshopper Sparrow

27 May

13 Jun

16 Jun

18 May

17 Jul

18 May-06 Aug

Baird's Spanow

30 May

1 1 Jun

9 Jun

14 May

21 Jul

14 May-10 Aug

Chestnut-collared Longspur

21 May, 6 Jun, 15 Jun

7 Jun

7 Jun

30 Apr

19 Jul

30 Apr- 10 Aug

Western Meadowlark

24 May, 2 Jun, 7 Jun

4 Jun

4 Jun

30 Apr

14 Jul

30 Apr-07 Aug

Savannah SpaiTOws and Western Meadowlarks;
the remaining species were unimodal. All CID
frequencies were positively skewed towards early-
.season breeding. On average, Sprague’s Pipits
were the earliest to nest (median date = 25 May)
and Grasshopper Sparrows were the latest (medi-
an date = 16 Jun) (Table 2). All species ended
clutch initiation by the end of July. The earliest
fledging date observed was 26 May (Chestnut-
collared Longspur) and the latest was 25 August
(Sprague’s Pipit). Variation in mean CID across
years was low for Chestnut-collared Longspurs
(SD = 2.9 days) and higher for Savannah
Sparrows (SD = 5.6 days). The trend in CID
across years was extremely weak for either
species {\b\ < 0.20 d, IH < 0.12).

Variability (CV < 0.18) in incubation and
nestling phase duration was low, regardless of
species (Table 3). Mean clutch size across all

years ranged from 4.1 eggs (Chestnut-collared
Longspurs) to 4.6 eggs (Sprague’s Pipits). Mean
clutch size did not vary considerably across years
for either Chestnut-collared Longspurs (CV =
0.04) or Savannah Sparrows (CV = 0.06). Clutch
size tended to decrease over the course of a
season, regardless of bird species, although
Sprague’s Pipits and, to a lesser extent. Chest-
nut-collared Longspurs had increased clutch sizes
during the first month of the breeding season
(Fig. 3). Hatching rate (% of total eggs laid that
hatched) across all years ranged from 88% for
Chestnut-collared Longspurs and Grasshopper
Sparrows to 78% for Savannah Sparrows. Hatch-
ing rates were similar for Sprague’s Pipits (85%),
Baird’s Sparrows (84%), and Western Meadow-
larks (82%).

Mean number of nestlings across all years
ranged between 3.5 for Savannah SpaiTows to 4.0

TABLE 3. Nest stage duration and reproductive parameters, by songbird species at Bowdoin National Wildlife Refuge.
Montana, 1997-2007. Values represent mean ± CV (n; range). Fledged/nest represents the number of young fledged across
all nests, regardless of nesting outcome.

Reproductive parameters

Nest stage (days)

Species

Incubation

Nestling

Clutch size

No. of nestlings

No. Hedged/
.successful nest

Fledged/nest

Sprague’s Pipit

12.2 ± 0.12

12.9 ± 0.18

4.6 ±0.17

4.0 ± 0.26

3.4 ± 0.35

1 .30 ± 1 .07

(85; 7-15)

(19; 9-17)

(129; 1-6)

(97; 1-6)

(49; 1-6)

Savannah Sparrow

1 1.6 ± 0.1 1

lO.I ± 0.16

4.3 ± 0.23

3.5 ± 0.37

3.2 ± 0.41

1.23 ± 0.94

(184; 8-15)

(73; 7-14)

(260; 1-6)

(191; 1-6)

(101; 1-6)

Grasshopper

10.9 ± 0.14

9.7 ± 0.17

4.3 ± 0.24

3.9 ± 0.28

3.6 ± 0.33

1.89 ± 1.56

Sparrow

(86; 7-13)

(26; 7-12)

(123; 0^'-6)

(95; 1-6)

(64; 1-5)

Baird’s Sparrow

1 1.0 ± 0.10

9.6 ± 0.1 1

4.3 ±0.19

3.7 ± 0.31

3.4 ± 0.36

1.52 ± 1.31

(62; 7-14)

(27; 8-11)

(90; 2-6)

(70; 1-5)

(40; 1-5)

Chestnut-collared

10.9 ± 0.12

11.1 ±0.16

4. 1 ± 0.20

3.6 ± 0.27

3.4 ± 0.32

1 .50 ± 1 .37

Longspur

(567; 7-15)

(185; 7-15)

(770; 1-7)

(627; 0“-6)

(342; 0^'-5)

Western

12.9 ± 0.12

12.7 ±0.10

4.4 ± 0.30

3.8 ± 0.40

3.2 ± 0.47

1.24 ± 0.82

Meadowlark

(78; 8-15)

(31; 9-16)

(119; 0“-7)

(89; 1-6)

(46; 1-5)

‘‘ Brown-headed Cowbird (Molothrus ater) only.

460

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3, September 2010

(D

N

tfl

SZ

o

o

c

cc

CD

6.5
6.0

5.5

5.0 -

4.5 -

4.0

3.5 -

3.0 -

2.5

— ♦ Sprague's Pipit

— ■ — Savannah Sparrow

ii Grasshopper Sparrow

— - X - - Baird's Sparrow
— X — Chestnut-collared Longspur
— • — Vttestern M eado \Mark

1-15

May

16-31

May

1-15

Jun

16-30

Jun

1-15 16-31
Jul Jul

Date

FIG. 3. Mean clutch size plotted against date by species across all years (1997-2007), Bowdoin National Wildlife
Refuge, Montana. Mean clutch size generally decreased for each species as the nesting season progres.sed.

for Sprague’s Pipits (Table 3). Grasshopper Spar-
rows had the highest fledging rate ( 1 .9) across all
nests in all years while Savannah Sparrows and
Western Meadowlarks had the lowest (1.2)
(Table 3). Variation in mean fledglings/nest
across years was low for Chestnut-collared
Longspurs (CV = 0.20) with a modest decreasing
trend across years (b = —0.05, r = —0.59). This
statistic was more variable for Savannah SpaiTows
(CV = 0.45), but with no discernable trend (r =
-0.09).

Nest Success. — DSR across all species and
years was estimated at 0.947 (SE = 0.002). There
was little statistical support for species-specific
variation in DSR (Table 4). Model-averaged
estimates for each species (Table 5) indicated
little variation in DSR among species. However,
variation in DSR across years was evident,
suggesting a decline during the study period (h
= —0.002, r = —0.54) (Fig. 4). There was little

TABLE 4. Model set exploring the relationship
between bird species (6 levels) and year (1997-2007) at
Bowdoin National Wildlife Refuge, Montana on nest daily
survival rate (DSR) using Akaike’s Information Criterion,
corrected for small sample sizes (AIC, ). Delta AIC,. (AlC,.
model i - AIC, minimum). Akaike weight (w,), and number
of parameters (K) are included for each model.

Model A AIC, vv, K

Year

().()

0.66

1 1

Specie.s -f year

1.-3

0.3 1

16

Intercept

6.8

0.02

1

.Specie.s

10.7

0.00

6

Species*ycar

10.7

0.00

63

statistical support for an interaction between
species and year (Table 4).

Nest depredation accounted for 82% of all
known nest failures with most (71%) during the
nestling stage. Desertion (abandonment) was
judged to be the cause for 18% of all known nest
failures. Cause of abandonment was discernable
in only 27% of these cases; 25% {n = 38) were
attributed to severe weather (e.g., heavy rain,
hailstorms), 1% (n = 2) to observer activities, and
1% (« = 2) to Brown-headed Cowbird brood
parasitism. Brown-headed Cowbird parasitism
rates ranged from 2% for Chestnut-collared
Longspurs to 26% for Western Meadowlarks
(Table 6).

DISCUSSION

Songbird community structure based on nesting
density was relatively stable across the 1 1-year
period of this study. Chestnut-collared Longspurs
were clearly the dominant nesting species every
year. Savannah Span'ows were the second most
abundant bird species during all but 2 years, while
Baird's Sparrows were least abundant among
focal species during all but 2 years. Few to no
nests of shortgrass prairie obligates, such as
Horned Lark or McCown’s Longspur {Calcarius
mccownii) were located. Chestnut-collared Long-
spurs generally breed in drier, shortgrass to short-
mixed-grass prairie, while Sprague’s Pipits and
Baird’s Sparrows range across the breadth of
mixed-grass prairie grassland types (Knopf 1996).
Our assemblage of breeding songbirds, in addition
to plant community composition and structure
(Dieni and Jones 2003), classify our study area as

Jones el at. • NESTING OF GRASSLAND SONGBIRDS

461

TABLE 5. Estimated nest DSR for six bird species across all years at Bowdoin National Wildlife Refuge, Montana,
1997-2007. Standard errors (SE), lower (LCI) and upper (UCI) 95% confidence limits, and number of nests (/?) are
included. Nest success equals DSR raised to the mean number of days of the nesting period (incubation and nestling stages
combined) for each species.

Species

DSR

SE

LCI

UCI

n

Nesting period

Nest success

Sprague's Pipit

0.947

0.003

0.942

0.952

128

25.1

0.27

Savannah Sparrow

0.946

0.004

0.938

0.953

255

21.7

0.25

Grasshopper Span'ow

0.948

0.004

0.940

0.955

121

20.6

0.39

Baird’s Sparrow

0.947

0.003

0.940

0.953

90

20.6

0.32

Chestnut-collared Longspur

0.947

0.002

0.943

0.951

759

22.0

0.31

Western Meadowlark

0.947

0.003

0.941

0.952

118

25.6

0.25

Totals

0.947

0.002

0.943

0.950

1,471

22.3

0.30

xeric short-mixed-grass prairie (Knopf 1996,
Askins et al. 2007).

Breeding Bird Survey data indicate long-term
(1966-2007) declines in rangewide populations
for all our focal bird species (Sauer et al. 2008). In
particular, rangewide populations of two mixed-
grass endemics, Sprague’s Pipit (-3.9, P = 0.00,
n = 169) and Baird’s Sparrow ( — 3.4, P = 0.01, n
= 141), as well as Chestnut-collared Longspur
(—2.8, P = 0.00, n = 160) have exhibited steep
declines (Sauer et al. 2008). We recorded general
declines in nest density for most of our study
species on undisturbed study sites, although no
discemable trend was observed for Sprague’s
Pipit or Grasshopper Sparrow. The decline in
nesting density of Baird’s Sparrow may be a
function of drier than normal climatic conditions
that prevailed over most of the study period
(National Climatic Data Center 2007). Our study
area possessed none of the habitat conditions that

aie commonly the result when the northern
mixed-grass prairie is left idle (Madden et al.
1999). The prevalence of invasive plant species
and woody-vegetation encroachment remained
low, while forbs were abundant during the course
of the study (S. L. Jones, unpubl. data). Moreover,
xeric mixed-grass prairie historically had lengthy
periods without grazing, and natural burn fre-
quencies that often ranged between 8 and 26 years
(Askins et al. 2007).

We used nest density as an index to bird
abundance to gauge community structure and
trend, but nest density clearly does not equate
directly to bird or even territorial density. Some
species we studied have been documented to have
multiple broods in a season. Our population was
marked for only a few species and years (Jones et
al. 2007); it was not possible in most cases to
distinguish single broods by different individuals
from multiple broods after a successful nest or re-

0.98 -

„ 0.97
CE
C/3

Q 0.96
B

2 0.95

CTJ

1 0.94

i_

3
CO

^ 0.93
(0

2 0.92

CO
OJ

^ 0.91

0.90 - T - — T r r- , ^ , ,

1997 1999 2001 2003 2005 2007

Year

FIG. 4. Annual variation in daily survival rates of songbird nests (DSR ± SE) at Bowdoin National Wildlife Refuge,
Montana. There was considerable annual variation in DSR (annual mean = 0.946 ± 0.01 1 CV). A declining trend in DSR
was evident (h = -0.002, r = -0.54).

462

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3, September 2010

TABLE 6. Nesting outcome and parasitism rates for each bird species across all years, 1997-2007 at Bowdoin National
Wildlife Refuge, Montana. Most monitored nesting attempts were unsuccessful with the majority of failures attributed to
nest predation. Weather events (e.g., hail) were collectively the leading known cause of nest desertion. Parasitism by
Brown-headed Cowbirds resulted in only a few nest failures.

Brown-headed Cowbird

parasitism Known causes of abandonment («)

Species

Successful

Depredated

Abandoned

Unknown

n

Relative

frequency

Total cowbirds
fledged

Weather

Observer

disturbance

Parasitism

Sprague’s Pipit

0.37

0.57

0.06

0.00

128

0.02

2

1

0

0

Savannah Sparrow

0.40

0.43

0.17

0.01

260

0.13

17

15

1

1

Grasshopper

Sparrow

0.52

0.35

0.12

0.01

123

0.04

2

3

0

0

Baird’s Sparrow

0.43

0.42

0.16

0.00

89

0.04

0

4

0

0

Chestnut-collared

Longspur

0.44

0.48

0.07

0.01

770

0.02

3

11

0

1

Western

Meadowlark

0.40

0.45

0.14

0.00

119

0.26

18

4

1

0

n

637

688

151

13

1,489

91

42

38

2

2

nesting after nest failure. However, Savannah
Sparrows, Chestnut-collared Longspurs, and
Western Meadowlarks all had multimodal CID
distributions suggesting at least some community
members may have had several clutches per year.
None of the species we studied has been
documented with more than two successful broods
per year (Wheelwright and Rising 1993, Lanyon
1994, Vickery 1996, Hill and Gould 1997).
Sprague’s Pipit, and Baird’s and Grasshopper
sparrows had only single frequency peaks in our
CID data, which suggests a tendency towards a
single brood per season. This contrasted with
Vickery (1996), who reported Grasshopper Spar-
rows in general have a protracted breeding season,
often producing >2 broods annually, even in the
northern portion of their range. Sprague’s Pipit
and Baird’s Sparrow have only rarely been
documented to have multiple broods in a season
(Robbins and Dale 1999, Green et al. 2002).
Second broods after successfully fledging young
have been documented twice for Sprague’s Pipit
(Sutter 1996, Davis 2009). Re-nesting after nest
failure has been rarely documented for this
species (Sutter 1996, Davis 2009), as has
polygyny (/? = 1) (Dohms and Davis 2009).
Baird’s Sparrow in southwest Manitoba (Davis
and Sealy 1998) and in our study had a .second,
smaller nest initiation peak in mid to late June,
suggesting .second broods or re-nesting attempts;
.second broods have only been documented twice
(Davis and Sealy 1998; S. L. Jones, unpubl. data).

Mean CID for Savannah Sparrows and Chest-
nut-collared Longspurs varied little across years.

Mean clutch size also had little variation among
seasons; however, a within-season declining
pattern was evident for this parameter for all
species. Similar patterns have been documented
elsewhere for these bird species (Vickery 1996,
Davis and Sealy 1998, Winter 1999, Davis 2003).
Sprague’s Pipits were a notable exception in our
study, where clutch size generally increased
during the first month of the breeding season. A
similar pattern was also reported for Sprague’s
Pipit in Saskatchewan (Davis 2003). Clutch size
in our study linearly decreased over the course of
the breeding season for Western Meadowlarks,
which contrasted with a study in Manitoba, where
Western Meadowlarks had peak clutch size during
mid-season (Lanyon 1994).

Brown-headed Cowbird parasitism rates, ex-
cept for Western Meadowlark, were relatively low
in our study (2-13%). Much higher rates were
reported for our focal species in southwest
Manitoba: Sprague’s Pipit = 18%, Savannah
Sparrow = 32%, Grasshopper Sparrow = 29%,
Baird’s Sparrow = 37%, and Chestnut-collared
Longspur = 14% (Davis and Sealy 2000). Our
rate for Western Meadowlark (26%) was low
compared to rates reported in Manitoba (43%)
(Davis and Sealy 2000), and North Dakota (47%)
(Koford et al. 2000), but were consistent with
reported rates in Saskatchewan (25%) (Davis
2003). Chestnut-collared Longspurs and Spra-
gue’s Pipits had the lowest parasitism rates in
our study, which is consistent with findings in
other studies (Hill and Gould 1997, Davis and
Sealy 2000, Davis 2003). Several factors may be

Jones el al. • NESTING OF GRASSLAND SONGBIRDS

463

responsible for the relatively low levels of
parasitism found in our study area (i.e., large
patches of continuous grassland, a landscape with
few potential perches, relatively little shrub and
tree cover, and no cattle grazing). All of these
factors directly enhance Brown-headed Cowbird
parasitism rates (Davis and Sealy 2000, Koford et
al. 2000).

Nest depredation is commonly the single most
important factor affecting nesting success in cup-
nesting passerines (Ricklefs 1969); depredation
rates have been generally high throughout the
mixed-grass prairie (Vickery et al. 1992; Winter
1999; Davis 2003, 2005; Winter et al. 2005). Nest
predation accounted for the majority of observed
nest failures in our study, mainly during the
nestling stage. This is a common pattern among
passerines; greater parental activity at the nest, as
well as greater olfactory cues, noise, and move-
ments by nestlings are thought to increase the
likelihood of nest detection by predators (Halupka
1998).

Overall nest success for Sprague’s Pipits was
estimated to be much higher in southwest
Manitoba (47%, Davis and Sealy 1998) than
estimated in our study area (27%). Grasshopper
Sparrow and Western Meadowlark nesting suc-
cess rates were comparable to those reported by
Lanyon (1994) and Vickery (1996). Our estimate
of nest success (32%) for Baird’s Sparrow was
higher than reported in Saskatchewan (26%,
Davis 2003), but lower than in Manitoba (54%,
Davis and Sealy 1998). Nest success for Chestnut-
collared Longspurs was higher in Alberta (56%,
Hill and Gould 1997) than estimated for our study
area (44%; Table 6).

DSR estimates did not differ markedly among
bird species, suggesting that factors such as
weather and the predator community were affect-
ing all bird species equally. Therefore, tho.se
elements affecting nest success that vary with year
(i.e., predator community or weather) probably
influenced the members of this grassland songbird
community in a similar manner in a given year.

ACKNOWLEDGMENTS

This project was funded by the U.S. Fish and Wildlife
Service, Nongame Migratory Bird Program, Region 6. Field
support and resources were provided by J. E. Comely, C. R.
Luna, D. M. Prellwitz. and the staff at Bowdoin National
Wildlife Refuge. We thank the many dedicated field
assistants who worked on this project over the years: Brice
Adams, A. C. Araya, B. S. Atchley, R. R. Conover, Michael
Friedrich, Jeanne Hammond, D. I. Hedegaard, Jessica Hinz,

B. G. Larson, Rebecca Lind, L. B. Lingohr, S. N. Luttich,
Kevin Payne, Heather Sauls, Caroline Stahala, Vicki
Trabold, P. N. Wiederrick, and J. K. Wood. We are also
grateful to M, T. Green, G. R. Guepel, and J. R. Ruth for
insights that contributed to the design and conduct of this
study. D. J. Ingold, D. H. Johnson, and an anonymous
reviewer provided edits and comments that substantially
improved this paper. We thank Mike Artmann for the map
in Fig. 1.

LITERATURE CITED

Ahlering, M. A. AND D. H. Johnson. 2006. Conspecific
attraction in a grassland bird, the Baird’s Sparrow.
Journal of Field Ornithology 77:365-371.

Askins, R. A., F. Chavez-Ramirez, B. C. Dale, C. A,
Haas, J, R. Herkert, F. L. Knopf, and P. D.
Vickery. 2007. Conservation of grassland birds in
North America: understanding ecological processes in
different regions. Ornithological Monographs 64: 1^6.

Basore, N. S., L. B. Best, and J. B. Wooley Jr. 1986.
Birds nesting in Iowa no-tillage and tilled cropland.
Journal of Wildlife Management 50:9-28.

Burnham, K. P. and D. R. Anderson. 2002. Model
selection and multi-model inference: a practical
information-theoretic approach. Second Edition.
Springer-Verlag, New York, USA.

Clark, R. G. and D. Shutler. 1999. Avian habitat
selection: pattern from process in nest-site use by
ducks. Ecology 80:272-287.

Davis, S. K. 2003. Nesting ecology of mixed-grass prairie
songbirds in southern Saskatchewan. Wilson Bulletin
115:119-130.

Davis, S. K. 2005. Nest-site selection patterns and the
influence of vegetation on nest survival of mixed-grass
prairie passerines. Condor 107:605-615.

Davis, S. K. 2009. Re-nesting intervals and length of
incubation and nestling periods of Sprague’s Pipit.
Journal of Field Ornithology 80:265-269.

Davis, S. K. and S. G. Sealy. 1998. Nesting biology of the
Baird’s Sparrow in southwestern Manitoba. Wilson
Bulletin 110:262-270.

Davis, S. K. and S. G. Sealy. 2000. Cowbird parasitism
and nest predation in fragmented grasslands of
southwestern Manitoba. Pages 220-228 in Ecology
and management of cowbirds and their hosts: .studies
in the conservation of North American passerine birds
(J. N. M. Smith, T. L. Cook. S. 1. Rothstein, S. K.
Robinson, and S. G. Sealy, Editors). University of
Texas Press, Austin, USA.

Dieni, j. S. and S. L. Jones. 2003. Grassland songbird nest
site selection patterns in north-central Montana.
Wilson Bulletin 115:388-396.

Dinsmore. S. j., G. C. White, and F. L. Knopf. 2002.
Advanced techniques for modeling avian nest survival.
Ecology 83: 3476-3488.

Dohms. K. M. and S. K. Davis. 2009. Polygyny and male
parental care by Sprague’s Pipit. Wikson Journal of
Ornithology 121:826-830.

464

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3. September 2010

Green, M. T., P. E. Lowther, S. L. Jones, S. K. Davis,
AND B. C. Dale. 2002. Baird’s Sparrow (Ammodra-
miis bairdii). The birds of North America. Number 638.

Halupka, K. 1998. Vocal begging by nestlings and
vulnerability to nest predation in Meadow Pipits
Anthiis pratensis: to what extent do predation costs
of begging exist? Ibis 140; 144—149.

Higgins, K. F. 1984. Lightning fires in North Dakota
grasslands and in pine-savanna lands of South Dakota
and Montana. Journal of Range Management 37:100-
103.

Hill, D. P. and L. K. Gould. 1997. Chestnut-collared
Longspur {Ccdcarius ornatiis). The birds of North
America. Number 288.

Houston, C. S. and J. K. Schmutz. 1999. Changes in bird
populations on Canadian grasslands. Studies in Avian
Biology 19:87-94.

iGL, L. D. AND D. H. Johnson. 1999. Le Conte’s Sparrows
breeding in Conservation Re.serve Program fields;
precipitation and patterns of population change.
Studies in Avian Biology 19:178-186.

Jones, S. L., J. S. Dieni, M. T. Green, and P. J. Gouse.
2007. Annual return rates in breeding grassland
songbirds. Wilson Journal of Ornithology 1 19:89-94.

JoNGSOMJiT, D., S. L. Jones, T. Gardali, G. R. Geupel,
AND P. J. Gouse. 2007. A guide to nestling
development and aging in altricial passerines. USDl,
Fish and Wildlife Service, Biological Technical
Publication FWS/BTP-R6008-2007. Washington,
D.C., USA.

Knopf, F. L. 1994. Avian assemblages on altered
grasslands. Studies in Avian Biology 15:247-257.

Knopf, F. L. 1996. Prairie legacies-birds. Pages 135-148 in
Prairie conservation: preserving North America’s most
endangered ecosystem (F. B. Sam.son and F. L. Knopf,
Editors). Island Press, Covelo, California, USA.

Koford, R. R., B. S. Bowen, J. T. Lokemoen, and A. D.
Kruse. 2000. Cowbird parasitism in grassland and
cropland in North Dakota. Pages 229-235 in Ecology
and management of cowbirds and their hosts: studies
in the conservation of North American passerine birds
(J. N. M. Smith, T. L. Cook, S. 1. Rothstein, S. K.
Robinson, and S. G. Sealy. Editors). University of
Texas Press, Austin, USA.

Lanyon, W. E. 1994. We.stern Meadowlark (Sturnella
neglecta). The birds of North America. Number 104.

Lokemoen, J. T. and R. R. Koford. 1996. Using candlers
to determine the incubation stage of passerine eggs.
Journal of Field Ornithology 67:660-668.

Madden, E. M., A. J. Han.sen, and R. K. Murphy. 1999.
InlJuence of pre.scribed fire history on habitat and
abundance of pas, serine birds in northern mixed-grass
prairie. Canadian Field-Naturalist 1 13:627-640.

Martin, T. E. and G. R. Geupel. 1993. Nest-monitoring
plots: methods for locating nests and monitoring
success. Journal of Field Ornithology 64:507-519.

National Climatic Data Center. 2007. Monthly surface
data, Malta, Montana. National Oceanic and Atmo-
spheric Administration, Asheville, North Carolina,
USA. http://lwfncdc.noaa.gov/oa/ncdc.html.

Ricklefs, R. E. 1969. An analysis of nesting mortality
in birds. Smithsonian Contributions in Zoology
Number 9. Washington, D.C., USA.

Robbins, M. B. and B. C. Dale. 1999. Sprague’s Pipit
(Antbus spragiieii). The birds of North America.
Number 439.

Rotella, j. j. 2007. Nest survival models. Pages 17.1-
17.22 in Program MARK: “a gentle introduction” (E.
G. Cooch and G. C. White, Editors). Sixth Edition.
www.phidot.org/software/mark/docs/book/.

Rotella, J. J., S. J. Dinsmore, and T. L. Shaffer. 2004.
Modeling nest-survival data: a comparison of recently
developed methods that can be implemented in MARK
and SAS. Animal Biodiversity and Conservation 27:
187-205.

Sauer, J. R., J. E. Hines, and J. Fallon. 2008. The North
American breeding bird survey, results and analysis
1966-2007. Version 5.15. USGS, Patuxent Wildlife
Research Center, Laurel, Maryland, USA. www.mbr-
pwrc . usgs .go v/bbs/.

Sutter, G. C. 1996. Renesting intervals in Sprague’s Pipit.
Anthns spragueii. Canadian Field-Naturalist 1 10:694-
697.

Vickery, P. D. 1996. Grasshopper Sparrow (Ammodramus
savanna rnin). The birds of North America. Number
239.

Vickery, P. D., M. L. Hunter, and J. V. Wells Jr. 1992.
Evidence of incidental nest predation and its effects on
nests of threatened gra.ssland birds. Oikos 63:281-288.

White, G. C. 2008. Program MARK. Mark and recapture
survival rate estimation (Version 5.1). www.wamercnr.
colostate.edu/~gwhite/mark/.

Winter, M. 1999. Relationship of fire history to territory
size, breeding density, and habitat of Baird’s Sparrow
in North Dakota. Studies in Avian Biology 19:171-
177.

Winter, M., D. H. Johnson, and J. A. Shaffer. 2005.
Variability in vegetation effects on density and nesting
success of grassland birds. Journal of Wildlife
Management 69:185-197.

Wheelwright, N. T. and J. D. Rising. 1993. Savannah
Sparrow (Pas.sercidiis sandwichensis). The birds of
North America. Number 45.

Zar, j. H. 1999. Biostatistical analysis. Fourth Edition.
Prentice-Hall, Upper Saddle River, New Jersey, USA.

The Wilson Journal ofOrnithology 1 22(3):465-474, 2010

MOUNTAIN BIKING TRAIL USE AFFECTS REPRODUCTIVE SUCCESS
OF NESTING GOEDEN-CHEEKED WARBLERS

CRAIG A. DAVIS,'"' DAVID M. LESLIE JR.,' W. DAVID WALTER,' AND ALLEN E. GRABER'*

ABSTRACT. — We evaluated toraging and nesting behavior, territory size, and nest success of Golden-cheeked Warblers
(Dendroica clirysoparia), a federally endangered songbird, relative to mountain biking trail use. We conducted our study at
two mountain biking sites and two control sites at Fort Hood Military Base and in Austin, Texas, in spring 2002 and 2003.
Teiritories ot male Golden-cheeked Warblers in biking sites (2.2 ha) were >1.5 times as large as those in non-biking sites
(1.4 ha). Mayfield nest success in biking sites (n = 33) was 35% compared to 70% in non-biking sites {n = 22). Nest
abandonment was three times greater in biking areas (15%) than non-biking areas (5%). Seven nests were depredated in
biking sites, but only two nests were depredated in non-biking sites. Texas rat snakes (EUiphe ohsoleta) were the most
frequent nest predator at biking sites, accounting for 71% of the predations. We conducted behavioral observations of male
Golden-cheeked Warblers in biking (n = 139) and non-biking (/t = 204) sites. Males spent similar amounts of time in
diurnal behaviors in biking and non-biking sites. We used video-camera systems to record female nesting behaviors at 17
nests in biking sites and 15 nests in non-biking sites. Nesting behaviors of females did not differ between biking and non-
biking sites. The cumulative effect of disturbance from mountain biking trail use on Golden-cheeked Warbler foraging and
nesting behavior appears to be minimal, but fragmentation and alteration of habitat by mountain biking trails may reduce
quality of nesting habitat for Golden-cheeked Warblers. Received 23 November 2009. Accepted 27 March 2010.

Mountain biking is one of the fastest-growing
outdoor recreational activities in the United
States, experiencing substantial growth in the last
20 years (Mosedale 2002, Pike Masteralexis et al.
2008). This increase has led to greater demands
for quality outdoor experiences on trail networks
in areas that often contain fragile environments
(Flather and Cordell 1995). Land managers have
raised concerns about the impacts of mountain
biking activities on natural resources (Chavez et
al. 1993, Symmonds et al. 2000). These concerns
led to restrictions in mountain biking activities at
all federal wilderness areas and many other public
lands in the United States despite little scientific
insight on their impacts to wildlife (Taylor and
Knight 2003).

The Golden-cheeked Warbler {Dendroica chry-
soparia) is a federally endangered species that
breeds exclusively in the mature juniper (Jimi-
perus ashei)-o'dk (Quercus spp.) woodlands of

' Department of Natural Resource Ecology and Manage-
ment. Oklahoma State University, Stillwater, OK 74078,
USA.

'U.S. Geological Survey, Oklahoma Cooperative Fish
and Wildlife Research Unit. Department of Natural
Resource Ecology and Management, Oklahoma State
University, Stillwater, OK 74078, USA.

■’Colorado Cooperative Fish and Wildlife Research Unit,
201 JVK Wagar Building, 1484 Campus Delivery,
Colorado State University, Fort Collins, CO 80523, USA.

■"SWCA Environmental Consultants. 130 Rock Point
Drive, Suite A. Durango, CO 81301, USA.

’Corresponding author;
e-mail: craig.a.davis@okstate.edu

central Texas (Ladd and Gass 1999). Some of its
breeding range is used by mountain bikers (e.g..
Dinosaur Valley State Park in Somervell County,
Lost Maples State Natural Area in Bandera
County, Belton Lake Outdoor Recreation Area
[BLORA] in Bell County, and Emma Long
Metropolitan Park [ELMP] in Travis County).
Increased levels of predation on passerine nests
near mountain biking trails have been documented
(Miller et al. 1998, Miller and Hobbs 2000).
Mountain biking also may alter distribution and
behaviors of Golden-cheeked Warblers. In partic-
ular, disturbance from mountain biking may
lessen the time and energy used for fitness-
enhancing activities including feeding, nest atten-
tiveness, mate attraction, and ten'itory defense
(Knight and Gutzwiller 1995).

Disturbance from mountain biking activities
may pose a threat to successful breeding of
Golden-cheeked Warblers; a Biological Opinion
issued by the U.S. Fish and Wildlife Service
(USDI 1993) and a regional “take” permit issued
by the U.S. Fish and Wildlife Service under the
umbrella oi the Balcones Canyonlands Con.serva-
tion Plan recommended eliminating and restrict-
ing mountain biking activities from Golden-
cheeked Warbler habitat in areas of the Fort
Hood Military Reservation and Travis County,
respectively. Those recommendations were con-
troversial, and further research of the impacts of
mountain biking on Golden-cheeked Warblers
was needed. Our objectives were to examine
effects of mountain biking trail use on foraging

465

466

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 3. September 2010

and nesting behavior, territory size, and nest
success of Golden-cheeked Warblers.

METHODS

Study Area. — We conducted this study at two
mountain biking sites and two control sites from
March to June, 2002 and 2003. The mountain
biking sites were at BLORA in Fort Hood
Military Reservation (31° 06' N, 97° 31' W) and
ELMP (30° 19' N, 97° 50' W) in Balcones
Canyonlands Preserve (BCP) in Austin, Texas.
Control sites were at Training Area 13A (13A) on
Fort Hood Military Base (30° 10' N, 97° 45' W)
and Forest Ridge Preserve (FRP; 30° 24' N, 97°
55' W) in BCP. Recreational activities were
restricted from both control areas, and military
activities were restricted on both sites at Fort
Hood. Mountain biking activities, as measured by
TrailMaster photoelectric trail counters placed
along trails, averaged 15 mountain bikers/day (5
bikers/day at BLORA and 17 bikers/day at
ELMP) and ranged from zero to 30 bikers/day at
biking sites (Davis and Leslie 2008). Mountain
bikers accounted for >80% of the trail users at
biking sites. Other users recorded at biking sites
included motorized bikers (<2%), hikers and
joggers (2%), and researchers (11%).

Fort Hood and BCP are within the Crosstimbers
and Southern Tallgrass Prairie and Edwards
Plateau ecoregions, respectively. Fort Hood con-
tains —21,850 ha of Golden-cheeked Warbler
habitat and supports the largest known population
of Golden-cheeked Warblers under one manage-
ment authority (USDI 1993); the BCP contains
— 10,600 ha of preserve properties with 5,300 ha
of Golden-cheeked Warbler habitat (Reidy et al.
2008). Mountain biking sites ranged in size from
116 (ELMP) to 192 ha (BLORA), and control
sites ranged in size from 190 (FRP) to 198 ha
(13A). All sites were woodlands associated with
steep limestone canyons and plateaus dominated
by Ashe Juniper and a variety of oaks.

Territory Size. — We used spot-mapping (Ralph
et al. 1993) to delineate territories for color-
banded and unbanded males at each site from 10
March to 24 June 2002 and 2003. Males were
color banded from previous studies (at the
BLORA site) and during our study. We used
conspecific playback to capture males with 6-m
mist nets. We banded males with unique color
combinations of two or three plastic color bands
and a numbered aluminum U.S. Geological
Survey band. Individual locations of males were

identified by standing directly under the location
where a male was observed and using a handheld
global positioning system unit (GPS; Garmin GPS
III-I-, Garmin International, Olathe, KS, USA) to
record Universal Transverse Mercator (UTM)
coordinates. We recorded one location of each
individual during a single visit, but attempted to
obtain locations at different time periods through-
out the day. We later plotted locations of
individuals on 1:4,000 digital orthoquad maps
with 1 .5-m resolution imagery and a 500 X 500-m
grid overlay to distinguish territories throughout
each year. We distinguished individual territories
by identifying color bands. We relied on counter-
singing by unbanded males that occupied adjacent
territories as well as a general familiarity of an
individual’s territory as the season progressed to
facilitate delineation of territories. We attempted
to obtain >30 locations per individual male
(Laundre and Keller 1984, Bibby and Burgess
1992, Seaman et al. 1999). We used the Home
Range extension in ArcView and 1 ,000 iterations
of the bootstrap procedure to calculate the
minimum number of male locations sufficient
for a reliable minimum convex polygon (MCP)
estimate (Rodgers and Carr 1998, ESRI 2000).
We calculated that 33 locations were sufficient for
estimating the MCP for males on the basis of the
bootstrap procedure. We estimated MCP using the
Home Range extension in ArcView for those
males with a minimum of 33 locations.

We characterized land cover types within each
study site and each teiritory using the National
Land Cover Database (NLCD) of 2001 to evaluate
whether teiTitories differed by habitat composi-
tion. Land-cover categories in the NLCD were
reclassified into eight broad categories (water,
bare ground, deciduous forest, evergreen forest,
mixed forest, shrub/scrub, grassland, and wetland)
to standardize our analysis across the four study
sites. The proportions of landcover types from
NLCD (i.e., forested, grassland) within each
bird’s home range were extracted in ArcView.
We walked all trails within each study area using
a GPS unit (Trimble Geoexplorer 3, Trimble
Navigation Ltd., Sunnyvale, CA, USA) to delin-
eate trails. We downloaded trail locations and
created a Geographic Information System (GTS)
cover layer of trails for each study site using
ArcView.

Time Activity Budgets of Territorial Males. —
We assessed the impact of mountain biking
activity on foraging activities by conducting

Davis et at. • MOUNTAIN BIKING AND GOLDEN-CHEEKED WARBLERS

467

behavioral observations of males on territories at
each study site from March to June, 2002 and 2003.
We conducted observations throughout the day.
We minimized sampling bias due to diurnal
variation in foraging behavior by collecting
behavioral data equally during three time periods:
early, sunrise to 1100 hrs; mid-day, 1101 to 1500
hrs; and late, 1501 hrs to sunset. Males were
located by listening for a Type A- or B-song. We
predominantly observed males using binoculars,
but, in some cases, observed males directly. We
attempted to observe each bird for 300 sec and
discarded observation sessions <30 sec in length.
We dictated behaviors into a handheld microcas-
sette tape-recorder. We avoided bias toward
conspicuous behaviors by not recording behaviors
until 5 sec after locating a bird (Wunderle and Latta
1998). We used focal individual sampling (Alt-
mann 1974, Martin and Bateson 1986) to collect
behavioral data. We avoided autocorrelation for
banded birds by not recording behavior from the
same individual more than once in a single day. We
did not collect behavioral observations during
inclement weather (rain or high wind).

We classified behaviors into nine categories
(Robinson and Holmes 1982, Remsen and Rob-
inson 1990): singing, preening, agonistic, mate
guarding, perching (resting), locomotion (flying
and hopping), foraging (searching, attacking, and
food handling), feeding fledgling, and time-out
(time when the bird was out-of-sight). The bird
was considered in “time-out” when a focal male
briefly disappeared behind vegetation during an
observation. The tape recorder continued to run
during the “time-out” period, and observations
resumed when the bird reappeared. We ended the
observation at the end of the last observation
before the “time-out” if the bird did not reappear
within the 300-sec period.

We further examined foraging activities in
2003 by recording the amount of time a male
was engaged in searching, attacking, and food-
handling. Search behaviors were considered to be
any movements leading up to locating and
acquiring food, and included hops (leg powered
movements <10 cm), short flights (<20 cm),
medium flights (20 cm to 2 m), and long flights
(>2 m). Attack maneuvers were any movements
directed at a food item or the substrate that
concealed the food item after it was located and
included gleaning, reaching, hanging, leaping,
sally-striking, sally-hovering, and flutter-chasing
(Remson and Robinson 1990).

Parental Behavior at Nests. — We assessed the
impact of mountain biking trail use on nesting and
parenting activities of Golden-cheeked Warblers
by video-recording nests using miniature video
camera systems (Fuhrman Diversified Inc., Sea-
brook, TX, USA). We relied on video-recordings
in 2002 provided by another researcher studying
nest predation of Golden-cheeked Warblers at Fort
Hood sites (M. M. Stake, University of Missouri,
Columbia, USA); we collected nesting and parent-
ing behavior in 2003 using eight miniature video
camera systems at the Fort Hood and Austin sites.
Protocols for video-recording were similar for
2002 and 2003. Each system consisted of a 32- by
32- by 60-mm camera with a small lens and six
infra-red diodes (950 nm) that allowed filming at
night, an articulating arm that supported the
camera, and a clamp at the base of the arm for
attachment to a substrate. A video recorder,
allowing 24 hrs of footage (~7 frames/sec) on a
120-min VHS cassette tape, was connected to the
camera with a 20-m cable and powered by a 12-V
rechargeable battery. We used a hand-held monitor
connected to the cable for positioning the lens at
the nest during set-up and connected to the base
set-up to check nest status each day without
approaching the nest. We changed tapes daily and
replaced batteries every other day.

We generally found nests opportunistically
during location of focal males. Most nests were
found by following an adult to the nest during
building and nestling stages. We set up cameras at
nests, if accessible, and if eameras were available;
nests found in the egg stage were prioritized for
cameras but nests in the nestling stage were used
as cameras became available. Cameras were
typically installed in the afternoon to reduce risk
of Brown-headed Cowbird (Molothrus ater)
parasitism. We only installed cameras after the
laying stage to reduce the risk of female
abandonment (Stake 2000, Stake and Cimprich
2003, Stake et al. 2004, Reidy et al. 2008) and
removed cameras if females did not resume
activity at the nest within I hr. We were only
able to monitor female activity at nests during
incubation and nestling stages. Nests were
monitored daily by video until nest fate was
known (e.g., fledged young, depredated, aban-
doned, failed due to weather).

We watched tapes using a multispeed video
player. Tapes for individual nests were viewed
until the nest was no longer active. We recorded
behaviors during the incubation stage, early

468

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. 3, September 2010

nestling stage, and late nestling stage. Behaviors
recorded included female incubating eggs, female
brooding nestlings, female or male provisioning
nestlings, female shading nestlings, female or
male perching and scanning at the nest, and male
feeding the female. Females brood recently
hatched nestlings for long periods during the first
3 days after hatching and continue brooding for 1-
2 days thereafter (Ladd and Gass 1999); thus, we
designated the early nestling stage as the first
5 days of the nestling stage. We recorded a variety
of behaviors, but focused on incubation, brooding
nestlings, and provisioning nestlings because
changes in these behaviors may affect nest
success (Gill 2007). Specifically, we focused on
percent time spent incubating, number of incuba-
tion boLits/day and length of incubation bouts
during the incubation period, and percent time
spent brooding and provisioning nestlings, num-
ber of brooding and provisioning bouts/day, and
length of brooding and feeding bouts during the
early and late nestling stage.

Nest Success. — We assessed effects of moun-
tain bike trail use on Golden-cheeked Warbler
nesting success by monitoring all nests located at
the study sites. Most of the nests (90%) were
monitored with video; we monitored nests without
cameras every 2-5 days until near fledging at
which time we visited daily to more accurately
monitor nests and correctly identify nest outcome.
We considered a nest to be successful if >1 host
young fledged. We considered a nest depredated if
a predator removed at least one egg or nestling
while the nest was active, and a nest abandoned if
the male and female were absent from the nest
after several days and the eggs or young were still
present. We were able to identify all predation
events to species-level from watching tapes.
Successful production of ^1 fledgling was
confirmed by locating fledglings being fed by a
parent (usually verified the day of or day after
fledging), or by recording fledging by video
surveillance.

Statistical Analyses.— We calculated area of
trails by multiplying the linear length of trails
within each MCP by 2.5 m, considered the
average width of trails (AEG. pers. obs.). We
used one-way analysis of variance (ANOVA) to
examine differences in territory size, amount of
trail area within each MCP, and habitat compo-
sition (% developed, % deciduous, % juniper, and
% grassland) between biking and non-biking
areas.

Some banded males had multiple foraging
observations within time periods, and we aver-
aged behaviors for those individuals within each
time period prior to conducting analyses. We
analyzed time-activity data using multivariate
analysis of variance (MANOVA) with a factorial
an'angement (SYSTAT 2004). Treatment (biking
and non-biking), time period (early, mid-day, and
late), and year (2002 and 2003) were independent
factors in the MANOVA. We used MANOVA
because the dependent variables (i.e., individual
behaviors) were not independent of each other
(i.e., the amount of time engaged in one behavior
influenced the amount of time engaged in other
behaviors). Wilks’ lambda (X.) was the test
criterion.

We used univariate analysis of variance (AN-
OVA) to examine differences in individual
behaviors between years if the overall MANOVA
was significant (Barker and Barker 1984). Behav-
ioral data met assumptions of normality and
homogeneity (Johnson and Wichern 1988, SY-
STAT 2004). We used a similar MANOVA with
factorial airangement to analyze foraging search
behavior data. We used Kruskal-Wallis tests,
because of our small sample size, to compare
food capture techniques (i.e., proportion of attack
modes used by each male during foraging) of
males between biking and non-biking sites.

There was considerable variability in amount of
time each nest was videotaped (which could
overemphasize the activities at some nests); thus,
we calculated parental behavior parameters on a
daily basis. We compared parental behavior
parameters using Kruskal-Wallis tests.

We used Mayfield’s (1961, 1975) method to
calculate daily nest survival rates (DNS) for
incubation and nestling stages, and nest success
for biking (incubation: 1 1 1 nest-days, nestling:
146 nest-days) and non-biking (incubation: 69
nest-days, nestling: 108 nest-days) sites. We
calculated standard errors for DNS following
Johnson (1979). All statistical tests were consid-
ered significant at P < 0.05.

RESULTS

Territory Size. — Territory size for male Golden-
cheeked Warblers averaged 1.5 times larger, in
biking than non-biking sites (Table 1). Territories
ranged from 0.48 to 7.27 ha in biking sites and
0.24 to 4.33 ha in non-biking sites. Trails
occupied twice the area within territories of
Golden-cheeked Warbler males in biking than

Davis el al. • MOUNTAIN BIKING AND GOLDEN-CHEEKED WARBLERS

469

TABLE 1 . Mean ± SE for minimum convex polygon (MCP) territory size (ha), trail area (ha) within MCPs, and habitat
composition (%) within MCPs for Golden-cheeked Warblers at biking and non-biking sites near Austin and on Fort Hood,
Texas, during spring 2()02-2()()3.

Variable

Biking III

= ,20)

Non-biking (« = 61)

F-

p

Territory size (MCP)

2.16 ±

0.19

1.43 ± 0.16

6.01

0.016

Trail area within MCP

0.068 ±

0.009

0.029 ± 0.004

9.77

0.002

Habitat composition

Developed

0.67 ±

0.42

0.00 ± 0.00

1.21

0.274

Deciduous

6.00 ±

1.50

3.48 ± 2.70

0.78

0.378

Juniper

92.00 ±

1.78

96.52 ± 2.70

2.05

0.155

Grassland

1.34 ±

0.83

0.00 ± 0.00

1.27

0.262

F and P values for analysis of variance comparisons between biking and non-biking sites: df = 1,89.

non-biking sites (Table 1), Habitat composition
of territories did not differ between sites
(Table 1),

Time Activity Budgets of Territorial Moles. —
We made 400 total observations (1-7 observa-
tions/male, 100-120 unique males) of male
Golden-cheeked Warblers during the 2 years of
study. Overall behavior did not differ between
biking and non-biking sites (Wilks’ T = 0,97, P =
0,265; Table 2) and time periods (Wilks’ X =
0,95, P = 0.365), but differed between years
(Wilks’ X = 0.87, P < 0.001). There were no
significant two- or three-way interactions (Wilks’
X > 0.93, P > 0.116). Singing and provisioning
nestlings were the only behaviors that differed
between years. Male Golden-cheeked Warblers
spent less time singing in 2002 than 2003 (8.3 vs.
16.0%) and more time provisioning nestlings in
2002 than 2003 (1.2 vs. 0.1%).

Prey search behaviors of male Golden-cheeked
Warblers did not differ between biking and non-
biking sites (Wilks’ X = 0.97, P = 0.677;
Table 3). The total number of search bouts during

each observation period also did not differ
between biking and non-biking sites (T|,93 =
0.03; P = 0.866; T ± SE = 26.2 ± 4.4 bouts vs.

27.2 ± 3.2 bouts). Male Golden-cheeked War-
blers spent similar amounts of time engaged in
attack techniques to capture prey at biking and
non-biking areas (Table 4).

Parental Behavior at Nests. — We video-record-
ed 32 Golden-cheeked Warbler nests (17 in biking
sites vs. 15 in non-biking sites) for 4,094.5
daytime hrs in 2002 and 2003 (incubation stage:

1.141.2 hrs for nests at biking sites and 828.2 hrs
for nests at non-biking sites; nestling stage:
1,109.7 hrs for nests at biking sites vs.
1,015.4 hrs for nests at non-biking sites). Average
recording time per nest was 110.6 ± 17.4 hrs
(range = 15.2-432.8 hrs). Golden-cheeked War-
blers spent similar amounts of time incubating
eggs, brooding nestlings, and provisioning nest-
lings in biking and non-biking sites (Table 5). The
number of bouts and length of the bout for each of
these behaviors also did not differ between biking
and non-biking sites (Table 5).

TABLE 2. Diurnal time activity budget (.v ± SE) for male Golden-cheeked Warblers at biking (n = 139) and non-
biking (/; = 204) locations near Austin and on Fort Hood, Texas during spring 2002-2003.

D.uy time (%)

2002 200.2

Behavior

Biking

Non-biking

Biking

Non-hiking

Perching

37.5

-h

4.9

26.8

4-

3.5

29.9

4-

2.6

31.4 ± 2.2

Singing

7.8

0.9

8.7

-h

0.9

15.6

4-

1.3

16.3 ± 1.2

Preening

10.2

H-

3.3

7.6

4-

2.4

5.9

4-

1.3

6.7 ± 1.3

Agonistic

0.0

-f-

0.0

0.2

4-

0.1

0.6

4-

0.5

0.2 ± 0.1

Mate guarding

0.2

0.2

0.0

4-

0.0

0.2

4-

0.2

0.0 ± 0.0

Locomotion

38.8

5.1

49.7

4-

4.2

44.6

4-

3.1

42.0 ± 2.5

Foraging

4.0

-+■

1.0

3.9

+

0.8

2.5

4-

0.5

3.0 ± 0.6

Food handling

0.9

-+■

0.5

1.4

4-

0.7

0.6

4-

0.4

0.3 ± 0.2

Feeding fledgling

0.6

0.3

1.7

4-

0.6

0.1

4-

0.1

0.1 ± 0.1

470

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 3, September 2010

TABLE 3. Food searching behaviors (x ± SE) for male
Golden-cheeked Warblers at biking {n = 41) and non-
biking {n = 54) sites near Austin and on Fort Hood, Texas
during spring 2003.

Percent of total food searching time

Behavior Biking Non-biking

Hop

59.4 ± 5.0

59.8 ± 4.0

Short flight. <20 cm

10.7 ± 1.9

11.6 ± 2.2

Medium flight, 20 cm-2 m

5.4 ± 1.2

7.4 ± 1.3

Long flight, >2 m

24.5 ± 5.0

21.1 ± 4.3

Nest Success. — We

monitored

33 nests in

biking sites and 22 nests in non-biking sites.
Twenty-one (64%) and 19 nests (86%) success-
fully fledged at least one host young in biking and
non-biking sites, respectively. Seven nests failed
due to predation in biking sites, compared to two
in non-biking sites. Texas rat snakes {Elaphe
obsoletci) accounted for most of the depredation
(71%) in biking sites, and American Crows
{Corviis branchy tynchos) and fire ants {Solenop-
sis invicta) each accounted for one depredation in
biking sites. Texas rat snakes and Cooper’s
Hawks (Accipiter cooperii) accounted for the
two depredations in non-biking sites. Nest aban-
donment was three times higher in biking sites (5
nests; 15%) than non-biking sites (1 nest, 5%).
The abandonments occurred either late in the
laying stage or early in the incubation stage. We
did not document parasitism by Brown-headed
Cowbirds.

Daily nest survival for biking sites was 0.964 ±
0.018 for the incubation stage and 0.945 ± 0.019
for the nestling stage, whereas DNS for non-biking
sites was 0.986 ± 0.014 for the incubation stage
and 0.981 ± 0.013 for the nestling stage. The
Mayfield nest success estimate in non-biking sites
(70%) was twice as high as in biking sites (35%).

DISCUSSION

Our results suggest mountain biking trail use
negatively impacted Golden-cheeked Warblers
during the breeding season. Male territories were
>1.5 times larger and nest success was substan-
tially lower in biking than non-biking sites. Nest
abandonment and predation were also higher in
biking than non-biking sites. Several studies have
suggested that habitat fragmentation negatively
impacts Golden-cheeked Warbler nest survival
(Fink 1996; Maas-Burleigh 1998; Peak 2007;
Reidy et al. 2008, 2009). Peak (2007) reported
nest survival of Golden-cheeked Warblers was
lower in woodlands with high forest edge.
Similarly, Reidy et al. (2009) reported that as
edge factors including open edge density and
amount of open edge increased, nest survival of
Golden-cheeked Warblers substantially declined.
Golden-cheeked Warbler temtories in biking sites
in our study occurred in habitats more fragmented
by trails than temtories in non-biking sites. It
appears mountain biking indirectly affected Gold-
en-cheeked Warblers by habitat alteration and
fragmentation caused by occurrence of mountain
biking trails.

Territory size in many bird species is inversely
related to food abundance (Mahler 1970, Gass et
al. 1976, Salmonson and Baida 1977, Smith and
Shugart 1987). The larger territories observed in
biking sites in our study suggests that habitat
quality was poorer in these sites than in non-
biking sites. Fragmentation and alteration of
habitats in biking sites may have reduced the
quality of these habitats for Golden-cheeked
Warblers. For example, arthropod abundance
may have been low within forest edges and near
biking trails because of altered microclimate and
vegetation structure (Jokimaki et al. 1998, Kilgo
2005). However, Davis and Leslie (2008) reported

TABLE 4. Foraging attack techniques {x ± SE) used by male Golden-cheeked Warblers at biking (n = 16) and non-
biking (n = 21 ) sites near Austin and on Fort Hood, Texas during spring 2003.

Percent

Attack technique

Biking

Non-biking

P‘

Gleaning

Reaching

Hanging

Sally-.striking

Sally-hovering

Flutter-chasing

57.4 ±

10.1

47.4

± 8.9

0.555

13.4 ±

7.2

29.3

± 8.1

0.192

3.1 ±

3.1

4.3

± 4.3

0.823

5.2 ±

3.6

1.1

± 1.1

0.323

20.8 ±

8.6

16.1

± 6.7

0.506

0.0 ±

0.0

1.7

± 1.2

0.232

“ /'-value from Kru.skal-Wallis tc.st with I df.

Davis et a!. • MOUNTAIN BIKING AND GOLDEN-CHEEKED WARBLERS

471

TABLE 3. Parental behavior (.v ± SE) for female Golden-cheeked Warblers nesting
non-biking sites near Austin and on Fort Hood, Texas during daylight hours in spring 2002 and 200.3.

in biking and

Parameter

Biking

Non-biking

/"

Incubation stage*’

Incubation. %

67.40 ± 0.83

74.84

± 0.93

0.105

No. of incubation bout.s/day

17.98 ± 2.38

14.05

± 2.12

0.355

Length of incubation bout, min

32.39 ± 6.86

38.44

± 3.22

0.298

Early nestling stage*"

Brooding, %

36.59 ± 8.47

25.46

± 7.42

0.257

No. of brooding bouts/day

16.82 ± 2.93

24.60

± 8.06

0.345

Length of brooding bout, min

16.63 ± 5.72

12.60

± 4.38

0.165

Feeding nestlings, %

7.46 ± 1.49

6.02

± 0.93

0.852

No. feeding bouts/day

67.47 ± 17.07

79.65

± 22.75

0.710

Length of feeding bout, min

1.38 ± 0.22

1.58

± 0.48

0.850

Late nestling stage*"

Feeding nestlings, %

4.93 ± 1.08

4.25

± 0.71

0.446

No. of feeding bouts/day

62.22 ± 9.91

76.24

± 18.04

0.608

Length of feeding bout, min

1.19 ± 0.19

0.87

± 0.20

0.440

““ P-value from Kruskal-Wallis test with 1 df.

Sample sizes for incubation stage, early nesting stage, and late nestling stage in biking and non-biking sites were 8 and 7, 7 and 8, and 9 and 11 nests,
respectively.

that arthropod abundance from sweep-net samples
of tree branches in biking and non-biking sites did
not differ, and our study found no difference in
foraging behaviors between biking and non-biking
sites, which suggests that habitat quality was
similar between biking and non-biking sites.
Davis and Leslie (2008) did not distinguish
between samples collected from forest edges or
near biking trails and the impact of habitat
fragmentation and alteration on arthropod abun-
dance in biking sites is unknown.

Male Golden-cheeked Warblers in our study
spent similar amounts of time engaged in diurnal
behaviors in biking and non-biking sites. Nesting
behaviors of female Golden-cheeked Warblers
also did not differ between biking and non-biking
sites. These observations suggest disturbance
from mountain biking does not affect daily
activity budgets of Golden-cheeked Warblers,
but our direct observations of Golden-cheeked
Warbler encounters with mountain bikers did
show the birds were disturbed by bikers. During
three of the four observed encounters with bikers,
the bird flushed >20 m in response to the passing
biker. Moreover, three of five nests abandoned at
biking sites were <2 m from the trail, suggesting
disturbance from mountain biking has a role in
nest abandonment. We cannot conclude that direct
disturbance from mountain biking is benign to
nesting Golden-cheeked Warblers. Several studies

have shown that direct human disturbance can
alter activity budgets of birds (Stalmaster 1983,
Kaiser and Fritzell 1984, Burger 1994, Mikola et
al. 1994, Knapton et al. 2000). Our limited
number of observations of direct encounters of
Golden-cheeked Warblers and mountain bikers
likely does not indicate that such encounters are
rare, but rather, they are difficult to observe. We
expect variations in biker behaviors (e.g., speed,
group sizes, and silent vs. talking bikers) will
elicit variable responses in warblers. These factors
should be considered when evaluating the impact
of mountain biking on Golden-cheeked Warblers.

Our estimate of nest success for Golden-
cheeked Warblers in biking sites was lower than
in non-biking sites (35 vs. 70%), but our nest
success estimate for biking sites is comparable to
other studies from Fort Flood (45% [Stake et al.
2004], 34% [Peak 2007], 40% [Reidy et al. 2009])
and Austin (40% [Reidy et al. 2009]). This lower
nest success e.stimate in biking sites suggests
those sites may provide lower quality nesting
habitats for Golden-cheeked Warblers compared
to non-biking sites. The nest success estimate for
biking sites is still above the rate necessary to
balance juvenile and adult mortality (25-30%;
Donovan and Thompson 2001). Reidy et al.
(2008) showed that Golden-cheeked Warblers
can successfully reproduce in large protected
urban sites, but they also suggested these habitats

472

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 3. September 2010

may not be able to sustain productive populations
if further disturbance in the landscape or breeding
patch continues. This also appears to be the case
for biking sites in our study, which are currently
able to maintain viable populations of Golden-
cheeked Warblers. Additional use, fragmentation,
and alteration of these habitats could impact
productivity by affecting densities of breeding
adults, pairing success, or predator abundance
which may result in these habitats becoming
inadequate for sustaining viable populations
(Reidy et al. 2008).

Reidy et al. (2009) suggested higher predator
abundance and activity near edges contributed to
low nest survival of Golden-cheeked Warblers near
edges. The main predator of Golden-cheeked
Warbler nests in our study was the Texas rat snake,
which is a skilled tree climber (Pierce et al. 2008,
Sperry et al. 2009) and major nest predator of
Golden-cheeked Warblers (Stake et al. 2004, Reidy
et al. 2008). Texas rat snakes prefer fragmented and
edge habitats (Blouin-Demers and Weatherhead
2001, Sperry et al. 2009), and respond positively to
increased edge and fragmentation (Sperry et al.
2009). Thus, it is likely that habitat fragmentation
and alteration of biking sites by trails increased
vulnerability of Golden-cheeked Warbler nests in
those areas to predation by rat snakes and other
edge-adapted predators. Sperry et al. (2009)
suggested that Golden-cheeked Warbler nests that
occurred in areas with many canopy openings, such
as areas traversed by mountain bike trails, may be
more at risk to rat snake predation.

The direct impact of mountain biking on
Golden-cheeked Warblers may be minimal, but
the indirect impact from fragmentation and
alteration of habitats from mountain bike trails
may reduce the quality of nesting habitat for
Golden-cheeked Warblers. In particular, Golden-
cheeked Warblers nesting in habitats fragmented
and altered by mountain biking trails may be more
vulnerable to nest predation (Reidy et al. 2009)
and possibly encounter lower prey abundance
(Jokimaki et al. 1998, Kilgo 2005). Conservation
efforts that curtail construction of new mountain
biking trails in Golden-cheeked Warbler habitat
and reduce the amount of forest open edge habitat
created by existing mountain biking trails should
promote recovery objectives.

ACKNOWLEDGMENTS

Funding was provided by the U.S. Army Fort Hood
Military Base and U..S. Army Corps of Engineers

Construction Engineering Research Laboratory. The U.S.
Army Corps of Engineers Construction Engineering
Research Laboratory, Jones Technologies, and the Okla-
homa Cooperative Fish and Wildlife Research Unit
(Oklahoma State University, Oklahoma Department of
Wildlife Conservation, U.S. Geological Survey, U.S. Fish
and Wildlife Service, and Wildlife Management Institute
cooperating) administered the grant and provided logistical
support. We also thank The Nature Conservancy of Fort
Hood Field Office for their support. We appreciate the
support of J. D. Cornelius, D. M. Herbert, T. J. Hayden, and
C. E. Pekins during data collection at Eort Hood field sites,
and D. M. Koehler and C. M. Abbruzzese for support
during data collection at Austin field sites. We thank J. L.
Sterling, M. M. Stake, K. T. Cutrera, V. D. Bump, and H.
M. Oswald along with numerous volunteers for data
collection; A. Reisch for recording parental behavior from
video tapes; B. D. Hebert and T. J. Richardson for recording
male behavior from microcassettes and entering behavior
data; A. J. Eritchie for sorting and identifying arthropods;
and A. D. Thomas for data entry.

LITERATURE CITED

Altmann, j. 1974. Observational study of behavior:

samplings methods. Behaviour 49:227-265.

Barker, H. R. and B. M. Barker. 1984. Multivariate
analysis of variance (MANOVA): a practical guide to
its use in scientific decision making. University of
Alabama Press, Birmingham, USA.

Bibby, C. j. and N. D. Burgess. 1992. Bird census
techniques. Academic Press, Harcourt Brace and
Company, London, United Kingdom.

Blouin-Demers, G. and P. J. Weatherhead. 2001.
Habitat use by black rat snakes (Elaphe obsoleta
obsoleta) in fragmented forests. Ecology 82:2882-
2896.

Burger, J. 1994. The effect of human disturbance on
foraging behavior and habitat use in Piping Plover
{Charadriiis melodus). Estuaries 17:695-701.

Chavez, D. J., P. L. Winter, and J. M. Baas. 1993.
Recreational mountain biking: a management perspec-
tive. Journal of Park and Recreation Administration
11:29-36.

Davis, C. A. and D. M. Leslie Jr. 2008. Effects of
mountain biking activity on foraging and nesting
behavior of Golden-cheeked Warblers. Final Report,
Project Number MIPR #W81EWF63530473. U.S.
Army Corps of Engineers Research and Development
Center, Construction Engineering Research Laborato-
ry, Champaign, Illinois, USA.

Donovan, T. M. and E. R. Thomp.son 111. 2001. Modeling
the ecological trap hypothesis: a habitat and demo-
graphic analysis for migrant songbirds. Ecological
Applications 11:871-882.

Environmental Systems Research Institute (ESRI).
2000. ArcView GIS, Version 3.2a. Environmental
Systems Re.search Institute, Redlands, California,
USA.

Eink, M. L. 1996. Factors contributing to nest predation
within habitat of the Golden-cheeked Warbler, Travis

Davi.s- Cl a/. • MOUNTAIN IMKING AND GOLDEN-Ci lEEKTD WARBEERS

473

County. Texas. Tliesis. Texas A&M University,
College Station, USA.

Fl.ather. C. H. .^no II. K. CORDELl,. 1995. Outdoor
reereation: historieal and antieipated trends. Pages 3-
16 in Wildlife and reereationists: eoexistenee through
management and researeh (R. U. Knight and K. ,1.
Gutzwiller, Editors). Island Press, Washington, D.C.,
USA.

Gass, C. L., G. Angeher, and J. Centra. 1976. Regulation
of food supply by feeding territoriality in the Rufous
Hummingbird. Canadian Journal of Zoology 54:2046-
2054.

GtLL, F. B. 2007. Ornithology. Third Edition. W. H.
Freeman and Company, New York, USA.

Johnson, D, H. 1979. Estimating nest success: the Mayfield
method and an alternative. Auk 96:651-661.

Johnson, R. A. and D. W, Wichern. 1988. Applied
multivariate statistical analysis. Second Edition. Pren-
tice Hall, Englewood Cliffs, New Jersey, USA.

JOKIMAKI, J., E. HUHTA, J. iTAMtES, AND P. RaHKO. 1998.
Distribution of arthropods in relation to forest patch
size, edge, and stand characteristics. Canadian Journal
of Forest Re.search 28:1068-1072.

Kaiser. M. S. and E. K. Fritzell. 1984. Effects of river
reereationists on Green-backed Heron behavior. Jour-
nal of Wildlife Management 48:561-567.

Kilgo, j. C. 2005. Harvest-related edge effects on prey
availability and foraging of Hooded Warblers in a
bottomland hardwood forest. Condor 107:627-636.

Knapton, R. W.. S. a. Petrie, and G. Herring. 2000.
Human disturbance of diving ducks on Long Point
Bay, Lake Erie. Wildlife Society Bulletin 28:923-930.

Knight, R. L. and K. J. Gutzwiller (Editors). 1995.
Wildlife and reereationists: coexistence through man-
agement and re.search. Island Press, Washington, D.C.,
USA.

Ladd, C. and L. Gass. 1999. Golden-cheeked Warbler
(Dendroicci chrysoparia). The birds of North America.
Number 420.

Laundre, j. W. and B. L. Keller. 1984. Home-range size
of coyotes: a critical review. Journal of Wildlife
Management 48: 1 27- 1 39.

Maas-Burleigh. D. S. 1998. Factors influencing demo-
graphics of Golden-cheeked Warblers (Dendroicci
chrysoparia) at Fort Hood Military Reservation,
Texas. Thesis. University of Oklahoma, Norman,
USA.

Mahler, W. J. 1970. The Pomarine Jaeger as a brown
lemming predator in northern Alaska. Wil.son Bulletin
82:130-157.

Martin, P. and P. Bate.son. 1986. Measuring behavior: an
introductory guide. Cambridge University Press, New
York, USA.

Mayfield, H. 1961. Nesting success calculated from
exposure. Wil.son Bulletin 73:255-261.

Mayfield, H. F. 1975. Suggestions for calculating nest
success. Wilson Bulletin 87:456-466.

Mikola, J., M. Miettinen, E, Lehikoinen, and K.
Lehtila. 1994. The effects of disturbance caused by
boating on survival and behavior of Velvet Scoter

Melanilta fusca ducklings. Biological Conservation
67:1 19-124.

Miller, J. R. and N. T. Hohils, 2000. Recreational trails,
human activity, and nest predation in lowland riparian
areas. Landscape and Urban Planning 50:227-236.

Miller, S. G., R. L. Knight, and C. K. Miller. 1998.
Innuence of recreational trails on breeding bird
communities. Ecological Applications 8:162-169.

Mosedale, j. 2002. Mountain biking in the Canadian
Rocky Mountains: a situational analysis. Mountain
Forum E-consultation for the UNEP/Bischkek Global
Mountain Summit. Bishkek, Kyrgystan.

Peak, R. G. 2007. Forest edges negatively affect Golden-
cheeked Warbler nest survival. Condor 109:628-637.

Pierce, J. B., R. R. Fleet, L. McBrayer, and D. C.
Rudolph. 2008. Use of trees by Texas rat snake
(Elaplie ohsoleta) in eastern Texas. Southeastern
Naturalist 7:359-366.

Pike Masteralexis, L., C. A. Barr, and M. A. Hums.
2008. Principles and practice of sport management. Third
Edition. Jones and Bartlett Publishers, New York, USA.

Ralph, C. J., G. R. Geupel, P. Pyle, T. E. Martin, and
D. F. Desante. 1993. Handbook of field methods for
monitoring landbirds. USDA, Forest Service, General
Technical Report PSW-GTR-144. Pacific Southwest
Research Station, Albany, California, USA.

Reidy, j. L., M. M. Stake, and F. R. Thompson III. 2008.
Golden-cheeked Warbler nest mortality and predators
in urban and rural landscapes. Condor 110:458-466.

Reidy, J. L., F. R. Thompson, and R. G. Peak. 2009.
Factors affecting Golden-cheeked Warbler nest sur-
vival in urban and rural landscapes. Journal of Wildlife
Management 73:407^13.

Remsen, j. V. AND .S. K. Robinson. 1990. A classification
scheme for foraging behavior of birds in terrestrial
habitats. Studies in Avian Biology 13: 144-160.

Robinson, S. K. and R. T. Holmes. 1982. Foraging behav-
ior of forest birds: the relationships among search tactics,
diet, and habitat stmeture. Ecology 63: 1918-1931.

Rodgers, A. R. and A. P. Carr. 1998. HRE: the home
range extension for AreView™: user's manual. Beta
te.st Version 0.9, July 1998. Ontario Mini.stry of
Natural Resources, Centre for Northern Forest Eco-
system Research, Thunder Bay, Ontario, Canada.

Salmonson, M. G. and R. P. Balda. 1977. Winter
territoriality of Townsend's Solitaires (Myadestes
townsendi) in a pinyon-juniper-ponderosa ecosystem.
Condor 49: 148-161.

Seaman, D. E., J. J. Millspaugh, B. J. Kernohan, G. C.
Brundige, K. j. Raedeke, and R. a. Gitzen. 1999.
Effects of sample size on kernel home range estimates.
Journal of Wildlife Management 6,3:739-747.

Smith, T. M. and H. H. Shugart. 1987. Territory size
variation in the Ovenbird: the role of habitat structure.
Ecology 68:695-704.

Sperry, J. H., D. A. Cimprich, R. G, Peak, and P. J.
Weatherhead. 2009. Is nest predation on two
endangered bird species higher in habitats preferred
by snakes? Ecoscience 16:1 1 1-1 18.

.Stake, M. M. 2000. Using video to identify Black-capped
Vireo nest predators at Fort Hood, Texas. Endangered

474

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 3, September 2010

species monitoring and management at Fort Hood,
Texas: 2000 Annual Report. The Nature Conservancy
of Texas, Fort Hood Project, Austin, USA.

Stake, M. M. and D. A. Cimprich. 2003. Using video to
monitor predation at Black-capped Vireo nests.
Condor 105:348-357.

Stake, M. M., J. Faaborg, and F. R. Thompson III. 2004.
Video identification of predators at Golden-cheeked
Warbler nests. Journal of Field Ornithology 75:337-
344.

Stalmaster, M. V. 1983. An energetics simulation model
for managing wintering Bald Eagles. Journal of
Wildlife Management 47:349-359.

Symmonds, M. C., W. E. Hammitt, and V. L. Quisen-

BERRY. 2000. Managing recreational trail environ-
ments for mountain bike user preferences. Environ-
mental Management 25:549-564.

SYSTAT. 2004. SYSTAT 11. SYSTAT Software Incorpo-
rated, San Jose, California, USA.

Taylor, A. R. and R. L. Knight. 2003. Responses of
wildlife to human activity: terminology and methods.
Wildlife Society Bulletin 31:1263-1271.

U. S. Department OF Interior (USDI). 1993. Biological
opinion. USDI, Eish and Wildlife Service, Austin,
Texas, USA.

Wunderle, J. M. and S. C. Latta. 1998. Avian resource
use in Dominican shade coffee plantations. Wilson
Bulletin 110:271-281.

The Wilson Jotinuil of Oniilliology 122(3);475-483, 2010

SURVIVAL, SITE FIDELITY, AND POPULATION TRENDS OF
AMERICAN KESTRELS WINTERING IN SOUTHWESTERN FLORIDA

DANIEL M. HINNEBUSCH,' ^ JEAN-FRANgOlS THERRIEN,' ^
MARC-ANDRE VALIQUETTE,' - BOB ROBERTSON,

SUE ROBERTSON,’ AND KEITH L. BILDSTEIN'

ABSTRACT. — The winter population ecology of American Kestrel (Falco span’erius), one of the most abundant and
widely distributed raptors in North America, is poorly understood. We systematically searched a 225-km^ area in Cape
Coral, southwestern Florida, for American Kestrels during 14 winters (1 Dec- 15 Mar, 1994-2008) to measure their annual
apparent survival and to see il individuals returned to the same wintering area. We recaptured 101 of 2,958 banded kestrels
during the study. We estimated annual apparent survival of 75% for males and 74% for females using a Cormack-Jolly-
Seber model. These estimates are considerably higher than previous estimates for American Kestrels, but are similar to
estimates reported for other species of Falco. Forty-six percent of the kestrels estimated to have survived were observed in
the study area 1 year after recapture, based on year-specific color banding. All but six of 101 kestrels were recaptured
within 1 km ot where they were banded. Four of five kestrels banded as nestlings and subsequently recaptured in the study
area were banded in southeastern Pennsylvania, suggesting migratory connectivity. Eighty percent of the kestrels trapped
were females, but the proportion of females decreased annually (-3 ± 1% per year). Overall, the population decreased by
an average ot 7 ± 2% per year. Recent land-use change accompanied by increased human density and suburban expansion
may be causing the observed trends. Received 29 October 2009. Accepted 9 March 2010.

American Kestrels [Falco sparverius) occur in
open habitats throughout most of North and South
America (Smallwood and Bird 2002). The
American Kestrel, historically abundant through-
out most of its range, has declined significantly
since the 1970s (Sullivan and Wood 2005, Farmer
et al. 2008). Potential causes for this decline
include West Nile virus, organophosphate poison-
ing, increased predation by growing populations
of Cooper’s Hawks [Accipiter cooperii), popula-
tion declines of Northern Flickers [Colaptes
auratus), a primary cavity nester; and land-use
changes (Sullivan and Wood 2005, Farmer et al.
2006, Medica et al. 2007, Farmer et al. 2008).

American Kestrels are partial migrants that
undertake leapfrog migration (Roest 1957, Small-
wood and Bird 2002). Individuals breeding at
northern latitudes are more likely to migrate and
move greater distances than individuals breeding
farther south (Goodrich and Smith 2008). Banding
returns indicate that most individuals that migrate
to southwestern Florida breed in the northeastern
United States or in southeastern Canada (Layne
1982). Individuals or, more rarely, pairs of

' Hawk Mountain Sanctuary, Acopian Center for Con-
servation Learning, 410 Summer Valley Road, Orwigsburg,
PA 17961, USA.

-Universite Laval, Departement de biologie and Centre
d'etudes nordiques, Quebec, QC GIK 7P4, Canada.

^ Deceased.

■'Corresponding author; e-mail: hinnebusch@hawkmtn.org
and ospreyl984@yahoo.com

kestrels maintain areas of exclusive use during
winter (Cade 1955, Mills 1975). Males and
females use different habitats in winter with
males occupying areas with more woody vegeta-
tion (Smallwood and Bird 2002). This may be the
result of later-migrating males selecting subopti-
mal habitat left vacant by females (Smallwood
1988), or because larger females dominate smaller
males (Ardia and Bildstein 1997).

Site fidelity by migratory birds can be advan-
tageous because individuals benefit from being
familiar with a previously occupied teiritory
(Spaans 1977, Nichols et al. 1983). Familiarity
with a territory likely allows for more efficient
foraging and predator avoidance (Hinde 1956,
Clarke et al. 1993, Yoder et al. 2004). Evidence
supporting winter site fidelity in American
Kestrels is limited to studies of small numbers
of band recoveries and observation of uniquely
marked individuals. Return rates of 19% (four of
21 individuals) and 17% (eight of 47) were
reported at wintering sites in central Ohio (Mills
1975) and .southern Texas (Bolen and Derden
1980), respectively, for kestrels marked with
patagial tags. Tabb (1977) reported a return rate
of 2% (17 of 842) for kestrels banded at several
sites in southern Florida, but that estimate is based
entirely on recaptures because individuals could
not be reliably identified without recapture.
Several other species of raptors demonstrate
winter site fidelity (Garrison and Bloom 1993,
Harmata and Stahlecker 1993, Powers 1996).

475

476

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 3. Seplemher 2010

Moreover, satellite tracking has provided evi-
dence for winter site fidelity by North American
falcons, including Peregrine Falcon (Falco pere-
grimis) and Prairie Falcon (F. mexicamis), based
on a small number of individuals (McGrady et al.
2002, Steenhof et al. 2005).

Survival estimates are essential for effective
species’ management and conservation (Crone
2001, Sandercock 2006). However, survival in the
wild remains poorly known for the American
Kestrel (Smallwood and Bird 2002). To date, two
studies have estimated mortality rates using band
recovery and territory occupancy (Roest 1957,
Henny 1972), a technique that does not consider
the probabilities of capture and detection biases
and is further biased by territory vacancies that
may not be the result of deaths of birds (Gould
and Fuller 1995).

We report the results of a 14-year study of
American Kestrels wintering in southwestern
Florida. Specifically, we: ( 1 ) estimate annual
apparent survival, (2) discuss a decrease in size
of the observed population characterized by a
decrease in number of females, (3) provide
evidence for strong winter site fidelity, and (4)
discuss evidence for migratory connectivity be-
tween winter territories in Florida and natal sites
in the northeastern United States.

METHODS

Study Area. — We marked American Kestrels
within —275 km^ containing 2,966 km of roads in
the city of Cape Coral and adjacent Pine Island,
west of Fort Myers, Lee County, Florida. The
study site is on a peninsula along the Gulf of
Mexico bordered to the east by the Caloosa-
hatchee River and to the west by Matlacha Pass.
Native cover includes slash pine (Pinii.s elliottii)
woodlands and tidal marshes (Wesemann 1986,
Millsap and Bear 2000). An extensive network of
canals and roads was built in the area in the late
I95()s (Bernard 1983, Zeiss 1983). The site now
consists largely of single-family residences and
unoccupied, but regularly mowed, lots (Millsap
and Bear 2000). Parallel roads are -100 m apart.
The human population in 2000 was 102,286
(376.1/knr). In 2006 it had grown 48% to
151,389 (556.7/km^) (USCB 2009).

Data Collection.— BR and SR .searched Cape
Coral for American Kestrels each winter ( 1 Dec-
15 Mar) for 14 years beginning in December
1994. Observers drove along each road in the
study area at <25 km/hr while .searching for

kestrels, which often hunted from prominent
perches and were easily observed (Bildstein and
Collopy 1987). We surveyed each road once each
winter and recorded male/female, location, and
band color of all kestrels observed on our initial
visit to each road.

We attempted to capture every kestrel observed
by using a bal-chatri trap (25-cm-diameter, 7-cm-
high) baited with two mice {Miis muscidus)
(Berger and Mueller 1959). Unless previously
banded, captured kestrels were marked with a
USGS aluminum band and a year-specific plastic,
coil-style, color band (red, blue, white, yellow,
black, or green), except for 1994-1995 and 1995-
1996, when color bands were not used. Each
color, except white, was used in 2 nonconsecutive
years with bands being placed on different legs in
different years. Wrap-around color bands were
manufactured by Gey Band & Tag Company,
Norristown, Pennsylvania, USA. A body-condi-
tion index was calculated for each captured bird
by dividing the cube root of body mass by wing
chord, which was used as a wing loading index for
Merlins {Falco cohunbarius) by Temple ( 1972). If
we were unable to capture a bird, we returned to
the same site later in the season, usually in early
March, and again attempted to trap it. However,
observations of kestrels observed on a road after
our initial survey of that road were not included in
analyses. We indicated sites at which we banded
or observed kestrels as the nearest intersection of
roads. We obtained information on banding and
recovery locations outside the Cape Coral study
area from the Bird Banding Laboratory, USGS,
Patuxent Wildlife Research Center, Laurel, Mary-
land, USA. We considered recovered banded
kestrels to have been found during the breeding
.season if they were found between 1 April and 3 1
August {sen.<ii4 Smallwood and Bird 2002).

Statistical Analyses. — We estimated annual
apparent survival (4)) and capture probability (p)
using recapture data with a Cormack-Jolly-Seber
(CJS) model of mark-recapture in MARK 5.1
software (White and Burnham 1999). The model
was based on a single study site with 14 recapture
occasions (1995-2008). We used a sing;le biolog-
ical state, treating the age of all individuals as
after-hatch-year. Our global model allowed ap-
parent survival and capture probability to vary
with year and male/female, as well as an
interaction between those covariates. We retained
all models for which AAIC,. was <2 and used
model averaging to estimate parameters.

Iliniwhusch d al. • WINTERING AMERICAN REST REES

477

The initial capture and recapture sites for each
recaptured bird were identified using ArcGIS
(ESRl 2009). We tested for a difference between
males and females in inter-capture distance, the
distance between the site of original capture and
the recapture site, using the Mann-Whitney U-
test. We used Spearman’s rank coirelation to test
for a relationship between inter-capture distance
and years between captures and body-condition
index at time of initial capture. We compared
average annual capture success rate for marked
and unmarked kestrels using a paired Mest.

Observed population was estimated as the
number of kestrels observed on the initial visit
to all roads. Effort did not vary among years and
we used observed population as an index of true
population size to compare among years. We used
the natural logarithms of the female proportion of
the population and the number of kestrels
observed each year to estimate the change in the
female proportion of the population, and the
annual percent change in the observed population,
respectively, using simple linear regression. All
statistical analyses used SAS 9.1 (SAS Institute
Inc. 2003). Tests were considered significant if P
< 0.05. All error terms, except as noted, represent
one standard error.

RESULTS

We captured 2,958 American Kestrels (211 ±
27/year), including 2,365 females, 592 males, and
one individual for which we did not record
gender. We recaptured 101 kestrels (80 females
and 21 males) at least 9 months after their initial
capture (Table 1). The mean time between
banding and recapture was 2.8 ± 0.2 years (range
= 1 1 months to 9 years). No bird was recaptured
more than once. Each year, an average of 192 ±
56 individuals marked in previous years with
year-specific color bands were observed, but these
birds were difficult to recapture. Overall annual
capture rates of unbanded kestrels were much
higher than recapture rates for banded birds (61 ±
2% vs. 4 ± 1%; t = 29.5, P < 0.001).

We retained three models for apparent survival
and recapture probability with AAIC,. < 2
(combined Akaike weight, vv, = 0.86; Table 2).
We estimated male apparent survival at 75 ± 5%
and female apparent survival at 74 ± 4% based on
model averaging. Overall, the year effect was not
retained as a covariate affecting apparent survival,
suggesting constant adult apparent survival during
the 14 years of study. Recapture probability, as

estimated with the CJS model, was low during the
study period. It was greatest in 1995-1996 (5%
for males and 4% for females), but remained
between I and 3% for both males and females in
every subsequent year until 2007-2008, when no
kestrels were recaptured.

The proportion of females in the observed
population (banded and unbanded) decreased by 3
± 1 % annually (R- = 0.79, F = 38.0, P < 0.001 ).
Eemales comprised 86% of the population in the
first 3 years of study and 64% of the population in
the last 3 years. The total number of kestrels
observed in the study area decreased an average of
7 ± 2% per year (R- = 0.54, F = 11.6, =

0.007). The number of females observed de-
creased an average of 10 ± 2% per year (R- =
0.63, F = 17.3, P = 0.002; Eig. 1) and the
number of males increased by 4 ± 1% per year
(R- = 0.37, F = 5.9, P = 0.035).

Eor kestrels color banded in year n, we
observed an average of 34 ± 1% in year // -l- 1,
18 ± 1% in year n + 2, and 9 ± 1% in year « -h 3
(Eig. 2). Considering our estimate of annual
apparent survival with a CJS model, we observed,
in year /? -I- 1, 46% of the kestrels banded in year n
that were still alive in year n -r 1 (Eig. 2).

The longest inter-capture distance that could
have occurred within our study area was 32 km.
Mean distance between initial and second capture
for recaptured kestrels was 488 ± 102 m (range =
0 to 7,887 m). All but six birds recaptured were
caught <1 km from the initial capture site, and
75% were recaptured within 423 m (Eig. 3).
There was no difference in inter-capture distance
between males and females {U = 1,208, P =
0.26), and inter-capture distance was not corre-
lated with years between captures (/y = 0.083, P
= 0.41). Inter-capture distance and body-condi-
tion index at time of initial capture exhibited a
nearly significant, weak coirelation (/y = 0. 1 9, P
= 0.061).

Twenty-five banded kestrels were observed
outside the study area (Eig. 4). We recaptured
five females in Cape Coral that had been banded
as nestlings, four of which had been banded in
eastern Pennsylvania and one of which had been
banded in central Nova Scotia. We also recap-
tured two females that had been banded as adults,
one in April in Connecticut and one in September
on the Delmarva Peninsula, Virginia. Eighteen
( 1 1 females and 7 males) kestrels banded in Cape
Coral were found dead outside the study area,
including 13 (7 females and 6 males) that were

TABLE 1. Percent of American Kestrel cohorts recaptured each year after banding in Cape Coral, southwestern Florida. indicates the number of kestrels banded each year.
Subsequent columns indicate the percent (above) and number (below) recaptured each year after banding.

478

THE WILSON JOURNAL OF ORNITHOLOGY • Voi 122, No. 3, September 2010

oooooooooooooooooooooooooo

OOOOOOOOOO — — OOOOOO oo — —

V

OOOOOOOO — — -^(N

V

V

— CN — (N O O

OOOOOOOO — — — — — — — — (N(NrO

V V

__0000 — — — — —

V V V

__00 — — — CN — — — m

V V

(N — (NOO — — OOOO — CN

V V

OOOO — — CNiri — CNCNOO

— CN — —

V

— CN — ro CN

OOOOOOOO

V

— C^l C'l ^

V

— — — in

V

.

•rt

O

OO

(N

Os

o

SO

OS

O

os

r--

r-

—

O

r4

sD

OO

Ov

o

ri

rY

»r)

so

r-*

00

Os

os

OS

O^

o

o

O

O

o

O

O

o

o

Q\

OS

OS

Os

o

o

O

O

o

O

O

o

o

(N

<N

fN

1

(N

1

<N

(N

1

(N

1

4

1

si

1

r-

1

OO

1

OS

ci

1

1

r^i

1

rn

liO

si

1

r-

os

o^

OS

OS

o

o

O

o

o

O

o

o

Os

Os

OS

Os

OS.

OS

o

o

O

o

o

o

o

o

(N

fN

(N

(N

<N

(N

IHwiehusch cl cil. • WINTERING AMERICAN KESTRELS

479

TABLE 2. Models of apparent survival (d)) and recapture probability (p) for recaptured American Kestrels (n = 101 ) in
southwestern Florida for which ve, > 0.10. Both apparent survival and recapture probability were allowed to vary with year
and sex, as well as an interaction between year and sex, in the global model. We tested all possible models nested within the
global model. The best model not included in model averaging [0(sex)/p(year + sex)| had AAIC<. = 2.25 and vt', = 0.13.

Model

AIC,

AAIC..

Parameters

Variance

0(.)/p(year)

1209.07

0.00

0.39

14

1 19.36

4)(.)/p(year + sex)

1209.64

0.57

0.30

15

1 17.92

0(sex)/p(year)

1210.72

1.65

0.17

15

1 18.99

recovered during the breeding season, although
we were not able to ascertain if kestrels died in
breeding areas or as migrants. Kestrels recovered
during the breeding season included five in
Quebec, one in New Brunswick, one in Maine,
two in New York, two in Pennsylvania, one in
Delaware, and one in South Carolina. Kestrels
recovered during the non-breeding season includ-
ed one each in New Brunswick, Ontario, New
York, North Carolina, and South Carolina. No
birds captured in Cape Coral were recaptured
outside the study area and reported to the Bird
Banding Laboratory.

DISCUSSION

The annual apparent survival estimates of 75%
for males and 74% for females are considerably

higher than those reported by Roest (1957) and
Henny (1972) (43 and 55%, respectively). Both of
these studies relied on band recoveries from dead
birds or nest-site occupancy to evaluate annual
mortality rates. Their estimates of annual apparent
survival are based on a time when direct
persecution by humans (Bildstein 2008) may have
resulted in greater mortality. At least 24% of the
recovered kestrel bands reported by Roest (1957)
came from birds killed by people. Human-caused
mortality of kestrels reported in Henny (1972)
ranged from 16% for 1958-1965 to 48% for
1925-1945. Our estimates are similar to estimates
reported for other species of Falco in North
America, including 75-89% for Prairie Falcons
(Enderson 1969, Steenhof et al. 2006), 62% for
Merlins (Lieske et al. 2000), and 79-93% for

Year

FIG. 1. Number of observations of female (circles), male (.squares), and total (triangles) American Kestrels in Cape
Coral, southwestern Florida, 1996-2008. “n" indicates number of kestrels and “Y” indicates the calendar year beginning
in a winter season in the regression equations.

480

THE WILSON JOURNAL OL ORNITHOLOGY • VuL 122, No. 3, September 2010

FIG. 2. Proportion of color-marked American Kestrel
cohorts observed in Cape Coral, southwestern Florida and
the estimated proportion surviving each year. Points
indicate the average proportion of marked cohorts observed
in the study area after each year. Vertical lines are 95%
confidence intervals. The curve indicates the estimated
surviving population based on annual adult apparent
survival estimates from the CIS model in this study.

Peregrine Falcons (Court et al. 1989, Tordoff and
Redig 1997). Survival estimates for Common
Kestrels (Falco tinnunciiliis) in the United King-
dom were 80% for males and 66% for females

(Dobson 1987). To our knowledge, this study
reports the first estimates of annual apparent
survival for American Kestrels using a CJS
model.

We recognize use of year-specific color bands
is not the most effective method to estimate
annual apparent survival, recapture probability,
and between-year inter-capture distances because
it did not allow us to record individual encounter
histories for kestrels that were not recaptured. Our
estimate of apparent survival with the CJS model
was based only on the 101 recaptured kestrels. We
believe this estimate is reliable because it
considers the low recapture probability.

The decline in the numbers of kestrels observed
in the study area does not seem to be attributable
to a decrease in adult survival over the study
period as no year effect was retained in the
models. However, the decrease in numbers of
kestrels observed started in 2002, 3 years after
discovery of West Nile virus in New York
(Lanciotti et al. 1999). American Kestrels are
susceptible to infection by West Nile virus
(Komar et al. 2003) and exhibit symptoms that
may decrease survival in wild birds (Nemeth et al.
2006). Ninety-five percent of tested American
Kestrels breeding in Pennsylvania had antibodies
indicating they had been exposed to the virus

03

■g

>

TD

C

a)

n

E

Z3

Inter-capture distance (m)

FIG. 3. Di.stances between capture and recapture sites between winter seasons for American Kestrels in Cape Coral,
southwestern Florida. The greatest inter-capture distance was 7.9 km.

Hiimehiisch cl a/. • WINTERINC} AMERICAN KESTRELS

481

(Medica et al. 2007). The population decrease in
our study area was characterized by a decrease in
females. We believe this change in sex ratio could
be due to land-use changes at the study site.
Female kestrels occupy more open habitat in
winter than males and, over the course of our
study, land use in our study area shifted from open
areas to more woody cover due to suburban
development. The increa.se in number of males
observed in the study area was relatively small
compared to the decrease in the number of
females observed. The observed increase in
human population in the study area, combined
with the suburban expansion of Cape Coral, has
reduced the occurrence of good wintering habitat
for kestrels. Those two factors seem to negatively
affect the overall Cape Coral kestrel population as
well as skew the sex ratio toward males.

American Kestrels wintering in Cape Coral,
Florida exhibited strong winter site fidelity. Most

kestrels that returned to the study area wintered a
short distance (<1 km) from their previous
wintering area, based on between-year inter-
capture distances. The relatively large numbers of
color-banded birds we saw, together with a small
recapture rate, suggest the small number of
recaptures was not due to emigration but rather to
capture avoidance. Our observations of numerous
color-banded kestrels in the area support strong
winter site fidelity by American Kestrels. Our
estimate of the proportion of birds observed in
winters subsequent to capture is likely conservative
due to loss of color bands (Nekson et al. 1980).
Seven percent of the ke.strels observed in the last
year of our study (2007-2008) had an aluminum
band and no color band. It is unlikely these kestrels
had been banded in years when we did not use
color bands (1994-1995 and 1995-1996) because
we estimate that >96% of the kestrels banded in
those years had already died by 2007-2008.

482

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 3, September 2010

Four of five kestrels recaptured after being
banded as nestlings outside the study area were
banded within 42 km of each other in southeastern
Pennsylvania, suggesting migratory connectivity
in populations we studied. American Kestrels
migrate singly or in small groups (Kerlinger et al.
1985), and juveniles migrate earlier than adults
(Smallwood 1988, Mueller et al. 2000); thus.
Juveniles do not have the opportunity to follow
adults on migration, as has been observed for
some flocking migrants (Maransky and Bildstein
2001). Strong connectivity between natal and
wintering areas has important conservation impli-
cations because a decrease in productivity in a
local breeding population can drastically affect
the population in wintering areas with which it is
connected and vice versa. Recoveries of kestrels
banded in the study area were distributed over a
much wider range than were banding locations of
nestlings recaptured in the study area, providing
little evidence of connectivity between wintering
and breeding areas for adults. We could not
ascertain whether a recovered bird had died in its
breeding area or as a migrant in most cases. We
cannot clearly state there is a lack of migratory
connectivity in adult kestrels.

ACKNOWLEDGMENTS

The unexpected death of Bob Robert.son provides us with
the opportunity to recognize his endless devotion of time
and energy to the study of American Kestrels. Danny
Bystrak, at the Bird Banding Laboratory, provided
information on banding and recovery locations for kestrels
outside our study area. We especially appreciate the
comments of C. E. Braun, C. J. Farmer, Brian Millsap,
and Karen Steenhof on earlier versions of this manuscript.
This is Hawk Mountain Sanctuary contribution to con.ser-
vation science number 189.

LITERATURE CITED

Ardia, D. R. and K. L. Bildstein. 1997. Sex-related
differences in habitat .selection in wintering American
Kestrels, Falca span'erius. Animal Behaviour 53:
1305-1311.

Berger, D. D. and H. C. Mueller. 1959. The bal-chatri: a
trap for the birds of prey. Bird-Banding 30:18-26.
Bernard, E. 1983. Lies that came true: tall tales and hard
sales in Cape Coral. Fla. Anna Publishing Inc., Ocoee,
Florida. USA.

Bii.dSTEIN. K. L. 2008. A brief history of raptor
conservation in North America. Pages 5-36 in State
of North America's birds of prey ( K. L. Bildstein, J. P.
Smith. E. Ruelas Inzunza. and R. R. Veit, Editors).
Nultall Ornithological Club, Cambridge. Massachu-
setts. USA. and American Ornithologists’ Union,
Washington. D.C.. USA.

Bildstein, K. L. and M. W. Collopy. 1987. Hunting
behavior of Eurasian (Falco tiimiinciilus) and Amer-
ican Kestrels (F. span’erius): a review. Pages 66-82 in
The ancestral kestrel (D. M. Bird and R. Bowman,
Editors). Raptor Research Report Number 6.

Bolen, E. G. and D. S. Derden. 1980. Winter returns of
American Kestrels. Journal of Field Ornithology
51:174-175.

Cade, T. J. 1955. Experiments on winter territoriality of the
American Kestrel, Falco sparverius. Wilson Bulletin
67:5-17.

Clarke, M. F., K. Burke da Silva, H. Lair, R.
Pocklington, D. L. Kramer, and R. L. McLaugh-
lin. 1993. Site familiarity affects escape behaviour of
the eastern chipmunk. Taniias striatus. Oikos 66:533-
537.

Court, G. S., D. M. Bradley, C. C. Gates, and D. A.
Boag. 1989. Turnover and recruitment in a tundra
population of Peregrine Falcons. Ibis 131:487^96.

Crone, E. 2001. Is survivorship a better fitness surrogate
than fecundity? Evolution 55:261 1-2614.

Dobson, A. P. 1987. A comparison of seasonal and annual
mortality for both sexes of fifteen species of common
British birds. Omis Scandinavica 18:122-128.

Enderson, j. H. 1969. Peregrine and Prairie falcon life
tables ba.sed on band-recovery data. Pages 505-509 in
Peregrine Falcon populations: their biology and
decline (J. J. Hickey, Editor). University of Wisconsin
Press. Madison, USA.

Environmental System Research Institute (ESRl).
2009. ArcGIS. Version 9.3. Environmental System
Research Institute, Redlands, California, USA.

Farmer, C. J., L. J. Goodrich, E. Ruelas Inzunza, and J.
P. Smith. 2008. Conservation status of North Amer-
ica’s birds of prey. Pages 303M19 in State of North
America’s birds of prey (K. L. Bildstein, J. P. Smith.
E. Ruelas Inzunza, and R. R. Veit, Editors). Nuttall
Ornithological Club, Cambridge, Massachusetts, USA,
and American Ornithologists’ Union, Washington,

D. C., USA.

Farmer, G. C.. K. McCarty, S. Robertson, B. Robert-
son, AND K. L. Bildstein. 2006. Suspected predation
by accipiters on American Kestrels {Falco span>erius)
in eastern Pennsylvania, U.S.A. Journal of Raptor
Research 40:294-297.

Garrison, B. A. and P. H. Bloom. 1993. Natal origins and
winter site fidelity of Rough-legged Hawks wintering
in California. Journal of Raptor Research 27:1 16-118.

Goodrich L. J. and J. P. Smith. 2008. Raptor migration in
North America. Pages 37-149 in State of North
America’s birds of prey (K. L. Bildstein, J. P. Smith,

E. Ruelas Inzunza, and R. R. Veit, Editors). Nuttall
Ornithological Club, Cambridge, Massachusetts, USA,
and American Ornithologists’ Union, Washington,
D.C., USA.

Gould, W. R. and M. R. Fuller. 1995. Survival and
population size estimation in raptor studies: a
comparison of two methods. Journal of Raptor
Research 29:256-264.

Harmata, a. R. and D. W. Stahlhcker. 1993. Fidelity of
migrant Bald Eagles to wintering grounds in southern

Iliimchiisch cl al. • WINTERING AMERICAN KESTRELS

483

Colorado and northern New Mexieo. Journal of Field
Ornithology 64:129-134.

Henny, C. J. 1972. An analysis of the population dynaniies
ot seleeted avian speeies: with speeial reterence to
ehanges during the eurrejit modern pesticide era.
Wildlife Re.search Report I. USDI, Fish and Wildlife
Service, Washington. D.C., USA.

Hinde. R. a. 1956. The biological significance of the
territories of birds. Ibis 98:340-369.

Kerlinger, P., V. P. Bingman. and K. P. Able. 1985.
Comparative flight behaviour of migrating hawks
studied with tracking radar during autumn in central
New York. Canadian Journal of Zoology 63:755-761.

Komar, N., S. Langevin, S. Hinten, N. Nemeth, E.
Edwards. D. Hettler. B. Davis, R. Bowen, and M.
Bunning. 2003. Experimental infection of North
American birds with the New York 1999 strain of West
Nile virus. Emerging Infectious Diseases 9:31 1-322.

Lanciotti, R. S., J. T. Roehrig, V. Deubel, J. Smith, M.
Parker, K. Steele, B. Crise, K. E. Volpe, M. B.
Crabtree, J. H. Scherret, R. A. Hall, J. S.
Mackenzie, C. B. Cropp, B. Panigrahy, E. Ostlund,
B. Schmitt, M. Malkinson, C. Banet, J. Weissman,
N. Komar, H. M. Savage, W. Stone, T. McNamara,
AND D. J. Gubler. 1999. Origin of the West Nile virus
responsible for an outbreak of encephalitis in the
northeastern United States. Science 286:2333-2337.

Layne, J. N. 1982. Analysis of Florida-related banding data
for the American Kestrel. North American Bird
Bander 7:94-99.

Lieske, D. j., I. G. Warkentin, P. C. James, U. W.
Oliphant, and R. H. M. Espie. 2000. Effects of popu-
lation density on survival in Merlins. Auk 1 17:184-193.

Maransky, B. P. and K. L. Bildstein. 2001. Follow your
elders: age-related differences in the migration behav-
ior of Broad-winged Hawks at Hawk Mountain
Sanctuary. Pennsylvania. WiLson Bulletin 1 13:350-353.

McGrady, M. j., T. L. Maechtle. J. J. Vargas, W. S.
Seegar, and M. C. Porras Pena. 2002. Migration
and ranging of Peregrine Falcons wintering on the
Gulf of Mexico coast. Tamaulipas, Mexico. Condor
104:39-48.

Medica, D. L., R. Clauser, and K. Bildstein. 2007.
Prevalence of West Nile virus antibodies in a breeding
population of American Kestrels (Fatco sparverius) in
Penn.sylvania. Journal of Wildlife Diseases 43:538-
541.

Mills, G. S. 1975. A winter population study of the
American Ke.strel in central Ohio. Wifson Bulletin
87:241-247.

Millsap, B. a. and C. Bear. 2000. Density and
reproduction of Burrowing Owls along an urban
development gradient. Journal of Wildlife Manage-
ment 64:33^1.

Mueller, H. C.. N. S. Mueller. D. D. Berger, G. Allez,
W. Robichaud, and j, L. Kaspar. 2000. Age and .sex
differences in the timing of fall migration of hawks
and falcons. Wifson Bulletin 112:214-224.

Nelson, L. J.. D. R. Ander,son, and K. P. Burnham.
1980. The effect of band loss on estimates of annual
survival. Journal of Field Ornithology 51:30-38.

Nemeih, N., D. Gould, R. Bowen, and N. Komar. 2006.
Natural and experimental West Nile virus infection in
five raptor species. Journal of Wildlife Diseases 42: 1-13.

Nichols, J. D., K. J. Reinecke, and J. E. Hines. 1983.
Eactors affecting the distribution of Mallards wintering
in the Mississippi Alluvial Valley. Auk 100:932-946.

Powers, L. R. 1996. Wintering Sharp-shinned Hawks
(Accipiter striatus) in an urban area of southwestern
Idaho. Northwestern Naturalist 77:9-13.

Roest, a. I. 1957. Notes on the American Sparrow Hawk.
Auk 74:1-19.

Sandercock, B. K. 2006. Estimation of demographic
parameters from live-encounter data: a summary
review. Journal of Wildlife Management 70: 1504-1520.

SAS Institute Inc. 2003. Statistical analysis system.
Version 9.1.3. SAS Institute Inc., Cary, North
Carolina, USA.

Smallwood, J. A. 1988. A mechanism of sexual segrega-
tion by habitat in American Kestrels (Falco sparverius)
wintering in south-central Florida. Auk 105:36^6.

Smallwood, J. A. and D. M. Bird. 2002. American
Kestrel (Falco sparverius). The birds of North
America. Number 602.

Spaans, a. L. 1977. Are starlings faithful to their
individual winter quarters? Ardea 65:83-87.

Steenhof, K., M. R. Fuller, M. N. Kochert, and K. K.
Bates. 2005. Long-range movements and breeding
dispersal of Prairie Falcons from southwest Idaho.
Condor 107:481-496.

Steenhof, K., K. K. Bates, M. R. Fuller, M. N.
Kochert, J. O. McKinley, and P. M. Lukacs.
2006. Effects of radiomarking on Prairie Falcons:
attachment failures provide insights about survival.
Wildlife Society Bulletin 34:1 16-126.

Sullivan, B. L. and C. L. Wood. 2005. A plea for the
common birds. North American Birds 59:18-30.

Tabb, E. 1977. Winter returns of American Kestrels in
southern Florida. North American Bird Bander 2:163.

Temple, S. A. 1972. Systematics and evolution of the North
American Merlins. Auk 89:325-338.

Tordoff, H. B. and P. T. Redig. 1997. Midwest Peregrine
Falcon demography, 1982-1995. Journal of Raptor
Research 31:339-346.

U. S. Census Bureau (USCB). 2009. State and county
QuickFacts. Cape Coral, Florida. U.S. Census Bureau,
Washington, D.C., USA, http://quickfacts.census.gov/
qfd/states/1 2/1 2 1 0275.html.

Wesemann, T. 1986. Factors influencing the distribution
and abundance of Burrowing Owls (Athene cunicu-
laria) in Cape Coral. Florida. Thesis. Appalachian
State University. Boone. North Carolina. USA.

White, G. C. and K. P. Burnham. 1999. Program MARK:
survival estimation from populations of marked
animals. Bird Study 46:120-139.

Yoder, J. M., E. A. Marsciiall. and D. A. Swanson.
2004. The cost of di.spersal: predation as a function of
movement and site familiarity in Ruffed Grouse.
Behavioral Ecology 15:469^76.

Zeiss, B. 1983. The other side of the river: historical Cape
Coral, B. Zeiss, Cape Coral. Florida. USA.

The Wilson Journal of Ornithology 122(3):484-493, 2010

TEMPORAL AND SPATIAL SHIFTS IN HABITAT USE BY BLACK
BRANT IMMEDIATELY FOLLOWING FLIGHTLESS MOLT

TYLER L. LEWIS,' -' PAUL L. FLINT,' JOEL A. SCHMUTZ,' AND DIRK V. DERKSEN'

ABSTRACT. — Each year thousands of Pacific Black Brant (Branta bernicla nigricans) undergo flightless wing molt in
the Teshekpuk Lake Special Area (TLSA), Alaska, in two distinct habitats: inland, freshwater lakes and coastal, brackish
wetlands. Brant lose body mass during wing molt and likely must add reserves upon regaining flight to help fuel their
2,500 km migration to autumn staging areas. We characterized movements and habitat use by Brant during post-molt (the
period immediately following the recovery of flight) by (1) marking individual Brant with GPS (global positioning system)
transmitters, and (2) conducting a series of replicate aerial surveys. Individuals molting in inland habitats promptly
abandoned their molt wetland during the post-molt and moved into coastal habitats. Consequently, inland habitats were
nearly deserted by early August when Brant had regained flight, a decrease of >5,000 individuals from the flightless period
of early July. Conversely, coastal molting Brant largely remained in coastal habitats during the post-molt and many coastal
wetlands were occupied by large flocks (>1,000 birds). Our results indicate that inland, freshwater wetlands were less
suitable post-molt habitats for Brant, while coastal wetlands were preferred as they transitioned from flightless molt. The
immediacy with which Brant vacated inland habitats upon regaining flight suggests that food may be limiting during molt
and they are not selecting inland molt sites strictly for food resources, but rather a balance of factors including predator
avoidance and acquisition of protein for feather growth. Our data clearly demonstrate that patterns of habitat use by Brant in
the TLSA change over the course of the molt season, an important consideration for management of future resource
development activities in this area. Received 27 July 2009. Accepted 17 February 2010.

Failed- and non-breeding individuals of many
Northern Hemisphere goose species migrate to
distantly-located molting areas to undertake a
flightless wing molt (Hohman et al. 1992, Kjellen
1994). The flightless molt period imposes a
number of constraints, including restricted ability
to escape predators or access unused habitats and
food sources (Fox and Kahlert 2000, Kahlert
2006). Growth of new flight feathers requires
intake of certain nutrients essential to feather
development, especially proteins and sulfur-con-
taining amino acids (Murphy and King 1984,
Sedinger 1984, Heitmeyer 1988, Fox and Kahlert
1999). Fox and Kahlert (2005) suggested that
geese alleviate these constraints by choosing wing
molting habitats that balance low predation risk
with food supplies high in protein and energy. The
constraints of the flightless period are immediate-
ly relaxed upon regaining flight, however, and
habitat attributes necessary for molting, including
open-water escape areas or availability of nutri-
ents for feather growth, may no longer drive
habitat selection.

Northern Hemisphere geese migrate soon after
completing their molt, often traversing consider-

' U..S. Geological Survey. Alaska Science Center, 4210
University Drive. Anchorage, AK 99508, USA.

^Current address: Institute of Arctic Biology, and
Department of Biology and Wildlife, University of Alaska,
31 1 Irving Building, Fairbanks, AK 99775, USA.
’Corresponding author; e-mail: tlewis@usgs.gov

able distances from arctic molting sites to more
southerly staging and wintering areas (Hohman et
al. 1992). However, because the flightless wing
molt is an energetically costly event and geese
often lose considerable body mass during the molt
(Williams and Kendeigh 1982, Taylor 1993, Fox
and Kahlert 2005, Portugal et al. 2007), geese
likely need to accumulate new fat reserves to fuel
their forthcoming migration. Thus, post-molt
habitats may be chosen largely for food availabil-
ity and, when a molt site offers sufficient food
availability for birds to build reserves, individuals
may opt to stay at that site beyond the time at
which flight was recovered. Conversely, abrupt
abandonment of a molt site following recovery of
flight may suggest food is limiting and alternate
sites offer superior food supplies, conferring a
selective advantage to birds that change sites.

Each year, thousands of Pacific Black Brant
(Brantct bernicla nigricans\ hereafter Brant)
migrate to the Teshekpuk Lake Special Area
(TLSA), on the Arctic Coastal Plain of Alaska,
where they undergo flightless wing molt (Derksen
et al. 1982, Taylor 1995). TLSA and sun'ounding
areas provide Brant with a choice beJween two
distinct molting habitats in close proximity:
coastal, brackish wetlands or inland, freshwater
wetlands (Flint et al. 2008). Each habitat offers
different plant species as potential forage for
molting Brant, as well as different types of open-
water escape habitat (Markon and Derksen 1994,

484

L('u/.v et al. • POST-MOLT ECOLOGY OF BLACK BRANT

485

70°50'0"N-

70"40'0"N'

153°30’0"W

153"0’0"W

152"30'0"W

-70°50’0'’N

-70“40’0"N

152WW

FIG. 1. Molting goose region of the Teshekpuk Lake Special Area, Alaska, during 2006-2008. Lakes at which Brant
were aerially surveyed are darkened and strata boundaries are depicted with bold lines.

Flint et al. 2008). We examined the post-molt
ecology of Brant in the TLSA to better understand
the potential differences in food availability and
quality between coastal versus inland habitats, as
well as preference of Brant for each habitat type
following the recovery of flight. We used a
combination of population-level aerial surveys
and individuals marked with GPS (global posi-
tioning system) transmitters to examine changes
in Brant distribution from molt to post-molt,
probability of remaining on a molt site following
recovery of flight, and types of habitats used post-
molt.

METHODS

Classification of Coastal versus Inland Molt-
er. — The TLSA is characterized by tundra vege-
tation, continuous permafrost, and large perma-
frost thaw lakes (Markon and Derksen 1994,
Jorgensen et al. 2006). Flightless molting Brant
occur primarily in the northeast section (Fig. 1)
and are primarily subadults, non-breeders, and
failed-breeders associated with breeding colonies

in Alaska, Russia, and western Canada (Bollinger
and Derksen 1996). Flint et al. (2008) divided the
TLSA molting area into three strata based on
distinct vegetative and landscape characteristics.
The majority of our research was conducted in
strata two (inland stratum) and three (coastal
stratum; Lig. 1 ). The coastal stratum is character-
ized by the occuirence of saltwater intrusion,
elevations <1.5 m, and salt-tolerant plant com-
munities, while the inland stratum has a lack of
saltwater intrusion, slightly greater elevations
(^4 m), and predominantly freshwater plant
communities. Brant marked with transmitters
were classified as either inland or coastal molters
depending upon the stratum in which they molted.
Additionally, some transmitter-marked Brant
molted in areas near the TLSA, but outside the
stratum boundaries. This latter group molted
solely in coastal habitats and is included in the
coastal molter classification.

GPS Transmitters. — Our first method for char-
acterizing movements and habitat use of Brant
during the post-molt was through use of GPS

486

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 3. September 2010

transmitters. We captured female Brant from
nesting colonies on the Arctic Coastal Plain of
Alaska, between Barrow (71° 18' N, 156° 44' W)
and the Colville River Delta (70° 25' N, 150° 23'
W), during two summers (2007-2008). We
captured birds on nests using spring-loaded bow
traps (Sayler 1962) and marked them with GPS
transmitters. We removed eggs from nests of
captured females changing their status to failed
breeder and significantly increasing the likelihood
they would migrate to the TLSA to undergo
flightless wing molt. We marked 39 Brant from
11 to 17 June 2007 and 39 Brant from 12 to 20
June 2008. We experienced high rates of trans-
mitter failure in both years of our study as verified
via incidental recaptures of Brant with failed
transmitters, recovery of defective transmitters
from hunter-harvested individuals, and sudden
loss of transmitter signals from flightless individ-
uals (i.e., birds that could not move from our
reception range). Thus, missing transmitters are
largely attributable to transmitter failure rather
than molt migration from our study area (Lewis
and Flint 2008).

GPS transmitters were externally affixed be-
tween the shoulder blades of captured Brant via
dual subcutaneous anchors, following Lewis and
Flint (2008). We used two models of GPS
transmitters in 2007 that differed only in power
source: solar (/? = 19) and battery powered (n =
20). All 39 GPS transmitters used in 2008 were
battery-powered. Estimated transmitter life was
45-60 days for battery-powered transmitters and
45-i- days for solar models. GPS transmitters were
programmed to collect and store a location every
6 hrs and offload all stored locations via a daily
VHF radio transmission. Offloaded data were
collected weekly from fixed-wing aircraft
equipped with wing-mounted two-element yagi
antennas. Data were downloaded using an
HR2600 receiver produced by HABIT Research
(Victoria, BC, Canada).

Aerial Surveys. — Aerial surveys, our second
method for characterizing Brant movements and
habitat use, encompassed most of the molting
Brant population in the TLSA. We conducted
replicate aerial surveys of 33 selected lakes within
the TLSA during summers 2006-2008 (Fig. 1).
The initial survey was during the first week of
July and repeated on a weekly basis for 6 weeks,
encompassing the core molting period of Brant
(Derksen et al. 1982). The survey protocol
followed that used by the U.S. Fish and Wildlife

Service (USFWS) in their annual mid-July survey
of molting geese in the TLSA as described by
Flint et al. (2008). This protocol assumes that
detection rate and observer error are constant on
all surveys. We used the USFWS survey data
from 1996 to 2005 to select the subsample of
lakes for our repeated survey, choosing only lakes
that were used by >200 Brant in >3 years. This
resulted in a subsample of 33 lakes which
typically contained >90% of the molting Brant
counted in these years. All surveyed lakes were
within the boundaries of either the inland or
coastal stratum and were classified accordingly
(Flint et al. 2008).

Statistical Analyses. — We calculated two vari-
ables using data from our birds marked with GPS
transmitters to explore Brant post-molt ecology:
probability of molt site abandonment and distance
from coastline. Both variables are time-depen-
dent. being measured in relation to time since
onset of molt, and thus required that start date of
molt be known. We used an algorithm, which
assumed movement rates of Brant are significant-
ly reduced during the flightless molt, to estimate
each individual’s start date of molt. The algo-
rithm, described by Lewis et al. (2010), identified
the date at which movement rates distinctly
changed, allowing us to objectively define a
precise date of molt initiation for each of our
transmitter-marked birds. We did not use this
algorithm to estimate the end date of molt because
many individuals did not increase their movement
rates or move among wetlands despite presumably
regaining flight. We used estimates from previous
Brant research in the TLSA which found that
partial flight is regained —23 days after shedding
of flight feathers (Taylor 1995). We assumed for
all analyses that Brant had regained partial flight
by 23 days from the onset of flightlessness.

We quantified timing of molt site abandonment
for each of our transmitter-marked individuals.
Most wetlands used during molt are separated by
wide stretches of upland tundra which are
essentially impassable to flightless Brant; thus,
individuals are often confined to one wetland, or
interconnected wetlands, until flight is regained.
Individuals were assigned a primary molt wetland
from the time molt started, as calculated by our
molt algorithm, and their daily status classified as
“present” until they moved to a different,
unconnected wetland, at which point their status
was classified as “abandoned”. This exercise
produced a binary variable for each of our

Lewis ei cil. • POST-MOLT ECOLOGY OF BLACK BRANT

487

transmitter-marked individuals, where each day
since the onset of molt was classified as either
present or abandoned. We used this binary
variable in Kaplan-Meier survival models to
estimate the daily probability of molt site
abandonment for coastal and inland molters. The
Kaplan-Meier analysis produced separate proba-
bility curves for coastal and inland molters with
each curve plotting the probability that Brant
remained on their primary molt wetland versus
time since onset of molt. We used a log-rank test
of equality to examine the null hypothesis (a =
0.05) that there was no difference in probability
curves of coastal versus inland molters (Pollock et
al. 1989).

We modeled the distance from coastline in
relation to time since onset of molt to examine if
coastal versus inland molters differed in their
post-molt habitat use. We fit a series of general
linear models (GLM) to describe the shape of time
trends for the explanatory variable of ‘distance
from coast’. Models examined included intercept,
linear, quadratic, and cubic time function with the
best-fitting model selected using Akaike’s Infor-
mation Criterion (AIC; Burnham and Anderson
2002). The start of molt was defined as time zero
with time (hrs) of all subsequent GPS locations
added onto zero and all prior GPS locations
subtracted from zero. The cubic function was the
best-fitting time model (AAIC = 0) and all other
models had a AAIC < 4.0. Thus, using ‘distance
from coast’ as our explanatory variable, we next
fit a series of general linear mixed models in
which the cubic time function was included in all
models. We used mixed models to account for
repeated measures of individuals marked with
transmitters by including subject as a random
effect (Littell et al. 2000). The mixed models
included the response variable ‘habitat’, defined
as either coastal or inland depending upon the
habitat strata in which each individual molted.
The model set included a time only model, habitat
as an additive variable with time, or all combi-
nations of habitat by time interactions for a total
of nine models. The best-fitting model was
selected using AAIC and the null model (inter-
cept-only) was compared against each model to
assess model fit (Burnham and Anderson 2002).

We examined the spatial distribution of Brant
across strata by summarizing each aerial survey as
the ratio of Brant counted inland to Brant counted
coastally and used this ratio as our response
variable to fit a candidate set of GLMs. We used

FIG. 2. The probability individuals remained on their
molt wetland by number of days since start of molt for
Brant marked with GPS transmitters which molted at
coastal versus inland wetlands, Teshekpuk Lake Special
Area, Alaska, during 2007-2008. Brant typically regain
flight 23-24 days after onset of molt.

year (2006, 2007, 2008) and Julian date of each
survey as our explanatory variables, and our
candidate model set consisted of single variable
models, the additive model, the interaction model,
and the intercept-only null model for a total of
five models. Akaike’s information Criterion
(AICc), adjusted for small sample size, was
calculated for each model in the candidate set,
AAICc and AIC weights (vc,) were used to infer
the relative support of each model, and the null
model (intercept-only) was compared against each
model to assess model fit. All analyses were
performed using SAS Version 9.1 statistical
software (SAS Institute 2003) and descriptive
statistics are presented as mean ± SE.

RESULTS

GPS transmitters on 23 Brant functioned for
^23 days from onset of flightlessness allowing us
to examine Brant behavior as they transitioned
from their flightless molt. Sixteen of these 23
birds were classified as coastal molters and seven
as inland molters. The probability that an
individual remained on its molt wetland on a
given day was significantly different for coastal
versus inland molters (log-rank test, = 7.73, df
= \, P = 0.005), as inland molters abandoned
their molt site sooner than coastal molters
(Fig. 2). Abandonment of molt sites by inland
molters began 22 days after molt initiation and the
probability of remaining at their molt site was
21% at 28 days. Abandonment of molt sites by
coastal molters began 25 days after molt initiation

488

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 3, September 2010

TABLE 1. Candidate general linear mixed models evaluating variation in distance from coast (m) for Brant with GPS
transmitters in the Teshekpuk Lake Special Area, Alaska, 2007-2008. The explanatory variables are time (t), included as a
cubic function (t -h F + F) in all models, and habitat (h) defined as coastal or inland. Models are listed in order of AAIC.

Model

Parameters

Log-likelihood

AAIC

Wj

h + t + ht + t- -b ht- + t-"'

10

-3,5927.29

0

0.37

h + t + ht + F + t-’ + hF

10

-3,5927.34

0.09

0.36

h + t + ht + F + hF + F + ht^

11

-3,5926.85

1.11

0.21

h + t + t- + t^ + ht^

9

-3,5930.88

5.17

0.03

h + t + t- + ht- + 0

9

-3.5931.61

6.65

0.01

h + t + t- + hF + F + ht^

10

-3,5930.83

7.09

0.01

h + t + ht + F + F

9

-3.5937.37

18.16

0.00

h + t + t- + t^

8

-3,5938.63

18.67

0.00

t + F + t'

7

-3,5947.15

33.71

0.00

and the probability of remaining at their molt site
was 86% at 28 days.

The best-fitting model (AAIC = 0) to explain
distance from the coast for Brant marked with
GPS transmitters contained the explanatory vari-
able of habitat, defined as coastal or inland
molter, in the habitat*time {ht) and habitat*
time*time (ht^) terms of the cubic function
(Table 1). Two other models also received
substantial support (AAIC 2): the model with
habitat in the ht and ht’ terms and the model with
habitat in all three terms of the cubic time
function (Table 1). The null model, in which the
habitat variable was not included, received no
empirical support indicating that inclusion of
habitat explains a significant amount of variation
in distance to coast (Table 1). Plots of predicted
values for each of the three well-supported models
indicate they differ only at the extreme tails of the
cubic functions. The biological inference for each
of the three well-supported models is identical and
we used model-averaged estimates, based on the

three top models, for all further inference. Model-
averaged parameter estimates (Table 2) indicate
that by 23 days after initiation of molt, when
many Brant have regained flight, inland molting
Brant promptly moved into coastal habitats, while
coastal molting Brant remained in coastal habitats
(Fig. 3). Inland and coastal molters were at
similar distances from the coast before the onset
of molt and were ~8 and 2 km from the coast
during the flightless molt, respectively (Fig. 3).

The best-fitting model (AAIC^ = 0) to explain
the proportion of Brant per habitat strata, as
counted from aerial surveys, contained only Julian
date as a response variable (Table 3). The additive
model containing year and Julian date received
moderate support, although its vv, was low relative
to the top model (Table 3). All other models,
including the null model, had AAIC^. values >20.
These results indicate the distribution of Brant
across habitat strata was strongly influenced by
Julian date. The proportion of Brant in the coastal
stratum increased by date beginning in mid to late

TABLE 2. Weighted parameter estimates and unconditional standard error (SE) for general linear mixed models
evaluating variation in distance from coast (m) for Brant with GPS transmitters in the Teshekpuk Lake Special Area,
Alaska. 2007-2008.

Explanatory variable'

Parameter estimate

Unconditional SE

Intercept

7,572.67

947.77

Habitat

-5,350.29

1.074.16

Time

3.48

1.69

Habitat*Time

-5.19

1.88

Time^

-1.48*10

5.87*10 ’

Habitat*Time^

6.21*10 ’

3.98*10 ■’

Time’

-3.47*10

6.81*10

Habitat*Time’

5.47*10

3.67*10

" Habital is a categorical variable (coastal or inland) with inland as the reference value.

Lewis el al. • POST-MOLT ECOLOGY Ol- BLACK BRANT

489

Pr^-moK Flightless moK Fllghtregained

Days since start cff molt

FIG. 3. Model averaged estimate of distance from
coastline (in) by days since start of molt for Brant marked
with GPS transmitters which molted at coastal versus inland
wetlands, Teshekpuk Lake Special Area, Alaska,
during 2007-2008.

FIG. 4. The proportion of Brant in coastal versus inland
strata based on aerial surveys, Teshekpuk Lake Special
Area, Alaska, conducted during July-early August, 2006-
2008. Aerial surveys were conducted six times per year
with each survey separated by ~1 week. Weekly surveys
are summarized across years, producing weekly means
± SE.

July as they began to regain flight (Fig. 4). This
pattern of movement into the coastal stratum did
not differ across years or by the interaction of
year*date. Most Brant had regained flight by early
August and >90% were in the coastal stratum, a
30% increase from the flightless molt period
(Fig. 4). The increased proportion of Brant in the
coastal stratum did not result from inland molters
leaving the TLSA altogether. Rather, the number
of birds in the coastal stratum increased or
remained constant during late July, despite some
coastal molters leaving the area. Thus, many
inland molters moved into coastal stratum habitats
upon regaining flight, similar to the pattern
observed for our individuals marked with trans-
mitters. Aerial surveys, however, demonstrated
this pattern for much larger numbers of Brant

TABLE 3. Candidate general linear models evaluating
variation in ratio of Brant counted inland versus those
counted coastally during aerial surveys conducted in the
Teshekpuk Lake Special Area, Alaska, 2006 to 2008.
Models are listed in order of AAIC.

Model

Parameters

R'

AAIC

B7

Julian date

3

0.86

0

0.68

Julian date + Year

Julian date + Year

5

0.90

1.53

0.32

+ Julian date*Year

8

0.91

21.41

0

Null

2

0

28.00

0

Year

4

0.02

34.38

0

(>15,000 annually) and over a larger spatial
area.

Post-molt habitat use varied among wetlands
within the coastal stratum with some being
completely abandoned by Brant during the post-
molt while others experienced increases of
hundreds to thousands of Brant. Numbers of
Brant either increased or remained stable at 40%
of the 22 coastal wetlands for which sufficient
aerial survey data were available during the post-
molt period; conversely, post-molt Brant numbers
decreased significantly at all nine inland stratum
wetlands surveyed. The coastal wetland complex
at the Smith River (Fig. 1 ) was particularly
important, supporting >5,000 Brant annually
during the post-molt, an increase of 2,000-3,000
birds above the wetland’s average annual molting
population.

DISCUSSION

Brant regain flight when their primaries reach
70-75% of their full-grown length, which typi-
cally occurs 23-24 days after onset of molt
(Taylor 1995). Using this timeline, we found that
individuals molting in inland, fre.shwater habitats
abandoned their molt site almost immediately
upon regaining flight, moving almost exclusively
into coastal habitats. Inland wetlands were nearly
abandoned by early August, a decrease of >5,000
individuals from the flightless period of early
July, and the only remaining individuals were
likely those which had yet to attain flight.

490

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3. September 2010

Conversely, Brant molting in coastal habitats
continued to reside on their molt wetlands for
several days after regaining flight and did not
switch habitats during the post-molt, remaining
within coastal habitats. Many coastal wetlands
were occupied by large flocks (>1,000 birds) of
flying Brant during the post-molt. Our results
indicate Brant almost universally prefer coastal
habitats immediately following recovery of flight
and inland, freshwater wetlands and lakes are less
suitable post-molt habitats. Our results also
corroborate earlier research from the 1980s, in
which a small sample (/? = 18) of radio-marked
Brant moved from inland to coastal sites during
the post-molt period (Jensen 1990). This suggests
the post-molt patterns observed in our study have
persisted over a timescale spanning no less than
multiple decades.

Brant molting in inland habitats chose to leave
their molt site sooner than those molting coastally,
but the differential timing of molt site abandon-
ment could alternatively be explained by differ-
ential molt durations between the two habitat
types. Habitat-specific factors, such as food
quality, competition for resources, or presence of
other stressors may influence the rate of feather
growth and, by extension, the duration of molt
(Hahn et al. 1992, Romero et al. 2005). Citril
Finches {Carduelis citrinella) in higher quality
habitats molted more rapidly than those in low
quality habitats, despite the two habitats being
only 5 km apart (Borras et al. 2004). Similarly, the
longer residency of Brant at coastal molt sites may
be due to hypothetically poorer habitat quality at
these sites, rather than individuals choosing to
remain past the recovery of flight. We cannot
completely discount this alternative hypothesis,
but we find it unlikely that quality differences
between coastal versus inland habitats would shift
molt durations to the extent that explains our
observed differences in timing of molt site
abandonment. Abandonment of coastal molt sites
lagged that of inland sites by >5 days, a
pronounced difference for a species that averages
23 days to regain flight, and >50% of coastal-
molting individuals had not moved by day 30. The
majority of Brant molted coastally and inland
molting Brant promptly moved to coastal habitats
following cessation ot molt, suggesting that
coastal habitat quality is at least equivalent with
inland habitats, if not superior. Overall, the
patterns of post-molt movements and habitat use
by Brant would not seem to support the

hypothesis that variation in time spent at molting
locations is negatively associated with habitat
quality.

Adult Brant collected in the TLSA from 1988
to 1989 lost >20% of their body mass and >80%
of their lipid reserves during the flightless wing
molt, completing molt with lipid reserves com-
prising only 2-4% of body mass (Taylor 1993).
More recent research (2005-2007) also docu-
mented considerable body mass losses for molting
Brant, although rates of mass loss were slower
than during 1988-1989 (PLF, unpubl. data).
Given significant declines in body mass and lipid
reserves, molting Brant must regain body reserves
during the post-molt period to fuel their forth-
coming 2,500 km migration to their primary
autumn staging area at Izembek Lagoon, Alaska
(Vangilder et al. 1986, Dau 1992, Reed et al.
1998). Post-molt distributions of Brant are likely
driven by availability and quality of food
resources, and the unidirectional movement of
Brant from inland habitats suggests that inland
food resources are less able to fulfill post-molt
requirements than coastal resources. The imme-
diacy with which Brant vacated inland stratum
habitats upon regaining flight also suggests food
is limiting during molt, potentially due in part to
Brant grazing pressure, and Brant are not selecting
inland molt sites strictly for food resources. Brant
are likely selecting inland molt sites for a balance
of factors which includes protein supplies for
feather growth and large open-water areas for
predator avoidance (Fox and Kahlert 2005).

The sedge Corex suhspathacea and grass
Pucciiiellio phyrognodes grow in saline-influ-
enced tundra habitats (Markon and Derksen
1994) and are preferred forage items of Brant
across the arctic and subarctic (Sedinger et al.
2001, Person et al. 2003). These plant species in
the coastal habitats of the TLSA often occur as
‘grazing lawns’, which are areas of high density
plant growth created by frequent goose grazing
(Person et al. 2003). Grazing lawns typically
experience short-term reductions in food abun-
dance due to the immediate effects of grazing, but
their long-term maintenance requires frequent
grazing to create positive feedbacks of increased
plant growth rates and nutrient content (Ruess. et
al. 1997, Person et al. 2003). The nitrogen
recycled through goose feces contributes to the
salinity tolerance of grazing lawns (Ruess et al.
1997). The pronounced post-molt shift of Brant
into coastal habitats likely contributes to mainte-

Lewis ct a!. • POST-MOLT ECOLOGY OF BLACK BRANT

491

nance and expansion of grazing lawns, increasing
coastal tood availability and further prompting
Brant to exploit coastal habitats. The coastal
plants grazed by Brant also have lower carbon to
nitrogen ratios (C:N), the main index of food
quality for geese (Riddington et al. 1997, Hassall
et al. 2001), than the common inland stratum
forage plants of Carex aquatilis and Deschompsia
caespitosa (JAS, unpubl. data). Brant, because of
their small body size and con'espondingly small
digestive tract, have high food passage rates and
require high quality foods from which nutrients
can be quickly extracted (Prop and Vulink 1992,
Sedinger 1997). Coastal areas in the TLSA are
heavily influenced by summer fogs, which locally
reduce air temperature and solar exposure,
potentially delaying plant phenology (Post and
Stenseth 1999, Menzel et al. 2006). Plant quality
decreases as the arctic summer progresses (i.e.,
C;N temporally increases), and delayed phenolo-
gy in coastal areas would result in higher quality
forage plants during the late season post-molt
period (Crawley 1983, Van der Wal et al. 2000).
The combination of higher quality plant species
and delayed plant phenology in coastal habitats
may also be important factors contributing to the
post-molt shift of Brant into coastal habitats.

The TLSA also supports large numbers of
molting Greater White-fronted {Anser albifrons),
Canada (Brcinta canadensis). Cackling (B. hutch-
insii), and Lesser Snow (Chen caendescens
caerulescens) geese along with Brant. The
number of molting geese counted in the TLSA
in recent years has ranged from 70,000 to 90,000,
collectively making this one of the most signif-
icant goose molting areas in the circumpolar arctic
(Flint et al. 2008). The TLSA is currently
designated for future leasing for oil and gas
extraction (2008 Northeast NPR-A Record of
Decision; available at http://www.blm.gov/ak/st/
en/prog/energy/oil_ga.s/npra.html). Planning to
minimize the effects of development of molting
areas on Brant populations will require a clear
understanding of patterns of habitat use and how
patterns may change seasonally. The only data
currently available to assess patterns of habitat use
of Brant in the TLSA comes from a single annual
survey conducted during the peak flightless period
in mid-July (Flint et al. 2008). These survey data
are not suited to describing intra-annual patterns
of habitat use, such as the post-molt habitat shifts
we describe. Our data clearly demonstrate that
patterns of habitat use by Brant in the TLSA

change over the course of the molt sea.son, and
that future management and development deci-
sions must consider seasonal variation in distri-
butions of Brant.

ACKNOWLEDGMENTS

The Bureau of Land Management and U.S. Geological
Survey, Alaska Science Center, provided funding and
assisted with logistics. D. A. Nigro provided logistic
support and helicopter management during captures. D. C.
Douglas developed our algorithm for estimating molt
initiation. Blake Bartzen assisted with captures. R. B.
Lanctot provided housing in Barrow. J. S. Hamilton of
Arctic Air Alaska provided aerial support. This paper
benefited from the reviews of D. A. Nigro and T. F.
Fondell. Use of trade, product, or company names is solely
for descriptive purposes and does not imply endorsement or
criticism by the U.S. government.

LITERATURE CITED

BoLLtNGER. K. S. AND D. V. Derksen. 1996. Demographic
characteristics of molting Black Brant near Teshekpuk
Lake, Alaska. Journal of Field Ornithology 67:141-
158.

Borras, a., T. Cabrera, J. Cabrera, and J. C. Senar.
2004. Interlocality variation in speed of moult in the
Citrial Finch Serinus citronella. Ibis 146:14—17.
Burnham, K. P. and D. R. Anderson. 2002. Model
selection and multi-model inference: a practical
information-theoretic approach. Second Edition.
Springer- Verlag, New York, USA.

Crawley, M. J. 1983. Herbivory: the dynamics of animal-
plant interactions. Blackwell Scientific Publications.
Oxford, United Kingdom.

Dau, C. P. 1992. The fall migration of Pacific Fly way Brent
Branta bernicla in relation to climatic conditions.
Wildfowl 43:80-95.

Derksen, D. V., W. D. Eldridge, and M. W. Weller.
1982. Habitat ecology of Pacific Black Brant and other
geese moulting near Teshekpuk Lake. Alaska. Wild-
fowl 33:39-57.

Fi.int, P, L., E. j. Mallek, R. j. King. J. A. Schmutz. K. S.
Bollinger, and D. V. Derksen. 2008. Changes in
abundance and spatial distribution of geese molting
near Teshekpuk Lake, Alaska: interspecific com-
petition or ecological change? Polar Biology 31:549-
556.

Fox, A. D. and .1. Kahlert. 1999. Adjustments to nitrogen
metabolism during wing moult in Greylag Geese.
Anser anser. Functional Ecology 13:661-669.

Fox. A. D. AND J. Kahlert. 2000. Do moulting Greylag
Geese An.ser anser forage in proximity to water in
respon.se to food availability and/or quality? Bird
Study 47:266-274.

Fox, A. D. and .1. Kahlert. 2005. Changes in body mass
and organ size during wing moult in non-breeding
Graying Geese Anser an.ser. Journal of Avian Biology
36:538-548.

Hahn, T. P., J. Swingle. J. C. Wingfield, and M.

492

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122. No. 3, September 2010

Ramenofsky. 1992. Adjustments of the prebasic
molt schedule in birds. Ornis Scandinavica 23:314-
321.

Hassall, M., R. Riddington, and A. Helden. 2001.
Foraging behaviour of Brent Geese, Branta b.
bernicla. on grasslands: effects of sward length and
nitrogen content. Oecologia 127:97-104.

Heitmeyer, M. E. 1988. Protein costs of the prebasic molt
of female Mallards. Condor 90:263-266.

Hohman, W. L., C. D. Ankney, and D. H. Gordon. 1992.
Ecology and management of postbreeding waterfowl.
Pages 128-189 in Ecology and management of
breeding waterfowl (B. D. J. Batt, A. D. Afton, M.
G. Anderson, C. D. Ankney, D. H. Johnson, J. A.
Kadlec, and G. L. Krapu, Editors). University of
Minnesota Press, Minneapolis, USA.

Jensen, K. C. 1990. Responses of molting Pacific Black
Brant to experimental aircraft disturbance in the
Teshekpuk Lake Special Area, Alaska. Dissertation.
Texas A&M University, College Station, USA.
Jorgensen, M. T., Y. L. Shur, and E. R. Pullman. 2006.
Abrupt increase in permafrost degradation in arctic
Alaska. Geophysical Research Letters 33:L02503,
doi: 10. 1029/2005GL024960.

Kahlert, j. 2006. Factors affecting escape behaviour in
moulting Greylag Geese An.ser anser. Journal of
Ornithology 147:569-577.

Kjellen, N. 1994. Moult in relation to migration in birds —
a review. Ornis Svecica 4:1-24.

Lewis, T. L. and P. L. Flint. 2008. Modified method for
external attachment of transmitters to birds using two
subcutaneous anchors. Journal of Field Ornithology
79:336-341.

Lewis, T. L., P. L. Flint, J. A, Schmutz, and D. V.
Derksen. 2010. Pre-moult patterns of habitat use
and moult site selection by Brent Geese Branta
bernicla nigricans’, individuals prospect for moult
sites. Ibis 152: 556-568.

Littell, R. C., G. a. Milliken, W. W. Stroup, and R. D.
WoLFiNGER. 2000. SAS System for mixed models.
Fourth Edition. SAS Institute, Cary, North Carolina,
USA.

Markon, C. J, and D. V. Derksen. 1994. Identification of
tundra land cover near Teshekpuk Lake, Alaska using
SPOT satellite data. Arctic 47:222-231.

Menzel, a., T. H. Sparks, N. Estrella, E. Koch, A.
Aasa, R. Ahas, K. Alm-Kubler, P. Bissolli, O.
Braslavska, a. Briede, F. M. Chmielewski, Z.
Crepinsek, Y. Curnel, a. Dahl, C. Defila, A.
DrtNNELLY, Y. Filella, K. Jatczak, F. Mage, A.
Mestre, 0. Nordli, j. Penuelas, P. Pirinen, V.
RemiSova. H. Scheifinger, M. Stritz, a. Susnik, a.
J. H. Van Vliet, F.-E. Wielgolaski, S. Zach, and A.
ZUST. 2006. European phenological response to
climate change matches the warming pattern. Global
Change Biology 12:1969-1976.

Murphy, M. E. and J. R. King. 1984. Sulfur amino acid
nutrition during molt in the White-crowned Sparrow.
Condor 86:314-332.

Pi-R.soN, B. T.. M. P. Herzog, R. W. Ruess, .1. S. Sedinger,
R. M. Anthony, and C. A. Babcock. 2003.

Feedback dynamics of grazing lawns: coupling
vegetation change with animal growth. Oecologia
135:583-592.

Pollock, K. H., S. R. Winterstein, C. M. Bunck, and P.
D. Curtis, 1989. Survival analysis in telemetry
studies: the staggered entry design. Journal of Wildlife
Management 53:7-15.

Portugal, S. J., J. A. Green, and P. J. Butler. 2007.
Annual changes in body mass and resting metabolism
in captive Barnacle Geese {Branta leucopsis): the
importance of wing moult. Journal of Experimental
Biology 210:1391-1397.

Post, E. and N. C. Stenseth. 1999. Climatic variability,
plant phenology, and northern ungulates. Ecology
80:1322-1339.

Prop, J. and T. Vulink. 1992. Digestion by Barnacle
Geese in the annual cycle: the interplay between
retention time and food quality. Functional Ecology
6:180-189.

Reed, A., D. H. Ward, D. V. Derksen, and J. S.
Sedinger. 1998. Brant (Branta bernicla). The birds of
North America. Number 337.

Riddington, R., M. Hassall, and S. J. Lane. 1997.
The selection of grass swards by Brent Geese
Branta b. bernicla’. interactions between food
quality and quantity. Biological Conservation 81:153-
160.

Romero, L. M., D. Strochlic, and J. C. Wingfield. 2005.
Corticosterone inhibits feather growth: potential
mechanism explaining seasonal down regulation of
corticosterone during molt. Comparative Biochemistry
and Physiology, Part A 142:65-73.

Ruess, R. W., D. D. Uliassi, C. P. H. Mulder, and B. T.
Person. 1997. Growth responses of Carex ramen.'ikii
to defoliation, salinity, and nitrogen availability:
implications for geese-ecosystem dynamics in western
Alaska. Ecoscience 4:170-178.

SAS Institute. 2003. SAS Version 9.1. SAS Institute,
Cary, North Carolina, USA.

Sayler, J. W. 1962. A bow-net trap for ducks. Journal of
Wildlife Management 26:219-221.

Sedinger, J. S. 1984. Protein and amino acid composition
of tundra vegetation in relation to nutritional require-
ments of geese. Journal of Wildlife Management
48:1128-1136.

Sedinger, J. S. 1997. Adaptations to and con.sequences of
an herbivorous diet in grouse and waterfowl. Condor
99:314-326.

Sedinger, J. S., M. P. Herzog, B. T. Person, M. T. Kirk,
T. Obritchkewitch, P. P. Martin, and A. A.
Stickney. 2001. Large-scale variation in growth of
Black Brant goslings related to food availability. Auk
1 18:1088-1095.

Taylor, E, J. 1993. Molt and bioenergetics of Pacific
Black Brant (Branta bernicla nigrican.s) on the Arctic
Coastal Plain, Alaska. Dissertation. Texas A&M
University, College Station, USA.

Taylor, E. J. 1995. Molt of Black Brant (Branta bernicla
nigricans) on the Arctic Coastal Plain, Alaska. Auk
1 12:904-919.

Van der Wal, R., N. Madan, S. van Lieshout, C.

Lewis el al. • POST-MOLT ECOLOGY OF BLACK BRANT

493

Dormann, R. Langvatn, and S. D. Albon. 2000.
Trading forage quality for quantity? Plant phenology
and patch choice by Svalbard reindeer. Oecologia
123:108-1 15.

Vangilder, L. D., L. M. Smith, and R. K. Lawrence.

1986. Nutrient reserves of premigratory Brant during
spring. Auk 103:237-241.

Williams, J. E. and S. C. Kendeigh. 1982. Energetics of
the Canada Goose. Journal of Wildlife Management
46:588-600.

The Wilson Journal of Ornithology 122(3);494-502, 2010

ARTHROPOD FORAGING BY A SOUTHEASTERN ARIZONA

HUMMINGBIRD GUILD

DONALD R. POWERS,'*' JESSAMYN A. VAN HOOK,'
ELIZABETH A. SANDLIN," AND TODD J. McWHORTER^-'

ABSTRACT. — We tested the hypothesis that foraging for arthropods may be a viable source of energy when
hummingbirds are competitively excluded from sources of nectar. We hypothesized that the Magnificent Hummingbird
(Eugenes fulgens) relies more upon arthropods than the Blue-throated Hummingbird (Lampornis clemenciae) or Black-
chinned Hummingbird (Archilochus alexandri) in southeastern Arizona. We were unable to quantify arthropod foraging by
A. alexandri, but measured frequent arthropod foraging by both E. fulgens and L. clemenciae. E. fulgens engaged in more
aerial llycatching than L. clemenciae, and their rate of flycatching attempts was higher than by L. clemenciae. Analysis of
gut contents showed that E. fulgens consumes the greatest diversity of arthropods. Respiratory quotient measurements
indicated E. fulgens catabolized a greater amount of fat/protein than the other species. Gut morphology of E. fulgens does
not appear to differ from other hummingbirds suggesting hummingbirds in general may have the ability to use arthropods as
an alternative energy source when access to floral energy is restricted. Our data are consistent with the hypothesis that the
diet of E. fulgens includes more arthropods than other species with which they compete. Received 10 November 2009.
Accepted 3 Eebruary 2010.

Floral nectar has been assumed to be the
primary energy resource of hummingbirds. This
is reasonable from both ecological and physio-
logical perspectives when one considers that
many aspects of hummingbird and floral natural
history are intertwined (Feinsinger 1983, Stiles

1995) , and that sugars in floral nectar are quickly
and easily digested. It is also known that
hummingbirds supplement their diets with small
arthropods as floral nectar lacks many important
nutrients (Baker 1977). However, arthropods are
generally considered to be of little energetic
importance for hummingbirds (Wolf and Hains-
worth 1971).

Hummingbirds often spend <10% of their day
foraging for arthropods when nectar is in
sufficient supply (Gass and Montgomerie 1981).
There is evidence that arthropod consumption
increases during specific periods of the natural
cycle of hummingbirds such as during reproduc-
tion, specifically by nesting females (Murphy

1996) , and during periods of low nectar availabil-
ity (Chavez-Ramirez and Dowd 1992). Time
spent on other activities, including nectar forag-
ing, is likely reduced when time spent foraging for
arthropods increases (Chavez-Ramirez and Dowd
1992). Stiles (1995) argued that hummingbirds

' Biology Department. George Fox University. 414 North
Meridian .Street. Newberg. OR 97132. USA.

^Department of Ecology and Evolutionary Biology.
University of Arizona. Tucson. AZ 85721, USA.

’School of Animal and Veterinary Sciences, University
of Adelaide, Roseworthy Campus, SA 5371, Australia.
'Corresponding author; e-mail: dpowcrs@gcorgefox.cdu

acquire a meaningful amount of energy from
consumption of arthropods, which would be
needed to meet energy demands when nectar
intake is reduced. This is not difficult to imagine
since arthropods have high energy content
(Weathers and Sullivan 1989) and appear to be
fully digested in a hummingbird’s digestive tract
(Remsen et al. 1986).

We know of no studies that examined the
possibility that hummingbirds might use arthro-
pods when excluded from a nectar source because
of local social interactions. For example, an
aggressive dominant species could restrict access
of subordinate species to clumps of flowers
contained within its territory forcing the subordi-
nate species to seek alternative energy sources
(Young 1971). This may reduce the energetic
impact of competition if the subordinate could use
arthropods as a primary energy source. This
attribute would provide ecological separation
between competitors so they can coexist (Rosen-
zweig 1985, Sandlin 2000b) in areas that do not
provide abundant nectar resources (Brown et al.
1978).

Eugenes fulgens (Magnificent Hummingbird) is
a member of a three-species hummingbird guild
seasonally inhabiting the Chiricahua Mountains of
southeastern Arizona. E. fulgens will readily feed
on nectar in artificial feeders, but use declines
when feeders are defended by dominant Blue-
throated Hummingbirds (Lampornis clemenciae)
(Pimm et al. 1985, Powers and Conley 1994,
Sandlin 2()00a. Powers et al. 2003). This behavior
is substantially different from that exhibited by

494

Powers el uL • AF^THROPOD FORAGING BY HUMMINGBIRDS

495

Black-chinned Hummingbirds {Archilochus alex-
andri), another subordinate species, which con-
sistently intrudes on L. clenienciae territories in
attempts to gain a nectar reward (Powers and
Conley 1994). There are few natural flowers
available in many portions of the Chiricahua
Mountains prior to onset of monsoon rains, and
arthropods may be the only nearby alternative
energy source for E. fulgens (Pimm et al. 1985).
The idea that E. fulgens has the ability to subsist
only on arthropods is not new. Marshall ( 1957:81 )
suggested that E. fulgens can inhabit pine-oak
(Piniis spp.-Quercus spp.) woodlands because it
“can dispense with moist habitat and flowers”
and switch to arthropod consumption.

We conducted this study to ascertain if
arthropods could be used as an alternative energy
source when nectar availability is restricted and if
the use of arthropods for energy is a feasible
strategy that allows E. fulgens to reduce compet-
itive interactions with territorial L. clemenciae.
Evidence from behavioral, morphological, and
physiological experiments are presented to dem-
onstrate the energetic importance of arthropod
foraging to hummingbirds.

METHODS

Study Animals. — Males of three species were
used in this study: E. fulgens (—7.5 g), L.
clemenciae (-8.0 g), and A. alexandri (—3.0 g).
All three species occur in southeastern Arizona
during summer and use different foraging strate-
gies (Pimm et al. 1985, Powers and Conley 1994,
Sandlin 2000a). L. clemenciae is a dominant
species, defending territories along borders of
riparian canyons. A. alexandri is non-territorial at
our study site and robs nectar from territories
defended by L. clemenciae . E. fulgens is also non-
territorial, but appears to forage as a trapliner
(Powers 1996), avoiding many competitive inter-
actions at dense flower aggregations.

Study Area. — This study was conducted at the
American Museum of Natural History’s South-
western Research Station in the Chiricahua
Mountains, Cochise County, Arizona (31° 50' N,
109° 15' W; 1,700 m asl). Habitat at the station is
largely riparian, bordered by oak woodland and a
mixed deciduous/coniferous forest. Pimm et al.
(1985) provide a more detailed description of the
habitat. This forest endures a pronounced dry
season from November until monsoon rains begin
in early July that can dramatically affect the
amount of nectar available (Li and Brown 1999).

Feeding Stations. — Twenty feeding stations in
areas with vegetation that provided many poten-
tial perching sites were established for behavioral
observations within the boundaries of the field
station. This increased the probability that arthro-
pod foraging could be observed in conjunction
with nectar foraging. Each individual feeding
station consisted of either a Perky-Pet Glass
Feeder (Model No. 203-CP, Woodstream Corp.,
Denver, CO, USA) with the perch and corolla
removed, or a triplet feeder constructed from three
smaller Perky-Pet plastic singlet feeders (Model
No. 214, Woodstream Corp., Denver, CO, USA)
connected with Velcro® strips. The feeders were
suspended from either a tree branch or a wire
hook attached to a vertical length of PVC pipe.
The height of the feeders (distance from the
ground to the bottom of the feeder) ranged from
0.74 to 2.10 m. A 0.52 M (18% weight/weight)
sucrose solution, a concentration typical of nectars
in many hummingbird flowers (Baker 1975), was
used at all the feeding stations.

Observations were made at four different time
intervals for 97 hrs throughout the day during
June and July 1998 and 1999. The timed intervals
were: early morning (0500-0730 hrs MST), late
morning (0800-1130 hrs), late afternoon (1500-
1800 hrs), and early evening (1815-1930 hrs).
The time of day each feeding station was visited
was randomized to ensure an unbiased examina-
tion of both nectar and arthropod foraging over all
times of day in all habitat types.

Arthropod Foraging. — The amount of time
birds spent foraging for arthropods near the
feeding stations was recorded. Arthropod foraging
was divided into two modes: perch foraging and
aerial foraging. Foliage gleaning or other known
modes of arthropod foraging (Stiles 1995) were
not detected. Perch foraging occurs when birds
forage for arthropods while sitting on a perch (no
night is involved). This mode of arthropod
foraging has been noted in a few hummingbird
species (Pitelka 1942, Brice 1992). Typical
behaviors include tongue extension, the bill
opening, “gulping”, or the head darting back
and forth. Arthropod foraging data were only
taken when it was clear that birds were attempting
to catch arthropods (often the vantage point made
it possible to see arthropod capture). Time spent
perch foraging was measured by starting a
stopwatch when a bird opened its bill, extended
the tongue, or when the bird gulped (indicating
the initiation of an arthropod foraging bout). The

496

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 3, September 2010

number of attempts the bird made for arthropods
was eounted with the stopwatch running and the
measurement was terminated when the bird flew,
vocalized, scratched, rubbed its bill, preened its
feathers, or lost active interest. The approximate
height of the perch where the event occurred was
recorded after recording the time and the number
of attempts.

The second major mode of arthropod foraging,
aerial flycatching, occurs during hovering or slow
forward flight (Stiles 1995). We measured the
duration of aerial flycatching bouts with a
stopwatch from the time the bird left its perch
until it returned to a perch or when the bout had
clearly ended. The number of foraging attempts
during a bout was estimated (we could not reliably
count actual captures) by counting the number of
head lunges (indicating a capture attempt). The
height of the perch from which the bout was
started was estimated visually.

Crop and Gizzard Analysis. — Crop and gizzard
data came from samples taken by dissection of 17
birds (// = 5 L. clemenciae , n = 6 E. fulgens, and
n = 6 A. alexandri) during June and July 1996.
Gut contents were preserved in 95% ethanol.
Whole or nearly whole arthropod specimens were
identified to the lowest possible taxon. Measure-
ments of the length of whole arthropods were
made to calculate an approximate size range for
arthropod prey. The crop and gizzard contents
were dried to a constant mass in a drying oven
(65° C) so we could compare the relative mass of
arthropod contents for each hummingbird species.

Measurements were standardized to compare
the amount of arthropod material in the gut
between species by dividing the mass of arthropod
material collected by bird metabolic body mass
(gO.?.-); exponent approximately describes the
relationship between body mass and metabolic
rate; Calder and King 1974).

Gut Allometry.—l\\Q gut mass and length of
each species were compared with 10 and eight
other hummingbird species, re.spectively (M.-V.
Lopez-Calleja, C. Martmez del Rio, and J. E.
Schbndube, unpubl. data). The intestines, from
proventriculus to cloaca, were removed from birds
killed by thoracic compression and immediately
placed in 1.02% saline. The inte.stine was flushed
with saline, blotted dry, and weighed to the
nearest 0.01 g after any fat and me.sentery were
removed. Total length was measured to the
nearest 0.1 mm with a digital caliper after the
intestine was gently straightened. The relationship

Paragona gigas

- Eugenes fulgens

Utmpomis clemenciae

Calypfe anna

I Selasphorus rufus

— Selasphorus plaryrercus

I Archilochus alexandri

— Archilochus colubris

Cynanthus larirvsrris

■ Hylocharis leucoiis

Amadlia ruiila

I Chlorostilhon canivetti

L Sephanoides sephanoides

Colibri thalossinus

0.2

FIG. 1. Presumed phylogeny of 14 hummingbird
species used in our phylogentic independent contrasts. A
0.2 branch-length scale is below the physlogenetic tree. The
phylogeny is based on McGuire et al. (2007) with
placement of C. latirostris based on Garcfa-Deras et
al. (2008).

between gut length, gut mass, and body mass was
examined using standard linear least-squares
regression as well as phylogenetically indepen-
dent contrasts (PIC; Felsenstein 1985, Garland et
al. 1992). We performed least-squares regressions
of log gut mass versus log body mass and log gut
length versus log body mass for standard analysis.
We calculated PIC for log body mass, log gut
mass, and log gut length (Garland et al. 1992).
The phylogeny and relative branch lengths used in
our PIC analysis were based on McGuire et al.
(2007) and Garcia-Deras et al. (2008) (Fig. 1).
The maximum branch length was arbitrarily set at
1.0 with shorter branch lengths scaled appropri-
ately. PIC analysis is robust to branch length
variation (Garland et al. 1999) and setting branch
lengths in this manner are unlikely to influence
PIC results. Absolute values of standardized
contrasts were correlated with their branch
lengths, and branch lengths were increased by a
factor of 10 and log transformed. This made
correlations non-significant and contrasts were

Powers el al. • ARTHROPOD FORAGING BY HUMMINGBIRDS

497

weighted equally in subsequent analyses. Stan-
dardized contrasts were positivized according to
Garland et al. { 1992). We perlormed least-.squares
regression through the origin on positivized
contrasts ot log gut mass versus log body mass
and log gut length versus log body mass.

Respiratory Quotient. — Respiratory quotient
(RQ) measures the proportion of carbohydrate
fuel an animal catabolizes during the measure-
ment process by comparing carbon dioxide
production with oxygen consumption (RQ =
VCO2W02). A measurement close to 1.0 represents
pure carbohydrate use, while for birds an RQ
close to 0.7 represents pure protein/fat use.
Comparing RQs between species provides a view
of the importance of nectar as a fuel at a given
point in time. A bird foraging more on arthropods
might be expected to have an RQ closer to 0.7.
We predicted RQs for wild-caught E. fulgens
would be lower than those for the other two
species.

Birds were captured in mist nets three times
each day (morning, midday, and late afternoon)
and randomly assigned to one of two groups: (1)
those caught from the wild and not fed, and (2)
those birds that were fed sucrose solution after
being caught but prior to RQ measurements. Fed
birds were allowed to drink their fill of 0.86 M
sucrose solution, and served as a control because
they were presumably catabolizing primarily
carbohydrate. The RQ of unfed individuals was
expected to reflect the “normal” metabolic
substrate for a bird at a certain time of day.

Measurements of O2 consumption (V02) and
CO2 production (VCO2) were made using an open-
circuit, positive-pressure respirometry system
(Powers 1991). Body mass was measured to the
nearest 0.1 g both before and after trials using a
portable balance (Model No. LS 200, Ohaus
Coip., Pine Brook, NJ, USA). The birds for each
trial were kept for about 0.5 hr in a metabolism
chamber within an environmental chamber at a
constant temperature (27° C) prior to making
measurements. Metabolism chambers consisted
of a large Mason jar (effective volume: 800 ml)
for E. fulgens and L. clemenciae or a small Mason
Jar (effective volume: 380 ml) for A. ale.xandri.
Metabolism chambers were placed in an environ-
mental control chamber (I-35L, Percival Scientif-
ic, Perry, lA, USA). Temperatures were recorded
to the nearest 0.1 °C using a Physitemp Bat- 12
(Physitemp Instruments, Clifton, NJ, USA) and a
Cu-Cn thermocouple. The flow rate of dry, CO2-

free air through the metabolism chamber was
500 ml/min. The inlet air passed through soda
lime and Drierite to remove ambient CO2 and
water vapor, respectively, prior to passing through
an O2 analyzer (Model No. S-3A, Applied
Electrochemistry, Naperville, IL, USA) and a
CO2 analyzer (Model No. LI-6262, LI-COR
Biosciences, Lincoln, NE, USA). Data acquisition
and analysis were with a Power Macintosh (Model
No. 7200, Apple Computer Corp., Cupertino, CA,
USA) using Warthog LabHelper and LabAnalyst
software (Mark Chappell, University of Califor-
nia, Riverside, CA, USA).

Statistical Analysis. — We used nonparametric
analysis of variance (Kuskal-Wallis test; Zar
1974) to evaluate differences among sample
means. Post hoc testing was by multiple compar-
isons using Mann-Whitney U-tests (Zar 1974)
when more than two groups were compared with
alpha values adjusted using the Bonferroni
coiTection. Curves illustrating allometric relation-
ships were plotted using simple linear regression
and the extent of variation explained by the
regression is reported as (Zar 1974). All
statistical calculations used SPSS 16.0.1 (SPSS
2007). Values reported are means ± SD.

RESULTS

Arthropod Eoraging. — Individual A. ale.xandri
were difficult to locate and observe foraging for
arthropods, possibly because the presence of one
or more aggressive L. clemenciae motivated these
birds to remain inconspicuous. E. fulgens and L.
clemenciae both exhibited a higher arthropod
foraging rate aerially than while perched (E.
fulgens: U\ ^\ = 21.5, P < 0.001; L. clemenciae:
^1.81 = 27.9, P < 0.001; Eig. 2). E. fulgens had a
higher aerial foraging rate than L. clemenciae
(^1.113 = 4.8, P = 0.03). L clemenciae perhaps
compensated for the lower aerial foraging rate by
spending more time in areal foraging than E.
fulgens (f/|.25 = 6.93, P = 0.01; T = 0.09 ±
0. 1 min/hr for E. fulgens, x = 0.52 ± 0.6 min/hr
for L. clemenciae). The majority of individual E.
fulgens engaged in aerial arthropod foraging (87%
of 93 observations) rather than perch foraging
(13% of 93 observations). Individual L. clem-
enciae, in contrast, divided arthropod foraging
activity more evenly with 45% engaged in perch
foraging and 55% foraging aerially (95 total
observations). These birds would typically drink
from a feeder and then perch in close proximity to
the feeder. L. clemenciae would often engage in

498

THE WILSON JOURNAL OL ORNITHOLOGY • Voi 122, No. 3, September 2010

Species

FIG. 2. Mean (± SD) number of insect capture
attempts per second for E. fulgens, and L. clemenciae
during a foraging bout. Numbers above the error bars are
.sample sizes. The indicates a significant difference
between the aerial and perch values for both E. fulgens and
L. clemenciae. The “t” indicates a significant difference
between aerial values for E. fulgens and L. clemenciae.

this behavior for a short time and then fly to
defend the nearest feeder or to feed from it.

There was a marked difference in the average
height from which the two species foraged for
arthropods with E. fulgens perching at a greater
height in both modes than L. clemenciae (aerial
foraging height f/|,8i= 32.2, P < 0.001; x = 9.9
± 4.7 m for E. fulgens, x = 3.6 ± 2.5 m for L.
clemenciae', perch foraging height; f/1,37 = 5.23,
P = 0.02; X = 5.9 ± 4.0 m for E. fulgens, x = 2.1

± 1.2 m for L. clemenciae). Both species engaged
in aerial foraging from a higher perch than when
perch foraging {E. fulgens'. (7i,59 = 5.39, P =
0.02; L. clemenciae'. Ui_s9 — 5.84, P = 0.02). E.
fulgens appeared to remain high in the tree canopy
(especially in Juglans spp. and Platanus spp.),
while L. clemenciae perched in lower tree limbs
or understory shrubs where it was closer to
feeders.

Crop and Gizzard Contents. — Guts of E.
fulgens contained the most arthropod taxa of our
three study species (Table 1). The number of taxa
represented (from 2 Classes and 4 Orders) is
conservative because only arthropods that could
be positively identified (to the lowest possible
taxon) are reported. This meant identifying nearly
whole specimens only. Absolute number of
individuals in each of the taxa could not be
calculated because most ingested arthropods were
fragmented. All three identifiable members of the
Order Hymenoptera were wasps. Several Homop-
teran specimens were the same type of leafhopper
(Family Cicadellidae). There were three insects
and one arachnid specimen that could not be
further identified.

Numbers and size range of whole arthropods as
well as the dry mass of all arthropod material
collected from the three hummingbird species
digestive tracts did not differ (dry mass: 6/2.16 =
2.096, P = 0.351; whole arthropods: //2.16 =
2.228, P = 0.328; Table 2).

Gut Allometry. — The linear regressions of gut
mass and body mass, and the phylogentically
independent contrasts (PIC) of gut mass and body

TABLE 1

Arthropod taxa found in crops and gizzards of hummingbird species in this study.

Species

A* B" C" D" E" F- G’ H" I'

E. fulgens

L. clemenciae
A. alexandri

X XXX XXX

XX X X

XX X

“ A = Cla.ss In.secta, Order Hymenoptera (length .1.4 - .1.8 mm); B = Cla.ss Insecta, Order Hymenoptera (length 2..1 mm); C = Class Insecta. Order Hymenoptera
(length 1.0 mm); D = Cla.ss Insecta. Order Homoptera, Family Cicadellidae; E = Cla.ss Insecta, Order Diptera; F = Class Insecta (unknown Order"); G = Class
Insecta (unknown Order"); H = Cla.ss In.secta (unknown Order"); I = Class Arachnida. Order Araneae.

Different from other listed categories.

TABLE 2. Arthropods {x ± SD) in crops and gizzards of hummingbirds in this study.

Species

n Whole arthropods (#/individual) Arthropod length (range, mm) Dry ma.ss (mg/g"’")

L. clemenciae
E. fulgens

A. alexandri

5 1.8 ± 3.0 2.47-2.55 0.9 ± 1.2

6 6.2 ± 7.3 1.02-5.75 1.6 ± 1.6

6 0.8 ±1.1 0.97-3.50 1 .6 ± 1 .0

Powers et cil. • ARTHROPOD FORAGING BY HUMMINGBIRDS

499

>' - 0.37UII - 1 .276 y = l)-UUO

C D

FIG. 3. The relationship between gut mass and length, and body mass in hummingbirds ranging in size from ~2 to 20 g.
A and B are the standard linear regression and the linear regression of independent contrasts for gut mass as a function of
body mass. C and D are the standard linear regression and the linear regression of independent contrasts for gut length as a
function of body mass. Triangles in A and C are hummingbird species in this study.

mass were both significant (Fig. 3A, B) indicating
body mass is a key factor affecting gut mass. The
linear regression and PIC of gut length and body
mass was also significant (Fig. 3C, D) indicating
gut length varied with body mass.

Respiratory Quotient. — Respiratory quotient
(RQ) within a species did not vary with time of
day and the data for each species were pooled
(Fig. 4). There was no difference between fed and
unfed RQ for A. alexandri (Cue = 2.91, P =
0.09). There was a difference between fed and
unfed RQ for both E. fulgens and L. clemenciae
(E. fulgens: f/1,19 = 13.7, P < 0.001; L.
clemenciae: (/| j9 = 12.39, P < 0.001). E. fulgens
also had a lower unfed RQ than A. alexandri or L.
clemenciae (7/2,39 — 6.88, P = 0.032). There was
little variation in unfed RQ in E. fulgens.

DISCUSSION

Nectar carbohydrates have been assumed to be
the primary energy source for hummingbirds, but
arthropods may also provide significant amounts
of energy in addition to protein and lipids. Most
insects from sweep net samples (Weathers and

E. fulgens L. clemenciae A. alexandri

Species

FIG. 4. Fed and unfed mean (± SD) respiratory
quotient (RQ) for A. alexandri . E. fulgens. and L.
clemenciae. The indicates a significant difference
between fed and unfed birds of the same species. The “t”
indicates significance between unfed RQ for E. fulgens and
the other species.

500

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 3. September 2010

Sullivan 1989) conducted in the Chiricahua
Mountains were from the Orders Coleoptera,
Homoptera, and Diptera. Weathers and Sullivan
(1989) showed these insects have an average
energy value of about 25 kJ/g (dry mass).
Karasov’s (1990) conservative estimate of the
metabolizable energy of insets is 19.3 kJ/g (dry
mass) compared to 16.4 kJ/g for nectar. Thus,
energetically, arthropods have the potential to be
more than a small supplement to a primarily
carbohydrate diet. Hainsworth (1977) showed that
if a hummingbird devotes equal time to nectar-
ivory and flycatching, even low efficiency rates of
~4()% can provide more energy than nectar
feeding.

E. fulgens seemed to emphasize aerial foraging
for arthropods. They spent the vast majority of
their arthropod foraging time in this mode, had the
highest frequency of capture attempts, and
generally foraged from higher perches. L clem-
encicie split their time more evenly between perch
and aerial foraging, but actually spent more total
time foraging for arthropods. It is possible that we
actually observed a relatively small portion of
their total arthropod-foraging activity because of
the traplining behavior of E. fulgens.

Sandlin (2000a) suggested that E. fulgens
switches to arthropod foraging to avoid the
con.sequences of competition. Successful territo-
rial species such as L. cletnenciae maintain
reliable nectar sources by defending them from
invaders thereby gaining an energetic advantage
(Powers et al. 2003). Thus, L. cletnenciae may
prefer to perch close to feeders, facilitating quick
responses to intruders. This would be an efficient
way to supplement their nectar diet for L.
cletnenciae foraging on arthropods while perched
in a defensive position. Powers and Conley ( 1994)
report that L. cletnenciae are perched about 70%
of the time, allowing them to both guard their
territory and to forage for passing insects when
available.

E. fulgens had the greatest diversity of ar-
thropod species in its crop and gizzard (Table 2).
The.se data corroborate the findings of Cottam and
Knappen (1939) who examined gut contents of E.
fulgens and L. cletnenciae from the nearby
Huachuca Mountains of Arizona. They found 13
different taxonomic specimens in the guts of E.
fulgens but only .seven in E. cletnenciae. The
reason for this difference is unclear. E. fulgens
forages for arthropods higher in the canopy than
E. cletnenciae and the greater arthropod diversity

in E. fulgens guts may reflect the diversity of
arthropods available in the upper canopy. It is also
possible the trapline-foraging behavior of E.
fulgens diversifies the arthropod component of
the diet (Colwell 1973, Feinsinger and Chaplin
1975). Our data show that E. fulgens and L.
cletnencae forage differently for arthropods but
there is no evidence for a difference in arthropods
consumed.

It is not possible to calculate the total
contribution of arthropods to the diet of the
hummingbirds in this study. It has been suggested
the types of arthropods we found in the guts are
digested quickly (Remsen et al. 1986, Stiles
1995). Thus, our measures of dry mass are
probably a good index of short-term arthropod
consumption and correspond reasonably well to
the maximum rate arthropods were captured
(range ~6 to 30 captures/hr assuming 100%
capture efficiency and total arthropod digestion in
1 hr). The total digestible energy of arthropods
consumed ranges from 0.2 to 0.4 kJ/day if
arthropod dry mass is turned over in the gut
hourly. Daily energy expenditure (DEE) is 82 kJ/
day for L. cletnenciae and 29 kJ/day for A.
alexandri (Powers and Conley 1994). Our data
suggest arthropods account for < 1 % of DEE
assuming DEE for E. fulgens is similar to L.
cletnencae. These data suggest there is no
difference in arthropod consumption or energy
derived from arthropods for our three humming-
bird species.

Our hummingbirds had gut masses and lengths
that corresponded with their body size suggesting
that none of the hummingbirds in this study has a
unique gut design. Any ability to subsist on
arthropods when nectar is scarce is likely a
common trait in hummingbirds.

All three hummingbird species had RQ values
<0.85 indicating they were not strictly cataboliz-
ing carbohydrate. The RQ of E. fulgens was
statistically lower than the other two species but
we are uncertain if the difference is truly enough
to argue a difference in use of arthropods for
energy. The low variability in E. fulgens RQ does
at least hint at a difference in how the three
species forage but the exact nature of this
difference is unclear. The higher variability in
the RQ of L. cletnenciae and A. ale.xandri could
correlate with time since their last nectar meal.
The lower variability in E. fulgens could support
more consistent arthropod consumption but it
could also be explained by the routine capture of

Powers el al. • ARTHROPOD FORAGING BY HUMMINGBIRDS

501

traplining individuals that have not yet fed. RQ
data alone are insufficient to draw conclusions
regarding the importance of different metabolic
substrates used by the hummingbirds in this study.
Hummingbirds can rapidly switch between fatty
acid and carbohydrate oxidation (Welch and
Suarez 2007) making interpretation of a depres-
sion in RQ difficult.

E. fiilgens differs behaviorally in arthropod
foraging from at least L. clemenciae but there is
no strong evidence that arthropod consumption
provides a disproportionate amount of energy
compared to the other hummingbird species in
this study. It is probable that all three species
energetically benefit from arthropods but quanti-
fying the total contribution of insects to their
energy budget is difficult. We still do not have a
complete understanding of the role arthropods
have in hummingbird nutrition but are hopeful
that future studies using techniques such as stable
isotope analysis might provide addition insight.

ACKNOWLEDGMENTS

We thank the American Museum of Natural History and
Southwestern Research Station for use of their research
facilities. We also thank the Murdock Charitable Trust and
Holman Endowment for the Sciences for providing funding
for this project. We thank Amber Hamilton and N. A.
Miller for assistance in collecting behavioral data in the
1999 field season. M. Victoria Lopez-Calleja, Carlos
Martinez del Rio, and J. E. Schondube generously shared
unpublished data on hummingbird gut morphology. We
thank C. A. Engman and Ryan Lapour for editing
assistance, and D. J. Kimberly and D. L. Swanson for
technical advice. We thank W. W. Weathers and an
anonymous reviewer for their constructive comments on
this manuscript.

LITERATURE CITED

Baker, H. G. 1975. Sugar concentrations in nectars from
hummingbird flowers. Biotropica 7:37-42.

Baker, H. G. 1977. Non-sugar chemical constituents of
nectar. Apidologie 8:349-356.

Brice, A. T. 1992. The essentiality of nectar and arthropods
in the diet of Anna’s Hummingbird (Ccilypte anna).
Comparative Biochemistry and Physiology 101 A: 151-
155.

Brown, J. H., W. A. Kodric-Brown, and A. Kodric-
Brown. 1978. Correlates and consequences of body
size in nectar-feeding birds. American Zoologist
18:678-700.

Calder III, W. A. AND J. R. King. 1974. Thermal and
caloric relations in birds. Pages 260-413 in Avian
biology (D. F. Earner and J. R. King, Editors).
Volume 4. Academic Press, New York, USA.

Chavhz-Ramirez, F. and M. Dowd. 1992. Arthropod
feeding by two Dominican hummingbird species.
Wilson Bulletin 104:74.3-747.

Cot.WELi., R. K. 1973. Competition and coexistence in a
simple tropical community. American Naturalist
107:737-760.

COTTAM, C. AND P. Knappen. 1939. Food of .some
uncommon North American birds. Auk 56:138-169.

Feinsinger, P. 1983. Coevolution and pollination. Pages
282-310 in Coevolution (D. J. Futuyma and M.
Slatkin, Editors). Sinauer Associates, Sunderland,
Massachusetts, USA.

Feinsinger, P, and S. B. Chaplin. 1975. On the
relationship between wing disc loading and foraging
strategy in hummingbirds. American Naturalist
109:217-224.

Felsenstein, j. 1985. Phylogenies and the comparative
method. American Naturalist 125:1-15.

Garci'a-Deras, G. M., N. Cortes, M. Money, A. G.
Navarro, J. Garci'a-Moreno, and B, E. Hernan-
dez-Banos. 2008. Phylogenetic relationships within
the genus Cynanthus (Aves: Trochilidae), with em-
phasis on C. doubledayi. Zootaxa 1742:61-68,

Garland Jr., T., P. H. Harvey, and A. R. Ives. 1992.
Procedures for the analysis of comparative data using
phylogenetically independent contrasts. Systematic
Biology 41:1 8-32.

Garland Jr., T., P. E. Midford, and A. R. Ives. 1999. An
introduction to phylogenetically based statistical
methods, with a new method for confidence
intervals on ancestral values. American Zoologist
39:374-388.

Gass, C. L. and R. D. Montgomerie. 1981. Hummingbird
foraging behavior: decision-making and energy regu-
lation. Pages 159-194 in Foraging behavior: ecolog-
ical, ethological and psychological approaches (A. C.
Kamil and T. D. Sargent, Editors). Garland STOP
Press, New York, USA.

Hainsworth, F. R. 1977. Foraging efficiency and parental
care in Colihri coruscans. Condor 79:69-75.

Karasov, W. H. 1990. Digestion in birds: chemical and
physiological determinants and ecological implica-
tions. Studies in Avian Biology 13:391-415.

Li, S.-H. and j. L. Brown. 1999. Influence of climate on
reproductive success in Mexican Jays. Auk 1 16:924-
935.

Marshall Jr., J. T. 1957. Birds of pine-oak woodland in
southern Arizona and adjacent Mexico. Pacific Coast
Avifauna 32:5-1 25.

McGuire, J. A., C. C. Witt, D. L. Altshuler, and J. V.
Remsen Jr. 2007. Phylogenetic systematics and
biogeography of hummingbirds: Bayesian and maxi-
mum likelihood analyses of partitioned data and
selection of an appropriate partitioning strategy.
Systematic Biology 56:837-856.

Murphy, M. E. 1996. Nutrition and metabolism. Pages 31-
60 in Avian energetics and nutritional ecology (C.
Cary, Editor). Chapman and Hall, New York, USA.

Pimm, S. L., M. L. Rosenzweig, and W. Mitchell, 1985.
Competition and food selection: field tests of a theory.
Ecology 66:798-807.

502

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3, September 2010

PiTELKA, F. A. 1942. Territoriality and related problems in
North American hummingbirds. Condor 44:189-204.

Powers, D. R. 1991. Diurnal variation in mass, metabolic
rate, and respiratory quotient in Anna’s and Costa’s
hummingbirds. Physiological Zoology 64:850-870.

Powers, D. R. 1996. Magnificent Hummingbird {Eugenes
fiilgens). The birds of North America. Number 221.

Powers, D. R. and T. M. Conley. 1994. Field metabolic
rate and food consumption of two sympatric hum-
mingbird species in southeastern Arizona. Condor
96:141-150.

Powers, D. R., A. R. Brown, and J. A. Van Hook. 2003.
Influence of normal daytime fat deposition on
laboratory measurements of torpor use in territorial
versus nonterritorial hummingbirds. Physiological and
Biochemical Zoology 76:389-397.

Remsen Jr., J. V., F. G. Stiles, and P. E. Scott. 1986.
Frequency of arthropods in stomachs of tropical
hummingbirds. Auk 103:436-439.

Rosenzweig, M. L. 1985. Some theoretical aspects of
habitat selection. Pages 517-540 in Habitat selection
in birds (M. L. Cody, Editor). Academic Press. New
York, USA.

Sandlin, E. A. 2000a. Cue use and resource subdivision

among three coexisting hummingbird species. Behav-
ioral Ecology 1 1 :550-559.

Sandlin, E. A. 2000b. Foraging information affects the
nature of competitive interactions. Oikos 91:18-28.

SPSS. 2007. SPSS 16.0 for Macintosh. SPSS Inc., Chicago,
Illinois, USA.

Stiles, F. G. 1995. Behavioral, ecological and morpholog-
ical correlates of foraging for arthropods, by the humming-
birds of a tropical wet forest. Condor 97:853-878.

Weathers, W. W. and K. A. Sullivan. 1989. Juvenile
foraging proficiency, parental effort, and avian reproduc-
tive success. Ecological Monographs 59:223-246.

Wolf, L. L. and F. R. Hainsworth. 1971. Time and
energy budgets of territorial hummingbirds. Ecology
52:979-988.

Welch, K. C. and R. K. Suarez. 2007. Oxidation rate
and turnover of ingested sugar in hovering Anna’s
(Calypte anna) and Rufous {Selasphonts ntfus) humming-
birds. Journal of Experimental Biology 210:2154—
2162.

Young, A. M. 1971. Foraging for insects by a tropical
hummingbird. Condor 73:36-45.

Zar, j. H. 1974. Biostatistical analysis. Prentice-Hall Inc.,
Englewood Cliffs, New Jersey, USA.

The IVilson Journal of Ornithology 1 22(3):5()3-5 1 2, 2010

ARMY ANT RAID ATTENDANCE AND BIVOUAC-CHECKING
BEHAVIOR BY NEOTROPICAL MONTANE FOREST BIRDS

SEAN O’DONNELL,''^ ANJALl KUMAR,' AND CORINA LOGAN-

ABSTRACT. We quantified resident and migrant bird attendance at army ant swarm raids {n = 48) in a neotropical
montane forest. All observations were during seasons when Nearctic migrant birds are present. Bird species differed in
army ant raid-attending behavior. Resident bird species attended 2 to 54% of raids, while migrants attended at lower
maximum frequencies (2 to 21% of raids affended per species). Some residenf and migrant bird species attended raids more
frequently than expected based on capture rates in mist-net studies and point-count density surveys. Army ant raid
attendance may be a regular element of foraging behavior for some resident species, and important in the wintering ecology
of some Nearctic migrant species. The bird species that attended raids most frequently were predicted to show behavioral
specializations for exploiting army ant swarms. Eight resident bird species (but no migrants) performed a specialized
behavior, bivouac checking, by which birds sample army ant activity. Resident bird species’ frequencies of raid attendance
were positively associated with frequency of checking bivouacs (/• = 0.68). We hypothesize the absence of obligate army
ant-following birds in montane forests has favored performance of specialized behaviors for exploiting army ant raids by
some resident birds. Received 3 October 2009. Accepted 9 March 2010.

Neotropical army ants (Formicidae: Ecitoninae)
are top predators, and a diverse array of animal
species associate with army ant colonies (Franks
1982, Franks and Bossert 1983, Brady 2003, Koh
et al. 2004). Birds attend army ant foraging-raids
to feed on arthropods and small vertebrates that
flee from the advancing ants. Birds primarily
attend the swarm raids of Eciton burchellii and
Labidus praedator (Willis and Oniki 1978, Wrege
et al. 2005). Bird flocks at army ant raids often
include multiple species, and their composition is
largely distinct from sympatric mixed foraging
flocks of insectivores (Willis 1972, Willis and
Oniki 1978, Otis et al. 1986, Willson 2004, Peters
et al. 2008).

Some bird species are obligate army ant raid
attendants that obtain most or all of their food at
army ant swarms (Willis and Oniki 1978, Swartz
2001, Willson 2004, Brumfield et al. 2007). Other
bird species attend raids opportunistically (Swartz
2001, Chaves-Campos 2003). Opportunistic army
ant raid-attending bird species vary in their
reliance on army ants (Willis 1972, Willis and
Oniki 1978). Obligate army ant-following birds
are agonistic toward other birds at raids in
lowland forests. This interference competition
reduces the value of ant raids as a food source

' Animal Behavior Program, Department of P.sychology,
Box 351525, University of Washington, Seattle, WA
98195, USA.

^University of Cambridge, Department of Experimental
Psychology, Cambridge CB2 3EB, United Kingdom.
^Corresponding author; e-mail: sodonnel@uw.edu

to Other birds (Willis 1966, Willis and Oniki 1978,
Brumfield et al. 2007).

Obligate army ant-following birds are poorly
represented or absent from montane forests
(Willis and Oniki 1978, Brumfield et al. 2007).
There are no obligate army ant-following birds at
our study site near Monteverde, Costa Rica
(Kumar and O’Donnell 2007). Birds from a
diverse array of families attend army ant raids in
the Monteverde area, including some resident and
Nearctic migrant species (henceforth migrants;
Vallely 2001, Kumar and O’Donnell 2007). We
hypothesized that some montane bird species
would exhibit behavioral specializations for
exploiting army ant raids in the absence of local
competition from obligate army ant-following
birds. We asked whether some montane bird
species attend raids more often than expected as a
first test of this hypothesis. The frequency of raid
attendance varies among Monteverde area birds,
but attendance frequency alone does not account
for possible effects of local abundance (Vallely
2001, Kumar and O’Donnell 2007, Peters et al.
2008). We extended our previous analyses
estimating the effects of species’ relative abun-
dance on army ant raid attendance in this study.
Bird abundance estimates were derived from
previously published mist-net captures and
point-count den.sities (Young et al. 1998, Jan-
kowski et al. 2009).

We predicted the most frequent raid attendant
birds would be more likely to exhibit specialized
behaviors for exploiting army ant swarms (Willis
1972, Willis and Oniki 1978, Swartz 2001,
Chaves-Campos 2003). We ascertained whether

503

504

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 3. September 2010

montane forest birds perform bivouac-checking
behavior to test this prediction. Bivouac-checking
birds visit Eciton burchellii temporary nests
(bivouacs) as the ant foraging raids start in the
morning, and again in the evening. Bivouac-
checking birds may assess army ant raid activity
and direction in the morning, or whether ants are
emigrating to a new bivouac site in the evening
(Swartz 2001, Chaves-Campos 2003). We corre-
lated bird species’ frequencies of bivouac check-
ing with their frequencies of attendance at army
ant raids to examine whether raid attendance
frequency and bivouac-checking behavior were
associated among bird species.

METHODS

Study Site. — Our sample sites spanned the
continental divide in the Tilaran Mountain Range
near Monteverde, Puntarenas Province, Costa
Rica (10° 18' N, 84° 47' W). The Monteverde
area includes forest reserves with adjacent pri-
vately held continuous forest and associated forest
fragments (Guindon 1997, Haber 2000, Harvey
2000, Jankowski et al. 2009). We collected data in
fore.sts between 1,175 m above mean sea level (m
asl) and 1,580 m asl elevation on the Pacific
Slope, and between 985 and 1,680 m asl elevation
on the Atlantic Slope. Our raid attendance
samples included four life zones (Holdridge life
zone system; Holdridge 1966, Guindon 1997,
Young et al. 1998, Young and MacDonald 2000).

Sampling Bird Flock Composition. — We col-
lected data during seasons when migrants occur in
the Monteverde area (Stiles and Skutch 1989,
Garrigues and Dean 2007). The four field trips
collectively spanned much of the period of
migrant presence (mid-Sep to late May): 1 1
January-4 March 2005 (dry season), 5 October-
I 1 December 2005 (late wet season/onset of dry
season), 19 December 2007-1 January 2008
(early dry season), and 7-10 April 2008 (late
dry season). We ob.served birds at 48 army ant
raids during these trips. The army ant raids
involved three swarm-raiding army ant species
(Eciton hurchellii, n = 40; Lahidus praedator, n
= 5; and L. spinninodis, n = 3). We did not
separate army ant species in our analyses as the
Lahidus species raids were distributed across all
life zones, and army ant species had no measur-
able effect on attending bird flock composition in
the Monteverde area (Kumar and O’Donnell
2007). We observed a maximum of one swarm
per day and alternated sampling dates among

elevations to minimize order effects. We treated
each army ant swarm as an independent data point
because observations were separated in space and
time. Observations were made during the diurnal
active raid period of surface-foraging army ants.
Start times ranged from 0900 to 1600 hrs local
time (CST).

We located flocks of army ant swarm-attending
birds or foraging army ant columns by searching
along established trails. We walked to swarm raid
fronts where birds fed and positioned ourselves at
the best location for unobstructed viewing of the
swarm front; this was usually off to one side and
facing in the direction the swarm was moving. If
the raid or the bulk of bird activity shifted location
during observations we walked to a new observa-
tion spot while attempting to avoid disturbing the
birds. We noted all bird species and the number of
individuals that were present upon our amval
(Coates-Estrada and Estrada 1989). The observa-
tion sessions lasted for 1 hr except when raid
activity ceased, heavy rainfall began, or ants
traversed impassable terrain (.y ± SD observation
time = 47 ± 18 min). We analyzed only those
flocks for which all attending birds were identi-
fied to species.

We collected four types of data on bird flock
composition and bird behavior at the army ant
swarms. (1) Start and end time to the nearest min.
(2) Latitude/longitude coordinates and elevation
to the nearest 10 m asl were taken with a hand-
held Geographic Positioning System (GPS) unit or
an air pressure altimeter. Elevations were con-
firmed from topographic maps. (3) We identified
all birds seen in attendance to species based on
plumage appearance, behavior, relative body size,
geographic range, and vocalizations (Stiles and
Skutch 1989, Garrigues and Dean 2007). A bird
had to be observed collecting prey that was
fleeing from ants to be counted as an attendant.
Birds were categorized as residents or migrants.
(4) We recorded the number of individuals of each
bird species present. The birds were not banded
and we could not distinguish individuals that left
the raid and returned from newly arriving swarm
attendants. Thus, we used the maximum number
of individuals that were observed simultaneously
as a conservative estimate of the number of
attending birds from each species (Coates-Estrada
and Estrada 1989).

Bivouac-checking Observations. — We conduct-
ed 15 systematic watches for bivouac-checking
behavior at E. burchellii bivouacs following

O’Donnell Cl al. • MONTANE BIRDS AT ARMY ANT SWARMS

505

Swartz (2001 ). We sal on (he ground or on folding
chairs at sites 5 to 10 ni from the bivouac, facing
the entrance of the bivouac shelter. We chose
locations that were partially concealed by vege-
tation but had an unobstructed view of the bivouac
entrance. Our observation durations ranged from
60 to 120 min (.v = 87.4 min/session; 21.9 hrs of
observation). Like Swartz (2001), we conducted
12 observation sessions in the morning (start
times 0440 to 0800 hrs) and extended the protocol
by conducting three bivouac watches in the
afternoon (start times 1533 to 1630 hrs). We
recorded bivouac-checking behavior and counted
a bird as checking the bivouac if it first flew
toward the bivouac site and perched within 5 m of
the bivouac. The bird then had to either peer at the
bivouac shelter entrance, or land among the ants
when ants were active outside the bivouac to be
counted. We noted bird arrival time to the nearest
minute and, whenever possible, recorded the
duration of each bird’s visit. We also recorded
the species identity of birds that were observed
performing bivouac-checking behavior while we
were collecting other data (n = 5 observations;
Swartz 2001). We used the same behavioral
criteria for bivouac checking as during systematic
observations. Dates of bivouac-checking observa-
tions were: 15 November to 3 December 2005, 22
to 29 December 2007, 7 to 1 0 April 2008, and 5 to
25 July 2009. We made six systematic and two
opportunistic bivouac-checking observations dur-
ing seasons when Nearctic migrants are present at
the field site.

We could not identify birds as individuals as
birds were not banded and many species were
sexually monomorphic in plumage. We used only
one visit/species/observation session to estimate
frequencies of bivouac checking. All but one of
our bivouac-checking observations were in the
premontane wet forest life zone. We sampled
bivouac checking at elevations from 1,100 to
1,430 m asl; one systematic ob.servation was
conducted at a higher elevation (1,575 m asl) in
the lower montane wet forest life zone, but no
birds were recorded bivouac checking at this site.
We also noted if birds vocalized while bivouac
checking in 2008 and 2009. We used observations
of bird attendance at a larger sample of army ant
raids {n = 54) for comparisons of raid attendance
with bivouac checking. These data were collected
during several field trips between 25 January 2005
and 10 April 2008. These raids partially overlap
with the migrant season raid sample, but all were

observed in the same life zone and over a similar
range of elevations as our bivouac-checking
observations (1,100 to 1,475 m asl).

Statistical Analyses. — We calculated two mea-
sures of bird species attendance at swarm raids
based on species richness and individual abun-
dance at the 48 migrant season raids (Coates-
Estrada and Estrada 1989): (1) the percent of raids
at which each species was present, and (2) the
percent of all raid-attending birds accounted for
by each species. We used two independently
collected data sets to estimate forest understory
activity and relative abundance of bird species.
We compared these with our abundance measure
of army ant raid attendance (i.e., each species’
percent of all birds at raids) to identify species
that were present at raids more often than
expected by chance. The estimate of relative
activity came from mist-net data from Monte-
verde area forests (Young et al. 1998). We
calculated the number of mist-net captures for
each attending bird species from Young et al.
(1998: table 2), summing total captures for each
species over the four life zones that overlapped
our sample area. Mist-net studies of bird abun-
dance must be interpreted with caution (Remsen
and Good 1996). For example, bird species may
differ in probability of being caught in nets in the
understory. We used an independently-derived
estimate of relative abundance from standardized
auditory and visual point-count data to comple-
ment the mist-net data. The point-count data also
came from sites largely overlapping with our
observations (Jankowski et al. 2009: supplemen-
tary table 1).

We followed Peters et al. (2008) to calculate
two indices of raid attendance for each bird
species. We regressed each species’ proportion of
the total birds observed at raids against its total
captures in the mist-net studies (Young et al.
1998), and against its total/ha abundance estimat-
ed from point counts (Jankowski et al. 2009). We
calculated residuals from these two linear regres-
sions and used the residuals as indices of raid
attendance. We identified species as high-fre-
quency army ant raid attendants when their index
value (regression residual) was higher than the
upper 95% confidence interval (Cl) value from
the con-esponding linear regression model. We
calculated the mist-net index separately for
residents and migrants because resident birds
were mist-netted year-round, including months
when migrants are absent. We could not calculate

506

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 3. September 2010

a point-count index for migrants as they were not
recorded in the point-count study, in part because
migrants rarely vocalize outside their breeding
ranges (Jankowski et al. 2009).

We asked whether bird species differences in
raid attendance were correlated with bivouac-
checking frequency. We calculated the Pearson
correlation between an index of raid attendance
frequency (i.e., the point-count attendance index)
and bivouac-checking frequency for the 31
resident bird species observed at army ant raids
in the bivouac-checking sample elevation range.

RESULTS

Bird Flock Composition at Ant Raids. — We
observed 54 species of birds attending raids
during the period of migrant presence (Table 1).
The number of bird species in flocks at army ant
raids ranged from one to 17 (x ± SD = 4.73 ±
3.27), while the number of migrant species ranged
from zero to six (0.79 ± 1.29). We recorded 11
species of migrants attending raids; two were
thrushes (Turdidae), one was a vireo (Vireonidae),
and eight were wood warblers (Parulidae) (Ta-
ble 1). We recorded 13 bird species at raids that
had not been noted as army ant raid attendants in
the Monteverde area previously ( 1 1 new residents
and 2 new migrants). Migrants participated in 21
(43.8%) of the flocks at army ant swarms. None
of the swarm-attending flocks was comprised only
of migrants.

Resident Species Differences in Raid Atten-
dance.— Resident bird species varied widely in
frequency of raid attendance. Resident species
ranged from 2.1 to 54.2% of army ant raids
attended, and from 0.2 to 10.8% of the individual
birds at army ant raids (Table 1 ). Species’ point-
count densities and mist-net captures were
positively but weakly correlated (r = 0.42, n =
54, P < 0.01). These measures provide different
information on bird species’ baseline rates of
occurrence in the habitat. Both measures of
baseline density were non-significantly related to
army ant raid attendance by the 43 resident bird
species. Point-count estimates of bird species’
abundances were poor predictors of species
occurrence at army ant raids {R^ = 0.004, df =
53^ P = 0.64). Similarly, capture frequency in
mist-net studies was weakly associated with
attendance at army ant raids {R^ = 0.06, df =
42, P = 0.10). Both indices of frequency of raid
attendance indicated strong species differences in
frequency of army ant raid attendance after

accounting for density effects. There was high
consistency between the two indices of which bird
species attended army ant raids at higher than
expected frequencies. Seven resident bird species’
frequencies at raids exceeded the 95% Cl value
predicted by linear regression of raid attendance
on mist-net capture frequency (Fig. 1). Six of
these seven species (and one other) were signif-
icantly more frequent at raids than expected based
on point-count densities (Fig. 1).

Migrant Species Differences in Raid Atten-
dance.— Relative abundances of migrants at raids
spanned an order of magnitude among species in
terms of percent of raids attended (2.1 to 20.8%)
and in percent of individual birds seen at raids
(0.19 to 1.87%; Table 1). The Kentucky Warbler
{Oporonis formosus) was the most frequent
migrant at army ant raids by both measures.
Mist-net capture frequency of migrant species was
a poor predictor of their proportion of raid-
attending birds {R~ = 0.07, n = 11, P = 0.43).
Examination of residuals from this analysis
suggests three migrant bird species attended raids
more frequently than expected by mist-net
captures (Fig. 2).

Bird Species that Check Bivouacs. — We ob-
served eight species of birds from eight families
checking bivouacs in the Monteverde area (Ta-
ble 1). Birds checked bivouacs during 10 of the 15
systematic watches (67%); birds were seen
bivouac checking during 67% of both morning
and afternoon watches. The number of species
seen during an observation session ranged from
zero to four {x — 1.47 species). We did not see
more than one individual of a given species airive
at the same time. Birds of three species were
present at a bivouac simultaneously on one
occasion, but were not seen interacting. Six
species checked bivouacs both in the morning
and afternoon (Table 1); the other two species
checked in the morning but were .seen only once.

Bird Behavior During Bivouac Checks. — We
obtained duration data for 17 bivouac checks by
six bird species (the bird was not identified on one
timed visit). Bivouac-check durations pooled
across species varied from a few seconds to
several minutes (range = 5 to 360 sec; -.v ± SD =
103 ± 126 sec). Visit lengths did not differ
significantly among species (Kruskal-Wallis test,
X- = 7.6, df = 5, P = 0.18).

Three species. Orange-billed Nightingale-
Thrush {Catharus aurantiirostris), Rufous-and-
white Wren (Thryothorus rufalhus), and Rufous-

O'Donnell ct al. • MONTANE BIRDS AT ARMY ANT SWARMS

507

TABLE 1 . Frequencies of occurrence of birds as foragers at army ant swarm raids in montane forests near Monteverde,
Costa Rica. Nearctic migrants are indicated by an asterisk *. Species not seen in previous studies of avian army ant raid
attendance in the Monteverde area are indicated by a plus symbol +.

Number of raids attended

% of raids

Number of birds

% of

Species

(H = 48)

attended

{n = 316)

birds

Orange-billed Nightingale-Thrush

Cathams aurantHrostris

26

54.2

58

10.8

White-eared Ground Sparrow

Melozone leucotis

18

37.5

29

5.4

Blue-crowned Motmot

Momolus momota

15

31.3

21

3.9

Brown Jay

Cyanoconix morio

14

29.2

37

6.9

Slaty-backed Nightingale-Thrush

Cathams fuscater

10

20.8

21

3.9

Rufous-and-white Wren

Thryothoms mfalbus

10

20.8

13

2.4

Rufous-capped Warbler

Basileiitenis rufifrons

10

20.8

11

2.1

Kentucky Warbler
*Oporornis fonnosus

10

20.8

10

1.9

Immaculate Antbird

Myrmeciza invnaculata

9

18.8

22

4.1

Wood Thrush

*Hylocichla mustelina

7

14.6

7

1.3

Golden-crowned Warbler

Basileutems culicivorus

6

12.5

8

1.5

Slate-throated Whitestart

Myiobonis miniatus

6

12.5

7

1.3

Yellowish Flycatcher

Empidonax flavescens

6

12.5

6

1.1

Three-striped Warbler

Basileutems tristhatus

5

10.4

12

2.2

Ruddy Woodcreeper

Dendrocincla homochroa

5

10.4

7

1.3

Wilson’s Warbler
*Wilsonia piisilla

5

10.4

7

1.3

Swainson’s Thrush
*Cathams ustulatus

5

10.4

5

0.9

Blue-throated Toucanet

Aulacorhynchus prasinus

4

8.3

6

1.1

Chestnut-capped Brush Finch

Arremon bmnneinucha

4

8.3

4

0.8

Ruddy-capped Nightingale-Thrush

Cathams frantzii

4

8.3

4

0.8

White-throated Thrush

Turdiis assimilis

3

6.3

8

1.5

Azure-hooded Jay

Cyanolyca cucuUata

3

6.3

5

0.9

10 species including one migrant: Black-and-white
Warbler *Mniotilta varia

2

4.2

2 or 3

0.4 or 0.6

22 species including 6 migrants*; Chestnut-sided
Warbler. +Dendroica pensylvanica'. Black-
throated Green Warbler, D. virens, Ovenbird,
+Seiums aiirocapilla', Golden-winged Warbler,

+ Vennivora chtysoptera; Canada Warbler
Wilsonia canadensis; Philadelphia Vireo, Vireo
philadelphicus

1

2.1

1 or 2

0.2 or 0.4

508

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. 3. September 2010

Total mist-net captures

0 100 200 300 400

FIG. I. Frequency of army ant raid attendance plotted against estimated local abundance for montane resident bird
species (Monteverde, Costa Rica). Solid line: linear regression best fit. Curved dashed lines: 95% confidence intervals (Cl).
Triangle symbols indicate species with raid attendance above the 95% Cl (0-b NT = Orange-billed Nightingale-Thrush; B
J = Brown Jay; W-e GS = White-eared Ground Sparrow; lA = Immaculate Antbird; B-c M = Blue-crowned Motmot; S-b
NT = Slaty-backed Nightingale-Thrush; R-c W = Rufous-crowned Warbler; R-w W = Rufous-and-white Wren).

capped Warbler (Basileitteru.s rufifrons), either
sang or called while bivouac checking. Ruddy
Woodcreepers (Dendrocincla homochroa) and/or
Orange-billed Nightingale-Thrushes called or
sang in the vicinity of bivouac sites during eight
observation sessions, even when birds were not
ob.served checking the bivouac. Individuals of
these species apparently circled the bivouac site at

distanees of lO to 1 5 m, calling or singing from
different perches.

Raid Attendance and Bivouac Checking. — Biv-
ouac checking was associated with high frequen-
cies of army ant raid attendance. We observed 42
species of birds attending 54 army ant raids in the
elevation range where we sampled bivouac
checking. All bird species that checked bivouacs

O’Donnell Cl al. • MONTANE BIRDS AT ARMY ANT SWARMS

509

Total mist-net captures - Migrants

FIG. 2, Frequency of army ant raid attendance plotted
against estimated local abundance for migrant bird species
(Monteverde, Costa Rica). Solid line: linear regression best
fit. Curved dashed lines: 95% confidence intervals (Cl).
Triangle symbols indicate species with raid attendance
above the 95% Cl (KW = Kentucky Warbler; WW =
Wilson’s Warbler; WT = Wood Thrush).

were also seen attending ant raids. The eight
species observed checking bivouacs were among
the most frequent raid attendants (Table 2).
Bivouac-checking birds represented 19.0% of
the bird species at these raids, but 57.7% of
individual birds at the raids. Furthermore, the
percent of bivouacs checked by a species
correlated positively (r = 0.68, /? = 31 , P <
0.01) with the point-count raid attendance index

lor the 31 resident bird species that attended raids
in the same elevation range (Table 2).

Six of our systematic watches and two
opportunistic observations were in months when
Nearctic migrant birds are wintering in the
Monteverde area (Nov, Dec, and Apr.). No
migrant birds were seen checking bivouacs
(Table 2).

DISCUSSION

Resident Bird Species Differences in Redd
Attendance. — Frequency of mist-net capture and
point-count densities did not predict variation in
raid attendance among bird species observed. Bird
species density was not an important determinant
of frequency of army ant raid attendance, and bird
species must vary in their frequency of raid
attendance for other reasons. The two indepen-
dently-derived indices of raid attendance for
residents identified similar lists of bird species
that were at raids more frequently than expected
(Peters et al. 2008). These species may depend
more heavily on army ants for food, and they may
exhibit specialized behaviors for exploiting army
ant swarms. These behaviors could include
mechanisms for locating and orienting to the ants
or to other attending birds, or abilities to track
army ant colony activity and movements such as
bivouac checking (Willis and Oniki 1978, Swartz
2001, Chaves-Campos 2003).

TABLE 2. Bird species observed attending army ant raids and checking
Costa Rica.

bivouacs in montane forests near Monteverde,

Common name

ScientiHc name (Family)

Percent of
bivouac check
ob.servaiions
in = 20)

Afternoon

bivouac-

checking

observed?

Percent of
raids
attended
(n = 54)

Percent of
birds
al raids
(n = -S27)

Raid

attendance
index rank
(n = .11 species')

Orange-billed Nightingale-
Thrush

Cathanis aurantiirostris
(Turdidae)

50

Yes

59.3

20.2

1

Ruddy Woodcreeper

Dendrocinchi homochroa
(Furnariidae)

30

Yes

5.6

0.9

14

Rufous-capped Warbler

Basilenterus rufifrons
(Parulidae)

15

Yes

35.2

7.6

5

Rufous-and-white Wren

Thryothorus rufalhiis
(Troglodytidae)

15

Yes

24.1

5.5

6

Blue-crowned Motmot

Momotiis moinota
(Momotidae)

10

Yes

38.9

8.9

4

Blue-throated Toucanet

Aukicorhynchus prasums
(Ramphastidae)

10

Yes

9.3

2.4

8

White-eared Ground Sparrow

Melozone leiicoti.s
(Emberizidae)

5

No

48.1

1 1.9

2

Chirqui Quail-Dove

Geotrygon ch i riquen.'iis
(Columbidae)

5

No

1.9

0.3

26

510

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 3. September 2010

Ecological Implications of Bivouac-checking
Behavior. — Only obligate army ant-following
birds checked bivouacs in lowland neotropical
forests, even at sites with opportunistic species
that frequently attended raids (Swartz 2001,
Chaves-Campos 2003). Swartz (2001) suggested
bivouac-checking behavior can be used to distin-
guish obligate army ant-following bird species
from opportunistic raid attendants. Performance
of this specialized behavior by Monteverde
resident birds indicates army ant swarms are an
important food resource to some montane bird
species. The relative importance of foraging at ant
raids to bird diets has not been quantified, but
species differences in raid attendance combined
with bivouac-checking behavior indicate some
specialization for foraging at army ant swarms.
Two bird species were seen checking bivouacs
only once, indicating that more montane birds
would be seen performing this behavior with
additional observations. Whether resident and
migrant montane birds exhibit other specialized
behaviors to enhance attendance at ant raids
requires further study. Some resident attendants
vocalized while checking bivouacs, including
nightingale-thrushes (Catharus spp.). Calling and
singing at bivouac sites by some residents may
recruit other birds to attend raids (Chaves-Campos
2003).

Bivouac-checking birds devote time and energy
to sampling army ant behavior without an
immediate food reward, and possibly expose
themselves to predators. Bivouac-checking bouts
were generally short in duration, consistent with
this behavior incurring some cost (Swartz 2001).
Some bird species check bivouacs in the morning
(at the start of raiding) and in late afternoon (when
most colony emigrations begin; O’Donnell et al.
2009), suggesting that birds track movements of
ant colonies as well as their daily raiding activity.
Individually marking and tracking birds will be
necessary to identify how many bivouacs the
montane resident birds check on a given day, and
how far they travel among bivouacs (Willson
2004, Chaves 2008).

We hypothesize that reduced (or absent)
competition with obligate army ant-following
birds in some montane forests removes biotic
constraints on specialized raid-attending behavior
for other species. Reasons tor obligate army ant-
following bird ab.sence in the Monteverde area
Pacific Slope are not known, but lack of local
competition with obligate army ant-tollowing

birds may enhance the value of army ant swarms
to opportunistic raid-attending birds. It remains to
be tested whether these behaviors represent local
evolutionary adaptations or plastic expression of
behavior based on learning or cultural transmis-
sion. We predict that observations at other sites
without obligate army ant-following birds will
reveal specialized army ant raid-attending behav-
iors in additional bird species.

Ecological Implications of Migrant Raid Atten-
dance.— Three species of migrants attended army
ant raids more often than expected according to
the mist-net capture-based index. The Kentucky
Warbler and two thmshes (Wood Thrush [Hylo-
cichla nmstelina] and Swainson’s Thrush, [Ca-
tharus utitulatus]) have been recorded as army
ant-raid attendants from Mexico to Colombia
(Willis 1966, 1984; Greene et al. 1984; Coates-
Estrada and Estrada 1989; Roberts et al. 2000;
Meisel 2004), as have several other migrants we
observed attending army ant swarms (Willis 1966,
Hardy 1974, Greene et al. 1984, Roberts et al.
2000, Meisel 2004, Rios et al. 2008). Army ant
raid attendance is apparently consistent across the
wintering ranges of some migrant birds. Swarm-
raiding army ants do not occur in the breeding
ranges of the migrants we recorded as raid
attendants (Watkins 1985). Attendance at army
ant raids by wintering migrants is a striking
example of plasticity in foraging behavior, and of
differences in foraging ecology between breeding
and wintering habitats. Changes in foraging
ecology are an important component of niche-
switching between breeding and wintering sites
(Nakazawa et al. 2004). Studies at South Amer-
ican sites are needed to learn if similar changes in
foraging ecology, including army ant raid atten-
dance, are exhibited by Austral migrant birds
(Jahn et al. 2004).

Migrants were not at raids without residents.
Residents may initiate raid-attending flocks that
are later joined by migrants, possibly in part
because residents are locally more abundant
(Young et al. 1998, Jankowski et al. 2009). Some
resident birds may also be more effective at
locating or tracking army ant raids, in part due to
specialized behaviors like bivouac checking.
Migrants did not check bivouacs but could use
other behavioral mechanisms for exploiting ant
raids. Data on young migrant bird interactions
with army ants during their first migration into the
geographic range of these ants could reveal
important developmental patterns, including roles

O Donnell et al. • MONTANE BIRDS AT ARMY ANT SWARMS

511

ot learning and social transmission in the
performance of raid-attending behavior.

All of the migrants we observed at raids are
known to forage on other resources while
wintering (Willis 1966, 1984; Di Giacomo and
Di Giacomo 2006; SO’D and AK, pers. obs.).
However, even occasional reliance on army ant
swarm-raids as a food source may have implica-
tions for migrant bird conservation. The extent of
reliance of migrants on army ants as a food source
may affect fitness of birds and winter survival by
migrants based on changes in army ant density.
Army ant-foraging behavior and population dy-
namics are strongly affected by forest clearing and
forest fragmentation (Franks 1982, Meisel 2006,
Kumar and O’Donnell 2009). Habitat change
effects on wintering migrants could be exacerbat-
ed through changes in army ant communities
(Koh et al. 2004).

ACKNOWLEDGMENTS

Thomas Scare and Sean Tully assisted with field
observations. We thank Frank Joyce and Katy Van Dusen
for continuing logistical support, and the many residents of
the Monteverde area who informed us of army ant swarm
activity and allowed access to their private lands for
research. The Tropical Science Center, Ecolodge San Luis/
University of Georgia, Monteverde Conservation League,
and the Monteverde Institute extended support and access
to protected lands for data collection. Research was
conducted under permits from the Ministry of the
Environment and Energy, Republic of Costa Rica (Scien-
tific Passports 0387 and 01667), and in accordance with the
laws of the Republic of Costa Rica. We thank the
Organization for Tropical Studies (OTS) for ongoing
research support, including assistance in obtaining permits.
This project was supported by a University of Washington
ALCOR Fellowship and an OTS Post-Course Award to AK,
and University of Washington Royalty Research Fund and
NSF grants (IBN-0347315) to SO’D. National Geographic
Television also supported the field research. Research was
partially supported by the National Science Foundation
while SO'D was working at the foundation. Any opinions,
findings, and conclusions or recommendations are those of
the authors and do not necessarily reflect the views of the
National Science Foundation. Two anonymous reviewers
made helpful comments that improved the paper.

LITERATURE CITED

Brady, S. G. 2003. Evolution of the army ant syndrome:
the origin and long-term evolutionary stasis of a
complex of behavioral and reproductive adaptations.
Proceedings of the National Academy of Sciences of
the USA 100:6575-6579.

Brumreld. R. T., J. G. Tello, Z. A. Cheviron. M. D.
Carling, N. Crochet, and K. V. Rosenberg. 2007.
Phylogenetic conservatism and antiquity of a tropical

specialization: army-anl-following in the typical ant-
birds (Thamnophilidac). Molecular Phylogeny and
Evolution 45:1-13.

Chaves, J. 2008. Benefits of cooperative food search in the
maintenance of group living in Ocellated Antbirds.
Dissertation. Purdue University, West Lafayette,
Indiana. USA.

Chaves-Campos, j. 2003. Localization of army-ant swarms
by ant-following birds on the Caribbean slope of Costa
Rica: following the vocalization of ant birds to find the
swarms. Ornitologia Neotropical 14:289-294.

Coates-Estrada, R. and A. Estrada. 1989. Avian
attendance and foraging at army-ant swarms in the
tropical rainforest of Los Tuxtias, Veracruz, Mexico.
Journal of Tropical Ecology 5:281-292.

Di Giacomo, A. S. and A. G. Di Giacomo. 2006.
Observations of Strange-tailed Tyrants (Alectrurus
risora) and other grassland birds following army ants
and armadillos. Journal of Eield Ornithology 77:266-
268.

Eranks, N. R. 1982. Ecology and population regulation in
the army ant Eciton burchelli. Pages 389-395 in The
ecology of a tropical forest: seasonal rhythms and long
term changes (E. G. Leigh Jr., A. S. Rand, and D. M.
Windsor, Editors). Smithsonian Institution Press,
Washington, D.C., USA.

Franks, N. R. and W. H. Bossert. 1983. The influence of
swarm raiding ants on the patchiness and diversity of a
tropical leaf litter ant community. Pages 151-163 in
Tropical rain forest: ecology and management (S. L.
Sutton, T. C. Whitmore, and C. A. Chadwick, Editors).
Blackwell Scientific Publications, Oxford, United
Kingdom.

Garrigues, R. and R. Dean. 2007. The birds of Costa
Rica. Cornell University Press, Ithaca, New York.
USA.

Greene, E., D. Wilcove, and M. McFarland. 1984.
Observations of birds at an army ant swarm in
Guerrero, Mexico. Condor 86:192-193.

Guindon, C. F. 1997. The importance of forest fragments
to the maintenance of regional biodiversity surround-
ing a tropical montane reserve, Costa Rica. Disserta-
tion. Yale University, New Haven, Connecticut, USA.

Haber, W. A. 2000. Plants and vegetation. Pages 39-94 in
Monteverde: ecology and conservation of a tropical
cloud forest (N. Nadkarni and N. Wheelwright,
Editors). Oxford University Press. New York, USA.

Hardy, J. W. 1974. Jays as army ant followers. Condor
76:102-1 19.

Harvey, C. 2000. Colonization of agricultural windbreaks
by forest trees: effects of connectivity and remnant
trees. Ecological Applications 10:1762-1773.

Holdridge, L. R. 1966. The life zone system. Adansonia
6:199-203.

Jahn, A. E., D. J. Levey, and K. G. Smith. 2004.
Reflections across hemispheres: a system-wide ap-
proach to New World bird migration. Auk 121 1005-
1013.

Jankowski, J. E., A. L. Ciecka, N. Y. Meyer, and K. N.
Rabenold. 2009. Beta diversity along environmental
gradients: implications of habitat specialization in

512

THE WILSON JOURNAL OL ORNITHOLOGY • Voi 122, No. 3, September 2010

tropical montane landscapes. Journal of Animal
Ecology 78:315-327.

Koh, L. P., R. R. Dunn, N. S. Sodhi, R. K. Colwell, H. C.
Proctor, and V. S. Smith. 2004. Species coextinc-
tions and the biodiversity crisis. Science 305:1632-
1634.

Kumar, A. and S. O’Donnell. 2007. Fragmentation and
elevation effects on bird-army ant interactions in
neotropical montane forest of Costa Rica. Journal of
Tropical Ecology 23:581-590.

Kumar, A. and S. O’Donnell. 2009. Elevation and forest
clearing effects on foraging differ between surface-
and subterranean-foraging army ants (Formicidae:
Ecitoninae). Journal of Animal Ecology 78:91-97.

Meisel, J. E. 2004. The influence of microclimate and
habitat area on the ecology of the army ant Eciton
burchellii in tropical forest fragments. Dissertation.
University of Wisconsin, Madison, USA.

Meisel, J. E. 2006. Thermal ecology of the neotropical
army ant Eciton burchellii. Ecological Applications
16:913-922.

Nakazawa, Y., a. T. Peterson, E. Martinez-Meyer,
AND A. G. Navarro-Siguenza. 2004. Seasonal niches
of Nearctic-neotropical migratory birds: implications
for the evolution of migration. Auk 121:610-618.

O’Donnell, S., J. Lattke, S. Powell, and M. Kasparl
2009. Species and site differences in neotropical army
ant emigration behavior. Ecological Entomology
34:476-482.

Otis, G. W., S. C. Santana, D. L. Crawford, and M. L.
Higgins. 1986. The effect of foraging army ants on
leaf-litter arthropods. Biotropica 18:56-61.

Peters, M. K., S. Likare, and M. Kraemer. 2008. Effects
of habitat fragmentation and degradation on flocks of
African ant-following birds. Ecological Applications
18:847-858.

Remsen Jr., j. V. AND D. V. Good. 1996. Misuse of data
from mist-net captures to assess relative abundance in
bird populations. Auk 113:381-398.

Rios, M., G. Londono, and L. Biancucci. 2008. Notes on
birds that follow army ants in the Northern Andes.
Ornitologia Neotropical 19:137-142.

Roberts, D. L., R. J. Cooper, and L. J. Petit. 2000. Use
of premontane moist forest and shade coffee agroeco-
systems by army ants in western Panama. Conserva-
tion Biology 14:192-199.

Stiles, F. G. and A. F. Skutch. 1989. A guide to the birds
of Costa Rica. Cornell University Press, Ithaca, New
York, USA.

Swartz, M. B. 2001. Bivouac checking, a novel behavior
distinguishing obligate from opportunistic species of
army-ant following birds. Condor 103:629-633.

Vallely, a. C. 2001 . Foraging at army ant swarms by fifty
bird species in the highlands of Costa Rica. Ornitho-
logia Neotropical 12:271-275.

Watkins II, J. F. 1985. The identification and distribution
of the army ants of the United States of America
(Hymenoptera, Formicidae, Ecitoninae). Journal of the
Kansas Entomological Society 58:479-502.

Willis, E. O. 1966. The role of migrant birds at swarms of
army ants. Living Bird 5:187-231.

Willis, E. O. 1972. The behavior of Plain-brown
Woodcreepers, Dendrocincla fuliginosa. Wilson Bul-
letin 84:377-420.

Willis, E. O. 1984. Neotropical thrushes (Turdidae) as
army ant followers. Ciencia y Cultura 36:1 197-1202.

Willis, E. O. and Y. Oniki. 1978. Birds and army ants.
Annual Review of Ecology and Systematics 9:243-
263.

Willson, S. K. 2004. Obligate army-ant-following birds: a
study of ecology, spatial movement patterns, and
behavior in Amazonian Peru. Ornithological Mono-
graphs 55:1-67.

Wrege, P. H., M. Wikelski, J. T. Mandel, T. Rasswei-
LER, and 1. D. CouziN. 2005. Ant birds parasitize
foraging army ants. Ecology 86:555-559.

Young, B. E. and D. B. MacDonald. 2000. Birds. Pages
179-222 in Monteverde: ecology and conservation of
a tropical cloud forest (N. Nadkarni and N. T.
Wheelwright, Editors). Oxford University Press, New
York, USA.

Young, B. E., D. Derosier, and G. V. N. Powell. 1998.
Diversity and conservation of understory birds in the
Tilaran Mountains, Costa Rica. Auk 1 15:998-1016.

The Wilson Journal of Omilhology 1 22(3): 5 1 3~5 1 7, 2010

MAROON-FRONTED PARROT {RHYNCHOPSITTA TERRISI) BREEDING
HOME RANGE AND HABITAT SELECTION IN THE NORTHERN
SIERRA MADRE ORIENTAL, MEXICO

SONIA GABRIELA ORTIZ-MACIEL,'- CONSUELO HORI-OCHOA,' AND

ERNESTO ENKERLIN-HOEELICH'

ABSTRACT. The Maroon-tronted Parrot (Rhynchopsitta lerrisi) is a threatened species endemic to pine-oak (Pinus
^'P^ -Quercus spp.) forests in “sky islands” of the Sierra Madre Oriental in Mexico. We measured parrot {n = 10) home
ranges during the breeding season (1999—2001) using radiotelemetry. Home ranges varied among years: 1999 =
12,379.05 ha, 2000 — 8,633.82 ha, and 2001 = 4,736.75 ha. Differences in home range size may reflect variation in
distribution, abundance, and patchiness of food. Daily movements in 1999 were estimated as 23.6 km versus 13 7 km in

2000 and 26.8 km in 2001. Habitat preferences within the P
was for Pinus-Abies-Pseiidotsuga forest and in 2000 and 20C
influenced by nesting behavior. Rapid landscape changes ma
areas for long term conservation of Maroon-fronted Parrots.

Maroon-fronted Parrots {Rhynchopsitta terrisi)
occupy temperate coniferous and mixed conifer-
ous-deciduous forests of the Sierra Madre Orien-
tal, Mexico, ranging from Nuevo Leon and
Coahuila in the north to Queretaro in the south.
Breeding occurs in a small area in Nuevo Leon
and Coahuila where parrots nest in cavities and
crevices in limestone cliffs that are often several
hundred meters high. Maroon-fronted Parrots nest
in aggregations and known nesting colonies have
a long history of use. Nesting occurs in summer
and early fall. Once breeding and nesting season
ends, usually in November, parrots migrate south
in the Sierra Madre Oriental (Enkerlin-Hoeflich et
al. 1999). Recent counts suggest a total population
of —3,500 individuals (Valdes-Pena et al. 2008).
These parrots are endemic and protected under
Mexican Government regulations, as well as
internationally (SEMARNAP 2000, Birdlife In-
ternational 2007, CITES 2007).

Maroon-fronted Parrots feed mainly on pine
(Pinus spp.) seeds (especially those of Pinus
strobiformis, P. montezumae, P. greggii, P.
cembroides, and P. culminicola), and flowers
and fruits of agave (Agave gentryi) during the
breeding season. They may fly long distances
searching for food, and annual movement patterns
vary considerably because of temporally and
spatially erratic nature of pine cone crops
(Enkerlin-Hoeflich et al. 1999). These parrots

' Centro de Calidad Ambiental, CEDES 5o pi.so, ITESM,
Avenida Eugenio Garza-Sada 2501 sur, C. P. 64849.
Monterrey, Nuevo Leon, Mexico.

^Corresponding author; e-mail: sgom@itesm.mx

'iius forest varied among nesting seasons. Preference in 1999
1 it was for Pinus forest and chaparral. This variation may be
/ necessitate planting and conserving pine forest to regenerate
Received 3 January 2008. Accepted 26 January' 2010.

eat acorns of several oak species (SGOM, pers.
obs.) during winter. Most pine forests in the
region are relicts and regeneration is scant or
absent once forests are destroyed by natural
causes or cut for use by humans.

Radiotelemetry has been used as a tool in
estimating parrot home range size because
monitoring movements of neotropical parrots is
difficult as they usually fly long distances, which
allows opportunities to select among different
forest habitats (Gilardi and Munn 1998). Our
objectives were to use radiotelemetry to: (1)
measure home range size, and (2) identify habitats
used by breeding Maroon-fronted Parrots during
three breeding seasons. We also estimated areal
requirements for different habitats during breed-
ing to effectively conserve and manage Sierra
Madre Oriental forests and habitats for this
important species.

METHODS

Our study was conducted in the northern
Sierra Madre Oriental from 100° 16' W, 25° 42'
N to 99° 50' W, 24° 50' N. This mountain chain is
600 km north-south and 80 km east-west.
Elevation varies from 2,000 to 2,500 m above
sea level with several parallel ,series of precipitous
limestone ridges along a northwest to southeast
direction. Climate varies with elevation, ranging
from extremely dry to semi-dry and hot to
subalpine climates at the highest elevations.
Annual precipitation averages 400 to 900 mm
(INEGI 1986).

We captured 10 adult Maroon-fronted Parrots
(4 in 1999, 6 in 2000) during the breeding season

513

514

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 3. September 2010

TABLE 1. Vegetation types, area, and number of visits by Maroon-fronted Parrots during the nesting seasons 1999,
2000, and 2001 in the Sierra Madre Oriental, Mexico.

Area used

Vegetation type 1999 2000 2001

Pinus- Abies-Pseudotsuga

4,739

ha

in =

-- 118)

29,350

ha

in =

= 28)

11,376

ha

in =

= 32)

Pinus

21,180

ha

in =

= 27)

3,504

ha

in --

= 37)

1,533

ha

in --

= 53)

Pinus cenibroides

2,571

ha

in =

= 20)

2,849

ha

in =

= 3)

744

ha

in =

= 3)

Pinus-Quercus

744

ha

in =

= 3)

9,986

ha

in =

= 0)

4,581

ha

in =

= 0)

Chaparral

9,666

ha

in =

= 14)

1,957

ha

in =

= 213)

660

ha

in =

= 53)

Totals

38,900

ha

in --

= 182)

47,645

ha

in =

= 281)

18,894

ha

in =

= 141)

with mist nets at the El Taray Sanctuary in
Coahuila and fitted them with telemetry neck-
collars (Holohil Systems Ltd., Carp, ON, Canada).
Radios weighed either 20 g (Model AI-2C, n = 9;
loop antennas) or 13 g (Model SI-2C, n = 1; whip
antenna) and represented 4.3 and 2.8% of average
bird weight, respectively. Both types of transmit-
ters had 24 months of expected battery life. The
average (± SD) distance for signal detection was
3.6 ± 1.3 km {n = 181; Ortiz-Maciel 2000).
These types of radios had no apparent negative
impact on parrots (Snyder et al. 1994, Meyers
1996, Salinas-Melgoza and Renton 2005). We
noted no apparent detrimental effects of transmit-
ter attachment during our study, and some birds
with non-functioning transmitters (dead batteries)
were observed at the nesting cliffs in 2007.

We attempted to locate and follow radio-
marked birds using 3-element yagi antennas and
TRX-2000 receivers (Wildlife Materials Inc.,
Carbondale, IL, USA) to measure home ranges.
We searched for parrots in high-elevation pine
forests along valley roads during their breeding
season, and also at nesting colonies and clay licks.
We radio-located birds from dawn to sunset for
39 days from August to November 1999, 47 days
from July to November 2000, and 43 days from
May to October 2001.

We established a bird’s position by recording
the direction of the maximum signal, taking a
bearing with a compass, and using a Global
Positioning System (GPS) unit to ascertain the
observer’s location. Only one observer took
bearings because of equipment limitations. We
moved along the road >3 km and took another
bearing to more precisely identity the location of
a bird. We did not observe radio-marked birds
directly at times, but usually heard vocalizations
of parrot Hocks, because daily tracking started at a

roosting site. We also located parrots at nesting
cliffs in their cavities. All bird locations were
recorded on 1:50,000 scale maps of the Sierra
Madre Oriental (McFarland 1991, Ortiz-Maciel
2000). Transmitter signals were not detected
during all searches because pan'ots may have
been out of the receiver’s range and behind ridges
that blocked the transmitter’s signal.

Site Characterization. — We plotted all loca-
tions on a landscape image (from Landsat-TM
1993) and processed it with Arcview GIS 3.2
(ESRl 1999). Forest habitats, characterized fol-
lowing Hori-Ochoa (1998), were: Pinits-Abies-
Pseudotsuga, Piniis, Pinus cemhroides, Pinus-
Quercus, Querciis-Pimis, and Quercus. Other
habitats included submontane shrub, microphyl-
lous desert shrub, desert shrub, chaparral, grass-
land, and agricultural zones.

We measured breeding home range size using
the 50% kernel method for each field season
(Worton 1989), and processed data using the
Animal Movement extension from Arcview GIS
3.2 (ESRI 1999). We followed some parrots for an
entire day and calculated total distances traveled
by adding the distances between each location.

Habitat selection was estimated for each
breeding season with a Chi-square independence
test by counting the number of visits to each
vegetation type and comparing the areal propor-
tions for the 95% minimum convex polygon
(White and Gairott 1990) (Table 1).

We calculated Bonferroni confidence intervals
to identify use by vegetation types (Neu et al.
1974, Byers and Steinhorst 1984, White and
Garrott 1990). Confidence intervals (95%) were
calculated using only habitats where birds were
observed including Piuus-Ahies-Pseiidotsuga, Pi-
nu.s, Pinu.s cemhroide.s, and Piuus-Quercits forest
and chaparral. Values are presented as means

Ortiz- Made! et al. • M A[^OON-FRONTED PARROT HOME RANGES

515

TABLE 2. Pioportioii ol Maroon-tronted Parrol locations by habitat type, habitat available, and Bonferroni confidence
intervals tor the 1999. 2()()(), and 2001 breeding seasons, Sierra Madre Oriental, Mexico

Year and habitat type

Pmportion of parrot
locations (P,,/,,)

Proportion of habitat
available (P„,,)

Bonferroni
confidence iniervals

1999

Pinus- Ahie.'i-P.seudot.uigu forest'

0.65

0.12

0.56

<

Pob. ^

0.74

Pinus forest

0.15

0.54

0.08

<

P,,!,, ^

0.22

Pinus cenibroides forest

0.1 1

0.07

0.05

<

Pnhs S

0.17

Pinus-Quercus forest

0.02

0.02

-0,01

<

Poh'^ ^

0.04

ChapaiTal

0.08

0.25

0.03

<

Pnbs ^

0.13

2000

Pinus-Ahies-Pseudotsuga forest

0.10

0.62

0.05

<

Pohs ^

0.15

Pinus forest"

0.13

0.07

0.08

<

Poh. ^

0,18

Pinus cenibroides forest

0.01

0.06

-0.01

<

A.Av ^

0.03

Pinus-Quercus forest

0.00

0.21

0.00

<

Poh, ^

0.00

Chaparral"

0.76

0.04

0.69

<

0.82

2001

Pinus-Ahies-Pseudotsuga forest

0.23

0.60

0.14

<

Pob, ^

0.32

Pinus forest"

0.38

0.08

0.27

<

Pob, ^

0.48

Pinus cenibroides forest

0.02

0.04

-0.01

<

Pobs -

0.05

Pinus-Quercus forest

0.00

0.24

0.00

<

Pobs ^

0.00

Chaparral"

0.38

0.04

0.27

<

Pobs ^

0.48

■“ Preferred habitat.

(± SD) with P < 0.5. The null hypothesis for
each bird was that habitat types were used in
accordance with their availability.

RESULTS

Home Range Size. — Home range size obtained
with 50% kernel methods varied among breeding
seasons. Home range size in 1999 was
12,379.05 ha [n = 3 birds), 8,633.82 ha [n = 7
birds) in 2000, and 4,736.75 ha (/? = 4 birds) in
2001.

Daily Movements. — We ob.served two parrots in
1999 move a mean (± SD) distance of 23.6 ±
5.7 km per day (« = 15 days). The mean distance
moved per day for three birds was 13.66 ±
7.64 km per day (n = 6 days) in 2000; the mean
distance was 26.82 ± 18.17 km (n = 5 days) for
two parrots in 2001. We also located a frequently-
used communal night roost in Sierra La Viga.

Habitat Selection. — Maroon-fronted Parrots
demonstrated preference for specific habitat types
within the pine forest mosaic ba.sed on frequency
of locations within 95% minimum convex poly-
gon home ranges. There was a difference between
habitat used and areal abundance (breeding
seasons: 1999, P < 0.001, X-4 = 493.5; 2000,
P < 0.001, = 3,721.9; 2001 P < 0.001, =

688.4). Bonferroni confidence intervals indicated

some of the six forests habitats were used more
than expected. Pinus-Ahies-Pseudotsuga forest
was used most in 1999, whereas Pinus forest
and chaparral were used most in 2000 and 2001
(Table 2).

DISCUSSION

Most Maroon-fronted Parrot locations were at
the highest elevations of the mountains, where
subfreezing temperatures persist, indicating they
are similar to Thick-billed Parrots (Rhvnchopsitta
pachyrhyncha) and are temperate in their habitat
preferences (Snyder et al. 1994, 1999). Breeding
activity of Maroon-fronted Pan'ots started nor-
mally in 1999, but no young fledged and all nests
failed by mid-September. Large home ranges that
year reflected food scarcity similar to that
reported for Thick-billed PaiTOts where breeding
success and effort appeared tightly linked to food
availability (Enkerlin-Hoeflich et al. 1999). Most
used habitats were in high elevation Pinits-Ahies-
Pseudot.suga forest generally considered to be less
suitable for foraging because the small reward
ratio of small seeds characteristic of tree species
in these forests (Enkerlin-HoeHich et al. 1999).

Average home ranges of breeding Maroon-
fronted PaiTots were larger than those of other
psittacines; juvenile Puerto Rican Amazon {Ama-

516

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 3, September 2010

zona vittata) have estimated home ranges of
1,372 ha (Lindsey et al. 1991), Ground Parrots
{Pezoponis wallicus) of Australia and Tasmania
have estimated home ranges of 9.2 ha (McFarland
1991), and juvenile Lilac-crowned Amazon
(Amazona finschi) have home ranges of 4,674 ha
(Salinas-Melgoza 2003). These home range dif-
ferences may reflect that Maroon-fronted Parrots
now inhabit disturbed areas and require larger
areas to satisfy their basic needs. Continental
parrot species have larger home ranges, compared
with island species, because of more availability
of habitats assuring refuge during periods of stress
(Snyder et al. 1987).

Maroon-fronted Parrots did not use all forest
types equally; all of their activities were in five
habitats (4 forest types and chaparral). Pinus-
Ahies-Pseudotsiigci forest was used more than
expected in 1999, whereas Pinus forest and
chaparral were more used than expected in 2000
and 2001. The El Taray nesting cliff is in
chaparral habitat which may influence apparent
habitat preferences.

The dependence of Maroon-fronted Parrots on
coniferous forests increases their vulnerability with
respect to habitat changes resulting from timber
harvest and fires as noted >25 years ago (Lawson
and Lanning 1981, Enkerlin-Hoeflich et al. 1999).
These temperate forests support different vegeta-
tion types and regeneration, management, or
restoration varies depending on each plant com-
munity. For example, after a fire, chaparral stands
of 20-30 years in age have nearly com-
plete shrub cover with few to no emergent
trees (Brown et al. 2000). Thus, the overall
productivity of habitats for parrots is decreasing
given the slow or absent forest regeneration and
replacement by oak chaparral with little food value
for parrots. Spatial and temporal patchiness of food
plants may be increasing which may lead to food-
induced stress on the population. Maroon-fronted
Parrots also have local pressure for one of their
preferred food sources, pinion pine seeds, which
are valuable at the local market and extensively
harvested as an income supplement by rural
families.

Recent fires have devastated several important
sites, including foraging and breeding areas;
>90% of standing trees died in 1998 in an area
of 5,000 ha (Brown et al. 2000). Another fire in
2006 destroyed 545 ha of standing trees
(CONAFOR 2006a, b) including >70% of the
El Taray Sanctuary (Manzano-Camarillo 2006).

These habitat changes have directly affected
Maroon-fronted Parrot nesting activities.

Regional forest managers should conserve and
re-establish Pinus forest communities. Manage-
ment should include control of timber harvests,
fires, reforestation, and restoration (Valdez-Ta-
mez 1981). Eorest management should be imple-
mented to manage and conserve critical habitats,
increase size of forested areas, and provide
alternative livelihood opportunities for local
inhabitants via ecotourism or provision of pay-
ment and incentives for ecosystem services and
watershed protection.

ACKNOWLEDGMENTS

This research was funded by ITESM Campus Monterrey
and Centro de Calidad Ambiental, E. Alexander Bergstrom
Awards Program, American Bird Conservancy’s William
Belton Grants Program, and the World Parrot Trust. We
appreciate the suggestions of Noel Snyder on this
manuscript and the knowledge he shared with us on
Maroon-fronted Parrots. We also thank Museo de las Aves
de Mexico for use of facilities at the El Taray Sanctuary;
Fabian Lozano for support with the Geographic Information
System Laboratory; Patricia Vela, Santiago Salazar, and
Fabiola Yepez for helping with image processing; Roberto
Mercado for help with statistics; and Javier Cruz, Jose
Manzano, Julio Ronquillo, Romualdo Martinez, and
Claudia Macias for field support.

LITERATURE CITED

Birdlife International. 2007. Maroon-fronted Parrot.
BirdLife species factsheet. www.birdlife.org/datazone/
.species/index. html.

Brown, S., M. Delaney, and D. Shoch. 2000. Assessing
the potential for carbon sequestration at three forest
fire restoration sites in Mexico. Environmental Policy
and Institutional Strengthening Indefinite Quantity
Contract Number PCE-I-00-96-00002-00. Prepared
by Winrock International team for U.S. Agency for
International Development Mission to Mexico, Little
Rock, Arkansas, USA.

Byers, C. R. and R. K. Steinhorst. 1984. Clarification on
a technique for analysis of utilization availability data.
Journal of Wildlife Management 48: 1050-1052.
CITES. 2007. Convention on International Trade in
Endangered Species of Wild Fauna and Flora.
Appendices I, II, and III. www.cites.org/eng/app/
appendices. shtml.

CONAFOR. 2006a. Comision Nacional Forestal. conafor.
gob.mx/portal/docs/secciones/comunicaciQn/B-052006.
pdf

CONAFOR. 2006b. Comision Nacional Forestal. conafor.
gob.mx/portal/docs/seccione.s/comunicacion/B- 1 22006.
pdf

Enkeri.in-Hoefijch, E. C., C. Macias-Caballero, T.
Monterrubio-Rico, M. A. Cruz-Nieto, N. F. R.
Snyder, D. Venegas-Holguin, J. Cruz-Nieto, G.

Ortiz-Mciciel Cl al. • MAROON-FRONTED PARROT HOME RANGES

517

Ortiz-Maciel, J. Gonzalez-Elizondo, and E.
Stone. 1999. Statu.s, di.stribucion, ecologia y con-
servacion dc la.s cotorras serranas (Rhynchopsitta
pachyrhyiiclui y R. terrisi) en el Norte de Mexico.
Final report. In.stituto Tecnologico y de Estudios
Siiperiores de Monterrey. Centro de Calidad Ambien-
tal. Programa de Manejo Sostenible de Ecosistemas,
Monterrey, Nuevo Leon, Mexico.

Environmental Systems Research Institute (ESRI).
1999. Arcview GIS 3.2. Environmental Systems
Research Institute Inc., Redlands, California, USA.

Gilardi, J. D. and C. A. Munn. 1998. Patterns of activity,
flocking and habitat use in parrots of the Peruvian
Amazon. Condor 100:641-653.

Hori-Ochoa. M. C. 1998. Variabilidad de las respuestas
espectrales de acuerdo a los cambios fenologicos de la
vegetacion del anticlinorio de Arteaga, Coahuila y
Nuevo Leon, Mexico. Thesis. Institute Tecnologico y
de Estudios Superiores de Monterrey, Nuevo Leon,
Mexico.

INSTITUTO NaCIONAL DE ESTADISTICA GeOGRAFIA E IN-
formatica (INEGI). 1986. Sintesis Geografica del
Estado de Nuevo Leon, Mexico. Monterrey, Nuevo
Leon, Mexico.

Lawson, P. W. and D. V. Panning. 1981. Nesting and
status of the Maroon-fronted Parrot (Rhynchopsitta
terrisi). Pages 385-392 in Conservation of New World
parrots (R. F. Pasquier, Editor). Parrot Working Group
Meeting, St. Lucia. Smithsonian Institution Press,
Washington, D.C., USA.

Lindsey, G. D., W. J. Arendt, J. Kalina, and G. W.
Pendleton. 1991. Home range and movements of
juvenile Puerto Rican Parrots. Journal of Wildlife
Management 55:318-322.

Manzano-Camarillo, M. 2006. Diagnostico de dafios por
incendios forestales del predio El Taray, municipio de
Arteaga, Coahuila. Technical report presented to
Comision Nacional de Biodiversidad. Institute Tecno-
logico y de Estudios Superiores de Monterrey. Centro
de Calidad Ambiental. Programa de Recursos Fore-
stales y Zonas Aridas, Monterrey, Nuevo Leon,
Mexico.

McFarland, D. C. 1991. The biology of the Ground
Parrot, Pezoporus wallicus, in Queensland. I. Micro-
habitat use, activity cycle and diet. Wildlife Research
18:169-184.

Meyers, J. M. 1996. Evaluation of 3 radio transmitters and
collar designs for Amazona. Wildlife Society Bulletin
24:15-20.

Nfiu, C. W., C. R. Byers, and J. M. Peek. 1974. A
technique for analysis of utilization availability data.
Journal of Wildlife Management 38: 541-545.

Ortiz-Maciel, S. G. 2000. Uso del paisaje por la cotorra
serrana oriental (Rhynchopsitta terrisi). Thesis. In-
stitLito Tecnologico y de Estudios Superiores de
Monterrey, Nuevo Leon, Mexico.

Salinas-Melgoza, A. 2003. Dinamica espacio-temporal
de individuos Juveniles del loro corona lila (Amazona
finschi) en el bosque seco de la costa de Jalisco.
Thesis. Universidad Nacional Autonoma de Mexico,
Mexico, D.F.

Salinas-Melgoza, A. and K. Renton. 2005. Seasonal
variation in activity patterns of juvenile Lilac-crowned
Parrots in tropical dry forest. Wilson Bulletin
117:291-295.

SEMARNAP. 2000. Secretaria de Medio Ambiente,
Recursos Naturales y Pesca. Proyecto para la Con-
servacion, Manejo y Aprovechamiento Sustentable
de los Psitacidos en Mexico. SEMARNAP, Mexico,
D.F.

Snyder, N. F. R., J. W. Wiley, and C. B. Kepler. 1987.
The parrots of Luquillo: natural history and conserva-
tion of the Puerto Rican Parrot. Western Foundation of
Vertebrate Zoology, Los Angeles, California, USA.

Snyder, N. F. R., E. C. Enkerlin-Hoeflich, and M. A.
Cruz-Nieto. 1999. Thick-billed Parrot (Rhynchopsitta
pachyrhyncha). The birds of North America.
Number 406.

Snyder, N. F. R., S. E. Koening, J. Koschmann, H. A.
Snyder, and T. B. Johnson. 1994. Thick-billed
Parrot releases in Arizona. Condor 96:845-862.

Valdes-Pena, R. a., S. G. Ortiz-Maciel, S. O. Valdez-
Juarez, E. C. Enkerlin-Hoeflich, and N. F. R.
Snyder. 2008. Use of clay licks by Maroon-fronted
Parrots (Rhynchopsitta terrisi) in northern Mexico.
Wilson Journal of Ornithology 120:176-180.

Valdez-Tamez, V. 1981. Contribucion al conocimiento de
los tipos de vegetacion, su cartografia y notas
floristico-ecologicas del municipio de Santiago, Nuevo
Leon, Mexico. Thesis. Universidad Autonoma de
Nuevo Leon, Monterrey, Nuevo Leon, Mexico.

White, G. C. and R. A. Garrott. 1990. Analysis of
wildlife radio-tracking data. Academic Press, London,
United Kingdom.

WORTON, B. J. 1989. Kernel methods for estimating the
utilization distribution in home-range studies. Ecology
70:164-168.

The Wilson Journal of Ornithology 122(3):5 18-531 , 2010

SHORT-TERM EFFECTS OF FIRE ON BREEDING BIRDS IN
SOUTHERN APPALACHIAN UPLAND FORESTS

NATHAN A. KLAUS,' SCOTT A. RUSH,"-^ TIM S. KEYES,'

JOHN PETRICK.^ AND ROBERT J. COOPER-

ABSTRACT. — We investigated how variation in fire severity (control or no fire; low, medium, and high severity fires)
and interval (1-2 years vs. 3-6 years after fires) affected habitat and avian abundance, species diversity, richness, and
evenness in the southern Appalachian Mountains. Fire severity and interval had significant implications for both habitat and
avian communities. Species richness within 2 years of fires was on average 26% higher in areas receiving medium and high
severity treatments than in unburned control units. Species diversity and species richness were markedly greater 3-6 years
after fires within high severity treatments (12 and 44%, respectively), compared to unbumed controls. Relative abundance
and species evenness did not vary with fire severity or time since fire. The short-term effects of low severity fires, or high
severity fires with short rotation periods (^2 years) may have limited positive effects on avian communities. Facilitation of
disturbance regimes including mid to high severity fires, which foster uneven-aged forests, can be an effective conservation
tool for restoring avian communities. Received 30 June 2009. Accepted 1 March 2010.

Disturbance is a fundamental ecological pro-
cess that has had a role in structuring and
maintaining diversity within many ecological
communities (Loucks 1970, Elliott et al. 1999,
Brawn et al. 2001, Bond and Keeley 2005, Bond
et al. 2005). Generally, disturbance increases
diversity by creating a mosaic of habitats or
SLiccessional stages within a landscape (Askins
2001, Brawn et al. 2001). Fire is one form of
disturbance that has had a large part in structuring
the ecological communities of the southern
Appalachian Mountains (Lorimer 1980, Van Lear
and Waldrop 1989, Waldrop et al. 1992).
Suppression of fire on public lands during the
past century has likely reduced diversity and
structure of animal and plant communities
(Abrams 1992, Lorimer 2001, Artman et al.
2005). Several federal land management agencies,
in an effort to restore diversity and ecological
function, now plan for increased use of prescribed
fire in the southern Appalachian Mountains.
However, effects of restoring fire on wildlife are
largely unknown in these mostly hardwood forest
systems. In particular, effects of varying fire
.severity and interval on wildlife are largely
unknown.

Many bird species associated with early
succession habitats and pine (Pinus spp.) savannas

' Nongame Con.servation Section, Georgia Department of
Natural Resources, 116 Rum Creek Drive, Forsyth, GA
31029. USA.

^ D. B. Warnell School of Fore.st Resources, University of
Georgia, Athens, GA 30602, USA.

’ Chattahoochee-Oconee National Forest, 1755 Cleveland
Highway, Gainesville. GA 30501, USA.

■‘Corresponding author; e-mail; srush@uwindsor.ca

would likely benefit from return of fire to the
southern Appalachian Mountains. Species of
concern include: Bachman’s Sparrow {Aimophila
aestivalis). Northern Bobwhite {Colinus virginia-
nus), Red-cockaded Woodpecker {Picoides bor-
ealis), Prairie Warbler (Dendroica discolor), and
Golden-winged Warbler {Vermivora chrysoptera)
(Brewster 1886, Burleigh 1958, Stupka 1963,
Hunter et al. 2001, Klaus 2004). Measured
responses for other bird species may not be
positive (Lang et al. 2002, Artman and Down-
hower 2003, Tomcho et al. 2006), and the benefits
of fire restoration may not integrate similarly
across all members of avian communities, or in all
habitats (Artman et al. 2001, Saab and Powell
2005, Tomcho et al. 2006, Greenberg et al. 2007).
Thus, reintroduction of fire should be based on
desired ecological condition.

The historical suppression of fire and changes
in agricultural practices within the eastern
United States have been implicated in loss of
early successional habitat (Askins 2001). Loss
of this habitat type has had a significant impact
on avian communities. Early succession song-
birds in the southern Appalachians have the
.strongest declines of any group of birds (Hunter
et al. 1999, Sauer et al. 2005). Many of the
species that require mature forest for nesting also
use early succession habitat as fledglings and
during molt (Anders et al. 1998, Vega Rivera et
al. 1999, Marshall et al. 2003, Rush and Stutch-
bury 2008). Restoration of early successional
habitat may be a key feature of policies directed at
conservation of many avian species (Dettmers
2003, Bulluck and Buehler 2006, Buehler et al.
2007).

518

Klaus et at. • FIRE AND BIRDS IN THE SOUTHERN APPALACHIAN MOUNTAINS 519

Prescribed fire is often cited as a likely tool for
restoring and providing suitable habitat for many
avian species (Partners in Flight Working Group
2002), but our current understanding of the
appropriate application of fire in shaping forest
ecosystems of the southern Appalachian Moun-
tains remains limited. We currently know of only
a few studies that have examined the effects of
prescribed fire on songbirds in hardwood forests
of the eastern United States (Aquilani et al. 2000;
Artman et al. 2001, 2005; Greenberg et al. 2007);
only one of these studies occurred in the southern
Appalachians. Our objectives were to document
how breeding season distributions of birds
changed with variation in fire severity and time
since fire in the southern Appalachians.

METHODS

Study Area. — This study was conducted within
the Brasstown and Tallulah ranger districts of the
Chattahoochee-Oconee National Forest (CNF),
Murray County, Georgia, USA. U.S. Forest
Service (USFS) personnel were consulted within
the CNF and 22 potential study sites were
identified at elevations from 600 to 1,500 m. All
bum units were >200 ha and had burned in the
last 8 years, either as a prescribed fire or as part of
a wildfire. Vegetation sampling sites were placed
into one of three fire severity groups on the basis
of the extent of canopy mortality: low, medium, or
high. Low severity burns were typical of most
‘fuel reduction’ prescribed fires in the southern
Appalachians. Flame lengths were generally less
than 1 m, resulting in <5% canopy mortality
throughout the burn unit and low to moderate
midstory mortality. Moderate to severe midstory
and 5 to 20% canopy mortality characterized
medium severity burns and flame lengths ranged
from <0.5 m in mesic sites to 4 m along dry ridge
tops, leaving a heterogeneous result. High severity
bums were characterized by 20 to 50% canopy
mortality, mostly along ridge tops, and severe to
total midstory mortality except in ravines.

We selected 12 of the 22 burn units, four of
each treatment, based on similarity of forest type,
elevation, and average aspect. All chosen burn
units had been burned <6 years earlier. Half the
12 burn units selected (2 of each treatment) were
in sites that were burned <2 years previously, and
two were in sites that were burned 3-6 years
previously. Forest types consisted primarily of
mixed pine (Finns spp.)-hardwoods and drier oak-
hickory (Qiiercus-Carya), although ridge tops

occasionally included small stands of F. taeda,
F. virginiana, F. rigida, and F. pungens. Four
control units were also identified with the aid of
USFS personnel. Control units were similar in
forest type, elevation, and average aspect but had
not burned in >20 years.

Bird Surveys. — Surveys were conducted by
USFS and Georgia Department of Natural Re-
sources staff. Participants received several weeks
of spring training on bird identification and point-
count techniques prior to surveys. All participants
had conducted bird surveys for several years prior
to this study. Two transects were placed for each
survey within each burn unit. Transects followed
contours and were at least 200 m apart. All
surveys were conducted between 0630 and 1100
hrs, 17 May-10 June 2004, following protocols
established by Hamel et al. (1996), except that
counts were conducted for 10 min and distance
bands of 0-10, 11-25, 26-50, 51-100, and
>100 m were used. Ten point counts were
conducted on each survey transect, 20 per burn
unit. Locations of individual point counts were:
(1) separated by at least 200 m, (2) at least 100 m
inside the burn unit, and (3) placed in upland
mixed pine hardwood stands or oak stands.
Ravines, road edges, and other cover types were
avoided. Four replications of each treatment were
surveyed for a total of 80 points per treatment.

Habitat Measurements. — We conducted vari-
able-radius vegetation surveys within each burn
unit centered on each point count location. We
measured the basal area using a 10-factor prism of
live trees >25 cm diameter breast height (DBH);
this metric is referred to as BASAL AREA. We
visually estimated average canopy (CANOPY
COVER) and herbaceous cover (HERBACEOUS
COVER) to the nearest 10% for a 25-m radius
around each plot center. We also measured shrubs
(woody plants 1 to 7 m tall and <12 cm DBH)
within a 3-m radius around the plot center by
counting all stems (SHRUB DENSITY) and
measuring average shrub height (SHRUB
HEIGHT) to the nearest decimeter. Each burn
unit was categorized based on known fire histories
into one of three fire histories: (1) control [not
burned within the last >20 years], (2) burned
within 1-2 years of surveys, and (3) burned within
3-6 years of surveys.

Data Analy.ds.-—V^e restricted our analysis to
limit the repeated counting of the same individual
at two adjacent survey sites and to include only
those birds detected within 100 m of each survey

520

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 3, September 2010

point. The relative abundance of each species for
each treatment was calculated as the average
number of individuals detected/ha. We restricted
each point-count survey to counts of individuals
detected per distance band during each time
interval for conservative estimates of relative
abundance of each species. Detections of the same
species in the same distance band but in different
time intervals were treated as a possible recount of
the same individual and were excluded from
further analysis. All bird species observed at each
survey point along each transect were used in
calculations of bird species diversity, species
richness, and species evenness. Species diversity
was calculated as the Shannon-Weaver Diversity
Index (//') (Shannon and Weaver 1963). Species
richness was calculated as the number of individ-
ual species observed per point during each point
count, and species evenness was calculated as H'
divided by the natural log of species richness
(Magurran 1988).

We used linear, mixed-effects models (Ime) in
R (Version 2.7.1; R Development Core Team
2008) to compare habitat measurements, diversi-
ty, species evenness, and species richness between
treatments using the burn unit as the experimental
unit. Mixed-effects models are appropriate for this
data structure because they partition information
into multi-level components. The data recom-
mends level- 1 (survey point), level-2 (survey
transect), and level-3 (burn unit) components in
our case. Each component is estimated with the
appropriate degrees of freedom, and the standard
errors of other parameter estimates are appropri-
ately adjusted (Raudenbush and Bryk 2002).
Transect and burn unit effects were accounted
for as random variables, and treatment as a fixed
effect. We used q-q plots prior to analysis to
examine the normality of each habitat metric. We
log transformed BASAL AREA, SHRUB DEN-
SITY, SHRUB HEIGHT, and HERBACEOUS
COVER and arcsine square-root transformed
CANOPY COVER to improve normality and
homoscedasticity. We used conditional t-tests for
each Ime to ascertain whether covariates were
significantly different from zero (Pinheiro and
Bates 2000). We also developed a series of
orthogonal contrasts to compare means between
the different levels of burn severity and times
since fire.

We used canonical correspondence analysis
(CCA) of data collected at each of the survey
points to examine the relationship between bird

community structure and measured environmental
variables (McCune and Grace 2002). This tech-
nique was used to produce graphical presentations
depicting relationships among abundance of
individual species, treatments, and measured
environmental gradients. Habitat variables ex-
plaining a significant amount of variation (P ^
0.10), as calculated by Monte Carlo permutation
tests (1,000 random permutations of samples in
the species data; package Vegan in R) (R
Development Core Team 2008), were included
in the CCA analyses. The means of these
variables are represented by the origin in the
resulting diagram. We constrained our analysis for
clarity in presentation of bird assemblages to
include only those species that were detected at
>20% of all survey points. We conducted two
separate analyses to facilitate comparisons, one
comparing habitat and bird assemblage between
control and treatments 1-2 years after fires, and a
separate analysis with similar comparisons for
control and treatments sampled 3-6 years after
fires. We log transformed BASAL AREA,
SHRUB DENSITY, SHRUB HEIGHT, and
HERBACEOUS COVER and arcsine square-root
transformed CANOPY COVER prior to using
them in the CCA.

RESULTS

Fire and Habitat. — Fire severity affected hab-
itat structure. Comparisons of treatments and
unburned controls revealed CANOPY COVER
and SHRUB DENSITY differed among treat-
ments (^6.9 ~ 19.1, P < 0.001; F6 9 = 4.0, P =
0.03, respectively), (Fig. 1). High severity treat-
ments had significantly less CANOPY COVER
than control units both 1-2 years it = 6.8, df = 9,
P < 0.001), and 3-6 years after fires (t = 7.9, df
= 9, P < 0.001). CANOPY COVER did not
differ relative to the number of years since fires
it = 1.0, df = 9, P = 0.33) but, SHRUB
DENSITY was significantly lower among high
severity treatments 1-2 years after fires than 3-
6 years after fires {t = —4.4, df = 9, P = 0.002).

Ejfect.s on Avian Diversity and Communities. —
Sixty-three species were detected among the four
treatments (Table 1 ). Species richness differed
among several treatments (Pf,.9 = 4.9, P = 0.02)
and was significantly higher 1-2 years after fires
relative to the low severity burn units. It did not
differ among medium and high severity burn units
(t = -2.5, df = 9, P = 0.03, t = -2.3, df = 9,
P = 0.05) (Fig. 2). Species richness 3-6 years

Herbaceous cover (%) Slmib stems/m^ Canopy cover (%) Basal ai'ea (mMia)

K/dns el al. • FIRE AND BIRDS IN THE SOUTHERN APf^ALACHIAN MOUNTAINS 521

o

o

o

0-1

00

C>'

■o

O'

O'

O'
O -I

D

Conti'ol

I

I Low Medium Higli

I 1 to 2 year's

I Low Medium Higli

* 3 to 6 years

Fire intensity/year's since fire

FIG. I. Box and whisker plots of habitat characteristics measured at point-count locations within control and burn
treatments. Whiskers represent maximum and minimum observations while boxes repre.sent the 25 and 75% quartiles.

522

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 3, September 201U

TABLE 1. Relative abundance of species detected during surveys in the southern Appalachian Mountains. Estimates
are individuals/ lOha (.v ± SE). Missing values indicate species not detected.

Species

Scientific name

Ruffed Grouse
Bonasa umbellus
Wild Turkey

Meleagris gallopavo
Black Vulture
Coragyps atratus
Turkey Vulture
Cathartes aura
Sharp-shinned Hawk
Accipiter striatus
Broad-winged Hawk
Buteo platyptenis
Red-tailed Hawk
B. jamaicensis
Mourning Dove
Zenaida macroura
Yellow-billed Cuckoo
Coccyzus americanus
Barred Owl
Strix varia
Chimney Swift
Chaetiira pelagica
Ruby-throated Hummingbird
Archilochus coliihris
Red-bellied Woodpecker
Melanerpes carolinus
Downy Woodpecker
Picoides puhescens
Hairy Woodpecker
P. villosiis
Northern Flicker
Colaptes aiiratus
Pileated Woodpecker
Dryocopus pileatus
Eastern Wood-Pewee
Contopus virens
Eastern Phoebe
Sayornis phoehe
Great Crested Flycatcher
Myiarchiis crinitus

Treatment (years after fires)

Low Medium High

Control 1-2 3-6 1-2 3-6 1-2 3-6

0.1 ± 0.1

0.2 ±0.1 0.1 ± 0.1

0.2 ± 0.2

0.2 ± 0.2 0.3 ± 0.2

0.2 ± 0.2

0.3 ± 0.2

0.2

± 0.2

0.1 ± 0.1

0.1 ±0.1

0.2 ± 0.1

0.1

± 0.1

0.3 ± 0.2

0.3

± 0.2

0.2 ± 0.1

0.6 ± 0.2

0.7 ± 0.2

1.0 ± 0.3

0.3

± 0.2

0.2 ± 0.1

0.5 ± 0.2

0.3 ± 0.2

0.5 ± 0.2

0.1

± 0.1

0.1

± 0.1

0.3

± 0.2

0.6 ± 0.2

0.2 ± 0.2

1.2 ± 0.4

0.6 ± 0.3

0.1

± 0.1

0.1 ± 0.1

0.1 ± 0.1

0.1 ±0.1

0.1

± 0.1

0.2 ± 0.1

0.2 ± 0.1

0.1 ± 0.1

0.1 ± 0.1

0.2

± 0.1

O.I ±0.1

0.2

± 0.1

0.6 ± 0.2

0.1 ± 0.1

1 .2 ± 0.3

0.6 ± 0.2

0.2

± 0.1

0.1 ±0.1

0.1 ± 0.1

0.7 ± 0.3

0.4 ± 0.2

0.9

± 0.3

1.6 ± 0.3

1.1

± 0.2

1.5 ± 0.4

1.6 ± 0.3

1 .3 ± 0.3

1.5 ± 0.3

0.7 ± 0.2

0.2 ± 0.2

1 .4 ± 0.3

2.4 ± 0.4

0.1

± 0.1

0. 1 ± 0. 1

0.2 ± 0.1

0.1

± 0.1

0. 1 ±0.1

0.1

± 0.1

0.2 ± 0.1

0.2 ± 0.1

0.5 ± 0.2

0.6 ± 0.2

Klaus et a!. • FIRE AND BIRDS IN THE SOUTHERN APPALACHIAN MOUNTAINS 523

TABLE 1. Continued.

Treatment (years after fires)

Species

Low

Medium

High

Scientific name

Control

1-2

.3-6

1-2

3-6

1-2

3-6

Yellow-throated Vireo

0.2 ± 0.1

0.4 ± 0.2

Vireo flavifrons

Blue-headed Vireo

0.6 ± 0.2

0. 1 ± 0. 1

1 .0 ± 0.3

0.7 ± 0.2

0.7 ± 0.2

1.3 ± 0.3

0.7 ± 0.2

V. solitariiis

Red-eyed Vireo

5.6 ± 0.4

4.0 ± 0.5

4.4 ± 0.5

4.0 ± 0.5

5.8 ± 0.5

3.5 ± 0.6

5.4 ± 0.5

V. olivaceus

Blue Jay

0.9 ± 0.3

0.6 ± 0.2

0.9 ± 0.3

0.9 ± 0.3

0.6 ± 0.2

0.6 ± 0.2

0.8 ± 0.3

Cyanocitta cristata

American Crow

1.5 ± 0.3

4.6 ± 0.6

3.7 ± 0.4

1.3 ± 0.3

1.4 ± 0.3

1.3 ± 0.3

2.7 ± 0.4

Con’us hrachyrhynchos

Carolina Chickadee

0.3 ± 0.2

0.5 ± 0.2

0.2 ± 0.1

0.7 ± 0.3

0.1 ± 0.1

0.4 ± 0.2

Poecile carolinensis

Tufted Titmouse

1.0 ± 0.2

1.4 ± 0.2

2.5 ± 0.4

1.7 ± 0.3

1.6 ± 0.3

0.6 ± 0.3

2.1 ± 0.3

Baeolophus bicolor

White-breasted Nuthatch

0.5 ± 0.2

0.2 ± 0.1

0.4 ± 0.2

0.4 ± 0.2

0.5 ± 0.3

0.6 ± 0.3

0.5 ± 0.2

SiUa carolinensis

Carolina Wren

0.2 ± 0.1

0.3 ± 0.2

0.2 ± 0.1

0.6 ± 0.2

0.4 ± 0.2

1.3 ± 0.3

1.4 ± 0.3

Thryothorus kidovicianus

Blue-gray Gnatcatcher

0.1 ± 0.1

0.2 ± 0.1

0.1 ± 0.1

0.6 ± 0.3

Polioptila caerulea

Wood Thrush

0.1 ± 0.1

0.6 ± 0.3

0.2 ± 0.1

0.2 ± 0.1

0.6 ± 0.3

0.4 ± 0.2

0.3 ± 0.2

Hylocichla mustelina

Brown Thrasher

0.1 ± 0.1

0.1 ± 0.1

0.1 ± 0.1

Toxostoma rufum

Cedar Waxwing

0.1 ± 0.1

0.9 ± 0.5

0.2 ± 0.2

Bombycilla cedrorum

Golden-winged Warbler

0.2 ± 0.2

Vennivora chrysoptera

Northern Parula

4.0 ± 0.4

0.1 ± 0.1

Parula americana

Chestnut-sided Warbler

0.4 ± 0.2

0.2 ± 0.2

1 .4 ± 0.4

0.8 ± 0.2

1 .5 ± 0.4

6.6 ± 0.5

Dendroica pensylvcmica

Black-throated Blue Warbler

0.6 ± 0.3

1 .0 ± 0.3

D. caendescens

Black-throated Green Warbler

1 . 1 ± 0.3

1.2 ± 0.3

1 .7 ± 0.4

1.4 ± 0.3

3.5 ± 0.5

0.8 ± 0.2

1 .0 ± 0.3

D. virens

Blackburnian Warbler

0.4 ± 0.2

0.9 ± 0.3

0.3 ± 0.2

0.4 ± 0.2

D. fusca

Yellow-throated Warbler

0.1 ±0.1

0.1 ± 0.1

0.9 ± 0.3

0.2 ± 0. 1

0.8 ± 0.3

D. dominica

524 THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 3. September 2010

TABLE 1. Continued.

Treatment (years after fires)

Low Medium High

Species

Scientific name

Control

1-2

3-6

1-2

3-6

1-2

3-6

Pine Warbler

0.2 ± 0.1

0.2 ± 0.1

0.3 ± 0.1

0.2 ± 0.1

D. piniis

Prairie Warbler

0.1 ±0.1

1.0 ± 0.3

D. di.tcolor

Cerulean Warbler

0.2 ± 0.2

D. cerulea

Black-and-white Warbler

1.8 ± 0.3

0.6 ± 0.2

1.4 ± 0.3

1.5 ± 0.3

1.3 ± 0.3

1.0 ± 0.2

0.9 ± 0.3

Mniotilta varia

American Redstart

0.1 ±0.1

0.1 ± 0.1

0.6 ± 0.2

Setophaga ruticilla

Worm-eating Warbler

0.6 ± 0.2

0.3 ± 0.2

0.3 ± 0.2

0.1 ± 0.1

0.5 ± 0.2

0.2 ± 0.2

Helmitheros vermivorum

Ovenbird

4.7 ± 0.5

1.1 ±0.3

2.3 ± 0.5

2.9 ± 0.4

1.2 ± 0.3

0.2 ± 0.2

Seiurus aurocapilla

Kentucky Warbler

0.2 ± 0.1

0.1 ± 0.1

0.1 ± 0.1

0.1 ± 0.1

0.2 ± 0.1

0.1 ± 0.1

Opororni.^ formosus

Common Yellowthroat

0.1 ± 0.1

Geothlypis trichas

Hooded Warbler

0.9 ± 0.3

1.8 ± 0.3

2.5 ± 0.4

0.6 ± 0.2

1.3 ± 0.3

1.4 ± 0.4

Wilsonia citrina

Canada Warbler

0.2 ± 0.1

W. canaden.sis

Yellow-breasted Chat

0.2 ± 0.2

0.1 ± 0.1

2.9 ± 0.5

Icteria vireos

Eastern Towhee

0.5 ± 0.2

0.4 ± 0.2

1.4 ± 0.4

0.8 ± 0.2

0.9 ± 0.3

4.2 ± 0.5

4.0 ± 0.5

Pipilo erythrophthcdmu.s

Chipping Sparrow

1.2 ± 0.4

0.2 ± 0.1

Spizella pcisserina

Field Sparrow

0.1 ± 0.1

S. pusilla

Song Sparrow

0.1 ± 0.1

Melospiza melodici

Dark-eyed Junco

0.1 ±0.1

0.2 ± 0.1

Jiinco hyertudis

Scarlet Tanager

1 .2 ± 0.3

1 .4 ± 0.3

2.4 ± 0.5

1 .9 ± 0.3

2.4 ± 0.4

2.6 ± 0.4

2.3 ± 0.4

Piranga oHvacea

Northern Cardinal

0.1 ±0.1

0.3 ± 0.2

0.8 ± 0.2

0.1 ± 0.1

0.1 ±0.1

0.6 ± 0.2

Ccirdbudi.'i cardimdis

Rose-breasted Grosbeak

0.2 ± 0.1

0.4 ± 0.2

0.3 ± 0.2

0.6 ± 0.2

Pheucticus litdovicianus

Klaus ei al. • FIRE AND BIRDS IN THE SOUTHERN APPALACHIAN MOUNTAINS 525

TABLE I. Continued.

Treatnienl (years after fires)

Species

Low

Medium

High

Scientific name

Control 1-2 3-6

1-2

3-6

1-2

3-6

Blue Grosbeak

Passerine! caeriilea

0.4

± 0.2

Indigo Bunting

P. cycinea

1.0 ± 0.3 0.7 ± 0.3 2.5 ± 0.5 2.6

± 0.4

2.1

± 0.3

4.0 ± 0.4

5.3

± 0.6

American Goldfinch

0.1

± 0.1

0.1

± 0.1

1.0 ± 0.3

0.6

± 0.3

Spilius tristis

after fires, relative to all other treatments, was
significantly higher among the high severity bum
units (Fig. 2). Generally, species diversity (//') did
not differ among treatments or relative to time
since fire (F(,_g = 2.4, P = 0.12). However, species
diversity was significantly greater in the high
severity bum units 3-6 years after fires in contrast
with the controls (/ = -3.0, df = 9, F = 0.1)
(Table 2). Species evenness did not differ among
treatments (F(, g = 0.43, P = 0.84) (Fig. 2).

Canonical correspondence analysis for bird
species assemblages and habitat measured 1-
2 years after fires indicated the overall relation-
ship between species and environmental variables
(all canonical axes) differed significantly from
those derived randomly (Table 2). The primary
axis (horizontal axis. Fig. 3) indicated strong
positive relationships with BASAL AREA,
SHRUB DENSITY, and CANOPY COVER, a
moderate positive effect with SHRUB HEIGHT,
and a moderate negative effect with HERBA-
CEOUS COVER (Table 2). The secondary axis
(vertical axis, Eig. 3), indicated a strong negative
relationship with SHRUB HEIGHT and weaker
relationships with SHRUB DENSITY, CANOPY
COVER, and BASAL AREA. The habitat in high
severity burn units 1-2 years after fires was best
characterized by reduced CANOPY COVER and
BASAL AREA (Eig. 3).

The avian community associated with high
severity burn units 1-2 years after fires was
largely represented by early successional species
including American Goldfinch (scientific names
are in Table 1 ), Eastern Towhee, Indigo Bunting,
and Eastern Wood-Pewee. These species were
along the negative side of the primary axis.
Ovenbirds, a species most closely associated with
habitat characterized by higher CANOPY COV-
ER, BASAL AREA, and HERBACEOUS COV-

ER tended to be in a positive position relative to
both the primary and secondary axes. Treatment
ellipses overlapped among other species indicat-
ing that habitat metrics and species pools were
similar among treatments (Fig. 3).

Relationships between species and habitat 3-
6 years after fires were significantly different for
all variables except SHRUB HEIGHT (Table 2).
This metric was omitted from further analysis.
The primary axis of the CCA had a strong
negative relationship with BASAL AREA and
CANOPY COVER, and a strong positive rela-
tionship with SHRUB DENSITY. The secondary
axis had a strong negative relationship with
HERBACEOUS COVER. High severity burn
units were best characterized by habitat with less
CANOPY COVER, lower BASAL AREA, and
higher SHRUB DENSITY. Several species along
the primary axis were depicted farthest from the
centroid and included Yellow-breasted Chat,
Chestnut-sided Warbler, Indigo Bunting, and
Eastern Towhee (Fig. 4). Species such as Hooded,
Black-and-White, and Black-throated Green war-
blers were mapped highest on the vertical axis
(Fig. 4).

DISCUSSION

The effects of fire severity can vary between
species and across avian communities. Our results
indicated both species richness and diversity
increased relative to fire severity and time since
fire. Relative abundance did not change consid-
erably among treatments or with time since fire
(Table 1, Fig. 2); this may have been a bias from
our conservative analysis of repeat observations.
Species evenness also remained stable among
treatments.

The species most negatively affected by fire
was the Ovenbird, which is associated with closed

526

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 3, September 2010

VI

VI

O

o

VI

o

•

o

(D

Dh

CO

o

A

o

o

(L> -H

CLh
W

d

o

o

8

9

O

o

Vi

(D

Vi

(D

O

<D

SUh

CO

00

d

d

d

o

Contiol I

Low Medium

1 to 2 yeai's

Higli j Low Medium

3 to 6 yeai'S

Higli

Fire intensity/yeai's since fire

FIG. 2. Species richness, diversity, and evenness among avian communities within the southern Appalachian
Mountains. Surveys were conducted at sites which received one of four levels of fire severity <2 years since fire or 3-
6 years after fire.

KlciiL'i el al. • FIRE AND BIRDS IN THE SOUTHERN APPALACHIAN MOUNTAINS 527

TABLE 2. Values tor the first two axes in canonical correspondence analysis (CCA) and correlations with
environmental variables used to constrain the ordination. CCA was based on estimates of relative avian abundance and
habitat measured at control (no tire) and varying levels of tire severity treatments within 2 years and 3 to 6 years after fires
in the southern Appalachian Mountains.

1-2 years post fire

.2-6 years post fire

Stati.stic

CCA A.xis I

CCA Axis II

r

p

CCA Axis 1

CCA Axis 11

r

p

Eigenvalues

0.34

0.28

0.33

0.16

Cumulative variance

0.18

0.33

0.23

0.33

Intraset correlations

BASAL AREA

0.95

-0.30

0.23

<0.001

-1.00

-0.013

0.16

<0.001

SHRUB HEIGHT

0.59

-0.80

0.19

<0.001

0.23

0.97

0.00

0.94

SHRUB DENSITY

0.93

-0.37

0.05

0.02

0.89

-0.45

0.20

<0.001

HERBACEOUS COVER

-0.58

0.82

0.04

0.07

0.12

-0.99

0.37

0.063

CANOPY COVER

0.96

-0.28

0.25

<0.001

-0.99

0.11

0.04

<0.001

canopy forests with limited understory (Neimi and
Hanowski 1984, Smith and Shugart 1987).
Availability of this habitat may be reduced for
many years following a high severity fire, a

relationship governed by increased growth of
herbaceous cover and lack of canopy protection
(Elliott et al. 1999, Greenberg et al. 2007).
Conversely, our results indicated that many

CAI

FIG. 3. Habitat associations relative to fire severity for species (AMGO = American Goldfinch, etc.) detected on point-
count surveys 1-2 years after fires in northern Georgia. Solid line represents boundary of habitat centroids measured using
linear combination scores derived for control units (no fire, thin solid line), low severity (thick solid line), medium .severity
(thin dashed line), and high .severity burn units (thick dashed line). Boxed habitat metrics are mapped with arrows indicating
relative locations within the community. The longer the di.stance between the metric and the community center (axis point
0, 0) the stronger the relation.ship with the community. Proximity of individual .species to habitat metrics indicates strength
of association between relative species abundance and habitat metric. Only those species detected on >20% of surveys
are shown.

528

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 3, September 2010

CAI

FIG. 4. Habitat associations relative to fire severity for species (BAWW = Black and White Warbler, etc.) detected on
point-count surveys conducted 3-6 years after fires in northern Georgia. Community mapped using Canonical
Correspondence Analysis (CCA) with habitat centroids based on linear combination scores derived from habitat measured
at control units (thin solid line), low .severity (thick solid line), medium severity (thin dashed line), and high severity (thick
dashed line). Boxed habitat metrics are mapped with arrows indicating relative locations within the community. The longer
the di.stance between the metric and the community center (axis point 0, 0) the stronger the relationship with the
community. Proximity of individual species to habitat metrics indicates .strength of association between relative abundance
and habitat metric. Only those species detected on >20% of surveys are shown.

species associated with disturbance and early
successional habitat responded positively to fire
severity.

Species that responded positively to increa.sed
fire severity included American Goldfinch, Chest-
nut-sided Warbler, Eastern Towhee, Indigo Bun-
ting, Prairie Warbler, and Yellow-breasted Chat.
Most of these species had higher abundance 3-
6 years after fires, tracking decreased canopy
cover and released growth of shrubs relating to
increa.sed shrub density. Also included among
these species were Golden-winged and Cerulean
warblers, species that are both endangered in
Georgia. Both species were encountered once and
were found exclusively in either medium (Gold-
en-winged Warbler) or high severity (Cerulean
Warbler) treatments (Table 1 ).

The physiognomic properties of habitats con-
tinued 3-6 years after high severity fires.
Continued canopy openness related positively

with abundance of several woodpecker species;
Hairy Woodpeckers and Northern Flickers had
highest abundance in high severity treatments.
Species characteristic of forests with more-open
canopy, such as Yellow-billed Cuckoo, Eastern
Wood-Pewee, and Blue-gray Gnatcatchers in-
creased relative to fire severity, responses that
may relate to suppressed understory, more-open
conditions, higher insect abundance, and im-
proved foraging habitat (Greenberg et al. 2007).

Mid-story nesting species are thought to be
relatively flexible in habitat .selection and may not
respond to fire-driven habitat change (Artman et
al. 2005). Our results indicated density' of several
of these mid-story nesting species including
Wood Thrush, Blue-headed Vireo, and Red-eyed
Vireo did not differ relative to either fire severity
or time since fire. However, American Redstart, a
species that forages and nests in shrubs and vines,
appeared to be absent from treatment burn units

K/au.s et al. • FIRE AND BIRDS IN THE SOUTHERN ARPALACHIAN MOUNTAINS 529

during the first 2 years after fires. This association
may reflect the temporary reduction of vertical
structure during the first few years after a fire
(Waldrop et al. 2007).

Uneven-age forests may represent natural
historical conditions within the southern Appala-
chian Mountains (Lorimer 1980). Our results
indicated that returning a fire regime to these
forests may be an effective conservation tool for
restoring avian diversity (Artman et al. 2005,
Greenberg et al. 2007). However, effectiveness of
fire restoration can vary (Artman et al. 2005).
Continued fire suppression, or frequent applica-
tion of low severity fires, may lead to forest
maturation, have limited effects on species
diversity, and can lead to declines in some species
(Artman et al. 2005, Klaus et al. 2005). Applica-
tion of high severity fires corresponding with
short rotation periods may not allow habitat
regeneration and can lead to decreased diversity.
Application of higher severity fires over broader
time intervals may be more effective in managing
uneven-age forests.

Changes in relative abundance of a particular
species may not equate directly with habitat
quality or persistence (Johnson 2007). We did
not examine the effects of treatment on reproduc-
tive success, survival, or other demographic
measures and cannot provide predictions relating
treatments to these population metrics. Our results
reflect acute numerical relationships among
individual species and across avian communities.
Longer-term studies focused on addressing vari-
ation in population demographics through several
seasons after different fire severity treatments
would greatly benefit our understanding of avian
communities.

CONSERVATION IMPLICATIONS

Management directed at conservation of avian
species should balance restoration of early
successional habitat with conservation of mature
forest (Dettmers 2003, Artman et al. 2005,
Bulluck and Buehler 2006, Buehler et al. 2007).
Our results, and those of others (Artman et al.
2005, Greenberg et al. 2007), indicate conserva-
tion activities that foster avian diversity within the
southern Appalachian Mountains may benefit
from occasional high severity fires and a fire
return interval of 3 < 10 years. We are not
suggesting all burns should be of high severity,
only that the full range of fire severity be used
through time and space.

Fire may be most beneficial to avian conserva-
tion when it is used over relatively large spatial
.scales (Artman et al. 2005, Greenberg et al. 2007).
Managed fires will help develop a matrix of
habitats and successional stages, and minimize the
isolation of populations (Askins 2001, Brawn et
al. 2001). A rationale against pre.scription of
smaller-scale fires is that larger fires may be more
easily controlled in habitat characterized by
terrain similar to that in the Chattahoochee-
Oconee National Forest. Larger burn units can
make use of creeks and roads for fire breaks
helping avoid undue disturbance such as plowing,
which may be a vector for invasion of exotic
species (Memam et al. 2006). Our study did not
examine the cumulative impacts of repeated low
intensity burns, or combinations of burn intensi-
ties. However, our results suggest that medium
and high intensity fires can have positive effects
on bird communities and should be considered a
valuable land management tool.

ACKNOWLEDGMENTS

We acknowledge support of the D. B. Warnell School of
Forest Resources of the University of Georgia, the
Nongame Conservation Section of the Georgia Department
of Natural Resources, and the U.S. Forest Service. We
thank J. S. Hender.son, and J. M. Wentworth for helping
select study sites, D. G. Gregory and C. E. Hans for
assisting with bird survey work, and M. N. Hayes for
assistance with data entry.

LITERATURE CITED

Abrams, M. D. 1992. Fire and the development of oak
forests in eastern North America; oak distribution
reflects a variety of ecological paths and disturbance
conditions. Bioscience 42:346-353.

Anders, A. D., J. Faaborg, and F. R. Thompson 111. 1998.
Postfledging dispersal, habitat use, and home-range
size of juvenile Wood Thrushes. Auk 1 15:349-358.
Aquilani, S. M.. D. C. Leblanc, and T. E. Morrell.

2000. Effects of prescribed surface fires on ground-
and shrub-nesting neotropical migratory birds in a
mature Indiana oak forest, USA. Natural Areas Journal
20:317-324.

Artman, V. L. and J. F. Downhover. 2003. Wood Thrush
(Hylodchla miistelina) nesting ecology in relation to
prescribed burning of mixed-oak forest in Ohio. Auk
120:874-882.

Artman, V. L., E. K. Sutherland, and J. F. Downhover.

2001. Prescribed burning to restore mixed-oak com-
munities in .southern Ohio: effects on breeding bird
populations. Conservation Biology 15:1423-1434.

Artman, V. L., T. F. Hutchinson, and J. D. Brawn.
2005. Fire ecology and bird populations in eastern

530

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 3, September 2010

deciduous forests. Studies in Avian Biology 30:127-
138.

Askins, R. a. 2001. Sustaining biological biodiversity in
early successional communities: the challenge of
managing unpopular habitats. Wildlife Society Bulle-
tin 29:407-412.

Bond, W. J. and J. F. Keeley. 2005. Fire as a global
‘herbivore’: the ecology and evolution of flammable
ecosystems. Trends in Ecology and Evolution 20:387-
394.

Bond, W. J., F. I. Woodward, and G. F. Midgley. 2005.
The global distribution of ecosystems in a world
without fire. New Phytologist 165:525-538.

Brawn, J. D., S. K. Robinson, and F. R. Thompson III.
2001. The role of disturbance in the ecology and
conservation of birds. Annual Review of Ecology and
Systematics 32:251-276.

Brewster, W. 1886. An ornithological reconnaissance in
western North Carolina. Auk 3:94-112.

Buehler, D. a., a. M. Roth, R. Vallender, T. C. Will,

J. F. Confer, R. A. Canterbury, S. B. Swarthout,

K. V. Rosenberg, and L. P. Bulluck. 2007. Status
and conservation priorities of Golden-winged Warbler
(Vennivora chrysoptera) in North America. Auk
124:1439-1445.

Bulluck, L. P. and D. A. Buehler. 2006. Avian use
of early successional habitats: are regenerating
forests, utility right-of-ways and reclaimed surface
mines the same? Forest Ecology and Management
236:76-84.

Burleigh, T. D. 1958. Georgia birds. University of
Oklahoma Press, Norman, USA.

Dettmers, R. 2003. Status and conservation of shrubland
birds in the northeastern U.S. Eorest Ecology and
Management 185:81-93.

Elliott, K. L., R. L. Hendrick, A. E. Major, J. M. Vose,
AND W. T. Swank. 1999. Vegetation dynamics after a
prescribed fire in the southern Appalachians. Forest
Ecology and Management 114:199-213.

Greenberg, C. H., A. L. Tomcho, J. D. Lanham, T. A.
Waldrop, J. Tomcho, R. J. Phillips, and D. Simon.
2007. Short-term effects of fire and other fuel
reduction treatments on breeding birds in a southern
Appalachian upland hardwood forest. Journal of
Wildlife Management 71:1906-1916.

Hamel, P. B., W. P. Smith, D. J. Twedt, J. R. Woeher, E.
Morris, R. B. Hamilton, and R. J. Cooper. 1996. A
land manager’s guide to point counts in the Southea.st.
USDA, Forest Service, General Technical Report SO-
120. Southern Re.search Station, New Orleans, Louisi-
ana, USA.

Hunter, W. C., R. Katz, D. Pashley, and B. Ford. 1999.
Partners in flight bird conservation plan for the
southern Blue Ridge (Physiographic Area 23). USDl,
Fish and Wildlife Service, Atlanta, Georgia. USA.

HtiNTER, W. C. D. A. Buehler, R. A. Canterbury, J. L.
Confer, and P. B. Hamel. 2001. Conservation of
disturbance-dependent birds in ea.stern North America.
Wildlife Society Bulletin 29:440-455.

.loilN.SON, M. D. 2007. Measuring habitat quality: a review.
Condor 1 09:489-504.

Klaus, N. A. 2004. Status of the Golden-winged Warbler
in north Georgia, and a nesting record of the
Lawrence’s Warbler. Oriole 69:1-7.

Klaus, N. A., D. A. Buehler, and A. M. Saxton. 2005.
Forest management alternatives and songbird breeding
habitat on the Cherokee National Forest, Tennessee.
Journal of Wildlife Management 69:222-234.

Lang, J. D., L. A. Powell, D. G. Krementz, and M. J.
Conroy. 2002. Wood Thrush movements, habitat use,
and effects of forest management for Red-cockaded
Woodpeckers. Auk 119:109-124.

Lorimer, C. G. 1980. Age structure and disturbance history
of a southern Appalachian virgin forest. Ecology
61:1169-1184.

Lorimer, C. G. 2001. Historical and ecological roles of
disturbance in eastern North American forests:
9,000 years of change. Wildlife Society Bulletin
29:425-439.

Loucks, C. G. 1970. Evolution of diversity, efficiency, and
community stability. American Zoologist 10:17-25.

Magurran, a. E. 1988. Ecological diversity and its
measurement. Princeton University Press, Princeton,
New Jersey, USA.

Marshall. M. R., J. A. DeCecco, A. B. Williams, G. A.
Gale, and R. J. Cooper. 2003. Use of regenerating
clearcuts by late-successional bird species and their
young during the post-fledging period. Eorest Ecology
and Management 183:127-135.

McCune, B. and J. B, Grace. 2002. Analysis of ecological
communities. MJM Software Design. Glenedon
Beach, Oregon, USA.

Merriam, K. E., j. E. Keeley, and J. L. Beyers. 2006.
Euel breaks affect nonnative species abundance in
California plant communities. Ecological Applications
16:515-527.

Neimi, G. J. and j. M. Hanowski. 1984. Relationships of
breeding birds to habitat characteristics in logged
areas. Journal of Wildlife Management 48:438-443.

Partners in Plight Working Group. 2002. Priority
research needs for the conservation of neotropical
migrant landbirds. Journal of Field Ornithology
73:329-339.

PiNHEiRO, J. C. AND D. M. BATES. 2000. Mixed-effects
models in S and S-Plus. Springer-Verlag. New York.
USA.

Raudenbush. S. W. and a. S. Bryk. 2002. Hierarchical
linear models: applications and data analysis methods.
Sage Publications, Thousand Oaks, California, USA.

R Development Core Team. 2008. R: a language and
environment for statistical computing. R Foundation
for Statistical Computing, Vienna. Austria.

Rush, S. A. and B. J. M. Stutchbury. 2008. Survival of
fledgling Hooded Warblers (WHsonia citriua) in small
and large forest fragments. Auk 125:183-191.

Saab, V. A. and H. D. W. Powell. 2005. Fire and avjan
ecology in North America: process influencing pattern.
Studies in Avian Biology 30:1-13.

Sauer. J. R., J. E. Hines, and J. Fallon. 2005. The North
American Breeding Bird Survey, results and analysis
1966-2005. Version 6.2.2006. USGS, Patuxent Wild-
life Re.search Center, Laurel, Maryland, USA.

Klaus et al. • FIRE AND BIRDS IN THE SOUTHERN APPALACHIAN MOUNTAINS 53 I

Shannon, C. E. and W. Weaver. 1963. Tlic muihemaiical
theory of communication. University ol' Illinois Press,
Urbana, USA.

Smith, T. M. and H. H. Shugart. 1987. Territory size
variation in the Ovenbird: the role of habitat structure.
Ecology 68:695-704.

Stupka, A. 1963. Notes on the birds of the Great Smoky
Mountains during the breeding season. Migrant 17:60-
62.

Tomcho, a. L., C. H. Greenberg, J. D. Lanham, T. A.
Waldrop, J. Tomcho, and D. Simon. 2006. Effects of
fuel reduction treatments on breeding birds in a southern
Appalachian upland hardwood forest. USDA, Forest
Service. General Technical Report PSW-GTR-203. Pa-
cific Southwest Research Station, Portland, Oregon, USA.

V.AN Lear, D. H. and T. A. Waldrop. 1989. History, use,
and effects of fire in the southern Appalachians.

USDA, Forest Service, General Technical Report SE-
54. Southeastern Forest Experiment Station, Asheville,
North Carolina, USA.

Vega Rivera, J. H., J. H. Rappole, W. J. McShea, and C.
A. Haas. 1999. Postbreeding movements and habitat
use of adult Wood Thrushes in northern Virginia. Auk
1 16:458-466.

Waldrop, T. A., D. L. White, and S. M. Jones. 1992. Fire
regimes for pine-grassland communities in the south-
eastern United States. Forest Ecology and Manage-
ment 47:195-210.

Waldrop, T. A., D. A. Yaussey, R. J. Phillips, T. A.
Hutchinson, L. Brudnak, and R. E. J. Boerner.
2007. Fuel reduction treatments affect stand structure
of hardwood forests in western North Carolina and
southern Ohio, USA. Forest Ecology and Management
255:31 17-3129.

The Wilson Journal of Ornithology 122(3); 532-544, 2010

AVIAN COMMUNITIES OF THE ALTAMAHA RIVER ESTUARY

IN GEORGIA, USA

ROSS A. BRITTAIN, ' VICKY J. MERETSKY,' AND CHRIS B. CRAET'

ABSTRACT. — We surveyed male breeding birds in five habitats (bottomland forest, maritime oak [Quercus spp.], pine
[Finns spp.] forest, maritime shrub, saltmarsh) of coastal Georgia, USA using distance-sampling methods to estimate
population densities. We examined species-habitat relationships using indicator species analysis (ISA). Acadian Flycatcher
{Empidona.x virescens) in bottomland forest. Northern Parula (Paritla americana) in maritime oak. Brown-headed Nuthatch
(Sitta piisilla) in pine forest. Clapper Rail (Rallus longirostris) in saltmarsh, and White-eyed Vireo (Vireo griseiis) in shrub
habitat ranked highest for Partners in Flight (PIF) priority species by densities. The ISA indicated fewer PIF priority species
were associated with saltmarsh, but more species (6) were unique to saltmarsh than any other habitat. Indicator species
occurred more often in maritime oak than bottomland forest (8 vs. 6). but both habitats had similar numbers of PIF priority
species (4). Shrub habitat covered the smallest area (~0.2%) and had three PIF priority species, including Painted Bunting
(Passerina ciris), the only PIF species with extremely high priority in this study. Received 27 November 2009. Accepted 18
March 2010.

Indicator species appeal to conservationists,
land managers, and governmental agencies be-
cause they provide an efficient method of
monitoring the impacts of environmental distur-
bances and management policies (Carignan and
Villard 2002). Thus, indicator species analysis
(ISA), based on combining frequency of occur-
rence and mean abundance (Dufrene and Le-
gendre 1997), is gaining use in conservation
studies (Graham and Blake 2001, Kirk and
Hobson 2001, Morissette et al. 2002, Mouillot et
al. 2002, Grundel and Pavlovic 2007). The use of
indicator species can be helpful if it includes
monitoring many species that represent different
taxa and life histories, the chosen species
monitored have excellent quantitative regional
data, and habitat-related trends can be distin-
guished from natural variation (Carignan and
Villard 2002).

Once a suite of indicator species has been
cho.sen to represent a community, monitoring
activities need to efficiently and reliably estimate
appropriate population parameters that signal
potential impacts to conservation stakeholders
(Thomas 1996). Two relatively simple, yet
meaningful, avian abundance parameter estima-
tions for modeling population dynamics are
species density and size (Buckland et al. 2001).
Population size is the density times the area of
appropriate habitat, and research has focused on

' Indiana University. School of Public and Environmental
Affairs, Room 410, Bloomington, IN 47405, USA.

^Current addre.ss: 3475 Winchester Drive, Greenwood,
IN 46143, USA.

’Corre.sponding author: e-mail: rabritta@indiana.edu

improving estimates of population density. Other
important parameters, including population de-
mographics and survival, are more difficult to
assess because they require capturing individuals
(Nisbet 2001).

A common method used to estimate avian
abundance has been count data (Hodges and
Krementz 1996, Bajema et al. 2001 , Rosenstock et
al. 2002). The point-count method is relatively
simple to conduct but assumes bird detection
remains constant across different habitats, observ-
er abilities, weather conditions, and species
characteristics (Rosenstock et al. 2002). Dis-
tance-sampling methods yield more precise esti-
mates of bird density than index methods, such as
point counts, by adjusting for detection (Rosen-
stock et al. 2002, Thompson 2002). A 7-year
comparison of point-count methods versus dis-
tance sampling found the latter method was more
robust in large-scale, multispecies surveys (Nor-
vell et al. 2003).

Distance-sampling methods have been used in
many studies to assess quantitative differences in
habitat use by avian species to distinguish species-
habitat relationships (Hodges and Krementz 1996,
Estades and Temple 1999, Fletcher and Koford
2002), but the method does not work well with
rare species (Buckland et al. 2001). Conversely,
ISA is valuable for assessing .species-habitat
correlations when species are rare, non-normality
exists, where distance sampling may be logisti-
cally difficult, and where there may be no
detection of a species at many sample sites
(Mouillot et al. 2002).

Our objective was to examine avian communi-
ties and densities of male breeding bird species in

532

Brinahi cl cil. • COASTAL GEORGIA AVIAN COMMUNI TIES

533

FIG. 1. Study area and sample sites on Sapelo Island, McIntosh County, Georgia and Clayhole Swamp, Glynn
County, Georgia.

five habitats of coastal Georgia, USA (tidal-
freshwater broadleaf deciduous forest, saltmarsh,
maritime scrub-shrub, maritime broadleaf ever-
green forest, and maritime narrowleaf evergreen
forest).

METHODS

Study Area. — We surveyed broadleaf deciduous
bottomland forest affected by tides, saltmarsh,
maritime scrub-shrub, maritime broadleaf ever-
green forest, and maritime narrowleaf evergreen
forest in coastal Georgia, USA (Fig. 1) for male
breeding birds. We selected a broadleaf deciduous
forest affected by tides (bottomland forest) at the
Clayhole Swamp Wildlife Management Area
adjacent to the Altamaha River, Glynn County,
Georgia, managed by the Department of Natural
Resources (GDNR) (Duberstein and Kitchens
2007). We selected four other habitats on 6,677-
ha Sapelo Island, McIntosh County, Georgia,
managed by GDNR. We selected saltmarsh
habitat primarily on the southwestern portions of
the island and maritime scrub-shrub habitat
(shrub) based on linear landscape features, such
as secondary dunes, on the edge between forest

and saltmarsh (Johnson et al. 1974). Maritime
broadleaf evergreen forests (maritime oak [Quer-
cus spp.]) were dominated by live oak {Q.
virginiana), primarily on the north end of the
island (Johnson et al. 1974). Maritime narrowleaf
evergreen forests (pine [Pinus spp.] forest)
consisted of loblolly pine {P. taeda) with scattered
slash pine (P. elliottii) and were distributed
throughout the island (Johnson et al. 1974).
GDNR manages Sapelo Island’s pine forests for
sawtimber (old growth) using fire and timber
harvest.

We selected bird sample points at least 250 m
apart. We used a grid in bottomland forest stalling
from the southernmost accessible point via
Honeygal Road. Saltmarsh sample points on
Sapelo Island were randomly selected using a
l-m^ grid overlain on a Geographic Information
System (GIS) shapefile. Sample points in shrub
habitat were in the ecotone between primary
dunes and terrestrial forests on Sapelo Island. Oak
and pine forest sample points were selected by
driving one-lane sand roads and randomly locat-
ing them between 10 and 200 m on a line directly
into forest patches that were relatively consistent
within a 100-m radius.

534

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 3, September 2010

Bird Sampling. — We conducted point counts
following Ralph et al. (1993). We located
10 points in each habitat in 2006 (50 total) and
an additional 20 points per habitat in 2007
(total of 150 points). Ten-minute counts were
conducted twice at least 2 weeks apart at each
point within 4 hrs of sunrise between 19 May and
9 June 2006, and between 17 May and 13 June
2007. The relative direction and distance of each
detected bird from the sample point during each
sampling event, excluding flyovers, were estimat-
ed. Distances were verified in the field on the way
to the next point by pacing the distance from the
point just sampled to a detected bird. Detections
of the same species only included birds detected
simultaneously or with obvious differences in
plumage. Observer bias issues were limited by
having a single observer conduct all sampling
events.

Population Densities. — The density (# of birds/
km") of species sampled using point counts was
estimated in program DISTANCE 5.0 (Thomas et
al. 2006) and equation D = n/kv, where
D — density of birds per unit area, n = number
of birds detected, k = number of points sampled,
and V = the detection function of the species with
distance from the sample point following Buck-
land et al. (2001). The detection function indicates
the probability of detecting an individual as a
function of distance from the sample points for a
particular species (Buckland et al. 2001). The
density of each species is calculated by fitting
the frequency of detection as a function of
distance into each of 12 theoretical models using
the best fitting model (with the lowest Akaike’s
Information Criterion [AIC] value) (Buckland et
al. 2001).

The frequency distribution of distance to
detected (seen or heard) males for each species
was pooled across all habitats to increase the
sample size of detections. The cutoff for exclusion
from DISTANCE analysis was <50 observations
for a species across all habitats and <10
individuals of a species in a given habitat. We
tested for differences in detection between habitat
types using DISTANCE without pooling the
frequency distribution across habitats, but only
pooled results are presented due to lack of
differences. Densities were considered to be
different when there was a lack of overlap of the
95% confidence intervals (Gardner and Altman
1986, Hodges and Krementz 1996, Anderson et al.
2001, Fletcher and Koford 2002).

DISTANCE requires detection functions to
monotonically decrease with distance from the
sample point so the detection function is either
flat or decreases with distance (Thomas et al.
2006). DISTANCE constrains parameters of the
detection function at fixed points to fit the
function to the model, if the detection function
increases with distance. These constraints distort
the detection function which may be better
adjusted by truncating the tail of the data or
altering the intervals to smooth the detection
function (Thomas et al. 2006). We also conducted
the analyses by truncating detections at 80- and
90-m, but only the non-truncated results are
reported due to lack of differences. Models were
chosen based on the least number of constrained
parameters to obtain monotonicity and the lowest
AIC value.

Statistics. — Species-habitat associations were
assessed using indicator species analysis and
9999 Monte Carlo iterations (ISA; McCune and
Grace 2002). ISA results include the relative
abundance, relative frequency, highest indicator
value, and P values of significant differences.
Statistical results were considered significant at a
< 0.05. Dufrene and Legendre (1997) recommend
a cutoff threshold indicator value (IV) of 25 for a
species to be considered truly indicative of a
habitat.

RESULTS

Species Abundance. — We detected 56 species
during the two breeding seasons (scientific names
and Partners in Flight priority status are in
Table 1). We encountered Northern Cardinals
and Carolina Wrens more often than any other
species (401 and 400 encounters for 5 habitats,
respectively). Red-eyed Vireos were most abun-
dant in bottomland forest (142), Northern Parula
in maritime oak (180), Red-bellied Woodpecker
in pine forest (116), Clapper Rail in saltmarsh
(150), and Northern Cardinal in shrub (140)
(Table 2).

Bird Densities. — We estimated breeding male
densities for 26 species overall (Table 2). None
of the 14 species for which we estimated densities
in bottomland forest was unique to this habitat,
but Prothonotary Warbler and Barred Owl were
only detected in bottomland forest. Tufted Tit-
mouse densities were significantly higher in
bottomland forest than all other species (172

Brittain et al. • COASTAL GEORGIA AVIAN COMMUNITIES

535

TABLE 1. Species detected during point counts in five habitats in coastal Georgia

during 2006 and 2007.

Species

Scienufic name

AOU code

Great Blue Heron

Ardeci heraciias*

GBHE

Great Egret

A. alha^

GREG

Snowy Egret

Egretta thiiUA

SNEG

Yellow-crowned Night-Heron

Nyctanassa violacea^

YCNH

Red-shouldered Hawk

Buteo lineatiis

RSHA

Red-tailed Hawk

B. jamaicensis

RTHA

Clapper Rail

Rallus longirostris^^^

CLRA

Wilson’s Plover

C ha rci drills wilsonia^^^

WIPE

Willet

Tringa semipahnata^^

WILL

Whimbrel

Numeniiis phaeopus

WHIM

Ruddy Turnstone

Arenaria interpres

RUTU

Mourning Dove

Zenaida macroura

MODO

Common Ground-Dove

Columbina passerina^^

COGD

Yellow-billed Cuckoo

Coccyzus aniericanus^^

YBCU

Barred Owl

Strix varia

BADO

Ruby-throated Hummingbird

Archilochus colubris

RTHU

Red-headed Woodpecker

Melanerpes eni’throcephalus^^

RHWO

Red-bellied Woodpecker

M. carolinus

RBWO

Downy Woodpecker

Picoides pubescens

DOWO

Pileated Woodpecker

Dryocopus pileatus

PIWO

Eastern Wood-Pewee

Coni opus V ire ns*

EAWP

Acadian Flycatcher

Empidonax virescens*

ACFL

Great Crested Flycatcher

Myiarchus crinitus

GCFL

Eastern Kingbird

Tyrannus tyrannus

EAKI

White-eyed Vireo

Vireo griseus**

WEVI

Yellow-throated Vireo

V. flavifrons*

YTVI

Red-eyed Vireo

V. olivaceus

REVI

Blue Jay

Cyanocitta cristata

BLJA

American Crow

Corvus brachyrhynchos

AMCR

Fish Crow

C. ossifragus

FICR

Carolina Chickadee

Poecile carolinensis**

CACH

Tufted Titmouse

Baeolophus bicolor

TUTI

Brown-headed Nuthatch

Sitta pusilla***

BHNU

Carolina Wren

Thryothorus ludovicianus

CARW

Marsh Wren

Cistothorus palustris

MAWR

Blue-gray Gnatcatcher

Polioptila caerulea

BGGN

Eastern Bluebird

Sialia sialis

EABL

Northern Mockingbird

Mimus polyglottos

NOMO

Brown Thrasher

Toxostoma rufum

BRTH

Northern Parula

Parula americana***

NOPA

Yellow-throated Warbler

Dendroica dominica***

YTWA

Pine Warbler

D. pinus**

PIWA

Prothonotary Warbler

Protonotaria citred**

PROW

Common Yellowthroat

Geothlypis trichas

COYE

Hooded Warbler

Wilsonia citrina***

HOWA

Yellow-breasted Chat

Icteria virens*

YBCH

Eastern Towhee

Pipilo erythrophthabnus**

EATO

Seaside Sparrow

Ammodramus maritimus***

SESP

Summer Tanager

Piranga rubra*

SUTA

Northern Cardinal

Cardinalis cardinalis

NOCA

Blue Grosbeak

Passerina caerulea

BLGR

Painted Bunting

P. ciris****

PABU

Red-winged Blackbird

Agelaius phoeniceus

RWBL

Boat-tailed Grackle

Quiscalus major

BTGR

Brown-headed Cowbird

Molothrus ater

BHCO

Orchard Oriole

Icterus spurius**

OROR

^ Local interest species (Hunter et al. 2001).

Moderate priority species.

High priority species.

tttt

Extremely high priority species.

TABLE 2. Total detections (counts) and breeding male population densities (birds/km^) with 95% confidence interval (in parentheses) calculated using DISTANCE loi birds
breeding in five habitats of coastal Georgia.

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 3, September 2010

CN CN On IT)

o

o

NO

S'

XI

o

O

m

(N

1

—

r-

ON 1

1

1

1

1

00

o

1—1

00

—

9—1

m

m

00 in

o

'3*

r4

CN

CN

r-

NO

CN "Tf CN

ON m On — '

— -m’^'^’^NOONm

^ ON

cn r-

(N

sD

00

lo

00

ON

o

in

ON On ON

o

(N

u

.c

•5-s.

s

.—K

V

S

_

•rf

s.

ON

in

o

NO

^ Tj-

ON

r-

CN

'3-

in

m

r-

00

CN

in

00

1

cn

00

—

—

1

ON

1

1

— • 1

1

1

On

1

—

1

1

00

1

On

o

1 On

ON

O

CN

1

CN

00

ON

r-

CN

tJ- —

^—

s

—

—1

in

r-

CN

r-

' — '

00 —

m

m

ON

ON

00

ON

oo

ON

CN

m

r-

CN

o

o

ON

00

—

—

CN

—

NO

(N

m On nO ^ m

— NO m
Tt — NO

— fNlOOOfOOOOCNNOON

o<N — — — TfinsooN

X)

X'

r--

00

J

ON

.a

m

00

a

in

cT

o

CN

CN

o

i»H

m 00

r-

— H

CN

—

00

1

1

NO

CN

1

T

T

T

4

NO

1

T

m

cn

T

r-

CN

ON

in

o —

m

ro

cn

NO

CN

2

NO

•—

CN —

00

’3'

ON

CN

ON

o

ON

ON

m

CN

ON r-

o

r-*

—

m

rj-

00

ON

r-

CN —

CN

in

r-

msooor^oor^cNoo

00 (N m

OcnONOONOmO — On
Tfm — — — '

ra

a

00

3"

s

a

(N 5^

3-

m

— «

3"

—4

r-

NO

in NO

3"

— (

— —

CN

O

r--

1

NO

1

CN

1

—

1 1

CN

1

1

NO

4

1

OO

T

»n

m

CN

cn r-

On

o

m

in

in

00 CN

—

in

00

w ^ w

CN

CN

00

CN

in

On

CN

3-

NO

m

— '

—4

CN

CN

r-

r--

ON

(N o m fN

ONOOCN — rosO r^(N(NfNr^

r- — m »n(N — rt— -

On NO ON

(N 00 O

C W)

0/j m

uu

c

o

Lm

<u

X

£

Oi)

^ -2
o —
c

a:

•o

o

D

2

3

o

SZ

? a>
X 2

T3 ^

1) t/5

1)

>

o

9

(U "O
C (U c
O > 3

E Q J
H 3 S

o

o

o

3

u

•o

OJ

(/) w o

O O c/5 >" U Oi

■a —

OJ C

02 ^

p C 5

E <u T3 p 5 o

“ -p-t O H —

O

-O

ou

c

c

e

3

X

-o

2

o

I- 1.

o u
<v a;

CL Cl

S Cl

> ^ ^
> ^ ■R

U D. O.

-o-R ^

T3 O ^ U
2 ^ ^

-o

fU

^ OJ

o

x:

o

-3 ^ '

H u-

3 _

O ^

>N 2

X

§ U

*

o

o

^ w

E .2

1)

2 ^ -
3 qj c/5 '—

^^o;Su>“oace:Q2o2Qa.cu<Ouj:>>

'B < a

OU 1^ «

■E "o 2
^ ^-S
c V j
a o
B

*

o

(U

O ^

-A o
^ 3

(U —

0^ X

o

u.

U

3 O

•c U

3

3

<

<D

1)

■o

3

O

2

U

Cd

3

"o

S3

-o

<u

c3

X

3

z

-o

0
T3
3
d)
JZ

1

3

O

X U H X

<u

C3

3

a

U

TABLE 2. Continued.

Brittain ct al. • COASTAL GEORGIA AVIAN COM M UN I'Ll ES

537

U

U

in

r-*

(N

OS

£

£>

o

x>

S— N

00

■5-v

OS

o

r--

<N

o

00

in

o

rsi

in

r-1

1

in

■rt

1

4

1

o

1

4

(N

in

1

1

in

7

1

in

OS

*—

00

•— <

so

in

'w'

' — '

o

r-

00

r-

o

o

o

00

so

r-

00

r^i

•— «

m

so

so

in

Tt

m

in

O (N

r--

— ■

—

m

00

Tt —

so

—

(N

os

o

O'

so

00 SO
00

in

n

SO

c>0

OS

5-v

a

in

CN

O

O

r- —

o

1

Os

— H

1

oo

m 1

CN

00

m

m

1

04

1

OS

CN

O

1

00

4 2:

1

SO

1

00

c-\

00

m

(N

CN

>r

so

r--

m

in

o m

r-.

00

Os

(N

in

m m

C^l

m

O m

cN in m

00

O'.

o —

so — so

r-

CN

in — '

so

Os

00

CN —

CN

in

I

m

oi

so

r-*

O'!

I

(N

<N

O „

Tt

— in

-I ^

O so

OS m

— lo

r)

ri

O — Os
00 (n

in

Os

I

r-

oo

Os

m o

T 7

00 Os
CN 00

»n Os
m Os

O CN

m 00

os

7

in

m

I

00

<N

in

7

os

I

00 so
(N r-'

o

ri

in

so

m

r-J

00

m

m in
m oo

m oo
m so

o ^

^ .b ^ oj _

2 _Cj U -C 3

O Sljl

^ E

r3

O

Xi x:

U. w

^ o

^ 2

>» no

U

o "O o

(U o

^ • - -

oij u 2^ c

i a -s ? -5

= ^ o £ c

CQ ttJ Z 03 Z

c3

O

O

c

o

x:

o -a
E ^

^ "O

E o
o o

Z H

^ E

o <u

_ r3

a, U X >- LU

^ <u

5 Of)

§ ^
o- f2
oo ^

0) o

E

CO fc
O 3
OO 00

*

•o

u.

15

* (J
bO ^

S CQ

o O

j=

C ^
o ^
Z CQ

c

3

X

■o

o

<o ^
OX) o
c :3

7 7

O

U

•o

•3

C3

<U

x:

c

o

•o Cb

o o

QC: CQ CQ

ra

x:

o

Species whose density models had parameters constrained to obtain monotonicity.

Groups of population densities with overlapping confidence intervals for each species among habitats.

538

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 3. September 2010

TABLE 3. Model type, goodness of fit (GOF), T-value of GOF, coefficient of variation (CV) and degrees of freedom
(df) for selected DISTANCE models, “hn” = half-normal key function, “hr” = hazard rate key function, “ne” =
negative exponential key function, and “un” = uniform key function, “cos” = cosine series expansion, “sp” = simple
polynomial series expansion, and “hp” = hermite polynomial series expansion.

Habitat Species

Model

GOF D„

P-valiie

CV

df

Maritime oak Red-bellied Woodpecker

hr-i-cos

0.1448

0.0001

6.70

307.92

Downy Woodpecker

ne-bhp

0.1685

0.0190

38.08

80.18

Pileated Woodpecker

hr-i-cos

0.1282

0.2284

23.49

34.28

Eastern Wood-Pewee

hr-t-cos

0.1437

0.0764

6.37

82.75

Acadian Flycatcher

hn-l-sp

0.1107

0.1318

17.91

70.84

Great Crested Flycatcher

hr-f-cos

0.1288

0.0048

7.81

200.44

White-eyed Vireo

hr-t-cos

0.0664

0.0908

6.46

256.53

Yellow-throated Vireo

hr-i-cos

0.1691

0.0726

19.33

67.18

Red-eyed Vireo

hr-bcos

0.1136

0.0268

24.47

31.77

Carolina Chickadee

hr-t-cos

0.0798

0.3429

18.79

140.03

Tufted Titmouse

hn-bsp

0.1186

0.0059

12.49

71.31

Brown-headed Nuthatch

hn-bcos

0.1210

0.1762

9.31

84.68

Carolina Wren

hr-bsp

0.0817

0.0104

5.36

295.43

Blue-gray Gnatcatcher

ne-bsp

0.0967

0.0816

20.35

167.47

Northern Parula

un-bcos

0.1504

0.0001

3.78

33.95

Yellow-throated Warbler

hr-bcos

0.0729

0.1975

8.28

244.66

Pine Warbler

hn+sp

0.1438

0.0027

9.10

172.12

Summer Tanager

hr-bsp

0.0589

0.8676

11.54

107.99

Northern Cardinal

hr-bsp

0.0694

0.0444

5.26

375.20

Painted Bunting

hn-bsp

0.161 1

0.0124

13.1 1

101.90

Brown-headed Cowbird

hn-bsp

0.1543

0.1937

34.54

43.19

Pine forest Red-bellied Woodpecker

hr-bcos

0.1448

0.0001

6.81

305.86

Downy Woodpecker

ne-bhp

0.1685

0.0190

38.10

80.34

Pileated Woodpecker

hr-bcos

0.1282

0.2284

29.95

32.12

Eastern Wood-Pewee

hr-bcos

0.1437

0.0764

6.43

85.46

Acadian Flycatcher

hn-bsp

0.1107

0.1318

28.37

40.39

Great Crested Flycatcher

hr-bcos

0.1288

0.0048

7.73

195.61

White-eyed Vireo

hr-bcos

0.0664

0.0908

6.15

324.93

Yellow-throated Vireo

hr-bcos

0.1691

0.0726

25.23

48.09

Red-eyed Vireo

hr-bcos

0.1136

0.0268

33.62

30.42

Carolina Chickadee

hr-bcos

0.0798

0.3429

18.72

137.96

Tufted Titmouse

hn-bsp

0.1186

0.0059

16.54

46.64

Brown-headed Nuthatch

hn-bcos

0.1210

0.1762

9.34

85.73

Carolina Wren

hr-bsp

0.0817

0.0104

5.21

343.74

Blue-gray Gnatcatcher

ne-bsp

0.0967

0.0816

20.36

167.50

Northern Parula

un-bcos

0.1504

0.0001

2.59

41.23

Yellow-throated Warbler

hr-bcos

0.0729

0.1975

8.18

242.31

Pine Warbler

hn-bsp

0.1438

0.0027

9.20

177.27

Eastern Towhee

hn-bsp

0.1553

0.0038

18.59

32.23

Summer Tanager

hr-bsp

0.0589

0.8676

1 1.43

104.20

Northern Cardinal

hr-bsp

0.0694

0.0444

5.54

290.03

Painted Bunting

hn-bsp

0.161 1

0.0124

13.15

102.91

Brown-headed Cowbird

hn-bsp

0.1543

0.1937

26.72

55.86

Saltmarsh Clapper Rail

ne-bsp

0.1060

0.0704

26.39

123.74

Marsh Wren

hn-bsp

0.1050

0.3962

33.63

46.69

Red-winged Blackbird

ne-bsp

0.1285

0.0040

29.21

195.84

Boat-tailed Grackle

hn-bsp

0.1874

0.0010

23.32

62.01

Brittain ct al. • COASTAL GEORGIA AVIAN COMMUNITIES

539

TABLE ?i. Continued.

Habitat Species

Model

GOF n„

P-value

cv

df

Shrub Red-bellied Woodpecker

hr-l-cos

0.1448

0.0001

6.76

308.85

Downy Woodpecker

ne-i-hp

0.1685

0.0190

38.09

80.23

Great Crested Flycatcher

hr+cos

0.1288

0.0048

7.89

204.97

White-eyed Vireo

hr-l-cos

0.0664

0.0908

6.26

305.61

Carolina Chickadee

hr-t-cos

0.0798

0.3429

18.82

140.79

Carolina Wren

hr-l-sp

0.0817

0.0104

5.02

403.37

Blue-gray Gnatcatcher

ne-l-sp

0.0967

0.0816

20.37

168.12

Northern Parula

un-i-cos

0,1504

0.0001

1.98

56.49

Yellow-throated Warbler

hr-l-cos

0.0729

0.1975

8.08

238.23

Pine Warbler

hnH-sp

0.1438

0.0027

9.13

174.24

Eastern Towhee

hn-fsp

0.1553

0.0038

13.24

37.09

Summer Tanager

hr-fsp

0.0589

0.8676

11.44

104.52

Northern Cardinal

hr+sp

0.0694

0.0444

5.61

274.44

Painted Bunting

hn-l-sp

0.161 1

0,0124

13.15

102.81

Red-winged Blackbird

ne-l-sp

0.1285

0.0040

34.25

119.98

Boat-tailed Grackle

hn-l-sp

0.1874

0.0010

67.34

32.58

Brown-headed Cowbird

hn-Hsp

0.1543

0.1937

27.90

54.75

Bottomland forest Red-bellied Woodpecker

hr-fcos

0.1448

0.0001

6.73

307.94

Downy Woodpecker

ne-fhp

0.1685

0.0190

38.08

80.20

Pileated Woodpecker

hr-t-cos

0.1282

0.2284

20.35

36.29

Acadian Flycatcher

hn-l-sp

0.1107

0.1318

15.26

100.71

Great Crested Flycatcher

hr-l-cos

0.1288

0.0048

7.68

192.55

White-eyed Vireo

hr-rcos

0.0664

0.0908

6.15

325.09

Red-eyed Vireo

hr-t-cos

0.1136

0.0268

8.33

71.85

Carolina Chickadee

hr-l-cos

0.0798

0.3429

19.16

149.10

Tufted Titmouse

hn-l-sp

0.1186

0.0059

11.09

95.60

Carolina Wren

hr-l-sp

0.0817

0.0104

5.15

363.55

Blue-gray Gnatcatcher

ne-l-sp

0.0967

0.0816

20.38

168.33

Northern Parula

un-l-cos

0.1504

0.0001

2.14

49.57

Summer Tanager

hr-l-sp

0.0589

0.8676

11.48

105.93

Northern Cardinal

hr-l-sp

0.0694

0.0444

5.38

339.58

males/km^). Red-eyed Vireo densities were sig-
nificantly higher in bottomland forest than any
other habitat. We estimated breeding male
densities for four species in saltmarsh, but seven
species were unique to saltmarsh (Table 2). Red-
winged Blackbirds had significantly higher den-
sities in saltmarsh than all other habitats, and had
the highest density of any saltmarsh species (125
males/km^).

We estimated densities of breeding males for
17 species in shrub habitat (Table 2). None of
these species was unique to this habitat, but
Common Ground-Doves were only detected in
shrublands. Eastern Towhee, Northern Cardinal,
and Painted Bunting males had significantly
higher densities in shrub than other habitats. We
estimated male densities for 21 species in
maritime oak and 22 species in pine forest, but
none of these species was unique to either habitat,
although Eastern Bluebirds were only detected in

pine forest. Northern Parula and White-eyed
Vireo had significantly higher densities in mari-
time oak than other habitats. Brown-headed
Nuthatch and Pine Warbler had significantly
higher densities in pine forest than other
habitats. The highest estimated density in shrub,
maritime oak, and pine forest was for Blue-gray
Gnatcatcher (319, 346, and 346 male.s/km%
respectively).

Most DISTANCE models used either the
half-normal key function with a simple poly-
nomial series expansion (31%) or the hazard-
rate key function with a cosine series expansion
(35%) (Table 3). The P-values for the Kolmo-
gorov-Smirnov Goodness of Eit tests of each
model were <0.10 for 17 of the 26 species
(65%).

Indicator Species Analysis. — We calculated
that 28 species were indicators of one of the
five habitats in coastal Georgia using Dufrene

540

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3, September 2010

90
80 ^
70

0)

O

w 60

0)

_D

CD

■D

C

50

o

S 30

20

REVI

HOWA

YBCU

YTWA-~

NOPA*-

BHNU*

PIWA*.

CLRA* J

RWBiy

BTGR/

- PROw\
ACFL

TUT,^

WEVl*

GCFbA

EAWP\\

CARWX^

YTVI

EGGisC;^

RBWQ^

EABL*\

RHWa\

MAWR^

EAKI

\

PABU*'^

EATO*-^

YBCH^^

NOCA*._^

IV=25 cutoff

- PiwO'^;^
SUTA^

RSHA «

■ BADO*;^
YCNH-""^

CACH^^

MODCU^
COYE^
BLJA ^
BRTH^^

WHIM*,

GREG*\

SESP*-^

WILL—

RUTUY/

WIPL'^

BHCO — ^

OROR-^

BLGR^(^

COGD*^

Bottomland

forest

Maritime

oak

Pine

forest

Saltmarsh

Shrub

FIG. 2. Indicator Value (IV) scores for species in each of five habitats in coastal Georgia, USA. Also shown is the
indicator cutoff value (25) of Dufrene and Legendre (1997). * = species unique to the habitat or that have higher densities
in the habitat in distance sampling.

and Legendre’s (1997) IV > 25 cutoff; an
additional 21 species were indicators accord-
ing to the ISA but had IV scores < 25 (Fig. 2).
Six species were indicative of bottomland
forest and five species were indicative of
saltmarsh. Shrub habitat had the fewest indi-
cator species (4), but the species with the highest
IV score in shrub habitat was the only extremely
high priority species in the region, Painted
Bunting. Eight species were indicators of mari-
time oak and five were indicators of pine
forest.

DISCUSSION

Implications for Conservation. — We estimated
a proportionately high number of PIF priority
species were indicative of bottomland forest (4),
which occupies —10.0% of the coastal Georgia
landscape (Brittain 2009). Prothonotary Warblers
occurred in bottomland forest sites with high
forest cover and basal area along tidal channels
(Petit 1999). Similarly, Acadian Flycatchers
and Yellow-billed Cuckoos were indicative of
wetland forests typical of their southeastern
habitats (Stevenson and Anderson 1994, White-
head and Taylor 2002); Acadian Flycatchers
were afso abundant in maritime oak. Yellow-
throated Warblers may have been deterred from

occupying bottomland forests, despite another
study that observed them in that habitat type
(Hall 1996), by the absence of large dead trees
(Hamel 1992).

Georgia saltmarshes had the lowest number of
priority indicator species among the habitats
studied (1), but the highest diversity of species
unique to the habitat. Seaside SpaiTows occurred
exclusively in tall smooth cordgrass {Spartina
alterniflora) stands on the banks of larger
estuaries with extensive mudflats (Post and
Greenlaw 1994). Clapper Rails were detected in
tall Spartina of the low marsh and the shrubby
vegetation in the high marsh-upland ecotone
(Eddleman and Conway 1998). Tall Spartina
may be important as habitat for Seaside Span'ows
and other saltmarsh birds, such as Marsh Wrens,
by creating areas of dense vegetation for secure
nests.

Eastern Towhees and Yellow-breasted Chats in
coastal Georgia, both PIF priority species, were
more abundant in shrub habitat than in open forest
(Greenlaw 1996, Eckerle and Thompson 2001).
Common Ground-Doves were found only in
coastal dunes despite known use of open pine,
hummocks, and forest edges (Bowman 2002)
common on Sapelo Island, likely due to the lack
of bare soil in forest habitats (Hamel 1992). The
PIF Conservation Plan defines quality shrub

Bhitain et al. • COASTAL GEORGIA AVIAN COMMUNITIES

541

habitat as “largely forested areas with some edge
and forest openings for buntings” (Hunter et al.
2001). However, shrub habitat, where Painted
Buntings, Eastern Towhees, and Yellow-breasted
Chats were far more common in this and another
study (Springborn and Meyers 2005), was largely
composed of coastal dunes. Maritime scrub-shrub
habitat deserves conservation priority over other
habitats as it occupies only —0.2% of the coastal
Georgia landscape (Brittain 2009) and is impor-
tant for the area’s only species of extremely high
conservation priority.

Northern Parula occupied mature hardwood
forests with Spanish moss {Tillandsia usneoides)
(maritime oak type) and floodplains with blue
palm (Brohea annata) (bottomland forest type),
as reported by Moldenhauer and Regelski (1996),
but were far more abundant in maritime oak.
Similarly, Eastern Wood-Pewee and Yellow-
throated Vireo occuined in all three forest types,
but were more common in maritime oak (McCarty
1996, Rodewald and James 1996). The ISA also
revealed White-eyed Vireos were indicators of
maritime oak rather than the shrub habitat for
which they are listed by PIP (Hunter et al. 2001).

A PIE Conservation Plan goal in pine forest
includes emphasis on late successional stands
with increasing disturbance regimes to enhance
understory habitat quality (Hunter et al. 2001).
Managed disturbances, such as prescribed fire and
logging on Sapelo Island, maintain the open
stands of mature pine habitat preferred by
Brown-headed Nuthatches (Withgott and Smith
1998). However, this conservation plan may aid
Brown-headed Nuthatches and Red-headed
Woodpeckers but may harm Pine Warbler popu-
lations dependent on a well-developed shrub layer
(Rodewald et al. 1999, Smith et al. 2000). Pine
forest occupies nearly half of the coastal Georgia
landscape (Brittain 2009), and there may be
opportunity for diverse strategies to protect
associated species.

Bird Densities. — Assumptions in distance sam-
pling include complete detection of all individuals
close to the sample point and that birds do not
leave the sampling area prior to detection (Buck-
land et al. 1993, Thompson 2002). The detection
rates in our study generally increased within the
first 5-10 m of the sample point before declining,
indicating birds may have moved from the sample
point as the observer approached, or the area of
near-perfect detection extended ~5 m beyond the
sample point itself.

DISTANCE could not find acceptable detection
functions without constraining parameters to
obtain monotonicity for 10 species (38%).
Constraining the parameters may have distorted
the detection functions and skewed the results.
However, alternative methods, such as truncation,
did not resolve the constrained parameter issues
and yielded results that were not significantly
different than those calculated using the original
data. The lack of good fit for 65% of the detection
functions increased concerns about the model
results, but truncation and grouping the data
into intervals made the fit worse, as reflected
by lower P-values in the Kolmogorov-Smirnov
tests, and did not significantly change the
results.

Eorest structure may hamper detection of
silent and sedentary individuals, whereas open
habitats, such as saltmarsh, should allow for near
complete detection (Pacifici et al. 2008). Thus,
breeding male density estimates in forested
habitats were likely underestimated. Detection
variations among different habitats should be
considered (Simons et al. 2007) by limiting
the pooling of data to create the detection
functions. However, we believed increased
sample size outweighed detection differences
between habitats. We tested each species
without pooling the detection frequency data
and the results showed no significant differences
between pooled and non-pooled data for most
species (not presented). The 95% confidence
intervals for each of the species were significantly
lower in the non-pooled data than pooled data
for White-eyed Vireo (maritime oak and pine
forest). Blue-gray Gnatcatcher (shrub), and Yel-
low-throated Warbler (shrub). Yellow-throated
Warblers also had a significantly higher density
estimate for maritime oak using the non-
pooled data. However, all models had to con-
strain parameters to obtain monotonicity using
non-pooled frequency data, and had lower
Goodness of Eit P-values, indicating these
models were less reliable than those using the
pooled data. The lack of improved models and
differences in density estimates between results
using pooled versus non-pooled frequency data
led us to conclude the pooled data were safer to
use.

Some breeding male densities appear high,
particularly Blue-gray Gnatcatchers, Brown-head-
ed Nuthatches in pine forest. Northern Parula in
maritime oak, and White-eyed Vireos on Sapelo

542

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 3. September 2010

Island. Male Brown-headed Nuthatches, found
almost exclusively in pine forests, were often
observed traveling in family groups that may have
skewed results by being more easily detected in
family clusters than individually. Northern Par-
ulas were definitively more abundant in maritime
oak and the number of males captured during bird
banding attempts was similar to the number of
detected males (Brittain 2009). White-eyed Vireos
and Blue-gray Gnatcatchers appeared curious
about the observer’s presence at the sample point,
which may have positively biased estimated
densities.

Our measurements of densities of six species
(Acadian Flycatcher, Blue-gray Gnatcatcher,
Northern Parula, Prothonotary Warbler, Red-eyed
Vireo, White-eyed Vireo) were greater than
those reported by Hodges and Krementz (1996).
These differences may be the result of variations
in study design or individual bird detection skills.
Northern Parula densities in upstream riparian
corridors were similar to densities in shrub
habitat in this study (Hodges and Krementz
1996).

Distance sampling indicated the PIF priority
species with the highest densities were; Acadian
Flycatcher in bottomland forest. Northern
Parula in maritime oak. Brown-headed Nuthatch
in pine forest. Clapper Rail in saltmarsh, and
White-eyed Vireo in shrub habitat. Shrub
habitat should have the highest conservation
value since it had the highest proportion of
indicator species that were also PIF priority
species (75%), the only extremely high priority
indicator species (Painted Bunting), and occupied
a much smaller portion of the coastal land-
scape (—0.2%) than other habitats. Maritime
oak, bottomland forests, and saltmarsh cover
moderate portions of coastal Georgia, but
forest habitats had more PIF priority indicator
.species than saltmarsh (5, 4, and 1, respectively).
Protections given tidal wetlands under Section
404 of the Clean Water Act (e.g., bottomland
forest and saltmarsh; PL 92-500, 33 USC
1251), combined with the high percentage of
priority species, suggests avian conservation
efforts should focus on conserving maritime
oak habitat. Pine forest is relatively plentiful
and was equal to shrub for the least number of
PIF priority indicator species in upland
habitat.

ACKNOWLEDGMENTS

We thank the National Oceanic and Atmospheric
Administration (NOAA) and the National Estuarine
Research Reserve System (NERRS) for funding this
research through a Graduate Research Fellowship; the
U.S. Environmental Protection Agency for providing funds
for research in the Clayhole Swamp; the Georgia
Department of Natural Resources, Wildlife Resource
Division, that manages the Sapelo Island NERR; and
Buddy Sullivan, Dorset Hurley, Aimee Gaddis, Fred Hay,
and Jon Garbisch.

LITERATURE CITED

Anderson, D. R., W. A. Link, D. H. Johnson, and K. P.
Burnham. 2001. Suggestions for presenting the results
of data analyses. Journal of Wildlife Management
65:373-378.

Bajema, R. a., T. L. DeVault, P. E. Scott, and S. L.
Lima. 2001. Reclaimed coal mine grasslands and their
significance for Henslow’s Sparrows in the American
Midwest. Auk 118:422^31.

Bowman, R. 2002. Common Ground-Dove (Columbina
passerina). The birds of North America. Number 645.

Brittain, R. A. 2009. Trophic status, habitat use and
climate change impacts on avian species of coastal
Georgia. Dissertation. Indiana University, Blooming-
ton, USA.

Buckland, S. T., D. R. Anderson, K. P. Burnham, and

J. L. Laake. 1993. Distance sampling: estimating
abundance of biological populations. Chapman and
Hall, London, United Kingdom.

Buckland, S. T., D. R. Anderson, K. P. Burnham, J. L.
Laake. D. L. Borchers, and L. Thomas. 2001.
Introduction to distance sampling: estimating abun-
dance of biological populations. Oxford University
Press, New York, USA.

Carignan, V. AND M. A. ViLLARD. 2002. Selecting
indicator species to monitor ecological integrity: a
review. Environmental Monitoring and Assessment
78:45-61.

Duberstein, J. AND W. KITCHENS. 2007. Community
composition of select areas of tidal freshwater forest
along the Savannah River. Pages 321-348 in Ecology
of tidal freshwater forested wetlands of the southeast-
ern United States (W. H. Conner, T. W. Doyle, and

K. W. Krauss, Editors). Springer-Verlag, New York,
USA.

Dufrene, M. and P. Legendre. 1997. Species assemblages
and indicator species: the need for a flexible
asymmetrical approach. Ecological Monographs
67:345-366.

Eckerle, K. P. and C. F. THOMP.SON. 2001. Yellow-
breasted Chat (Icteria virens). The birds of North
America. Number 575.

Eddleman, W. R. AND C. J. Conway. 1998. Clapper Rail
{Rcillus longirnstris). The birds of North America.
Number 340.

Britniiu et al. • COASTAL GEORGIA AVIAN COMMUNITIES

543

Estades, C. F. and S. a. Temple. 1999. Dcciduous-foresl
bird conimunitie.s in a fragmented land.scape dominat-
ed by exotic pine plantations. Ecological Applications
9:573-585.

Fletcher, R. J. and R. R. Koford. 2002. Habitat and
landscape associations of breeding birds in native and
restored grasslands. Journal of Wildlife Management
66:101 1-1022.

Gardner, M. J. a.nd D. G. Altman. 1986. Confidence
intervals rather than P values: estimation rather than
hypothesis testing. British Medical Journal 292:746-
750.

Graham, C. H. and J. G. Blake. 2001. Influence of patch-
and landscape-level factors on bird assemblages in a
fragmented tropical landscape. Ecological Applica-
tions 1 1: 1709-1721.

Greenlaw, J. S. 1996. Eastern Towhee (Pipilo ery-
throphthalmus). The birds of North America.
Number 262.

Grundel, R. and N. B. Pavlovic. 2007. Distinctiveness,
use, and value of midwestern oak savannas and
woodlands as avian habitats. Auk 124:969-
985.

Hall, G. A. 1996. Yellow-throated Warbler (Dendroica
dominica). The birds of North America. Number
223

Hamel, P. B. 1992. Land manager’s guide to birds of the
south. The Nature Conservancy, Southeast Region,
Chapel Hill, North Carolina, USA.

Hodges, M. F. and D. G. Krementz. 1996.
Neotropical migratory breeding bird communities
in riparian forests of different widths along the
Altamaha River, Georgia. Wilson Bulletin 108:496-
506.

Hunter, W. C., L. H. Peoples, and J. A. Collazo.
2001. Partners in flight bird conservation plan for
the south Atlantic Coastal Plain (physiographic area
#03). American Bird Conservancy, The Plains,
Virginia, USA.

Johnson, A. S., H. O. Hillestad, S. F. Shanholtzer,
and G. F. Shanholtzer. 1974. An ecological
survey of the coastal region of Georgia. USDI,
National Park Service Scientific Monograph
Number 3.

Kirk, D. A. and K. A. Hobson. 2001. Bird-habitat
relationships in jack pine boreal forest. Forest Ecology
and Management 147:217-243.

McCarty, J. P. 1996. Eastern Wood-Pewee (Contopus
Virens). The birds of North America. Number
245.

McCune, B. and j. B. Grace. 2002. Analysis of ecological
communities. MJM Software Design, Gleneden Beach,
Oregon, USA.

Moldenhauer, R. R. and D. j. Regelski. 1996. Northern
Parula (Parula americana). The birds of North
America. Number 215.

Morissette, j. L., T. P. Cobb, R. M. Brigham, and
P. C. James. 2002. The response of boreal forest
songbird communities to fire and post-fire harvest-
ing. Canadian Journal of Forest Re.search 32:2169-
2183.

Mouillot, D., j. M, Culioli, and T. D. Chi. 2002.
Indicator species analysis as a test ol non-random
distribution of species in the context of marine
protected areas. Environmental Conservation 29:385-
390.

Nisbet, I. C. T. 2001. Detecting and measuring senescence
in wild birds: experience with long-lived seabirds.
Experimental Gerontology 36:833-843.

Norvell, R. E., E. P. Howe, and J. R. Parrish. 2003.
A seven-year comparison of relative-abundance
and distance-sampling methods. Auk 120:1013-
1028.

PaCIFICI, K., T. R. StMONS, AND K. H. POLLOCK. 2008.
Effects of vegetation and background noise on the
detection process in auditory avian point-count
surveys. Auk 125:600-607.

Petit, L. J. 1999. Prothonotary Warbler (Protonotaria
citreci). The birds of North America. Number
408.

Post, W. and J. S. Greenlaw. 1994. Seaside Sparrow
(Ammodramus maritimus). The birds of North Amer-
ica. Number 127.

Ralph, C. J., G. R. Geupel, P. Pyle, T. E. Martin, and
D. F. DeSante. 1993. Handbook of field methods
for monitoring landbirds. USDA, Forest Service,
General Technical Report PSW-GTR-144. Pacific
Southwest Research Station, Albany, California,
USA.

Rodewald, P. G. and R. D. James. 1996. Yellow-throated
Vireo {Vireo flavifrons). The birds of North America.
Number 247.

Rodewald, P. G., J. H. Withgott, and K. G. Smith. 1999.
Pine Warbler {Dendroica pinus). The birds of North
America. Number 438.

Rosenstock, S. S., D. R. Anderson, K. M. Giesen, T.
Leukering, and M. F. Carter. 2002. Landbird
counting techniques: current practices and an alterna-
tive. Auk 1 19:46-53.

Simons, T. R., M. W. Alldredge, K. H. Pollock, and J.
M. Wettroth. 2007. Experimental analysis of the
auditory detection process on avian point counts. Auk
124:986-999.

Smith, K. G., J. H. Withgott, and P. G. Rodewald.
2000. Red-headed Woodpecker (Melanerpes eiyrhro-
cephalus). The birds of North America. Number
518.

Springborn, E. G. and J. M. Meyers. 2005. Home range
and survival of breeding Painted Buntings on Sapelo
Island, Georgia. Wildlife Society Bulletin 33:1432-
1439.

Stevenson, H. M. and B. H. Anderson. 1994. The birdlife
of Florida. University Press of Florida. Gainesville,
USA.

Thomas, L. 1996. Monitoring long-term population
change: why are there so many analysis methods?
Ecology 77:49-58.

Thomas, L., J. L. Laake, S. Strindberg, E. E. C.
Marques, S. T. Bucki.and. D. L. Borchers,
D. R. ANDER.SON, K. P. Burnham. S. L. Hedley,
J. H. Pollard, J. R. B. Bishop, and T. A.
Marques. 2006. DISTANCE 5.0 Release 2.

544

THE WILSON JOURNAL OL ORNITHOLOGY • Voi 122, No. 3, September 2010

Research Utrit for Wildlife Population Assess-
ment, University of St. Andrews, United
Kingdom.

Thompson, W. L. 2002. Towards reliable bird surveys:
accounting for individuals present but not detected.
Auk 119:18-25.

Whitehead, D. R. and T. Taylor. 2002. Acadian
Flycatcher (Empidonax virescens). The birds of North
America. Number 614.

WiTHGOTT, J. H. AND K. G. SMITH. 1998. Brown-headed
Nuthatch {Sitta pusilla). The birds of North America.
Number 349.

The Wilson Jounml oj Ornithology 122(3):545-555, 2010

INFLUENCE OF COVER AND EOOD RESOURCE VARIATION ON
POST-BREEDING BIRD USE OF TIMBER HARVESTS WITH

RESIDUAL CANOPY TREES

MOLLY E. MCDERMOTT'- AND PETRA BOHALL WOOD'

ABSTRACT. — We investigated avian use of clearcuts and two-age harvests during the post-breeding period in 2006 in
the central Appalachians, West Virginia, USA with an intormation-theoretic approach to model selection. Cover variables
appeared to be most important; e.g., vegetative vertical complexity had a strong positive relation with capture rates of
mature forest birds and molting adults, as well as physical condition which supports a predator-avoidance hypothesis for
habitat use. Basal area was a poor predictor of captures; residual trees near nets tended to depress capture rates. Food
variables best explained capture rates for some species groups (e.g., early-successional insectivores and granivores, mature
forest nesting adults, molting birds), but post-breeding habitat quality was based primarily on vegetative cover. Habitat use
may depend on the bird s physical condition and molt status, and we found evidence for age-specific differences which may
impact survival. Our study suggests important links between post-breeding habitat quality, molt status, physical condition,
and bird age, and indicates a variety of response variables (relative abundance, survival, body condition) should be
measured to assess avian habitat quality during the post-breeding period. Received 19 March 2009. Accepted 6 February
2010.

Recent declines in populations of migratory
birds that nest in both mature forest and early-
successional habitats have stimulated interest in
limiting factors and potential mortality sources
throughout the year (Robbins et al. 1989, Hunter
et al. 2001). Habitat loss and fragmentation in
breeding and wintering areas are probable factors
(Askins et al. 1990), but little is known about
factors affecting survival during the post-breeding
period (Baker 1993, Pagen et al. 2000), the
interval between fledging of young and migration.
Many bird species that nest in early-successional
or mature forest habitats use early-successional
habitat extensively during post-breeding. For
example, many species of mature forest breeding
.songbirds move into nearby shrublands after
young have fledged (Pagen et al. 2000, Marshall
et al. 2003, Vitz and Rodewald 2007). Juvenile
Wood Thrush (Hylocichla mustelina) and Swain-
son’s Thrush (Catharus ustulatus) are also known
to move >1 km from their mature forest natal
area into early-successional habitat before migra-
tion (Anders et al. 1998, Vega Rivera et al. 1998,
White et al. 2005, Dellinger 2007).

One explanation for this behavior is predator
avoidance; early-successional habitats may pro-
vide increa,sed protection from predators (White et

'West Virginia Cooperative Fish and Wildlife Research
Unit, P. O. Box 6125, 322 Percivai Hall, Division of
Forestry and Natural Resources. West Virginia University,
Morgantown, WV 26506, USA.

^Corresponding author; e-mail:
mollyemcderrnott@gmail.com

al. 2005). A dense understory provides cover for
fledglings that, lacking the knowledge to detect
and escape predators, often have high post-
fledging mortality (Sullivan 1989, Anders et al.
1997). Adults may also seek protection and cover
during the post-breeding period as they undergo
pre-basic molt and are temporarily more vulner-
able to predators (Pagen et al. 2000). More
understory vegetation and increased vertical
structure were associated with increased survival
for fledgling Ovenbirds {Seiurus aurocapilla)
(Vitz 2008), even where prey abundance was
low (King et al. 2006).

A second explanation is the re,source-selection,
or optimal-foraging hypothesis (i.e., bird use of
early-successional habitats is driven by greater
food resource abundance) (White et al. 2005).
Abundant food resources are needed by both
adults and juveniles for accumulation of fat
reserves in preparation for migration and comple-
tion of molt (Murphy and King 1992, DeGraaf
and Yamasaki 2003). Molt is nutritionally de-
manding and requires energy and protein not only
for synthesizing feathers, but for increasing
metabolism to offset reduced insulation and flight
efficiency (Jenni and Winkler 1994). Juvenile
Swainson’s Thrush u.se of coastal scrub during
post-fledging was best explained by fruit abun-
dance variables, supporting the resource-selection
hypothesis (White et al. 2005).

Two-age harvesting or green-tree retention,
where canopy trees are retained to more closely
resemble natural disturbances, is becoming more
common (Franklin et al. 1997, Rose and Muir

545

546

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 3, September 2010

1997, Vanha-Majamaa and Jalonen 2001, Vergara
and Schlatter 2006), and is frequently used as an
alternative to clearcut harvests in central Appala-
chian hardwood forests (Miller et al. 2006).
Mature residual trees and a regenerating age class
provide vertical stratification for both shrub and
canopy-associated birds (Duguay et al. 2000). The
effects of two-age harvesting on breeding birds
have been demonstrated (Baker and Lacki 1997,
Boardman and Yahner 1999, Duguay et al. 2000,
McDermott and Wood 2009), but no studies of
which we are aware have examined avian use of
stands with residual tree retention during the post-
breeding period.

The objectives of this study were to; (1)
quantify how vegetative structure and food
availability in timber harvests with a gradient of
residual canopy trees affect post-breeding bird
habitat use, and (2) examine the impacts of
resource variation on avian body condition and
molt status. Our study is unique in that we
examine both mature forest and early-succession-
al nesting bird species, investigate multiple types
of food resources at the nest-level, and integrate
molt and body condition variables.

METHODS

Study Areas. — Our study was conducted in
West Virginia on the Monongahela National
Forest (MNF) in Pocahontas and Tucker counties
(38° 31' N, 79° 44' W), and the Wildlife and
Ecosystem Research Forest (WERF) in Randolph
County (38° 43' N, 80° 03' W). These managed
forests in the central Appalachian Mountains are
characterized by narrow, low valleys dissected by
northeast-southwest ridges.

We selected nine stands with a gradient of
retained basal area ranging from 0 to 7 m^/ha for
sampling. Stand age (4-7 years post-harvest) was
a selection criterion, because there were few
harvests in an early-successional stage. We chose
stands on the MNF {n = 5) that were in close
proximity to those on the WERF (/? = 4). Harvest
area was 4.2-21 ha and elevation was 630-
1,090 m.

The principal forest types were northern
Allegheny hardwoods, cove hardwoods, mixed
mesophytic, and oak-hickory (Quercus spp.-
Carya spp.). Dominant overstory tree species
included red oak (Q. rubra), red maple (Acer
ruhruni), chestnut oak (Q. primis), and yellow
poplar (Liriodendron tulipifera). The most abun-
dant understory plants were blackberry (Ruhus

spp.), black birch (Betula lenta), greenbrier
{Smilax spp.), American beech (Fagus grand-
ifolia), maples, and yellow poplar.

Mist-Netting. — We used mist nets to sample
bird habitat use because surveying birds by vision
is difficult in dense vegetation, and many birds are
secretive during the post-breeding period (Ralph
et al. 2004). Our goal was to relate bird capture
rates to microhabitat food and cover variables. We
placed 10 mist nets (12 X 2.6 m with 30 mm
mesh) in each stand, often along skid roads
created during timber harvest and usually oriented
perpendicular to the prevailing aspect. We
arranged nets so that distance from edge was
variable; a minimum of 20 m separated ends of
nets, and each net was within 60 m of another net.
We first sampled stands randomly and conducted
repeat visits in the same order at least 3 weeks
apart. A stand (n = 3) from each harvest type
(clearcut, low-leave two-age, and high-leave two-
age) was randomly chosen for the first round, and
alternated for each sampling round.

We sampled 2 consecutive days at each stand
and visited each twice from 25 June to 15 August
2006. We opened nets at sunrise and closed after
4-5 hrs except in inclement weather, when mist
nets were not operated. We checked nets every
30 min and recorded species, age, gender, wing
chord, mass, fat, breeding condition (extent of
brood patch or cloacal protuberance), molt status,
and time of capture for each captured individual.
Captured birds, except for Ruby-throated Hum-
mingbirds {Archilochus colubris), were fitted with
uses aluminum bands. We filled out molt cards
per the British Trust for Ornithology for actively
molting adults to obtain pre-basic molt scores.

Food Resource Sampling. — We sampled fruits
twice in each stand, once per round immediately
after the first day of mist-netting. We obtained an
index of fruit abundance by counting individual
fruits along 12-m transects within 1 m of either
side of each mist-net lane (Levey 1988). All fruits
were counted, identified to species, and classified
as unripe or ripe.

We also collected arthropod samples, concur-
rently with fruit sampling, to obtain biomass
availability estimates (Hutto 1990, Smith and
Rotenberry 1990). We used biomass as an index
of food availability, not an absolute measure' of
food consumed by birds (Hutto 1990, Duguay et
al. 2000). We obtained four arthropod samples at
each net location (// = 90), one from each of four
saplings, each round. Saplings within 10 m of the

McDevmon and W ood - POST-BREEDINCJ BIRD USE OF TIMBER HARVESTS

547

nets were beaten with a rubber mallet to dislodge
arthropods (Schaulf 1997). This capture method
was most suitable for sampling arthropods
available to foliage-gleaning birds, the most
common foraging guild captured in the pilot year
(McDermott 2007). Samples were taken from two
random tree saplings within mist-net height on
each side of the net. We often used the same
individual trees in the second round, as the time
interval (~3 weeks) allowed sufficient time for
repopulation (Linda Butler, Entomologist, West
Virginia University, pers. comm.).

We used a sturdy, light-colored sheet, 1 m“ in
area to capture the arthropods. The sheet was held
under the branches of one side of the sapling, and
one person beat four limbs that were above the
sheet from top to bottom (four beats per limb).
Once specimens fell onto the sheet, small
arthropods were collected with an aspirator into
a glass vial of ethanol. We picked up larger
insects with forceps and placed them into a vial.
Specimens were stored in 70% ethanol.

We identified all arthropods to Order, and
classified each individual into one of three size
categories: <3, 3-10, and >10 mm (Van Horne
and Bader 1990). Specimens were subsequently
placed on a tray and inserted into a drying oven at
60° C. We weighed each sample (by taxonomic
Order, size, net, and round) to the nearest 0.1 mg
to obtain an estimate of biomass after 48 hrs or
when specimens achieved a constant mass.
Arthropod richness at the Order level was
calculated as a measure of arthropod diversity.

Vegetation Sampling. — We recorded presence
or absence of vegetative cover on four 12-m
transects, 3 m and 6 m from the net paralleling
each side of the net lane (Schemske and Brokaw
1981), at five equidistant intervals along each
transect in height categories: 0-1 , >1-2 , >2-3 ,
>3-6 , >6-12 , >12-18 , >18-24 , and >24 m.
A vertical complexity index (VCI) was calculated
as a sum of all height cover class percentages
divided by 100 and multiplied by a constant
(Wood et al. 2005). We quantified the number of
stems <7.6 cm dbh and percent ground cover
categorized as leaf litter, woody, rock/bare
ground, or herbaceous in two 2 X 2 m plots
randomly placed on the tran.sect along each side of
the net lane. We measured basal area with a 10-
factor prism from the center of the net lane.

Data Analyses. — We standardized captures for
each species per 100 mist-net hrs, where 1 mist-
net hr = 1 net open for 1 hr. Bird species that did

not breed locally were excluded from analyses,
and only first captures of individuals were used.
We assigned species to two breeding habitat
guilds: early-successional or mature forest ba.sed
on Ehrlich et al. (1988) and observations from
earlier studies in West Virginia (e.g., Duguay et
al. 2000). Species were placed into foraging
guilds (insectivorous or seasonally frugivorous)
ba.sed on available life history data (e.g.. Birds of
North America species accounts) and a cursory
examination of fecal samples from captured birds.
We performed separate analyses for hatch year
(HY = born that spring) and adult (AHY >1 year
of age) birds; juveniles are usually more suscep-
tible to capture and may use habitats differently
than adults. We occasionally caught family
groups, but it was a small percent of overall
captures (~5%; similar to Marshall et al. 2003),
and these age groups were treated independently.

We selected explanatory variables to model
bird responses to resource availability that repre-
sented the vegetative cover and food resources
near nets, the main hypotheses for habitat use. We
tested for pairwise correlation among all variables
using Pearson’s product-moment correlation (Proc
CORR; SAS Institute 2003). Percentage variables
were arcsine transformed to approximate normal-
ity. We reduced the number of explanatory
variables within a category (fruit, arthropod, and
vegetative cover) by dropping strongly coirelated
variables {P > 0.7) to avoid multicollinearity. We
selected the one that was less correlated with other
model variables or most biologically meaningful
when choosing between two coiTelated variables.
Nine variables were retained for modeling (Ta-
ble 1).

We developed sets of a priori candidate models
from these nine variables to explain avian capture
rates and other metrics of interest based on a
review of recent literature on post-breeding bird
ecology. The 8-13 models developed for each
species or group are described in McDermott
(2007). Arthropod biomass was analysis-specific:
prey size categories were chosen based on what is
known of prey items consumed by each bird
species. Total arthropod biomass included all prey
sizes when multiple bird species were combined
for a particular analysis.

Response variables included capture rates of all
adult and HY birds for species captured at >30%
of nets or the top four species captured in each
habitat guild, and capture rates of each age group
classified by diet and breeding habitat (all

548

THE WILSON JOURNAL OL ORNITHOLOGY • Vd. 122, No. 3, September 2010

TABLE 1. Net-level vegetative and food resource variables used to model bird use of early-successional habitats
during the post-breeding period in northcentral West Virginia, 2006. Resources were sampled concurrently with mist-
netting from late-June through mid-August.

Variable

Code

Value range

-V ^

t SE

Vegetative cover

Understory stem density''

DENSE

34-282

121.42

± 3.74

Residual basal area, m'/ha

BA

0-11.48

3.02

± 0.24

Vertical complexity index*’

VCI

42-117

74.92

± 0.98

Fruit resources'’

Fruit abundance

ERUIT

95-4,408

802.07

± 41.57

Total ripe fruit

RIPE

0-846

18.65

± 5.29

Arthropod resources'*

Arthropod richness, number of Orders

RICH

5-12

8.33

± 0.1 1

Total arthropod biomass, mg

ARTH

15.3^07.9

100.04

± 5.46

Lepidopteran larval biomass, mg

LEP

0-373.3

16.19

± 2.25

Seed resources

Herbaceous ground cover, %

SEED

3-100

47.6

± 1.97

“ Number of .stems <7.6 cm dbh at two 2 X 2 m plots on either .side of each mist-net lane.

^ Vertical complexity index calculated as a .sum of all canopy cover class percentages divided by 100 and multiplied by 20.
Fruit was measured along two 12 X I m transects on either side of each mist-net lane.

Arthroptxis were sampled with a beating technique at 4 shrubs/net.

seasonal frugivores and insectivores by habitat
guild). We only analyzed age classes with >20
captures. A small percentage of adults met strict
criteria for post-breeding (only one-third were not
in full breeding condition and were actively
molting); we included all adults captured during
this period including late-breeders (30%) for both
breeding habitat guilds. We analyzed molt score
index and body condition in addition to the
individual species and groups.

Adult molt score (0-250; Cherry and Canned
1984) was reclassified to values of 0-6. We did
not analyze molt .score as continuous because
birds with a score of 0 or 250 are not molting
flight feathers; these scores were reclassified as 0
because these birds would be similarly capable of
flight. Adults in the heaviest stage of molt (.score
= 100-140) are most likely to have compromised
flight and higher energetic demands (Jenni and
Winkler 1994); they received the highest recla.s-
sified value (6). Each HY individual was given a
molt score of 0-3 depending on the number of
feather tracts in pin.

Body condition was calculated by dividing
mass (g) by wing length (mm) to account for body
size (Winker 1995). Mass was regressed against
wing length for each species, and the residuals
were u.sed as the condition index. These values
were reclassified to a score of 0^.

We used Poisson regression (Proc NLMIXED;
SAS Institute 2003) to model each response
variable except body condition, which had a
nomial distribution and was analyzed with linear
regression. This procedure allows for a hierarchical
(i.e., nested) approach to account for within-stand
variability, and both Poisson and linear distribu-
tions can be modeled. Analyses were conducted at
the net scale because of variability in food and
vegetative resources within stands and to examine
microhabitat use patterns.

We used an information-theoretic approach to
select the best approximating models from each
set of candidate models. The global model for
each response variable was constructed using all
variables (range = 7 to 9) relevant to each species
or group. We used Akaike’s Information Criterion
adjusted for small sample sizes (AlC,..) regardless
of the sample sizeiparameter ratio, because
second-order AIC,. converges to AIC as n
becomes large (Burnham and Anderson 2002).
We did not use QAIC to re-rank models, instead
including a random effect in the regression
models to account for any dispersion problems
in the data (SAS Institute 2003). We re.scaled the
resulting information criteria for each candidate
model to obtain A, AIC^. (differences between the
top model and every other model) and calculated
Akaike weights (h’,) to better interpret the relative

McDermott ami Wood • POST-BREEDING BIRD USE OF TIMBER HARVESTS

549

TABLE 2. Number ot iiuliviiluals captured for individual .species or groups of species analyzed (in bold) in regression
models. Birds were captured by mist nets in northcentral West Virginia, June-August 2006.

Species

Foraging guild'

Total captures

Captures’’

AHY

HY

Early-successional species

Chestnut-sided Warbler

I

309

123

186

Common Yellowthroat

I

42

26

16

Gray Catbird

F

75

39

36

Indigo Bunting

I

63

41

22

Early-successional insectivores

I

520

245

275

Mature-forest species

American Redstart

I

58

26

32

Canada Warbler

I

41

18

23

Hooded Warbler

I

49

34

15

Red-eyed Vireo

I

59

51

8

Veery

F

68

39

29

Mature-forest insectivores

I

440

227

213

Frugivores

F

225

136

89

“ Foraging guilds analyzed were: I = primarily insectivores post-breeding and F = seasonal frugivores.
HY = hatch year (juveniles), AHY = after hatching year (adults).

likelihood of a model. Models with a A,- AIC < 2
were given equal consideration as competing
models.

RESULTS

We captured 1,189 individuals: 579 HYs and
610 adults of 15 early-successional and 38 mature
forest species. Insectivores predominated; season-
al frugivores comprised 20% of species and
individuals captured. Total captures per species
(by age) or group of species analyzed ranged from
22 to 275 (Table 2).

Early-Successioncil Bird Habitat Use. — The
same cover and food variables were important
for both age groups of early-successional insec-
tivores (Table 3). Capture rates were negatively
related to vertical complexity (Fig. 1 ), understory
stem density, and basal area. Age groups had
opposite relationships to food variables. Adults
were positively related to arthropod biomass
(Fig. 2) and negatively to Lepidopteran larval
biomass, the opposite of HYs.

We analyzed four early-successional bird
species including three in.sectivores (Table 2).
Chestnut-sided Warbler (Dendroica pensylvanica)
captures were poorly predicted by all variables
(Table 3). Five models with different variables
had a similar likelihood of being selected for
adults, while HY capture rates were best
explained by the global model. Associations were
the same as for all early-successional HY captures

pooled because this species comprised 67% of
these captures. A negative association with
residual basal area best explained Common
Yellowthroat adult {Geothlypis trichas) captures
(Table 3, Fig. 3). Captures of adult Indigo
Buntings (Passerina cyanea) were best explained
by increases in vertical complexity. Several
models competed to explain captures of HY
Indigo Buntings. Of these, models with vertical
complexity and basal area (negative relationships)
had the highest collective weights.

Models containing food variables were impor-
tant for some species (Table 3). Adult Indigo
Bunting captures were positively associated with
herbaceous ground cover, which may represent
availability of seeds. Lepidopteran larval biomass
and total arthropod biomass were positive predic-
tors of Chestnut-sided Warbler HY and adult
captures, respectively. HY Indigo Bunting cap-
tures were positively associated with Lepidopter-
an larval biomass.

We analyzed one seasonal frugivore in the
early-successional habitat guild: Gray Catbird
{Duinetella carolinensis). Ripe fruit was nega-
tively related to Gray Catbird capture rates
(Table 3), and basal area and stem density were
positively related to adult captures. The global
model best explained HY catbird captures;
vertical complexity and basal area were positively
related to captures whereas nearly all food
variables were negatively related.

550

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 3. September 2010

TABLE 3. Regression analysis of bird capture rates by species, guild, and age group in northcentral West Virginia,
2006. Relationships with explanatory variables in competing models are presented.

Response variable

Explanatory variables

Early-successional species

Chestnut-sided Warbler, adults
Chestnut-sided Warbler, HY
Common Yellowthroat, adults
Gray Catbird, adults

Gray Catbird, HY

Indigo Bunting adults

Indigo Bunting HY

LEP (-), ARTH (-b), BA {+). DENSE (+), VCl (-)

GLOBAL"*

BA (-)

BA {+), VCI (+), RIPE (-), DENSE (+)

GLOBAL

VCl (-0, SEED (-b), RICH (-b), ARTH (-b)

BA (-), VCl (-), DENSE (-), RICH (-), LEP (-b), SEED (-),

ARTH (-)

Early-successional insectivores

Adults

HY

ARTH (-b), LEP (-), VCl (-), BA (-), DENSE (-)

GLOBAL

Mature forest species

American Redstart, adults
American Redstart, HY

Canada Warbler, HY

Hooded Warbler, adults
Red-eyed Vireo, adults

Veery, adults

Veery, HY

DENSE (-), ARTH (-b), VCl (-b)

VCI (-b), ARTH (-), DENSE (-b), LEP (-b), BA (-b)

BA (-), ARTH (-), DENSE (-), VCI (-)

ARTH (-b), DENSE (-b), LEP (-b)

BA (-), RICH (-b), DENSE (-), VCI (-b)

ARTH (-b). FRUIT (-b)

VCI (-), RIPE (-b)

Mature forest insectivores

Adults

HY

ARTH (-b), DENSE (-), VCl (+). LEP (-b), BA (-)

GLOBAL

All frugivores

Adults

HY

BA (-b), RIPE (-), DENSE (-) FRUIT (-b), VCI (-)

RIPE (-b), BA (+), DENSE (-), VCI (-b). FRUIT (-b)

Molt index

Adults

HY

RIPE (-b), ARTH (-), VCI (-b), BA (+), FRUIT (-b), LEP (-b), DENSE (-b)
FRUIT (-b), ARTH (-b)

Body condition index

DENSE (-), LEP (-), VCI (-b), FRUIT (-)

“ The global model, in which all variables were u.sed, wa.s the only competing model.

Mature Forest Bird Habitat Use. — Adult insec-
tivore capture rates were best explained by both
cover and food variables (Table 3). Capture rates
increased with increasing vertical complexity
(Fig. 1), total arthropod biomass (Fig. 2), and
Lepidopteran larval biomass, and decreased with
increasing basal area (Fig. 3) and stem density.
The global model best predicted capture rates of
HYs. Cover variables had the same relationships
as with adults, but arthropod biomass was
inversely related to captures of HYs.

Four of five species we analyzed were primary
in.sectivores (Table 2). Cover variables dominated
the top models for these species. Vertical
complexity was one ot the top explanatory
variables for both HY and adult American

Redstart (Setophaga ruticilla) captures; (positive
relationship). Red-eyed Vireo (Vireo olivaceus)
adults were also positively associated with
vertical complexity. The basal area model re-
ceived the most support for captures of HY
Canada Warblers (Wilsonia canadensis) and adult
Red-eyed Vireos (Table 3) showing a negative
relationship for these species (Fig. 3). The stem
density model received a weight of 13% for Red-
eyed Vireo adult captures and 20% for Hooded
Warbler (Wilsonia citrina) adult captures, but
showed opposite patterns: Red-eyed Vireo capture
rates declined while Hooded Warbler rates
increased with increasing understory stem density.

Food variables were positively associated with
captures of several species (Table 3): Lepidopter-

McDermott am! Wood • POST-BREEDING BIRD USE OF TIMBER HARVESTS

551

(A)

“liarly-successional

HY

- Early-successional
adults

• Mature forest HY

-Mature forest
adults

Vertical complexity index

-Body condition
index

^ ^ — Adult molt index

40 60 80 100 120

Vertical complexity index

FIG. 1. Relationship of vertical complexity of the
vegetation in early-successional habitats of northcentral
West Virginia in 2006 with (A) capture rate (captures/100
net hrs) of adult and hatch year (HY) birds of mature-forest
and early-successional insectivorous species, and (B) body
condition and molt index of captured adults.

an larvae with adult Hooded Warbler and HY
American Redstart captures; arthropod richness
with Red-eyed Vireo adult captures; and arthro-
pod biomass with American Redstart and Hooded
Warbler adults.

The Veery (Catharus fuscescens) was the only
seasonal frugivore analyzed in the mature forest
habitat guild. Adult Veery capture rate was

FIG. 2. Captures/100 net hrs as a function of arthropod
biomass for all mature-forest and early-successional adult
insectivores, adult Veeries (seasonal frugivores), and early-
successional hatch year (HY) birds in early-successional
habitats of northcentral West Virginia in 2006.

FIG. 3. Relationship between capture rate (captures/
100 net hrs) and basal area for HY and adult frugivores,
mature forest adults. Red-eyed Vireo adults, Canada
Warbler hatch year (HY) birds, and Common Yellowthroats
in early-successional habitats of northcentral West Virginia
in 2006.

positively associated with arthropod biomass
(Fig. 2) and total fruit (Table 3). The model with
vertical complexity (negative association) was
ranked only 1.5 times better than that with ripe
fruit (positive association) for HYs. However,
these relationships were weak suggesting inade-
quate model structure.

Frugivore Habitat Use. — Nine species from
different breeding habitat associations represented
the seasonal frugivores. Total fruit abundance was
positively associated with both age groups but
was not a strong predictor in top models
(Table 3). Age groups showed opposite relation-
ships with ripe fruit: negative for adults and
positive for HYs. The model with basal area
(positive association) received 33% of the weight
for adults, and basal area was the highest ranked
cover parameter when considering all HY bird
models (Fig. 3).

Molt Score. — Molting HYs were positively
associated with total fruit abundance (Table 3).
This model was —2.5 times as likely to be
selected as the model with arthropod biomass {w
= 0.11). Abundance of ripe fruit best explained
molt score of adults (vv = 0.69), but the global
model also received support. All cover variables
in the global model were positively related to
adult molt, and vertical complexity was the
highest ranked cover variable when considering
all models. Complex vertical structure (a high
index value) was associated with individuals in
the most compromising stage of molt (Fig. 1 ).

Body Condition Index. — Cover variables best
explained variation in body condition index
(Table 3). Low stem density (w = 0.15) and high

552

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3, September 2010

vertical complexity {w — 0.15; Fig. 1) best
predicted high body condition index values.
Arthropod and fruit variables were negatively
related to body condition.

DISCUSSION

Predator- Avoidance Hypothesis. — Cover vari-
ables were important for post-fledging juveniles
and adults using timber harvests. Birds may be
selecting habitats with increased vegetative com-
plexity that offers protection from predators. King
et al. (2006) found that fledgling Ovenbirds had
higher survival rates where understory vegetative
structure was greatest. Thus, vertical complexity
may be a good representation of protective cover.

Thicker cover as measured by understory stem
density had a negative influence on bird captures.
Similarly, Vitz and Rodewald (2007) found a
negative relationship between mature forest bird
capture rates and density of shrub vegetation.
Birds in our study were often traveling in foraging
flocks that contained several species, and an open
understory would facilitate movement of individ-
uals.

Adult Chestnut-sided Warblers, Gray Catbirds,
and Hooded Warblers, species that nest in the
harvested areas (McDermott 2007), had higher
capture rates at nets with more understory stems.
Thus, a higher understory stem density could
serve as protective cover during nesting and post-
breeding for these species. Early-successional bird
habitat use outside of breeding tends to be less
flexible than for those species nesting in mature
forest (Rodewald and Brittingham 2004). Stand-
level variables such as area and amount of edge
(McDermott 2007, Schlossberg and King 2008)
may explain more variation in habitat use by those
species nesting in shrub-scrub.

Residual trees in two-aged stands inhibit
understory growth and limit stem density (Miller
et al. 2006), but leave-trees provide raptor perches
and residual trees may not offer predator protec-
tion. This suggestion is supported by our data as
captures were negatively associated with basal
area; within stands, captures were greater at nets
with fewer nearby trees.

Resource-Selection Hypothesis. — This study
provides evidence that post-breeding habitat use
is influenced by food resource availability. For
example, arthropod biomass positively predicted
captures of adult insectivores and Veeries, which,
like other seasonal frugivores, also consume
arthropods. Lepidopteran larval biomass was a

positive predictor of mature forest (e.g., Canada
and Hooded warblers) and early-successional HY
captures. Lepidopteran larvae are an important
food source for nestling and adult songbirds (Van
Horne and Bader 1990, Jones et al. 2003). Shrub-
dwelling Lepidopteran larvae may have higher
biomass in clearcuts than in mature forest because
deciduous shrubs have high amounts of foliage
(Van Horne and Bader 1990, Keller et al. 2003).
Thus, some birds, especially mature forest indi-
viduals, may be keying on this rich energetic
source within early-successional habitats. Arthro-
pods may be a fleeting resource because biomass
is significantly higher in young clearcuts when
they become fully stocked (by year 6 or 7) versus
later serai stages (Keller et al. 2003).

We found few strong relationships with total
fruit abundance. Similarly, Vitz (2008) did not
find evidence that mature forest birds use clear-
cuts for fruit resources. White et al. (2005) found
a positive relationship between thrush habitat use
and fruit abundance, but we found only a weak
positive relation for the Veery, the only thrush
species analyzed individually in our study.

We included herbaceous cover as a crude index
of seed availability for seed-eating birds in the
models. Indigo Bunting captures, particularly
adults, were explained best by herbaceous cover.
Thus, herbaceous plants may serve as a cue for
seed availability for granivores.

We found limited evidence for the resource
selection hypothesis. We do not know accurately
what constitutes available food for a bird without
behavioral observations to quantify food resource
selection (Hutto 1990). Foliage-associated arthro-
pod biomass may have misrepresented available
food, despite adjusting for potential prey size
differences. Birds may not be able to key in on
food resources at the net scale because of high
within-stand variability in arthropod communities.
We do not know how or if food is limiting during
the post-breeding period. Food limitation likely
varies from year to year and is more likely to
occur during breeding because of food require-
ments of rapidly growing nestlings (Nagy and
Holmes 2005).

Food and Cover Related to Condition, Molt,
and Age. — The physical condition of a bird in
theory is positively related to food resources and
deposited body fat. However, food resource
availability in our study was negatively associated
with body condition of captured birds. Instead, a
high index value was associated with high vertical

McDermott and Wood - POST-BREEDING BIRD USE OF TIMBER HARVESTS

553

complexity. One explanation is that birds with a
higher body condition index may use the most
structurally complex habitat, which we infer
offers the best protection and increases likelihood
of survival. Birds in poor condition may select
habitat with rich food resources to replenish fat
reserves, while birds in better condition select
microhabitat that provides protection from pred-
ators (Moore and Aborn 2000). Habitat use may
depend on physical condition, and both cover and
food resources are important for post-breeding
survival, which is influenced by body condition of
fledglings in some species (Krementz et al. 1989,
Naef-Daenzer et al. 2001 ). There was no evidence
that body condition increased over the season; we
sampled prior to pre-migratory fattening, and the
mean condition index decreased over time due to
a concomitant increase in HY captures.

Age-specific differences were evident for most
of the groups analyzed, particularly with food
variables. Adults had a higher body condition
index than juveniles for most species analyzed
(MEM, unpubl. data), which may have contribut-
ed to these differences. HY capture rates were
negatively associated with arthropod biomass,
while adult capture rates had a positive relation-
ship. HYs are likely inexperienced in recognizing
arthropod availability cues and may have a lower
foraging efficiency than adults (Stevens 1985,
Weathers and Sullivan 1989, Vanderhoff and
Eason 2008) and thus, lower survival. These data
bolster the argument that adults and HYs may be
keying on different features within a stand, which
may account for age-specific survival rates during
the post-breeding period.

Molting birds likely take advantage of both
food and cover. Adults in heavy molt with
diminished flight capabilities were associated
with high vertical complexity, which likely
provides protection from predators. Molting birds
were positively associated with fruit variables.
Changes in food choice have been ob.served in the
field for molting passerines, and food availability
may influence the timing and place of molt (Jenni
and Winkler 1994). Increased energy require-
ments during molt could be fulfilled by an
abundance of fruits for the 13% of molting adults
and 15% of molting Juveniles that were frugivo-
rous or, for insectivores, an abundance of
arthropods that are attracted to fruits (Sallabanks
and Courtney 1992).

Early-successional habitats, including those
retaining canopy trees such as the two-age

harvests we sampled, should be recognized for
their conservation value. Avian survival may be
compromised by ignoring bird habitat require-
ments outside of breeding (King et al. 2006).
Habitat use does not imply habitat quality; a
variety of parameters (relative abundance, surviv-
al, body condition) should be measured to
evaluate post-breeding habitat quality for birds.

ACKNOWLEDGMENTS

This work was funded by the USDA Forest Service,
Monongahela National Forest. The U.S. Geological Survey,
West Virginia Cooperative Fish and Wildlife Research Unit
provided field vehicles, equipment access, and logistical
support. West Virginia University, USDA Forest Service,
and P. D. Keyser ahso provided logistical support. We thank
MeadWestvaco for allowing access to the Wildlife and
Ecosystem Research Forest and USDA Forest Service for
access to sites on the Monongahela National Forest. We are
grateful to K. G, Fleyden, P. M. McElhone, J. S. Rehar, E.
E. Samargo, M. B. Shumar, and J. C. Walker for help
collecting data in the field and P. M. McElhone and M. B.
Shumar for help sorting insects in the laboratory. We thank
P. J, Herron for allowing us to work under his master
banding permit. J. T. Anderson, C. M. Johnson, and Sue
Olcott provided comments on an earlier version of this
manuscript.

LITERATURE CITED

Anders, A. D., J. Faaborg, and F. R. Thompson III. 1998.
Postfledging dispersal, habitat use. and home-range
size of juvenile Wood Thrushes. Auk 1 15:349-358.
Anders, A. D., D. C. Dearborn. J. Faaborg, and F. R.
Thompson 111. 1997. Juvenile survival in a population
of neotropical migrant birds. Conservation Biology
11:698-707.

Askins, R. a., j. F. Lynch, and R. Greenberg. 1990.
Population declines in migratory birds of eastern North
America. Current Ornithology 7:1-57.

Baker, M. D. and M. J. Lacki. 1997. Short-term changes
in bird communities in response to silvicultural
prescriptions. Forest Ecology and Management
96:27-36.

Baker, R, R, 1993. The function of post-lJedging
exploration: a pilot study of three species of pas.serines
ringed in Britain. Ornis Scandinavica 24:71-79.
Boardman, L. a. and R. H. Yahner. 1999. Wildlife
communities associated with even-aged reproduction
stands in two stale forests of Pennsylvania. Northern
Journal of Applied Forestry 16:89-95.

Burnham, K. P. and D. R. Ander.son. 2002. Model
.selection and multimodcl inference: a practical-
theoretic approach. Second Edition. Springer-Verlag.
New York, USA.

Cherry, J. D. and P. F. Cannell. 1984. Rate and timing of
prebasic molt of adult Boreal Chickadees. Journal of
Field Ornithology 55:487^89.

DeGraaf, R. M. and M. Yama.saki. 2003. Options for
managing early-successional forest and shrubland bird

554

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3, September 2010

habitats in the northeastern United States. Forest
Ecology and Management 185:179-191.

Dellinger, T. A. 2007. Post-fledging ecology and survival
of neotropical migratory songbirds on a managed
Appalachian forest. Thesis. West Virginia University,
Morgantown, USA.

Duguay, J. P., P. B. Wood, and G. W. Miller. 2000.
Effects of timber harvests on invertebrate biomass and
avian nest success. Wildlife Society Bulletin 28:1123-
1 131.

Ehrlich, P. R., D. S. Dobkin, and D. Wheye. 1988. The
birder’s handbook. Simon and Schuster, New York,
USA.

Franklin, J. F., D. R. Berg, D. A. Thornburgh, and J. C.
Tappeiner. 1997. Alternative silvicultural approaches
to timber harvesting: variable retention harvest
systems. Pages 111-139 in Creating a forestry for
the 21st Century (K. A. Kohm and J. F. Franklin,
Editors). Island Press, Washington, D.C., USA.

Hunter, W. C., D. A. Buehler, R. A. Canterbury, J. L.
Confer, and P. B. Hamel. 2001. Conservation of
disturbance-dependent birds in eastern North America.
Wildlife Society Bulletin 29:440-455.

Hutto, R. L. 1990. Measuring the availability of food
resources. Studies in Avian Biology 13:20-28.

Jenni, L. and R. Winkler. 1994. Moult and ageing of
European passerines. Academic Press Limited, Lon-
don, United Kingdom.

Jones, J., P. J. Doran, and R. T. Holmes. 2003. Climate
and food synchronize regional forest bird abundances.
Ecology 84:3024-3032.

Keller, J. K., M. E. Richmond, and C. R. Smith. 2003.
An explanation of patterns of breeding bird species
richness and density following clearcutting in north-
eastern USA forests. Forest Ecology and Management
174:541-564.

King, D. 1, R. M. Degraaf, M. L. Smith, and J. P.
Buonaccorsi. 2006. Habitat selection and habitat-
specific survival of fledgling Ovenbirds (Seiurus
auroeapilla). Journal of Zoology 269:414-421.

Krementz, D. G., j. D. Nichols, and J. E. Hines. 1989.
Post-fledging survival of European Starlings. Ecology
70:646-655.

Levey, D. 1988. Spatial and temporal variation in Costa
Rican fruit and fruit-eating bird abundance. Ecological
Monographs 58:251-269.

Marshall, M. R., J. A. Dececco, A. B. Williams, G. A.
Gale, and R. J. Cooper. 2003. Use of regenerating
clearcuts by late-successional bird species and their
young during the post-fledging period. Forest Ecology
and Management 183:127-135.

McDermott, M. E. 2007. Breeding and post-breeding
forest bird community dynamics in regenerating
clearcuts and two-age harvests in the central Appala-
chians. Thesis. West Virginia University, Morgan-
town, USA.

McDermoit, M. E. and P. B. Wood. 2009. Short- and
long-term implications of clearcut and two-age
silviculture for con.servation of breeding fore.st birds
in the central Appalachians, USA. Biological Conser-
vation 142:212-220.

Miller, G. W., J. N. Kochenderfer, and D. B.
Fekedulegn. 2006. Influence of individual reserve
trees on nearby reproduction in two-aged Appalachian
hardwood stands. Forest Ecology and Management
224:241-251.

Moore, F. R. and D. A. Aborn. 2000. Mechanisms of en
route habitat selection: how do migrants make habitat
decisions during stopover? Studies in Avian Biology
20:34^2.

Murphy, M. E. and J. R. King. 1992. Energy and nutrient
use during moult by White-crowned Sparrows,
Zonotrichia leiicophrys gcimbelii. Ornis Scandinavica
23:304-313.

Naef-Daenzer, B., F. Widmer, and M. Nuber. 2001.
Differential post-fledging survival of Great and Coal
tits in relation to their condition and fledging date.
Journal of Animal Ecology 70:730-738.

Nagy, L. R. and R. T. Holmes. 2005. Food limits annual
fecundity of a migratory songbird: an experimental
study. Ecology 86:675-681.

Pagen, R. W., F. R. Thompson III, and D. E. Burhans.
2000. Breeding and post-breeding habitat use by forest
migrant songbirds in the Missouri Ozarks. Condor
102:738-747.

Ralph, C. J., E. H. Dunn, W. J. Peach, and C. M.
Handel. 2004. Recommendations for the use of mist
nets for inventory and monitoring of bird populations.
Studies in Avian Biology 29:187-196.

Robbins, C. S., J. R. Sauer, R. S. Greenberg, and S.
Droege. 1989. Population declines in North American
birds that migrate to the Neotropics. Proceedings of
the National Academy of Science of the USA
86:7658-7662.

Rodewald, P. G. and M. C. Brittingham. 2004. Stopover
habitats of landbirds during fall: use of edge-
dominated and early-successional forests. Auk
121:1040-1055.

Rose, C. R. and P. S. Muir. 1997. Green-tree retention:
consequences for timber production in forests of the
western Cascades. Ecological Applications 7:209-217.

Sallabanks, R. and S. P. Courtney. 1992. Frugivory,
seed predation, and insect-vertebrate interactions.
Annual Review of Entomology 37:377^00.

SAS Institute. 2003. Version 9.1. SAS Institute Inc., Cary,
North Carolina, USA.

SCHAUFF, M. E. 1997. Collecting and preserving insects and
mites: techniques and tools. NHB-168. USDA, Sys-
tematic Entomology Laboratory, National Museum of
Natural History. Washington, D.C., USA.

SCHEMSKE, D. W. AND N. Brokaw. 1981. Treefalls and the
distribution of understory birds in a tropical forest.
Ecology 62:938-945.

SCHLOSSBERG, S. AND D. 1. KING. 2008. Are shrubland birds
edge specialists? Ecological Applications 18:1325-
1330.

Smith, K. G. and J. T. Rotenberry. 1990. Quantifying
food resources in avian studies: present problems and
future needs. Studies in Avian Biology 13:3-5.

STEVEN.S, J. 1985. Foraging success of adult and juvenile
starlings Sturnus vulgari.'i: a tentative explanation for

McDermott cnid Wood - POST-BREEDING BIRD USE OF TIMBER HARVESTS

555

the preference of juveniles for cheiries. Ibis 127:341-
347.

Sullivan, K. A. 1989. Predation and starvation: age-
specific mortality in juvenile Juncos (Jitnco phaeno-
tiis). Journal of Animal Ecology 58:275-286.

V.ANDERHOFF, E. N. AND P. K. Eason. 2008. Comparisons
between juvenile and adult American Robins foraging
for mulberry fruit. Wilson Journal of Ornithology
120:209-213.

Vanha-Majamaa, I. and J. Jalonen. 2001. Green tree
retention in Fennoscandian forestry. Scandinavian
Journal ot Forest Research Supplement 3:79-
90.

Van Horne, B. and A. Bader. 1990. Diet of nestling
Winter Wrens in relationship to food availability.
Condor 92:413-420.

Vega Rivera, J. H., J. H. Rappole, W. J. McShea, and C.
A. Haas. 1998. Wood Thrush postfledging movements
and habitat use in northern Virginia. Condor 100:69-
78.

Vergara, P. M. and R. P. Schlatter. 2006. Aggregate
retention in two Tierra del Fuego Nothofagus forests:

short-term elfects on bird abundance. Forest Ecology
and Management 225:213-224.

ViTZ, A. C. 2008. Survivorship, habitat use, and movements
for two species of mature forest birds. Dis.sertation.
Ohio State University, Columbus, USA.

ViTZ, A. C. AND A. D. Rodewald. 2007. Vegetative and
fruit resources as determinants of habitat use by
mature forest birds during the postbreeding period.
Auk 124:494-507.

Weathers, W. W. and K. A. Sullivan. 1989. Juvenile
foraging proficiency, parental effort, and avian repro-
ductive success. Ecological Monographs 59:223-246.

White, J. D., T. Gardali, F. R. Thompson III, and J.
Faaborg. 2005. Resource selection by juvenile
Swainson’s Thrushes during the postfledging period.
Condor 107:388-401.

Winker, K. 1995. Autumn stopover on the Isthmus of
Tehuantepec by woodland Nearctic-Neotropic mi-
grants. Auk 112:690-700.

Wood, P. B., J. P. Duguay, and J. V. Nichols. 2005.
Cerulean Warbler use of regenerated clearcuts and two-
age harvests. Wildlife Society Bulletin 33:851-858.

The Wilson Journal of Onhihology 1 22(3):556-562, 2010

DISTRIBUTION, ABUNDANCE, AND STATUS OF CUBAN SANDHILL
CRANES {GRUS CANADENSIS NESIOTES)

XIOMARA GALVEZ AGUILERA'-' AND EELIPE CHAVEZ-RAMIREZ-^

ABSTRACT. — We conducted the first country-wide survey between 1994 and 2002 to examine the distribution,
abundance, and conservation status of Sandhill Crane (Gnis canadensis nesiotes) populations throughout Cuba. Ground or
air surveys or both were conducted at all identified potential areas and locations previously reported in the literature. We
define the current distribution as 10 separate localities in six provinces and the estimated total number of cranes at 526
individuals for the country. Two populations reported in the literature were no longer present and two localities not
previously reported were discovered. The actual number of cranes at two localities was not possible to evaluate due to their
rarity. Only four areas (Isle of Youth, Matanzas, Ciego de Avila, and Sancti Spiritus) each support more than 70 cranes. The
remaining locations each have less than 25 individuals. Sandhill Cranes appear to be declining and have almost disappeared
in Pinar del Rio and Granma provinces, and in northern Matanzas Province. Identified threats to the remaining populations
include habitat modification (woody plant encroachment, agricultural expansion, and fire suppression), predation due to
wild hogs (Sus scrofa), dogs [Canis lupus familiaris), mongoose (Crossarchus spp.), and poaching. Received 3 November
2009. Accepted II February’ 2010.

The Cuban subspecies of Sandhill Crane (Grus
canadensis nesiotes) is one of many endemic bird
subspecies and species present in Cuba. Little
information regarding distribution or overall
ecology (Meine and Archibald 1996) has been
published recently or historically on this rare
crane despite being one of the largest birds in the
country, and the Caribbean. The Cuban Sandhill
Crane has unique characteristics relative to the
other subspecies. For example, most nests are
constructed on dry land (Walkinshaw 1953,
Galvez Aguilera et al. 2005), which differs from
all other subspecies which nest primarily in
association with wetlands (Tacha et al. 1992).
The Cuban subspecies of Sandhill Crane is listed
as critically endangered (lUCN 1994). Only four
populations were known in the early 1990s and
the population for Cuba was estimated at between
100 and 200 individuals by different authors.

The first records of Sandhill Cranes in Cuba are
from Poey (1851-1855) and Gundlach (1875,
1876) who described the distribution of this
species in large savannahs in Cienega de Zapata
and pine (Finns spp.) -dominated savannahs east
of Guamutas and on the Isle of Pines (cuiTently

' Empresa Nacional Para la Protcccion de Flora y Fauna,
Ministerio de Agricultura. Havana. Cuba.

^Platte River Whooping Crane Maintenance Trust Inc..
661 1 West Whooping Crane Drive. Wood River, NE
68X8.3, USA.

’Current address: Conservacion de Flamencos y Aves,
Ninos y Cn'as AC, C. 33 #503 x 6 y 72, Merida, Yucatan,
C.P. 90070, Mexico.

■‘Corresponding author; e-mail:
fchavez@whoopingcrane.org

named Isle of Youth). Barbour (1943) reported
cranes from south of Matanzas Province near
Alacranes and Union de Reyes. Bangs and Zappey
(1905), Read (1913), Walkinshaw and Baker
(1946), and Garrido (1985) described the distri-
bution on the Isle of Youth (Pines) as limited to
open plains north of Cienega de Lanier. Walk-
inshaw (1953) defined the distribution on the Isle
of Youth as encompassing the region north of
Cienega de Lanier from Siguanea westward to
West Port and eastward to Pasadita. They are
reported to extend to Sabana Grande on the Isle of
Youth during winter. Garcia (1987) summarized
the status of endemic subspecies in Cuba and
considered cranes rare in all regions where they
were previously considered abundant, such as in
Guane, Mendoza, and Vinales. Garcia (1987)
emphasized the need to protect cranes where they
were still present at that time, including Pinar del
Rio, Ciego de Avila, Camaguey, Cienega de
Zapata, and the Isle of Youth.

The presence of Sandhill Cranes has been
documented over time in different regions of
Cuba; however, few authors have made quantita-
tive estimates of their abundance. The most
documented population is that on the Isle of
Youth. The status of Cuban Sandhill Cranes in
Cuba at the beginning of our field work was not
well known and many different estimates of crane
numbers had been put forward from a low of 100
to a high of “about” 300 individuals (Meine and
Archibald 1996). There have been reports over the
past several years on the population size of cranes
by different authors, although they are not in
agreement. Cuban Sandhill Cranes, from the time

556

Galvez Aguilera and Chavez- Ramirez • CUBAN SANDHILL CRANE STATUS

557

Gundlach (1875) made his observations when
cranes were believed to have been “plentiful”,
are considered to have decreased (Walkinshaw
1953). Cranes were considered to be more
common during the mid 1980s than formerly
(GaiTido 1985), suggesting that numbers had
increased between the 1950s and 1980s. Walk-
inshaw (1953) suggested that during the early
1950s, Sandhill Cranes had almost disappeared
from mainland Cuba. Numbers reported for
Cuban Sandhill Cranes have fluctuated from 200
in 1953 (Walkinshaw 1953) to 120 in 1975
(Gan-ido and Garcia 1975) to more than 200
individuals in the early 1990s (XGA, pers obs.).
Most reports citing numbers of cranes in Cuba are
estimates based on only limited field research or
tield surveys, which we believe underestimated
actual crane numbers.

Many of the 15 crane species, and several
subspecies of more common cranes, are consid-
ered to be threatened or endangered (Meine and
Archibald 1996). The Cuban Sandhill Crane at the
beginning of our work was considered to be
critically endangered per the lUCN Red List
(lUCN 1994), likely because no reliable informa-
tion on numbers and distribution was available.
The critically endangered designation was based
on the belief the populations present were highly
fragmented and each numbered less than 50
individuals. We undertook a study to evaluate
the current status of the Cuban subspecies of
Sandhill Crane in 1995 due to lack of reliable
recent information and disagreement among
different literature sources on distribution and
abundance. Our overall objective was to define
the distribution and estimate abundance of the
Cuban Sandhill Crane throughout the country.
Our specific objectives were to: (1) identify the
presence and distribution of different popula-
tion(s) or subpopulations of this subspecies in
Cuba, (2) estimate the number of individuals
present at each .separate location or population,
and (3) evaluate the status and potential conser-
vation problems for each of the distinct popula-
tions.

METHODS

Country-wide Surveys. — We initially planned a
nation-wide search; however, the entire country is
not suitable for cranes and we limited our field
surveys to areas deemed appropriate based on
vegetation characteristics, topography, and per-
sonal knowledge of different areas. A preliminary

evaluation of Sandhill Crane distribution was
defined by contacting field personnel of the
Empresa Nacional para la Proteccion de Flora y
Fauna (National Agency for the Protection of
Flora and Fauna, hereafter Flora and Fauna) to
identify areas where cranes had been observed in
the previous 5 years. Areas reported in the
literature as having cranes were included for
preliminary evaluation, regardless of whether new
or recent observations had been reported. Each
site believed to have potential for cranes was
visited and presence was confirmed through
actual sighting of cranes from air or ground
surveys. The number of cranes present at selected
sites in each area was estimated from intensive
and extensive ground counts.

Aerial surveys were conducted throughout a
large portion of the country in areas described on
vegetation maps as having characteristics of
potentially suitable habitat for Sandhill Cranes
(open areas, savannahs, and sparsely forested
areas). Open habitat types in Cienega de Zapata,
and the provinces of Isle of Youth, Sancti Spiritus,
and Camaguey were surveyed at low altitudes
(40-60 m) during October 1995 using a fixed-
wing airplane. Areas in the provinces of Pinar del
Rio, Cienega de Birama, and the northern regions
of Matanzas were surveyed from the air in
January 1996.

Areas in each province with appropriate
vegetation characteristics, or actual sightings
following the aerial surveys, were visited to
conduct personal interviews with local inhabi-
tants. Three hundred and forty interviews were
conducted of between 10 and 20 individuals in
each potential region who had lived in the area at
least 20 years. Those interviewed included
employees of Flora and Fauna, forest rangers,
fishermen, farmers, cattle producers, and others
who spent much of their time outdoors in areas
likely to be used by cranes. Areas where
interviews indicated cranes were likely to be
present were surveyed extensively by driving and
walking routes to estimate the number of cranes
present. More intensive and standardized counts
were conducted in areas where cranes appeared to
be abundant, or widespread.

Population Estimates. — Coordinated point
counts were conducted in four areas that appeared
to support the largest numbers of cranes to define
their distribution and estimate population size. All
counts were conducted during February between
1995 and 2002 with a different area sampled each

558

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 3, September 2010

TABLE 1. Cuban Sandhill Crane populations by location and province.

Province

Location

Population'

Estimated number

Status

Protected by'’

Pinar del Rio

Guane

1

<10

Stable

ENPFF/CGB

Consolacion Sur

2t

0

Extirpated

NA

Matanzas

Cienega de Majaguillar

3

<12

Unknown

ENPFF

Cienega de Zapata

4

120

Increasing

EFI/(AP)

Sancti Spiritus

Cienega de las Guayaberas

5

71

Stable

ENPFF (AP)

Ciego de Avila

Moron Norte

6

102

Stable

ENPFF (AP)

Jucaro

7

Present

Stable?

ENPFF

Camaguey

Sabana de Lesca

8

24

Stable

CGB

Cayo Romano

9

16

Unknown

ENPFF (AP)

Granma

Cienega Birama

lot

7

Not confirmed

ENPFF (AP)

Isle of Youth

Los Indies (11) and Sabana Grande

11, 12

171

Increasing

ENPFF (AP)

(12)

“ Figure 1.

^ ENPFF = Empresa Nacional para la Proteccion de Flora y Fauna (National Agency for the Protection of Flora and Fauna), EFI = Empresa Forestal Integral
(Integrated Forestry Agency). CGB = Cuerpo de Guardabosques (Forest Ranger Corps). AP = Protected Area,
t = not confirmed.

year. Methods used for counts were those
developed by the International Crane Foundation
and used in Wisconsin (Meine and Archibald
1996) and Mississippi (Hereford et al. 2001) for
other subspecies of Sandhill Cranes. Counts were
conducted by systematically establishing obser-
vation points 1 .5-2 km apart throughout the entire
area to be sampled. Ground counts conducted in
the four areas with the largest numbers of cranes
consisted of between 16 and 40 points. Two to
four observers at each point recorded the number
of cranes observed, time of observation, direction
from observation point, and direction of flight.
Between 300 and 500 persons participated in the
counts at each location surveyed. Two hours of
observation were conducted before sunset during
an evening followed by 2 hrs of observation the
next morning beginning at sunrise.

We recorded number of birds seen, time of

observation, and direction of flight at each sample
location to eliminate duplicate observations. The
maximum number of birds, after eliminating
duplicate counts, was considered the total for that
area.

RESULTS

Distribution and Abundance. — We defined the
distribution of Cuban Sandhill Cranes as consist-
ing of 10 separate localities and presumed to
consist of nine separate populations in six
provinces (Table 1, Fig. 1). We estimated 526
cranes for the entire country. The total cranes
counted in each of the areas with largest numbers
of cranes varied from 71 in Guayaberas to 171 on
the Isle of Youth (Table 2) for a total of 464
cranes in the four areas with the largest popula-
tions. We estimated all other sites combined to
have 62 cranes (Table 1) based on site visits.

FIG. I. Distribution of Sandhill Crane populations in Cuba. Numbers represent the location of populations
corresponding with data in Table 1.

Galvez Aguilera am! Chavez- Ramirez • CUBAN SANDHILL CRANE STATUS

559

TABLE 2.

Estimates of Sandhill Crane

abundance in the four largest populations in

Cuba.

Categor>’

Isle of Youth

Cienega de Zapata

Guayaberas

Moron Norie

Survey period

February 1998

February 1999

February 1998

February 2(K)2

# Count stations

52

26

16

32

Cranes obs. /point

5.3

6.3

9.1

6.5

Range

1-22

1-21

4-16

2-18

Estimated total cranes

171

120

71

102

ground-based surveys, direct observations, and
knowledge ot local people. We confirmed crane
presence for Jucaro (#7, Fig. 1) but have no
estimate of abundance.

We could not confirm presence of cranes
during our study or based on local people’s
knowledge, for two locations previously reported
(Walkinshaw 1949). These included the region of
Consolacion del sur in Pinar del Rio Province and
Cienega de Birama in Granma Province (#’s 2 and
10, respectively; Fig. 1 ). Only a rough estimate of
numbers is available for two other localities due to
the rarity of crane observations in those areas.
These localities are Cienega de Majaguillar in
northern Matanzas Province and Cayo Romano
key, off the northern coast of Camaguey Province
(#’s 3 and 9, respectively; Fig. 1).

Provincial Reports

Pinar Del Rio Province. — Interviews with 30
local residents suggested crane presence in a small
area over the entire time period they could
remember. We confirmed crane presence in
1996 in the area of Guane (#1, Fig. 1) with no
more than 10 individuals likely remaining. Locals
suggested cranes were more common through the
1990s but had decreased significantly in recent
years. Large areas were planted with Carribean
pine (Pinus caribaea) in the 1970s, which
apparently caused many areas to become wooded
and less suitable for cranes. Our last visit to the
area in 2002 confirmed the presence of one chick;
however, even fewer cranes were apparent than in
1996.

Threats to cranes in this province include
habitat modification, presence of large agricultur-
al operations including hog {Siis scrofa) farming,
and extensive hunting activities, some of which
are reported to be illegal. The area where cranes
are present is under management and protection
by Forestry Agency authorities, and hunting has
been limited due to concerns expressed to
authorities regarding the critical situation for

cranes in the area. However, the low number of
cranes present and the small area of habitat
available will make it difficult for cranes to
increase in this locality in the future.

Matanzas Province. — Cienega de Zapata (#4,
Fig. 1) in the southern portion of the province has
long been known to support cranes. Crane
presence was confirmed from aerial surveys in
1995 and, in 1996, five nests were found in
different sections of the wetlands. We estimated
120 cranes during the ground point counts
conducted in Cienega de Zapata in 1998. We
recorded a new locality in the northern area of the
province not previously reported as supporting
cranes called Cienega de Majaguillar (#3, Fig. 1)
with an estimate of <12 cranes. The situation of
these cranes appears to be similar to that of the
precarious population in Pinar del Rio.

There are several threats to cranes in this area.
Thei'e is loss of open savannah due to increases in
woody shrubs and trees such as cayput (Galaleuca
quinquenerva), believed to be a consequence of
decreased use of prescribed fires since the 1 970s.
Significant populations of wild hogs and dogs
(Canis lupus familiaris) are present and could
threaten Sandhill Cranes.

Sancti Spiritus Province. — Sandhill Cranes are
present in the Cienega de Guayaberas and,
according to those interviewed, have been present
for at least 40 years (#5, Fig. 1 ). A ground point
count conducted in 1998 estimated 71 cranes. The
area suitable for cranes is <12 knr and is
decreasing due to invasion by exotic woody
plants such as marabou (Dichrostachys glomer-
ata), which eliminates open spaces. Large areas
have also been modified for rice cultivation and
this practice continues to increase. The presence
ot water buffalo [Buhalus huhalis) has caused
overgrazing in some areas further contributing to
invasion of exotic woody plants.

Ciego de Avila Province.— SandhiW Cranes
have been reported in this province since the
beginning of the 20th century. Crane presence was

560

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. 3, September 2U 10

confirmed via aerial and ground surveys in 1995
and 1996 in the northern portion of the province
(#6, Fig. 1), but not in the south (#7, Fig. 1). We
did not obtain a count of cranes in the south but
did confirm their presence. We estimated 102 and
101 cranes were present in the northern area of the
province based on point counts conducted in 1997
and 2002, respectively. Chicks were observed in
this area in 1995 and 1996. Northern Ciego de
Avila supports one of the largest populations of
cranes in Cuba and is a priority area for
protection. The case for protection was made to
government authorities. A 250,000-ha protected
area was established in summer 2004 that will
limit and control expansion of habitat alterations
in the entire region. However, agriculture, cattle
grazing, and tree plantations continue and could
affect crane habitat in the future.

Potential threats in this area include extensive
livestock grazing, exotic plants and animals, and
habitat fragmentation due to agriculture and tree
plantations. Cattle grazing is expanding and there
is an effort to establish non-native grasses for
livestock forage. Feral water buffalo are present,
which has led to overgrazing and erosion in some
areas. Invasive plants include casuarina {Casiia-
rina equisetifolia) and marabou, which are rapidly
increasing and causing reduction of open savan-
nahs. The hydrological regime of this area has
been affected by deforestation of native plants,
road construction, and sugar cane fields, which
could flood crane nests and drown chicks. The
presence of sugar cane fields provides cover for
mongoose (Crossarchus spp.) and black rats
(Rattiis rattus), which depredate crane eggs.

Ccwuigiiey Province. — Historical Sandhill Crane
localities (#8, Fig. 1) in this province included
Sabanas de Lesca and Me.seta de San Felipe.
Surveys of local people suggested cranes were no
longer present but we observed six cranes flying in
1994. At least 12 pairs were observed in a 2001
ground survey in the Meseta de San Felipe. A
previously unreported area was key Cayo Romano
off the northern coast (#9, Fig. 1), which is
believed to have had cranes for at least 30 years.
No extensive count was possible during our work
in this area but biologists who visited the site
estimated that —16 cranes are present on this key.

Principle threats in this area include possible
future mining activities and military exercises
(Cayo Romano) that occur in the area occupied by
cranes. Wild dogs are known to be present in the
area.

Granma Province.— A single recent visual
observation was reported in interviews for this
province in the region of Birama in the delta of
Rio Cauto (#10, Fig. 1). Interviews suggested
cranes were common in the 1970s through the mid
1980s, but decreased in number through the early
1990s and are now believed to be completely
absent. Our ground searches conducted yearly
between 1999 and 2002 could not confirm the
presence of cranes in this area. Habitat modifica-
tions in areas of previous crane use include
development of wetlands for rice production.
We believe cranes have been extirpated from this
province.

Isle of Youth Province. — Isle of Youth is a well
known area for Cuban Sandhill Cranes (#’s 11 and
12, Fig. 1). The type specimen for this subspecies
was collected from the Isle of Youth (Bangs and
Zappey 1905). Two ground counts were conduct-
ed on the Isle of Youth using the same points in
1995 and 1998, yielding estimates of 1 15 and 171
cranes, respectively. The 1995 value is similar to
that reported by Walkinshaw (1953) of 100 cranes
for the Isle of Youth. We believe the 1998 count is
more accurate as it was conducted over 2 days
during one morning and one afternoon, compared
to the 1995 count which was conducted only
during the morning.

Threats present include cattle grazing activity
and herding during the nesting period. Illegal
hunting and fishing activities cause disturbance to
the cranes in this area.

Conservation Status of Populations

Eight of 10 areas supporting Sandhill Cranes
are under some form of protection, either as
protected areas or other schemes managed by
Flora y Fauna. These include Birama, Cayo
Romano, Norte de Moron, Cienega de Guayaber-
as, Cienega de Zapata, Majaguillar, Jucaro, and
Los Indios/Sabana Grande. Two more were under
management by the Forestry Agency in Matanzas
Province (Empresa Integral Forestal de Cienega
de Zapata). Areas where Sandhill Cranes were
present in Guane in Pinar del Rio Province and
Lesca in Camaguey Province were also managed
by the Forestry Agency.

DISCUSSION

We confirmed Cuban Sandhill Cranes to be
more widespread than previously reported (Gar-
rido 1985, Garcia 1987, Meine and Archibald
1996). However, presence of cranes in two areas

Gcdvez Aguilera and Chavez- Ramirez • CUBAN SANDHILL CRANE STATUS

561

that previously supported them (Walkinshaw
1949) could not be confirmed. The abundance of
cranes estimated during this study for Cuba is
higher than all previously reported estimates,
leading to reclassifying the Cuban Sandhill Crane
from critically endangered in 1994 to vulnerable
in 1997. Our estimate of 526 cranes is signifi-
cantly greater than the estimated 103 individuals
of the endangered Mississippi Sandhill Crane
{Gnis canadensis pulla) (Hereford et al. 2001),
but lower than the 4,000-6,000 estimated for
Florida Sandhill Cranes (G. c. pratensis) (Tacha et
al. 1994).

The present distribution is broader than expected,
but at least two previously documented sites appear
to no longer support cranes. We have no explana-
tion for causes of the disappearance of Sandhill
Cranes from these areas, except to note significant
habitat changes have occurred in those areas. A
series of threats was identified for all of the present
populations. At least two populations, those in
Guane in Pinar del Rio and Birama in Granma, are
so low that it is likely they have already disappeared
or will do so soon. The estimates are unreliable for
other localities and are possibly optimistic, and
should be re-evaluated in the field (Cayo Romano
and Sabana de Lesca in Camaguey, and Majaguillar
in northern Matanzas Province).

We could not identify causes of population
disappearance or decreases, but important factors
may include population isolation effects and
degradation and decreases in available habitat.
Decreases in size or fragmentation of habitat
could contribute to population declines in some
areas as alterations are visible throughout the
entire country. The Mississippi Sandhill Crane’s
low productivity and endangered status is attrib-
uted in large part to habitat fragmentation
(Valentine 1970; S. G. Hereford, pers. comm.).
All existing populations are separated by distanc-
es greater than the longest known dispersal
movement by non-migratory Sandhill Cranes (up
to 48.3 km in Florida; Nesbitt et al. 2002). We
have limited data on potential movements and
dispersal of Cuban Sandhill Cranes but, of 10
radio-marked birds on the Isle of Youth, a single
crane moved >18 km during 1 year (Galvez
Aguilera 2002). Even shorter movements (10.5 km
by Florida Sandhill Cranes in Georgia; Bennett
1989) are reported for other non-migratory Sand-
hill Cranes.

Four significant populations appear to be stable
and perhaps increasing; (1) Isle of Youth, (2)

northern Ciego de Avila, (3) Cienega de Zapata in
Matanzas, and (4) Cienega de Guayaberas in Sancti
SpiritLis. Delineation of a new 250,()()()-ha protect-
ed area in 2004 that encompasses all crane areas in
northern Ciego de Avila Province resulted in the
four largest populations being at least partially
within the boundaries of a protected area. Howev-
er, that does not guai'antee protection as several
threats are present regardless of whether they are in
or outside a protected area boundary. Woody plant
encroachment continues in many areas within
Reserves, due to elimination of natural fire
regimes, and is a serious problem of concern for
several federal agencies (FC-R, pers. obs.). Feral
domestic (hogs and dogs) and introduced animals
(mongoose) continue to be threats to nesting cranes
in all four areas; however, the exact effect of each
one of these predators has not been quantified.
Predation is a problem that affects other non-
migratoi'y populations and is the major cause of
mortality for the Mississippi Sandhill Crane (S. G.
Hereford, pers. comm.)

We have been working directly in the Isle of
Youth over the last few years and recently in Ciego
de Avila to better evaluate the specific threats to
cranes and develop appropriate or responsive
management plans. Some Reserve personnel are
trying to implement counts at regular intervals to
document population trends. The trend and conser-
vation status of Cuban Sandhill Cranes may not be
entirely known until we can ascertain whether
numbers are increasing or decreasing over time.
Possible direct actions to enhance declining popu-
lations could include exchange of individuals from
one population to another or from potential captive
breeding efforts. Captive breeding programs for
cranes do not exist in Cuba, but are present and
successful in other countries. Indirect actions to
expand or improve nesting habitat could include
management activities including instituting pre-
scribed fire programs, and elimination and manage-
ment of invasive woody plants, as has been done for
the Mississippi Sandhill Crane (Valentine and
Hereford 1997). Limiting agriculture and forestry
plantation expansion in protected areas where cranes
are present could also help expand or improve
nesting habitats. Control of potential domestic and
introduced predators is another action that could be
taken in areas where predators are a problem.

ACKNOWLEDGMENTS

Many individual.s contributed to this effort. Some of
those that provided support over several years include

562

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 3, September 2010

Miguel Magraner Fernandez, Jose Rivera Rosales, Fidel
Quialia Gongora, Jose Osorio, Vladimir P. Fleites, Jose R.
Rodriguez, Daniel Reit. Jorge Jerez, Amar Perez, Duniet
Marrero, Pavel Martinez Arredondo, Leandro Torrella,
Odey Martinez, and Frank Moreira. We thank the Empresa
Nacional para la Proteccion de Flora y Fauna, territorial
delegations of the Ministerio de Agricultura, the provincial
delegations of CITMA in Sancti Spritus and Ciego de
Avila, Cuerpo de Guardabosques Nacional, The Interna-
tional Crane Foundation, and Jessica Dooley and Luis
Ramirez for assistance with the manuscript. We thank the
many individuals, organizations, and agencies that helped
in a variety of ways. Funding for this project was provided
in part by grants from the Tom D. and Catherine T.
MacArthur Foundation, the National Geographic Society,
and the World Watch Institute. We greatly appreciate the
comments and suggestions by the editor and two anony-
mous reviewers.

LITERATURE CITED

Bangs, O. and W. R. Zappey. 1905. Birds of the Isle of
Pines. American Naturalist 39:179-215.

Barbour, T. 1943. Cuban ornithology. Memoirs of the
Nuttall Ornithological Club Number 9. Cambridge,
Massachusetts, USA.

Bennett, A. J. 1989. Movements and home range of
Florida Sandhill Cranes. Journal of Wildlife Manage-
ment 53:830-836.

Galvez Aguilera, X. 2002. Distribucion y abundancia de
Grus canadensis nesiotes en Cuba. Uso de habitat y
reproduccion de una poblacion de esta specie en la
Reserva Ecologica Los Indios, Isla de la Juventud.
Dissertation. Universidad de la Habana, Cuba.
Galvez Aguilera, X., V. Berovides, and E. Chavez-
Ramirez. 2005. Nesting ecology and productivity of
the Cuban Sandhill Crane on the Isle of Youth, Cuba.
Proceedings of the North American Crane Workshop
9:225-236.

Garcia, F. 1987. Las aves de Cuba, subespecies endemicas.

Tomo II. Gente Nueva, La Habana, Cuba.

Garrido, O. H. 1985. Cuban endangered birds. Pages 992-
999 in Neotropical ornithology (P. A. Buckley, M. S.
Foster, E. S. Morton, R. S. Ridgely, and F. G. Buckley,
Editors). Ornithological Monographs Number 36.
Garrido, O. H. and F. GarcIa. 1975. Catalogo de las aves
de Cuba. Academia de Ciencias, La Habana. Cuba.

Gundlach, J. 1875. Neue beitrage zur ornithologie Cubas.
Journal fur Ornithologie 23:293-340.

Gundlach, J. 1876. Contribucion a la omitologi'a cubana.
Impresora La Antilla, La Habana, Cuba.

Hereford, S. G., T. E. Grazia, and J. N. Phillips. 2001.
The 2000 Mississippi Sandhill Crane annual popula-
tion census. USDI, Fish and Wildlife Service,
Mississippi Sandhill Crane National Wildlife Refuge
Report. Gautier, Mississippi, USA.

lUCN. 1994. lUCN red list categories. lUCN, Gland,
Switzerland.

Meine, C. D. and G. W. Archibald. 1996. The cranes:
status survey and conservation action plan. lUCN,
Gland, Switzerland.

Nesbitt, S. A., S. T. Schwikert, and M. J. Folk. 2002.
Natal dispersal in Florida Sandhill Cranes. Journal of
Wildlife Management 66:349-352.

POEY, F. 1851-1855. Apuntes sobre la fauna de la Isla de
Pinos. En Memorias sobre la Historia Natural de la Isla
de Cuba, acompanadas de sumarios Latinos y extractos
en Erances, I. Imprenta de Barcina, La Habana: 424-431.

Read, A. C. 1913. Birds observed in the Isle of Pines,
Cuba. Oologist 30:130-131.

Tacha, T. C., S. a. Nesbitt, and P. A. Vohs. 1992.
Sandhill Crane {Grus canadensis). The birds of North
America. Number 3 1 .

Tacha, T. C., S. A. Nesbitt, and P. A. Vohs. 1994.
Sandhill Crane. Pages 76-94 in Migratory shore and
upland game bird management in North America
(T. C. Tacha and C. E. Braun, Editors). International
Association of Eish and Wildlife Agencies, Washing-
ton, D.C., USA.

Valentine, J. M. 1970. A colony of Sandhill Cranes in
Mississippi. Journal of Wildlife Management 34:761-
768.

Valentine, J. M. and S. G. Hereford. 1997. History of
breeding pairs and nesting sites of the Mississippi
Sandhill Crane. Proceedings of the North American
Crane Workshop 7:1-9.

Walkinshaw, L. H. 1949. The Sandhill Crane. Bulletin
Number 29. Cranbrook Institute of Science, Bloom-
field Hills, Michigan, USA.

WALKIN.SHAW, L. H. 1953. Nesting and abundance of the
Cuban Sandhill Crane on the Isle of Pines. Auk 70: 1-10.

Walkinshaw, L. H. and B. W. Baker. 1946. Notes on the
birds of the Isle of Pines, Cuba. Wilson Bulletin
58:133-142.

The Wilson Jot<r?ial of Ornithology 122(3):563-568, 2010

THE FEATHER STRUCTURE OF DIPPERS:

WATER REPELLENCY AND RESISTANCE TO WATER PENETRATION

ARIE M. RIJKE' 2 AND WILLIAM A. JESSER'

ABSTRACT. — Dippers (Cinclidae) are the only passerines that find their prey almost exclusively under water. We
examined the breast feathers of the five species of dippers for barb diameter and spacing. These results were compared with
those measured for other bird families, both aquatic and terrestrial, to detect the presence of any adaptation in the structure
of their plumage to their aquatic lifestyle. Barb diameters (2r) were about the same for all dipper species at ~25 pm, but
spacing {2d) varied from ~ 150 to 215 pm. Dipper feathers exhibited large values for (r+d)/r, the structural requirement for
excellent water repellency, comparable to that of many other tenestrial birds including starlings (Stumidae), swifts
(Apodidae), and nightjars (Caprimulgidae). A slightly improved resistance to water penetration, indicated by a smaller
value for (r + d)/r, was observed for White-throated Dipper {Cinclus cinclus), whereas increased water repellency was
found for both American (C. mexicanits) and Brown dippers (C. pallasii). Each of these three species is known to dive. No
adaptations were observed for White-capped (C. leucocephalus) and Rufous-throated (C. schuizi) dippers, neither of which
dive. The breast feathers of dippers have evolved excellent water repellency, but diving dippers show further adaptations for
improved resistance to water penetration and water repellency to allow under-water foraging. Received 2 November 2009.
Accepted 8 March 2010.

Dippers are the only passerines that find their
prey almost exclusively under water. The White-
throated Dipper {Cinclus cinclus). Brown Dipper
(C. pallassi), and American Dipper (C mexica-
nus) habitually plunge into shallow streams and
lakes to search for food among rocks, boulders,
and vegetation whereas the White-capped Dipper
(C. leucocephalus) and Rufous-throated Dipper
(C. schuizi) only dip their heads under water to
catch prey detected from their rocky perch. All
species, however, are exposed to continuous spray
and traverse sheets of falling water with surprising
abandon (Ormerod and Tyler 2005).

That feathers of aquatic birds are well adapted
to specific foraging habits has been known for
some time (Rijke and Burger 1985). For example,
the water repellency of feathers, i.e., the ability of
water drops to ‘pearl’ and roll off, is reduced
relative to that of terrestrial birds. Further,
resistance of feathers to water penetration, i.e.,
the pressure required to force water between the
barbs and barbules, is often increased. The extent
of both water repellency and resistance to water
penetration can be ascertained by measuring the
diameters of the barbs and barbules and their
respective separations, which have been quanti-
fied for ~ 1 50 aquatic and terrestrial species of 45
families (Rijke 1970, Elowson 1984).

Water repellency can be conveniently ex-

‘ Department of Materials Science and Engineering,
University of Virginia, 116 Engineer’s Way, P. O. Box
400745, Charlottesville, VA 22904, USA.

^Corresponding author; e-mail: amr@virginia.edu

pressed by the contact angle 0, defined as the
angle between the tangent to the curved water
surface at the point of contact with the solid
surface and the plane of the surface on which the
drop is resting, measured through the water
(Fig. lA). If the surface is not smooth, but rough
or porous, the drops may entrap air in the hollows
and interstices, forming additional air-water
interfaces, which will cause considerable increase
in the contact angle (Fig. IB), according to:

COS0a=/iCOS0— /2 (1)

where f\ is the area of solid- water interface and/2
is the area of the water-air interface per unit of
apparent surface area. For water drops on barbs,/)
and /2 can be expressed (Cassie and Baxter 1944)
as:

f^={n-Q)r/{rA-d) {la)

and

f2 = \—rsinQ/{r + d) {2b)

where 2r represents the diameter of the barbs
measured in the plane of the long axes of the barbs
which are separated by distance 2d (Moilliet
1963). The increase in apparent contact angle is
ascertained only by the parameter {r + d)/r and not
by the separate values of r and d. Thus, the
apparent contact angle for values of this parameter
ranging between 2.4 (penguins, Spheniscidae) and
6.0 (ducks, Anatidae) would vary between —126“'
and 146°. These values are significantly higher

563

564

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3, September 20 lU

FIG. 1. Schematic of a drop of water, sufficiently small to not be perturbed by gravitational forces, on a smooth surface
(A) and a porous surface (B) covered with a repellent waxy fdm. A contact angle of 90° is measured on the smooth surface
covered with preening oil. This angle is increased to ~1 15° on the porous surface as a direct result of the formed air-
water interfaces.

than those attained for the most repellent of
smooth surfaces, which equal 114° (Moilliet
1963).

An expression for the pressure {P), required to
force water between the barbs and barbules can be
derived from premises similar to those above
(Moilliet 1963) and is:

P = Y/r|cos9+ \J [(r + r/)/r]^ — sin20| (3)

where y represents the surface tension of water.
Equation 3 shows that P will increase if both rand
(r + d)/r are decreased. The requirement of
relatively small values for (r + d)/r to provide
sufficient resistance to water penetration is
opposed by the need for large values for this
parameter to attain good water repel lency. Thus,
the structural characteristics compatible with
optimal water repellency are, at least in part, in
conflict with the requirements of resistance to
water penetration. Contour feathers can be
expected to exhibit a balance between these two
opposing functions.

Our objectives were to investigate: (I) whether
dippers have contour feathers similar to those of
aquatic or terrestrial bird families in terms of
water repellency and resistance to water penetra-
tion, and (2) if dippers that dive for their prey
have a different feather structure from those that
do not dive.

METHODS

Feathers from the breasts of the five species
of dippers (White-throated: catalog # USNM

412413; Brown: # 582988; American: # 420411;
White-capped: # 412413; Rufous-throated: #
2645 1 1 ) were made available through the coop-
eration of the Division of Birds, U.S. National
Museum (USNM) at the Smithsonian Institution,
Washington, D.C., USA. Only these museum
specimens were used. They were examined under
a Motic BA400 microscope (Motic Instruments
Inc., Richmond, BC, Canada) equipped with a
Jenoptik (Jenoptik AG, Jena, Germany) ProgRes®
C12 digital camera to measure the diameters of
the barbs (2r) and their separations {2d). Photo-
micrographs were made and adjusted using
ProgRes® C12 (Jenoptik AG, Jena, Germany),
Auto-Montage® Essentials® (Syncroscopy, Fre-
derick, MD, USA), and Adobe® Photoshop® CS2
Version 9 (Adobe Systems Inc., San Jose, CA,
USA) software.

The feathers examined were not covered by a
cover slide to not distort the structure and
arrangement of barbs and barbules (Rijke 1970).
Parts in which the hamuli of the barbules had
become unhooked and other elements that had
obviously suffered from damage or abrasive
effects were ignored. Only breast feathers were
examined because these had been previously
found to be representative of contour feathers,
whereas rectrices and remiges, as a rule, do not
reflect the water shedding capabilities, of body
feathers (Rijke et al. 1989, 2000).

At least five measurements were recorded for
statistical analysis from the mid-distal section of
the dorsal aspect of each feather. Five or more
breast feathers from one individual of each
species were measured with the total number of

Rijkc amUcsscr • FEATHER S TRUCTURE OF DIRI^ERS

565

TABLE I. Barb diameters, spacing and values for (/■ + d}/r for breast feathers of dippers compared to other groups
(Rijke and Burger 1985).

n

2r

SD

2d

SD

(r + dj/r

SD

A/ik hyUroslalic

jam

jam

111'

White-throated

30

24.3

0.9

149

8

7.1

0.2

0.084

Brown

51

23.6

1.0

214

9

10.1

0.2

0.060

Rufous-throated

30

23.6

0.8

160

4

7.8

0.1

0.077

American

34

24.1

0.7

186

3

8.7

0.1

0.073

White-capped

45

26.6

0.9

185

12

7.9

0.4

0.074

Starlings

34

206

7.1

0.061

Swifts

27

182

7.7

0.071

Nightjars

26

181

7.9

0.072

Meters.

measurements n for each species ranging between
30 and 51. Previous work has shown that variation
in barb diameter and spacing among individual
birds and between males and females of each
species is not statistically significant (Rijke 1970,
Elowson 1984). The arithmetic mean of these
measurements and their standard deviations (SD)
are reported (Table 1 ).

RESULTS

Photomicrographs of the breast feathers of all
five species of dippers are quite similar to the
naked eye with barbs and barbules clearly
delineated (Fig. 2). They are also similar in
appearance to scanning electron microscope
(SEM) micrographs of feathers of Reed Cormo-
rant (Phalacrocorax africamis), Double-crested
Cormorant (P. auritus). Brown Pelican (Peleca-
inis occidentalis), and Mallard {Anas platy-
rhynchos). Only after measuring the barb diam-
eters and their separations did the differences in
microstructure become apparent. The photomi-
crographs also illustrated the parallelism of the
barbs in regions where barbules are normally
zipped up and their divergence where the hooks
and flanges of the barbules have become undone.
These latter regions were discarded in our
measurements.

The dimensions of the dippers’ breast feathers
and the pressures P, required to force water
through the feather as calculated using equation 3
for 0 = 90°, were compared with data from earlier
studies (Table 1). The barb diameter 2r of the five
species varied little among species and was
essentially the same at ~25 pm. The spacing
between barbs 2d varied, ranging from —149 pm

for the White-throated Dipper to 214 pm for the
Brown Dipper. However, no correlation was
detected between the little variation in barb
diameter and the larger variation in spacing. The
corresponding values for (r + d)/r ranged from 7. 1
to 10.1. Their SD values indicate statistically
significant differences between species, but with
overlap for White-capped and Rufous-throated
dippers (Table 1 ).

DISCUSSION

The values measured for 2r, 2d, and {r -H d)/r
indicate that all five dipper species exhibit
excellent water repellency comparable with that
of typical terrestrial birds like swifts and nightjars.
Spacing between the barbs in all dipper feathers
studied was slightly smaller near the rachis and
wider near the periphery of the feather, in
agreement with previous findings for other birds
(Rijke 1970). This feature is thought to increase
the value of (r 5- d)/r distally as a possible further
adaptation to shedding water (Kennedy 1970,
1972).

Unfortunately, no data from studies by other
authors are available in the literature. Only
Elowson (1984:375), using SEM micrography,
studied the feathers of one dipper species (C.
cincliis) as part of a larger study on spread-wing
postures and found that “the barbule and ramus
structure was so very different from the 14 non-
passerine species... .that it was not possible to
measure the r and d dimensions....’’. Elow.son did
not provide a SEM micrograph of this dipper
specimen, but it is possible that differences in
measuring technique are the source of di.screpancy
with the pre.sent study. This may also explain why

566

THE WILSON JOURNAL OL ORNITHOLOGY • Vul. 122. No. 3, September 2010

White-capped

White- throated

Brown

throated

f'lG. 2. Photomicrographs of the breast feathers of the five species of dippers. All micrographs arc displayed at the
same magnification having calibration bars of length 300 pm.

her (r -i- d)/r value for the Mallard (—10) is very
different from tho.se of Cassie and Baxter (1944),
Rut.schke (1960) and Rijke (1968), all of whom
found 5.9 by transmission light microscopy.

The value for the parameter (r -b d)/r varied

among the five dipper species resulting in
relatively increased resistance to water penetra-
tion in the White-throated Dipper and increased
water repel lency for American and Brown dippers
(Fig. 3). The calculated values for the resistance

Rijke cinclJcssey • FEATHER SIRUCTURE OF DIPPERS

567

including the five species of dippers. A best-fitting straight
line is drawn for the truly aquatic and truly terrestrial
families ( 1, 2, 3, 5, 8 and 11, 12, 13) and is expressed by the
equation 2r = 87.8-7.83(r-f c/)//-. 1 = Penguins, 2 =
Auks (Alcidae), 3 = Divers (Gaviidae), 4 = Gannets, 5 =
Cormorant, 6 = Shearwaters, 7 = American White
Pelican, 8 = Ducks, 9 = Rails, (Rallidae), 10 = Brown
Pelican, 11 = Starlings, 12 = Swifts, 13 = Nightjars.
Dippers; L = White-capped, C = White-throated, S =
Rufous-throated, M = American, and P = Brown.

to water penetration (Pcaic) show the inverse
relationship with the r and fr -f d)/r values that
follows directly from equation 3. They were
calculated based on the diameters and spacing of
the barbs of single breast feathers without
considering the contributions of the barbules or
the effect of stacking multiple breast feathers
(Rijke et al. 1989). Thus, the actual water
resistance of the entire plumage cannot be
ascertained from measurements or calculations
on single feathers. Therefore, the Pcaic values
shown (Table 1) serve as a relative measure
among species.

There are many examples of families, and
species within certain families, that have opti-
mized the balance between water repellency and
resistance to water penetration in terms of barb
diameter and spacing, to suit their particular
habitats and behavioral patterns (Rijke and Burger
1985). Values for 2r and 2d of breast feathers
appear to depend on each other. Specifically, for
truly aquatic and terrestrial families (Fig. 3), an
essentially linear relationship exists between 2r
and (r + d)/r with aquatic species having wide
barb diameters and small values for (/• -i- d)/r, and
terrestrial species having narrow barbs and large
values for (/- -b d}/r. There is no apparent a priori
reason or physical basis for this linearity. These
data (Fig. 3) are useful for comparing a variety of

behavioral patterns with the structural dimensions
of dipper feathers, and tho.se of typical aquatic and
terrestrial families and species including families
that do not obviously belong to either.

An example of the relationship between 2r, {r +
d)/r, and behavioral pattern is provided by the
data point for gannets (Sulidae) which shows a
barb diameter of 50 pm, about the same as that of
cormorants (Phalacrocoracidae) and shearwaters
(Procellariidae), but a smaller value for (/■ -l- d)/r,
indicating a greater resistance to water penetra-
tion. The latter may be an adaptation to the
gannet’s habit of diving from the air and then
pursuing prey under water. The breast feathers of
Brown Pelicans have much smaller barb diame-
ters and higher (/- + d)/r values than those of
American White Pelicans (Peleccinus eryphror-
hynchos). The resulting increased water repellen-
cy may represent an adaptation to its habit of
diving for food from the air. Brown Pelicans,
unlike gannets, however, do not pursue their prey
under water. American White Pelicans find their
prey while swimming when resistance to water
penetration will be the more important, which is
reflected by the smaller values for (r + d)/r. Storm
petrels (Hydrobatidae) with a (r + d)/r of ~7. 1 are
another example. This high value suggests
relatively poor resistance to water penetration,
possibly a reason why storm petrels feed from the
water surface without alighting, but when they do
alight, they float buoyantly high, a condition
consistent with low pressure for water penetration
(Rijke and Burger 1985).

These examples suggest the specific structural
dimensions of contour feathers are an evolution-
ary adaptation to a variety of behavioral patterns.
The relatively high values for (r + d)/r of dippers
suggest their contour feathers, like those of truly
terrestrial birds, have evolved primarily to repel
water. This is reflected by typical dipper behavior:
they move quickly in and out of shallow waters,
rarely remain under water for >15 sec and seldom
dive deeper than ~2 m. Once out of the water, a
quick shake removes any remaining water drops
from their feathers (Attenborough 1998).

The White-capped and the Rufous-throated
dippers have contour feathers that show the same
structural dimensions as contour feathers of
typical terrestrial birds, whereas those of the
White-throated Dipper appear to have developed a
slight increase in water resistance. In contrast, the
American Dipper and, in particular, the Brown
Dipper with their larger (r + d)/r values have

568

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 3, September 2010

attained a substantial increase in water repellency
at some expense of their water resistance. The
first two species, which peck their under-water
prey from a rocky perch and do not dive, show no
difference in feather structure from that of truly
terrestrial birds, whereas the last three are true
plungers with breast feathers that show adapta-
tions to an aquatic lifestyle. The White-throated
Dipper, among these three, has evolved an
adaptation to resisting water penetration; the other
two have developed greater water repellency, in
particular the Brown Dipper. These diverging
adaptations may be explained by differences in
habitat; the White-throated Dipper is more
frequently found along water courses at lower
altitudes, where currents are relatively slow and
the water is deeper. This would favor greater
resistance to water penetration. American and
Brown dippers, in contrast, typically inhabit
higher mountain ranges, where currents are fast
and shallow, which would favor water repellency
over resistance to water penetration. The geo-
graphical range of the White-throated Dipper, in
its most eastern distribution, partly overlaps that
of the Brown Dipper, which has probably evolved
primarily in the isolation of the high Himalayan
mountain ranges. Brown Dippers tend to breed at
altitudes well over 5,000 m, whereas the White-
throated Dipper prefers the 2,000 to 3,500 m
range in Afghanistan and somewhat higher in the
Himalaya. However, there is overlap in altitude,
particularly at the western end of their overlap-
ping ranges and during the non-breeding season
(Ormerod and Tyler 2005).

Adaptations of dipper feathers from their
terrestrial origins to their under-water feeding
habits in terms of both water repellency and
resistance to water penetration are consistent with
our hypothesis including the observation that no
similar adaptations have been made by their non-
diving congeners. It is likely that plunging is a
recent behavior given the dippers’ passerine
origins and that their feathers are primarily water
repellent, and only partially adapted to either
prevent water penetration or extend water repel-
lency. Plunging is apparently, an evolutionary

step that White-capped and Rufous-throated
dippers have not completed.

ACKNOWLEDGMENTS

The authors express their gratitude to the Curators,
Division of Birds, National Museum of Natural History,
Smithsonian Institution, Washington, D.C. for permission
to use dipper feathers for this study. The authors are also
indebted to C. J. Dove for assistance with measurements.

LITERATURE CITED

Attenborough. D. 1998. The life of birds. Video®
Episode 5. BBC Books, London, England.

Cassie, a. B. D. and S. Baxter. 1944. Wettability of
porous surfaces. Transactions of the Faraday Society
40:546-551.

Elowson, a. M. 1984. Spread-wing postures and the water
repellency of feathers; a test of Rijke’s hypothesis.
Auk 101:371-383.

Kennedy, R. J. 1970. Directional water-shedding properties
of feathers. Nature 227:736-737.

Kennedy, R. J. 1972. The probable function of flexules.
Ibis 1 14:265-266.

Moilliet, j. L. (Editor). 1963. Water proofing and water
repellency. Elsevier, New York, USA.

Ormerod, S. J. and S. J. Tyler. 2005. Family Cinclidae
(Dippers). Pages 332—355 in Handbook of the birds of
the world. Volume 10 (J. del Hoyo, A. Elliott, and
D. A. Cristie, Editors). Lynx Edicions, Barcelona,
Spain.

Ruke, a. M. 1968. The water repellency and feather
structure of cormorants, Phalacrocoracidae. Journal of
Experimental Biology 48: 185-189.

Ruke, A. M. 1970. Wettability and phylogenetic develop-
ment of feather structure in water birds. Journal of
Experimental Biology 52:469^79.

Ruke, A. M. and E. H. Burger. 1985. Wettability of
feathers and behavioural patterns in water birds.
Proceedings of the Pan-African Ornithological Con-
gress 6: 153-158.

Ruke, A. M., W. A. Jesser, and S. A. Mahoney. 1989.
Plumage wettability of the African Darter, (Anhinga
melanogaster) compared with the Double-crested
Cormorant, (Phalacroconix aiirifiis). O.strich 60:128-
132.

Ruke, A. M., W. A. Jesser, S. W. Evans, and H.
Bouwman. 2000. Water repellency and feather
structure of the Blue Swallow (Hinmdo atrocaerulea).
Ostrich 7 1 ; 143-145.

Rut.SCHKE, E. 1960. Untersuchungen iiber Wasserfestigkeit
und Struktur des Gefieders von Schwimmvogeln.
Zoologische Jahrbiicher 87; 441-506.

The Wilson Journal oj Ornithology l22(3):569-576, 2010

ANNUAL BIRD MORTALITY IN THE BITUMEN TAILINGS PONDS IN
NORTHEASTERN ALBERTA, CANADA

KEVIN P. TIMONEY'-' AND ROBERT A. RONCONP

ABSTRACT. Open pit bitumen extraction i.s capable of cairsing mas.s mortality event.s of resident and migratory birds.
We investigated annual avian mortality in the tailings ponds of the Athabasca tar sands region, in northeastern Alberta,
Canada. We analyzed three types of data: government-industry reported mortalities; empirical studies of bird deaths at
tailings ponds; and rates ot landing, oiling, and mortality to quantify annual bird mortality due to exposure to tailings ponds.
Ad hoc selt-reported data from industry indicate an annual mortality due to tailings pond exposure in northeastern Alberta
of 65 birds. The self-reported data were internally inconsistent and appeared to underestimate actual mortality. Scientific
data indicate an annual mortality in the range ot 458 to 5,029 birds, which represents an unknown fraction of true mortality.
Government-overseen monitoring within a statistically valid design, standardized across all facilities, is needed. Systematic
monitoring and accurate, timely reporting would provide data useful to all concerned with bird conservation and
management in the tar sands region. Received 17 November 2009. Accepted 5 May 2010.

Global demand for unconventional energy
sources such as coal bed methane, heavy oil,
and bitumen has grown in recent years. Bitumen
in northern Alberta, Canada, is extracted by two
methods, in situ well-based approaches and truck
and shovel open pit mining. The latter method
produces “tails” during separation of bitumen
from the sand. The tails, a mixture of process-
affected water, residual hydrocarbons, brine, silts
and clays, and metals are discharged into tailings
ponds. The extent of tailings ponds in northeastern
Alberta grew by 422% between 1992 and 2008
(Timoney and Lee 2009). The Athabasca tar sands
development is one of the largest energy projects
in the world. Production of bitumen is predicted to
rise from the current 1.3 million barrels/day to
three million barrels/day by 2018 (Alberta Energy
2009).

Water bodies along migration routes attract
many bird species as they afford foraging,
roosting, nesting, and resting opportunities (Ron-
coni 2006). A variety of deterrents have been used
to discourage waterbirds from landing in tailings
ponds such as floating and beach effigies, propane
scare cannons, and sound-producing systems
(Boag and Lewin 1980, Colder Associates Ltd.
2000, Ronconi and St. Clair 2006). Some birds
that land at tailings ponds become oiled and a
proportion of the oiled birds later die. Bird
mortality rates from oiling have not been precisely

'Treeline Ecological Research. 21551 Township Road
520. Sherwood Park, AB T8E IE3, Canada.

^Department of Biology, Dalhousie University, 1355
Oxford Street. Halifax. NS B3H 4J1. Canada.

■’Corresponding author;
e-mail: ktimoney@interbaun.com

measured, but casualties appear to be high for
gregarious species, particularly for diving birds
(Clark 1984). Bird migration is affected by
weather as birds are more likely to land when
they encounter headwinds, low temperatures, and
precipitation (Newton 2007). Storms may increase
the likelihood of bird oiling at tailings ponds
(Ronconi 2006), and inclement weather may
increase the probability of mass mortality events.

Oiled ducks may suffer from reduced insula-
tion, increased metabolic rate, and hypothermia
even from small amounts of oil (Hartung 1967,
McEwan and Koelink 1973). Survival rates of
rehabilitated birds may be as low as 1 to 20% for
some species (Mead 1997). Birds from 43 species
have died due to exposure to tailings ponds in the
area, mostly waterbirds such as dabblers and
divers: Mallard (Anas platyrhynchos). Common
Goldeneye {Bucephala clangula). Northern Shov-
eler {Anas clypeata). Lesser Scaup {Aythya
affinis), American Coot {Fulica americana),
grebes, mergansers, geese, and shorebirds includ-
ing Semipalmated Sandpiper {Calidris piisilla).
Pectoral Sandpiper (C. melanotos). Stilt Sandpiper
(C. himanlopits), and Lesser and Greater Yellow-
legs {Tringa flavipes, T. melanoleiica) (Shaip et
al. 1975, Dyke et al. 1976, Gulley 1980, Ronconi
2006). Deaths of birds of prey, gulls, passerines,
and other groups have also been documented.
Mortality rates may be high even at small ponds:
27 dead birds were found at a 0.4-ha tailings pond
lacking deterrents (Dyke et al. 1976). There may
be continual “incidental take" of birds during the
open water .season, especially at night when
human observations are impractical. Oiled birds
in tailings ponds have been observed to sink out of

569

570

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. 3. September 2010

sight (Dyke et al. 1976), minimizing the chance of
detection.

Our objective was to provide estimates of
annual bird mortality resulting from bitumen
tailings pond exposure in northeastern Alberta,
Canada through synthesis and analysis of avail-
able data. These data included numbers reported
by industry to government and scientific data on
mortality and landing rates at tailings ponds.

METHODS

Study Area. — We studied avian mortality in the
Athabasca bitumen (tar) sands region (geographic
center at 57° 03' N, 111° 31' W, Fig. 1) in the
lower Athabasca River watershed north of the city
of Fort McMurray, Alberta, Canada. The
120.6 km^ of tailings ponds within the area of
open pit mining, as of March 2008, covered 1.4
times the area of natural water bodies (84.9 kni')
composed of the surface of the Athabasca River
(50.4 km-) and lakes, ponds, and other river
surfaces (34.5 km-) (Timoney and Lee 2009; KPT
and RAR, unpubl. data). The area lies within a
convergence zone of North American waterfowl
llyways; millions of birds migrate through
northeastern Alberta en route to and from local
and distant breeding areas in northern Alberta, the
Peace- Athabasca Delta, Mackenzie River Valley,
and the arctic (Butterworth et al. 2002, Thomas
2002, USDI 2009b). Thirty-five species and
species groups of waterbirds, and 29 other species
have been observed on one lease (Syncrude # 17)
at the natural water body Mildred Lake (Sharp et
al. 1975). More than 16,000 birds were observed
flying over one tailings pond during spring
migration (Ronconi and St. Clair 2006) while
more than 25,000 swans, geese, ducks. Sandhill
Cranes (Grus canadensis), and gulls were ob-
served in daylight during a fall migration at
Syncrude Lease # 17 (McLaren and McLaren
1985). The total number of migratory birds
passing through the lower Athabasca River Valley
is unknown.

Data Collection and Analyses.—Spot censuses
and shoreline searches for dead birds of varying
duration, extent, and frequency at tailings ponds
of known areal extent during the open-water
season (Gulley 1980, Van Meer and Arner 1985)
were used to calculate bird mortalities per knr
from which mortalities were adjusted to the 2008
areal extent ot tailings ponds. Mortality data were
collated from three companies with tailings ponds
(Suncor 1990-2008, Syncrude 2000-2007, and

FIG. 1. Study area (modified from Timoney and Lee
2009). Areas (as of 19 March 2008) undergoing bitumen
extraction are hachured; tailings ponds are black. Tailings
pond names are MLSB = Mildred Lake Settling Basin;
ANTP = Aurora North tailings pond; SATP = Shell Albian
tailings pond; TIP = Tar Island Ponds 1 and 1 A; 8AEML =
Suncor Millennium tailings ponds 8A and EML.

Shell Albian 2000-2008). Data were obtained
from reports produced by the companies (Syn-
crude 2008), and from the Alberta government
(Sustainable Resources Development) under a
freedom of information request (K. P. Timoney,
October 2008). These data, reported by company,
year, and mortality type were analyzed to obtain
mean annual mortality. We estimated the total
number of birds landing and subjected to oiling
during spring migration at the Albian Sands
Muskeg River Mine: landings/hr (from Ronconi
and St. Clair 2006) and oiled birds/day (from
Ronconi 2006).

RESULTS

Mortality Rates Estimated by Systematic Sur-
veys.— Systematic surveys for dead birds at
tailings ponds (Table 1), used to calculate mor-
tality per km^ (range 7.2 to 145.2 birds), were
extrapolated to estimate total potential mortality
based on 120.6 km^ of tailings ponds in 2008. An
estimate based on the lowest observed mortality at

Tinumcy ciiul Roiiconi • BIRD MORTALITY AT TAILINGS PONDS

571

TABLE 1. Estimated annual bird mortality/kmVyear in the Athabasca tar sands tailings ponds based on spot counts and
systematic shoreline surveys.

Site

Area (km’)”

Year

Dead

birds

Dead

birds/knr'

Reference’’

Cumment.s

MLSB

12.25

1984

94

7.68

L 3

scare cannons, human effigies

MLSB

1 1.74

1985

84

7.15

1, 3

scare cannons, human effigies

MLSB

1 1.58

1980-1983*’

189.5

16.36

1, 3

scare cannons, human effigies

Pond 1

1.86

1977

77

41.40

2

human effigies; fresh tailings received 95% of
days (Apr-Ocl)

Pond 1

1.86

1978

79

42.47

2

human effigies; fresh tailings received 100% of
days (Apr-Oct)

Pond 1

1.86

1979

270

145.16

2

human effigies with artificial lighting at night;
fresh tailings received 92% of days (Apr-Oct)

Pond lA

0.56

1977

43

76.79

2

deterrents?‘‘; fresh tailings received 34% of days
(Apr-Oct)

Pond lA

0.56

1978

31

55.36

2

deterrents?'*; fresh tailings received 25% of days
(Apr-Oct)

Pond lA

0.56

1979

33

58.93

2

deterrents?'*; fresh tailings received 0% of days
(Apr-Oct)

“ .Xreas of MLSB derived from planimetry of airphotographs ( 1 980, AS2 1 65- L
(1980).

3; 1984, AS3051

-5; 1986, AS3356-280; Ponds 1 and 1 A areas derived from Gulley

Dead birds/year 1980-1983 derived mathematically from reported values for 1984 and 1980-1984 (Van Meer and Arner 1985): 1980-1984 average mortality of
170.4 birds/year; total birds dying 1980-1984 = 170.4 X 5 = 852 birds; 1984 mortality of 94 birds; 1980-1983 average mortality = (852 — 94)/4 = 189.5 birds/
year, or 16.36 birds/km’; the average weighted mean mortality 1980-1985 = ((16.36 X 4) + 7.68 + 7.15)/6 = 13.38 birds/km’. The high estimate of annual mortality
is the weighted mean mortality per km=; it is the sum of 500.38 birds/km’ for 12 years of data (1980-1983 comprises 4 years of data), 500.38/12 = 41.70 birds/km^
References: 1 = Van Meer and Amer (1985); 2 = Gulley (1980); 3 = Colder Associates Ltd. (2000).

Queries sent to Suncor (17 Nov to 12 Dec 2008) regarding deterrents in use on Pond lA during 1977-1979; no reply received to date (4 May 2010).

Syncrude’s Mildred Lake Settling Basin (MLSB)
of 7.2 birds/km- in 1985 yielded an annual
mortality of 863 birds. A medium estimate based
on Syncrude’s MLSB (1980-1985) average mor-
tality of 13.38 birds/km^ yielded an annual
mortality of 1,614 birds. A high estimate based
on the weighted mean mortality rate for all years
at Syncrude’s MLSB and Suncor’ s Tar Island
Ponds 1 and lA of 41.7 birds/km^ yielded an
annual mortality of 5,029 birds.

Industry-based Annual Mortality Reported
to Government. — Annual mortality attributed to
oiling over the period 2000 to 2007 ranged from
17 to 201 birds. The weighted mean (± SD)
annual mortality was 65 ± 59 birds. (Table 2).
Additional annual mortalities attributed to ’other’
and to ‘unknown’ causes (details in Table 2)
averaged (± SD) 13 ± 9 (max 3 1 in 2007) and 16
± 9 birds, respectively. Industry data had poor
agreement with the government data released
under the freedom of information request (Ta-
ble 3); the mean difference was 19%.

Annual Bird Mortality Estimated by Landing
and Oiling Rates. — A spring landing rate of
121.44 birds/day in a 3.5 km- tailings pond was
calculated, resulting in an estimated rate of 34.69
landings/km^/day during daylight hours only (low

estimate. Table 4). We calculated 54.93 landings/
day (109.86 landings/km-/day) (high estimate.
Table 4) from observations at a 0.50-km- area
where deten'ent testing occuired (RAR, unpubl.
data not previously reported in Ronconi and St.
Clair 2006). Scaling for the total area of tailings
ponds in 2008, about 125,513 to 397,408 birds
may land during a 30-day spring migration period.
Thirteen oiled waterbirds and shorebirds were
found during the same period and at the same site
(Ronconi 2006), most of which were covered in
>50% oil, from which oilings/day and the
proportion of landed birds becoming oiled were
calculated (Table 4). We estimate that 286 to 905
birds may be oiled during spring migration at an
overall oiling rate of 0.2278% for birds that
landed on ponds. About 229 to 815 birds may die
each spring due to oiling if an oiled bird is unable
to land more than once, and 80 to 90% of oiled
birds die (Discussion).

DISCUSSION

Uncertainties in Mortality Estimates. — Bird
oilings may peak in August and September rather
than in spring (Van Meer and Arner 1985). and it
is reasonable to double the spring mortality to
derive an annual mortality of ~458 to 1,630 birds.

572

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 3, September 2010

TABLE 2. Bird mortalities attributed to oiling, ‘other’ and ‘unknown’ causes released by the Alberta Government for
petroleum companies with tailings ponds in northeastern Alberta."

Year

Oiling"

Other

Unknown

Suncor

Syncrude

Albian"

Suncor

Syncrude

Albian

Suncor

Syncrude

Albian

1990

103

0

0

1991

93

0

0

1992

194

0

2

1993

135

0

4

1994

87

0

1

1995

43

0

0

1996

72

0

4

1997

71

0

6

1998

80

0

3

1999

48

0

10

2000

193

8

0

2

1

12

7

0

2001

2

15

0

2

0

23

3

0

2002

17

20

1

6

3

1

2

1

2003

15

16

17

2

23

3

2

5

0

2004

10

33

2

0

5

2

2

9

1

2005

3

8

14

2

18

2

1

16

1

2006

3

57

3

4

8

7

3

8

4

2007

9

10

26

1

7

6

6

19

6

2008"

16

4

0

2

1

0

2

4

Recent Mean

± 31.5 ±

20.9 ±

12.4 ±

1.2 ±

8.9 ±

3.0 ±

6.2 ±

8.6 ±

i.6 ±

SD'’

65.5

16.7

10.0

1.4

7.6

2.4

7.7

6.0

2.2

“ 'Other' includes electrocution, collisions, predation, fights with other birds, and natural causes; 'Unknown' includes incidents where company was not able to
identify cause of death and incidents where cau.se of death was not reported.

Tailings pond at Shell Albian began to fill in 2003; mortality due to oiling not expected prior to 2003. Calculations of recent tailings pond mortalities used years
2000-2007 for Suncor and Syncrude and years 2003-2007 for Shell Albian. Mortality means for 'other' and 'unknown' use the period 2000-2007. Mortality means
are for each company and mortality type. Average mortalities by year, 2000-2007. oiling 64.8 ± 58.7. other 13.1 ± 9.3. and unknown 16.5 ±9.1.

Values for 2008 were 'year to date' current to July 2008 with the exception of Syncrude for which the death of 1,606 ducks at the Aurora North tailings pond in
April 2008 was not made public until 2009.

This adjustment to the mortality estimate may be
conservative as it does not include mortalities that
occur before spring, between spring and fall
migration, and after fall migration. Annual
tailings pond mortality estimates derived from
mortality surveys (863 to 5,029 birds) and
landing-oiling rates (458 to 1,630 birds) are
roughly of the same magnitude. Self-reported
oiling mortality data from industry provide the
lowest estimate (65 bird deaths/year) whereas
Wells et al. (2008) provide a high estimate of

8,676 to 156,168 bird deaths/year. Wells et al.
(2008) assumed that all birds that land at tailings
ponds are oiled and that peak landing rates exist
24 hrs/day for 100 days. Our mortality estimates
may be conservative given that 500,000 to one
million birds die annually at oilfield wastewater
ponds in the United States (USDI 2009a). Those
wastewater ponds are similar to bitumen tailings
ponds in their mixture of water, residual oil or
bitumen, and salts.

The presence of extensive tailings ponds

TABLE 3. Annual bird mortality at Syncrude"

as reported by the Alberta government and by Syncrude (2008).

Source

Year

2000

2001

2002

2fX)3

2004

2005

2006

2007

Alberta Government

17

20

28

44

47

42

73

36

Syncrude

20

21

31

44

69

55

46

35

DifferenceL %

-18

-5

-11

0

-47

-31

37

3

“ Combined Mildred Lake and Aurora leases.

Difference = (Govemmeni - Syncrude/Govemment) X 100; mean difference without regard to sign = 19%.

Timoney and Ronconi • BIRD MORTALITY AT TAILINGS RONDS

573

TABLE 4. Rates of landing and the proportion of birds that subsequently become oiled at Shell Albian Sands tailings
pond during April-May 2003.''

Number of
birds landing

Landings/hr

Landings/day'

Oiled birds

Oiliiigs/day

% Landed birds
that became oiled

Ducks

536

3.99

63.69

7

0.149

0.23

Shorebirds

444

3.30

52.76

4

0.085

0.16

Geese/Swans

10

0.07

1.19

1

0.021

1.79

Gulls

13

0.10

1.54

1

0.021

1.38

Other waterbirds

19

0.14

2.26

0

0.000

0.00

Overall

L022

7.60

121.44

13

0.277

0.23

Data compiled from spring migration studies in the 3.5 knr main tailings pond at Shell Albian Sands, Muskeg River Mine {Ronconi 2006. Ronconi and St. Clair
2006) yielding a low estimate of 34.69 landings/km’/day; there were 54.93 landings/day within the 0.50 km* observation area yielding a high estimate of 109.86
landings/knr/day; 47 days of ob.servaiion for oiled birds.

Loons, grebes, cranes, herons, cormorants, and coots.

^ Obser\ations for daylight hours only; average 15.97 hrs of daylight between 18 April and 29 May, 134.4 hrs of observation (source: www.almanac.com/rise for
Fort McMurray, AB, Canada).

containing bitumen, polycyclic aromatic hydro-
carbons (PAHs), naphthenic acids, brine, heavy
metals, and ammonia along an internationally
significant migratory bird comdor poses long-
term threats to migratory and resident birds
(Schick and Ambrock 1974, Wells et al. 2008).
Tailings ponds may pose the greatest threat in
spring when warm effluent-fed tailings ponds
provide open water at a time when natural water
bodies remain frozen; however, a high risk of
oiling may extend throughout the open water
season (Van Meer and Amer 1985).

There are four aspects of our estimates that
influence their accuracy. First, no nocturnal
observations of migrating or landing birds were
made. Many birds migrate at night (Richardson
1971, Blokpoel 1973, Blokpoel and Burton 1973),
but landing rates during darkness are unknown.
Many birds migrate at night and mortality rates
might be higher if data from night-time observa-
tions were available. There are also no observa-
tions for November through early April, when
natural water bodies are frozen but large areas of
tailings ponds remain unfrozen due to addition of
warm tailings. The frequency of landings then is
unknown, but is presumably greater than zero for
resident birds. Higher rates of nocturnal landings
and landings in non-migratory periods would
increase our estimates.

A second source of estimation error is that
numbers of birds flying over, landing, becoming
oiled, or being found dead are an unknown
fraction of the true parameter values. Some of
these parameters were estimated from the most
recent and systematically collected data available
(e.g., Ronconi 2006, Ronconi and St. Clair 2006);
however, without data from other sites and years.

there is no means to assess variation in rates of
birds landing, becoming oiled, or dying.

Third, we assumed a mortality rate of 80 to
90% for birds that came in contact with oil.
Mortality rates of oiled birds are unknown (Clark
1984), but even very small amounts of oil may kill
birds (Hartung 1967, McEwan and Koelink 1973)
and survival rates of rehabilitated oiled birds may
be as low as 1 to 20% (Mead 1997). Our estimates
assume a small proportion (10-20%) of oiled
birds may survive oiling.

Finally, by scaling our estimates from individ-
ual tailings ponds to the areal extent of tailings
ponds in the region, we assumed that bird use and
associated mortalities are similar across sites. This
assumption remains untested and, for some
species such as shorebirds, the extent of shoreline
contaminated with bitumen may be a better
predictor of mortality than extent of open water.

Individual events may result in large variations
in mortality. A migratory waterfowl mortality
event at the Syncrude Aurora North tailings pond
occuiTed in April 2008 at which 1,606 dead
waterfowl were later found (CBC 2008, 2010).
Provided that all dead waterfowl were found and
no non-waterfowl died, the single event resulted
in a mortality of 162 birds/km“ well in excess of
our highest estimate (Table 1). The frequency of
mass mortality events is unknown.

Inconsistencies and Deficiencies in Reporting
Bird Mortality. — We note three major shortcom-
ings in the data provided by government and
industry. First, mortality estimates based on
mortality surveys and landing/oiling rates are far
higher than those reported by government.
Second, industry-reported mortalities often do
not match mortalities reported by the government

574

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 3, September 20 10

(Table 3), even though government and industry
numbers should be identical. Third, the bird
mortality data released by government lack detail.
Only company name and total bird mortality for
each year and general cause of death are reported;
this results in loss of valuable data on location,
date, and circumstances of specific incidents.

Sampling design, including appropriate sample
size, sampling effort, and accurate detection and
identification of species is a critical aspect of an
effective monitoring protocol (McComb et al.
2010). Numbers of bird mortalities are directly
related to search effort and sampling design.
Industry-reported data on bird deaths are prob-
lematic as they are not systematic, repeatable, and
statistically robust. Review of mortality data in
Syncrude annual reports indicates that few of the
ob.servations come from tailings ponds, which is
likely where most oiled birds die. Syncrude (2006,
2008) reported underestimating mortality when
explaining an increasing trend in bird mortality in
recent years as partially attributable to improved
monitoring and reporting practices.

The Need for Improved Data. — Currently,
neither the total number of birds migrating
through the region nor the total annual bird
mortality attributable to tailings ponds are known
with sufficient scientific rigor. Data on mortalities
during extreme weather events are lacking. The
fate of lightly-oiled birds that continue migration,
in particular to summer breeding areas, is
unknown. Important questions remain about the
variability in landing and oiling probability with
season, weather conditions, time of day, and pond
size and location. Questions also remain about
mortality detection efficiency in relation to
sampling effort, monitoring protocols, and tailings
pond size. The ad hoc monitoring by industry,
sanctioned by government, is inconsistent, cannot
answer these questions, and undoubtedly under-
estimates actual mortality.

CONSERVATION IMPLICATIONS

The pace and scale of development of the
Athabasca tar sands is unprecedented in North
American history. The industrial footprint and
resultant habitat loss may double in 15 years and
will certainly increase bird mortality rates. Open
pit bitumen extraction may exert population-level
impacts upon migratory and resident birds, and is
capable of causing mass mortality events. Existing
natural water bodies should be protected to help
offset landings of birds in tailings ponds (Ronconi

2006) . Production of liquid tailings should be
phased-out before this expansion of the industry
occurs.

Tailings ponds exceed the extent of natural
water bodies in the area, continue to increase in
extent, and lie along an internationally significant
fly way; thus, they may pose a significant regional
mortality risk. Harmful effects of tailings ponds
are not limited to oiling of waterbirds. Ingestion
of bitumen grit by waterfowl may be a significant
route of exposure to contaminants (King and
Bendell-Young 2000). Nesting Tree Swallows
(Tachycineta bicolor) exposed to process-affected
wetlands have higher mortality, hormonal stress,
nestling parasitism, and reduced nesting success
(Wayland and Smits 2004, Gentes 2006). The
effects were attributed to PAH exposure, possibly
through feeding on contaminated insects.

A variety of strategies have been tested to
reduce the attraction of birds to industrial
developments (Brough and Bridgman 1980,
Stevens and Clark 1997, Read 1999; and reviews
by Bomford and O’Brien 1990, Donato et al.

2007) . The tar sands industry in northeastern
Alberta has been using landing deterrents such as
propane cannons and scarecrows for >30 years
(Boag and Lewin 1980, Gulley 1980, Colder
Associates Ltd. 2000). One of the long-standing
problems of bird deterrents is habituation (Bom-
ford and O’Brien 1990, Stickley et al. 1995,
Conover 2001). A recent comparison of modem
deten'ent techniques found the odds of landing at
“protected” bitumen tailings ponds remained
high relative to non-deterrent controls (Ronconi
and St. Clair 2006). Overall, these authors
observed no significant difference in the deter-
rence value of industry standard versus radar-
activated systems. The effectiveness of existing
deterrents may be enhanced with development of
compensation ponds (Read 1999, Donato et al.
2007), providing clean water and a positive
stimulus for birds deterred from tailings ponds.

Government should assume responsibility for
development of systematic monitoring and re-
search on tailings pond bird landing, oiling, and
mortality rates. The work should be conducted by
independent scientists using a statistically valid
sampling design with emphasis on spring and fall
migration. A rigorous and systematic monitoring
plan standardized across all facilities is likely to
yield a better understanding of the factors
contributing to avian mortality at tailings ponds
than ad hoc monitoring. These data would be

Timoncy and Romani • BIRD MOR TALITY AT TAILINGS PONDS

575

valuable lo development and rerinenienl of
effective mitigation strategies (e.g., deterrents
and compensation ponds) as a component of an
adaptive management approach towards reducing
avian mortalities. Well-designed monitoring pro-
grams help managers and policy makers to reach
informed decisions based on facts (McComb et al.
2010). Systematic monitoring and accurate, time-
ly reporting would provide data useful to all
concerned with bird conservation and manage-
ment in the tar sands region.

ACKNOWLEDGMENTS

We thank C. C. St. Clair, the editor, and two anonymous
reviewers for helpful comments that improved this
manuscript. RAR was supported by the Killam Trust,
Dalhousie University, during the writing of this paper.

LITERATURE CITED

Alberta Energy. 2009. Oil sands. Alberta Energy,
Edmonton, Alberta, Canada.

Blokpoel, H. 1973. Bird migration forecasts for military
air operations. Occasional Paper 16. Canadian Wildlife
Service, Ottawa, Ontario, Canada.

Blokpoel, H. and J. Burton. 1973. Weather and height of
nocturnal migration in east-central Alberta; a radar
study. Bird Banding 46:31 1-328.

Boag. D. a. and V. Lewin. 1980. Effectiveness of three
waterfowl deterrents on natural and polluted ponds.
Journal of Wildlife Management 44:145-154.
Bomford, M. and P. H. O’Brien. 1990. Sonic detenents in
animal damage control: a review of device tests and
effectiveness. Wildlife Society Bulletin 18:411^22.
Brough T. and C. J. Bridgman. 1980. An evaluation of
long grass as a bird deterrent on British airfields.
Journal of Applied Ecology 17:243-253.
Butterworth, E., a. Leach, M. Gendron, B. Pollard,
and G. R. Stewart. 2002. Peace-Athabasca Delta
Waterbird Inventory Program: 1998-2001. Ducks
Unlimited Canada, Ottawa, Ontario, Canada.
Canadian Broadcasting Corporation (CBC). 2008.
Few survivors after 500 ducks take dip in Alberta oil
sands waste. CBC News, Edmonton, Alberta, Canada.
www.cbc.ca/canada/north/story/2008/04/30/ducks-
follo.html?ref= rss

Canadian Broadcasting Corporation (CBC). 2010.
Syncrude ducks death trial. CBC News, Edmonton,
Alberta, Canada, http://www.cbc.ca/canada/cdmonton/
story/201 0/03/24/f-edm()nton-i ndepth-syncrude-ducks-
trial.html

Clark, R. B. 1984. Impact of oil pollution on seabirds.

Environmental Pollution (Series A) 33:1-22.

Conover, M. 2001. Resolving human-wildlife conHicts:
the science of wildlife damage management. CRC
Press LLC, Boca Raton, Florida, USA.

Donato, D. B., O. Nichols, H. Possingham, M. Moore,
P. F. Ricci, and B, N. Noller. 2007. A critical review

of the effects of gold cyanide-bearing tailings solutions
on wildlife. Environment International 33:974-984.

Dyke, G. R., D. A. Birdsall, and P. L. Sharp. 1976. Tesi
of a bird deterrent device at a tailings pond, Athabasca
Oil Sands, 1974. Professional Paper 1976-1. Syncrude
Canada Ltd., Calgary, Alberta, Canada.

Gentes, M.-L. 2006. Health assessment ol Tree Swallows
(Tachycineta hicolor) nesting on Athabasca oil sands,
Alberta. Thesis. University of Saskatchewan, Saska-
toon, Canada.

Colder Associates Ltd. 2000. Report on oil sands
tailings pond bird deterrent system.s-a review of
research and current practices. Suncor Energy Inc.
(Oil Sands), Syncrude Canada Ltd., and Albian Sands
Energy Inc., Calgary, Alberta, Canada.

Gulley, J. 1980. Factors influencing the efficacy of human
effigies in deterring waterfowl from polluted ponds.
Thesis. University of Alberta, Edmonton, Canada.

Hartung, R. 1967. Energy metabolism in oil-covered
ducks. Journal of Wildlife Management 31:798-804.

King, J. and L. I. Bendell-Young. 2000. The toxicolog-
ical significance of grit ingestion to juvenile Mallard
ducklings. Journal of Wildlife Management 192:181-
193.

McComb, B., B. Zuckerberg, D. Vesely, and C. Jordan.
2010. Monitoring animal populations and their habi-
tats: a practitioner’s guide. CRC Press, Boca Raton,
Florida, USA.

McEwan, E. H. and a. F. C. Koelink. 1973. The heat
production of oiled Mallards and scaup. Canadian
Journal of Zoology 5 1 :27-3 1 .

McLaren, M. A. and P. L. McLaren, 1985. Bird
migration watches on Crown Lease 17, Alberta, Fall
1984. Environmental Research Monograph 1985-2.
Syncrude Canada Ltd., Calgary, Alberta, Canada.

Mead, C. 1997. Poor prospects for oiled birds. Nature
390:449-450.

Newton, I. 2007. The migration ecology of birds. Elsevier
Ltd., New York, USA,

Read, J. L. 1999. A strategy for minimizing waterfowl
deaths on toxic ponds. Journal of Applied Ecology
36:345-350.

Richardson, W. J. 1971. Spring migration and weather in
eastern Canada: a radar study. American Birds
25:684-690.

Ronconi, R. a. 2006. Predicting bird oiling events at oil
sands tailings ponds and assessing the importance of
alternate waterbodies for waterfowl: a preliminary
assessment. Canadian Field-Naturali.st 120:1-9.

Ronconi, R. A. and C. C, St. Clair. 2006. Efficacy of a
radar-act ivatcd on-demand system for deterring wa-
terfowl from oil sands tailings ponds. Journal of
Applied Ecology 43:1 1 1-1 19.

Schick, C. D. and K. R. Ambrock. 1974. Waterfowl
investigations in the Athabasca Tar Sands Area.
Canadian Wildlife Service, Ottawa, Ontario, Canada.

Sharp, P. L., D. A. Bird.sai.l, and W. J. Richardson.
1975. Inventory studies of birds on and near Crown
Lea.se Number 17. Athabasca Tar Sands, 1974.
Environmental Re.search Monograph 1975-4. Syn-
crude Canada Ltd., Calgary, Alberta, Canada.

576

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 3. September 2010

Stevens, R. G. and L. Clark. 1998. Bird repellents:
development of avian-specific tear gases for resolution
of human-wildlife conflicts. International Biodeterio-
ration and Biodegradation 42:153-160.

Stickley, a. R., D. F. Mott, and J. O. King. 1995. Short-
term effects of an inflatable effigy on cormorants at
catfish farms. Wildlife Society Bulletin 23:73-77.

Syncrude. 2006. Annual report of oil sands development
in 2005 and projected for 2006, Mildred Lake Oil
Sands Mine. Alberta Environment, Edmonton, Al-
berta, Canada.

Syncrude. 2008. 2007 Annual reclamation progress
tracking report, Mildred Lake and Aurora North oil
sands mines. Alberta Environment, Edmonton, Canada.

Thomas, R. 2002. An updated, provisional bird inventory
for the Peace-Athabasca Delta, northeastern Alberta.
BC Hydro, Burnaby, British Columbia, Canada.

Timoney, K. P. and P. Lee. 2009. Does the Alberta tar
sands industry pollute? The scientific evidence. The
Open Conservation Biology Journal 3:65-81.

U.S. Department of Interior (USDI). 2009a. Migratory
bird mortality in oilfield wastewater disposal facilities.
USDI, Pish and Wildlife Service, Cheyenne, Wyo-

ming, USA. http://www.fws.gov/mountain-prairie/
contaminants/contaminants 1 b.html

U.S. Department OF Interior (USDI). 2009b. Waterfowl
breeding population and habitat survey strata esti-
mates, Strata 13-18, 20, and 77. USDI, Pish and
Wildlife Service, Patuxent, Maryland, USA. http://
mbdcapps.fws.gov/

Van Meer, T. and B. Arner. 1985. Bird surveillance and
protection programme, summary of 1984 and 1985
activities. Syncrude Canada Ltd., Calgary, Alberta,
Canada.

Wayland, M. and J. Smits. 2004. The ecological viability
of constructed wetlands at Suncor: population and
health-related considerations in birds. Task 5. Pages
48-61 in Northern Rivers Ecosystem Initiative: collec-
tive findings. Assessment of natural and anthropogenic
impacts of oil sands contaminants within the Northern
River Basins (F. M. CONLY, Compiler). Environment
Canada, Saskatoon, Saskatchewan, Canada.

Wells, J., S. Casey-Lefkowitz, G. Chavarria, and S.
Dyer. 2008. Impact on birds of tar sands oil
development in Canada’s boreal forest. Natural
Resources Defense Council, New York, USA.

The Wilson Journal of Ornilhology 122(3):577, 2010

ABUNDANCE AND DISTRIBUTION OF WATERBIRDS IN THE

ELANOS OF VENEZUELA

FRANCISCO J. VILELLA,'^ MARK S. GREGORY AND GUY A. BALDASSARRE-

ABSTRACT. — The Llanos is a significant waterbird site in the Western Hemisphere, but abundance and distribution of
waterbirds across this vast region are poorly known, which hampers conservation initiatives. We used point counts along
road routes in the Llanos region of Venezuela to examine abundance and distribution of waterbirds during 2000-2(X)2
within five ecoregions across the Llanos. We detected 69 species of waterbirds and recorded 283,566 individuals, of which
10 species accounted for 80% of our observations. Wading birds (Ciconiiformes) represented the largest guild both in
numbers of species (26) and individuals (55%), followed by waterfowl (26%), and shorebirds (11%). Five species
comprised 62% ot all individuals: Cattle Egret (Bnhulciis ibis). White-faced Whistling Duck (Dendrocygna vidiiata).
Black-bellied Whistling Duck (D. autiimnalis). Great Egret (Ardea alba), and Wattled Jacana (Jacana jacana). Wading
birds were particularly ubiquitous with at least 21 of 26 species recorded in each of the ecoregions. Species richness (66),
proportion of waterbirds detected (54%), and mean number of birds per route (1,459) were highest in the Banco-Bajio-
Estero savanna ecoregion. Our study provides the most comprehensive data set available on waterbirds in the Llanos of
Venezuela and highlights regions of special conservation concern. Received 20 April 2009. Accepted 7 October 2009.

' U.S. Geological Survey, Cooperative Research Unit,
Department of Wildlife and Fisheries, Mississippi State
University, Mississippi State, MS 39762, USA.

‘ SUNY College of Environmental Science and Forestry,
Syracuse, NY 13210, USA.

^ Current address: Fundacion Jardln Botanico de Merida,
Apartado 52, Merida 5101, Venezuela.

■* Corresponding author; e-mail: fvilella@cfr.msstate.edu

Erratum

The second author’s name was inadvertently
left off of the article published in the March 2010
Issue (Volume 122: 102-115). We apologize for
this error. The corrected author sequence is shown
above.

Short Communications

The Wilson Journal of Ornithology 122(3):578-582, 2010

Piping Plover Foraging Distribution and Prey Abundance in the

Pre-laying Period

Jonathan B. Cohen'-^ and James D. Fraser*

ABSTRACT. — Migratory birds arriving in breeding
areas should select territories that maximize reproduc-
tion and survival. Prey available prior to egg laying may
be as important as prey availability for chicks later in
the season. We sampled benthic and terrestrial prey
items in Piping Plover (Charadrins melodiis) foraging
habitat soon after they arrived in breeding areas in New
York during 2001-2003. Benthic invertebrates in the
sand flats were abundant and available to adults,
whereas terrestrial arthropods typically used later in
the season were sparse in all cover types. Foraging
adults selected intertidal sand Bats over other micro-
habitats. One benefit of nesting near sand flats
apparently is abundant food upon arrival in breeding
areas. Protecting habitat between arrival in breeding
areas and territory establishment is uncommon but
warranted for this threatened species. Received 4
September 2009. Accepted 17 March 2010.

Choice of a breeding territory is a key
determinant of avian fitness and is made based
on proximate cues that, in good habitat, will be
correlated with fitness benefits (Fretwell and
Lucas 1970). These cues may be related to cover
from predators (Benson et al. 2009), intraspecific
social attraction (Harrison et al. 2009), nest site
availability (Rhodes et al. 2009), and food
availability (Earnst and Rothe 2004). Food
abundance during egg-laying may be critical both
to current and future avian reproductive success
via effects on adult energy reserves (Martin 1987).
Lack (1954) proposed that breeding should be
timed so chicks should hatch at the time of
greatest food abundance. However, habitat quality
during territory establishment (i.e., between the
start of the breeding season and egg-laying) and
during brood-rearing may not be correlated
(Nooker et al. 2005). Thus, adults cannot forecast
availability of food for chicks by prey abundance
leading up to and during egg-laying (Kirstan et al.
2007).

' Department of Fisheries and Wildlife Sciences, Virginia
Tech, Blacksburg, VA 24061. USA.

^Corresponding author; e-mail: jocohenl@vt.edu

Selecting habitat that does not provide adequate
food for chicks is not a viable strategy, but food
availability for the young may not necessarily be a
proximate cue for decisions about territory
location and size. Prey used by adults and chicks
during the breeding season may be entirely
different for some bird species, (Wilson et al.
2004) or partially overlap (Shealer 1998, Jiguet
2002). Diets of adult and chick Eurasian Dotterels
(Charadrius morinellus) differed early in the
breeding season and became more similar as
soft-bodied prey became more prevalent (Gal-
braith et al. 1993).

Precocial birds may benefit more than altricial
species from settlement decisions based on habitat
conditions at time of teiritory establishment,
rather than on a forecast of future conditions,
because they can move their broods to high-
quality foraging habitat soon after hatch (Koszto-
lanyi et al. 2007). Our objective was to examine
the types and distribution of prey available to
Piping Plover (Charadrius melodus) adults upon
arrival in breeding areas. Foraging adults tend to
concentrate in intertidal flats upon arrival at East
Coast breeding sites where those habitat features
are available (Fraser et al. 2005), whereas after the
breeding season chicks forage in a range of upland
and intertidal habitats (Elias et al. 2000). We
predicted that intertidal flats would provide
abundant prey between arrival and egg-laying.

METHODS

Our study area was West Hampton Dunes, New
York, USA (40 46.5' N, 72 43' W) from 2001 to
2003. This 2.8-km long barrier island village was
200-500 m wide, bounded by the Atlantic Ocean
to the south and Moriches Bay to the north. The
ocean shoreline of the island was characterized by
a relatively linear high-wave energy intertidal
zone with fresh wrack (washed-up organic debris)
at the daily high tide line. The bay shoreline of the
island was a low-wave energy intertidal zone,
including portions of linear shoreline interrupted

578

SHORT COMMUNICATIONS

579

by 27 ha of irregularly shaped sand flats with
fresh wrack at the daily high tide line.

Piping Plover adults arriving to breed on the
Atlantic Coast forage primarily on intertidal sand
Hats and mudflats that may or may not be
contiguous with sandy beaches used for nesting
(Cairns 1982, Fraser et al. 2005). Nesting pair
density is higher at sites that contain sand flats
adjacent to nesting areas than at sites that do not
(Patterson et al. 1991, Elias et al. 2000, Cohen et
al. 2009), suggesting a fitness benefit of nesting
near sand flats. Adults forage on benthic poly-
chaetes and crustaceans (Cairns 1977) and
terrestrial arthropods (Shaffer and Laporte
1994). Adults lead chicks to foraging sites after
the eggs hatch where they prey primarily on adult
arthropods (Cuthbert et al. 1999); chicks generally
do not prey on benthic invertebrates until they are
near fledging. Sand flats provide abundant
arthropods in summer (Cohen et al. 2009), but
nesting Piping Plovers arrive before insects
typically emerge. Thus, prey for chicks cannot
be a proximate influence on territory selection.

We collected 10-cm diameter X 2-cm deep
sediment cores from randomly-selected sites from
the bay side intertidal flats using sections of PVC
pipe. We collected all samples on 1 day, within a
week of the first large annval of Piping Plovers.
Piping Plovers forage on invertebrates at or just
under the surface, and our cores represented the
zone with the most readily available prey.
Sediment samples were stored in 1-L Nalgene
jars filled with 100% ethanol. We later sorted
organisms from the sediment samples and counted
them by general category (polychaete worms,
crustaceans, insect larvae, and other organisms).

We sampled adult arthropods in several cover
types on the ocean and bay side of the island using
paint stirrers coated, except for the handles, with
Tanglefoot Insect Trap Coating (The Tanglefoot
Company, Grand Rapids, MI, USA). These cover
types included intertidal zone (the zone between
the water and the day’s high tide line, which on
the bay side contained the sand flats), fresh wrack
(washed up organic debris from the most recent
high tide), old wrack (clumps of sand-covered
wrack deposited by earlier high tides), backshore
(flat, dry sandy area between the mean high tide
line and the dune, including bare and vegetated
substrates), and dune (a low sandy ridge). We
placed one pair of paint stirrers in each cover type
on each of six transects, the first of which was
placed at random within 100 m of the edge of the

site and the rest were evenly-spaced at 42()-m
intervals. We placed one stirrer vertically in the
sand with the uncoated handle buried and the flat
surface facing the water’s edge. The other stirrer
was horizontal on the ground 10 cm south of the
vertical stick with its long axis parallel to the
water’s edge. The area exposed was 64.5 cm^
(21.5 X 3 cm) for the horizontal stick (coated on
the upper side only) and 1 29 cm^ for the vertical
stick (coated on both sides, Loegering and Fraser
1995). We collected sticky trap samples in 1 day
within 1 week of collecting sediment samples.
The sticks were exposed for 3 hrs between mid-
falling and mid-rising tide, starting between 0700
and 1000 hrs, after which we recorded the number
of arthropods in different taxa.

We performed a census of Piping Plovers in the
study area concuirent with arthropod sampling,
and recorded the numbers that were foraging in
2002 and 2003. We calculated the proportion of
birds in each cover type on the bay and ocean side
of the island, including upland and intertidal
areas. We also measured the area of the intertidal
zones and upland zones (the aggregation of non-
intertidal cover types) from aerial photographs
taken before 5 May each year by tracing polygons
on digitized versions of the photographs in
ArcView 3.1 (ESRI, Redlands, CA, USA). We
then calculated the percent area of those cover
types.

RESULTS

We observed the first Piping Plovers at the site
between 12 and 1 5 March with larger numbers (~
20) arriving between 18 and 21 March. The
earliest recorded nest at the site was on 17 April.

We found low numbers of ten'estrial arthro-
pods, and many sticky traps caught no arthropods
(Table 1). By contrast, there were abundant
benthic invertebrates in the bayside intertidal
zone (Table I ). No .sediment cores were totally
devoid of organisms. The benthos in the cores
comprised mainly polychaete worms, crustaceans,
and insect larvae with the proportions of each
depending on year (Fig. I ).

We observed 93 ± 1% (x ± SE) (/? = 46) of
foraging Piping Plovers in the bay intertidal zone
in 2002, which comprised 32% of the habitat area.
The other 7 ± 1% were using the ocean intertidal
zone which comprised 8% of the habitat area. The
remaining 60% of the area was unused upland.
We only observed seven foraging Piping Plovers

580

THE WILSON JOURNAL OL ORNITHOLOGY ♦ Vol. 122. No. 3, September 2010

Year

■ Polychaetes □ Insect larvae □ Crustaceans

FIG. 1. Mean percent abundance of polychaete worms, insect larvae, and crustaceans in benthic cores from intertidal
sand flats in a Piping Plover breeding area. West Hampton Dunes, New York, 2001-2003.

in 2003, all of which were in the bay intertidal
zone, which comprised 38% of the habitat area.

DISCUSSION

The prey base when Piping Plovers arrived in
breeding areas largely consisted of intertidal zone

benthos. We did not have the resources to sample
and analyze the ocean intertidal zone benthos, but
in casual observations we noted Piping Plovers
foraging on worms there. The plovers were
distributed between two habitats (sand flats and
ocean intertidal zone) that provided benthic prey.
The substrate in upland areas was dry sand and

TABLE 1. Prey organisms in Piping Plover foraging
.samples. West Hampton Dunes, New York, 2001-2003.

habitat, caught in terrestrial sticky trap samples and benthic core

Prey type

Cover type

Date

n

No. >0-'

Organism counts

Min Max Median

Terrestrial

Bay ITZ"

19 Mar 2001

5

0 (0%)

0

0

0

24 Mar 2002

5

2 (40%)

0

5

0

25 Mar 2003

5

0 (0%)

0

0

0

Ocean ITZ

19 Mar 2001

6

0 (0%)

0

0

0

24 Mar 2002

6

1 (17%)

0

1

0

25 Mar 2003

6

0 (0%)

0

0

0

UplantF

19 Mar 2001

40

1 (3%)

0

1

0

24 Mar 2002

43

20 (47%)

0

8

0

25 Mar 2003

48

4 (8%)

0

2 •

0

Benthic

Bay ITZ

18 Mar 2001

6

6 (100%)

3

18

1 1

24 Mar 2002

6

6 (100%)

4

135

43 '

25 Mar 2003

9

9 (100%)

1 1

349

34

" No. >0 = miniber of samples (sticky traps or sediment cores) with at least one prey organism.

Uphmd = pooled samples from all stratii (cover types) at or above the tide line. These include fresh wrack, old wrack, backshorc (bare and vegetated strata), and
dune.

SHORT COMMUNICATIONS

581

would not be expected to host benthos. Terrestrial
arthropods were virtually absent from all zones,
even though large numbers are caught in sticky
traps later in spring and summer (Cohen et al.
2009). Thus, early season prey abundance and
distribution could affect Piping Plover territory
location and size, and could help explain why
nesting pair density of Piping Plovers is high
where sand flats are available (Patterson et al.
1991, Elias et al. 2000, Cohen et al. 2009). There
often is a tidal lag between the bay and ocean side
of baiTier islands as water passes through tidal
inlets, and the presence of bay side intertidal flats
could increase the potential time each day that
benthic organisms are available as prey.

A connection between early season foraging
opportunities and fitness may not require adults to
forecast prey availability for chicks. Reproductive
success in many avian studies decreases with time
from onset of breeding (Martin 1987). Insect
abundance during egg-laying and chick-rearing
were uncon-elated for Tree Swallows (Tachyci-
neta bicolor), but reproductive success was
correlated with insect abundance during egg-
laying by affecting timing of breeding and egg
mass (Nooker et al. 2005). Thus, it likely benefits
female Piping Plovers to nest adjacent to high-
quality foraging habitat that allows them to
quickly enter breeding condition. Maternal nutri-
tion has been shown to influence quality of
offspring of many avian species with offspring
ranging from fully altricial to fully precocial
(Blomqvist et al. 1997, Parker 2002, Reynolds et
al. 2003, Verboven et al. 2003).

Perhaps more importantly in our study area,
intertidal flats usually host a higher abundance of
arthropods in late spring than other cover types
(Loegering and Fraser 1995, Elias et al. 2000,
Cohen et al. 2009). Thus, the same cover type
provides prey for adults upon amval and different
but still abundant prey for chicks later in the
season. However, chicks in .some breeding areas
cannot access intertidal sand flats, and there is a
difference between pre-laying and chick-rearing
foraging areas. Piping Plover broods in Virginia
without access to bay side flats starved, becau.se
the ocean beaches were arthropod-poor and only
chicks that accessed intertidal flats survived
(Loegering and Fraser 1995). Some adults in that
case settled within easy flight range of early-
season foraging habitat but, when their eggs
hatched, there was insufficient food for their
chicks. Sand flats were used by foraging adults in

a site adjacent to our study area in New York, but
were not available to any broods (Cohen et al.
2009). Chick survival was unaffected in that ca.se
(Cohen et al. 2009).

Most strategies commonly used for Piping
Plover conservation focus on enhancing survival
of eggs and chicks to improve productivity
(fledglings produced per breeding pair), and
protections often are established after birds have
settled on their territories (USDI 1996). However,
true productivity is number of recruits into the
breeding population per breeding pair (Martin
1987). The correlation between recruitment and
fledglings produced depends in part on natal site
philopatry which can be a function of density and
habitat quality (Sedgwick 2004, Wrege et al.
2006), and which was low (~ 12%) at our site
(Cohen et al. 2006). Site fidelity of adults and
immigration of new breeders may depend on
habitat conditions during pre-laying (Wrege et al.
2006). Thus, protecting habitat from disturbance
or degradation during the pre-laying period is
important.

ACKNOWLEDGMENTS

The U.S. Army Coips of Engineers, New York District,
provided funding for this study, Amy Alsfeld, Elizabeth
Eorbus, Sarah Gibson, Stephen Hartsfield, Jed Hayden,
Emilie Masiello, Melissa Neely, Dawn Romaine, and
Donald Wardwell assisted with data collection. Steven
Deso assisted with laboratory work.

LITERATURE CITED

Benson, T. J., N. M. Anich, J. D. Brown, and J. C.
Bednarz. 2009. Swainson's Warbler nest-site selec-
tion in eastern Arkansas. Condor 1 1 1:694-705.
Blomqvist, D., O. C. Johansson, and E. Gotmark. 1997.
Parental quality and egg size affect chick survival in a
precocial bird, the Lapwing Vanelliis vanelliis. Oeco-
logia 110:18-24.

Cairns, W. E. 1977. Breeding biology and behaviour of the
Piping Plover (Charadrius melodits) in southern Nova
Scotia. Thesis. Dalhousie University. Halifax, Nova
Scotia, Canada.

Cairns, W. E. 1982. Biology and behavior of breeding
Piping Plovers. Wilson Bullelin 94:531-545.

Cohen, J. B., J. D. Eraser, and D, H. Catlin. 2006,
Survival and site fidelity of Piping Plovers on Long
Island, New York. Journal of Field Ornitholosy
77:409^17.

Cohen, J. B., L. M. Houghton, and J. D. Fraser. 2009.
Nesting density and reproductive success of Piping
Plovers in relation to storm- and human-created habitat
changes. Wildlife Monographs 173.

CUTHBERT, F. J.. B. SCHOLTENS. L. C. WeMMER, AND R.
McLain. 1999. Gizzard contents of Piping Plover

582

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3, September 2010

chicks in northern Michigan. Wilson Bulletin 111:
121-123.

E.arnst, S. L. and T. C. Rothe. 2004. Habitat selection by
Tundra Swans on northern Alaska breeding grounds.
Waterbirds 27:224-233.

Eli.as, S. P., J. D. Fraser, and P. A. Buckley. 2000.
Piping Plover brood foraging ecology on New York
barrier islands. Journal of Wildlife Management 64:
346-354.

Fr.yser. J. D., S. E. Keane, and P. A. Buckley. 2005.
Prenesting use of intertidal habitats by Piping Plovers
on South Monomoy Island, Massachusetts. Journal of
Wildlife Management 69: 1731-1736.

Fretwell, S. D. and H. L. Lucas Jr. 1970. On territorial
behavior and other factors intluencing habitat distri-
bution in birds. Acta Biotheoretica 19:16-36.
Galbraith. H., S. Murray, K. Duncan, R. Smith, D. P.
Whitfield, and D. B. A. Thompson. 1993. Diet and
habitat use of the Dotteral Charadrhis morinellus in
Scotland. Ibis 135: 148-155.

Harrison, M. L., D. J. Green, and P. G. Krannitz. 2009.
Conspecifics influence the settlement decisions of
male Brewer’s Sparrows at the northern edge of their
range. Condor 1 1 1:722-729.

JiGUET, F. 2002. Arthropods in the diet of Little Bustards
Tetrax tetrax during the breeding season in western
France. Bird Study 49:105-109.

Kirstan 111. W. B., M. D. Johnson, and J. T.
Rotenberry. 2007. Choices and consequences of
habitat selection for birds. Condor 109:485-488.
Kosztolanyi, a., T. Szekely, and I. C. Cuthill. 2007.
The function of habitat change during brood-rearing in
the precocial Kentish Plover Charadriits cdexandrinus.
Acta Ethologica 10:73-79.

Lack, D. 1954. The natural regulation of animal numbers.

Oxford University Press, Oxford, United Kingdom.
Loegering, J. P. AND J. D. Fraser. 1995. Factors affecting
Piping Plover chick survival in different brood-rearing
habitats. Journal of Wildlife Management 59: 646-655.
Martin, T. E. 1987. Food as a limit on breeding birds: a
life-history perspective. Annual Review of Ecology
and Systematics 18:453-^87.

Nooker, J. K., P. O. Dunn, and L. A. Whittingham.
2005. Effects of food abundance, weather, and female
condition on reproduction in Tree Swallows {Tacliy-
cineta bicolor). Auk 122:1225-1238.

Parker, T. H. 2002. Maternal condition, reproductive
investment, and offspring sex ratio in captive Red
Junglefowl (Gallus gallus). Auk 119:840-845.

Patterson, M. E., J. D. Fraser, and J. W. Roggenbuck.
1991. Factors affecting Piping Plover productivity on
Assateague Island. Journal of Wildlife Management
55: 525-531.

Reynolds, S. J., S. J. Schoech, and R. Bowman. 2003.
Diet quality during pre-laying and nestling periods
inOuences growth and survival of Florida Scrub-Jay
(Aphelocoma coeridescens) chicks. Journal of Zoolo-
gy, London 261:217—226.

Rhodes, B., C. O’Donnell, and I. Jamieson. 2009.
Microclimate of natural cavity nests and its implica-
tions for a threatened secondary-cavity-nesting pas-
serine of New Zealand, the South Island Saddleback.
Condor 111: 462^69.

Sedgwick, J. A. 2004. Site fidelity, territory fidelity, and
natal philopatry in Willow Flycatchers (Empidonax
trailli). Auk 121:1 103-1 121.

Shaffer, F. and P. Laporte. 1994. Diet of Piping Plovers
on the Magdelen Islands, Quebec. Wilson Bulletin
106: 531-536.

Shealer, D. a. 1998. Differences in diet of chick
provisioning between adult Roseate and Sandwich
terns in Puerto Rico. Condor 100:131-140.

U. S. Department of Interior (USDI). 1996. Piping
Plover (Cluiradriiis melodu.s), Atlantic Coast Popula-
tion. Revised Recovery Plan. USDI, Fish and Wildlife
Service, Hadley, Massachusetts, USA.

Verboven, N., P. Monaghan, D. M. Evans, H. Schwabl,
N. Evans, C. Whitelaw, and R. G. Nager. 2003.
Maternal condition, yolk androgens and offspring
performance: a supplemental feeding experiment in
the Lesser Black-backed Gull (Lcinis fuscus). Proceed-
ings of the Royal Society of London. Series B 270:
2223-2232.

Wrege, P. H., W. D. Shuford, D. W. Winkler, and R.
Jellison. 2006. Annual variation in numbers of
breeding California Gulls at Mono Lake, California:
the importance of natal philopatry and local and
regional conditions. Condor 108: 82-96.

Wilson, L. J., F. Daunt, and S. Wanless. 2004. Self-
feeding and chick provisioning diet differ in the
Common Guillemot Uria aedge. Ardea 92: 197-208.

SHORT COMMUNICATIONS

583

The Wilson Journal of Ornithology 1 22(3):.‘i83-5K7, 2010

Interspecific Song Imitation by a Cerulean Warbler

Than J. Boves,'-* David A. Buehler,' and Phillip C. Massey^

ABSTRACT. — We report a male Cerulean Warbler
(Dendroica cernlea) singing a highly accurate version
of another species’ song (Hooded Warbler; Wilsonia
citrina), as well as a typical Cerulean Warbler song. We
performed playback experiments to examine the manner
in which this bird used, and responded to, the different
songs. He responded aggressively (wing flicks and
garbled song), and with his Cerulean Warbler song, to
recorded versions of two different types of Cerulean
Warbler songs, including a type that he was not
observed singing. He did not respond in any manner
to playback of Hooded Warbler song. Received 5
October 2009. Accepted 2 February’ 2010.

Few cases of high-quality interspecific imita-
tion exist in the wild, outside of natural voice
mimics. Differences in acquired or innate auditory
templates are thought to limit interspecific song
learning (Marler 1976, 1984). Interspecific imita-
tion, when it does occur, typically is between
closely-related species or species that compete for
space and resources (Kroodsma 2004). Members
of the Parulidae, of non-mimicking species, are
among those best known to learn and copy songs
of other species (Payne et al. 1984, Spector 1992,
Martin et al. 1995).

Males in many genera of Parulidae (including
Dendroica) sing two distinct song types that may
.serve different functions (Kroodsma 1981, Spec-
tor 1992). Type I songs, in the case of Cerulean
Warblers (Dendroica certdea), are primarily used
in an intersexual context (e.g., mate attraction)
and are sung more often early in the breeding
season when the male is unmated (Woodward
1997). Type II songs are tho.se that typically have
an intrasexLial function (e.g., territorial defense)
and are sung more often later in the breeding
season, after a mate is secured (Woodward 1997).
We describe a case of a male Cerulean Warbler
singing an indistinguishable version of another

' Department of Forestry. Wildlife, and Fisheries,
University of Tennessee, Knoxville, TN 37996, USA.

■701 White Pine Drive. Hendersonville, NC 28739,
USA.

^Corresponding author; e-mail: tboves@utk.edu

species’ song (Hooded Warbler; Wilsonia citrina),
as well as a typical Cerulean Warbler song. We
also describe the results of playback experiments
performed with this individual.

OBSERVATIONS

On 25-27 June 2009, we observed an after-
second-year, unpaired male Cerulean Warbler in
the North Cumberland Wildlife Management Area
(36° 12' 41.1" N, 84° 16' 55.1" W), in the
Cumberland Mountains of Tennessee, USA, sing-
ing both a Cerulean Warbler song and an accurate
copy of a Hooded Warbler song. We delineated this
individual’s temtory and ascertained his pairing
status based on extensive spot-mapping, nest
searches, and banding efforts. Two other paired
male Cerulean Warblers maintained territories
adjacent to this individual but, to the best of our
knowledge, this individual was unpaired and did
not breed during the 2009 season. We did not
observe him singing a Hooded Warbler song until
25 June, although he had been defending this
territory since early May. The high density of
Hooded Warblers, and the quality of this individ-
ual’s imitation, served to mask this behavior from
our observations until this occasion.

We first ob.served this individual on 25 June
2009 at 1500 hrs singing a Cemlean Warbler song
at a rate of six/min over a 5-min period (Fig. 1 ). At
1505 hrs, he began singing a Hooded Warbler song
at a rate of eight/min over a 1 5-min period (Fig. 2).
At 1525 hrs, he began to sing the Cerulean Warbler
song again for a 10-min period. He continued to
alternate between songs for the next 15 min. We
then attempted to capture the bird using playback
ot a similarly structured Cerulean Warbler song, a
Cerulean Warbler decoy placed at 2 m, and a mist-
net. We placed the speakers at ground level and
adjusted the volume of the speakers to the
approximate intensity level produced by the bird.
The bird immediately responded with a typical
Cerulean Warbler vocal response to conspecific
playback (a garbled, softly sung. Cerulean Warbler
song). He also performed the same Cerulean
Warbler song that he was observed singing earlier.

584 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 3. September 2010

7 500-

7 000

6 500

6 000

5500-

5000

^ 4500-

I

5“ 4 000-

c

0
3

1 3 500-

u.

3 000-

2500-

2 000

1 500

1 000

0 500

—\ r

01 02 03 04 05 06 07

09 1 11 12 13 14 15 16 17 18 19

Time (s)

FIG. 1. Spectroaram of Cerulean Warbler song, performed by the bird in question. This song was indistinguishable by
ear from songs performed by other male Cerulean Warblers. A previously recorded version of this song type was played to
the male Cerulean Warbler in question on 25 June 2009 and he responded aggressively.

Physically, he responded typically as well, with
nervous wing flicks and foraging imitations as he
approached within 5 m of the speakers and decoy.
The bird was caught shortly thereafter, banded,
photographed, and released. He appeared to be a
phenotypically normal, after-second-year male

Cerulean Warbler (based on Pyle 2000), displaying
traits that may be indicative of age and good
condition (e.g., a strong breast band and large
amount of white spotting on his tail; TJB, unpubl.
data). The bird returned to singing a Hooded
Warbler song 4 min after release.

FIG. 2. .Spectrogram of Hooded Warbler song performed by the Cerulean Warbler in question. This song was
indistinguishable by ear from actual Hooded Warbler songs.

SHORT COMMLINICATIONS

585

8 500
6 000
7 500
7 000-
6 500
6 000
5 500-
5 000

f 4000-1

3 500-
3 000
2 500
2 000
1 500-

1 000-
0 500

09 1 11

Time (s)

12 13 14 15 16 17 18 19

FIG. 3. Spectrogram of alternative Cerulean Warbler song type used in playback experiment with the bird in question
on 26 June 2009. He responded to this song type in the same aggressive manner as he responded to the Cerulean Warbler
song type that he was heard singing.

We returned to the location on 26 June and made
sound recordings of both songs using a Dan Gibson
Parabolic Microphone and a M-Audio Microtrack II
digital recorder. We detected the bird at 0800 hrs,
without use of playback, singing the same type of a
Cerulean Warbler song. He sang this song for 15 min
at a rate of six/min. After that, he sang his Hooded
Warbler song for 9 min, at a rate of nine/min.
Following our recording of the bird’s repertoire, we
broadcast a recorded version of a similar Hooded
Warbler song for 10 min and he did not respond,
vocally or physically. We then broadcast an
alternate version of a Cerulean Warbler song with
no decoy (Fig. 3). The bird again responded within
2 min, approaching the speakers and di.splaying
similar aggressive behavior (wing flicks, garbled
song). He then retreated into the canopy and sang his
Cerulean Warbler song (Fig. 1 ). We returned on 27
June to see if he would re.spond to Hooded Warbler
playback and, once again, he did not. We did,
however, observe him again singing the same type
of Cerulean and Hooded warbler songs.

We recorded individual male Cerulean War-
blers earlier in the season for unrelated research
and, after these observations, we recorded neigh-
boring Hooded Warblers for comparison (Figs. 4,
5). We created spectrograms of all songs using
Raven Pro acoustic analysis software (Cornell
Laboratory of Ornithology, Ithaca, NY, USA).

DISCUSSION

This individual likely learned his song reper-
toire from both Cerulean and Hooded warbler
tutors. This would be possible at many locations
in the Appalachian Mountains, as both Hooded
and Cerulean warblers are relatively common in
similar habitats. These two species, however, use
different strata of the forest canopy (Cerulean
Warblers use the overstory and Hooded Warblers
use the understory), belong to separate, well-
resolved genera (Lovette and Birmingham 2002),
and have distinctly different songs. Our observa-
tions differ from the pattern of interspecific
learning suggested by Kroodsma (2004); imitators
usually belong to closely related species or
compete for space and resources. There are
several species of wood-warblers within the same
genus and occupying similar forest strata, which
seem more likely to act as interspecific tutors for
Cerulean Warblers (e.g.. Black-throated Green
[Dendroica virens] or Blackburnian warblers [D.
fusca\).

Our playback experiment results suggest this
individual used, and responded to. Cerulean
Warbler song in an intrasexual (Type 11) manner.
His response to Cerulean Warbler playback (wing
nicks, garbled songs) was typical of teiTitorial
behavior by Cerulean Warblers and, during
playback bouts, he only performed his Cerulean

586

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 3, Septe))}her 2010

10 500-
10 000-
9 500-
9 000-
8 500-
8 000-
7 500-
7000-
6 500-

5 6 000-

c 5 500-

4

D

I 5 000-

u.

4 500-
4 000-
3 500-
3 000-
2 500-
2 000-
1 500-
1 000-
0 500-

0 1

^2 ^ oT ~04 05 06 o'? 08 ^ o'b 1 ^ 77

Time (s)

— , . 1 1 . 1 . j—

12 13 14 15 16

1 9

FIG. 4. Spectrogram of Cerulean Warbler song performed by a neighboring Cerulean Warbler.

Warbler song. This male was unpaired and singing
at a late date (highly unusual for after-second-year
birds in this region), suggesting that his unusual
song repertoire was the cause of his lack of
reproductive success. We were unable to ascertain
how he used his Hooded Warbler song or how he

perceived other Hooded Warbler songs. He did not
respond (physically or vocally) to Hooded Warbler
song, suggesting that he did not use or perceive
Hooded Warbler song in a Type II manner.

This individual may be at a disadvantage when
defending a high-quality territory because his use

N

X

(B

D

CT

7 000:
6 500-
6.000:

5.500-
5.000:

4.500
4 000:

3.500
3 000-

2.500-
2 000-
1 500-
1 000:
0 500

02

— I —

04

— I

06

— I —

0.8

1

Time (s)

— I —

12

— I —

1 4

■ '* I

-■ 1 '

1.6

1,8

FIG. 5. Spectrogram of Hooded Warbler song performed by a neighboring Hooded Warbler.

SHORT COMMUNICATIONS

587

of an interspecific song may increase the chance
ot conspecilic fights, therefore increasing the
costs of defense (Martin et al. 1995). This was the
case during the 2009 season; the territory he
occupied was adjacent to only two conspecific
teiTitories, which is low for appropriate habitat in
the densely-populated Cumberland Mountains.
His teiTitory also had lower than average DBH
and canopy height, similar to areas often unoccu-
pied by (or occupied by younger male) Cerulean
Warblers in the area (TJB, unpubl. data).

In the event this individual returns in 2010, we
will attempt to observe his interactions with
neighboring Hooded Warblers, record changes in
song preference throughout the breeding season,
and observe changes in responsiveness to play-
back of different songs at different times during
the season.

ACKNOWLEDGMENTS

This study was performed in conjunction with the
Cerulean Warbler Forest Management Experiment, which
is funded by the National Fish and Wildlife Foundation,
U.S. Fish and Wildlife Service, the Nature Conservancy,
and the National Council for Air and Stream Improvement.
We thank D. K. Dawson and M. G. Efford for use of the
microphone and recorder. We thank T. M. Freeberg, D. A.
Spector, and an anonymous reviewer for helpful comments
on this manuscript.

LITERATURE CITED

Kroodsma, D. E. 1981. Geographic variation and functions
of song types in warblers (Parulidae). Auk 98:743-751 .

Kroodsma, D. E. 2004. The diversity and plasticity of
birdsong. Pages 108-131 in Nature’s music: the
science of birdsong. Elsevier Academic Press, San
Diego, California, USA.

Lovette, 1. J. AND E. Birmingham. 2002. What is a wood-
warbler? Molecular characterization of a monophyletic
Parulidae. Auk 119:695-714.

Marler, P. 1976. Sensory templates in species-specific
behavior. Pages 314-329 in Simpler networks and
behavior (J. Fentress, Editor). Sinauer, Sunderland,
Massachusetts, USA.

Marler, P. 1984. Song learning: innate species differences
in the learning process. Pages 289-309 in The biology
of learning (P. Marler and H. S. Terrace. Editors).
Springer-Verlag, Berlin, Germany.

Martin, P. R., J. R. Fotheringham, and R. J. Robertson.
1995. A Prairie Warbler with a conspecific and
heterospecific song repertoire. Auk 112:770-774.

Payne, R. B., L. L. Payne, and S. M. Dohlert. 1984.
Interspecific song learning in a wild Chestnut-sided
Warbler. Wilson Bulletin 96:292-294.

Pyle, P., S. N. G. Howell, R. P. Yunick, and D. F.
Desante. 2000. Identification guide to North American
passerines. Slate Creek Press, Bolinas, California, USA.

Spector, D. A. 1992. Wood-warbler song systems: a
review of paruline singing behaviors. Current Orni-
thology 9:199-238.

Woodward, R. L. 1997. Characterization and significance
of song variation in Cerulean Warblers. Thesis.
Queen’s University, Kingston. Ontario, Canada.

The Wilson Journal of Ornithology 122(3):587-591, 2010

Nests, Nest Placement, and Eggs of Three Philippine Endemic Birds

Luis A. Sanchez-Gonzalez,'-^’^ Carl 01iveros,‘ “ Nevong Puna,-^ and Robert G. Moyle'

ABSTRACT. — We describe the nest and eggs of two
Philippine endemic passerines, the Rusty-faced Babbler
{Robsoniits rabori) and the Blue-breasted Blue Fly-

' Natural History Museum and Biodiversity Research
Center, Dyche Hall, University of Kansas, 1345 Jayhawk
Boulevard, Lawrence, KS 66045, USA,

~9 Bougainvillea Street, Manuela Subdivision, Las Pinas
City, Philippines 1740.

^46-C Brgy Maningning, Lacao Street, Puerto Princesa
City, Palawan, Philippines 5300.

■‘Museo de Zoologia “Alfonso L. Herrera’’, Departa-
mento de Biologia Evolutiva, Facultad de Ciencias,
Universidad Nacional Autonoma de Mexico, Apartado
Postal 70-399, Mexico D.F., 04510, Mexico.

^Corresponding author; e-mail: lasg@ku.edu

catcher (Cyornis herioti). We also describe a novel type
of nest and nest placement for the Short-crested
Monarch (Hypothymis helenae), and provide the first
description of the eggs of this species. We discuss
similarities among eggs and nests of these species with
their relatives, and the need for more information
regarding natural history and breeding habits for the
Philippine avifauna. Received 13 October 2009. Ac-
cepted 10 February 2010.

Ornithologists have long been aware of the
distinctiveness of the Philippine avifauna, which
is reflected in the high percentage of endemism in
the archipelago (Dickinson et al. 1991). However,

588

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 3, September 2010

FIG. 1. Left. Nest of Robsonius rahori. Arrow shows the entrance location (Photograph by Nevong Puna). Right. Egg
of Robsonius rabori (Photograph by Luis A. Sanchez-Gonzalez). Site: ~12 km WSW of Baler. Municipality of San Luis in
Aurora Province (15° 39.22' N, 121° 30.44' E), June 2009.

the Philippine avifauna is one of the most poorly
known in the world (Kennedy et al. 2000).
Breeding habits for some endemic species are
well known, and the first descriptions of the
breeding habits or nestlings of other .species are
starting to appear in the literature (Strijk 2004).
However, breeding habits for a large portion of
the Philippine avifauna remain undocumented.
Dickinson et al. (1991) estimated that —395
species may breed in the country and 157 of
these apparently lack any breeding information.
Not surprisingly, 77 of the.se poorly known
species are endemic, which repre.sents about one
half of the Philippine endemics.

Knowledge of species natural history and
nesting biology is fundamental for development
of sound protection strategies. This information is
also useful for taxonomic purposes. For example,
studies on nest architecture can provide a new set
of taxonomic characters used for phylogenetic
analysis (Zyskowsky and Prum 1999), while data

on clutch size, egg color, and egg patterns can be
used to infer evolutionary relationships (e.g.,
Mayr 1931, Ford 1981, Zelenitsky and Modesto
2003). We report the first description of the nest,
nest placement, and eggs for two species of
Philippine endemic birds, Cyornis herioti and
Robsonius rabori, and provide the first description
of the eggs of Hypothymis helenae as well as a
completely new type of nest placement for this
species.

METHODS

Study Area. — All data for this paper were
collected during a survey conducted from mid
May to late June 2009 in the Sierra Madre
Mountains of Aurora Province, in the northeastern
part of Luzon Island. The main vegetation type in
the area is dipterocarp forests, although montane
forests are found above 1,200 m. The three sites
were at elevations ranging from 50 m at Baler to
530 m at the two other sites visited. The average

SHORT COMMUNICATIONS

589

FIG. 2. Left. Location of the nest cavity of Cyornis herioti. The white arrow shows the cavity in the rock (Photograph by
Luis A. Sanchez-Gonzalez). Right. Nest and eggs of Cyornis herioti', Aurora National Park, Aurora Province (15° 67.80 N,
121° 33.61 E), May 2009. Photograph by Robert G. Moyle).

rainfall is 300 mm per year and is evenly
distributed throughout the year (Dickinson et al.
1991).

Field Procedures. — Data were gathered during
observation walks spanning the hours of avian
activity. Observations were made with binoculars
(7 X 35, 10 X 50) and we used a specialized field
guide (Kennedy et al. 2000) for bird identifica-
tions.

NEST DESCRIPTIONS

Robsoniiis rabori. — The Rusty-faced Babbler
(Timaliidae) is an uncommon and local species
known only from the northern half of Luzon
Island (Collar and Robson 2007), where it occurs
in rocky or part-rocky areas in lowland forest and
second growth up to 1,000 m (Kennedy et al.
2000). Almost nothing is known about the
breeding habits of this elusive species, and its
breeding dates have only been extrapolated from
observations of fledglings in February and March,
and in May and August (Kennedy et al. 2000).

We found a nest of this species on 1 8 June 2009
along an overgrown dirt logging road in second
growth forest at 525 m, ~12 km WSW of Baler,
Municipality of San Luis in Aurora Province (15°

39.22' N, 121° 30.44' E). The nest consisted of a
large clump of dry sticks, branches, and leaves
with a front entrance (8 cm in diameter). The
dome was 19 cm wide and propped up by twigs
50 cm from the ground against a bank of rock and
mud (Fig. 1 Left). The nest was <1 m from a
0.5 m high rock slide. A small stream was —30 m
from the nest down a steep slope. The nest
contained a single elongated egg (28.9 X
18.55 mm) with a whitish-cream base color and
gray and light-brown blotches (Fig. 1 Right). The
female occupant of the nest was collected (catalog
number KUNHM 114678); another egg about to
be laid was in its oviduct indicating a clutch size
of two eggs. Both are now in the Natural History
Museum of the University of Kansas (KUNHM).

The nest architecture of this species is similar to
that reported for a R. .sorsogoueusis (Collar and
Robson 2007); they build large balls of clumped
dried twigs and leaves. This architecture differs
from babblers in the genera Trichastoma, Keiio-
pio, and Turdinus, which currently stand close to
Robsoniiis in linear classifications, but is similar
to that of Gypsophilo and Napothera, which also
build dome nests (Collar and Robson 2007).
However, the eggs of R. rabori superficially

590

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 3. September 2010

FIG. 3. Nest and eggs of Hypothymi.s helenae inside a tree-fern stump. Baler, Aurora Province ( 14° 44' N, 121° 33' E),
June 2009. Photograph by Luis A. Sanchez-Gonzalez.

resemble those of the genus Tiirdinus, as illus-
trated in the case of T. macrodactylus from Java
(Hoogerwerf 1949).

Cyornis herioti. — The Blue-breasted Blue Fly-
catcher (Muscicapidae) is endemic to Luzon and
Catanduanes islands where it inhabits primary and
selectively-logged forests between 100 and
1,200 m (Taylor 2006). This sexually dimorphic
species occurs in mixed species flocks, solitary or
in pairs, and its breeding habitats are mostly
unknown. The breeding season, which may occur
in May, has been inferred from observations of
individuals with enlarged gonads, nestlings, and
recently Hedged juveniles. However, both the nest
and eggs are unknown (Kennedy et al. 2000,
Taylor 2006).

A nest and eggs of this flycatcher were found on
23 May 2009 in Aurora National Park, Aurora
Province (15 67.80 N, 121° 33.61 E). The nest was
in a shallow cavity in a 4-m tall rock along the
course of a river (Fig. 2 Left). The nest was cup-
shaped and had thick walls (inside dimensions 60

X 50 mm; outer dimensions 100 X 120 mm). The
inside was lined with fern fronds and the outside
was covered with moss. Depth of the nest cup was
35 mm, whereas the total depth of the nest was
—55 mm. The nest contained two broken eggs and
one whole egg (21 X 16 mm), which were light
brown with slightly darker specks (Fig. 2 Right).
The female was observed incubating the eggs
during the night. All attributes of the nest including
placement in a shallow cavity and clutch size seem
to be typical of nesting habits in the genus, as well
as in other genera of Muscicapidae, such as
Cyanoptila and Eiimyias (Taylor 2006).

Hypothymi.s helenae. — The Short-crested Mon-
arch is an endemic species of the Monarchidae
that inhabits humid forests up to 1,000 m, where it
usually occurs singly or in pairs in mixed species
flocks. Breeding has been reported to occur
around August, although some individuals taken
in April and May also seemed to be in breeding
condition as inferred from enlargement of the
gonads (Kennedy et al. 2()()(), Coates et al. 2006).

SHORT COMMUNICATIONS

591

Allen et al. (2006) described the nest of this
species from Camiguin Norte Island but eggs
were not observed.

We found a nest of this species on 8 June 2009
in secondary growth forest at Baler, Aurora
Province (14° 44' N, 121° 33' E). The nest was
inside the broken trunk of a tree fern —70 cm
trom the ground. The cavity was 55 mm deep and
65 mm in diameter. Nest material included dry
leaves and small branches of tree fern (Fig. 3), all
roughly shaped into a cup. The nest contained two
eggs, one of which was laid the day after
discovery of the nest. The eggs (21 X 16 mm)
were bluish green with brown and dark brown
blotches; the distal part of the egg was heavily
blotched. The female, which has a different
plumage from the male, was flushed from the
nest at dusk of the second day.

Nest architecture and placement of this species
differs somewhat from its conspecific H. helenae
personam from Camiguin Norte and its close
relative H. azurea, for which nests are typically
cup-shaped structures decorated outside with
lichen and moss, usually in the fork of a tree
branch (Hoogerwerf 1949, Allen et al. 2006,
Coates et al. 2006), whereas the heretofore
undescribed nest of H. helenae helenae from
Luzon was placed in a hollow trunk. Nest
architecture and nest placement within the genus
Hypothymis, and specifically within H. helenae,
needs further study. Natural history and phyloge-
netic studies of the genus are also needed to
examine the significance of these differences.

ACKNOWLEDGMENTS

LASG thanks Consejo Nacional de Ciencia y Tecnologi'a
(CONACyT) for a postdoctoral fellowship in the program
“Estancias postdoctorales y sabaticas al Extranjero para la
consolidacion de grupos de investigacion” Segunda Etapa
(Expediente 93730) at the University of Kansas. This study
was conducted with financial support from the National
Science Foundation (DEB 0743491 to RGM). Michael
Andersen greatly improved the English. We afso thank C.
E. Braun. Bob Kennedy, and an anonymous reviewer whose
comments considerably improved this manuscript.

LITERATURE CITED

Allen, D., C. Espanola, G, Broad, C. Oijvhros, and J.
C. T. Gonzalez. 2006. New bird records for the
Babuyan Islands, Philippines, including two first
records for the Philippines. Forktail 22:57-70.

Coates, B. J., G. C. L. Dutson, C. E. Filardi, P.
Clement, P. A. Gregory, and C. W. Moeliker.
2006. Family Monarchidae (monarch-flycatchers).
Pages 244-329 in Handbook of the birds of the world.
Volume 11. Old World flycatchers to Old World
warblers (J. Del Hoyo, A. Elliott, and D. A. Christie,
Editors). Lynx Edicions, Barcelona, Spain,

Collar, N. J. and C. Robson. 2007. Family Timaliidae
(babblers). Pages 70-291 in Handbook of the birds of
the world. Volume 12. Picathartes to tits and
chickadees (J. Del Hoyo, A. Elliott, and D. A. Christie,
Editors). Lynx Edicions, Barcelona, Spain.

Dickinson, E. C., R. S. Kennedy, and K. C. Parkes. 1991 .
The birds of the Philippines. An annotated check-list.
B.O.U. Check-list Number 12. British Ornithologists’
Union, Tring, United Kingdom.

Ford, J. 1981. Evolution, distribution and stage of
speciation of Rhipidura fuliginosa complex in Aus-
tralia. Emu 81:128-144.

Hoogerwerf, A. 1949. Een bijdrage tot de oologie van het
eiland Java. Uitgave van de Kon. Plantentuin van
Indonesie, Buitenzorg, Java, Indonesia.

Kennedy, R. S., P. C. Gonzales, E. C. Dickinson, H. C.
Miranda, and T. H. Fisher. 2000. A guide to the
birds of the Philippines. Oxford University Press,
Oxford. England.

Mayr, E. 1931. Birds collected during the Withney South
Sea Expedition XVI. Notes on the genus Rhipidura.
American Museum Novitates 502:1-21,

Striik, j. S. 2004. Description of the nest and nestling of
Great Eared Nightjar Eurostopodus macrotis from
Luzon. Philippines. Forktail 20:128-129.

Taylor, B. 2006. Family Mu.scicapidae (Old World
flycatchers). Pages 56-163 in Handbook of the birds
of the world. Volume 1 1 . Old World flycatchers to Old
World warblers (J. Del Hoyo, A. Elliott, and D. A.
Christie, Editors). Lynx Edicions. Barcelona. Spain.

Zelenitsky, D. K. and S. P. Modesto. 2003. New
information on the eggshell of ratites (Aves) and its
phylogenetic implications. Canadian Journal of Zool-
ogy 81 :962-970.

Zyskowsky, K. and R. Prum. 1999. Phylogenetic analysis
of the nest architecture of neotropical ovenbirds
(Furnariidae). Auk 118:891-911.

592

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3. September 2010

The Wilson Journal of Ornithology 122(3):592-597, 2010

Nests, Eggs, and Young of the Azure-crowned Hummingbird

{Amazilia cyanocephala)

Juan Francisco Ornelas'

ABSTRACT. — The first of several nests (previously
undescribed) of the Azure-crowned Hummingbird
{Amazilia cyanocephala) was found on 20 Lebruary
2009 along the Rio Pixquiac in Coatepec, Veracruz on
the Gulf of Mexico side of the Sierra Madre Oriental of
Mexico. The cup-shaped nest consisted of fibers and
scales of tree ferns with the outside covered with
liverworts and a few mosses and lichens. It was saddled
on a thin, horizontal branch and had one “tail” of
hanging liverworts, longer than the height of the nest
cup, draped beneath the nest. It eventually contained
two white, non-glossy eggs that were long-elliptical in
shape, measuring roughly 13.5 X 7.5 mm. Other nests
of the species discovered subsequently were similar in
construction and placed on branches or substrates of
different plant species. Observations of nest building,
eggs, and incubation behavior at the first nest were
generally consistent with descriptions for other Amazilia
hummingbirds. Received 29 September 2009. Accepted
21 Januaty 2010.

Little is known about the nests, eggs, incuba-
tion, fledglings, and breeding behavior of mem-
bers of the genus Amazilia (Skutch 1931,
Harverschmidt 1952, Atwood et al. 1991), which
comprises ~30 species in Central and South
America (AOU 1998, Remsen et al. 2008).
Published records of nests and eggs are available
for less than half of the species (del Hoyo et al.
1999). Thus, knowledge of nesting of Amazilia is
far from complete (Ornelas 1995, Weller 1998).

The Azure-crowned Hummingbird (Amazilia
cyanocephalct) is a common resident in Mexico
from Tamaulipas south through the Isthmus of
Tehuantepec to Chiapas, and through Central
America to Honduras and northern Nicaragua
(Howell and Webb 1995, Johnsgard 1997). This
medium-sized hummingbird is characterized by
the immaculate white throat and belly with golden
green Hanks and indicated breast band, uniformly
green to bronzy upperparts and olive-green

' Dcpartamento de Biologi'a Evoluliva, In.stitulo de
Hcologi'a. AC. Carrctera antigua a Coatcpcc No. 351, El
Haya. 91070 Xalapa. Veracruz., Mexico; e-mail: franciseo.
()rnela.s@inecol.edu.mx

centers of under tail coverts, and a bright,
glittering blue forehead and crown (more tur-
quoise in female-type plumage). It is similar in
size and general appearance to the Violet-crowned
Hummingbird (A. violiceps), but A. cyanocephala
has a blackish upper mandible and a reddish lower
mandible with dark tip, whereas the bill of A.
violiceps is fleshy red except for the dark tip
(Weller 1998). Both taxa are apparently entirely
allopatric (Johnsgard 1997); locally sympatric, the
White-bellied Emerald (A. Candida) is much
smaller and easily distinguished in the field by
the dark subterminal band of its tail.

The species is fairly common in oak (Quercus
spp.) and pine (Finns spp.) woodlands, cloud
forests, humid evergreen forest edges, pine
savannas, gardens, and brushy and second growth
habitats from 500 to about 2,400 m elevation
(Davis 1972, Peterson and Chalif 1973, Howell
and Webb 1995. Johnsgard 1997). Birds on the
Atlantic Slope from southern Tamaulipas to
Veracruz in Mexico, and in the interior from
eastern Oaxaca and Chiapas to western Nicaragua
are assigned to the cyanocephala subspecies.
Resident birds of the Mosquitia coastal region in
Honduras are assigned to the chlorostephana
subspecies with crown glittering green to tur-
quoise-green (Howell and Webb 1995). Range of
the chlorostephana subspecies below 100 m is
tied to the distribution of pine (P. caribae)
savanna, stands of pine close to gallery forest,
and edges of isolated evergreen rainforest (del
Hoyo et al. 1999). Birds from Belize south to
northern Nicaragua, formerly separated as guate-
malensis, show strong intergradation with nomi-
nate birds from Chiapas, Mexico (Howell and
Webb 1995).

Little information on nesting behavior of the
Azure-crowned Hummingbird is available, but
two species of the widespread genus Amazilia
have been studied (Skutch 1931, Harverschmidt
1952), and major differences from these are
unlikely (Johnsgard 1997). Azure-crowned Hum-
mingbirds with enlarged gonads have been

SHORT COMMUNICATIONS

593

FIG. 1. Ne.sl placement, nests, eggs, and nestlings of Azure-crowned Hummingbird (AmazUia cxanocephala)
photographed along the Rio Pixquiac near Xalapa, Veracruz. (A) Completed nest first under construction on 20 February
2009 (nest #1, Table 1 ). (B) Two eggs laid in an interval of 2 days, 25 and 27 February (alternate laying). (C) Eggs hatched
almost .synchronously after 15 days of incubation, the first egg hatched by noon on 15 March. (D) Eight-day old ne.stlincrs.
(E) Incomplete nest (nest #2) built after failure of nest #1. (F) Ne.st #2 was not completed but the first egg was laid on 5
April. (G) and (H) present two views of nest #5. Photographs by the author.

594

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3. September 2010

TABLE 1. Measurements (mm) of Amazilia cyanocephala nests and eggs (min and max exterior and interior diameter
of cup-shaped nests).

Nest measurements Egg measurements

Nest #

Total

depth

External

diameter

Inside diameter
of cup

Depth
of cup

Egg 1

Egg 2

Reference

1

45

45 X 45

28 X 32

20

13 X 7

14 X 8

This Study

2

30

44 X 45

27 X 28

17

This study

3

50

42 X 55

25 X 32

22

This study

4

40

51 X 57

30 X 34

22

This study

5

51

43 X 55

24 X 32

21

This study

6

45

44 X 51

25 X 32

20

This study

7

45

38 X 45

25 X 30

22

This study

8

50

45

27

22

14.2 X 9.0

14.2 X 9.2

Rowley (1984)

reported from February to August (del Hoyo et al.
1999) and two nests were found in April in
Oaxaca, Mexico (Rowley 1984). Nests of A.
cyanocephala have been found before, but none
has been described and observed in detail;
detailed nesting information is unavailable for
most species in this genus (del Hoyo et al. 1999).
The objective of this paper is to describe the nest,
eggs, and incubation of the Azure-crowned
Hummingbird (A. c. cyanocephala) in Veracruz,
Mexico.

OBSERVATIONS

The study area consisted of cloud forest and
secondary riparian growth bordering the Rio
Pixquiac (19° 30' 25" N, 96° 57' 39" W; 1,348 m
above sea level) 6 km west of the city of Xalapa,
Veracruz, Mexico. Rio Pixquiac is a small
tributary of Rfo Consolapa, which traverses
Coatepec on the eastern foothills of the Cofre de
Perote volcano on the Atlantic side of the Sierra
Madre Oriental. 1 observed a hummingbird at
0715 hrs on 20 February 2009 fly directly into a
Phvllostachys bamboo bush at a distance of about
15 m. 1 trained binoculars at where the bird
appeared to land and observed it sitting on a nest
1.5 m above the ground adding nesting material. 1
watched (he hummingbird for ~ 15-20 min,
during which time it visited the nest twice. I
was able to confirm the bird was building the nest
at 0730 hrs the following morning. 1 watched the
female building the nest for —30 min, during
which time she made two visits. I returned later in
the morning (at 1015 hrs) to obtain photographs of
the nest. The female was not .seen during this 15-
min visit. 1 checked the nest daily at —0800 hrs.

The female made one trip to the nest every

30 min during the course of the early morning
observation on 20 February, when the nest was
first found, typically remaining for a minute or
more. Her building rate early on the following
morning was similar and the nest was nearing
completion (Fig. lA). However, the female when
watched from much closer range, showed signs of
nervousness and flew from the nest at each visit. I
stopped close-range observations on nest building
to avoid cessation of activity and to reduce risk of
nest abandonment.

The female, at times, spent a short time during
nest-building visits sitting motionless in the nest
cup in an incubation posture on 25 February. I
checked the nest in her absence by noon, and the
nest contained one egg. The female amved a few
minutes later, perching 5 m from the nest and
caiTying material for the nest (lichens). I moved
away from the nest after taking photographs and
egg measurements, and the female entered the
nest and placed nesting material on the outer wall.
No signs of incubation occun'ed that day or the
following day. A second egg (Fig. IB) was found
(egg measurements in Table 1) on 27 February.
Eggs were laid in the early morning and there was
an interval of 2 days before the second egg was
laid. Incubation began after the second egg was
laid, after which the nest and eggs were checked
twice daily, early in the morning and in the
afternoon until the eggs hatched. Building (— 8--
10 days) and maintenance of the nest, which
consisted primarily of attaching lichens to the
outside of the nest (there were few lichens at the
start of observation), continued throughout the
incubation period.

The incubation period lasted 15 days and eggs
hatched almost synchronously. The first egg

SHORT COMMUNICATIONS

595

FIG. 2. Illustration of the construction of Aniazilia cyaiioceplictia nests collected along the Rio Pixquiac near Xalapa
Veracruz. Open cup-shaped nest at the two ends of the supporting, horizontal Phyllosuichys bamboo branch kept the top of
the nest cup level with respect to the ground (below). Nest cup on the top ol an old Helicoiiici infructescence (above)
Drawing by Edmundo Saavedra.

596

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 3. September 2010

hatched (Fig. 1C) by noon on 15 March and the
second hatched —1800 hrs that day. Eggshell
remains were removed from the nest the following
day. The female brooded the young after feeding
and during the night on subsequent days. Nest-
lings were photographed when they were 8 days
of age (Fig. ID). 1 checked the nest early in the
morning on 23 March, when the nestlings were
9 days of age, and found it empty. The nest was
intact with no signs of disturbance or stealing of
nesting material by other hummingbirds (Skutch
1931). 1 believe Social Flycatchers {Myiozetetes
similis) attacked the nestlings, as 1 observed
several perched near by and had witnessed their
attacks on nestlings of Northern Rough-winged
Swallow {Stelgidopteryx serripeuuis) in the same
area. The nest was collected 2 days later to make a
full description and illustration (Fig. 2). The
female remained in the area for several days.

The cup-shaped nest was built in a 2 m tall
Phyllostachys bamboo, isolated by young second
growth and Heliconia plants near the Rfo
Pixquiac. It was saddled on a lower, slightly
downward sloping branch 1.5 m above the
ground. The horizontal supporting branch was
imbedded within the nest wall and provided
additional support on the lower side of the nest.
One “tail” of nest material hung beneath the
supporting branch (Fig. 2, nest measurements in
Table 1). The nest cup was constructed in two
layers like most hummingbird nests. The outer
layer was comprised of strips of grasses, small
dried pieces of leaves and flowers of Mimo.sa spp.,
scales from tree ferns {Alsophila firnia, Cyathea
hicrenotci and Cyathea aff. fidva) (Cyatheaceae),
whereas the somewhat thicker inner layer was
comprised of Tillandsia deppeana seed down
(Bromeliaceae) and many hairs from horses. The
outside of the nest was heavily decorated with
many pieces of greenish-white lichen, liverworts
Calypogeja spp. (Calypogejaceae), and less so
with mosses; all were adhered with spider web.

1 discovered the female building a second nest
(nest #2; Table 1 ) on 2 April, only 30 m to the
west of nest #1, fastened on the outmost,
unsheltered branch of a Bougainvillea shrub,
1 .7 m above the ground. The nest was half
completed and had no lichens on the outer wall
(Fig. 1 E). I continued my observations of this nest
by daily checking it at -0800 hrs. The female
continued steadily at her growing nest during
consecutive days. The first egg was laid on 5
April, but the nest was not completed (Fig. IF). A

period of inclement weather was initiated the
following day that presumably caused the female
to interrupt building and nest attendance, resulting
in failure to lay a second egg on 7 April. The
single egg lay cold in the nest for an additional
3 days and on 10 April it was destroyed, leaving
only eggshell fragments. The nest was collected
2 days later for measurements. The female was
observed in the area for several days, but a third
nest attempt was not observed.

I am confident the female, which built the
second nest, was the same individual whose
nestlings were depredated. A female (apparently
the same) nested in this area for 3 consecutive
years; nest #3 was placed on a dried Heliconia
infructescence (Fig. 2) and nest #4 was placed on
a Bougainvillea shrub. I searched the Rio Pixquiac
area on 10 April and located three more A.
cyanocephala nests similar to those described.
Nests were placed on a branch of a palm
{Acrocomia spp.) tree (Fig. IG-H), a Heliconia
infructescence, and on a branch of a Quercus tree,
all no more than 3 m above the ground. The three
nests were each active with two nestlings. Nests
were collected and measured (Table 1) after
young fledged.

This study provides the first descriptions and
measurements of the nests of AmazUia cyanoce-
phala. Evidence is presented that supports the
belief that predation and weather may be the main
causes of nesting failure of tropical hummingbirds
(Skutch 1931, FieiTO-Calderon and Martin 2007),
as well as temperate hummingbirds (e.g., Baltos-
ser 1986, 1996).

ACKNOWLEDGMENTS

I thank Mayra Ledezma for showing me the second nest.
Eduardo Ruiz Sanchez for identifying the bamboo.
Francisco Lorea Hernandez for identifying the plant
material. Edmundo Saavedra for the nest illustration, and
Fernando Gonzalez Garci'a, Elisa Peresbarbosa, C. E.
Braun, and two anonymous reviewers for manuscript
review and suggestions. 1 especially thank Julieta Ornelas,
Ana Belen Ornelas, Elisa Peresbarbosa, and Jo.se Luis
Cordova for assistance in the field. This study was partly
funded by a grant (Ref. 61710) from the Consejo Nacional
de Ciencia y Tecnologi'a (CONACyT) and research funds
from the Dcpartamento de Biologi'a Evolutiva. Instituto de
Ecologi'a, A.C. (Ref. 902-12-563).

LITERATURE CITED

American Ornithologist.s’ Union (AOU). 1998. Check-
list of North American birds. .Seventh Edition.
Washington, D.C.. USA.

SHOr^T COMMUNICATIONS

597

Atwood. J. L., V. L. Frrz, and J. E. Bamesberger. 1991.
Temporal patterns of singing activity at leks of the
White-bellied Emerald. Wilson Bulletin 103:373-386.

Baltosser, W. H. 1986. Nesting success and productivity
of hummingbirds in southwestern New Mexico and
southeastern Arizona. Wilson Bulletin 98:353-367.

Baltosser, W. H. 1996. Nest attentiveness in humming-
birds. Wilson Bulletin 108:228-245.

D.avis. L. I. 1972. A field guide to the birds of Mexico and
Central America. University of Texas Press, Austin, USA.

DEL Hoyo, j., A. Elliott, and J. Sargatal (Editors).
1999. Handbook of the birds of the world. Volume 5.
Barn-owls to hummingbirds. Lynx Editions, Barce-
lona, Spain.

Fierro-Calder6n, K. and T. E. Martin. 2007. Repro-
ductive biology of the Violet-chested Hummingbird in
Venezuela and comparisons with other tropical and
temperate hummingbirds. Condor 109:680-685.

Haverschmidt, E. 1952. Notes on the life history of
AmaziUa fimbriata in Surinam. Wilson Bulletin 64:69-
79.

Howell, S. N. G. and S. Webb. 1995. A guide to the birds
of Mexico and northern Central America. Oxford
University Press, New York, USA.

Johnsgard, P. A. 1997. The hummingbirds of North

America. Second Edition, Smithsonian Institution
Press, Washington, D.C., USA.

Ornelas, J. F. 1995. Radiation in the genus Aniazilicr. a
comparative approach to understanding the diversifi-
cation of hummingbirds. Dissertation. Univeristy of
Arizona, Tucson, USA.

Peterson, R. T. and E. L. Chalif. 1973. A field guide to
Mexican birds. Houghton Mifflin Co., Boston, Mas-
sachusetts, USA.

Remsen Jr, j. V., C. D. Cadena, A. Jaramillo, M. Nores,
J. F. Pacheco, M. B. Robbins, T. S. Schulenberg, F.
G. Stiles, D. F. Stotz, and K. J. Zimmer. 2008. A
classification of the bird species of South America.
Part 4. Apodiformes. http://www.museum.lsu.edu/
~Remsen/SACCBaseline04.html

Rowley, J. S. 1984. Breeding records of some land birds in
Oaxaca, Mexico. Proceedings of the Western Founda-
tion of Vertebrate Zoology 2:74-224.

Skutch, A. F. 1931. Life history of the Rieffer’s
Hummingbird (AmaziUa tzacatl tzacatl) in Panama
and Honduras. Auk 48:481-500.

Weller, A. A. 1998. Biogeographie, geographische
variation und taxonomie der gattung AmaziUa (Aves,
Trochilidae). Dissertation. Rheinische Friedrich-Wil-
helms-Universitat, Bonn, Germany.

The Wilson Journal of Ornithology 122(3):597-599, 2010

Nest Description of the Garden Emerald {Chlorostilbon assimilis) from

Costa Rica

Luis Sandoval''^ and Ignacio Escalante’

ABSTRACT. — The Garden Emerald (Chlorostilbon
assimilis) is endemic to southwestern Costa Rica and
Panama, and knowledge about its’ reproductive habits is
limited. We describe the nest and nestlings of the
Garden Emerald based on a nest found in La Amistad
International Park. The nest was built on an anthropo-
genic substrate, and was similar to nests described for
other emerald species. However, unlike other emeralds,
the nest contained no lichens, rnos,ses, or ferns. The
nestlings resembled adult female plumage, similar to
that for other nestling emerald species. Received 24
August 2009. Accepted 3 February' 2010.

The neotropical genus Chlorostilbon includes
17 species of small hummingbirds (AOU 1998,

' Escuela de Biologi'a, Universidad de Costa Rica. 2060
San Pedro de Montes de Oca, San Jose, Costa Rica.

^Corresponding author; e-mail: biosandoval(®hotmaiI.
com

Remsen et al. 2009), 1 1 of which have described
nests (Wolf 1964, Hilty and Brown 1986, Stiles
and Skutch 1989, Thomas 1994, Howell and
Webb 1995, Stiles 1996, Oniki and Antunes
1998, Schuchmann 1999, Gamdo and Kirkcon-
nell 2000). The Garden Emerald (Chlorostilbon
a.ssimilis) is endemic to southwestern Costa Rica
and Panama to western Darien, including pacific
islands, and rarely from Bocas del Toro along
the Caribbean coast of Panama (Ridgely and
Gwynne 1989, Stiles and Skutch 1989). This
hummingbird inhabits secondary forest and open
areas with isolated shrubs at elevations up to
1,500 m (AOU 1998, Garrigues and Dean 2007).
The nest of the Garden Emerald is undescribed
(Stiles and Skutch 1989, Schuchmann 1999) and
we provide the first nest description for this
species. We also include a brief description of
the nestlings.

598

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 3. September 2010

FIG. 1 . Nest of Garden Emerald found on 28 March 2009 near the Altamira Ranger Station. La Amistad International
Park. Costa Rica. (Collected on 18 April 2009; photograph by Luis Sandoval).

OBSERVATIONS

We found a nest of Garden Emerald on 28
March 2009 at the Altamira Ranger Station in La
Amistad International Park, Buenos Aires, Pun-
tarenas, Costa Rica (09 02.9' N, 83° 00.8' W;
1,375 m asl). The site is an open area with a park
ranger station, surrounded by a garden, which
included a live fence of Stachytarpheta spp.
{ Verbenaceae) bushes, Heliconia spp. patches,
and young secondary forest. The nest was
collected on 18 April 2009 after the chicks
fledged, and was deposited in the Museo de
Zoologia. Universidad de Costa Rica.

The nest was a cup built on a hanging portion of
steel mesh under the elevated floor of the ranger
station, I m above ground. The egg cup was 28.0
X 23.5 mm in diameter, and 15.8 ± 2.9 mm
(mean ± SD) deep. The outside of the nest
measured 37.1 X 40.7 mm in diameter and 19.7 ±
3.5 mm tall. The nest was comprised of two
layers. The outer layer included tree bark, small

pieces of dry leaves, twigs, spider web, and
downy seeds (mostly Asteraceae). The inner layer
was comprised mostly of dry fine plant fibers and
downy plant material (Pig. 1).

The nest contained two completely feathered
nestlings when found. The bills of the nestlings
were black with reddish in the mandible base. The
throat and breast of the nestlings were gray, and
the backs were green with buffy-tipped feathers.

DISCUSSION

The nest structure, including the . two-layer
construction, is similar to nests described for
other species of emeralds (Schuchmann 1999).
However, the nest of Garden Emerald, unlike
other emerald nests, contained no lichens, mosses,
or fern materials (Oniki and Antunes 1998,
Schuchmann 1999, Garrido and Kirkconnell
2000). The feather color pattern of the Garden
Emerald nestlings was similar to that described
for others emerald species where the nestlings

SHORT COMMUNICATIONS

599

resemble the adult female plumage (Wolf 1964,
Stiles and Skutch 1989, Schuchmann 1999).

The female did not attend the chicks during the
night of 28 March, likely because the chicks filled
the entire nest. We observed the female attending
the nest and feeding both fledglings the following
morning. The nest was built on an anthropogenic
substrate, but we postulate that, in natural
conditions. Garden Emeralds would build nests
in the lower branches of bushes along trails, forest
edge, or in forest gaps. Our observations reveal
that Garden Emeralds can take advantage of
human structures and successfully use them for
nesting similar to wrens, swallows, or flycatchers
(Stiles and Skutch 1989, Baicich and Harrison
2005, Sandoval and Barrantes 2009).

ACKNOWLEDGMENTS

We thank H. F. Greeney, J. B. Hams, C. E. Braun, and an
anonymous referee for useful comments on an early version of
the manuscript, and the Altamira Ranger Station administra-
tive personnel in La Amistad International Park for support
during our field work. We especially thank Sebastian Bonilla
for bringing us the nest after the chicks fledged.

LITERATURE CITED

A-MERican Ornithologists’ Union (AOU). 1998. Check-
list of North American birds. Seventh Edition. Amer-
ican Ornithologists’ Union, Wa.shington, D.C., USA.
Baicich, P. J. and C. O. Harrison. 2005. Nest, eggs, and
nestlings of North American birds. Princeton Univer-
sity Press, Princeton, New Jersey, USA.

Garrido, O. and a. Kirkconnell. 2000. Field guide to
the birds of Cuba. Cornell University Press, Ithaca,
New York, USA.

Garrigues, R. and R. Dean. 2007. The birds of Costa
Rica; a field guide. Zona Tropical Publication, San
Jose, Costa Rica.

Hii.ty, S. and W. Brown. 1986. A guide to the birds of
Colombia. Princeton University Press, Princeton, New
Jersey, USA.

Howell, S. and S. Webb. 1995. A guide to the birds of
Mexico and northern Central America. Oxford Uni-
versity Press, New York. USA.

Oniki, Y. and a. Z. Antunes. 1998. On two nests of the
Clittering-bellied Emerald Chlorostilbon aureoventris
(Trochilidae). Ornitologfa Neotropical 9:71-76.

Remsen Jr., j. V., C. D. Cadena, A. Jaramillo, M.
Nores, j. E. Pacheco, M. B. Robbins, T. S.
SCHULENBERG, F. C. StILES, D. F. STOTZ, AND K. J.
Zimmer. 2009. A classification of the bird species of
South America. American Ornithologists’ Union.
www.museum.lsu.edu/~Remsen/SACCBaseline.html.

Ridgely, R. S. and j. a. Gwynne Jr. 1989. A guide to the
birds of Panama with Costa Rica, Nicaragua, and
Honduras. Princeton University Press, Princeton, New
Jersey, USA.

Sandoval, L and G. Barrantes. 2009. Nest usurping
occurrence of Piratic Flycatcher (Legatus leucophaius)
in southwestern Costa Rica. Ornitologfa Neotropical
20:401-407.

Schuchmann, K. L. 1999. Family Trochilidae (humming-
birds). Pages 468-680 in Handbook of the birds of the
world. Volume 5. Barn-owls to hummingbirds (J. del
Hoyo, A. Elliott, and J. Sargatal, Editors). Lynx
Edicions, Barcelona, Spain.

Stiles, F. G. 1996. A new species of Emerald hummingbird
(Trochilidae, Chlorostilbon) from the Sierra de
Chiribiquete, southeastern Colombia, with a review
of the C. mellisiigus complex. Wilson Bulletin 108:1-
27.

Stiles, F. G. and A. F. Skutch. 1989. A guide to the birds
of Costa Rica. Cornell University Press, Ithaca. New
York, USA.

Thomas, B. T. 1994. Blue-tailed Emerald hummingbird
(Chlorostilbon mellisiigus) nesting and nestling devel-
opment. Ornitologia Neotropical 5:57-60.

Wolf, L. L. 1964. Nesting of the Fork-tailed Emerald in
Oaxaca. Mexico. Condor 66:51-55.

600

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 3. September 2010

The Wilson Journal of Ornithology 122(3):600-603, 2010

The Nest of the Cipo Canastero {Asthenes luizcie), an Endemic Furnariid from

the Espinhago Range, Southeastern Brazil

Henrique Belfort Gomes’’^ and Marcos Rodrigues"

ABSTRACT. — The Cipo Canastero [Asthenes liiizae)
is a recently described species from the Espinha^o
Range, southeastern Brazil. We describe the nest, eggs,
and nestlings of this species. Six nests were found in
four different territories, two of which were active. All
nests were in Vellozia niveo (Velloziaceae). Mean
measurements were: height above ground = 21.8 cm,
entrance diameter = 4.3 cm, nest length = 23.5 cm,
nest height = 23.3 cm, and nest width = 20.3 cm. Both
active nests were parasitized by Shiny Cowbirds
(Molothriis bonariensis). Received 30 July 2008.
Accepted 5 March 2010.

The Cipo Canastero [Asthenes luizae Vielliard
1990; Furnariidae) is one of the four species of
endemic birds of the Espinha90 Range (Vascon-
celos 2008), an endemic bird area in South
America (Stattersfield et al. 1998). Due to its
restricted range and occurrence in a habitat
subjected to significant anthropogenic (fire and
cattle) and biological pressures (brood parasit-
ism), this species has been classified globally as
‘vulnerable’ (Vielliard 1990, Collar et al. 1992,
BirdLife International 2000). The recent discov-
ery of the Cipo Canastero, an isolated Asthenes in
southeastern Brazil, had great biogeographical
significance, suggesting a historical relationship
between the mountains of southeastern Brazil and
the Andean-Chacoan-Pampean-Patagonian re-
gions (Remsen 2003).

Anedoctal data regarding six nests of the Cipo
Canastero were briefly di.scussed by Studer and
Teixeira (1993). They reported one active nest
with two eggs and five empty nests. The species is
considered high priority for study and conserva-
tion due to its rarity, lack of biological data, and
biogeographical importance (Stotz et al. 1996).
Our objective is to provide detailed descriptions

' Rua Patagonia. 1155. Apartamento 101, 30320-080,
Belo Horizonte, MG, Brazil.

’ Laboratorio de Ornitologia, Dcpartamento de Zoologia,
ICB. Universidade Federal de Mina.s Gerai.s, Caixa Po.stal
486. 31270-901, Belo Horizonte, MG. Brazil.

’Corresponding author; e-mail: hbeltort@gniail.coni

of the nest, eggs, and nestlings of the Cipo
Canastero.

METHODS

Fieldwork was conducted at Alto da Boa Vista
(19° 17' S, 43° 35' W), an area of rocky fields at
an elevation of 1,230 to 1,355 m above sea level
in the Serra do Cipo Range, District of Santana do
Riacho, State of Minas Gerais, southeastern
Brazil. The vegetation of the rocky fields is
marked by sparse treelets and scrubs often with
thick bark. Species of Velloziaceae, Eriocaula-
ceae, Compositae, Rubiaceae, and Orchidaceae
are typical of rocky substrates in the area
(Giulietti et al. 1997). This vegetation type is
characterized by high floristic diversity and an
elevated number of endemic and endangered plant
species (Giulietti et al. 1997).

We randomly searched for nests in September-
November 2005 within an area of ~ 100 ha where
at least five pairs of adults had been daily
observed vocalizing and foraging. The nests found
were periodically monitored and their status
recorded. We used calipers (0.05 mm precision)
to measure the following parameters for each nest:
length, height, and width of the nest, height above
ground and entrance diameter (Fig. 1). Eggs were
weighed with a 10 g spring balance (0.5 g scale),
measured with calipers, and described regarding
their format and color. We report mean ±SD for
all varibles.

RESULTS

Six nests were found in four different teirito-
ries, two of which were active and four were
empty. Empty nests were found in territories with
active nests. All nests were in ‘canela-de-ema!
shrubs [Vellozio niveci) and had a spherical or
cylindrical shape (Fig. 1). The external part of
the nests was built of sticks without thorns (15-
55 cm in length), dry material of Barhacenio spp.
(Velloziaceae) and Vellozia spp., and moss. The
incubation chambers were covered by pilose
material of a Cactaceae, feathers, dry fibers of

SHORT COMMUNICATIONS

601

FIG.

Measurement points for nests of Cipo Cana.stero: (I) height above ground, (2) entrance diameter.

r w. V* / cilllunCC U Iclll IC' ICl , (3) PCSt

length, (4) nest height, and (5) nest width. Two white eggs of the Cipo Canastero and one of Shiny Cowhird are depicted
below the nest.

Velloziaceae, and pilose material of a gall that
develops in a shrub of Erythroxylum spp.

Nests were 21.8 ± 8.2 cm (/? = 6) above
ground and had an entrance diameter of 4.3 ±
0.5 cm (n = 4). Nest measurements were: length
= 23.5 ± 4.3 by 20.3 ± 4.2 cm in width and a

height of 23.3 ± 6.7 cm (// = 6). Both active
nests were parasitized by Shiny Cowbirds
(Molothms honariensis). One active nest had
three eggs, one of Cipo Canastero and two of the
cowbird, and the other had two Cipo Canastero
eggs and one cowbird egg. The eggs of Cipo

602

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3, September 2010

Canastero were oval-shaped and completely
white, while those of the cowbird varied in color
from pink with brown stains, to blue with brown
spots (Fig. 1). Cipo Canastero’ s eggs measured
21.8 ± 0.2 X 17.8 ± 0.1 mm and had a mass of
3 ± 0.5 g (n = 3).

One active nest was discovered on 21 Septem-
ber and was under construction. This nest had
three eggs (Cipo Canastero = 1, Shiny Cowbird
= 2) on 5 October. The Shiny Cowbirds eggs
hatched between 10 and 12 days later. A nestling
cowbird was found alive and the other was dead
on 18 October, and the Cipo Canastero egg
continued in the nest. This egg was found broken
on 28 October, without development. The nestling
cowbird was no longer in the nest on 2 November.
Both adults were marked with colored rings at this
nest. Only one individual Cipo Canastero was
observed on the nest during incubation, and the
other was next to the nest. Both adults were
observed delivering food for the nestling cowbird.

The other active nest was found on 26 October
and had three eggs (Cipo Canastero = 2, Shiny
Cowbird = 1). The nest contained one cowbird
nestling on 1 1 November and two nestlings of
Cipo Canastero on 17 November. One nestling
weighed 2.5 g and was 3.7 cm in length, and the
other 2 g and 3.9 cm in length. The Cipo
Canastero’s nestlings had orange skin and yel-
lowish beak and palate, and were covered with
gray down. Both Cipo Canastero’s nestlings had
their eyes shut. The nestling Shiny Cowbird
weighed 16 g on 17 November. Only the nestling
Shiny Cowbird was in the nest on 20 November.
The nestling Shiny Cowbird was not in the nest on
29 November and it was .seen next to two adults
about 1 50 m from the nest on 1 3 December.

DISCUSSION

The nests and eggs of Cipo Canastero we
de.scribe are similar to those briefly discussed by
Studer and Teixeira (1993). These authors also
found six nests of this species in Vellozici nivea.
These findings suggest Cipo Canastero may
specialize in using this plant species as a substrate
for nesting. However, V. nivea does not occur in
the entire range of the Cipo Canastero including
the northern mountains in Minas Gerais, north of
the Jequitinhonha River Valley (Vasconcelos
2008; R. Mello-Silva, pers. comm.). A possible
explanation for occurrence of Cipo Canastero in
areas where this plant is absent is that several
other species of Vellozia are common and

abundant in the rocky fields throughout the
Espinhago Range (Giulietti et al. 1997). Several
other shrubs of different families, but with
architecture similar to V. nivea, also occur in the
rocky fields and it is possible that Cipo Canasteros
also build their nests in them.

Zyskowski and Prum (1999) presented a
phylogenetic analysis of nest architecture in
Furnariidae. These authors had incomplete data
about the nest of Cipo Canastero and they grouped
this species in the same operational taxonomic
unit as Canyon Canastero (A. pudibunda). Cactus
Canastero (A. cactorum). Dusky-tailed Canastero
(A. humicola), Rusty-vented Canastero (A. dor-
bignyi), Berlepsch’s Canastero (A. berlepschi),
Steinbach’s Canastero (A. steinbachi), Short-
billed Canastero (A. baeri), and Patagonian
Canastero (A. patagonica). Nests of this group
are characterized by their domed vegetative
structure, built of sticks, and the presence of an
entrance tunnel. The detailed nest description
presented here supports this arrangement. How-
ever, it is possible that Asthenes is polyphyletic
(Zyskowski and Prum 1999, Remsen 2003,
Irestedt et al. 2006) and new analyses including
the Cipo Canastero are needed to better under-
stand the systematic position and the biogeo-
graphical affinities of this species (Vasconcelos et
al. 2008).

Adult Shiny Cowbirds were observed around
nests of Cipo Canastero at distances varying
from 3 to 15 m. The Cipo Canastero seemed to
ignore these species, but we observed a pair of
Cipo Canastero chasing two individual Rufous-
collared Sparrows {Zonotrichia capensis) that
approached their nest. The lack of reaction of
Cipo Canastero to presence of Shiny Cowbirds
may suggest they do not recognize this species
as a nest parasite.

The number of nests we found is low, but the
presence of Shiny Cowbird eggs in the nests
demonstrates the Cipo Canastero is a host. Shiny
Cowbird eggs had a shorter hatching period, from
10 to 12 days for the first nest and 15-16 days in
the second nest. This .shorter hatching period
allows Shiny Cowbird nestlings to develop
sooner, providing them with higher chances of
success. Early hatching in the first nest enabled
the nestling of the parasite to make a hole in the
egg of the host and, in the second, allowed the
nestling cowbird (6.5 times larger in mass) to
probably eject the Cipo Canastero nestlings from
the nest or caused death by starvation.

SHORT COMMUNICATIONS

603

Knowledge of the nesting substratum and
reproductive period of species is fundamental for
establishment of conservation programs. The Cipo
Canastero remains common in Serra do Cipo,
despite environmental impacts occurring during
recent years within its area of occurrence (Bird-
Life International 2000). Nest records, however,
are rare and more detailed studies on reproductive
seasonality and factors related to the Cipo
Canastero’ s success are needed to optimize the
conservation of this threatened and restricted-
range species.

ACKNOWLEDGMENTS

We thank Geraldo Fernandes, Tadeu Guerra, Adilson and
Maria Gomes. Marcelo Vasconcelos, and Denise Sarfar for
support in the field. Virginia Braga drew the nest depicted
(Fig. 1). Fabio Olmos, Flavio Rodrigues, Pedro Teixeira,
and two anonymous reviewers helped improve the manu-
script. Renato Mello-Silva provided important information
on the distribution of Vellozici nivea. This work was
supported by ‘Funda^ao O Boticario de Prote^ao a
Natureza’ and Idea Wild. MR was supported by the
Brazilian Research Council (CNPq) and HBG received a
student fellowship from the same organizations. We thank
the staff of the Brazilian Environmental Agency (IBAMA
and CEMAVE), especially Katia Ribeiro and Joao
Nascimento.

LITERATURE CITED

BirdLife International. 2000. Threatened birds of the
world. Lynx Edicions, Barcelona, Spain.

Collar, N. J., L. P. Gonzaga, N. Krabbe, A. Madrono
Nieto, L. G. Naranjo, T. A. Parker III, and D. C.
Wege. 1992. Threatened birds of the Americas: the
ICBP/IUCN red data book. Third Edition. Smithsonian
Institution Press, Washington, D.C., USA.

Giulietti, A. M., J. R. PiRANi, AND R. M. Harley. 1997.
Espinha9o Range region, eastern Brazil. Pages 397-
404 in Centres of plant diversity: a guide and strategy

for their conservation. Volume 3. (S. D, Davis, V. H.
Heywood, O. Herrera-MacBride, J. Villa-Lobos, and
A. C. Hamilton, Editors). Information Press, Oxford,
United Kingdom.

Irestedt, M., j. EjeldsA, and P. G. P. Eric.son. 2006.
Evolution of the ovenbird-woodcreeper assemblage
(Aves: Furnariidae) - major shifts in ne.st architecture
and adaptive radiation. Journal of Avian Biology
37:260-272.

Remsen, j. V. 2003. Family Furnariidae (ovenbirds). Pages
162-357 in Handbook of the birds of the world.
Volume 8. Broadbills to tapaculos (J. del Hoyo, A,
Elliott, and D. A. Christie, Editors). Lynx Edicions,
Barcelona, Spain.

Stattersfield, A. J., M. J. Crosby, A. J. Long, and D. C.
Wege. 1998. Endemic bird areas of the world:
priorities for biodiversity conservation. BirdLife
International, Cambridge, United Kingdom.

Stotz, D. F., j. W. Fitzpatrick, T. A. Parker HI. and D.
K. Moskovits. 1996. Neotropical birds: ecology and
conservation. University of Chicago Press, Chicago.
Illinois, USA.

Studer, A. AND D. M, Teixeira. 1993. Notas sobre a
biologia reprodutiva de Asthenes luizae Vielliard. 1990
(Aves, Furnariidae). Page 44 in Resumos do III
Congresso Brasileiro de Omitologia (M. P. Cime,
Editor). Editora da Universidade Catolica de Pelotas.
Pelotas, Brazil.

Vasconcelos, M. F. 2008. Mountaintop endemism in
eastern Brazil: why some bird species from campos
rupestres of the Espinha^o Range are not endemic to
the Cerrado region? Revista Brasileira de Omitologia
16:348-362.

Vasconcelos, M. F., S. D’Angelo Neto, and J. FjeldsA.
2008. Redescription of Cipo Canastero Asthenes
luizae, with notes on its systematic relationships.
Bulletin of the British Ornithologists’ Club 128:179-
186.

Vielliard, J. 1990. Uma nova especie de Asthenes da serra
do Cipo, Minas Gerais, Brasil. Ararajuba 1:121-122.

Zyskowski. K. and R. Prum. 1999. Phylogenetic analysis
of the nest architecture of neotropical ovenbirds
(Furnariidae). Auk 116:891-911.

604

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 3, September 2010

The Wilson Journal of Ornithology 122(3):604-608, 2010

Nest Box Use by Great Tits in Semi-arid Rural Residential Gardens

Motti Charter, ‘ Yossi Leshem,' Shay Halevi," and Ido Izhaki-^

ABSTRACT. — We studied use of nest boxes by
Great Tits [Parus major) in rural village gardens in a
semi-arid area. Great Tits occupied 46.6% of the nest
boxes, and used nest boxes within higher tree densities
and with more tree species in the vicinity. Breeding
success was greater in nest boxes with higher plant
density, more plant species, and greater height of trees
in the vicinity of the nest. The presence of children or
dogs (Canis lupus familiaris) near nest boxes did not
affect breeding parameters. Syrian Woodpeckers (Den-
drocopos syriacus) enlarged 38.0% of nest box
entrances during the first year. House Sparrows (Passer
domesticus) occupied 41.0% of the nest boxes with
enlarged holes and none of those with normal holes.
Great Tits occupied both types, but significantly fewer
pairs breeding in nest boxes with enlarged holes
succeeded in fledging at least one young, probably
due to their eviction by the larger House Sparrows.
Received 19 October 2009. Accepted 19 January 2010.

Great Tits (Pants major) are one of the most
studied passerines due to their wide distribution
and use of nest boxes. Most studies have been in
northern Europe (Sanz 1998) and, to a much lesser
extent, in Mediterranean Europe (Beldal et al.
1998, Lambrechts et al. 2008), but rarely in the
Middle East (Yavin 1987).

More studies have been of Great Tits breeding
in deciduous (van Balen 1973, Riddington and
Gosler 1995) and coniferous forests (Orell and
Ojanen 1983, Mand et al. 2005), than with those
in suburban (Dhondt et al. 1984, Cowie and
Hinsley 1987) and rural gardens (Perrins 1965,
Riddington and Gosler 1995). Residential gardens
around homes vary greatly in number of plant and
tree species; studies in the.se locations facilitate a
deeper understanding of the environmental factors
birds use for nest site selection.

Great Tits in Mediterranean regions are the
smallest species of cavity breeders and may lose

' Zoological Department. Tel-Aviv University, Ramat-
Aviv. Tel-Aviv 69978. Israel.

’ Moshav Ram-On 19205. Israel.

department of Evolutionary and Environmental Biolo-
gy. University of Haifa, Haifa 31905, Israel.

■•Corresponding author; e-mail: charterm@post.tau.ac.il

nest sites to larger species, mainly House
Sparrows (Passer domesticus) (Barba and Gil-
Delgado 1990), unlike in Europe where Great Tits
compete mainly with smaller species for nest sites
and are the dominant species (Dhondt and
Eykerman 1980, Minot 1981, Alatalo et al.

1986, Minot and Perrins 1986, Alatalo et al.

1987, Kempenaers and Dhondt 1991).

We studied nest box use by Great Tits in rural
village gardens in a Mediterranean climate and
examined environmental factors, in particular, the
presence of children, dogs (Canis lupus famil-
iaris), plants, trees, grass, and size of entrance
holes of nest boxes, that may be related to
occupation and breeding success of Great Tits.

METHODS

The study site was in Moshav Ram-On in the
semi-arid Jezreel Valley, Israel (32° 31' 55" N,
35° 15' 25" E), 80-90 m above sea level with an
average annual precipitation of 453 mm. The 50-
ha study site comprised 170 houses, each with a
garden of 500 m^ and other small buildings used
for storage. These were surrounded by agricultural
areas, mainly fields of sweet corn, alfalfa, oats,
wheat, grape vines, almond plantations, and olive
groves.

Eifty-eight wood nest boxes (15 cm wide X
15 cm long X 24 cm high) with an entrance of
30 mm diameter, at a height of 1 .5 to 2.0 m above
the ground, were affixed during August 2006 to
trees randomly with no more than one nest box
per garden. Active nests (defined as a nest box in
which Great Tits or Hou.se Sparrows built nests)
were visited each week from 1 January 2007 to 1
June 2007. Laying date, clutch size, mean number
of young per nest box after hatching, mean
number of fledglings per laying (all pairs that
laid at least 1 egg were included), and successful
pairs (pairs that fledged at least 1 young) were
recorded for each pair of Great Tits.

The following data were recorded for each nest
box within a 30-m radius; presence of dogs (yes/
no), presence of children (yes/no), plant density
(rated I lowest to 5 highest, not including trees).

SHORT COMMUNICATIONS

605

TABLE 1. Characteristics (x ± SE) of areas where Great Tits built nests

in nest boxes in senii-arid rural gardens, 2007.

Built ne.st

Yes

No

Mann-Wtiilney U

Plant density

3.04 ± 0.2 (/;

= 23)

2.74 ± 0.2 (n

= 23)

U =

230.0, n = 46, P = 0.43

Tree density

3.26 ± 0.2 (n

= 23)

2.43 ± 0.2 (/;

= 23)

U =

164.5, n = 46, P < 0.05

Number of grass species

0.94 ± 0. 1 (/(

= 18)

1 .00 ± 0.0 in

= 19)

U =

161.5, n = 37, P = 0.30

Number of tree species

3.96 ± 0.5 (/;

= 23)

2.26 ± 0.3 (;;

= 23)

U =

152.5, n = 46, P < 0.05

Number of plant species

6. 1 7 ± 1.4 (/;

= 23)

4.55 ± 0.6 (/I

= 22)

U =

228.0, n = 45, P = 0.57

Height ot trees

3.65 ± 0.4 (n

= 23)

2.96 ± 0.3 (n

= 23)

U =

184.5, 11 = 46, P = 0.07

Distance from homes

3.45 ± 0.3 (n

= 22)

3.52 ± 0.3 (/!

= 23)

U =

284.5, 11 = 46, P = 0.91

Distance closest neighboring nest box

4.04 ± 0.3 (u

= 23)

3.57 ± 0.3 (u

= 23)

U =

212.5, n = 46, P = 0.23

Distance to street

3.50 ± 0.4 (n

= 22)

2.61 ± 0.4 (n

= 23)

U =

186.5, n = 45, P = 0.1 1

tree density (1 to 5), number of grass species,
number of tree species, number of plant species,
height of trees (1 to 5), distance from the house (1
to 5), distance to closest neighboring nest box (1
to 5), distance to street (1 to 5), and whether the
entrance hole had been enlarged by Syrian
Woodpeckers (Dendrocopos syriacus) (yes/no).
Plant density, tree density, height of trees,
distance from the house, distance to closest
neighboring nest box, and distance to street were
also rated.

All statistical tests were two-tailed and all tests
were non-parametric. Descriptive breeding data
were analyzed using the Mann-Whitney t/-test.
Spearman’s test was used to analyze rank coirela-
tions among breeding parameters. Chi-square and
Fisher’s exact tests were used for comparing rate of
occupation and nest success, between nests of
different entrance sizes, and nest boxes within the
vicinity of dogs and/or children. Levels of
significance were set at P < 0.05. Statistical
analyses were performed using Statistica (Version
7.1) software and SPSS (Version 17).

RESULTS

Great Tits built nests in 46.6% of the nest boxes
{n = 58), and laid eggs in 85.2% (n = 27) of
these. They succeeded in Hedging at least one
young in 74.1% of the nest boxes in which they
laid eggs (n — 20). The mean laying date was 20
March 2007 (n = 23, SE 0.2), mean (± SE) clutch
size was 7.13 ± 0.4 (/? = 15), mean number of
Hedglings per laying pair was 4.9 ± 0.6 [n = 23),
and mean number of fledglings per successful pair
was 5.7 ± 0.4 (/? = 20). The breeding density was
5.4 pairs/ 10 ha.

Great Tits built more nests in nest boxes that
were in an area with higher tree density and a

higher number of tree species, but no differences
were found between nest box location and other
habitat attributes (Table 1). Greater breeding
success was observed (fledged at least 1 nestling)
in nest boxes in the vicinity of higher plant
density, more plant species, and greater height of
trees (Table 2). Clutch size was not correlated
with nest box location or any habitat attribute;
number of Great Tit fledglings was negatively
coiTelated with tree height in the vicinity of nests
(Table 3).

The presence of dogs did not affect clutch size
(Mann- Whitney U = 4.0, n = S, P = 0.38),
number of fledglings (Mann-Whitney U = 22.0, n
= 15, P = 0.55), or rate of occupation of nest
boxes ix- = 0.04, n = 30, P = 0.92). The
presence of children also did not affect Great Tit
clutch size (Mann-Whitney U = 16.0, n = 14, P
= 0.57), number of fledglings (Mann-Whitney U
= 28.0, /? = 1 8, P = 0.33) or rate of occupation of
nest boxes {x~ = 0.46, n = 44, P = 0.46).

Syrian Woodpeckers enlarged 38.0% of nest
box (/? = 58) entrances during the first year. Great
Tits equally occupied nest boxes with both
enlarged and normal hole entrances (/’ = 0.16,
n = 58, P = 0.69), but significantly fewer pairs
breeding in nest boxes with enlarged holes
succeeded in fledging at least one young
(Fig. 1). House Sparrows occupied 41.0% of the
nest boxes with enlarged entrance holes and none
with normal holes (Fig. 1).

DISCUSSION

Differences in occupation and breeding success
varied with local habitat attributes around the nest
box even though 46.6% of the nest boxes overall
were occupied by Great Tits. Tree number/
density/height and plant number/density had a

606

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. 3. September 2010

TABLE 2. Characteristics (x ± SE) of areas that affected nest success of Great Tits fledging at least 1 young per nest in
nest boxes in semi-arid rural gardens, 2007.

Succeeded to fledge young

Yes No Mann-Whiiney U

Plant density

3.35 ± 0.3 (n =

17)

2.17 ± 0.3 (72

= 6)

U =

23.0,' 72 = 12, P < 0.05

Tree density

3.47 ± 0.3 (72 =

17)

2.67 ± 0.4 (72

= 6)

U =

32.5, 72 = 23, P = 0.18

Number of grass species

0.93 ± 0.1 (72 =

14)

1.00 ± 0.0 (72

= 6)

U =

26.0, 72 = 18, P = 0.60

Number of tree species

4.47 ± 0.7 (72 =

17)

2.50 ± 0.4 (72

= 6)

U =

26.5, 72 = 22, P = 0.08

Number of plant species

7.53 ± 1.7 (72 =

17)

2.33 ± 0.6 (72

= 6)

U =

10.0, 72 = 22, P < 0.01

Height of trees

3.18 ± 0.2 (72 =

17)

5.00 ± 0.0 (72

= 6)

U =

3.0, 72 = 23, P < 0.001

Distance from homes

3.25 ± 0.4 (72 =

16)

4.00 ± 0.6 (72

= 6)

U =

36.0, 72 = 22, P = 0.34

Distance closest neighboring nest box

4.12 ± 0.3 (72 =

17)

3.83 ± 0.5 (72

= 6)

U =

44.5, 72 = 23, P = 0.62

Distance to street

3.69 ± 0.4 (72 =

16)

3.00 ± 0.9 (72

= 6)

U =

40.5, 72 = 22, P = 0.56

role in nest selection and success of breeding
pairs. Trees are important for Great Tits, as their
reproductive parameters increase in accordance
with size of the woodland (Hinsley et al. 1999).
Nestling diet in woodlands contained more
caterpillars (the preferred prey of Great Tits) than
in marginal habitats with fewer and younger trees
(Riddington and Gosler 1995). Children, dogs,
and proximity to streets did not appear to affect
Great Tit breeding. In contrast. Park et al. (2005)
found a difference in breeding near streets. Diet
was not monitored in the present study but may be
important, as both tree and plant densities may
influence prey abundance. Gardens with a higher
abundance of plants and trees are normally
irrigated more, providing green leaves for cater-
pillars.

Breeding densities and success were within the
range of other studies in gardens (Dhontdt et al.
1984, Cowie and Hinsley 1987), but lower than
for pairs in deciduous forests (Cowie and Hinsley
1987). Clutch size was smaller than in most

studies in Europe, and fits within the quadratic
relation suggested by Sanz (1998).

Syrian Woodpeckers were the primary cavity
builders in the study site and, in some cases,
enlarged natural cavities that were too small for
them but which were used by Great Tits.
Consequently, the enlarged entrances become
available for larger species including House
Sparrows. Great Tit and House SpaiTow diets do
not overlap, but both use cavities and may
compete for them. House SpaiTows in our study
occupied 41.0% of the nest boxes with enlarged
holes and none with normal holes. Great Tits
occupied nest boxes with both enlarged and
normal hole entrances similarly, but significantly
fewer pairs breeding in nest boxes with enlarged
holes succeeded in fledgling at least one young,
most likely due to their eviction by larger House
Sparrows.

Barba and Gil-Delgado (1990) reported Great
Tits occupied a diminishing number of nest boxes
over a period of 3 years and suggested this was

TABLE 3. The relation between habitat attributes, clutch size, and number of fledglings in semi-arid rural
gardens, 2007.

Clutch size" Number of fledglings"

r = -0.12, n = 12, P = 0.71
r = -0.22, /; = 12, P = 0.30
r = -0.51, /; = 8, P = 0.19
r = -0.14, n = 12, P = 0.66
r = 0.29, // = 1 2, P = 0.37

r = 0.10, /I = 20, P = 0.67

Plant density

Tree density

Number of grass species

Number of tree species

Number of plant species

Height of trees

Distance from homes

Distance close.st neighboring nc.st box

Distance to street

r = -0.31, n = 12, P = 0.32
r = 0.28, n = 12, P = 0.38
r = -0.02, n = 12, P = 0.96
r = -0.57, n = 12, P = 0.054

r = 0.07, n = 20, P = 0.76
r = -0.35, // = 15, P = 0.21
r = 0.10, n = 20. P = 0.69

r = 0.23, II = 20. P = 0.34

r = -0.51, n = 20. P < 0.05
r = 0.14, n = 19, P = 0.56

/• = 0.19, n = 20. P = 0.41

r = 0.01, n = 19, P = 0.97

■* Spearman’s lest.

SHORT COMMUNICATIONS

607

Normal entrances (r? = 36) Enlarged entrances (n = 22)

Nest entrance size

■ Percent of Great Tit pairs « Nest boxes occupied by House Sparrovvs

FIG. 1. Percent of Great Tit pairs that succeeded in fledging young = 7.92, n = 21, P < 0.01) and percent nest
boxes occupied by House Sparrows {x~ = 13.91, n = 58, P < 0.001) from nest boxes with normal entrance hole (30 mm)
and with entrance holes enlarged by Syrian Woodpeckers.

due to an increase in House Sparrows and black
rats (Rattus rattus) in the nest boxes. The normal
entrance holes in our study prevented House
Sparrows from entering and Great Tits were able
to breed, whereas in the entrances that were
enlarged, sparrows also were able to enter. Both
this study and that of Barba and Gil-Delgado
(1990) suggest that competition may be the reason
for decreased occupation (Barba and Gil-Delgado
1990) and breeding success (this study), but both
studies were purely observational. Experimental
studies manipulating hole entrance size are needed
to better understand whether competition between
the two species affects their breeding success.

ACKNOWLEDGMENTS

We thank the residents of Moshav Ram-On for assistance
and Ran Avrami for weather data. We especially thank
Hava and Uri Ravid for technical assistance in the field, and
Naomi Paz for editorial assistance.

LITERATURE CITED

Alatalo, R. V., L. Gustafsson, and A. Lundberg. 1986.
Interspecific competition and niche shifts in tits Pants:
evaluation of non-experimental data. American Natu-
ralist 127:879-834.

Alatalo, R. V., D. Erickson, L. Gustafs,son, and K.
Larson. 1987. Exploitation competition intluences the
use of foraging sites by tits: experimental evidence.
Ecology 68:284-290.

Barba, E. and J. A. Gil-Delgado. 1990. Competition for
nest-boxes among four vertebrate species: an experi-
mental study in orange groves. Ecography 13:183-186.
Beldal. E. J., E. Barba, J. E. Gil-Delgado, D. J.
Iglesias, G. M. Lopez, and J. S. Monros. 1998.

Laying date and clutch size of Great Tits (Pants major)
in the Mediterranean region: a comparison of four
habitat types. Journal of Ornithology 139:269-276.

CowiE, R. J. AND S. A. Hinsley. 1987. Breeding success of
Blue Tits and Great Tits in suburban gardens. Ardea
75:81-90.

Dhondt, A. A. AND R. Eyckerman. 1980. Competition
and the regulation of numbers in Great and Blue tits.
Ardea 68:121-132.

Dhondt, A. A., R. Eyckerman, R. Moermans, and J.
Huble. 1984. Habitat and laying date in Great and
Blue tits. Ibis 126:388-397.

Hinsley, S. A., P. Rothery, and P. E. Bellamy. 1999.
Influence of woodland area on breeding success in
Great Tits Pants major and Blue Tits Parus caeritleus.
Journal of Avian Biology 30:271-281.

Kempenaers, B. and A. A. Dhondt. 1991. Competition
between Blue and Great tits for roosting sites in
winter: an aviary experiment. Ornis Scandinavica
22:73-75.

Lambrechts, M. M., A. Rielix. M. J. Galan, M. Cartan-
SON, P. Perret, and j. Blondel. 2008. Double-
brooded Great Tits (Pants major) in Mediterranean
oak habitats: do first broods always perform better
than second broods? Russian Journal of Ecology
39:516-522.

Mand, R.. V. Tilgar, A. Lohmus. and A. Leivits. 2005.
Providing nest boxes for hole-nesting birds-Does
habitat matter? Biodiversity and Conservation
14:1823-1840.

Minot, E. O. 1981. Effects of interspecific competition for
food in breeding Blue and Great tits. Journal of
Animal Ecology 50:375-385.

Minot. E. O. and C. M. Perrins. 1986. Interspecific
interference competition-nest sites for Blue and Great
tits. Journal of Animal Ecology 55:331-350.

Orell, M. and M. Ojanen. 1983. Effect of habitat, date of
laying and density on clutch size of the Great Tit

608

THE WILSON JOURNAL OF ORNITHOLOGY • Vul. 122, No. 3. September 2010

Panis major in northern Finland. Holarctic Ecology
6:413-423.

Park, Y., W. Lee, and S. Rhim. 2005. Influence of forest
road on breeding of tits in artificial nest boxes. Journal
of Forestry Research 16:301-302.

Perrins, C. M. 1965. Population fluctuations and clutch-
size in the Great Tit, Pams major L. Journal of Animal
Ecology 34:601-647.

Riddington, R. and a. G. Gosler. 1995. Differences in
reproductive success and parental qualities between

habitats in the Great Tit (Pams major). Ibis 137:371-
378.

Sanz, J. J. 1998. Effects of geographic location and habitat
on breeding parameters of Great Tits. Auk 1 15:1034-
1051.

van Balen, j. H. 1973. A comparative study of the
breeding ecology of the Great Tit Pams major in
different habitats. Ardea 61:1-93.

Yavin, S. 1987. Nest site selection of the Great Tit (Pams
major). Thesis. Tel Aviv University, Israel.

The Wilson Journal of Ornithology 122(3):608-61 1, 2010

Analysis of Nest Sites of the Resplendent Quetzal (Pharomachrus mocinno):

Relationship between Nest and Snag Heights

Dennis G. Siegfried,'"^ Daniel S. Linville,' and David Hille^

ABSTRACT. — The Resplendent Quetzal (Pharoma-
chrus mocinno) is of particular conservation concern
becau.se of its iconic status in Central American culture.
This species is a secondary cavity nester and modifies
abandoned woodpecker nest sites in dead tree trunks
(i.e., snags). We used 1 1 historical nest sites, reported in
1969, from Atitlan, Guatemala and 10 recent nest sites
from San Gerardo de Dota. Costa Rica to examine if a
relationship exists between nest and snag height. There
were significant differences between Costa Rica and
Guatemala in both nest height (6.3 vs. 10 m, respec-
tively; /-test|4 = -2.49, P = 0.042) and snag height
(8.1 vs. 14.0 m, respectively; t-testn = —2.39, P =
0.033). There was no difference in nest heights relative
to snag heights for Costa Rica (0.76) and Guatemala
(0.77; /-test|7 = —0.20, P = 0.84). One aspect of
conservation efforts for this species has been placement
of nest boxes to provide nesting sites for additional
pairs. Our results provide a better understanding of
placement requirements for nest boxes to encourage
their use anywhere within the range of the species.
Received 3 December 2009. Accepted 5 April 2010.

Cavity nesting-.species have challenges from
finding the limited re.sottrce of a nest cavity
(Cornelius et al. 2008) to predation events on
eggs, chicks, or adults (Martin and Li 1992,
Brightsmith 2005). Many cavity-nesting species
in both temperate and tropical latitudes display

' Department of Biology, .Southern Nazarene University,
6729 Northwest 39"’ Expressway, Bethany, OK, 73008,
USA.

^Quetzal Education Research Center, 1913-7050, Car-
tago. Costa Rica.

’Corresponding author; e-mail: dsiegfri@snu.edu

preferences for cavity height in relation to height
of the tree or snag (i.e., relative nest height).
Primary cavity-nesting species, including Red-
headed Woodpecker (Melanerpes erythrocepha-
lus). Northern Flicker (Colaptes ait rants), and
Acorn Woodpecker (M. formicivonts), place nests
at relative nest heights of 0.49-0.58 (Sedgwick
and Knopf 1990, Hooge et al. 1999). Some
secondary-cavity nesters in the temperate zone
select relative nest heights with more variation:
Black-capped Chickadees {Poecile atricapillits)
and House Wren {Troglodytes aedon) at 0.37-
0.38, European Starlings {Slit nuts vulgaris) and
American Kestrels (Falco sparveriits) at 0.5
(Sedgwick and Knopf 1990), and Eastern
Screech-Owls {Megascops asio) at 0.31 (Belthoff
and Ritchison 1990).

Less information is available in tropical areas
about selection of relative nest heights by
.secondary-cavity nesters. The Resplendent Quet-
zal {Pharomachrus mocinno) is an exclusive
secondary-cavity nester (Johnsgard 2000) that,
as an iconic species, has become the focus of
con.servation efforts at many locations throughout
Central America. More efficient planning and
implementation has occurred with concomitant
con.servation benefits for many other species in
focusing on one vulnerable and highly mobile
species (Wheelwright 1983, Powell and Bjork
1995). The Resplendent Quetzal throughout its
range uses natural cavities created from stubs of
fallen branches or abandoned nest cavities of
other species, primarily woodpeckers, which

SHORT COMMUNICATIONS

609

Resplendent Quetzals occupy and enlarge
(Wheelwright 1983). The Acorn Woodpecker is
the primary excavator in the Talamanca Mountain
Range ot southern Costa Rica. This species
excavates several locations for nesting and then
abandons these sites as the level of decay
increases (Koenig et al. 1995). We examined nest
heights of currently active (in Costa Rica) and
historical (in Guatemala; Bowes and Allen 1969)
Resplendent Quetzal nests and compared them to
heights of snags where they occur. We sought to
ascertain if there were: ( 1 ) any differences in nest
placement between countries, (2) consistent
relative nest heights for Resplendent Quetzals,
and (3) any implications for future conservation
efforts.

METHODS

Study Site. — The San Gerardo de Dota Valley is
on the Pacific Slope of the Talamanca Mountain
Range, 85 km southeast of San Jose below the
Cerro del Muerte on the Pan-American highway.
The Quetzal Education Research Center (QERC;
9° 33' 08.67" N, 83° 48' 23.18" W; 2,200 m asl) is
part of a 400-ha preserve consisting of home-
steads pioneered in 1952 for dairy farming. Much
of the pasture land was converted to fruit-tree
orchards starting in 1984 and the last pasture areas
were allowed to begin the process of succession to
forest in 1996 (Neuenschwander 2001). The
primary forest is dominated by oaks {Qiiercus
spp.) with an understory dominated by Lauracea,
Moracea, and Fabacea. Secondary forest areas
have similar composition. The Savegre River
bisects the valley and most human activity occurs
in the associated riparian area. This area has been
planted with eucalyptus (Eucalyptus spp.) trees
for erosion control to supplement the remaining
native species.

Data Collection. — Our initial search for both
active and past nests from January through April
2008 involved interviewing local guides and
individuals in San Gerardo de Dota, Costa Rica
who were familiar with nest locations. We
searched along established trails to locate these
and other nests. Eight of the 10 sites that we
located were active and two were not being used
although observations in previous years indicated
these were active nest sites. We observed pair
activity including exchange of occupancy or
delivery of food items at a site to document it
was active. We measured nest and snag height
using a standard metric tape measure and

clinometer (Suunto PM-5). We calculated relative
nest height as the height of the nest divided by the
height ot the snag. This calculation was also done
lor the 1 1 reported sites from Guatemala (Bowes
and Allen 1969). We used 21 nest sites for
analysis, 10 from Costa Rica and 11 sites in
Guatemala reported by Bowes and Allen (1969).
All 21 sites were in trunks of dead trees (i.e.,
snags).

Statistical Analysis. — We had unequal number
ot sites in each country and a recognized
difference in subspecies. Thus, we used two-
sample r-tests assuming unequal variance to
compare nest and snag heights, and the ratios
between countries (Van Emden 2008). We used
Pearson product-moment correlation to examine if
the nest-to-snag relationship was linear across
sites. We used Systat 1 1 (Systat 2004) for all of
our calculations.

RESULTS

There was only one Resplendent Quetzal nest
cavity per tree at each Costa Rica site. There
were many woodpecker holes in a snag, but the
majority, based on shape and depth, were relic
foraging holes. There were two additional large
holes in the case of tree # 4; one was 1 m above
the active nest while the other was 650 cm
below it facing a different direction. Tree # 2
had one hole 500 cm above the nest that was
used by Spot-crowned Woodcreepers (Lepidoco-
laptes ajfini.s). Nesting success was difficult to
ascertain and it remains unclear how many nest
sites were successful. However, we ob.served a
nestling that left nest # 1 prematurely and died,
and that nest # 5 was abandoned by the adults
prior to fledging.

Snag heights in Costa Rica averaged 8.1 m (5 =
2.78, range = 1.9—10.8 m, n = 10) and were
lower than those used in Guatemala, where the
average was 14.0 m (s = 7.57, range = 5.8-
29.0 m, n = 11; /-test,.-, = -2.39, P = 0.033).
The average nest height of snags used in Costa
Rica (6.3 m, .v = 2.53, range = 1.4-10.1 m, n =
10) was also lower than in Guatemala (10.8 m, 5
= 6.15, range = 4.0-24.4 m, n = 11; Mesti4 =
— 2.49, P = 0.042). The product-moment corre-
lation between snag height and nest height was
high (/- = 0.976, P < 0.0001; Fig. 1). No
difference was found in nest heights relative to
snag heights for Guatemala (0.77) and Costa Rica
(0.76; Mest|7 = -0.20, P = 0.84).

610

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 3, September 2010

D)

0)

(/)

(U

25 -

1 1

1 1

o

20 -

r= 0.976

o

-

15 -

o

-

10 -

OO

•

o

o

-

5 -

•

•

•

Costa Rica

•

o

Guatemala

0 -

1 1

1 1 ’

1 ^

0 5 10 15 20 25 30

Snag height (m)

FIG. 1. Relationship of snag versus nest height of Respiendent Quetzais in two popuiations in Costa Rica and
Guatemaia (/■ = 0.976, P < 0.000 i).

DISCUSSION

Successful nesting by Resplendent Quetzals is
dependent on numerous factors; from males
having sufficient covert lengths and display
flights (Unger 1988) to defending and provision-
ing chicks. One aspect that is crucial to successful
nesting is finding suitable cavities. Resplendent
Quetzals in the northern portion of their range
excavate entire nests as there are no woodpeckers
or their allies that produce a cavity of suitable size
(Renner 2005). We observed Resplendent Quet-
zals at San Gerardo de Dota making small holes in
dead trees prior to mating, but nest cavities were
developed from cavities initially excavated by
Acorn Woodpeckers. The difference may be
related to most nest sites at San Gerardo de Dota
being in trunks of oaks, which have relatively
hard wood.

Finding abandoned Acorn Woodpecker nests
appears to be important, but relative nest height
may be of primary importance in .selecting a nest
site. Average relative nest height was 0.65
(Renner 2005) when no previously excavated
cavities were available, primarily due to a lack of
sufficiently large woodpecker species. Average
relative nest heights were 0.76 and 0.77, respec-
tively in our study and the earlier study in
Guatemala (Bowes and Allen 1969). Both our
Costa Rica site and the Guatemala (Bowes and

Allen 1969) site are within the range of a
woodpecker species that is sufficiently large to
excavate a cavity that may then be enlarged by the
Resplendent Quetzal (Skutch 1944, Bowes and
Allen 1969, Johnsgard 2000).

The primary-cavity nester at San Gerardo de
Dota is the Acorn Woodpecker. This species nests
roughly half way up in a snag (Hooge et al. 1999).
As the snag ages, material is lost from the top
reducing the overall relative nest height. It is only
after sufficient decay that the site is abandoned by
woodpeckers and becomes available for Resplen-
dent Quetzals. Nests are often used by Resplen-
dent Quetzals for several years until the snag
finally collapses.

The continued presence of Resplendent Quet-
zals is of primary interest to the inhabitants of San
Gerardo de Dota as there are more than 12,000
visitors annually to the Quetzal Education Re-
search Center. Conservation efforts at numerous
locations throughout the range of the Resplendent
Quetzal have included addition of artificial nest
boxes (Bowes and Allen 1969, LaBastille 1974).
Approximately four nest boxes per year, including
replacements, have been in place since 1993 at
San Gerardo de Dota. These boxes potentially
could provide nesting sites for additional pairs,
but none has been used. Thus, there may be
sufficient numbers of natural sites for the number
of nesting pairs. However, it is possible the

SHORT COMMUNICATIONS

existing boxes are inappropriately positioned.
These boxes are mounted at the top of 2.5 m
posts, which do not extend above the boxes like
the snags above the natural nesting cavities.
Further, the absolute height of the nest boxes is
considerably below the average heights of natural
nesting cavities recorded in our study.

Conservation efforts for the Resplendent Quet-
zal have implications for many other species in
the tropical cloud forests of Central America
following succession, which provides new areas
tor nesting. Completion of the proposed Meso-
American biological corridor in this area could be
beneficial for Resplendent Quetzals and other
species (Kaiser 2001).

ACKNOWLEDGMENTS

We thank the many individuals in San Gerardo de Dota
who helped locate nest sites for this project. We also thank
two anonymous reviewers who made many useful sugges-
tions as well as the National Aeronautics and Space
Administration (NASA) for providing funding for this
research.

LITERATURE CITED

Belthoff, J. R. and G. Ritchison. 1990. Nest-site
selection by Eastern Screech-Owls in central Ken-
tucky. Condor 92:982-990.

Bowes, A. and D. G. Allen. 1969. Biology and
conservation of the quetzal. Biological Conservation
1:297-305.

Brightsmith, D. J. 2005. Competition, predation and nest
niche shift among tropical cavity nesters: ecological
evidence. Journal of Avian Biology 36:74—83.
Cornelius, C., K. Cockle, N. Politi, I. Berkunsky, L.
Sandoval, V. Ojeda, L. Rivera, M. Hunter Jr., and
K. Martin. 2008. Cavity-nesting birds in neotropical
forests: cavities as potentially limiting resource.
Omitologia Neotropical 19:253-268.

Hooge, P. N., M. T. Stanback, and W. D. Koenig. 1999.

61 1

Nest-site selection in the Acorn Woodpecker. Auk
1 1 6:45-54.

JOHNSGARD, P. A. 2000. Trogoiis and quetzals of the world.
Smithsonian Institution Press, Washington, D.C.,
USA.

Kaiser, J. 2001. Conservation biology: bold corridor
project confronts political reality. Science 293:2196-
2199.

Koenig, W. D., R, L. Mumme, M. T. Stanback, and F. A.
PiTELKA. 1995. Patterns and consequences of egg
destruction among joint-nesting Acorn Woodpeckers.
Animal Behaviour 50:607-621.

LaBastille, A. 1974. Use of artificial nest-boxes by
quetzals in Guatemala. Biological Conservation 1 :64-
65.

Martin, T. E. and P. Li. 1992. Life history of open- vs.
cavity-nesting birds. Ecology 73:579-592.

Neuenschwander, D. E. and L. R. Finkenbinder. 2001.
The chainsaw and the white oak: from astrobiology to
environmental sustainability. Radiations 7:5-10.

Powell, G. and R. Bjork. 1995. Implications of
intratropical migration on reserve design: a case study
using Pharomachrus mocinno. Conservation Biology
9:354-362.

Renner, S. C. 2005. The Resplendent Quetzal {Pharoma-
chrus mocinno) in the Sierra Yalijux, Alta Verapaz,
Guatemala. Journal of Ornithology 146:79-84.

Sedgwick, J. A. and F. L. Knopf. 1990. Habitat
relationships and nest site characteristics of cavity-
nesting in cottonwood floodplains. Journal of Wildlife
Management 54:1 12-124.

Skutch, A. 1944. Life history of the quetzal. Condor
46:213-235.

Systat. 2004. Systat Software, Version 11.00.01. Systat
Software Inc., Chicago, Illinois. USA.

Unger. D. 1988. Welche function hat der extreme
sexuladimorphismus de Resplendent Quetzal {Phar-
omachrus mocinno)? Universitiit Tubingen, Tubingen.
Germany.

Van Emden, FI. F. 2008. Statistics for terrified biologists.
Blackwell Publishing Ltd., Malden, Massachusetts. USA.

Wheelwright. N. T. 1983. Fruits and ecology of
Resplendent Quetzals. Auk 100:286-301.

612

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 3. September 2010

The Wilson Journal of Ornithology 122(3):612-614, 2010

Parasitism of a Blue-winged Teal Nest by a Northern Shoveler in South Dakota

Thomas E. Lewis ' “ and Pamela R. Garrettson'

ABSTRACT. — Nest parasitism in dabbling ducks is
uncommon. Northern Shovelers (Anas clypeata) are
considered infrequent parasites and Blue-winged Teal
(A. discors) are rarely hosts to nest parasitism. We
documented parasitism of a Blue-winged Teal nest by a
Northern Shoveler. We reviewed 3,003 records from
nests located in 1994 to 1997 in North Dakota and
documented only one duck nest parasitized by a
Northern Shoveler. Only nine of 1,494 Blue-winged
Teal ne.sts were parasitized. We found no evidence of
Northern Shovelers parasitizing Blue-winged Teal nests
in our nest records and only one account in the
literature. Received 12 January 2010. Accepted 24
March 2010.

Northern Shovelers (Anas clypeata) and Blue-
winged Teal (A. discors) are common upland
nesting ducks in the northcentral United States
and prairie Canada (Dubowy 1996, Rohwer et al.
2002). The estimated abundance of Northern
Shoveler and Blue-winged Teal in the traditional
survey area was well above the long-term average
in 2009 (USDl 2009).

Nest parasitism occurs when a female purpose-
ly lays eggs in a nest to be raised by an individual
that is not her mate (Sayler 1992). Nest parasitism
is relatively uncommon among dabbling ducks,
except for cavity nesting or ducks that crowd
together on nesting islands (Sayler 1992). North-
ern Shovelers are documented to parasitize nests
of other ducks (Glover 1956, Weller 1959, Sayler
1992, Zimmerman et al. 2003). Blue-winged
Teal nests may rarely be parasitized by other
ducks, but documentation is lacking (Rohwer et
al. 2002). We document the second reported case
of a Northern Shoveler parasitizing a Blue-winged
Teal nest.

OBSERVATIONS

We Hushed a Northern Shoveler hen from a
nest on the morning of 1 2 May 2009. The nest was

' U.S. Fish and Wildlife Service, Division of Migratory
Bird Management, I 1510 American Holly Drive, Laurel.
MD 20708. USA.

^Corresponding author; thom_lewis@fws.gov

in 0.3-m high grass, —0.4 m from a gravel
railroad right-of-way and 20 m from a wetland
near Aberdeen, South Dakota (45° 27' 28" N, 98°
24' 58" W). Examination of the nest revealed two
small (—46^9 mm) and two medium (—52-
55 mm) eggs. The small eggs were tan in color
and the medium eggs were olive-tan and lighter in
color. During later visits the nest contained four
small and four medium eggs on 14 May 2009, six
small and four medium eggs on 16 May 2009, and
seven small and four medium eggs on 17 May
2009. A hen Blue-winged Teal flushed from the
nest on 17 May 2009 during our approach and
down and breast feathers diagnostic of the species
had been added to the nest. The ultimate fate of
the nest could not be ascertained due to the
temporary nature of our visit in South Dakota.

DISCUSSION

Northern Shoveler eggs are described as pale
olive-buff to a pale greenish gray in color (Bent
1923b, Palmer et al. 1976b, Bellrose 1980).
Average length for Northern Shoveler eggs ranges
from 52.2 to 52.85 mm (Bent 1923b, Palmer et al.
1976b, Dubowy 1996). Blue-winged Teal eggs are
reported as light buff-tan (Bennett 1938), creamy
tan (Glover 1956, Palmer et al. 1976a), or cream
(Rohwer et al. 2002) in color and average length
ranges from 46.4 to 47.1 mm (Bent 1923a, Glover
1956, Palmer et al. 1976a). The color and length
of the eggs we observed suggests the medium
eggs belonged to Northern Shoveler and the small
eggs to Blue-winged Teal. Broley (1950) de-
scribes Blue-winged Teal down feathers with
conspicuous white centers and dark tips and breast
feathers with an elongated dark sub-terniinal area
and light proximal portion. Feathers in our nest
matched this description. Observation of the hen
Northern Shoveler on 12 May 2009, the hen Blue-
winged Teal and feathers in the nest on 17 May
2009, along with the egg differences strongly
suggests the Northern Shoveler parasitized the
nest of the Blue-winged Teal.

Blue-winged Teal prefer nesting in upland sites
(Bent 1923a, Bennett 1938, Glover 1956, Johns-

SHORT COMMUNICATIONS

613

gard 1975, Palmer et al. 1976a, Bellrose 1980).
Rohwer el al. (2002) reported Blue-winged Teal
rarely nest in high densities on islands or on
muskrat {Ondatra zihethicns) houses and nests are
dispersed, not readily accessible, and nest para-
sitism is rare, but documentation is lacking. Bent
(1923a) described an account of a Blue-winged
Teal nest containing four eggs of the teal and five
eggs of another duck species. Bellrose (1980)
included Blue-winged Teal as victims of Redhead
(Aythya americana) parasitism. Lokemoen (1991)
reported that Blue-winged Teal nests had parasit-
ism in eight of 1 3 nests on islands, but zero of 54
nests on peninsulas. Zimmerman et al. (2003)
found Blue-winged Teal nests parasitized at rates
from 24 to 72% over a 7-year period on small
islands in North Dakota. Glover (1956) reported
only one of 186 Blue-winged Teal nests examined
was parasitized by another duck which was a
Northern Shoveler.

We examined data for 3,003 nests collected as
described by the Northern Prairie Wildlife Re-
search Center ( 1993) that we visited in northcentral
North Dakota from May to July 1994-1996
(Garrettson and Rohwer 2001). Of these nests,

1 ,494 were Blue-winged Teal and only nine (0.6%)
were reported as parasitized. None was parasitized
by a Northern Shoveler. We found only one case of
Northern Shoveler parasitism, and egg morphology
and color suggested conspecific parasitism (1 of
1 29 Northern Shoveler nests).

Weller (1959) listed parasitic egg-laying by
Northern Shoveler in nests of Mallard (Anas
platyrhynchos), American Wigeon (A. america-
na), Cinnamon Teal (A. cyanoptera), and Red-
head. Fournier and Hines (2001 ) found two of 669
.scaup nests parasitized by Northern Shovelers.
Zimmerman et al. (2003) reported Northern
Shovelers parasitized three of 1,642 nests, but
did not identify the parasitized species. Only
Glover (1956) reported parasitic egg-laying by a
shoveler in the nest of a Blue-winged Teal.
Considering the abundance of these species and
overlap in breeding range and nesting habitat, it is
surprising that parasitism of Blue-winged Teal
nests by Northern Shovelers is so rarely reported.

ACKNOWLEDGMENTS

Louisiana State University and Delta Waterfowl Foun-
dation supported nest record collection and associated
research. The manuscript benefited from the comments of
W. P. Johnson, P. I. Padding, two anonymous referees, and
the editor. The findings and conclusions in this article are

those of the authors ajid do not necessarily repre.sent the
views of the U.S. Fish and Wildlife Service.

LITERATURE CITED

Bellrose, F. C. 1980. Ducks, geese and swans of North
America. Third Edition. Stackpole Books, Harrisburg,
Pennsylvania, USA.

Bennett, L. J. 1938. The Blue-winged Teal; its ecology
and management. Collegiate Press Inc., Ames, Iowa,
USA.

Bent, A. C. 1923a. Blue-winged Teal. Pages 111-121 in
Life histories of North American wild fowl. Part 1.
U.S. National Museum Bulletin Number 126.

Bent, A. C. 1923b. Northern Shoveler. Pages 135-143 in
Life histories of North American wild fowl. Part I.
U.S. National Museum Bulletin Number 126.

Broley, J. 1950. Identifying nests of the Anatidae of the
Canadian prairies. Journal of Wildlife Management
14:452M56.

Dubowy, P. j. 1996. Northern Shoveler (Anas clypeata).

The birds of North America. Number 217.

Fournier, M. A. and J. E. Hines. 2001. Breeding ecology
of sympatric Greater and Lesser scaup (Aythya inarila
and Aythya affinis) in the subarctic Northwest
Territories. Arctic 54:444M56.

Garrettson, P. R. and F. C. Rohwer. 2001. Effects of
mammalian predator removal on production of nesting
ducks in North Dakota. Journal of Wildlife Manage-
ment 65:398-405.

Glover, F. A. 1956. Nesting and production of the Blue-
Winged Teal (Anas discors Linnaeus) in northwest
Iowa. Journal of Wildlife Management 20:28^6.
JoHNSGARD, P. A. 1975. Waterfowl of North America.

Indiana University Press, Bloomington. USA.
Lokemoen, J. T. 1991. Brood parasitism among waterfowl
nesting on islands and peninsulas in North Dakota.
Condor 93:340-345.

Northern Prairie Wildlife Research Center. 1993.
Northern Prairie Wildlife Research Center: Center nest
tile cooperator’s instniction manual. USDI, Northern
Prairie Wildlife Re.search Center, Jamestown, Nonh
Dakota. USA.

Palmer, R. S., F. C. Bellrose, and D. S. Farner. 1976a.
Blue-winged Teal (Anas discors). Pages 463^82 in
Handbook of North American birds. Volume 2.
Waterfowl (R. S. Palmer, Editor). Yale University
Press. New Haven, Connecticut. USA.

Palmer. R. S., F. C. Bellrose. L. K. Sowls. and A. W.
SCHORGER. 1976b. Northern Shoveler {Anas clvpeaia).
Pages 498-515 in Handbook of North American birds.
Volume 2. Waterfowl (R. S. Palmer. Editor). Yale
University Press, New Haven, Connecticut. USA.
Rohwer, F. C., W. P, John.son. and E. R, Loos. 2002.
Blue-winged Teal (Anas di.scor.'i). The birds of North
America. Number 625.

Sayler, R. D. 1992. Ecology and evolution of brood
parasitism in waterfowl. Pages 290-322 in Ecology
and management of breeding waterfowl (B. D. J. Batt.
A. D. Afton. M. G. Anderson, C. D. Ankney, D. H.

614

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 3. September 2010

Johnson, J. A. Kadlec, and G. L. Krapu, Editors).
University of Minnesota Press, Minneapolis, USA.
U.S. Department of Interior (USDI). 2009. Waterfowl
population status, 2009. USDI, Fish and Wildlife
Service, Washington, D.C., USA.

Weller, M. W. 1959. Parasitic egg laying in the Redhead

(Aythya americana) and other North American Anati-
dae. Ecological Monographs 29:333-365.

Zimmerman, A. L., M. A. Sovada, T. K. Kessler, and R.
K. Murphy. 2003. Nest parasitism on constructed
islands in northwestern North Dakota. Prairie Natural-
ist 35:65-80.

The Wilson Journal of Ornithology 122(3):614-617, 2010

Nest Predators of Ground-nesting Birds in Montane Forest of the Santa

Catalina Mountains, Arizona

Chris Kirkpatrick’ and Courtney J. Conway--^

ABSTRACT. — We used time-lapse video cameras
and track plates to identify nest predators of Red-faced
Warblers (Cardellina riihrifrons) and Yellow-eyed
Juncos (Junco phaeonotus) in high-elevation (>
2,300 m) forests of the Santa Catalina Mountains in
southeastern Arizona. Mammals, especially gray fox
(Urocyon cinereoargenteus) and cliff chipmunk {Ta-
niias dorsalis), were the principal nest predators of Red-
faced Warblers and Yellow-eyed Juncos within our
study system, accounting for 89% of all nest depreda-
tions. Our study is one of the first to use video cameras
at real nests to document the prevalence of nest
predators in montane forest ecosystems. Additional
research is needed to learn if mammals are the dominant
nest predators in other montane environments. Received
23 October 2009. Accepted 24 March 2010.

Nest depredation is the most common cause of
reproductive failure of small land birds (Ricklefs
1969, Martin 1992). However, we lack reliable
data on the identity and relative importance of
nest predator species for most birds (Thompson
2007). Information on nest predators is particu-
larly sparse for birds that breed in high-elevation
(>2,300 m) forests in sky island mountain ranges
of the southwestern United States. For example,
>89% of nest failures of ground-nesting birds in
high-elevation forests of the Santa Catalina

' Arizona Cooperative Fi.sh and Wildlife Research Unit,
.School of Natural Resources and the Environment, 325
Biological Sciences Ea.st, University of Arizona, Tucson,
AZ 85721, USA.

^U.S. Geological Survey, Arizona Cooperative Fish and
Wildlife Research Unit, School of Natural Resources and
the Environment, 325 Biological Sciences East, University
of Arizona, Tucson, AZ 85721, USA.

’Corresponding author; cconway@usgs.gov

Mountains in southeastern Arizona is attributed
to nest depredation (Kirkpatrick and Conway
2010). However, the identity of even the most
common nest predators remains unknown. Iden-
tification of nest predators is a critical first step in
managing breeding populations of birds and in
discerning how nest depredation shapes nest-site
selection, life history strategies, and breeding
behavior in birds (Weatherhead and Blouin-
Demers 2004, Thompson 2007).

We used time-lapse video cameras (henceforth
“video cameras’’) to monitor nests of Red-faced
Warblers (Cardellina rubrifrons) and Yellow-
eyed Juncos (Junco phaeonotus) in montane
forests of the Santa Catalina Mountains in
southeastern Arizona. Use of video cameras to
monitor bird nests provides a reliable method to
identify common nest predators (McQuillen and
Brewer 2000, Williams and Wood 2002), but the
technology is expensive and labor intensive and
often only a small sample of nests can be
monitored (Thompson 2007). We augmented
video camera monitoring with track plates
(Wilson and Delahay 2001) to identify potential
nest predator species within our study system.

METHODS

Study Area. — We conducted research in five
16-20 ha study plots (2,300-2,800 m elevation)
within several forested drainages (Kirkpatrick and
Conway 2010) and one forested ridge (Kirkpatrick
and Conway 2005) in the Santa Catalina Moun-
tains, Pima County, Arizona, USA. Temperatures
within our study area varied from an average of
6.5° C in mid-April to an average of 17.6° C in
mid-July (Decker and Conway 2009).

SHORT COMMUNICATIONS

615

Nest Monitoring. — We searched lor and mon-
itored nests ot Red-faced Warblers and Yellow-
eyed Juncos in onr five study plots from late April
to mid-July in 2005-2006. Red-faced warblers
and Yellow-eyed Juncos are the two most
common ground-nesting birds within our study
area: they breed concurrently, share similar nest-
site characteristics, and nest in close association
(Kirkpatrick and Conway 2010). Thus, we were
reasonably confident that nests of both species
would be susceptible to the same community of
nest predators.

Video Cameras.— Wc monitored 16 Red-faced
Warbler and 23 Yellow-eyed Junco nests with
continuously-operated, time-lapse (5 frames/sec),
video cameras equipped with infrared illumina-
tion for night-time photography (Fieldcam TLV,
Fuhrman Diversified Inc., Seabrook, TX, USA;
McQuillen and Brewer 2000). The nesting stage at
which we first set out video cameras at nests
varied: building (19% of nests), egg laying (12%),
incubation (54%), and nestling (15%). We
attached a small (5X2X2 cm) camera via a
small articulated arm to a small wooden stake
buried in the ground ~ 1 m from the nest and
positioned the camera so that it was 40-60 cm
from the nest. We concealed the camera and its
articulated arm with camouflage fabric and ran a
cable (hidden under leaf litter) > 15 m from the
video camera to a concealed location where we
placed the video recorder and battery.

We visited each video recorder every 24 hrs to
change video tapes and batteries, and to check the
status of the nest remotely with a hand-held video
monitor. We occasionally approached nests along
a defined trail to verify nest contents or to adjust
cameras. We made no effort to mask our scent
trail during these nest checks. Human scent trails
leading to artificial ground nests in our study area
did not increase the probability of nest depreda-
tion (Kirkpatrick and Conway 2009). We left
video cameras at nests until young fledged or
nests failed; we then reviewed video tapes of all
depredated nests to identify the species of nest
predator and the date and time of each depreda-
tion event. We assumed that multiple visits to a
nest by the same predator species were the result
of a single individual of that species.

Track Plates. — We placed 38 track plates at
100-m intervals along the center of four of our
five study plots during the middle of the bird
breeding season (3 Jun 2007), alternating loca-
tions of the track plates from one side of the

drainage to the other. We placed the plates on
relatively flat ground at a random distance <50 m
from the drainage bottom (T ± SD: 26 ± 10 m).
We chose a 50-m distance interval becau.se Red-
faced Warblers and Yellow-eyed Juncos select
nest sites that average 26 and 21 m, respectively,
Irom drainage bottoms within our study area
(Kirkpatrick and Conway 2010).

We used a protocol similar to Cain (2001) to
construct track plates from galvanized steel
sheeting (60 X 91 cm) and taped a piece of
contact paper (30 X 46 cm) to the middle of each
plate. We sprayed a mixture containing two parts
isopropyl alcohol and one part blue carpenter’s
chalk (Drennan et al. 1998) on the track plate
around the contact paper and placed a Japanese
Quail {Coturnix japonica) egg covered in chicken
{Callus gallus) egg yolk on the center of the
contact paper. We calculated an index of predator
activity based on the percentage of the 38 track
plates that had > 1 identifiable track of a predator
species following a 6-day exposure period (Dren-
nan et al. 1998).

RESULTS

Video Cameras. — Nineteen of 39 nests with
video cameras were successful, 17 were depre-
dated (total of 18 depredation events; Table 1),
two were abandoned due to inclement weather,
and one was abandoned for an unknown reason.
Both Red-faced Warbler and Yellow-eyed Junco
nests were depredated by gray foxes {Urocyon
cinereoargenteiis), cliff chipmunks {Tamias dor-
salis), and Steller’s Jays {Cyanocitta stelleri). One
Yellow-eyed Junco nest was depredated partially
by a gray fox and a woodrat (likely Neotoma
mexicana). Gray foxes and the woodrat depredat-
ed nests at night (67% of all nest depredation
events), whereas cliff chipmunks and Steller’s
Jays depredated nests during the day (33% of all
nest depredation events; Table 1). Gray foxes
depredated video camera nests both early and later
in the avian breeding season (27 Apr-19 Jul),
whereas cliff chipmunks depredated video camera
nests only later in the .season (1 Jun-1 Jul).

Track Plates. — We identified tracks of six
mammal and one bird species on track plates in
early June 2007. Indices of predator activity were
50% for gray fox, 32% for mice (Peromyscus
spp.), 32% for cliff chipmunk, 8% for Abert’s
squirrel {Sciurus aberti), 5% for Steller’s Jay, 3%
for bobcat {Lynx rufus), and 3% for domestic dog
{Canis lupus familiaris). Pilot work using track

616

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 3, September 2010

TABLE 1. Depredation events at Red-faced Warbler and Yellow-eyed Junco video camera nests by four nest predator
species in high-elevation (>2.300 m) forest within the Santa Catalina Mountains, Arizona (late Apr to mid Jul, 2005-
2006).

Depredation events

Nest predator

Red-faced Warbler nests

Yellow-eyed Junco nests

Totals

Time range (hrs) of depredations'*

Gray fox

3

8

1 1

2004-0330

Cliff chipmunk

2

2

4

0645-1911

Steller’s Jay

1

1

2

0800-1551

Woodrat*’

0

1

1

0200-0420^

Totals

6

12

18

Mountain Standard Time.

Likely N. me.xicuna (Kirkpatrick and Conway 2006).

'■ A woodrat depredated the brooding female Yellow-eyed Junco at 0200 hrs and then returned to the nest to depredate one of three nestlings at 0420 hrs. A gray
fox depredated the two remaining ne.stlings (which were still alive) at 0218 hrs the next day (Kirkpatrick and Conway 2006).

plates in early May 2007 also recordecJ rock
squirrel (Spermophilus variegatiis) and Common
Raven {Connis corax) within our study area (C.
Kirkpatrick, unpubl. data).

DISCUSSION

Mammals, especially gray fox and cliff chip-
munk, were the principal predators of Red-faced
Warbler and Yellow-eyed Junco nests within our
study system, accounting for 89% of all nest
depredations documented on video. Gray fox and
cliff chipmunk are common in montane forests
throughout southeastern Arizona (Hoffmeister
1956, Lange 1960) and are known to depredate
bird nests (Burt and Grossenheider 1976, Hart
1992). Other potential mammalian nest predators,
including mice, Abert’s squirrels, bobcats, do-
mestic dogs, and rock squirrels were present
within our study area (documented by track plates
and visual observations by field personnel) but
were not recorded depredating nests by video
cameras (perhaps because some of these species
occur at low densities; C. Kirkpatrick, pers. obs.).

Additional video camera monitoring may be
necessary to ascertain whether some of these
mammals (especially mice; Bradley and Marzluff
2003, King and DeGraaf 2006) are occasional nest
predators within our study system. Some nest
predators may avoid or be attracted to nests with
video cameras and the track plates provide a
means by which one can identify nest predators
present in the study system that were not captured
by video surveillance. Data from track plates do
not provide an unbiased estimate of density or
movement rates of nest predators because they are
baited and some predators may avoid or be
attracted to them, but they do provide information

on the suite of potential nest predators and their
relative levels of activity (relative to other
locations where track plates are placed with
similar methods).

The prevalence of mammalian nest predators in
our study system is surprising given that snakes
are considered to be the principal nest predator of
many New World passerine birds (Weatherhead
and Blouin-Demers 2004), especially in the
southern USA (Thompson 2007). We believe we
did not record snakes as nest predators because
our study area was: ( 1 ) >2,300 m above sea level
in an environment thermally inhospitable to most
snake species (M. J. Goode, University of
Arizona, pers. comm.); and (2) beyond the
northern range limits of several montane rattle-
snake species (Stebbins 1985), one of which
(twin-spotted rattlesnake [Crotahis pricei]) is
known to occasionally depredate Yellow-eyed
Junco nests at high elevation (2,560 m; Gumbart
and Sullivan 1990). We observed only two snakes
(Sonoran mountain kingsnake [Lampropeltis pyr-
ometana] and Arizona black rattlesnake [Crotalus
cerheriis]) near our study plots and none within
the boundaries of our study plots despite thou-
sands of person hours in the field from 2002 to
2009 (C. Kirkpatrick, unpubl. data).

Our study is one of the first to use video
cameras at real nests to identify nest predators and
measure their relative importance in montane
forest ecosystems (similar results were obtained in
montane forests of northern Arizona where
mammals caused 62% of diurnal nest depreda-
tions; T. E. Martin, University of Montana, pers.
comm.). Additional research is needed to ascer-
tain if mammals are the principal nest predators in
other high-elevation environments. Selection

SHORT COMMUNICATIONS

617

should favor adaptations by birds to reduce
exposure to numerous olfactory-oriented, noctur-
nal nest predators (e.g., gray foxes, woodrats,
mice) in montane environments, just as selection
is thought to favor behaviors that reduce exposure
to visually-oriented, diurnal predators in other
ecosystems (Weatherhead and Blouin-Demers
2004). Identifying the commonalities in the suite
of primary nest predators within different ecosys-
tems may help explain differences in behavior and
life history traits among species of birds. Montane
forests in the Santa Catalina Mountains provide an
ideal environment to examine these issues given
that breeding populations of ground-nesting birds
at higher elevations (>2,300 m) are contiguous
with breeding populations at lower-elevations
(>1,800 m) where snakes and other nest predators
are likely to be more common.

ACKNOWLEDGMENTS

This research was funded in part by the U.S. Fish and
Wildlife Service. We thank S. J. Sferra (U.S Bureau of
Reclamation) for the loan of video cameras and M. H. Ali,
M. M. Eastwood, D. D. LaRoche, N. M. Nardello, A. M.
Purdy, E. T. Rose, T. A. Selvidge, D. D. Tracy, and Anna
Westling-Douglass for assistance locating nests and oper-
ating video cameras. R. L. Peterson and the University of
Arizona Steward Observatory provided field housing and J.
D. Taiz (U.S. Forest Service) provided logistical support.
Any use of trade names is for descriptive purposes only and
does not imply endorsement by the U.S. Government.

LITERATURE CITED

Bradley, E. and J. M. Marzluff. 2003. Rodents as nest
predators: influences on predatory behavior and
consequences to nesting birds. Auk 120:1 180-1 187.
Burt, W. H. and R. P. Grossenheider. 1976. A field
guide to mammals: field marks of all North American
species found north of Mexico. Houghton Mifflin
Company, Boston, Massachusetts, USA.

Cain, J. W. 2001. Nest success of Yellow Warblers
(Dendroica petechia) and Willow Flycatchers {Empi-
donax iraillii) in relation to predator activity in
montane meadows of the central Sierra Nevada,
California. Thesis. California State University, Sacra-
mento, USA.

Decker, K. L. and C. J. Conway. 2009. Effects of an
unseasonable snowstorm on Red-faced Warbler nest-
ing .success. Condor 1 1 1:392-395.

Drennan, j. E., P. B. Beier, and N. L. Dodd. 1998. Use of
track stations to index abundance of sciurids. .lournal
of Mammalogy 79:352-359.

Gumbart, T. C. and K. a. .Sullivan. 1990. Predation on

Yellow-eyed Junco nestlings by twin-spotted rattle-
snakes. Southwestern Naturalist 35:367-368.

Hart, B. E. 1992. Tamais dorsalis. Mammalian Species 399.

Hoffmeister, D. F. 1956. Mammals of the Graham
(Pinaleno) Mountains, Arizona. American Midland
Naturalist 55:257-288.

King, D. I. and R. M. DeGraaf. 2006. Predators at bird
nests in a northern hardwood forest in New Hamp-
shire. Journal of Field Ornithology 77:239-243.

KiRKPATRtCK, C. AND C. J. CoNWAY. 2005. Reproductive
success and habitat associations of riparian birds
within the sky island mountains of southeastern
Arizona. Wildlife Research Report #2005-07. USGS,
Arizona Cooperative Fish and Wildlife Research Unit,
Tucson, USA.

KtRKPATRiCK, C. AND C. J. CoNWAY. 2006. Woodrat
(Neotoma) depredation of a Yellow-eyed Junco (Junco
phaeonotus) nest. Southwestern Naturalist 51:412-
414.

Kirkpatrick, C. and C. J. Conway. 2009. Effects of
researcher disturbance on nesting success of Red-faced
Warblers: a species of conservation concern. Wildlife
Research Report #2009-01. USGS, Arizona Coopera-
tive Fish and Wildlife Research Unit, Tucson, USA.

Kirkpatrick, C. and C. J. Conway. 2010. Importance of
montane riparian forest and influence of wildfire on
nest-site selection of ground-nesting birds. Journal of
Wildlife Management 74:729-738.

Lange, K. 1. 1960. Mammals of the Santa Catalina
Mountains. American Midland Naturalist 64:436^58.

Martin, T. E. 1992. Interaction of nest predation and food
limitation in reproductive strategies. Current Ornithol-
ogy 9:163-197.

McQuillen, H. L. and L. W. Brewer. 2000. Methodo-
logical considerations for monitoring wild bird nests
using video technology. Journal of Field Ornithology
71:167-172.

Ricklefs, R. E. 1969. An analysis of nesting mortality in
birds. Smithsonian Contributions to Zoology 9:1-48.

Stebbins, R. C. 1985. A field guide to western reptiles and
amphibians. Houghton Mifflin Company, Boston,
Massachusetts, USA.

Thompson, F. R. 2007. Factors affecting nest predation on
forest songbirds in North America. Ibis 149:98-109.

Weatherhead, P. J. and G. Blouin-Demers. 2004.
Understanding avian nest predation: why ornitholo-
gists should study snakes. Journal of Avian Biology
35:185-190.

Williams, G. E. and P. B. Wood. 2002. Are traditional
methods of determining nest predators and nest fates
reliable? An experiment with Wood Thrushes (Hvlo-
cichhi inioitelina) using miniature video cameras. Auk
1 19:1 126-1 132.

Wilson, G. J. and R. J. Delahay. 2001. A review of
methods to estimate the abundance of terrestrial
carnivores using field signs and observation. Wildlife
Research 8:151-164.

618

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 3. September 2010

The Wilson Journal of Ornithology 122(3):6 18-620, 2010

Feeding a Foreign Chick: a Case of a Mixed Brood of Two Tit Species

Toshitaka N. Suzuki^’^ and Yuko Tsuchiya^

ABSTRACT. — We report on a mixed brood of Great
(Pants major) and Varied (Poecile varius) tits. The
brood consisted of one Great Tit and five Varied Tits.
Incubation and food provisioning to nestlings were by
adult Varied Tits, but not by adult Great Tits. The
foreign Great Tit nestling tended to receive more food
than the average Varied Tit nestling. Provisioning of the
foreign Great Tit by Varied Tits lasted even after
fledging. We discuss the causes and outcomes of this
rare phenomenon. Received 5 December 2009. Accepted
9 March 2010.

Nestlings of different species are rarely present
in the same nest (Shy 1982), apart from cuckoo-
host interactions (Payne 1977, Rothstein 1990).
Mixed broods have been reported among several
species of passerines including titmice, nuthatch-
es, flycatchers, and thrushes (e.g., Mackenzie
1950, 1954; Slagsvold 1998; Dolenec 2002;
Robinson et al. 2005; Govoni et al. 2009). Two
types of mixed brood rearing are known: only one
species raises foreign young along with their own
(Slagsvold 1998), and both pairs from different
species raise the mixed brood together (i.e., nest
sharing; Robinson et al. 2005, Govoni et al. 2009).
The major cause of mixed broods in both cases
has been explained as interspecific competition
for a desirable nest site.

The occurrence of mixed broods provides an
opportunity to investigate the evolutionary origin
of interspecific brood parasitism (Slagsvold
1998), and to examine the role of learning in
species recognition (Hansen et al. 2008, 2009).
However, detailed observations ot mixed broods
have not been reported with the exception of
studies in which researchers experimentally cross-
foster eggs between different species (e.g.,
Slagsvold 1998). This .seems to be due to the

' Department of Life Seienee, Graduate Sehool ot
.Science, Rikkyo University, 3-.34-1 Nishi-Ikebukuro,
Toshima, Tokyo 171-8501. Japan.

^PREC Institute Inc., 1-7-15 Kyomachibori, Nishi,
Osaka 5.5()-00().3, Japan.

'Corresponding author; e-mail: toshi.n.suzuki@gmail.
com

rarity of occurrence of mixed broods as well as
insufficient opportunity for video recordings in
past decades.

We report a case of a naturally-occurring mixed
brood of two tit species. The mixed brood
consisted of one Great Tit {Parus major) and five
Varied Tits {Poecile varius). The Varied Tit pair
raised the foreign Great Tit with their own
nestlings in the same nest box. Feeding events
inside the nest box were video recorded.

OBSERVATIONS

We placed 25 nest boxes (17 X 17 X 30 cm, 3 cm
entrance) in a mixed deciduous-coniferous forest in
2006 in Kamizawa, Nagano, Japan (36° 21' N,
138° 35-36' E) and monitored them during the
breeding season. The forest consisted mainly of
Japanese larch {Larix leptolepis), Mongolian oak
{Qitercus mongolica), and giant dogwood {Swida
controversa). Nine of the 25 boxes were used by
Great Tits, one box by Varied Tits, and one
contained a mixed brood of Great and Varied tits.

We visited the mixed brood nest 1 1 times in the
period of 1 month (5 times during the egg stage, 5
times during the nestling stage, and once on the
day of fledging). Incubation by one of the Varied
Tit adults (probably the female) was confirmed
during the nest visitations, but not by either of the
Great Tits. We identified six eggs in the nest and
all eggs hatched synchronously on 16 May. The
Varied Tit also sat on the naked chicks on the day
of hatching. We recorded feeding events inside
the nest box during the nestling period by placing
a KPC-EX500 CCD camera (KT&C Co., Seoul,
Korea) on the ceiling of the nest box connected to
a GR-DF590 digital video camera (Victor Co.,
Kanagawa, Japan). Only the adult Varied Tits
provisioned food (caterpillars, bees or spiders) for
the nestlings for the entire 805 min recorded, and
no adult Great Tits visited the nest box.

We analyzed the provisioning behavior of the
adult Varied Tits using the video recordings from
inside the nest box. We could easily discriminate
between the foreign Great Tit and Varied Tit
chicks by morphology in the late nestling stage.

SHORT COMMUNICATIONS

619

(/)

D)

c

'c

g

w

■>

o

L_

Q.

0)

E

D

FIG. 1. Provisionings by adult Varied Tits to a foreign
Great Tit nestling and Varied Tit nestlings in 480 min of
video records. The bar representing the Varied Tit nestlings
indicates the mean value for the five nestlings. The broken
line indicates the expected number of provisionings if all
nestlings are provided with food randomly.

and used the video data recorded at this stage (14,
15, and 16 days of age) for analysis. We analyzed
154 provisioning events during 480 min of video
recordings. The adult Varied Tits fed several
different nestlings at each provisioning visit (x ±
SD: 2.2 ± 0.7 nestlings; ranging from 1 to 4). The
mean rate of provisioning visits was 8.8/hr.

We tested whether food provisionings by adult
Varied Tits were biased between the Great Tit
nestling and the Varied Tit nestlings using a G-
test (Sokal and Rohlf 1995). A significant
difference was found (G-test with Williams’
correction: Gj = 6.36, P = 0.012; Fig. 1) when
comparing the observed number of provisionings
received by the Great Tit nestling and the five
Varied Tit nestlings with the expected number of
provisionings (i.e., the value if all nestlings obtain
food randomly). This implies the adult Varied Tits
provided more food to the Great Tit young than
they did to young Varied Tits on average.

All young fledged sequentially between sunrise
and noon on 1 June when they were 17 days of
age. The Great Tit and Varied Tit fledglings were
confirmed to be fed by the adult Varied Tits after
fledging.

DISCUSSION

This is the first report of an occurrence of a
Great and Varied tit mixed brood, and the first
video analysis of a naturally occurring mixed

brood. The foreign Great Tit nestling appeared to
be smaller than Varied Tit chicks, but it tended to
obtain more food than the average Varied Tit
nestling (Fig. I). This appears to be the result of
active begging behavior by the Great Tit; it
begged for food in 100% of the provisioning visits
by the adults, whereas Varied Tit young begged
only in 85% of the provisioning visits, on average.
The Varied Tit adults fed the Great Tit even after
fledging, suggesting they learned to recognize the
foreign chick as if it was their own.

This observed mixed brood is unlikely to be
due to a new breeding strategy of Great Tits in this
population because this is the only known case in
our 4-year study (breeding seasons from 2006 to
2009), during which time we monitored 106 nest
boxes of cavity-nesting birds. A previous report
suggests that Great Tits have not evolved
intraspecific brood parasitism (Kempenaers et al.
1995). It is more likely the mixed brood is a
nonadaptive by-product of interspecific nest-site
competition. Interspecific nest usurpations in our
study site were often observed between Great Tits
and Varied Tits during the beginning of the
breeding season. It may be possible the observed
mixed brood occurred either when Varied Tits
took over the Great Tits’ nest in which one egg
had already been laid, or when Great Tits failed to
usurp the nest of Varied Tits.

Feeding of a foreign chick is obviously costly
for adult Varied Tits, but the fitness consequence
for the foreign Great Tit remains unclear. The
Great Tit nestling tended to obtain more food than
the average Varied Tit nestling, but we cannot
reject the possibility that quality and quantity of
food may be insufficient for the foreign Great Tit.
A previous report suggests the dietary needs of
nestlings may be different between these two tit
species (Mizutani and Hijii 2002). The Great Tit
may also have other costs, such as problems with
species recognition when subsequently trying to
mate and breed. Species recognition is imprinted
early in life, and may be irreversible once it has
been establi.shed (Hansen et al. 2008, 2009).

The occurrence of mixed broods may represent
an initial step in evolution of interspecific brood
parasitism (Slagsvold and Hansen 2001). The
pre.sent observation provides an example that a
foreign chick can be raised to fledging by a
different species. Future field observations may
reveal further examples, which may give us
valuable insights into the evolution of interspe-
cific brood parasitism.

620

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3. September 2010

ACKNOWLEDGMENTS

We are grateful to Masayoshi Kamioki, O. K. Mikami,
Kihoko Tokue, Keisuke Ueda, and two anonymous
reviewers for helpful comments on this manuscript. We
also thank the Lorestry Agency and the Picchio Wildlife
Research Center for permitting us to study at this site.

LITERATURE CITED

Dolenec, Z. 2002. A mixed brood of nuthatch (Sitta
europaea) and Great Tit (Parus major) species. Natura
Croatica 1 1 : 103-105.

Govoni, P. W., K. S. Summervill, and M. D. Eaton.
2009. Nest sharing between an American Robin and a
Northern Cardinal. Wilson Journal of Ornithology
121:424-426.

HAN.SEN, B. T., L. E. JOHANNESSEN, AND T. SLAGSVOLD. 2008.
Imprinted species recognition lasts for life in free-living
Great Tits and Blue Tits. Animal Behaviour 75:92 1-927.
Hansen, B. T., L. E. Johannessen, and T. Slagsvold.
2009. Interspecific cross-fostering affects mate guard-
ing behaviour in Great Tits (Pams major). Behaviour
146:1349-1361.

Kempenaers, B., R. Pinxten, and M. Eens. 1995.
Intraspecific brood parasitism in two tit Pams species:
occurrence and responses to experimental parasitism.
Journal of Avian Biology 26:1 14-120.

Mackenzie, J. M. D. 1950. Competition for nest sites among
hole breeding species. British Birds 43:184-185.

Mackenzie, J. M. D. 1954. Redstarts reared in tits’ nests.
Scottish Naturalist 66:146-154.

Mizutani, M. and N. Huh. 2002. The effects of arthropod
abundance and size on the nestling diet of two Pams
species. Ornithological Science 1:71-80.

Payne, R. B. 1977. The ecology of brood parasitism in
birds. Annual Review of Ecology and Systematics
8:1-28.

Robinson, P. A., A. R. Norris, and K. Martin. 2005.
Interspecific nest sharing by Red-brested Nuthatch
and Mountain Chickadee. Wilson Bulletin 117:400-
402.

Rothstein, S. 1. 1990. A model .system for coevolution:
avian brood parasitism. Annual Review of Ecology
and Systematics 21:481-508.

Shy, M. M. 1982. Interspecific feeding among birds: a
review. Journal of Field Ornithology 53:370-393.

Slagsvold, T. 1998. On the origin and rarity of
interspecific nest parasitism in birds. American
Naturalist 152:264-272.

Slagsvold, T. and B. T. Hansen. 2001. Sexual imprinting
and the origin of obligate brood parasitism in birds.
American Naturalist 158:354—367.

SOKAL, R. R. AND F. J. Rohlf. 1995. Biometry. Third
Edition. W. H. Freeman, New York, USA.

The Wilson Journal of Ornithology 122(3):620-622, 2010

Infanticide by an Eastern Phoebe

Gary Ritchison'-^ and Brandon T, Ritchison'

ABSTRACT. — We videotaped an Eastern Phoebe
(Sayornis phoebe) nest in central Kentucky on 22 June
2008 and documented a case of infanticide. An adult
male phoebe, likely the resident male, was captured at
the nest site on 21 June 2008. The resident female fed
nestlings (3, ~8 days of age) 69 times during a 4.1 -hr
taping .session (0815-1221 hrs EOT) the next day, but
the resident male was not observed at the nest.
Beginning at 0923 hrs, an intruding phoebe (likely a
male) occasionally visited the nest, sometimes pecking
at. but not feeding, the nestlings; this phoebe initiated a
vigorous attack at 1151 hrs, pecking and pulling the
head of one of the nestlings for 25 min before pulling it
out of the nest. The intruding phoebe returned after a
short absence, just before the end of the videotape, and
began attacking another nestling. The nest was empty
when checked the next day. Suitable territories and nest
sites may be limiting factors for Eastern Phoebes, and

' Department of Biological Sciences, Eastern Kentucky
University, Richmond, KY 40475, USA.

^Corresponding author; e-mail: gary.ritchison@eku.edu

infanticide by non-breeding males may enhance their
reproductive success if the resident female or a new
female subsequently pairs with the male and initiates a
new nest. Received II Januarv 2010. Accepted 24
March 2010.

Infanticide has been reported in many species of
birds, primarily in cooperative breeders, colonial
breeders, and polygamous species (Freed 1986).
Infanticide has been reported among socially
monogamous species of birds, but primarily species
where breeding opportunities may be limited by
biased .sex ratios or availability of needed resources
such as nest sites (Freed 1986; Robertson and
Stutchbury 1988; Mpller 1988, 2004). Infanticide
may be adaptive either by providing access to a
needed resource, such as a nest site (Robertson and
Stutchbury 1988), or by shortening the time until a
new breeding attempt (Freed 1986).

SHORT COMMUNICATIONS

621

Most reports ot infanticide among socially
monogamous species have involved secondary
cavity-nesting species where availability of nest
sites may limit breeding opportunities (Freed
1986, Robertson and Stutchbury 1988, Kermott
et al. 1991 ). We report a case of infanticide by an
Eastern Phoebe {Sayornis phoebe), a non-cavity-
nesting species with specific nest-site require-
ments that may limit breeding opportunities of
some individuals.

OBSERVATIONS

We studied Eastern Phoebes at the Blue Grass
Army Depot (BGAD) from mid-March to mid-
July 2008. The BGAD is in Madison County,
Kentucky and encompasses 5,907 ha of pastures
with trees and woodlands interspersed throughout.
Concrete shelters (about 2.5 X 5 X 2.5 m; « =
54), each with two open doors (1 X 2 m) at
opposite corners occur throughout the depot to
provide depot personnel with protection in case of
emergencies. The shelters are rarely visited by
Army personnel, but phoebes readily enter the
shelters to nest.

We located a phoebe nest on 21 June 2008 with
three nestlings ~8 days of age in a shelter that, for
security reasons, we visited infrequently. An adult
phoebe entering the shelter was captured in a mist
net on the same day and banded with a USGS
aluminum band and a unique combination of three
colored-plastic bands. The phoebe was identified
as a male based on the presence of a cloacal
protuberance. Standard measurements were taken
and the male, in apparently good condition, was
released.

The nest was videotaped from 08 1 5 to 1221 hrs
EDT on 22 June 2008 with a camcorder placed
2 m from the nest. The camcorder was protected
from the elements in the shelter so we did not
retrieve the videotape until the next day (23 Jun
2008). We subsequently reviewed the tape and
noted all visits to the nest by adult phoebes as well
as any other recorded behaviors or vocalizations.

An unbanded Eastern Phoebe, apparently the
resident female, provisioned nestlings at a rate of
16.8 visits/hr during the 4.1 -hr videotape; the
phoebe banded the previous day (assumed to be
the resident male) was not observed at the nest.
The female phoebe visited the nest and provi-
sioned nestlings 35 times from 0815 to 1015 hrs,
and always exhibited the same behavior, landing
at the edge of the nest and quickly feeding a
nestling. Another phoebe, also unbanded, but

behaving differently, was first observed at 0923
hrs. Over the next 42 min, this phoebe hovered
near the nest twice, briefly landed on the edge of
the nest three times, and pecked the head of a
nestling during two of those nest visits.

During the third hour of taping (1015-1115
hrs), the resident female fed nestlings 16 times;
the other phoebe entered the shelter and perched
on the camcorder four times, but did not visit the
nest. The resident female continued provisioning
nestlings during the fourth hour of taping (1115-
1215 hrs). The ‘intruding’ phoebe entered the
shelter and perched on the camcorder 21 times
from 1115 to 1145 hrs, uttering ‘bipeaked
vocalizations’ (Smith 1969) on multiple occasions
and ‘initial peaked vocalizations’ (Smith 1969)
twice. The female visited the nest three times
while the other phoebe was on the camcorder; the
presence of the intruding phoebe had no apparent
effect on the female’s behavior.

The intruding phoebe flew to the nest at 1151
hrs and began vigorously pecking and pulling
one of the nestlings, focusing its attack on the
nestling’s head. This phoebe pecked or pulled at
the nestling about 1,250 times (~50 times/min)
over the next 25 min. The resident female came
to the nest ~5 sec after the attack was initiated,
hovered briefly at the edge, then landed with
beak agape and moved to within ~2 cm of the
attacker for ~3 sec. The female then flew from
the nest, returning 5 sec later and again briefly
hovered near the nest. As the attack continued,
the female left and re-entered the shelter and
perched on the camcorder three times (but
uttered no calls), and flew to the nest and
provisioned nestlings five times, uttering no
vocalizations and not interacting with the
attacking phoebe. The intruding phoebe pulled
the apparently dead nestling out of the nest at the
conclusion of the 25-min attack, and returned to
the nest —4.5 min later and began attacking a
second nestling. The videotape ended 40 sec
after this second attack began.

The nest was empty when visited the next day
with no sign ot adult or nestling phoebes in the
shelter. We re-visited the shelter 12 days later; the
nest was still empty and we observed no adult
phoebes in or near the shelter.

DISCUSSION

Male Eastern Phoebes typically provision
nestlings and defend nest sites and territories
from conspecifics (Weeks 1994). Thus, the

622

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 3. September 2010

absence of the male at our focal nest plus the
presence of the intruding phoebe suggests the
resident male was not present. The male we
banded at the nest site, likely the resident male,
appeared to be in good condition when released.
The reason for the male’s absence is unknown, but
could have been due either to abandonment of the
territory (e.g., as a result of being captured in a
mist net) or predation.

Our observations indicate adult Eastern Phoe-
bes at times engage in infanticide. Beheler et al.
(2003:996) also noted instances of infanticide
(‘probably by nonparental adults’) by Eastern
Phoebes, and Weeks (1994) reported that infan-
ticide at times occurred after a territorial male
phoebe disappeared. Phoebes are sexually mono-
morphic and both males and females utter
‘bipeaked’ and ‘initial peaked’ vocalizations
(Smith 1969). Thus, we cannot be certain the
intruding phoebe in our study was a male.
However, the absence of the resident male
suggests the infanticidal phoebe was a male.
Infanticide by male phoebes may enhance repro-
ductive success if the resident female or a new
female subsequently pairs with the infanticidal
male and initiates another nest. This behavior has
been reported in other species of songbirds,
including House Wrens {Troglodytes aedon;
Freed 1986, Kermott et al. 1991), Tree Swallows
(Tachycineta bicolor, Robertson and Stutchbury
1988), and Barn Swallows (Hirundo nistica',
Mpller 2004).

Suitable territories and nest sites may be
limiting factors for Eastern Phoebes, and com-
petition for these resources may be intense
(Beheler et al. 2003). Phoebes occupy a variety
of woodland and edge habitats (Graber et al.
1974), but the availability of suitable nest sites
may be limited in many areas (Weeks 1994).
Thus, some adult phoebes may be non-breeding
floaters. Supporting this assumption is the rapid
replacement of dead mates and rapid settlement
of territories when new nest sites become
available (Beheler et al. 2003). The disappear-
ance of a territorial male repre.sents an oppor-
tunity for non-breeding males to acquire the
resources needed for reproduction. Phoebes are
typically double-brooded (Weeks 1994) and, in
our study area, second broods are generally
initiated in late May or early June; new nests are

rarely initiated after late June (GR, pers. obs.).
A male phoebe acquiring a territory with 8-day-
old nestlings on 22 or 23 June, as in our study,
would likely have no opportunity to breed
during the cun'ent breeding season if those
nestlings developed normally and fledged (gen-
erally when 16-18 days of age; Weeks 1994).
Infanticide could potentially provide a male
phoebe with a breeding opportunity. However,
the nest in our study was empty when checked
12 days later and we observed no phoebes in or
near the shelter, indicating the infanticidal male
phoebe in our study did not subsequently breed
in that shelter.

ACKNOWLEDGMENTS

We thank the U.S. Army for allowing us to conduct
re.search at their facility, the EKU Research Committee for
financial support, and C. E. Braun, H. P. Weeks Jr., and an
anonymous reviewer for helpful comments. A short video
clip of the infanticidal phoebe attacking and removing the
nestling is posted at http://www.youtube.com/watch?-
v=4mVQMpyqlrs.

LITERATURE CITED

Beheler, A. S., O. E. Rhodes Jr., and H. P. Weeks Jr. 2003.
Breeding site and mate fidelity in Eastern Phoebes
(Sayomis phoebe) in Indiana. Auk 120:990-999.

Freed, L. A. 1986. Territory takeover and sexually selected
infanticide in tropical House Wrens. Behavioral
Ecology and Sociobiology 19:197-206.

Graber, R. R., J. W. Graber, and E. L. Kirk. 1974.
Illinois birds: Tyrannidae. Illinois Natural History
Survey Biological Notes Number 86.

Kermott, L. H., L. S. Johnson, and M. S. Merkle. 1991.
Experimental evidence for the function of mate
replacement and infanticide by males in a north-
temperate population of House Wrens. Condor
91:630-636.

Moller, a. P. 1988. Infanticidal and anti-infanticidal
strategies in the swallow Hirundo ru.sticci. Behavioral
Ecology and Sociobiology 22:36.3-371.

Moller, A. P. 2004. Rapid temporal change in frequency
of infanticide in a passerine bird associated with
change in population density and body condition.
Behavioral Ecology 15:462^68.

Robertson, R. J. and B. J, Stutchbury. 1988.
Experimental evidence for sexually .selected infanti-
cide in Tree Swallows. Animal Behaviour 36:749-753.
Smith, W. J. 1969. Displays of Sayornis phoebe (Aves,
Tyrannidae). Behaviour 33:283-322.

Weeks, Jr., H. P. 1994. Ea.stern Phoebe (Sayornis phoebe).
The birds of North America. Number 94.

SnOr^T COMMUNICATIONS

623

The Wilson Journal of Ornithology 122(3):623-624. 2010

Marsh Wren Eats Small Fish

Andrea J. Ayers'’^ and James W. Armacost Jr.'

ABSTRACT. — We report an observation of a Marsh
Wren [Cistothorus palustris) eating a small fish. On 29
November 2008, a Marsh Wren was observed catching
and consuming a small fish, probably a mosquitofish
(Ganibusia affinis), in a coastal marsh in Jefferson
County, Texas. Marsh Wrens are known to eat
invertebrates, primarily insects and spiders, and we
believe this is the first record of a Marsh Wren
capturing and consuming vertebrate prey. Received 24
August 2009. Accepted 22 January 2010.

The Marsh Wren {Cistothorus palustris) is a
small passerine that inhabits a variety of coastal
and interior marshland habitats. Prior observations
of the feeding habits of Marsh Wrens indicate
they feed strictly on invertebrates, foraging in
dense patches of shrubs or along the base of cattail
(Typha spp.) stalks or other marsh vegetation, on
or near the ground or water surface (Bent 1948,
Kroodsma and Verner 1997). Marsh Wrens
typically move constantly as they forage, gleaning
prey from plant surfaces or from on or just below
the water surface. Most prey items are too small
for a person to identify from a few meters
(Kroodsma and Verner 1997). While conducting
fieldwork as part of a project to assess avian
mortality and population-level effects associated
with a transmission line in coastal marshes at the
J.D. Murphree Wildlife Management Area
(WMA) in Jefferson County, Texas, AJA made
an opportunistic observation of a foraging Marsh
Wren that captured and consumed a small fish.
After reviewing Bent (1948) and Kroodsma and
Verner (1997), we believe this is the first report of
a Marsh Wren taking live vertebrate prey.

OBSERVATIONS

The J.D. Murphree WMA contains a diverse
community of coastal wetlands, including fresh-
water, intermediate, brackish, and saline wetlands
(Texas Parks and Wildlife Department 2008). On

' Department of Biology, Lamar University, Beaumont,
TX 77710, USA.

^Conesponding author; e-mail: andreajayers@hotmail.
com

29 November 2008, at —1000 hrs CST, AJA
observed a Marsh Wren perched above a roadside
ditch that drained water from a former construc-
tion site for the transmission line. A section of the
ditch was partially overhung by matted vegetation
that was residue from Hurricane Ike, which made
landfall on the southeast Texas coast on 13
September 2008.

The Marsh Wren was perched on a small dead
limb —20 cm above and diagonally crossing the
running water, looking downstream. The wren sat
perched over the water observing small fish
swimming below for a short time. While still
perched, the wren bent down and plucked a fish
from the water, never leaving its perch. The wren
captured and handled the fish adeptly, positioning
the fish in its beak to swallow it head-first. The
wren proceeded to swallow the fish in small
increments, alive and whole. We estimated the
size of this fish to be approximately twice the
length of the Marsh Wren’s beak. The exposed
CLilmen of an adult Marsh Wren ranges in size
from 12 to 14 mm (Ridgway 1904); thus, we
estimate the fish to have been 24-28 mm in
length. Afterwards the bird flew from the roadside
ditch into denser marsh vegetation. This observa-
tion was made from a distance of —4 m. using 8
X 40 binoculars, and the prey item, which had a
silver, reflective body and a noticeable tail fin,
was easily identifiable as a fish rather than an
invertebrate. We speculate the fish was a
mosquitofish {Ganibusia affinis), based on the
type of marshland habitat.

DISCUSSION

This observation occurred 1 1 weeks after
Hun-icane Ike made landfall on the southeast
Texas coast. The J.D. Murphree WMA was
covered by a storm surge that peaked at >4 m
and persisted for several days (Berg 2009). This
.salt water intrusion killed many plants in Texas
coastal marshes, and abundance of aquatic inver-
tebrates may have been reduced. A possible
explanation for this Marsh Wren’s behavior may
have been the lack of invertebrate prey, leading it

624

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3. September 2010

to capture and consume a small vertebrate. Marsh
Wrens typically forage in dense vegetation, making
their foraging behavior difficult to observe
(Kroodsma and Vemer 1997), and it is possible
that vertebrates are a regular, but previously
undocumented, component of their diet.

We believe our observation of a Marsh Wren
feeding on a small fish is the first record of this
behavior and contributes new knowledge about the
ecology of this species. Pieman et al. (1993)
observed Marsh Wrens to destroy eggs and
nestlings of competing birds in deep-marsh habi-
tats, while Bump (1986) reported Marsh Wrens
appeared to have sampled the contents in four of
five cases when they punctured eggs. However, we
consider consumption of egg contents incidental,
and not primarily a foraging behavior. Larger
species of wrens are known to take vertebrate prey
on occasion. For example. Cactus Wrens {Campy-
lorhynchus hrunneicapilliis) have been known to
feed on amphibians and reptiles (Storer 1920,
ProLidfoot et al. 2000). It isn’t surprising that Marsh
Wrens, which feed mainly on aquatic invertebrates
(Bent 1948, Kroodsma and Vemer 1997), would
capture and consume small fish, given that other
species of wrens are known to take vertebrate prey
on occasion.

ACKNOWLEDGMENTS

We thank Golden Pass LNG Terminal LLC and Port
Arthur LNG Terminal LLC for access and funding our
research on bird interactions with transmission lines. We
al.so thank the Texas Parks and Wildlife Department, J.D.

Murphree Wildlife Management Area, for access to study
sites. We thank Paul Nicoletto of Lamar University for
assistance with fish identification and Steven Cardiff of the
Louisiana State University Museum of Natural Science for
assistance with beak measurements. We thank L. S.
Johnson, Andrew Kasner, and an anonymous reviewer for
comments that improved the manuscript.

LITERATURE CITED

Bent, A. C. 1948. Life histories of North American
nuthatches, wrens, thrashers, and their allies. U.S.
National Museum Bulletin Number 195.

Berg, R. 2009. Tropical cyclone report: Hurricane Ike
(AL092008) 1-14 September 2008. National Hurri-
cane Center, Miami, Florida, USA.

Bump, S. 1986. Yellow-headed Blackbird nest defense:
aggressive responses to Marsh Wrens. Condor 88:328-
335.

Kroodsma, D. E. and J. Verner. 1997. Marsh Wren
(Cistothonis palustris). The birds of North America.
Number 308.

PiCMAN, J., M. L. Milks, and M. Leptich. 1993. Patterns
of predation on passerine nests in marshes: effects of
water depth and distance from edge. Auk 1 10:89-94.
Proudfoot, G. a., D. a. Sherry, and S. Johnson. 2000.
Cactus Wren (Campylorhynclius hrunneicapillus). The
birds of North America. Number 558.

Ridgway, R. 1904. The birds of North and Middle
America: a descriptive catalogue of the higher groups,
genera, species, and subspecies of birds known to
occur in North America. Part 3. U.S. National Museum
Bulletin Number 50.

Storer, T. 1. 1920. Lizard eaten by Cactus Wren. Condor
22:159.

Texas Parks and Wildlife Department. 2008. J.D.
Murphree WMA. www.tpwd.state.tx.us/huntwild/
hunt/wma/find_a_wma/list/?id=40.

SnOF^T COMMUNICATIONS

625

The Wilson Jounuil of Ornithology 1 22(3);625-626, 2010

Occasional Mimicry and Night-time Singing by the Western Curve-billed
Thrasher (Toxostoma curvirostre pahneri)

R. Roy Johnson' - and Lois T. Haight'

ABSTRACT. — The first instance of vocal mimicry is
reported for the western subspecies of Curve-billed
Thrasher (Toxostoma curvirostre palmeri). A Curve-
billed Thrasher engaged in countersinging with a
migrating Black-headed Grosbeak (Pheucticus melano-
cephahts) near Tucson, Arizona. Night-time singing by
Curve-billed Thrasher is also documented for the first
time. At least three responding Curve-billed Thrashers
engaged in spontaneous song near Tucson, Arizona.
Additional night-time singing was elicited by playback
recordings Received 13 Fehritaiy 2010. Accepted 24
April 2010.

The Curve-billed Thrasher {Toxostoma curvi-
rostre) is placed in the Mimidae, a family that
includes several species that mimic other species
and/or sing at night (Bent 1948). Perhaps most
widely known is the ability of Northern Mock-
ingbird (Mimiis polyglottos) to mimic other
species, and also sing both day and night during
the breeding season (Demckson and Breitwisch
1992). Mimicry has been reported for the
nominate subspecies group of the Curve-billed
Thrasher in Texas (Rylander 1981, Tweit 1996),
but neither vocal mimicry nor night-time song has
been previously documented for the western
subspecies (T. c. palmeri; Bent 1948, Tweit
1996). The only reference found for possible
night-time singing by T. curvirostre was by
Oberholser (1974; 654), “it seldom .sings at
night,” also referenced by Tweit (1996) but with
no specific documentation or reference to source
by either publication.

OBSERVATIONS

The western Curved-billed Thrasher has a
varied and melodious repertoire, but mimicry of
another species was recorded only recently. A
Curved-billed Thrasher mimicked a migrant
Black-headed Grosbeak {Pheucticus melanoce-
phalus) on a desert plot —23 km east of Tucson,

' John.son & Haight Environmental Con.sultants, 3755
South Hunters Run, Tucson, AZ 85730, USA.
"Corresponding author; e-mail: rroyloi.s@aol.com

Arizona. The Black-headed Grosbeak was singing
near the top of a large ornamental eucalyptus
{Eucalyptus spp.) tree on 23 April 2009 at —0830
hrs MST. Another apparent Black-headed Gros-
beak was answering from a patch of Sonoran
Desert scrub —35 m away. Singing during spring
migration is not unusual for Black-headed Gros-
beak but countersinging by the species in desert
lowlands is unusual. The grosbeak singing in the
eucalyptus tree behaved in typical fashion,
commonly giving one or more chuck notes
between songs, but the bird in desertscrub was
not interspersing songs with chuck notes. On
closer examination with binoculars, the second
bird was observed to be a Curve-billed Thrasher
near the top of a foothill palo verde {Parkinsonia
microphylla) tree singing a song that was audibly
indistinguishable from that of the grosbeak.

T. c. palmeri has sung both spontaneously and
elicited night-time song on this same desert plot.
A Curve-billed Thrasher began spontaneous song
on 21 February 2005 at 2212 hrs (sunset 1815
hrs). It sang two refrains of several notes, the
refrains separated by -15 sec. The thrasher at site
A was answered by a short song from a second
thrasher -150 m distance (site B). Thrasher A
sang another short song and was answered by a
third thrasher (site C) -150 m in a different
direction at 2213 hrs. Thrashers singing at sites A,
B, and C roughly formed an equilateral triangle
with spacing of -150 m between thrashers. This
appioximates the spacing between T. curvirostre
males engaged in diurnal temtorial song in this
area during the breeding season.

The moon was bright on the night of 21
February (full moon 24 Feb) but an -5 min
shower had earlier produced a trace of rain at
—2145 hrs. A thrasher at site A again sang
spontaneously on the following night, 22 Febru-
ary, and was answered by a thrasher at site B
between 2127 and 2128 hrs. The sky was .similar
to the previous night with .scattered clouds but the
moon was bright and visible most of the time.

626

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3. September 2010

DISCUSSION

Diurnal song at this time of year is common for
this early nesting species for which breeding
readiness peaks in early February (Vleck 1993,
Tweit 1996). The Curved-billed Thrasher was not
among the 53 listed avian species known to
respond to playback tapes by 1980 (Johnson et al.
1981). After not hearing spontaneous singing
during the nights following 22 February, a
Curved-billed Thrasher song tape was played
(Colver et al. 1999) on 27 February. Following
five playings of the tape, beginning at 2229 hrs
with hesitation between each, a thrasher at site A
responded. It sang a refrain of ~six notes at 2233
hrs. A similar song again came from site A at
2234 hrs. after playing the tape two more times.
Despite further playing of the tape there was no
additional response. A similar response from a
thrasher at site A resulted when a tape was played
on 28 February, beginning at 2325 hrs. Answers
of a half-dozen notes were sung again at 2328 hrs
and at 2329 hrs. No further attempts were made to
elicit songs.

Neither behavior discussed here has been
recorded previously for this species. Because of
its close relationship to other mimids that exhibit
mimicry and/or nocturnal song, it would be of
phylogenetic and general ornithological interest to
devise a project, using playback tapes, to further

investigate these phenomenon among the subspe-
cies of T. curx’irostre.

ACKNOWLEDGMENTS

We thank R. C. Tweit, C. E. Braun, and two unknown

reviewers for critiques of an early manuscript.

LITERATURE CITED

Bent, A. C. 1948. Life histories of North American
nuthatches, wrens, thrashers, and their allies. U.S.
National Museum Bulletin Number 195.

Colver, K. J., D. Stokes, and L. Stokes. 1999. Stokes
field guide to bird songs: western region. Time Warner
Audio Books, New York, USA.

Derrickson K. C. and R. Breitwisch. 1992. Northern
Mockingbird (Mimus polyglottos). The birds of North
America. Number 7.

Johnson, R. R., B. T. Brown, L. T. Haight, and J. M.
Simpson. 1981. Playback recordings as a special avian
censusing technique. Studies in Avian Biology 6:68-
75.

Oberholser, H. C. 1974. The bird life of Texas. University
of Texas Press, Austin, USA.

Rylander, M. K. 1981. Vocal mimcry in the Curve-billed
Thrasher. Bulletin of the Texas Ornithological Society
14: 29-30.

Tweit, R. C. 1996. Curve-billed Thrasher {Toxostoma
cun’irostre). The birds of North America. Number 235.

Vleck, C. M. 1993. Hormones, reproduction, and behavior
in birds of the Sonoran Desert. Pages 73-86 in Avian
endocrinology (P. J. Sharp, Editor). Journal of
Endocrinology Ltd., Bristol, United Kingdom.

The Wilson Journal of Oniilhology 122(3):627-633, 2010

Ornithological Literature

Robert B. Payne, Review Editor

HANDBOOK OF THE BIRDS OF THE
WORLD. Volume 14. Bush shrikes to Old World
span-ows. Edited by J. del Hoyo, A. Elliott, and D.
Christie, 2009. Lynx Edicions, Barcelona, Spain.
2009: 893 pages, 51 color plates, many photo-
graphs and maps. ISBN: 978-84-96553-50-7. 212
Euros (—$310.00 US) (cloth). — The volume
continues the series on birds of the world with
general accounts on all families and species of
birds. This volume includes the Malaconotidae
(bushshrikes), Prionopidae (helmetshrikes), Van-
gidae (vangas), Dicruridae (drongos), Callaeidae
(New Zealand wattlebirds), Notiomystidae
(Stitchbird), Grallinidae (mudlarks), Struthideidae
(formerly Corcoracidae, Australian mudnesters),
Artamidae (woodswallows), Cracticidae (butcher-
birds), Pityriaseidae (Bristlehead), Ptilonorhyn-
chidae (bowerbirds), Paradisaeidae (birds-of-para-
dise). Corvidae (crows), Buphagidae (oxpeckers),
Stumidae (starlings), and Passeridae (Old World
sparrows).

The text includes a summary of recent molecular
studies on systematic relationships, descriptions of
each species and subspecies, their behavior and
ecology, a list of references, and an index to the
common and scientific names. The text was written
by ornithologists who have conducted fieldwork on
these birds and are familiar with them. All species
are well illustrated in the color plates and many are
shown in great supporting photographs. The family
accounts describe and compare the highlights of
the species accounts, which include taxonomy,
subspecies and distribution including a map,
habitat, food and feeding, breeding, movements,
survival, status and conservation concern, and a
bibliography. The length of the species accounts
varies with what is known about the birds and with
the style of the authors.

The Madagascar vangas are amazingly diverse
in morphology and most species are not well
known. Biological information on these endan-
gered birds might help in conservation plans; but
the continuing loss of habitat in Madagascar, even
in sites with nominal protection, is the main
challenge to their survival.

Cooperative breeding is widespread in butcher-
birds, woodswallows, and Australian mudnesters.

and the text describes the Rowleys’ previously
unpublished study of Australian Magpie {Gym-
norhina tibicen). Courtship displays of bowerbirds
and birds-of-paradise are well described and
illustrated in photographs. The extreme display
structures and decorations of bowerbirds include
the maze-like four-walled bower of Yellow-
breasted Bowerbird (Chlaniydera lauterbachi)
and the ornamental feathers of King-of-Saxony
Bird-of-Paradise (Pteridophora alberti) in the
bower of Archbold’s Bowerbird (Archboldia
papuensis). Bowerbird young remain in the nest
for 17-21 days, except Archbold’s Bowerbird
which live in cold and wet high-altitude habitats
and remain nestlings for 30 days. The 42 species
of birds-of-paradise are fascinating in their sexual
dimorphism and elaborate displays. None has
become extinct since 1600, but three are globally
threatened and others are declining in numbers in
the wild, especially due to hunting. Conservation
efforts have had success in protecting certain
populations from hunting and habitat loss. Cap-
tive-breeding programs now are feasible and are
recommended while the wild populations are still
healthy.

The crows are the most species-rich group in
Volume 14 with 24 genera, 123 species, and 359
taxa. The section on this family could stand alone
as a readable book. Corvids have more than their
share of cooperatively-breeding species; the best
known is Elorida Scrub-Jay {Aphelocoma coer-
idecens). Some corvids have a complex layer of
social organization with the pair, the family
group, and the flock (Red-billed Chough [Pvr-
rhocorax pyrrhocorax]). Resident singles or pairs
of Common Raven (Con’iis corox) chase single
intruders from carcasses but, when the intruders
recruit other vagrants to join them, the group can
overcome the aggressiveness of the resident
territorial birds. Ravens, American Crow (C.
brachyrhynchos), and Rook (C. fnigUegus) each
have more than 20 distinct calls of known
function. The geographic variation in calls of
Large-billed Crow (C. macrorhynchos) requires
further study. Some corvids including the Com-
mon Raven are good vocal mimics of other
species in captivity and others are mimics in the

627

628

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122. No. 3, September 2010

wild (Eurasian Jay [Garnilus glandariiis])\ the
function of mimicry in wild corvids remains a
mystery. Common Ravens associate with gray
wolf (Canis lupus) in Yellowstone National Park,
USA and the scavenging corvids get early access
to the carcass at wolf kills. Food-storing and
retrieval is common in several corvids; one,
Clark’s Nutcracker (Nucifraga columbiana), bur-
ies pine (Pinus spp.) seeds in places that lose their
snow cover at different times in spring allowing
nutcrackers to have a steady supply of seeds
through a season. Some crows are common and
widespread, but the Hawaiian Crow (Cor\ms
hawaiiensis) is now extinct in the wild; new birds
from a captive breeding group may be re-
introduced in the future, although birds reared
and released in the 1990s soon died from
predation and disease.

Starlings have a wide range of breeding
organizations, some are cooperative breeders,
some are densely colonial, and some are polyg-
ynous, all much like the unrelated New World
blackbirds (Icteridae). Some Passeridae sparrows
(here treated as 40 species), are highly colonial, as
up to 20 nests of Spanish Sparrow {Passer
hispaniolensis) have been built in a single nest
of White Stork (Ciconia ciconia), and colonies of
this sparrow have as many as 800,000 pairs.

This volume includes a Foreword that discusses
the influence of bird watchers on the interest in
and concern for birds. “Birding past, present and
future — a global view” compares the roles of
great men and social change in times past,
advances in field guides and the web as
information resources, advances in optics, cam-
eras, and recording devices, and the availability of
local and overseas tours to see birds. The
demographics and economic impact of birders
have been surveyed mainly in the U.S. and U.K.
In the U.S., 20 million people consider themselves
birders, getting into the field at least 50 times a
year. The economics of birding may amount to
US$31.7 billion, counting vehicles, fuel, guides,
food and lodging including ecotours, plus $1.9
billion on optics and $2.6 billion on nest boxes,
feeders and bird food. Birds benefit from birders
in projects for conservation, and a concern for
responsible ecotourism includes training local
guides and support of birding places. Birding is
promoted as exciting and intellectually challeng-
ing, and a passion for the natural world that is
thought to benefit our physical and emotional
health and wellbeing.

The volumes of the Handbook of the Birds of
the World are written clearly and well illustrated,
and they are the highest standard source of
information on birds. When Volume 16 is
published in late 201 1, the series will be the first
work to describe in words and illustrations all
species of birds in the world. — ROBERT B.
PAYNE, Professor Emeritus, University of
Michigan, 1306 Granger Avenue, Ann Arbor,
MI 48104, USA; e-mail: rbpayne@umich.edu

GROUSE OF PLAINS AND MOUNTAINS:
THE SOUTH DAKOTA STORY. By Lester D.
Flake, John W. Connelly, Thomas R. Kirschen-
mann and Andrew J. Lindbloom. South Dakota
Department of Game, Fish and Parks, Pierre,
South Dakota, USA. 2010: xv -i- 246 pages, and
~173 color photographs and 27 graphs. ISBN:
9780615350158. $15.00 (Soft cover). — This lav-
ishly illustrated book with color photographs and
graphs is intended for the general public but is
prepared to also appeal to biologists, scientists,
managers, and anyone interested in grouse in the
prairies of North America. The authors are well
known for their interest in grouse and several have
impressive research and teaching credentials. This
is a book about four species of grouse native to
South Dakota: Sharp-tailed Grouse {Tympanuchus
phasianellus). Greater Prairie-Chicken (T. cu-
pido). Greater Sage-Grouse {Centrocercus uro-
phasianus), and Ruffed Grouse {Bonasa umbel-
lus). The latter is the least expected as one does
not associate aspen {Populus spp.) forests with the
grasslands and shrublands of the Great Plains.
Greater Sage-Grouse have the smallest distribu-
tion at present in South Dakota followed by
Ruffed Grouse, Greater Prairie-Chicken, and
Sharp-tailed Grouse with the most extensive range
across the state. All four species are hunted with
modest (sage-grouse) to expansive seasons (the
other 3 species).

This book is based on science and accumulated
field experience. It has 13 chapters, six appendi-
ces, and an extensive Literature Cited. The
pre.sentation is pleasant and easily understood
but the information is supported by citation of the
scientific literature. The chapters are: 8outh
Dakota’s Grouse Species, Physical Characteris-
tics, Behavior, Mobility and Habitat, Nutritional
Needs and Food Habits, Nesting Ecology, Broods,
Survival and Mortality, Population Monitoring,
Hunting Seasons, Grouse Hunting, Habitat Loss,

OR N ITHOLOG ICA L LITER A I'U R E

629

and Further Thoughts. Each chapter is heavily
illustrated (only color photographs and graphs are
used), covers each of the species, and ends with a
Review (= summary).

Each chapter is interesting and well prepared
and there is something for everyone. I was
interested in all of the material but especially
the material (pages 3-4) on Dusky Grouse
(Demimgapiis obscurus), which may have been
present historically, although poorly conducted
transplants failed to establish a population after
releases in 1969 and 1974. Greater Sage-Grouse
are at the extreme eastern edge of their current
(only 3 counties in South Dakota) and historical
range in South Dakota. One wonders how this
species survives in areas with low densities of big
sagebrush {Artemisia tridentata subspp.). Silver
sagebrush {A. cana) obviously will support sage-
grouse; the presence of robust, taller grasses with
more cover is likely important in supporting sage-
grouse at the eastern and northern peripheries of
their distribution. Ruffed Grouse are also an
anomaly as they appear to have low dispersal
capability but somehow found their way to the
isolated Black Hills (most likely from the west).
Management of public lands (USDA, Eorest
Service) in the Black Hills for ponderosa pine
{ Finns ponderosa) does not benefit Ruffed
Grouse, which has more affinity for species of
aspen and other deciduous shrubs. Relatively little
is known about Ruffed Grouse in South Dakota,
which the authors acknowledge but try to fill the
data gaps with published information from
scientific studies in other states.

The authors are candid about the role of hunting
and are cautious about possible population effects
(with literature citations) on grouse. There are two
chapters (10 and II) that discuss data that can be
obtained from hunter-harvested birds, as well as the
joy of being afield with hunting dogs in the fall.
They describe how data are collected (Chapter 9)
and provide actual data on counts in spring as well
as harvest estimates. They are quick to note the
inadequacies of some surveys but do not discuss
how state wildlife agencies place priorities on how
funds are spent (obviously on species where hunter
interest is high and where grouse populations are
large). The photographs amply demonstrate that
grouse also have value for bird watchers, especially
in spring during breeding seasons.

Scattered throughout the text and focused in the
final chapter (13), the authors discuss some of the
habitat issues involved with grouse populations.

These include everything from West Nile virus
(pond development for livestock and energy
production which may enhance mo.squito distri-
bution) to livestock grazing, federal land conser-
vation programs, prairie and forest management,
fences and roads to wind turbine farms. These
issues are not restricted to South Dakota but occur
throughout the Great Plains and the Intermountain
West.

I was hard pressed to find factual errors and any
editorial glitches, but did find Dalke italicized on
page 61, and a missing parenthesis on page 73.
Use of Jargon such as ‘input’ is relatively minor
throughout. With rare exceptions (i.e., page 208,
hectares and acres), the authors do not use metric
terms and we read about pounds and inches and
miles. Several references in the Literature Cited
(T. D. I. Beck and J. L. Beck, M. M. McDonald
and J. E. McDonald) are not alphabetical. A nicer
problem is that there may be too many similar
photographs. One weakness is there is no Index.

The stated goal of this book (Eoreword) is “to
foster increased interest and appreciation for
South Dakota’s four grouse species and the
habitats that support them.’’ The book clearly
meets this goal as it is very attractive, the material
is accurate and current, it is well written and easy
to read, and the color photographs keep inviting
you to turn the next page. This book is a must for
everyone interested in grouse and the northern
Great Plains, and I highly recommend it. —
CLAIT E. BRAUN, Grouse Inc., 5572 North
Ventana Vista Road, Tucson, AZ 85750, USA;
e-mail: sg-wtp(5)j uno.com

FARMLAND BIRDS ACROSS THE WORLD.
Edited by Wouter van der Weijden, Paul Terwan,
and Adriaan Guldemond. Lynx Edicions, Barce-
lona, Spain. 2010: 144 pages and 500 -t- color
photographs, maps, and figures. ISBN: 13-978-
84-96553-63-7. $39.00 (hardcover). — Books that
aim to highlight conservation issues can .some-
times be downers that bring to mind Aldo
Leopold’s famous quote: “Being an ecologist is
living in a world of wounds.” Others can tend to
act as generally uniriendly encyclopedias of
inlormation, useful only to con.servation planners.
Farmland Birds Across the World is a unique
creation. It highlights the importance of agricul-
tural habitats to global birdlife in a way that is
accessible, informative, and engaging to the
casual reader, ecologist, and conservationist alike.

630

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 3, September 2010

It does so using more than 500 stunning large
format photographs of birds and the agricultural
landscapes they inhabit that invite the reader on a
journey around the globe. It is authored by seven
experts in biology and links the beauty and natural
history of birdlife to the challenging and often
complex conservation issues facing birds in
agricultural landscapes. One unique way they
convey these stories is by peppering the chapters
with separate boxed-off panels highlighting spe-
cific conservation issues (e.g., declines in Asian
vultures due to veterinary drugs), conservation-
related natural history stories (e.g., the journey of
the Black-tailed Godwit [Limosa lirnosa]) and
interesting on-the-ground conservation efforts
(e.g., novel planting schemes that benefit sky-
larks). This makes the book easy to pick up, even
for just a moment.

The introduction is a short overview that
highlights the complex, mutual relationships
between farming and birds that can be both
beneficial and negative. It highlights interesting
paradoxes. For example, agriculture and its
intensification represent perhaps the most impor-
tant threat to bird populations globally, yet
agricultural lands also represent the sole remain-
ing habitat for many critically endangered species.
Chapter 2 effectively begins the body of this book
and represents an overview of issues related to
farming such as factors affecting birds in crops,
resources that crops can provide to birds, the
ecosystem services provided by birds to crops,
crop damage by birds, and how farming can
adversely impact birds.

The next six chapters focus on birdlife in specific
agricultural environments: grasslands, cereal and
other annual crops, rice fields, orchards and
plantations, coffee and cacao, and farmyards. The
first page of each chapter contains a global map of
the distribution of each agricultural crop and some
brief statistics impressively illustrating the global
nature and extent of agriculture. For example,
cereals are the world’s most important calorific
crop, occupying approximately one-third of all
arable land. The authors highlight how changes in
the area and cultivation intensity of crops, the
birdlife exploiting the.se crops, and the threats to
bird populations vary by region. Notably, they
draw on massive global bird and agricultural data
bases to create figures illustrating the worldwide
trends in human activity shaping bird populations.

Each chapter ends with another unique feature,
a bulleted .section outlining different conservation

challenges and opportunities associated with
issues like global climate change, agricultural
policy, farming practices and, at times, a surpris-
ing and conspicuous lack of information. For
example, the authors (page 67) mention, “Al-
though China is the world’s leading rice producer,
little is known on birdlife in its rice fields than the
occurrence of herons and egrets.’’

The final chapter discusses the future prospects
for farmland birds, summarizing some of the
grand conservation challenges and highlighting
interesting and unique solutions to those prob-
lems. Collectively, this volume manages to
convey the authors’ excitement at the richness
of farmland birdlife, their confusion due to the
paradoxes associated with bird conservation and
agricultural production, their fear at the often
overwhelming threats facing farmland bird popu-
lations, and finally relief there are potential
solutions to improve the fortunes of farmland
birds.

My only significant complaint about this
volume was the limited treatment given to the
potential importance of biofuel production on
avian biodiversity. The authors do give significant
mention of the widespread impacts that oil-palm
{Elaeis spp.) is having on tropical birds. However,
second generation cellulosic biomass crops that
may include cereals, trees and woody plants, and
other crops have potential to radically reshape
agricultural landscapes worldwide. To be fair,
little is known about the ecological effects of
many of these crops, but I felt the threat of
biofuels to farmland birds should have made lists
of conservation threats.

It is estimated that over one-third of the earth’s
surface has now been converted to agricultural
uses. This makes the book a particularly timely
resource. A notable asset of this volume is that it
does a seamless job of integrating scientific ideas
into the chapters in a way that is accessible to
non-scientists. It is this feature that makes the
book so broadly appealing, yet useful even to
con.servation scientists. Biologists often think of
themselves as storytellers and the authors of
Farmland Birds Across the World have succeeded
in telling fascinating and important stories that
instill a .sense of global responsibility for the birds
that are impacted by the growing of our food,
fiber, and fuel. — BRUCE A. ROBERTSON,
Research Associate, Great Lakes Bioenergy
Research Center, Kellogg Biological Station,
Michigan State University, 3700 East Gull

ORNITHOLOGICAL LITERATURE

631

Lake Drive, Hickory Corners, Ml 49060, USA;
e-mail: brucerobertson@hotmail.eom

BIRDS OF EUROPE, SECOND EDITION. By
Lars Svensson. Princeton Field Guides, Princeton
University Press, Princeton, New Jersey, USA.
2009: 448 pages, 3,500+ color illustrations.
Illustrations and captions by Killian Mullarney
and Dan Zetterstrom. ISBN: 978-0-691-14392-7.
$29.95 (paperback). — The text covers all 772
species ot birds occuiTing in Europe as well as 32
introduced species or variants and 118 vagrant
species. In the United Kingdom, where it was
published by Harper/Collins, it is known as the
"Collins Bird Guide." The book is a substantial
revision of the author’s first edition (1999). The
book includes all bird species that breed or
regularly occur in Europe, Africa north of 30°
N, the Sinai Peninsula, the Middle East through
Turkey, Israel, Jordan, Syria, Georgia, and
Azerbaijan, and the Canary Islands and Madeira.
Europe is bordered on the east by the Ural
Mountains, Ural River, and Caspian Sea.

The text and distribution maps are on one page
and the illustrations are on the facing page. For
each species the text has a description of shapes,
male and female, young and old, breeding and
nonbreeding, and geographic variation where
appropriate. The text is clear, concise, and
accessible, and generally more detailed and
comprehensive than in North American field
guides. For example, in one of the longer species
accounts, the description of Rough-legged Hawk/
Rough-legged Buzzard (Biiteo lagopus) has 441
words, each well chosen. Key identification
features are highlighted in italics, variations in
plumage within a population are shown and
described, and differences between similar species
are noted; e.g., the wing pattern in flight is helpful
in identification of larks. Habitat, range, and
breeding status are described with notes on
frequent hybridization. Many species have useful
descriptions of feeding behavior and breeding
biology. The Middle Spotted Woodpecker (Den-
drocopos medium), unlike other related woodpeck-
ers, calls but does not drum in territorial assertion.
Voice is described in words, often with cadence-
and pitch-perfect transliteration: the species-
typical call of Rock Ptarmigan {Lagopus niuta)
is “here comes the bride’’. Species status
(breeding, resident/migratory, abundance) is noted
in short form for Britain and Ireland. Each family

has an informative account with comments such
as the night vision of owls is not “surreal’’ but
requires light from the moon or stars. Other
useful sections are included on seawatching (and
how to point out birds on a featureless sea.scape to
other watchers), raptors in flight, and general
advice for identification of waders/shorebirds and
gulls.

The quality of the illustrations is very good, the
birds lifelike, the details precise, the selection
comprehensive. Many birds are illustrated in more
than one posture and in flight. The color plates
include in small type many useful comments on
identification. Because more than one standard
taxonomic source is currently used in ornithology,
the species recognized in the book reasonably
follow “author’s preference’’ in recent revisions.
For example, in the Heiring Gull (Larus argen-
tatus) species complex, where 2 lA pages of color
plates have 67 images, several species are
recognized that previously were known as sub-
species; these are American Herring Gull (L.
smithsonianus). Yellow-legged Gull (L. luicha-
hellis), Armenian Gull (L. armenicus), Audoin’s
Gull (L. audouinii), and Caspian Gull (L.
cachinnans). Some distinctive forms recognized
as subspecies receive detailed accounts as well;
e.g., “Steppe Buzzard’’ {Buteo buteo vulpinus).
The owls are particularly well illustrated with new
images for this edition, and the comments on
these birds (“stately look’’, “deceptively gentle
look’’, “grim look’’, “astonished look’’, “aus-
tere look’’) capture the impression so the reader
can identify the live owl in the field. Wallcreeper
(Tichodroma murario) is described as “The
Hoopoe \Upupa epops] of the rock face!”
Hawfinch {Coccothraustes coccothraustes) has a
big head; “the Hawfinch is a flying pair of nut-
crackers!’’ A few plumages are shown that do not
appear in North American field guides, e.g..
summer plumage of female Long-tailed Duck
(Clangula hyemolis), and some compari.sons of
North American species go beyond what is said in
North American field guides, e.g., adult Buff-
breasted Sandpiper (Tryngites suhruficolli.s) adults
are similar to (and distinguishable from) iuvenile
Ruff (Philomachus pugna.x). Most North Ameri-
can Cotharus thrushes are illustrated with spotted
Juvenile wing coverts, perhaps the appropriate
plumage if most vagrants that appear in Europe
are birds in their first autumn, but the text should
indicate the age in these birds, as it does for first-
winter New World warblers. Two pages on the

632

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 3. September 2010

pochards (Aythya spp.) show hybrids and the most
similar non-hybrids with notes for identification.

Vagrant species are illustrated along with
similar resident or breeding species, or are in the
back of the book, and accidentals are listed but not
described. Introduced species are in the main
section of the book (e.g., Canada Goose [Branta
canadensis}) or in the back of the book (e.g., the
estrildid finches with established populations).

The book is an excellent field guide, small
enough to use in the field, a clear and informative
read on birds, and clearly the best book for
identification of the birds of Europe. It also will
be useful in North America especially in identi-
fication of northern Palearctic vagrants in Alaska
and elsewhere. It is highly recommended. —
ROBERT B. PAYNE, Professor Emeritus,
University of Michigan, 1306 Granger Avenue,
Ann Arbor. MI 48104, USA; e-mail: rbpayne(s>
umich.edu

CONSERVATION BIOLOGY OE HAWAI-
IAN EOREST BIRDS. Edited by Thane K. Pratt,
Carter T. Atkinson, Paul C. Banko, James D.
Jacobi, and Bethany L. Woodworth. Yale Uni-
versity Press, New Haven, Connecticut, USA.
2009: 707 pages, 97 black and white plus 32 color
illustrations. ISBN: 978-0-300-14108-5. $85.00
(cloth). — Hawaiian forest birds are famous among
evolutionary biologists for their spectacular bill
morphology and bright plumage. They are equally
famous among conservation biologists for extinc-
tion, declines, and numerous threats. This book
touches on the former but focuses on the latter. It
appears to largely summarize the USGS, Biolog-
ical Research Discipline efforts from a recent
grant. This may explain why they ignore much
other research on Hawaiian birds.

The book is organized into five parts. Part 1 is
introductory and has three chapters that deal with
origins, historic decline and extinction, and forest
birds and Hawaiian culture. Part 2 on status,
biology, and limiting factors has 1 I chapters on
monitoring, trends, habitat loss and degradation,
food exploitation, life history and demography,
malaria and pox virus, genetics, mammalian
predators. Hawaiian raptors, introduced birds,
and epidemiology of malaria and pox. Part 3 is
about applying research to management, including
controlling invasive species, land.scape ecology,
managing disease, controlling small mammals,
and captive propagation. Part 4 deals with

recovery programs for four species, one of which
is likely extinct, a second is close behind, a third
is rapidly declining even though it has been
extensively studied for more than 20 years, and
the fourth is a success story. Part 5 deals with the
future, and covers the social and political
obstacles and asks the question “Can Hawaiian
forest birds be saved?” Space limits chapters in
the book that can be reviewed.

The origins and evolution chapter by Thane
Pratt is masterful historical biogeography, both in
understanding where Hawaiian birds came from
as well as initial conditions they likely encoun-
tered. We understand why these are the main
vertebrates of Hawaiian forests, and a non-random
set of potential colonists. They competed among
themselves, which led to adaptive radiation, but
did not compete with amphibians, reptiles,
mammals, or social insects.

There are two excellent chapters on disease.
One deals with modeling the epidemiology of
avian malaria and pox; the other concerns
managing disease. The former is a particularly
clear exposition of compartmentalized modeling
for understanding transmission dynamics in a
vector-transmitted disease such as malaria. The
chapter on managing disease considers diverse
techniques for controlling mosquitoes and pigs
(Sus scrofa) to break transmission, along with
immunizing the birds themselves and genetically
modifying mosquitoes. Unfortunately, these chap-
ters, as well as the one on genetics and
conservation, ignored our prize-winning and
widely cited 2005 paper (Condor 107: 753-764)
documenting an increase in avian malaria in
highest elevation forest where pigs were managed.
This is a clear example that mosquitoes docu-
mented in the forest may have originated from
unmanaged lands, a point emphasized by the
book.

No specific adaptations were identified in the
book, despite intensive studies that have been
performed by several research groups. Banko and
Banko, in their chapter on evolution and ecology
of food exploitation, deal with comparative bill
morphology as adaptation for particular diets, but
they ignore work documenting that the liwi
(Vestiaria coccinea) has undergone a niche shift
with accompanying changes in bill morphology.
Woodworth and Pratt, in their chapter on life
history and demography, ignore nestling over-
growth and seasonal variation in sex allocation in
the Hawaii Akepa (Loxops coccineus coccineiis).

ORNITHOLOGICAL LITERATURE

633

clear adaptations tor breeding during a seasonal
decline in food. These have been dismantled by
competition with introduced Japanese White-eyes
(Zosterops japonicus), documented in Evolution-
ary Ecology Research 10: 931-965 (2008) ( =
EER 2008), referenced and dismissed in the book.

The Banko and Banko chapter on food
essentially ignored arthropod prey in Metrosi-
deros polymorpha trees for insectivorous birds.
This is the most important tree in wet to mesic
Hawaiian forests, and EER 2008 shows that most
birds use foraging substrates associated with the
tree. They refer to an old dissertation, not the
papers that document the diversity of prey with
seasonal pattern of variation. A reader might think
that Scotorythrci caterpillers are the most impor-
tant food resource for all bird species. Like other
islands, Hawaii has a preponderance of spiders.

The chapters in the book on monitoring forest
birds and identifying trends are thorough but
misleading. The authors fail to recognize analysis
of residuals as essential when modeling changes
in populations over time. A regression model
requires residual analysis to ensure the model fits
the data. If it does not, the model needs to be
corrected. The trends reported for Hakalau Eorest
National Wildlife Refuge are incon'ect, as shown
in our paper in the May 2010 issue of Condor
(112: 213-221), because their residuals indicate
significant lack of fit. The endangered Hawaii
Akepa and the liwi, a species of concern, are
significantly declining on the refuge. An example
in the book that clearly indicates the lack of fit is
Eig. 25.6 in the chapter on saving Hawaiian birds.
The endangered Hawaii Creeeper (Oreoinystis
mcmci) has a 21 -year trend that looks positive, but
residuals are negative in the early and late years of
the series, and positive in the middle, indicating

lack of lit. The creeper is declining, not
increasing, in models with an additional param-
eter with appropriate residuals.

The chapter by Jeff Eoster on introduced birds
provides a marvelous historical and ecological
perspective. Unfortunately, he dismisses EER 2008
because of the trend analysis. He argues that food
limitation can only be established by measuring
food. However, ornithologists have long recog-
nized that food limitation can be inferred by
stunted growth of nestlings, fault bars in wing
and tail feathers that reflect nutritional stress during
molt, lower fat, and changes in begging patterns.

Woodworth and Pratt in the life history and
demography chapter have a section on human-
caused changes in demographic rates, but only
report on nesting success. We (EER 2008),
document significant change in juvenile survival
and adult survival in the Hawaii Akepa associated
with increase in numbers of introduced Japanese
White-eyes. Chewing lice (Phthiraptera) are not
even mentioned in the book although they
contribute to parasite-mediated competition.

Most chapters in this book have to be read
critically, because they ignore important literature
and accept misleading trend analysis. The answer
to “Can Hawaiian forest birds be saved?” is
“No” unless the immediate threat is recognized
and managed. That threat is an introduced bird,
which has stunted growth and lowered survival
throughout the community (2009. Cun-ent Biolo-
gy 19: 1736-1740). Malaria will kill remaining
birds that are food limited because they will be
unable to mount energetically expensive immune
responses.— LEONARD A. FREED, Depart-
ment of Zoology, University of Hawai’i,
Honolulu, HI 96822, USA; e-mail: Ifreedt®
hawaii.edu

THE WILSON JOURNAL OF ORNITHOLOGY

Editor CLAIT E. BRAUN

5572 North Ventana Vista Road
Tucson, AZ 85750-7204
E-mail: TWILSONJO@comcast.net

Editorial NANCY J. K. BRAUN
Assistant

Editorial Board RICHARD C. BANKS

JACK CLINTON EITNIEAR
SARA J. OYLER-McCANCE
LESLIE A. ROBB

Review Editor ROBERT B. PAYNE

1306 Granger Avenue
Ann Arbor, MI 48104, USA
E-mail: rbpayne@umich.edu

GUIDELINES FOR AUTHORS

Please consult the detailed "Guidelines for Authors” found on the Wilson Ornithological Society web site
(http://www.wilsonsociety.org). All manuscript submissions and revisions should be sent to Clait E. Braun,
Editor, The Wilson Journal of Ornithology, 5572 North Ventana Vista Road, Tucson, AZ 85750-7204, USA.
The Wilson Journal of Ornithology office and fax telephone number is (520) 529-0365. The e-mail address is
TWilsonJO@comcast.net

NOTICE OF CHANGE OF ADDRESS

Notify the Society immediately if your address changes. Send your complete new address to Ornithological
Societies of North America, 5400 Bosque Boulevard, Suite 680, Waco, TX 76710.

The permanent mailing address of the Wilson Ornithological Society is: %The Museum of Zoology, The
University of Michigan, Ann Arbor, MI 48109, USA. Persons having business with any of the officers may
address them at their various addresses given on the inside of the front cover, and all matters pertaining to the
journal should be sent directly to the Editor.

MEMBERSHIP INQUIRIES

Membership inquiries should be sent to Timothy J. O'Connell, Department of Natural Resources Ecology and
Management, Oklahoma State University, 240 AG Hall, Stillwater, OK 74078; e-mail: oconnet@ okstate.edu

THE JOSSELYN VAN TYNE MEMORIAL LIBRARY

The Josselyn Van Tyne Memorial Library of the Wilson Ornithological Society, housed in the University of
Michigan Museum of Zoology, was established in concurrence with the University of Michigan in 1930. Until
1947 the Library was maintained entirely by gifts and bequests of books, reprints, and ornithological magazines
from members and friends of the Society. Two members have generously established a fund for the purchase
of new books; members and friends are invited to maintain the fund by regular contribution. The fund is
administered by the Library Committee. Jerome A. Jackson, Florida Gulf Coast Univeristy, is Chairman of the
Committee. The Library currently receives over 200 periodicals as gifts and in exchange for The Wilson Journal
of Ornithologv. For information on the Library and our holdings, see the Society’s web page at http://
www.wilsonsociety.org. With the usual exception of rare books, any item in the Library may be borrowed by
members of the Society and will be .sent prepaid (by the University of Michigan) to any address in the United
States, its possessions, or Canada. Return postage is paid by the borrower. Inquiries and requests by borrowers,
as well as gifts of books, pamphlets, reprints, and magazines, should be addressed to: Josselyn Van Tyne
Memorial Library, Mu.seum of Zoology, The University of Michigan, 1109 Geddes Avenue. Ann Arbor, Ml
48109-1079. USA. Contributions to the New Book Fund should be sent to the Treasurer.

This issue of The Wilson Journal of Ornithology was published on I September 2010.

Continued from outside back cover

569 Annual bird mortality in the bitumen tailings ponds in northeastern Alberta, Canada
Kevin P. Timoney and Robert A. Ronconi

577 Abundance and distribution of waterbirds in the Llanos of Venezuela
Francisco J. Vilella, Mark S. Gregory, and Guy A. Baldassarre

Short Communications

578 Piping Plover foraging distribution and prey abundance in the pre-laying period
Jonathan B. Cohen and James D. Fraser

583 Interspecific song imitation by a Cerulean Warbler

Than J. Boves, David A. Buehler, and Phillip C. Massey

587 Nests, nest placement, and eggs of three Philippine endemic birds

Luis A. Sdnchez-Gonzdlez, Carl Oliveros, Nevong Puna, and Robert G. Moyle

592 Nests, eggs, and young of the Azure-crowned Hummingbird {Amazilia cyanocephala)

Juan Francisco Ornelas

597 Nest description of the Garden Emerald {Chlorostilbon assimilis) from Costa Rica
Luis Sandoval and Ignacio Escalante

600 The nest of the Cipo Canastero {Asthenes luizae), an endemic furnariid from the Espinha90 Range,
southeastern Brazil

Henrique Belfort Gomes and Marcos Rodrigues

604 Nest box use by Great Tits in semi-arid rural residential gardens
Motti Charter, Yossi Leshem, Shay Halevi, and Ido Izhaki

608 Analysis of nest sites of the Resplendent Quetzal {Pharomachrus mocinno)\ relationship between nest
and snag heights

Dennis G. Siegfried, Daniel S. Linville, and David Hide

612 Parasitism of a Blue- winged Teal nest by a Northern Shoveler in South Dakota.

Thomas E. Lewis and Pamela R. Garrettson

614 Nest predators of ground-nesting birds in montane forest of the Santa Catalina Mountains, Arizona
Chris Kirkpatrick and Courtney J. Conway

618 Feeding a foreign chick; a case of a mixed brood of two Tit species
Toshitaka N. Suzuki and Yuko Tsuchiya

620 Infanticide by an Eastern Phoebe

Gary Ritchison and Brandon T. Ritchison

623 Marsh Wren eats small fish

Andrea J. Ayers and James W. Armacost Jr.

625 Occasional mimicry and night-time singing by the western Curve-billed Thrasher ( Toxostoma
curvirostre palmer!)

R. Roy Johnson and Lois T Haight

627 Ornithological Literature

Robert B. Payne, Book Review Editor

The Wilson Journal of Ornithology

(formerly The Wilson Bulletin)

Volume 122, Number 3 CONTENTS September 2010

Major Articles

417 Reproductive success and nestling growth of the Baywing parasitized by Screaming and Shiny cowbirds
Maria C. De Mdrsico, Bettina Mahler, and Juan C. Reboreda

432 Bobolink egg mass variability and nestling growth patterns
Barbara Frei, David M. Bird, and Rodger D. Titman

439 Breeding phenology and nesting success of the Yucatan Wren in the Yucatan Peninsula, Mexico
Jesus Vargas-Soriano, Javier Salgado Ortiz, and Griselda Escalona Segura

447 Breeding biology and natural history of the Slate-throated Whitestart in Venezuela

Roman A. Ruggera and Thomas E. Martin — ^

455 Reproductive biology of a grassland songbird community in northcentral Montana

Stephanie L. Jones, J. Scott Dieni, and Paula J. Gouse - j

465 Mountain biking trail use affects reproductive success of nesting Golden-cheeked Warblers 3

Craig A. Davis, David M. Leslie Jr., W. David Walter, and Allen E. Graber 1

475 Survival, site fidelity, and population trends of American Kestrels wintering in southwestern Florida J

Daniel M. Hinnebusch, Jean-Frangois Therrien, Marc-Andre Valiquette, Bob Robertson, Sue Robertson, and 4
Keith L. Bildstein

484 Temporal and spatial shifts in habitat use by Black Brant immediately following flightless molt
Tyler L. Lewis, Paul L. Flint, Joel A. Schmutz, and Dirk V Derksen

494 Arthropod foraging by a southeastern Arizona hummingbird guild

Donald R. Powers, Jessamyn A. Van Hook, Elizabeth A. Sandlin, and Todd J. McWhorter

503 Army ant raid attendance and bivouac-checking behavior by neotropical montane forest birds
Sean ODonnell, Anjali Kumar, and Gorina Logan

513 Maroon-fronted Parrot {Rhynchopsitta terrisi) breeding home range and habitat selection in the
northern Sierra Madre Oriental, Mexico

Sonia Gabriela Ortiz-Maciel, Consuelo Hori-Ochoa, and Ernesto Enkerlin-Hoejlich

518 Short-term effects of fire on breeding birds in southern Appalachian upland forests
Nathan A. Klaus, Scott A. Rush, Tim S. Keyes, Johyi Petrick, and Robert J. Cooper

532 Avian communities of the Altamaha River estuary in Georgia, USA
Ross A. Brittain, Vicky J. Meretsky, and Chris B. Craft

545 Influence of cover and food resource variation on post-breeding bird use of timber harvests with
residual canopy trees

Molly E. McDermott and Petra Bohall Wood

556 Distribution, abundance, and status of Cuban Sandhill Cranes {Grus canadensis nesiotes)

Xiomara Galvez Aguilera and Felipe Chavez-Ramirez

563 The feather structure of Dippers; water repellency and resistance to water penetration
Arie M. Rijke and William A. Jesser

Continued on inside hack cover

Wl/-

Wilson Journal

of Or ni tholog^

Volume 122y Number 4, December 2010

Published by the
Wilson Ornithological Society

THE WILSON ORNITHOLOGICAL SOCIETY
FOUNDED 3 DECEMBER 1888

Named after ALEXANDER WILSON, the first American ornithologist.

President — E. Dale Kennedy, Biology Department, Albion College, Albion, MI 49224, USA; e-mail:
dkennedy@aIbion.edu

First Vice-President— Robert C. Beason, P. O. Box 737, Sandusky, OH 44871, USA; e-mail;
Robert.C.Beason@gmaiI.com

Second Vice-President — Robert L. Curry, Department of Biology, Villanova University, 800 Lancaster
Avenue, Villanova, PA 19085, USA; e-mail: robert.curry@viIIanova.edu

Editor — Clait E. Braun, 5572 North Ventana Vista Road, Tucson, AZ 85750, USA; e-mail: TWILSONJO@
comcast.net

Secretary — John A. Smallwood, Department of Biology and Molecular Biology, Montclair State University,
Montclair, NJ 07043, USA; e-mail: smallwoodj@montclair.edu

Treasurer — Melinda M. Clark, 52684 Highland Drive, South Bend, IN 46635, USA; e-mail: MClark@tcservices.biz

Elected Council Members — Jameson F. Chace, Sara R. Morris, and Margaret A. Voss (terms expire
2011); Mary Bomberger Brown, Mary Garvin, and Mark S. Woodrey (terms expire 2012); Mark
Deutschlander, Paul G. Rodewald, and Rebecca J. Safran (terms expire 2013).

Living Past-Presidents — Pershing B. Hofslund, Douglas A. James, Jerome A. Jackson, Clait E. Braun,
Mary H. Clench, Richard C. Banks, Richard N. Conner, Keith L. Bildstein, Edward H. Burtt Jr., John
C. Kricher, William E. Davis Jr., Charles R. Blem, Doris J. Watt, and James D. Rising.

Membership dues per calendar year are: Regular, $40.00; Student, $20.00; Family, $50.00; Sustaining,
$100.00; Life memberships, $1,000.00 (payable in four installments).

The Wilson Journal of Ornithology is sent to all members not in arrears for dues.

THE WILSON JOURNAL OF ORNITHOLOGY
(formerly The Wilson Bulletin)

THE WILSON JOURNAL OF ORNITHOLOGY (ISSN 1559-4491) is published quarterly in March, June,
September, and December by the Wilson Ornithological Society, 810 East 10th Street, Lawrence, KS 66044-8897. The
subscription price, both in the United States and elsewhere, is $40.00 per year. Periodicals postage paid at Lawrence, KS.
POSTMASTER: Send address changes to OSNA, 5400 Bosque Boulevard, Suite 680, Waco, TX 76710.

All articles and communications for publication should be addressed to the Editor. Exchanges should be addressed
to The Josselyn Van Tyne Memorial Library, Museum of Zoology, Ann Arbor, MI 48109, USA.

Subscriptions, changes of address, and claims for undelivered copies should be sent to OSNA, 5400 Bosque
Boulevard, Suite 680, Waco, TX 76710, USA. Phone: (254) 399-9636; e-mail: business(gosnabirds.org. Back issues or
single copies are available for $ 1 2.00 each. Most back issues of the journal are available and may be ordered from OSNA.
Special prices will be quoted for quantity orders. All issues of the journal published before 2000 are accessible on a free
web site at the University of New Mexico library (http://elibrary.unm.edu/sora/). The site is fully searchable, and full-text
reproductions of all papers (including illustrations) are available as either PDF or DjVu files.

© Copyright 2010 by the Wilson Ornithological Society
Printed by Allen Press Inc., Lawrence, KS 66044, USA.

COVER: Wilson’s ?\ovev {Characlrius wilsonia). Illustration by Don Radovich.

@ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).

MCZ

library

DEC 0 9 2010

HARVARD

UNIVERSITY

FRONTISPIECE. Henslow's Sparrows (Ammodramus henslowii) arc tallgrass prairie specialists in North America.
Analysis of yearly resettlement and abundance patterns from Breeding Bird Survey data indicated this species is less likely
to return to previously used breeding areas than other grassland sparrows. Restricting analyses to single observer-collected
data resulted in significant effects not detected in data aggregated from multiple observers during the same study period.
Acrylic painting by Marco Antonio Pineda M of Puebla, Mexico.

Wilson Journal

of Ornithology

Piihlished by the Wilson Ornithological Society

VOL. 122, NO. 4 December 2010 PAGES 635-858

The Wilson Journal of Ornithology 122(4):635-645, 2010

BREEDING PATTERNS OE HENSLOW’S SPARROW AND SYMPATRIC

GRASSLAND SPARROW SPECIES

L. LYNNETTE DORNAK'

ABSTRACT. Henslow’s Sparrows (Ammodramiis henslowii) are reported to have irregular patterns of return to
breeding areas. I present data supporting these reports at rangewide extents, while testing potential biases inherent in the
North American Breeding Bird Survey data. Two measures of population variability were used to show that Henslow's
Sparrows are less likely to use breeding areas predictably and consistently, but have similar variance in numbers at
occupied sites relative to other sympatric grassland sparrow species. 1 illustrate how restricting analyses to single-observer-
collected Breeding Bird Survey data results in subtle but significant effects not detected in data aggregated from multiple
observers through the study period. The most conservative analysis (single-observer, restricted distribution) showed that
Henslow s Sparrows exhibited lower prevalence of occurrence than Grasshopper (Ammodramiis savannarum) (P < 0.001 ) and
Savannah (Passercuhis sandwichensis) (P < 0.001) sparrows but no difference in variation of abundance (P > 0.05). These
results suggest Henslow’s Sparrows are not returning to previously used breeding habitat from year-to-year. Grassland
management should consider the behavior documented in this study and attempt to incorporate this facet of Henslow's Spairow
biology into decisions that involve broad-scale land.scape de.sign. Received 27 Janiiaiy 2010. Accepted 8 May 2010.

Artificial grazing regimes, drainage of wet-
lands, large-scale agriculture, and alteration of
natural fire regimes have changed North Ameri-
can landscapes, and left behind only relict tracts
of native prairie (Hyde 1939; Knopf 1988, 1994).
The full extent of these impacts on native flora and
fauna has only recently begun to be appreciated.
Shifts are likely occumng continent-wide, but the
most dramatically impacted habitat has been native
grasslands. Since European settlement, 99.9% of
native prairies have been lost (Samson and Knopf
1994) or altered (Vickery et al. 1994) in North
America, and they are now termed a critically
endangered habitat (Noss et al. 1995). Thirteen
bird species within this biome have incuired

' Department of Geography, University of Kansas,
Lawrence, KS 66045, USA; e-mail: picoide.s@ku.edu

serious declines, far suipassing those of any other
North American biome (Knopf 1994, Peterjohn
and Sauer 1999, Robbins et al. 2002, Powell
2006), probably the result of continued conversion
of prairie habitat to an artificial landscape.

Henslow’s {Amwodramu.s henslowii). Grass-
hopper [A. savanna rum), and Savannah (Pas.ter-
ciiliis sandwichensi.s) sparrows are obligate
grassland nesters (Vickery et al. 1994) with
population declines across part or all of their
breeding ranges (Peterjohn and Sauer 1999, Wells
and Rosenberg 1999). None has national threatened
or endangered species status, but Henslow’s and
Grasshopper sparrows have been listed as Birds of
Conservation Concern (USDl 2008) and Species
ot Continental Importance (Rich et al. 2004).
However, the species’ behavior and habitat use, as
it relates to landscape configuration, must be

635

636

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 4, December 2010

understood more fully before successful manage-
ment can be achieved.

Hyde (1939:23) was the first to note the
unpredictability of Henslow’s Sparrows appear-
ance in breeding areas, writing “its presence in a
given season cannot be certainly predicted,” a
sentiment that has been repeated by other authors
(Wiens 1969, Skipper 1998, Ingold et al. 2009).
Most studies report low (Skipper 1998, Monroe
and Ritchison 2005) or nonexistent site fidelity for
Henslow’s Sparrows in breeding areas (J. L.
Zimmerman cited in Pruitt 1996). This behavior is
atypical compared to Grasshopper and Savannah
sparrows which exhibit higher nest site fidelity
(Bedard and LaPointe 1984, Wheelright and
Rising 1993). These studies have used site fidelity
as a measure of return rates to breeding areas in con-
•secutive years, and have necessarily concentrated at
only a few sites. Thus, a broader-scale assessment
of annual resettlement patterns in breeding areas
is necessary, but is lacking in the literature.

Henslow’s Sparrows may be also erratic and
opportunistic in selection of sites for breeding on
a broad spatial scale. Previous research on this
species has focused at local and regional scales
with a few exceptions (e.g., Herkert 2007). The
only studies that used landscape-level analysis
were limited to sections of the species’ range
(Bajema and Lima 2001, Cunningham and
Johnson 2006, Thogmartin et al. 2006), and little
information exists to provide a range-wide
perspective for this species.

1 examined abundance data from the entire
breeding range for Henslow’s Sparrows, and
compared it to similar data for Grasshopper and
Savannah sparrows. The focus was not to compare
habitats used by each species; instead, 1 analyzed
patterns of resettlement in habitat the birds
already had assessed as suitable for nesting.
Grasshopper and Savannah sparrows were chosen
for comparison, not because of similarity or
dissimilarity of habitat choices, but because they
are obligate grassland species that nest within the
breeding range of Henslow’s Sparrow. 1 u.sed
North American Breeding Bird Survey (BBS) data
(Sauer et al. 2007) to address two objectives. First,
I examined how Henslow’s Sparrow populations
vary in time and space. A secondary objective was
to assess the impact of using different subsets of
BBS data because these data have been criticized
for inherent biases. This information is critical to
understanding this species’ natural history, as well
as conservation implications that may be derived.

1,700

FIG. I . Henslow’s (A), Grasshopper (B), and Savannah
(C) sparrow distributions from Breeding Bird Survey route
data collected during 2000-2007. Henslow’s, Grasshopper,
and Savannah sparrows were observed on 1 32; 1, 387; and
1,765 routes, respectively.

METHODS

Study Area. — The study area included the entire
breeding ranges of Henslow’s, Grasshopper, and
Savannah sparrows (Fig. 1). All bird abundance
data were derived from the BBS, a monitoring

Donutk • POPULATION DYNAMICS OF HENSLOW’S SPARROWS

637

system created in the 1960s with the goal of
understanding long-term trends in North Ameri-
can breeding bird populations (Sauer et al. 2007).
BBS data are collected annually, on fair-weather
June mornings, on 4,100 standardized roadside
census routes across the United States, Canada
and, most recently, Mexico. Sampling points are
spaced evenly along the survey route (Robbins et
al. 1986), which is 39.4 km in length; 50 sampling
points are located every 0.8 km along the route.
Observers record all birds seen or heard during
3 min at each stop. Data are available for all years
between 1966 and the present, but I used data
from 2000-2007 to assure maximal route density
and consistency.

Count Data. — I used two measures to describe
the magnitude of yearly fluctuations of resettle-
ment across ranges of these species: prevalence of
occurrence (proportion of years present) and
variation in abundance. The former describes
how consistently a species returns to a given BBS
route year after year, and is calculated as the ratio
of the number of years in which a species was
detected on a route to the number of times during
the study period the route was surveyed. The latter
measures year-to-year variation in number of
individuals of a species at the site and is estimated
as the coefficient of variation: the standard
deviation data divided by the mean abundance
for each route across all years in which the route
was surveyed. Routes sampled in only 1 year of the
study period were excluded from analyses because
the coefficient of variation was undefined. These
two variables describe the consistency of occur-
rence and abundance, but saturation effects of
variation in abundance may influence my results.

The BBS data were divided and analyzed in
four groups to consider potentially inherent biases
(geographic and observer). (1) BBS routes were
considered across the entire breeding range of
each species. (2) Analysis was constrained to BBS
routes within the recorded breeding range of
Henslow’s Sparrows. 1 also examined BBS route
data collected by single observers across all years
in the study period, removing possible biases
originating from differences in observer consis-
tency. These analyses were conducted (3) across
the entire range of each species, and (4) only
within the distributional area of Henslow’s
Sparrow. The smallest data set had 56 routes
and, to have equal samples size for each analysis,
I randomly selected 50 routes per species for each
data set.

Statistical Analysis. — Shapiro-Wilk normality
tests and Levene’s test for homogeneity of
variances were performed on all data sets. At
least one sample within each data set had non-
normally distributed data and unequal variances.
Both negative and positive skews were common
within data sets, and transformation of the data
was not possible. Thus, I used Kruskal-Wallis
nonparametric analysis of variance to test for
differences among means, which provided con-
servative and consistent testing for differences
among all groups. Kruskal-Wallis test statistics
were evaluated with ot = 0.05. Mann-Whitney
(7-tests were performed to differentiate between
significant groups whenever significant x~ values
were obtained. I used the non-parametric Ken-
dall’s tau statistic to analyze the relationship
between prevalence of occurrence and variation in
abundance. A Bonferroni correction was applied
to Mann-Whitney (7-tests and Kendall’s tau
con'elations, and all effects were reported at a =
0.0167. Statistical analyses were performed using
SPSS, Version 16.0 (2007). All maps were created
in ArcGIS, Version 9.2 (ESRI 2009).

RESULTS

Prevalence of Occurrence. — Significant differ-
ences were found among species in all data sets
tested for prevalence of occuirence (P < 0.001).
There were significant differences between spe-
cies for the multiple-observer data set, both at the
full extent of the species’ ranges and within the
distribution of Henslow’s SpaiTOws (Table 1).
Mann-Whitney (7-tests for the full extent of the
species’ ranges indicated the median prevalence
of occuiTence for Henslow’s SpaiTows (0.50) was
lower than for Grasshopper (0.78, P = 0.006), and
Savannah ( 1.00, P < 0.001) spaiTows; Grasshop-
per Span'ows, did not differ significantly in
prevalence from Savannah Sparrows (P =
0.022, Table 2). Results constrained to within
the distributional area of Henslow’s Sparrow were
similar, and significant for all comparisons
(Tables 1, 2; Fig. 2).

There were significant among-group differenc-
es in the single-observer data sets for both the full-
range extent and within the distributional area of
Henslow’s Sparrow (P < 0.001, Table 1). Median
Henslow’s Sparrow prevalence (0.50) for full-
range extents was significantly lower than for
Grasshopper (0.88, P < 0.001), and Savannah
(1.00, P < 0.001) sparrows (Table 2). Grasshop-
per Sparrows had lower prevalence than Savannah

638

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

TABLE 1. Comparisons (Kruskal-Wallis tests) of Henslow's. Grasshopper, and Savannah sparrows between multiple-
and single-observer groups, and between data collected across the full extent of each species’ breeding range versus
restricted to Henslow's Sparrow’s distributional area. Breeding Bird Survey data were collected from 2000 to 2007 on 50
randomly selected routes. (*) indicates P < 0.05.

Full extent of species’ ranges

Henslow’s Sparrow distributional area

•Multiple observers

Single obsen'er

Multiple observers Single observer

H if P

« df P

HUP H di P

Prevalence of

occurrence

26.9

2

<0.001*

39.9

2

<0.001*

45.4 2

<0.001*

45.1

2

<0.001*

Variation in

abundance

9.0

2

0.01 1*

12.6

2

0.002*

2.7 2

0.26

8.1

2

0.018*

Sparrows {P — 0.005, Table 2). The same results
occurred when these data were restricted to
the distributional area of Henslow’s Sparrow
(Tables 1, 2; Fig. 3).

Variation in Abundance. — Results were less
consistent between data sets than for prevalence
comparisons. There were differences between
species (P = 0.011, Table 1) for the multiple-
observer data set across the full extent of the
species’ ranges. Grasshopper Sparrows had higher
variability in abundance (median = 0.54) than
Henslow’s Sparrows (0.42, P = 0.004), but no sig-
nificant differences existed between Henslow’s and
Savannah (0.50, P — 0.070) sparrows, or between
Savannah and Grasshopper spaiTows (P = 0.18,
Table 2). No differences were apparent between
species (Table 1, Fig. 2) when the data were limited
to the distributional area of Henslow’s Span'ow.

The single-observer data sets revealed a similar
pattern, but with significant differences between

species in analysis of the full extent (P — 0.002)
and within the distribution of Henslow’s Span'ow
(P = 0.018, Table 1). There was no difference
between Henslow’s SpaiTows and Grasshopper
(0.44, P = 0.090) or Savannah {P = 0.21,
Table 2) sparrows. Grasshopper Sparrows,
however, did (median = 0.47) have higher
variability than Savannah Sparrows (0.39, P <
0.001). The same results were obtained for
analyses limited to single-observer routes within
the distribution of Henslow’s SpaiTow (Tables 1,
2; Fig. 3).

Correlation of Prevalence and Variation. —
Prevalence of occuiTence was significantly and
positively correlated to variation in abundance for
Henslow’s Span'ows in the multiple-observer (i =
0.37, P < 0.001) and single-observer (x = 0.32, P
= 0.003) data sets. This relationship did not hold
for either Grasshopper or Savannah spaiTOws in
any data set (all P > 0.05).

TABLE 2. Comparisons (Mann- Whitney U-tests) of Henslow’s (HESP), Grasshopper (GRSP), and Savannah (SASP)
sparrows between multiple- and single-observer groups, and between data collected across the full extent of each species’
breeding range and restricted to Henslow’s Spaixow’s distributional area. Breeding Bird Survey data were collected from
2000 to 2007 on 50 randomly selected routes. One comparison for variation in abundance is lacking because of
nonsignificant results (Kruskal-Wallis analysis). (*) indicates P < 0.0167.

r-'ull e.xtent of .species' ranges

Henslow’s .Sparrow distributional area

Multiple observers

Single ob.server

Multiple observers

Single observer

U z

p

u

p

u

: P

u

p

Prevalence of occurrence

HE.SP X GRSP

856.5 -2.7

0.006*

649.5

-4.2

<0.001*

582.0

-4.7 <0.001*

625.5

-4.4

<0.001*

HESP X SASP

505.0 -5.2

<().()() 1*

422.5

-5.9

<0.001*

364.5

-6.3 <0.001*

347.0

-6.4

<0.001*

GRSP X SASP

938.0 -2.3

0.022

879.0

-2.8

0.005*

998.5

- 1 .9 0.045

933.5

-2.5

0.014*

Variation in abundance

HESP X GRSP

8.30.5 -2.9

().()()4*

1,003.5

-1.7

0.09

1,152.5

-0.7

0.50

HESP X SASP

987.5 - 1 .8

().()7()

1,068.0

-1.3

0.21

1 .()09.()

-1.7

0.10

GRSP X SASP

1 .056.5 - 1 .3

0.18

697.0

-3.8

<0.001*

815.0

-3.0

0.003*

Dornuk • POPULATION DYNAMICS OF IIENSLOW’S SPARROWS

639

Coefficient of variation
of abundance

0.000-0.125
0.125-0.321
0.321 -0.502
0.502-0.707
0.707- 1.404

Prevalence of
occurrence

o 0.250-0.333
O 0.333-0.500
O 0.500-0.714

Q 0.714-0.875
Q 0.875- 1.000

N

FIG. 2. Henslow's (A), Gra.sshopper (B), and Savannah (C) sparrow distributions showing prevalence (circle size) and
abundance (shading of circles). Data are restricted to within the Henslow's Sparrow's breeding distribution and were
collected by multiple observers from 2000 to 2007 on 50 randomly selected Breeding Bird Survey routes.

640

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

Coefficient of variation
of abundance

0.000-0.125
0.125-0.322
0.322-0.514
0.514-0.774
0.774- 1.404

Prevalence of

occurrence

O 0.250-0.333
O 0.333-0.500
O 0.500-0.714
O 0.714-0.875
O 0.875- 1.000

N

FIG. 3. Hen.slow’s (A), Grasshopper (B), and Savannah (C) sparrow distributions showing prevalence (circle size) and
abundance (shading of circles). Data are restricted to within the Henslow’s Sparrow’s breeding distribution but were
collected by single observers only for given routes from 2()()() to 2007 on 50 randomly selected Breeding Bird Survey routes.

Doniak • POPULATION DYNAMICS OF HENSLOW’S SPARROWS

641

DISCUSSION

Resettlement. — Site fidelity describes the like-
lihood ot an individual’s return to a particular site
from year to year. Often, the decision for adult
birds to return to a particular site is based on the
individual’s experiences, or the experiences of
conspecitics, at the site in previous years (Hilden
1965). This measurement is ideal for finer-scale
analyses, but broad-scale studies require consid-
eration of resettlement of an area by groups of
individuals. More importantly, global patterns
exhibited by these groups across space and time
can provide insight into differences across their
geographic ranges.

Previous studies have evaluated nest site
fidelity in Henslow’s, Grasshopper, and Savannah
sparrows (Bedard and La Point 1984, Skipper
1998, Jones et al. 2007), although, to my
knowledge, this study is the first to examine
variation in resettlement patterns across their
breeding ranges. Henslow’s Sparrows, in all
analyses, had the lowest prevalence of occun'ence
among the three species. They were, as a group,
less likely to return to a BBS route from I year to
the next. Henslow’s Sparrows when detected, with
the exception of one test, had variability in
abundance not distinguishable from those of
Grasshopper and Savannah sparrows. Both Grass-
hopper and Savannah sparrows had greater
numbers of high-prevalence routes than Hen-
slow’s Sparrows. These across-species differences
in variation in occurrence might be best explained
by habitat choices and social behavior of each
.species.

Henslow’s Sparrows require large, open grass-
lands (Hyde 1939, Smith 1968, Thogmartin et al.
2006). Vegetation structure, including aspects of
vegetation height (Wiens 1969, Herkert 1994a) and
density (Wiens 1969, Zimmerman 1988), is an
important factor for nest site selection by Hen-
slow’s Sparrows (Graber 1968, Bajema and Lima
2001, Powell 2006). Fields must remain relatively
undisturbed for several consecutive years to
achieve vegetation structure typical of Henslow’s
Sparrow’s breeding habitat. Thus, this sparrow
avoids fields frequently disturbed by haying
(Graber 1968, Cully and Michaels 2000), burning,
or grazing on frequent rotations (Bollinger 1995).
Similarly, Grasshopper and Savannah spaiTows
prefer natural habitat to managed landscapes
(Owens and Myers 1973, Dale et al. 1997). These
species are less strict in their habitat preferences.

unlike Henslow’s Sparrows, and take advantage of
a greater variety of habitats, including grazed,
cultivated (Owens and Myers 1973), and hay fields
(Graber 1968, Dale et al. 1997).

These differences in habitat preferences among
the species may explain differences in prevalence
of occurrence. The narrowness of breeding habitat
preferences of Henslow’s Sparrow within the
dynamic grassland biome possibly motivates
groups of individuals to seek unused habitat when
previously settled locations are no longer suitable
(Reinking et al. 2000). For example, an area that
is optimal habitat in 1 year may experience a
disturbance, such as a late summer burn, resulting
in unsuitable habitat for subsequent breeding
seasons, thereby encouraging Henslow’s Spar-
rows to seek new nesting locations. Settling
suitable habitat when encountered upon arrival
in the breeding area, rather than homing to a
previously used site and then relocating, would
also be advantageous for species that use highly
variable resources (Johnson and Grier 1988).
Grasshopper and Savannah span'ows use a wider
variety of habitats (Smith 1968, Wiens 1969,
Owens and Meyers 1973, Dale et al. 1997), and
may resettle an area in successive breeding
seasons, even though the habitat has been altered
from previous years. The ability to be opportu-
nistic and flexible when choosing a nesting
location may be advantageous for species whose
nesting habitat is unpredictable and patchy (Wiens
1973, Cody 1985, Johnson and Grier 1988).

Henslow’s Span'ows, in the relatively few areas
where they occurred consistently, had higher
variation in abundance. They had less variation
in abundance in areas where they were not
consistently pre,sent creating a positive association
between prevalence and variability. This pattern
might be explained by the effect of clustering,
combined with opportunistic settling. The earliest
literature on Henslow’s Sparrows noted the
tendency of the species to nest in loose colonies,
especially when occupying large patches of
habitat (Hyde 1939, Graber 1968, Wiens 1969).
Individuals of nomadic species, such as Hen-
slow’s Sparrows, arriving to an area may use the
presence of conspecifics which have already
.settled there as a means to evaluate habitat
quality, especially when knowledge of nest
success from previous years within that nesting
location is lacking (Bollinger and Gavin 1989,
Ahlering et al. 2006). Thus, clustering when
paired with opportunistic location selection may

642

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

produce a pattern of iiTegularity of settlement with
regularity of abundance. Unlike Henslow’s Spar-
rows, no significant relationship exists between
variability, prevalence, and abundance for either
Grasshopper or Savannah sparrows.

Biases in BBS Data. — Use of BBS data is a
timely and cost-effective way of answering
relatively short-term, broad-scale questions (Sauer
et al. 2005, Winter et al. 2006). The BBS is the
only source for range-wide, standardized data in
North America; it is commonly used to estimate
abundances and year-to-year fluctuations across
species’ breeding ranges (Bibby et al. 2000,
Diefenbach et al. 2003, Herkert 2007). Although
widely used, the BBS is not without potential bias.
One criticism of the BBS is that of change of
observers across time. Link and Sauer (1998)
suggested that differences in ability among
observers may influence trends detected along
routes over time. I found results differed when
data were limited to single-observer routes.
Specifically, Savannah Sparrows had significantly
higher prevalence rates than Grasshopper Sparrows
in the single-observer data set, a difference not
detected in analyses of the multiple-observer data
set. The difference was significant in both the full
extent of the species’ ranges and in the area; thus,
reduced to the distributional area of Henslow’s
Sparrow, the effect was not a result of geographic
extent. Removing routes surveyed by multiple
observers also affected variation in abundance.
There were contrasting differences between
species across the full extents of the species’
ranges. Grasshopper Sparrows had the highest
variability, but the ranks of the other two species
shifted. The multiple-observer data set in ana-
ly.ses constrained to within the distribution of
Henslow’s Sparrow showed no species differences,
but the single-observer data set revealed signifi-
cant differences.

Thus, results of several analyses changed when
route data collected by multiple observers were
removed. This variation is likely a result of change
in observers and their relative abilities (Link and
Sauer 1998). Sauer et al. (1994) suggested abun-
dance patterns are best represented by analyses of
single-observer data only. However, even single
observers may, through time, improve in identifi-
cation and detection skills, lor example by learning
a song (Link and Sauer 1998) or the opposite (e.g.,
with declining hearing abilities); these elfects
cannot be measured, but may be reduced by
comparisons with multiple-observer data.

Detection. — Other concerns not addressed with-
in the analyses of this paper should be considered.
It is possible that Henslow’s SpaiTows are not
detected as easily as Grasshopper or Savannah
sparrows. The species may be too rare in some
areas, resulting in detection difficulties (Wells and
Rosenberg 1999, DeVault et al. 2002). In other
locations where individuals are more abundant,
lack of activity (Diefenbach et al. 2007, Confer et
al. 2008), or failure to detect individuals may
cause discrepancies, thereby affecting count data.

Henslow’s SpaiTow’s cryptic appearance and
secretive behavior (Hyde 1939), in addition to its
insect-like song (Leftwich and Ritchison 2000),
may lead to detection problems during surveys.
Detection may be further restricted when singing
declines after mate pairing (Leftwich and Ritch-
ison 2000) or as a result of Henslow’s Sparrows’
proclivity for nocturnal singing (Walk et al.
2000). Henslow’s SpaiTows may not nest near
roadsides as a response to woody vegetation
(Patten et al. 2006) or traffic volume (Forman et
al. 2002), yet BBS data are collected entirely from
roadside routes. It is possible that populations of
birds are not surveyed accurately because of
species’ behavior and surveying techniques.
Bajema et al. (2001) suggest applying a correction
factor to estimate Henslow’s Span'ow abundance
to counter these detection problems.

CONSERVATION IMPLICATIONS

Henslow’s SpaiTow was once thought to be
prevalent throughout the western extent of its
breeding range, although it now occupies less than
1% of this original area (Robbins et al. 2002). Its
need for regular, although infrequent, habitat
disturbance to maintain a particular serai stage
may facilitate this decline in present human-
dominated landscapes (Pruitt 1996). This habitat
in the pre-European landscape was maintained by
fire (natural and artificial), and grazing by large
herbivores, creating a mosaic of habitats on the
landscape (Knopf 1994, Umbanhowar 1996).
Burning, grazing, and mowing have all been
recommended as suitable management practices
(Pruitt 1996), but the timing, extent, and frequency
of these disturbances are critical factors that affect
suitability of the habitat for breeding Henslow’s
Sparrows (Herkert 1994a, Powell 2006). The
patterns documented in this study are strongly
suggestive of a species that has adapted to be able
to track optimal sites in a shifting habitat mosaic.

Henslow’s Sparrows are also considered area-

Dornak • POPULATION DYNAMICS OF IIENSLOW'S SPARROWS

643

sensitive and affected by patch size of breeding
habitat (Herkert 1994b, Walk and Warner 1999,
Oleary and Nyberg 2000, Thogmartin et al. 2006).
The CLiiTent approach of conserving large isolated
patches of high-quality habitat may be too narrowly
focused on this concept. Renfrew and Ribic (2008),
in a study of Bobolinks (Dolichony.x on’zivonis),
and Grasshopper and Savannah spairows, conclud-
ed that patch-size might only be particularly
restrictive when the entire landscape is heavily
fragmented. Thus, conserving smaller areas in
addition to large, contiguous habitat patches within
the landscape matrix would create more total
grassland habitat and better suit Henslow’s Spar-
rows, as it enhances the overall quality of the
landscape and fosters its nomadic movements and
colonial behavior (Hora and Koford 2006, Ribic et
al. 2009). Acquiring smaller, but functional, plots
may be easier and more economical, thus producing
more immediate conservation impacts. Ultimately,
losses of suitable grassland nesting habitat on
regional scales may result in extinction of obligate
grassland species (Vickery et al. 1994), and
particularly Henslow’s Sparrows, which appear to
move much more broadly across regional land-
scapes than the other two species analyzed.

ACKNOWLEDGMENTS

I could not have accomplished this project without the
aid of A. T. Peterson. 1 express my gratitude for his support
with project design and manuscript refinements. I also
thank S. L. Egbert for professional and academic guidance,
and Y. J. Nakazawa, L. L. Rausch, and Xingong Li for
valuable technical support. Data were made available by the
Breeding Bird Survey. Financial support was provided by
the Department of Geography and by the Experimental
Program to Stimulate Competitive Research (EPSCoR).

LITERATURE CITED

Ahlering, M. A., D, H. Johnson, and J. Faaborg. 2006.
Conspecific attraction in a grassland bird, the Baird’s
Sparrow. Journal of Field Ornithology 77:36.6-371,
Bajema, R. a. and S. L. Lima. 2001. Landscape-level
analysis of Henslow’s Sparrow (Ammodramus heiislo-
wii) abundance in reclaimed coal mine grasslands.
American Midland Naturalist 145:288-298.

Bajema. R. A., T. L. DeVault, P. E. Scott, and S, L.
Lima. 2001. Reclaimed coal mine grasslands and their
significance for Henslow’s Sparrows in the American
Midwest. Auk 118:422-431.

Bedard, J. and G. LaPointe. 1984. Banding returns,
arrival times, and site fidelity in the Savannah
Sparrow. Wilson Bulletin 96:196-205.

Bibby, C. j., N. D. Burgess, D. A. Hill, and S. H.
Hustoe. 2000. Bird census techniques. Academic
Press, London, United Kingdom.

Bollinger, E. K. 1995. Successional changes and habitat
selection in hayfield bird communities. Auk 1 12:720-730.

Bollinger, E. K. and T, A. Gavin. 1989. Succe.ssional
changes and habitat selection in hayfield bird com-
munities. Auk 106:584-594.

Cody, M. L. 1985. Habitat selection in open-country birds.
Pages 191-226 in Habitat selection in birds (M. L.
Cody, Editor). Academic Press, Orlando, Florida.
USA.

Confer, J. L., R. E. Serrell, M. Hager, and E. Lahr.
2008. Field tests of the Rosenberg-Blancher method
for converting point counts to abundance estimates.
Auk 125:932-938.

Cully Jr., J. F. and H. L. Michaels. 2000. Henslow’s
Sparrow habitat associations on Kansas tallgrass
prairie. Wilson Bulletin 112:115-123.

Cunningham, M. A. and D. H. Johnson. 2006. Proximate
and landscape factors influence grassland bird distri-
butions. Ecological Applications 16:1062-1075.

Dale, B. C., P. A. Martin, and P. S. Taylor. 1997.
Effects of hay management on grassland songbirds in
Saskatchewan. Wildlife Society Bulletin 25:616-626.

DeVault, T. L., P. E. Scott, R. A, Bajema, and S. L.
Lima. 2002. Breeding bird communities of reclaimed
coal-mine grasslands in the American Midwest.
Journal of Field Ornithology 73:268-275.

Diefenbach, D. R., D. Brauning, and j. a. Mattice.
2003. Variability in grassland bird counts related to
observer differences and species detection rates. Auk
120:1168-1179.

Diefenbach, D. R., M. R. Marshall, J. A. Mattice, and
D. Brauning. 2007. Incorporating availability for
detection in estimates of bird abundance. Auk 124:96-
106.

Environmental Research Systems (ESRI). 2009. Arc-
Map, Version 9.2. Environmental Research Systems
Inc., Redlands, California, USA.

Forman, R. T. T., B. Reineking, and A. M. Hersperger.
2002. Road traffic and nearby grassland gird patterns
in a suburbanizing landscape. Environmental Manaae-
ment 29:782-800.

Graber, j. W. 1968. Passerherhidus heuslowii hendslowii
(Audubon): Western Henslow’s Sparrow. Pages 779-
788 in Life histories of North American cardinals,
grosbeaks, buntings, towhees, finches, sparrows, and
allies (A. C. Bent, Editor). Smithsonian Institute, U.S.
National Museum Bulletin 237.

Herkert, J. R. 1994a. Status and habitat selection of the
Henslow’s Sparrow in Illinois. Wilson Bulletin
I06:35-J5.

Herkert, J. R. 1994b. The effects of habitat fragmentation
on midwestern grassland bird communities. Ecological
Applications 4:461^71.

Herkert, J. R. 2007. Conservation Reserve Program
benefits on Henslow’s Sparrows within the United
States. Journal of Wildlife Management 7 1 -2750-
2751.

Hidden, O. 1965. Habitat selection in birds. Annales
Zoologici Fennici 52:53-75.

Horn, D. J. and R. R. Koford. 2006. Could the area-
sensitivity of some grassland birds be affected by

644

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, December 2010

landscape composition? Proceedings of the North
American Prairie Conference 19:109-116.

Hyde, A. S. 1939. The life history of Henslow’s Sparrow,
Passerherbiilus henslowii (Audubon). University of
Michigan, Ann Arbor, Museum of Zoology Miscella-
neous Publications 41:1-79.

iNGOLD, D. J., J. L. Dooley, and N. Lavender. 2009.
Return rates of breeding Henslow’s Sparrows on
mowed versus unmowed areas on a reclaimed surface
mine. Wilson Journal of Ornithology 121:194-197.

Johnson, D. H. and J. W. Grier. 1988. Determinants of
breeding distributions of ducks. Wildlife Monographs
100.

Jones, S. L., J. S. Dieni, M. T. Green, and P. J. Gouse.
2007. Annual return rates of breeding grassland
songbirds. Wilson Journal of Ornithology 119:89-94.

Knopf, F. L. 1988. Conservation of steppe birds in North
America. Ecology and Conservation of Grassland
Birds 7:27-H.

Knopf, F. L. 1994. Avian assemblages on altered
grasslands. Studies in Avian Biology 15:247-257.

Leftwich, C. and G. Ritchison. 2000. Singing behavior of
male Henslow’s Sparrows (Ammodmmus henslowii).
Bird Behavior 18:1-7.

Link, W. A. and J. R. Sauer. 1998. Estimating population
change from count data: application to the North
American Breeding Bird Survey. Ecological Applica-
tions 8:258-268.

Monroe. M. S. and G. Ritchison. 2005. Breeding biology
of Henslow’s Sparrows on reclaimed coal mine
grasslands in Kentucky. Journal of Field Ornithology
76:143-149.

Noss, R. F., E. T. Large, and J. M. Scott. 1995.
Endangered ecosystems of the United States: a
preliminary assessment of loss and degradation. USDl,
National Biological Service 06II-R-0I (MR). Wash-
ington D.C., USA.

O’Leary, C. H. and D. W. Nyberg. 2000. Treelines
between fields reduce the density of grassland birds.
Natural Areas Journal 20:243-249.

OwEN.S, R. A. AND M. T. Myers. 1973. Effects of
agriculture upon populations of native passerine birds
of an Alberta fescue grassland. Canadian Journal of
Zoology 51:697-713.

Paiten, M. a., E. Shochat, D. L. Reinking, D. H. Wolfe,
AND S. K. Sherrod. 2006. Habitat edge, land
management, and rates of brood parasitism in tallgrass
prairie. Ecological Applications 16:687-695.

Peterjohn, B. G. and j. R. Sauer. 1999. Population status
of North American grassland birds from the North
American Breeding Bird Survey. 1966-1996. Studies
in Avian Biology 19:27^4.

Powell, A. F. L. A. 2006. Effects of prescribed burns and
bison (Bo.s bison) grazing on breeding bird abundances
in tallgrass prairie. Auk 123:183-197.

Pruht. L. 1996. Henslow's Sparrow: status assessment.
USDl. Fish and Wildlife Service. Bloomington,
Indiana. USA.

Reinking. D. L.. D. A. Wiedenfeld, D. H. Wolfe, and R.
W. Rohrbaugh Jr. 2()()(). Distribution, habitat use.

and nesting success of Henslow’s Sparrow in Okla-
homa. Prairie Naturalist 32:219-232.

Renfrew, R. B. and C. A. Ribic. 2008. Multi-scale models
of grassland passerine abundance in a fragmented
system in Wisconsin. Landscape Ecology 23:181-193.

RtBic, C. A., R. R. Koford, j. R. Herkert, D. H. Johnson,
N. D. Niemuth, D. E. Naugle, K. K. Barker, D. W.
Sample, and R. B. Renfrew. 2009. Area sensitivity in
North American grassland birds: patterns and process-
es. Auk 126:233-244.

Rich, T. D., C. J. Beardmore, H. Berlanga, P. J.
Blancher, M. S. W. Bradstreet, G. S. Butcher, D.
W. Demarest, E. H. Dunn, W. C. Hunter, E. E.
Inigo-Elias, J. a. Kennedy, A. M. Martell, A. O.
Panjabi, D. N. Pashley, K. V. Rosenberg, C. M.
Rustay, J. S. Wendt, and T. C. Will. 2004. Partners
in Flight North American landbird conservation plan.
Laboratory of Ornithology, Cornell University, Ithaca,
New York, USA.

Robbins, C. S., D. Bystrak, P. Henslow, and P. H.
Geissler. 1986. The Breeding Bird Survey: its first
fifteen years, 1965-1979. USDl, Fish and Wildlife
Service, Washington, D.C., USA.

Robbins, M. B., A. T. Peterson, and M. A. Ortega-
Huerta. 2002. Major negative impacts of early
intensive cattle stocking on tallgrass prairies: the case
of the Greater Prairie-Chicken {Tyinpaniichiis ciipido).
North American Birds 56:239-244.

Samson, F. and F. L. Knopf. 1994. Prairie conservation in
North America. BioScience 44:418^21.

Sauer, J. R., B. G. Peterjohn, and W. A. Link. 1994.
Observer differences in the North American Breeding
Bird Survey. Auk 1 1 1:50-62.

Sauer, J. R., J. E. Hines, and J. Fallon. 2007. The North
American Breeding Bird Survey, results and analysis
1966-2006. Version 10.13.2007. USGS, Patuxent
Wildlife Research Center, Laurel, Maryland, USA.

Sauer, J. R., J. Casey, H. Laskowski, J. D. Taylor, and
F. Jane. 2005. Use of survey data to define regional
and local priorities for management on national
wildlife refuges. USDA, Forest Service. General
Technical Report PSW-GTR-191. Pacific Southwest
Re.search Station, Albany, California, USA,

Skipper, C. S. 1998. Henslow’s Sparrows return to previous
nest site in western Maryland. North American Bird
Bander 23:36^1 .

Smith, W. P. 1968. Passerherbiilus henslowii susiirrans
(Brewster): Eastern Henslow’s Sparrow. Pages 776-
778 in Life histories of North American cardinals,
grosbeaks, buntings, towhees, finches, sparrows and
allies (A. C. Bent. Editor). Smith.sonian Institute, U.S.
National Museum Bulletin 237.

SPSS Institute Inc. 2007. SPSS for Windows, Version
16.0. SPSS Institute Inc., Chicago, Illinois. USA.

Thogmartin. W. E., M. G. Knut.son, and j. R. Sauer.
2006. Predicting regional abundance of rare grassland
birds with a hierarchical spatial count model. Condor
108:25^6.

Umbanhowar Jr., C. E. 1996. Recent fire history of the
northern Great Plains. American Midland Naturalist
135:1 15-121.

Donuik • POPULATION DYNAMICS OF HENSLOW’S SPARROWS

645

U.S. Department of Interior (USDD. 2008. Birds of
conservation concern 2008. USDI. Fish and Wildlife
Service, Division of Migratory Bird Management,
Arlington, Virginia, USA.

Vickery, P. D., M. L. Hunter, and S. M. Melvin. 1994.
Effects of habitat area on the distribution of grass-
land birds in Maine. Conservation Biology 8:1087-
1097.

W.ALK, J. W. AND R. E. Warner. 1999. Effects of habitat
area on the occurrence of grassland birds in Illinois.
American Midland Naturalist 141:339-344.

Walk, J. W., E. L. Kershner, and R. E. Warner. 2000.
Nocturnal singing in grassland birds. Wilson Bulletin
1 12:289-292.

Wells, J. V. and K. V. Rosenberg. 1999. Grassland bird
conservation in northeastern North America. Studies in
Avian Biology 19:72-80.

Wheelwright, N. T. and J. D. Rising. 1993. Savannah
Sparrow (Passerculiis sundwichensis). The birds of
North America. Number 45.

Wiens, J. A. 1969. An approach to the study of ecological
relationships among grassland birds. Ornithological
Monographs 8:1-93.

Wiens, J. A. 1973. Intertenatorial habitat variation in
Grasshopper and Savannah sparrows. Ecology 54:877-
884.

Winter, M., D. H. Johnson, J. A. Shaffer, T. M.
Donovan, and W. D. Svedarsky. 2006. Patch size
and landscape effects on density and nesting success of
grassland birds. Journal of Wildlife Management
70:158-172.

Zimmerman, J. L. 1988. Breeding season habitat selection
by the Henslow’s Sparrow (Ammod ramus henslowii)
in Kansas. Wilson Bulletin 100:17-24.

The Wilson Journal of Ornithology 122(4):646-654, 2010

INFLUENCE OF WOODY VEGETATION ON GRASSLAND BIRDS
WITHIN RECLAIMED SURFACE MINES

BRET M. GRAVES,'-' AMANDA D. RODEWALD,' AND SCOTT D. HULL-

ABSTRACT. — We examined the influence of woody vegetation on reclaimed surface mines on relative abundance of
Grasshopper Sparrows {Aminocl ramus savannarum), Henslow’s Spanows (A. henslowii). Eastern Meadowlarks (Stiirnella
magna). Savannah Sparrows (Passerculus sandwichensis). Bobolinks (Dolichonyx oiyzivoriis), and Dickcissels (Spiza
americana) as well as nest-site selection and nesting success of Grasshopper and Henslow’s sparrows and Eastern
Meadowlarks. Grasshopper and Henslow's sparrows were the most abundant grassland species on reclaimed mines.
Numbers of Grasshopper. Henslow’s, and Savannah spanows. and Bobolinks were negatively a.ssociated with percent cover
of woody vegetation within 100 m of survey locations. Only Grasshopper Sparrows responded to woody vegetation at nest-
patch scales, as random locations had >2.5 times as much woody cover as nest locations. Daily nest survival (DNS) was
negatively associated with amount of woody vegetation within 100 m of Grasshopper (DNS 0.76 ± 0.001 SE) and
Henslow’s sparrow nests (DNS 0.94 ± 0.020 SE). but only marginally negatively related to daily nest survival of Eastern
Meadowlark nests (DNS 0.87 ± 0.006 SE). Avoidance of woody vegetation by grassland birds and the comparatively lower
daily nest survival of Grasshopper and Henslow’s sparrow nests near woody vegetation suggests managers of reclaimed
surface mines who manage to conserve grassland birds should direct efforts towards reducing woody encroachment.
Received 22 June 2009. Accepted 14 July 2010.

Grassland birds have declined more rapidly
than any other group of birds in the Midwest over
the past 30 years (Peterjohn and Sauer 1999, Cully
and Michaels 2000, Vickery and Herkert 2001,
Sauer et al. 2005). Among the most precipitously
declining are area-sensitive species including
Grasshopper Sparrow (Ammodramiis savan-
nanini), Henslow’s Sparrow (A. henslowii), and
Eastern Meadowlark (Stiirnella magna) (Herkert
1994). These declines have been largely attributed
to changes in agricultural practices (Johnson and
Igl 2001), habitat loss (Askins 1993), and habitat
degradation and fragmentation (Herkert 1994,
Cully and Michaels 2000, Vickery and Herkert
2001). Large gains in native and managed
grassland habitats in the Midwest are unlikely to
occur and successful conservation of grassland
birds will require consideration of new opportu-
nities to provide quality grassland habitat for area-
sensitive grassland-breeding species. Reclaimed
surface mines represent one opportunity to
manage for grassland bird species (Whitmore
and Hail 1978, Whitmore 1980, Wray et al. 1982,
Bajema et al. 2001, DeVault et al. 2002, Monroe
and Ritchison 2005. Galligan et al. 2006).
Reclaimed surface mines represent a conservation

' .School of Environment and Natural Resource.s, The
Ghio State University. 2021 Coffey Road. 210 Kottman
Hall, Columbus. OH 4.^210, USA.

Hfureau of Wildlife Management, Wisconsin Depart-
ment of Natural Resources, 101 South Webster Street, P. O.
Box 7921. Madison. W1 53707, USA.

'Corresponding author; e-mail: graves. 1 2@o.su.edu

paradox in that grassland birds occupy them even
though they are highly disturbed, dominated by
exotic grasses, and vulnerable to invasion by
exotic woody vegetation such as autumn olive
(Elaeagmis iimbellata).

Woody encroachment is a key management
issue with the potential to impact the conservation
value of reclaimed surface mines for grassland-
breeding birds. Reforestation of mined areas to the
original deciduous forest landscape is often not
po.ssible due to poor soil conditions (Wray et al.
1982, Brothers 1990, DeVault et al. 2002, Scott et
al. 2002), but invasion by woody species such as
autumn olive and black locust (Robinia pseiidoa-
cacia) commonly occurs if left unmanaged (Rum-
mel and Brenner 2003). The decision to maintain
open grasslands by removing woody vegetation
remains controversial despite known negative
effects of woody vegetation on grassland birds
(Bajema et al. 2001, Ribic and Sample 2001,
Bakker et al. 2002, Fletcher and Koford 2002,
Grant et al. 2004). Removal of woody vegetation
within reclaimed surface mines may alleviate nest
predation by redistributing movements of common
mammalian nest predators including common
raccoon (Procyon lotor), striped skunk (Mephitis
mephitis), and Virginia opossum (Didelphis vir-
giniana) (Winter et al. 2000). However, some
grassland species such as Vesper (Pooecetes
gramineiis) and Clay-colored (Spizella pallida)
sparrows, and Common Yellowthroats (Geothlypis
trichas) may be tolerant of woodland encroach-
ment at the landscape .scale (Grant et al. 2004).

646

Graves el a/. • GRASSLAND BIRD CONSERVATION ON RECLAIMED MINES

647

Effective encouragement and conservation of
grassland birds on reclaimed surface mines
requires an understanding of the influences that
unmanaged woody plant encroachment can have
on distribution, nest-site selection, and nesting
success ot obligate grassland birds. We focused
on six obligate grassland bird species receiving
national and regional conservation attention
(Walk and Warner 2000, Bajema et al. 2001,
Vickery and Herkert 2001, DeVault et al. 2002);
Henslow's SpaiTow, Grasshopper Sparrow, East-
ern Meadowlark, Savannah Span'ow {Passerciiliis
sandwichensis). Bobolink (Dolichonyx oryzi-
voriis), and Dickcissel (Spiza americana). We
examined: ( 1 ) how woody vegetation features
were related to habitat use and distribution of
Grasshopper, Savannah, and Henslow’s spaiTows,
Eastern Meadowlarks, Bobolinks, and Dickcissels
breeding within reclaimed surface mines; (2) the
extent woody habitat features were related to nest-
site selection of Grasshopper and Henslow’s
sparrows, and Eastern Meadowlarks; and (3) if
these woody habitat features affected daily nest
survival of Grasshopper and Henslow’s spaiTows,
and Eastern Meadowlarks.

METHODS

Study Area. — This study was conducted on
reclaimed surface mines within wildlife manage-
ment areas in eastern Ohio, USA, managed by the
Ohio Department of Natural Resources (ODNR),
Division of Wildlife. Collectively, these wildlife
management areas represent the range of re-
claimed surface mine habitats available to grass-
land birds in eastern Ohio. Tri-Valley Wildlife
Management Area (WMA) in northern Muskin-
gum County is —6,100 ha comprised of —50%
open land (grasslands, wetlands, and food plots),
40% woodland, and 10% brushland. Approxi-
mately 2,600 ha of reclaimed grasslands are
within Tri-Valley WMA. Woodbury WMA in
southern Coshocton County is —7,600 ha of
—35% openland (grasslands, wetlands, and food
plots), 8% brushland, and 57% woodland. Ap-
proximately 1,200 ha of reclaimed grasslands
are within Woodbury WMA (ODNR 2007).

Dominant vegetation types in the study area
were non-native cool .season grasses including
fescue (Festiica spp.), redtop (Agrostis gigantea),
timothy (Phleum pratense), and orchard grass
(Dacty'lis glomerata) as well as native warm
season grasses including switchgrass (Pauiciim
virgatum), Indian grass (Sorghastnim nutans), big

bluestem (Andropogon gerardii), and little blue-
stem {Schizachyriuni scopariurn). Native and non-
native forbs {Melilotus spp., Trifoliuin spp.,
Solidago spp., Lotus corniculatus), and woody
vegetation including autumn olive, black locust,
blackberry (Ruhus allegheniensis), multiflora rose
(Rosa inultiflora), lespedeza (Lespedeza spp.),
(lowering dogwood (Cornus florida), and pines
(Pinus spp.) were also well represented.

Surveys. — Grassland birds were surveyed from
May through July 2005 and 2006 at 101 different
point-count locations in Tri-Valley (24 in 2005
and 29 in 2006) and Woodbury (24 in 2005 and 24
in 2006). We used a systematic random sampling
design for point-count locations. Starting loca-
tions (6 per study site) of point-count survey
transects were randomly selected using a Geo-
graphic Information System (GIS); subsequent
point-count locations were systematically placed
250 m apart until it was not possible to place the
next point 250 m from the previous location (i.e.,
bander such as a woodland area or large wetland
complex). All point-count locations (n = 101)
were recorded with a handheld Global Positioning
System (GPS) unit. We performed grassland bird
surveys at each of the 101 point-count locations
three times each from 15 May to 15 July in both
2005 and 2006 between 0630 and 1100 hrs EST
on days without fog, precipitation or winds
>25 kph. Two trained observers first allowed a
2-min period for birds to adjust to surveyor
presence at each point-count location followed by
a 6-min survey period when all singing male
grassland birds were recorded within a 100 m
radius. The time of detection of each bird was
recorded during each point-count survey along
with species code and straight-line distance (m) to
each bird detected with the use of a laser
rangefinder (Ransom and Pinchak 2003).

Nest Survival. — We randomly selected eight of
16 5-ha study plots similar in topography within
the grassland units containing varying amounts of
woody encroachment for nest searches (4 within
Woodbury and 4 within Tri-Valley). Field teams
used a .spot-mapping protocol (Bibby et al. 2000)
to identify breeding territories to facilitate locat-
ing nests. Grids were visited eight times during
morning hours as part of the spot-mapping efforts
from 1 May to 16 June 2005 and 2006, and an
observer walked the entire 5-ha plot on parallel
transects (.separated by 75 m). These spot-maps
helped field teams extensively search plots for
nests of Grasshopper and Henslow’s sparrows.

648

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, December 2010

and Eastern Meadowlarks from 1 May to 1
August 2005 and 2006. We initially conducted
nest searches using rope-dragging techniques
where a 30-m rope was dragged behind two
individuals while a third walked several meters
behind the rope looking for birds that flushed
from nests. We also relied on behavioral cues to
find nests and carefully followed adults that were
carrying food and nesting material until a nest was
located. Observers systematically searched known
breeding territories obtained from spot maps to
locate nests. Some nests were located by chance
during bird surveys and while taking vegetation
measurements. We marked nest locations with a
handheld GPS unit as well as with one colored
flag placed 5 to 10 m from the nest depending on
how concealed the nest was to avoid flushing the
attending adult. We checked nests every 3^ days
during the incubation period and every 2-3 days
when young neared the fledging stage. Nests were
visited from different angles during each nest
check to avoid creating paths that may attract
predators. The number of eggs and/or nestlings,
date the nest was checked, and the species name
were recorded. A nest was considered active if at
least one egg was present. A typical nest period
for the three focal species was 26 days with 8-
10 days post-hatching to fledge young (Ingold
2002). We considered a nest successful if at least
one young fledged or if it was active for >8 days
post-hatching (Vickery 1996) and there were no
obvious signs of predation. A nest was considered
lost to predation if the eggs were removed, cold or
broken or if the nest was disturbed and the
nestlings were no longer in the nest during the
first week post hatching (Ingold 2002).

GIS Analysis.— 'We estimated the amount of
woody vegetation within 100 m of nests and
survey points using National Agricultural Imagery
Program (NAIP) aerial photographs from 2005
and 2006 of Coshocton and Muskingum counties,
which were projected using ArcGIS 9.1 (ESRl
2005). The percentage of woody vegetation and
number of woody patches were estimated for
circular plots with a lOO-m radius centered on
each nest (n = 81) and on all GPS .survey
locations in = 101). We visited plots and directly
compared vegetation shown in images to actual
vegetation on plots observed during field visits to
verify the size of woody plants detectable in aerial
photographs. Ground-truthing revealed woody
vegetation patches <4 m^ in area were not
reliably detected from aerial photographs; conse-

quently patches of woody vegetation >4 m" in
size were defined as woody vegetation patches.
Each woody patch of vegetation >4 m“ within
100 m of each nest and survey point was digitized
and the area of each polygon was used to calculate
the total area of woody vegetation within a 100-m
radius suiTOunding each nest and survey location.
The total percent of woody vegetation was
calculated by dividing the total area (m‘) of
woody vegetation by the total area of the circle
(31,415 m") multiplied by 100. Aerial photo-
graphs, projected in ArcGIS 9.1, were used to
measure the distance (m) from each nest and each
survey location to the nearest woodland edge.

Statistical Analyses. — We compared an a priori
set of woody vegetation variables (percent cover
of woody vegetation, number of woody patches,
and distance to woodland edge) on bird survey
locations to identify which woody vegetation
features were used as cues by grassland birds in
selecting suitable habitat. We u.sed 101 survey
locations as replicates to identify woody habitat
features used by Grasshopper, Henslow’s, and
Savannah spairows. Eastern Meadowlarks, Bobo-
links, and Dickcissels as cues in selecting suitable
habitat for breeding locations. The relationship
between relative abundance and woody vegetation
was tested using regression analysis in Proc
GENMOD (SAS Version 9.1; SAS Institute
2002) using generalized linear models. Bird
point-count data were transformed when appro-
priate to minimize non-normality and variance
heterogeneity. A log transformation was applied
to counts of Grasshopper and Henslow’s spar-
rows, and Eastern Meadowlarks and a normal
distribution was used. A Poisson distribution was
u.sed for counts of Savannah Sparrows, Bobolinks,
and Dickcissels.

Percent woody vegetation, number of woody
vegetation patches >4 m^, and the distance from
plot center to the nearest woodland edge were
moderately correlated (r < 10.51) and were
incorporated into a principal components analysis
(Proc FACTOR in SAS Version 9.1; SAS Institute
2002) to characterize woody vegetation surround-
ing survey locations using a single parameter. The
first factor of the three principle components
explained >56% of the variation in woody
vegetation among survey points (Eigenvalue =
1 .69). This first factor loaded positively for
percent woody cover (0.88) and number of woody
patches (0.70), and loaded negatively for distance
to woodland edge (—0.65).

(iraves cl at. • GRASSLANr:) HU^D CONSERVATION ON RECLAIMED MINES

649

We examined an a priori set of woody
vegetation variables (percent cover of woody
vegetation, number of woody patches, distance to
nearest woody patch, and distance to woodland
edge) at individual nests and random locations to
learn which woody vegetation variables influenced
grassland birds during nest-site selection. We used
81 nests as replicates to identify woody habitat
features used by individual birds (Grasshopper and
Henslow's spaiTows, and Eastern Meadowlarks) to
select nest-site locations for each species indepen-
dently during individual nest attempts. We trans-
formed the variables when appropriate to minimize
non-normality and variance heterogeneity. Percent
woody vegetation, number of woody vegetation
patches, and distance from plot center to the nearest
woody patch and woodland edge were used in a
discriminant function analysis (Quinn and Keough
2002) (DFA; Proc CANDISC in SAS Version 9.1;
SAS Institute 2002) to examine which variables
best discriminated between nest and random plots.
Variables that best discriminated between nest and
random plots were interpreted as potential habitat
cues used by grassland birds in selecting locations
for nest placement (Quinn and Keough 2002).

We examined how woody habitat features in
nest plots and at nests influenced nest success by
incorporating the first principal component de-
scribing woody vegetation suiTounding nest sites
(PCwood) into the logistic-exposure method
(Shaffer 2004) to model daily nest survival rates
of Henslow’s and Grasshopper sparrows, and
Eastern Meadowlarks. This approach models the
success or failure of nests during each interval
between nest checks and evaluates the probability
of nest success over a range of values for influential
categorical and continuous explanatory variables.
Nest losses to all sources were classified as failures.
Ne.sts abandoned prior to laying or found after
depredation had already occurred were excluded.
We fit the model (PCwood interacting with
species) with Proc GENMOD (SAS Version 9.1;
SAS Institute 2002) using a binomial response
distribution (interval nest fate = 1 if success, and 0
if fail) and provided the user-defined logit link
function (g(6) = loge (0'Vl 1 — 0'^'])) where t = the
length of the interval (Shaffer 2004).

RESULTS

Grasshopper and Henslow’s sparrows were the
most abundant species on Tri-Valley and Wood-
bury WMAs during 2005 and 2006. We detected
1,491 birds on point counts during the study.

including 732 Henslow’s Sparrows, 465 Grass-
hopper Sparrows, 209 Eastern Meadowlarks, 39
Bobolinks, 31 Dickcissels, and 15 Savannah
Sparrows. Mean relative abundance ranged from
2.33 birds per survey point for Henslow’s
Sparrows at Tri-Valley to 0 for Savannah
Sparrows and Bobolinks at Woodbury. The
amount of woody vegetation was negatively
associated with abundance of Grasshopper Spar-
rows (Beta estimate = -0.14 ± 0.041 SE; x~ =
11.34, P < 0.001), Henslow’s Sparrows (Beta
estimate = —0.15 ± 0.051 SE; = 7.73, P =
0.005), Savannah Sparrows (Beta estimate =
-2.47 ± 0.975 SE; x' = 7.38, P = 0.007), and
Bobolinks (Beta estimate = —2.35 ± 0.745 SE;
X“ = 1 1.48, P < 0.001 ), but not significantly for
Eastern Meadowlarks (Beta estimate = —0.11 ±
0.241 SE; x^ = 0.21, P = 0.65) or Dickcissels
(Beta estimate = -0.57 ± 0.704 SE; x“ = 0.84, P
= 0.36) (Fig. 1).

Forty-five Grasshopper Span'ow, 18 Henslow’s
Sparrow, and 1 8 Eastern Meadowlark nests were
monitored during the nesting seasons. Habitat
variables discriminated between random locations
and nest sites of Grasshopper Span'ows (Canon-
ical con-elation = 0.25, Likelihood ratio = 0.936;
Wilks’ Lambda F3 ,42 = 3.24, P = 0.024)
(Table 1). Random locations had >2.5 times as
much woody vegetation as ne.st locations of
Grasshopper Sparrows (F| 144 = 5.41, P =
0.021). Nest placement by Grasshopper Sparrows
was not associated with number of woody patches
(F| 144 = 1.75, P = 0.19) or distance to woodland
edge (F| 44 = 0.96, P = 0.33). Habitat variables
did not discriminate between random plot loca-
tions and nests of Henslow’s SpaiTows (Wilks'
Lambda F3 j 14 = 1.12, P = 0.35) or Eastern
Meadowlarks (Wilks’ Lambda F3 , ,4 = 1.35, P =
0.26) (Table 1).

Twenty of 45 (44%) Grasshopper Spairow
nests, sixteen of 18 (89%) Henslow's Sparrow
nests, and nine of 18 (50%) Eastern Meadowlark
nests were successful (i.e., fledged at least 1
young or young remained in the nest >8 days
post-hatching). The amount of woody vegetation
surrounding nest locations was significantly
associated with daily nest survival of Grasshopper
and Henslow’s spaiTows, and Eastern Meadow-
larks (Beta estimate = -1.19 ± 0.457 SE; x" =
6.57, P = 0.010), despite all three species
differing in daily nest survival rates (Beta
estimates: Eastern Meadowlark = -1.55 ± 1.17
SE, Grasshopper Spari'ow = —2.39 ± 1.13 SE,

650

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 4, December 2010

FIG. 1. Relative abundance of grassland bird species (A-F) within 100-m radius of 101 point-count locations on
reclaimed surface mines in relation to amount of woody vegetation (PC wood) in eastern Ohio, 2005-2006. Amount of
woody vegetation increases from —3 to 4 along the x axis.

Hen.slow’.s Sparrow = 0.00; “ 11.71, P —

0.003). There was no evidence of significant
interactions and all three species responded
similarly to woody vegetation surrounding their
nests = 4.32, P = 0.12).

Mean ± SE daily nest survival rate was 0.76 ±
0.001 for Grasshopper Sparrows, 0.94 ± 0.020 for
Henslow’s Sparrows, and 0.87 ± 0.006 for
Eastern Meadowlarks. Daily nest survival of
Grasshopper and Henslow’s sparrows was nega-

Graves el al. • CJRASSLAND liIRD CONSERVATION ON RECLAIMED MINES

651

TABLE I. Mean percent woody vegetation, number of woody patches, and distance to woodland edge (± SH) for nest-
site locations of Grasshopper and Henslow’s sparrows, and Eastern Meadowlarks and random vegetation plots.

Grasshopper Sparrow Herislow's Sparrow Eastern Meadowlark Random

Variable v ± SE F P .v ± SE F P J ± SE F P ^ ± SE

% Woody

vegetation 1 .5

H-

0.29

5.4

0.021

2.3

-F

1.38

1.0

0.325

1.5

-F

0.64

2.2

0. 1 37

4.0

-F

0.70

# Woody patches 3. 1
Distance to

H-

0.49

1.8

0.188

2.7

1.04

1.4

0.233

1.9

-F

0.42

3.6

0.061

4.2

-F

0.51

woodland 165.3

-F

1 1.70

1.0

0.331

161.1

16.09

0.7

0.396

201.8

-+■

19.39

0.7

0.391

181. 1

H-

9.49

lively associated with amount of woody vegeta-
tion (PCwood) surrounding nest locations
(Fig. 2). Eastern Meadowlark daily nest survival
was marginally associated with amount of woody
vegetation in proximity to nest locations (Fig. 2).

DISCUSSION

Our findings suggest woody vegetation nega-
tively influenced abundance, nest placement, and
nesting success of some grassland bird species on
reclaimed surface mines. Our results are consis-
tent with previous studies of the effects of woody
vegetation on grassland bird occurrence in the
midwestern U.S.; however, our study is unique as
we specifically examined reclaimed surface mines

(but see Bajema et al. 2001 and Galligan et al.
2006). The abundance of obligate grassland birds
has been shown to be negatively related to woody
vegetation in Minnesota and North Dakota (Grant
et al. 2004, Winter et al. 2006, respectively),
southwestern Missouri (Winter and Faaborg
1999), Illinois (O’Leary and Nyberg 2000),
eastern South Dakota (Bakker et al. 2002), and
West Virginia (Wray et al. 1982). Our failure to
detect an association between woody vegetation
and abundance of Eastern Meadowlarks and
Dickcissels was possibly an artifact of the
relatively large territory size of these species
and use of fixed radius sampling (detections
>100 m were not included in the analysis).

Daily nest survival by species

PCwood

FIG. 2. Daily survival rates of Grasshopper and Henslow’s sparrow, and Eastern Meadowlark nests in relation to
amount of woody vegetation (PCwood) in proximity to nest-site locations at Tri-Valley and Woodbury Wildlife
Management areas in eastern Ohio, 2005-2006. Amount of woody vegetation increa.ses from - 1 to 3 along the x axis.

652

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. 4. December 2010

Winter and Faaborg (1999) also found no
significant association between woody vegetation
and density of these two species. The small
number of observations on surveys may have
limited our ability to identify patterns in cases
where we did not detect significant associations
(e.g.. Savannah Sparrow and Dickcissel).

Numbers of Grasshopper, Henslow’s, and
Savannah sparrows, and Bobolinks were nega-
tively related to percent cover of woody vegeta-
tion within 100 m of survey locations, but only
Grasshopper Sparrows showed strong evidence of
selecting nest patches with lower amounts of
woody vegetation than random locations. Our
analysis failed to discriminate between Eastern
Meadowlark and Henslow’s Sparrow nests and
random locations, perhaps because of small
sample size {n = 18 nests) or the conservative
nature of our comparison of used plots to random
plots versus used plots to non-used plots. Nest
locations of Eastern Meadowlarks had more than
two times less woody cover and fewer woody
patches within 100 m than random locations
(Aldredge and Griswold 2006, Buskirk and Mill-
spaugh 2006, Johnson et al. 2006, Thomas and
Taylor 2006). The literature on effects of woody
vegetation on nest-site selection of grassland birds
has yet to reveal consistent relationships, and
investigations show active avoidance of woody
vegetation during nest-site selection of Eastern
Meadowlarks, Henslow’s and Grasshopper spar-
rows, and Dickcissels (Hull 2000, Winter et al.
2000, Hubbard et al. 2006), positive associations
with trees for Grasshopper Spairows (Sutter and
Ritchison 2005), or no association between nest-
site .selection and presence of woody vegetation
for Vesper and Clay-colored sparrows (Grant et
al. 2006). Grassland nesting passerines may use
the ab.sence of woody vegetation as a cue during
nest-site selection given that certain predators
(i.e., mice) and brood parasites (i.e., Brown-
headed Cowbird [Molothru.s ater]) frequently use
woody vegetation both as cover and visual
perches (With 1994).

Daily nest survival was related to the amount of
woody vegetation surrounding nests of Grasshop-
per and Henslow’s sparrows and marginally
related to Eastern Meadowlark nests, which is
consistent with other studies conducted elsewhere
in the Midwest (Winter et al. 2000, Scheiman et
al. 2003, Bollinger and Gavin 2004, Davis 2005;
but see Grant et al. 2006). Woody vegetation and
woodland edges may support a diverse predator

community not otherwise present in grassland
systems (Renfrew and Ribic 2003). Woody
elements are generally thought to attract predators
because they provide perches for avian predators
and promote use by other predators due to cover
and other resources (Thompson and Burhans
2003, Renfrew et al. 2005).

We suggest efforts to remove woody vegetation
will enhance the value of reclaimed surface mines
for grassland birds given the apparent avoidance
of woody vegetation and the overall low daily nest
survival of Grasshopper Span'ows and Eastern
Meadowlarks, coupled with results from other
studies relating low nesting success to presence of
woody vegetation. Removal of encroaching
woody plants will increase grassland area and
reduce edges and fragmentation. It may also
decrease numbers of woodland predators by
eliminating cover or perches and reducing move-
ment corridors. This should make grassland units
more productive and attractive to a diverse
community of grassland-nesting birds. Our study
emphasizes the role of woody vegetation in
grassland bird management on reclaimed surface
mines, but managers should also remain attentive
to landscape and patch-scale issues given their
known influence on grassland birds (Ribic and
Sample 2001, Bakker et al. 2002, Winter et al.
2005).

ACKNOWLEDGMENTS

R. .1. Gates and Daniel Shiistack provided statistical
advice, and Jacob Straub provided GIS expertise. We thank
Kacy Ray, Michelle Schroeder, Beth Geboy, and Jennifer
Haney for assistance in the field. Financial support for this
project was provided by the Ohio Department of Natural
Resources, Division of Wildlife and by the Ohio Biological
Survey through their small grants program.

LITERATURE CITED

Aldredge, .1. R. and J. Griswold. 200b. Design and
analysis of resource selection studies for categorical
resource variables. Journal of Wildlife Management
70:337-346.

Askins, R. a. 1993. Population trends in grassland,
shrubland, and forest birds in eastern North America.
Current Ornithology I 1:1-34.

Bajema, R. a., T. L. DeVault, P. E. Scott, and S. L.
Lima. 2001. Reclaimed coal mine grasslands and their
significance for Henslow's Sparrows in the American
Midwest. Auk 1 18:422-431.

Barker, K. K.. D. E. Nauct-e, and K. F. Higgins. 2002.
Incorporating landscape attributes into models for
migratory grassland bird con.servation. Conservation
Biology 16:1638-1646.

Graves et al. • GRASSLAND BIRD CONSERVATION ON RECLAIMED MINES

653

Bibby, C. J., N. D. Burgess, D. A. Hill, and S, H.
Mustoe. 2000. Bird census techniques. Academic
Press, San Diego, California, USA.

Bollinger, E. K. and T. A. Gavin. 2004. Responses ol
nesting Bobolinks (DoUchonyx oryzivonts) to habitat
edges. Auk 121:767-776.

Brothers, T. S. 1990. Surface-mine grasslands. Geograph-
ical Review 80:209-225.

Buskirk, S. W. and J. J. Millspaugh. 2006. Metrics for
studies of resource selection. Journal of Wildlife
Management 70:358-366.

Cully, J. F. and H. L. Michaels. 2000. Henslow’s
Sparrow habitat associations on Kansas tallgrass
prairie. Wilson Bulletin 112:115-123.

D.avis, S. K. 2005. Nest-site selection patterns and the
influence of vegetation on nest survival of mixed-grass
prairie passerines. Condor 107:605-616.

DeVault, T. L., P. E. Scott, R. A. Bajema, and S. L.
Lim.a. 2002. Breeding bird communities of reclaimed
coal-mine grasslands in the American Midwest.
Journal of Field Ornithology 73:228-275.

ESRl. 2005. ArcGIS Version 9.1. Environmental Systems
Research Institute, Redlands, California, USA.

Fletcher, R. J. and R. R. Koford. 2002. Habitat and
landscape associations of breeding birds in native and
restored grasslands. Journal of Wildlife Management
66:101 1-1022.

Galligan, E. W., T. L. DeVault, and S. L. Lima. 2006.
Nesting success of grassland and savanna birds on
reclaimed coal mines of the midwestern United States.
Wilson Journal of Ornithology 118:537-546.

Grant. T. A., E. Madden, and G. B. Berkey. 2004. Tree
and shrub invasion in northern mixed-grass prairie:
implications for breeding grassland birds. Wildlife
Society Bulletin 32:807-818.

Grant, T. A., E. Madden, T. L. Shaffer, P. J. Pietz, G. B.
Berkey, and N. J. Kadrmas. 2006. Nest survival of
Clay-colored and Vesper sparrows in relation to
woodland edge in mixed-grass prairies. Journal of
Wildlife Management 70:691-701.

Herkert, j. R. 1994. The effects of habitat fragmentation
on midwestern grassland bird communities. Ecological
Applications 4:461^7 1 .

Hubbard, R. D,, D. P. Althoff, K. A. Blecha, B. A.
Bruvold. and R. D. Japuntich. 2006. Nest site
characteristics of Eastern Meadowlarks and Grasshop-
per Sparrows in tallgra.ss prairie at the Fort Riley
Military Installation, Kansas. Transactions of the
Kansas Academy of Science 109:168-174.

Hull, S. D. 2000. Effects of management practices on
grassland birds: Eastern Meadowlark. USDI, Geolog-
ical Survey, Northern Prairie Wildlife Research
Center, Jamestown. North Dakota. USA.

Ingold, D. J. 2002. Use of reclaimed stripmine by
grassland nesting birds in east-central Ohio. Ohio
Journal of Science 102:56-62.

Johnson, C. J., S. E. Nielsen, E. H. Merrill, T. L.
McDonald, and M. S. Boyce. 2006. Re.source
selection functions based on use-availability data:
theoretical motivation and evaluation methods. Journal
of Wildlife Management 70:347-357.

Johnson, D. H. and L. D. Icjl. 2001. Area requirements of
grassland birds: a regional perspective. Auk 118:24-
34.

Monroe, M. S. and G. Rlichison. 2005. Breeding biology
of Henslow’s Sparrow on reclaimed coal mine
grasslands in Kentucky. Journal of Field Ornithology
76:143-149.

Ohio Department of Natural Resources (ODNR).
2007. Woodbury Wildlife Management Area. ODNR,
Division of Wildlife, Columbus, Ohio, USA.

O’Leary, C. H. and D. W. Nyberg. 2000. Treelines
between fields reduce the density of grassland birds.
Natural Areas Journal 20:243-249.

Peterjohn, B. G. and j. R. Sauer. 1999. Population status
of North American grassland birds from the North
American Breeding Bird Survey, 1966-1996. Studies
in Avian Biology 19:27^4.

Quinn, G. P. and M. J. Keough. 2002. Multivariate
analysis of variance and discriminant analysis. Pages
425-442 in Experimental designs and data analysis for
biologists (G. P. Quinn and M. J. Keough, Editors).
University Press, Cambridge, United Kingdom.

Ransom, D. and W. E. Pinchak. 2003. Assessing accuracy
of a laser rangefinder in estimating grassland bird
density. Wildlife Society Bulletin 31:460-463.

Renfrew, R. B. and C. A. Ribic. 2003. Grassland passerine
nest predators near pasture edges identified on
videotape. Auk 120:371-383.

Renfrew, R. B., C. A. Ribic, and J. L. Nack. 2005. Edge
avoidance by nesting grassland birds: a futile strategy
in a fragmented landscape. Auk 122:618-636.

Ribic, C. A. and D. W. Sample. 2001. Associations of
grassland birds with landscape factors in southern
Wisconsin. American Midland Naturalist 146:105-
121.

Rummel, S. M. and F. j. Brenner. 2003. Relationships
between vegetation and bird populations in reclaimed
mine grasslands and native grasslands. Journal of The
Pennsylvania Academy of Science 76:56-61.

SAS Institute. 2002. SAS/STAT Software. Release 9.1.
SAS Institute Inc., Cary, North Carolina, USA.

Sauer, J. R., J. E. Hines, and J. Fallon. 2005. The North
American Breeding Bird Survey, results and analysis
1966-2004. Version 2005.2. USGS, Patuxent Wildlife
Research Center, Laurel. Maryland, USA.

SCHEIMAN, D. M., E. K. Bollinger, and D. H. Johnson.
2003. Effects of leafy spurge infestation on grassland
birds. Journal of Wildlife Management 67:1 15-121.

ScoiT, P. E., T. L. DeVault. R, A. Bajema. and S. L.
Lima. 2002. Grassland vegetation and bird abundances
on reclaimed midwestern coal mines. Wildlife Society
Bulletin 30:1006-1014.

Shaffer, T. L. 2004. A unified approach to analyzing nest
success. Auk 121:526-540.

Sutter, B. and G. Ritchison. 2005. Effects of grazing on
vegetation structure, prey availability, and reproduc-
tive success of Grasshopper Sparrows. Journal of Field
Ornithology 76:345-351.

Thomas, D. L. and E. J. Taylor. 2006. Study designs and
tests for comparing resource use and availability. II.
Journal of Wildlife Management 70:324-336.

654

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, December 2010

Thompson, F. R. and D. E. Burhans. 2003. Predation of
songbird nests differs by predator and between field
and forest habitats. Journal of Wildlife Management
67:408^16.

Vickery, P. D. 1996. Grasshopper Sparrow {Ammodramus
savannarum). The birds of North America.
Number 239.

Vickery, P. D. and J. R. Herkert. 2001. Recent advances
in grassland bird research: where do we go from here?
Auk 1 18:1 1-15.

Walk, J. W. and R. E. Warner. 2000. Grassland
management for the conservation of songbirds in the
midwestem USA. Biological Conservation 94:165-
172.

Whitmore, R. C. 1980. Reclaimed surface mines as avian
habitat islands in the eastern forest. American Birds
34:13-14

Whitmore, R. C. and G. A. Hall. 1978. The response of
passerine species to a new resource: reclaimed surface
mines in West Virginia. American Birds 32:6-9.

Winter, M. and J. Faaborg. 1999. Patterns of area

sensitivity in grassland-nesting birds. Conservation
Biology 13:1424-1436.

Winter, M., D. H. Johnson, and J. Faaborg. 2000.
Evidence for edge effects on multiple levels in
tallgrass prairie. Condor 102:256-266.

Winter, M., D. H. Johnson, and J. A. Shaffer. 2005.
Variability in vegetation effects on density and nesting
success of grassland birds. Journal of Wildlife
Management 69:185-197.

Winter, M., D. H. Johnson, J. A. Shaffer, T. M.
Donovan, and W. D. Svedarsky. 2006. Patch size
and landscape effects on density and nesting success of
grassland birds. Journal of Wildlife Management
70:158-172.

With, K. A. 1994. The hazards of nesting near shrubs for a
grassland bird, the McCowns’ Longspur. Condor
96:1009-1019.

Wray, T., K. A. Strait, and R. C. Whitmore. 1982.
Reproductive success of grassland sparrows on a
reclaimed surface mine in West Virginia. Auk 99:157-
164.

The Wilson Journal of Ornithologv 122(4):655 665, 2010

EFFECTS OF BREEDING STAGE AND BEHAVIORAL CONTEXT ON
SINGING BEHAVIOR OF MALE INDIGO BUNTINGS

MATTHEW D. BECKETT' AND GARY RITCHISON'-

ABSTRACT. — We studied the effects of breeding stage and behavioral context on the singing behavior of male Indigo
Buntings (Passerina cyanea\ n = 15) during the 2004 breeding season in Madison County, Kentucky, USA, to better
understand how males with a single-song repertoire vary the characteristics of their song to convey different information.
Playback experiments were conducted in 2005 in territories of focal males (n = 14) to further examine the possible effect of
male-male interactions on singing behavior. We analyzed 10,919 songs of 15 male Indigo Buntings with songs consisting of
a series ot figures that were usually paired (i.e., phrases). Mean song duration was 2.30 ±0.13 (SE) sec (range = 1.44-
3.40 sec) with males varying song duration by varying the number of figures and phrases in each song. Singing rates varied
significantly (P < 0.0001) among breeding stages and were highest prior to pairing, suggesting singing has a role in mate
attraction. Singing rates also differed (P = 0.013) during playback experiments with rates higher during playback and post-
playback periods (Jf = 4.4 songs/min) than during the pre-playback period (x = 2.9 songs/min). These results suggest that
singing also has a role in territory defense. Songs of male buntings tended to be shorter prior to pairing and were generally
longer in duration after pairing. Playback experiments revealed that bunting songs were longer {P = 0.03) during and after
playback (x = 2.6 sec) than during the pre-playback period (x = 2.0 sec). These results suggest male Indigo Buntings vary
singing rates and song duration to convey different information, appearing to use shorter songs uttered at high rates to
attract mates and longer songs to convey aggression during male-male interactions. Received 21 December 2009. Accepted
J June 2010.

Males of most species of passerines have multi-
song repertories (MacDougal-Shackleton 1997).
These repertoires may be important in male-male
interactions (song matching; Stoddard et al.
1992), preventing habituation (Krebs et al. 1978,
Yasukawa 1981), providing information about
motivational state (Falls 1969), and repelling
intruders (Krebs et al. 1978, Yasukawa 1981).
Song repertoires may also have intersexual
functions. For example, female Atlantic Canaries
{Serinus canaria) exhibit more nest building
behavior when subjected to playback of larger
repertoires (Kroodsma 1976).

Males of some species, such as Ovenbirds
(Seiurus aurocapillcv. Weary and Lemon 1988) and
Henslow’s Sparrows {Ammodramus henslowii, Left-
wich and Ritchison 2008), have a single song type.
Males with single-song repertoires could potentially
use singing to convey different information, but may
do so by altering the characteristics of their single
song (Morton and Young 1986, Leftwich and
Ritchison 2008). However, few investigators have
examined how male songbirds with single-song
repertoires vary their singing behavior or the
characteristics of their songs during the breeding
season and in different behavioral contexts.

Male Indigo Buntings (Passerina cyanea) have
a repertoire of one song type or theme (Payne

' Department of Biological Sciences, Eastern Kentucky
University, Richmond, KY 40475, USA.

^Corresponding author; e-mail: gary.ritchison@eku.edu

2006), and little is known about how their singing
behavior varies during the breeding season and in
different social contexts (Thompson 1972, Shio-
vitz 1975). We examined the effects of breeding
stage and behavioral context on their singing rates
and singing behavior to better understand ( 1 ) the
possible functions of singing by male Indigo
Buntings, and (2) how males with a single-song
repertoire might vary their singing behavior to
convey different information.

METHODS

We studied Indigo Buntings from 26 April to
20 August 2004 and 1 9 May to 5 June 2005 at the
Central Kentucky Wildlife Management Area,
17 km southeast of Richmond, Madison County,
Kentucky, USA. Male buntings were observed
and recorded in 2004, and playback experiments
were conducted in 2005. The study area consisted
primarily of wooded edges, scattered trees, and
lields in early stages of succession. Male buntings
were captured using playback of conspecific
songs to lure them into mist nets. Captured
buntings were banded with a U.S. Geological
Survey aluminum band plus a unique combination
of three colored plastic bands to aid in individual
recognition. Male age (second-year [SY] or after-
,second-year [ASY]) was a.ssigned using plumage
differences (Pyle 1997, Payne 2006).

Teiritory boundaries were delineated by mon-
itoring movements of focal males and noting

655

656

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, December 2010

locations of interactions with conspecifics. Focal
individuals were observed and recorded 2-3 times
per week during observation periods of 20-30 min.
All observations were conducted during the
period from just after sunrise to 1100 hrs (EDT)
when singing rates of Indigo Buntings are
relatively constant (Thompson 1972). Inclement
weather may influence singing behavior so
buntings were not observed when it was raining.
The date, time, and focal bird’s identity were
noted on tape prior to each session. Recordings
were made using a Sony (TCM-50DV) cassette
recorder with a Sennheiser shotgun microphone
(Model ME 88), typically at a distance of 10-30 m
from focal males.

All songs of focal males were recorded during
each observation period and the breeding stage
was noted. Songs were categorized during each
observation, as: ( 1 ) spontaneous advertising (no
conspecific males heard or observed), (2) long-
range interaction, where a conspecific male was
singing in an adjacent or distant territory >50 m
away, and (3) short-range interaction, where a
conspecific male was singing <50 m away. Eocal
male songs were classified as long- and short-
range up to 1 min after a conspecific was last
heard singing.

We divided the breeding season into six stages:
(1) pre-pairing (no female present on male’s
territory), (2) nest building (period of nest
construction by the female), (3) egg-laying (from
the day the first egg was laid to the day the
penultimate egg of a clutch was laid), (4)
incubation (from the day the last egg was laid to
the day before first egg hatched), (5) nestling
(from the day the first egg hatched to the day
before young fledged), and (6) post-fledging
(from the day the young leave the nest to the
time when fledglings leave male’s territory).
Female buntings frequently produce two broods
during the breeding season and often began new
nests several days after young from a previous
nest Hedged. Thus, the post-Hedging period ended
when a female began building a new nest.

An attempt was made to locate and monitor all
nests in the territories of focal males to identify
breeding stages. Nests were located by observing
movements of females during nest building and
nestling provisioning, and by searching likely nest
sites in focal territories. Nests were found at
several stages and we, at times, backdated to
ascertain the dates of breeding stages. We used
the following durations for backdating: 8 (lirst

nest of the season) or 2 (later nests) days for nest
building, 1 day for each egg in a clutch for egg
laying, 12 days for incubation, and 9 days for the
nestling period (Bradley 1948, Payne 2006).

Playback experiments were conducted in 2005
in territories of focal males (/? = 14) to further
examine the possible effect of male-male interac-
tion on the singing behavior of male Indigo
Buntings. All songs used for the playback
experiments were taken from commercially-avail-
able recordings (Common Birds and Their Songs,
Houghton Mifflin Company; More Binding by Ear
— Eastern/Central, Houghton Mifflin Company,
Boston, MA, USA). Three tapes, each with songs
of a different male, were used in the experiments
(Tape 1 means: song length = 1.8 sec, number of
figures = 4.3, number of phrases = 5.3; Tape 2
means: song length = 2.9 sec, number of figures
= 12.5, number of phrases = 7.5; Tape 3 means:
song length = 2.8 sec, number of figures = 14.7,
number of phrases = 9.3; definitions of figure and
phrase are below). Tapes I, 2, and 3 had three,
four, and three different versions of each male’s
song, respectively. Male buntings on all three
tapes sang at a rate of four songs/min. Recordings
of males singing at higher rates would have
increased the probability of the songs of focal
males overlapping those on the recordings,
making subsequent song analysis more difficult.
A speaker was placed near the center of the focal
male’s territory to simulate singing by an
intruding male during each experiment. Each
experiment consisted of a 3-min pre-playback
period and a 6-min playback/post-playback period
(with songs played back for 3 min and observa-
tions continuing for an additional 3 min after
playback ended).

All songs recorded during each observation
period were analyzed. Characteristics of songs
measured included song duration, number of
figures (“a sound which produces a single,
complete, and distinct impression uninterrupted
by silence greater than two centiseconds”;
Shiovitz 1975:133), and the number of phrases
(“a subdivision of a song based upon recogniz-
able groups of sequential figures”; Shiovitz
1975:133; Fig. 1). In addition, we noted if songs
had “squeaky” notes added or were uttered at
higher than normal volume during the playback
experiments. High-frequency “squeaky” notes
are occasionally inserted between phrases in
.songs of male Indigo Buntings during territorial
disputes (Emlen 1972, Thompson 1972, Margo-

Beckett and Ritehisott • SINGING BEHAVIOR Ol' INDIGO BUN TINGS

657

( ^ r

Phrases

or n I 1 nrn

A/ \

N

X

o

c;

(U

cr

0)

Time (sec)

FIG. 1. Spectrograph of a typical song of a male Indigo Bunting showing the terminology used.

Hash et al. 1994). Song volume was subjectively
categorized as either normal or high volume. We
attempted to maintain a constant distance from
focal males during playback experiments to
permit better estimation of song volume. Record-
ings of the songs of focal males were subsequent-
ly analyzed using sound-analysis software (Raven
Version 1.2.1, Cornell Laboratory of Ornithology,
Ithaca, NY, USA).

Possible variation in characteristics of songs of
male Indigo Buntings with male age, breeding stage,
and behavioral context were examined using
repeated measures analysis of variance with Stu-
dent-Newman-Keuls (SNK) post-hoc tests to exam-
ine possible differences among means. All analyses
were conducted using the Statistical Analysis
System (SAS Institute 1999). All values are
presented as mean ± standard emor. Mean values
were calculated using the means for each focal male
to avoid bias due to different sample sizes.

RESULTS

We found no differences in characteristics of
songs of SY and ASY male Indigo Buntings (song
duration, number of figures, and number of
phrases; all P > 0.17). Thus, songs of all males
(SY and ASY) were combined for subsequent
analyses.

All male Indigo Buntings (/? = 15) in our study
had a repertoire of a single song type. Mean song

duration was 2.30 ± 0.13 sec and the mean
number of figures per song was 9.95 ± 0.5 1 . The
mean number of phrases per song was 6.15 ±
0.46.

Songs of male buntings exhibited both intra-
(Fig. 2) and inter-individual (Fig. 3) variation.
Each male had a typical song that included a
series of figures (phrases), usually paired, and
usually uttered in the same order. The number of
figures in each phrase occasionally varied, and
figures (phrases) typically used later in songs
were often omitted (Fig. 2).

Songs of some males (5 of 15, 33%) were
unique and shared few or no song figures with
other males. Males in adjacent temtories typically
shared some and, at times, all of the figures used
in their songs and formed song neighborhoods
where males used either the same or similar song
figures (10 of 15, 67%; Fig. 4). One SY male in
our study altered his song shortly after arriving in
the study area (Fig. 5).

Effects of Breeding Stage and Natural Interac-
tions on Stngitig Behavior. — Singing rates (songs/
min) of male Indigo Buntings varied significantly
among breeding stages (F5.45 = 15.1. P < 0.001)
with rates highest during the pre-pairing stage
(SNK test, P < 0.05; Fig. 6). Singing rates during
the other breeding stages (nest-building, egg-
laying, incubation, nestling, and post-fledging)
did not differ (SNK te.st, P > 0.05; Fig. 6).

658 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4. December 2010

Time (sec)

FIG. 2. Spectrographs illu.strating the typical variation that exists in the single song of a male Indigo Bunting.

8

6

4

2

8

6

4

2

8

6

4

2

8

6

4

2

8 ■

6 ■

4 ■

2 -

ahs

tchison • SINGING BEHAVIOR OE INDIGO BUNTINGS

659

0.5 1 0 1 5 2.0 2.5 3.0 3.5 4 0

Time (sec)

istrating the typical variation among songs of different male Indigo Buntings.

8

6

4

2

8

6

4

2

8

6

4

2

8

6

4

2

r'.»

M JOURNAL OF ORNITHOLOGY • VoL 122, No. 4. December 2010

J I I I L

Male 1

Male 16

0.5

1.0

1.5 2.0 2.5 3.0

Time (sec)

3.5 4.0

the close matching of songs of .several neighboring male Indigo Buntings in a

Becken and RUchison • SINGING BEHAVIOR OF INDIGO BUNTINGS

66 1

N

>^

O

c

0

13

O"

0

Time (sec)

FIG. 5. Spectrographs of two songs uttered by a second-year (SY) Indigo Bunting male illustrating song switching. The
first song was used immediately after arrival on his territory. Several days later, the second song incorporated figures and
phrases from those of a neighboring male.

Differences in the mean duration of songs
uttered by male Indigo Buntings during different
breeding stages approached significance (Fs^i =
2.2, P = 0.075) with longer songs during the nest-
building period (Fig. 7). Songs were shorter in
duration during post-fledging and pre-pairing
periods than during other breeding stages (Fig. 7).

Differences among breeding stages in the mean
number of figures (F^2,\ = 2.1, F = 0.099) and
phrases (F531 = 2.1, P = 0.094) per song also
approached significance with more figures and
phrases per song during nest-building and egg-
laying stages and fewer during post-fledging and

FIG. 6. Mean singing rates (± SE) of male Indigo
Buntings during different breeding stages.

pre-pairing stages (Fig. 8). None of the charac-
teristics of songs of male buntings, including
duration and number of figures and phrases per
song differed among behavioral contexts (short-
range interaction, long-range interaction, or spon-
taneous singing; P > 0.24).

Effects of Song Playback on Singing Behav-
ior.— We compared singing behavior of male
Indigo Buntings prior to presentation of the
stimulus (pre-playback period) to that during and
after presentation of the stimulus (playback and
post-playback periods combined). Singing rates
(songs/min) differed (F,jo = 8.3, P = 0.013)
between periods with higher rates during the

4

Pre-pairing Nest- Egg- Incubation Nestling Post-
building laying fledging

Breeding stage

FIG. 7. The mean duration of .songs (± SE) uttered by
male Indigo Buntings during different breeding stages.

662

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, Nu. 4. December 2010

FIG. 8. Mean number of figures and phrases (± SE) in
songs of male Indigo Buntings during different
breeding stages.

playback and post-playback periods (x = 4.4 ±
0.4 songs/min) than during the pre-playback
period (.v = 2.9 ± 0.6 songs/min). The character-
istics of songs also differed between periods with
longer songs (/^uo = 6.4, P = 0.03) that included
more figures iF\,\o = 6.6, P = 0.028) and phrases
(^1.10 = 5.0, P = 0.049) uttered during and after
playback (x duration = 2.6 ± 0.1 sec; x number of
figures = 1 1.4 ± 0.1; x number of phrases = 6.8
± 0.2) than prior to playback (x duration = 2.0 ±
0.1 sec; X number of figures = 8.8 ± 0.4; x
number of phrases = 5.5 ± 0.2). Male buntings
also uttered a significantly higher percentage
(6’i.io = 6.3, P — 0.026) of high-volume songs
during the post-presentation period (x = 11.4 ±
3.7%) than the pre-presentation period (none
uttered). Comparison of pre- and post-presenta-
tion periods revealed no difference in use of songs
including high-frequency ‘squeaky’ notes (Fijo
= 0.02, P = 0.9).

DISCUSSION

Song Matching. — All male Indigo Buntings in
our study had a repertoire of one song type. Songs
of some male buntings were unique, as also noted
by previous investigators (Thompson 1970, Payne
et al. 1988). Most males with adjacent territories
in our study shared some, if not all, of the figures
in their songs with conspecifics, and formed song
neighborhoods where males in adjacent territories
used the same or similar song figures. Song
neighborhoods have been reported in other
populations of Indigo Buntings (Thompson
1970, Payne and Westneat 1988). Song neighbor-
hoods may develop because incorporating figures
Li.sed by resident males into their .songs may be
beneficial for young male buntings with SY males
that song match more likely to successfully

acquire and defend a territory, acquire a mate,
and fledge young (Payne 1982, 1983).

Song Function. — Our results indicate singing
by male Indigo Buntings functions in mate
attraction with singing rates declining significant-
ly after pairing. Thompson (1972) also reported
male Indigo Buntings sang more songs per bout
and more bouts/hr when unmated. Previous
studies suggest male singing rates represent a
performance-related trait under positive sexual
selection (Gil and Gahr 2002) that can influence
mate choice and timing of mate acquisition
(Alatalo et al. 1990, Hoi-Leitner et al. 1995).
Otter et al. (1997) reported that singing rates
represented accurate signals of male quality in
Black-capped Chickadees {Poecile atricapillus).
Singing rates of male buntings and males of other
species of songbirds (e.g., Albrecht and Oring
1995, Ritchison 1995, Balsby 2000) decline after
pairing. Thus, high singing rates prior to pairing
may provide females with information concerning
male status (paired or not) and their quality.

Male Indigo Buntings in our study continued
singing after pairing, but at lower rates. Singing
by male buntings, both before and after pairing, as
for males of many species of songbirds (e.g.,
Krebs et al. 1978, Nowicki et al. 1998), is likely
important in territorial defense. Male Indigo
Buntings typically airive in breeding areas ~1-
2 weeks before females (Payne 2006; GR, pers.
obs.) and establish breeding territories. During
this period, and throughout our study, males were
at times observed counter-singing with neighbor-
ing males and chasing conspecifics near temtory
boundaries while singing; males responded to
playback by increasing their singing rates. These
observations indicate that singing by male Indigo
Buntings has an important role in establishing and
defending territories.

Singing rates of male Indigo Buntings in our
study were lowest during the fertile periods of
their mates (nest-building and egg-laying periods)
and remained relatively low during the incuba-
tion, nestling, and post-fledging periods. Some
investigators have suggested singing may be
important for stimulating female reproductive
cycles (Hinde and Steel 1976) or guarding mates
(Mpller 1988). Singing by male Indigo Buntings
does not appear to serve the.se functions because
singing rates in our study were lowest during
female fertile periods. Males of most songbird
species do not sing or sing al low rates during the
fertile periods of their mates (Gil et al. 1999).

Beckett and Ritehison • SINGING BEHAVIOR OF INDIGO BUN I INGS

663

Song Characteristics. — Songs of male Indigo
Buntings in our study tended to be shorter prior to
pairing and were generally longer in duration after
pairing. In addition, bunting songs were signifi-
cantly longer during and after playback than
during the pre-playback period. These results
suggest male Indigo Buntings may use shorter
songs to attract females, and longer songs to
convey aggression during male-male interactions.
Although females in some species appear to prefer
males that sing longer songs (e.g., White-throated
Sparrows, Zonotrichia albicollis', Wasserman and
Cigliano 1991), females in other species prefer
males that sing at faster rates (e.g., European Pied
Flycatcher, Ficedula hypoleuca\ Gottlander 1987,
Alatalo et al. 1990). If mate choice decisions by
female Indigo Buntings are influenced by singing
rates, then one possible explanation for the
tendency of males to sing shorter songs prior to
pairing is that such songs allow males to sing at
higher rates (i.e., more songs per unit time). In
addition, however, given that longer songs appear
to convey an increased tendency for aggression,
male buntings may also sing shorter songs prior to
pairing to signal their non-aggressive intentions to
potential mates. Similarly, Staicer (1996) suggest-
ed that male Adelaide’s Warblers (Dendroica
adelaidae) may use songs with characteristics that
convey less aggression (‘appeasing songs’) during
interactions with females.

Male buntings in our study often added
additional phrases to their songs during encoun-
ters with other males, both in response to
playback and when chasing conspecifics. Emlen
(1972) noted that buntings used ‘lengthened’
songs during agonistic encounters, and Saunders
(1929:48) witnessed “male Indigo Buntings
chasing each other about from tree to tree, and
both of them singing prolonged songs in flight.’’
In addition, Shiovitz (1975) found that playback
of long variations of songs elicited stronger
responses by male buntings than shorter varia-
tions. Shiovitz (1975) also suggested the first few
figures of the songs of male buntings likely
functioned as a “sign on’’ in gaining the attention
of conspecifics, whereas the remainder of the song
seemed to convey aggression. Thus, male Indigo
Buntings appear to increase song duration to
convey aggression, and perhaps the likelihood of
interacting, during intrasexual encounters. Males
of other species also appear to use longer songs to
convey an increased likelihood of aggression
during intrasexual encounters (e.g., McGregor

and Horn 1992, Balsby and Dabelsteen 2001,
Leitao et al. 2006, Lattin and Ritehison 2009).

We found male Indigo Buntings at times added
‘squeaky notes’ to songs and sang with increa.sed
volume during male-male interactions. Thompson
(1972) also indicated male Indigo Buntings
inserted high-frequency (about 9 kHz), ‘squeaky’
notes into songs when an intruding male was
present in a territory. Emlen (1972) noted that
male buntings at times added high-pitched
‘.squeak’ notes when responding aggressively to
playback of conspecific songs. These results
suggest male buntings alter the characteristics of
their single song to convey an increased likeli-
hood of aggression, i.e., singing longer, louder
songs that at times include ‘squeaky’ notes. In
further support of this hypothesis, male buntings
in our study at times inserted high-frequency
‘squeaky notes’ into songs during territorial
chases with conspecifics and uttered songs at
higher than normal volume when countersinging
at territory boundaries (MDB and GR, pers. obs.).
Changes in characteristics of songs during in-
trasexual encounters have also been reported in
other species of songbirds. For example, male
Barn Swallows (Hintndo rustica) emphasize
‘rattles’ in their songs during aggressive encoun-
ters, indicating that ‘rattles’ are an important
component of competitive interactions between
males (Galeotti et al. 1997). Similarly, American
Pipits {Anthus rubescens) incorporate ‘snarr’
notes, a rasping element with a broad frequency
range, into their songs during territorial disputes
(Rehsteiner et al. 1998).

Our results suggest the single song type of male
Indigo Buntings serves several functions, includ-
ing mate attraction and temtory defense. In
addition, male buntings vary song rates and song
duration during the breeding season and in
different behavioral contexts, appearing to use
shorter songs uttered at high rates to attract mates
and longer songs to convey aggression during
male-male interactions.

ACKNOWLEDGMENTS

We (hank E. R. A. Cramer, C. E. Braun, and an
anonymous reviewer for comments that improved our
paper, and the Eastern Kentucky University Research
Committee for providing funds to support our research.

LITERATURE CITED

At.ATALO, R. V., C. Glynn, and A. Lundberg. 1990.

Singing rate and female attraction in the Pied

664

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4. December 2010

Flycatcher: an experiment. Animal Behaviour 39:
601-603.

Albrecht. D. J. and L. W. Oring. 1995. Song in Chipping
Sparrows, Spizella passerimr. structure and function.
Animal Behaviour 50:1233-1241.

B.alsby, T. J. S. 2000. The function of song in Whitethroats
Sylvia communis. Bioacoustics 11:17-30.

B.alsby. T. J. S. and T. Dabelsteen. 2001. The meaning
of song repertoire size and song length to male
Whitethroats Sylvia communis. Behavioural Processes
56:75-84.

Bradley, H. L. 1948. A life history study of the Indigo
Bunting. Jack-Pine Warbler 26:103-1 13.

Emlen, S. T. 1972. An experimental analysis of the
parameters of bird song eliciting species recognition.
Behaviour 41:130-171.

Falls, J. B. 1969. Functions of territorial song in the
White-throated Sparrow. Pages 207-232 in Bird
vocalizations (R. A. Hinde, Editor). Cambridge
University Press, Cambridge, United Kingdom.

Galeotti, P., N. Saino, R. Sacchi, and A. P. Mdller.
1997. Song correlates with social context, testosterone
and body condition in male Barn Swallows. Animal
Behaviour 53:687-700.

Gil, D. and M. Gahr. 2002. The honesty of bird song:
multiple constraints for multiple traits. Trends in
Ecology and Evolution 17:133-141.

Gil, D., j. a. Graves, and P. J. B. Slater. 1999. Seasonal
patterns of singing in the Willow Warbler: evidence
against the fertility announcement hypothesis. Animal
Behaviour 58:995-1000.

Gori lander, K. 1987. Variation in the song rate of the
male Pied Elycatcher Ficedula hypoleuca: causes and
consequences. Animal Behaviour 35:1037-1043.

Hinde, R. A. and E. Steel. 1976. The effect of male song
on an oestrogen dependent behavior in the female
Canary {Serinus canarius). Hormones and Behavior
7:293-304.

Hoi-Leitner, M., H. Nechtelberger, and H. Hoi. 1995.
Song rate as a signal for nest site quality in Blackcaps
(Sylvia atricapilla). Behavioral Ecology and Sociobi-
ology 37:399-405.

Krebs, J. R., R. Ashcroft, and M. Webber. 1978. Song
repertoires and territory defen,se in the Great Tit.
Nature 271:539-542.

Kroodsma, D. E. 1976. Reproductive development in a
female songbird: differential stimulation by quality of
male song. Science 192:574-575.

La ttin, C. and G. Ritchi.son. 2009. Intra- and intersexual
functions of singing by male Blue Grosbeaks: the role
of within-.song variation. Wilson Journal of Ornithol-
ogy 121:714-721.

Leftwich. C. and G. Ritchi.son. 2008. Singing behavior of
male Henslow's Sparrows (Ammoclramus lienslowii):
description and causation. Bird Behavior 18:1-7.

Leitao. a.. C. T. Cate, and K. Riebei,. 2006. Within-.song
complexity in a .songbird is meaningful to both male
and female receivers. Animal Behaviour 71:1289-
1 296.

MacDou(;ai.-Shackleton, S. A. 1997. Sexual selection

and the evolution of song repertoires. Current
Ornithology 14:81-124.

Margoliash, D., C. Staicer, and S. A. Inoue. 1994. The
process of syllable acquisition in adult Indigo Buntings
(Passerine! cyanea). Behaviour 131:39-64.

McGregor, P. K. and A. G. Horn. 1992. Strophe length
and response to playback in Great Tits. Animal
Behaviour 43:667-676.

Moller, a. P. 1988. Spatial and temporal distribution of
song in the Yellowhammer Emheriza citronella.
Ethology 78:321-331.

Morton, E. S. and K. Young. 1986. A previously
undescribed method of song matching in a species
with a single song "type”, the Kentucky Warbler
(Oporornis formosus). Ethology 73:334-342.

Nowicki, S., W. a. Searcy, and M. Hughes. 1998. The
teiTitory defense function of song in Song Sparrows: a
test with the speaker occupation design. Behaviour
135:615-628.

Otter, K. A., B. Chruszcz, and L. Ratcliffe. 1997.
Honest advertisement and song output during the dawn
chorus of Black-capped Chickadees. Behavioral Ecol-
ogy 8:167-173.

Payne, R. B. 1982. Ecological consequences of song
matching: breeding success and intraspecific song
mimicry in Indigo Buntings. Ecology 63:401-41 1.

Payne, R. B. 1983. The social context of song mimicry:
song-matching dialects in Indigo Buntings (Passerina
cyanea). Animal Behaviour 31:788-805.

Payne, R. B. 2006. Indigo Bunting (Pas,serina cyanea).
The birds of North America. Number 4.

Payne, R. B. and D. F. Westneat. 1988. A genetic and
behavioral analysis of mate choice and song neigh-
borhoods in Indigo Buntings. Evolution 42:935-947.

Payne, R. B., L. L. Payne, and S. M. Doehlert. 1988.
Biological and cultural success of song memes in
Indigo Buntings. Ecology 69:104-1 17.

Pyle, P. 1997. Identification guide to North American
birds. Part I. Slate Creek Press, Bolinas, California,
USA.

Reh.steiner, U., H. Geisser, and H. U. Reyer. 1998.
Singing and mating success in Water Pipits: one
specific song element makes all the difference. Animal
Behaviour 55:1471-1481.

Ritchison, G. 1995. Characteri.stics, use, and possible
functions of the perch songs and chatter calls of male
Common Yellowthroats. Condor 97:27-38.

SAS iN.STtTUTE. 1999. SAS user's guide: statistics. SAS
Institute Inc., Cary, North Carolina, USA.

Saundf.r.s, a. a. 1929. Bird song. The University of the
State of New York, Albany. USA.

Shiovitz, K. a. 1975. The process of species-specific song
recognition by the Indigo Bunting, Passerina cyairea,
and its relationship to the organization of avian
acoustical behavior. Behaviour 55: 128-179.

Staicer, C. A. 1996. Acoustical features of .song categories
of the Adelaide's Warbler (Denilroica aclelaiclae). Auk
1 13:771-783.

.Stoddard, P. K., M. D. Beecher, S. E. Campbell, andC.
L. Horning. 1992. Song-type matching in the Song
Sparrow. Canadian Journal of Zoology 70:1440-1444.

Beckett and Ritehison • SINGING BEHAVIOR OF INDIGO BUNTINGS

665

Thompson, W. L. 1970. Song variation in a population of
Indigo Buntings. Auk 87:58-71.

Thompson, W. L. 1972. Singing behavior of the Indigo
Bunting, Passerina cyanea. Zeitschrift fur Tierpsy-
chologie 31:39-59.

W.’XSSERMAN, F. E. AND J. A. CiGLIANO. 1991. Soilg output
and stimulation of the female in White-throated

Sparrows. Behavioral Ecology and Sociobiology
29:55-59.

Weary, D. M. and R. E. Lemon. 1988. Evidence against
the continuity-versatility relationship in bird song.
Animal Behaviour 36:1379-1383.

Yasukawa, K. 1981 . Song and territory defense in the Red-
winged Blackbird. Auk 98:185-187.

The Wilson Journal of Ornithology 122(4):666-673, 2010

DEVELOPMENT OE INCUBATION TEMPERATURE AND BEHAVIOR
IN THRUSHES NESTING AT HIGH ALTITUDE

MARTIN L. MORTON' - AND MARIA E. PEREYRA'

ABSTRACT. — Onset of incubation was studied in three Hermit Thrushes (Catharus guttatus) and one American Robin
(Turclus migratorius), all with four-egg clutches, at a high altitude site in the Sierra Nevada, California, USA. Behavior of
laying females at the nest was measured from continuous recordings of internal egg temperatures of first-laid eggs. Full
nocturnal nest attentiveness began immediately with the first egg. Daytime attentiveness increased steadily during laying as
toraging time away from the nest decreased. On-off bouts by tending females in daytime increased in frequency and
decreased in duration until the last egg was laid. Time on the nest could not be directly equated to occurrence of incubation
because eggs were not uniformly warmed to exceed the temperature threshold required for embryonic development
(physiological zero). Incubation began, both day and night, after laying of the second egg. It increased steadily thereafter
with percentage of daytime devoted to incubation lagging well behind that of nighttime. Nest attentiveness and incubation
temperatures reached maxima about the time of clutch completion and were continued during following days. Received 17
December 2009. Accepted 2 April 2010.

Development of nest-attentiveness and incuba-
tion behaviors in female birds is a physiological-
ly-driven process that begins during pre-laying
and laying periods as females spend progressively
greater amounts of time on the nest, and less time
in foraging or other outside activities. The actual
onset of the incubation phase, in which eggs reach
temperatures sufficiently high to sustain embry-
onic development, corresponds with development
of a functional brood patch and sustained
application of heat to the eggs.

The rate at which these behaviors develop is
influenced by underlying physiological mecha-
nisms associated with gonadal maturation, follic-
ular development and ovulation, and subsequent
morphological changes with the appearance of a
functional incubation patch. Subsequent behaviors
surrounding the start of a clutch, although in large
part driven by gonadal development initially
stimulated by lengthening days (Wingfield 1984,
Sharp et al. 1998, Dawson 2008), can be
modulated by environmental conditions including
availability and suitability of nest sites (Morton
1978, 2002), food supply (Hahn et al. 1995) and,
in montane settings, storms (Morton 2002, Hahn
et al. 2004). Once egg laying begins, the transition
from laying to incubation progresses in an orderly
fashion under the influence of a changing
hormonal environment that eventually becomes
dominated by prolactin (Vleck 1998).

The development of incubation behavior marks

' Department of Biological Sciences, University of Tulsa,
TuLsa, OK 74104, USA.

^Corresponding author;
e-mail: marty-morton@utulsa.edu

the transition between pre-nesting and maternal
phases of reproduction, and is usually considered
to occur gradually (Haftom 1978, 1981, 1988;
Zerba and Morton 1983; Oppenheimer and
Morton 2000; Wang and Weathers 2009). Be-
cause embryonic development does not occur
until eggs are warmed above a threshold of 25-
27° C (White and Kinney 1974), termed the
physiological zero temperature (PZT), the amount
of time the female applies heat during laying, and
prior to completion of a clutch, becomes a critical
determinant of embryonic growth, the length of
the incubation period, and subsequent amounts of
hatching asynchrony within a clutch (Mead and
Morton 1985). Much importance has been at-
tached to the age-size hierarchy of hatchlings and
many hypotheses have been advanced to explain
its assumed adaptive value (Clark and Wilson
1985; Stoleson and Beissinger 1995, 1999). The
widespread occurrence of hatching asynchrony
and potential for maternal shifts in timing of
incubation onset to exaggerate or suppress age-
size hierarchies in hatchlings makes accurate
measurements of events that accompany shifts in
nest attentiveness between laying and incubation
particularly important.

We demonstrate using egg temperature (Tegg)
data from two species of thrush (Turdidae) that,
although daytime attentiveness is frequently used
in ecological studies to mark the start of the
incubation period (and hence embryonic develop-
ment), the.se are not equivalent conditions. In
particular, we show the behavioral (daytime and
night-time attentiveness) and physiological com-
ponents of incubation (heat application) are
decoupled early in laying, and only gradually

666

Morton and Pcrevra • ONSET OF INCUBATION IN THRUSHES

667

converge into actual incubation, as defined by the
maintenance of eggs at temperatures above the
threshold for development.

METHODS

Nests from three Hermit Thrushes {Cathariis
guttatiis) and one American Robin {Turdus
migratohits) were used in a study conducted at
Tioga Pass (37.8° N, 1 19.2° W; 3,000 m asl), in
the Sierra Nevada, California, USA. All nests
were in lodgepole pine (Piniis contorta). We
inserted the tip of a 36-ga copper-constantan
thermocouple into the center of the first egg
within a few hours after it was laid, then glued the
insulated leads to the shell. Lead wires were
threaded through the nest bottom and connected to
a hidden strip-chart recorder set to move at 2 cm/
hr and calibrated to record between —5 and 45° C
over a chart width of 10 cm. We moved well away
from the nest during thermocouple installation
(about 30 min), but left behind a dummy egg of
the size and coloration of the thrush egg in case
the laying female returned. Air temperature (Ta)
was measured with a hygrothermograph placed in
a vented box, in the shade, near the base of the
nest tree. Temperature data for both Tegg and Ta
were collected during the first 10 days of June in
1982 or 1985. Each nest was visited daily for
quick inspections of nest contents and of the
recording equipment.

RESULTS

All four thrushes laid three additional eggs on
successive days after the first egg was implanted.
Tegg data indicated that none of the birds returned
to their nests on the first day (Day 1 ) until after
sunset (Fig. 1). They remained on the nest
throughout the night (~8 hrs) then left, presum-
ably to forage, before returning to lay their second
egg. One individual (Fig. lA), for example, first
returned to her nest on Day 1 at 2001 hrs PST,
remained there for the entire night and left at 0356
hrs to begin Day 2. She was away from the nest
for 268 min then began an attentive period at 0824
hrs. She laid the second egg during the next
168 min and made three on-off bouts before
leaving the nest until 1945 hrs when night-time
attentiveness was resumed. Behavior at the nest
was much the same on Day 3 except the daytime
attentive period began earlier (0754 hrs) and
lasted longer (366 min). Continuous daytime
attentiveness by the female on Day 4 began at
0936 hrs whereupon she laid her fourth egg, but

left the nest thereafter only for relatively brief
periods of foraging; full time attentiveness, both
day and night had begun. Our method did not
permit us to record time spent on the nest by
females laying their first egg, but it can be
deduced (Fig. 1) that daytime attentiveness in-
creased with laying order. The same pattern of
long periods away from the nest on laying days
(Fig. lA), also occurred in the other two Hermit
Thrushes (Fig. IB, C) except that one female (B)
omitted the lengthy afternoon period of foraging a
day earlier. The American Robin exhibited an
attentiveness pattern during laying that resembled
Hermit Thrushes. She was on the nest full time at
night after the first egg was laid, but did not
continue with an extensive late afternoon absence
after Day 2 or in the early morning after Day 3
(Fig. ID).

Females of both species did not sit on the nest
continuously throughout their daytime periods of
nest attentiveness (hatched bars), but left period-
ically to feed, as indicated by a sawtooth pattern
to Tegg tracings. Frequency of these feeding or
on-off bouts increased as laying proceeded and
stabilized once the birds were in full-time
incubation (Table 1).

Night-time attentiveness ended before sunrise
as indicated in the four mornings shown for each
bird (Fig. lA, B, C, D). It ended 48.3 ± 1 1 .5 min
{x ± SD) before sunrise for the three Hermit
Thrushes, and 27.2 ± 5.2 min for the American
Robin. Night-time attentiveness resumed after
sunset for the five nights shown. The mean was
36.2 ± 8.5 min after sunset for Hermit Thrushes
and 44.2 ± 5.2 min for the American Robin. A
Hermit Thrush, on the day of her first egg, settled
for the night at 1824 hrs, nearly an hour before
sunset (Fig. 1C, Day 1) when a storm with hail
and cold rain occurred.

Females were on their nests during the first
night after laying had begun, but chart recordings
of Tegg indicated they were applying only enough
heat to keep the egg between Ta and the threshold
for development (Fig. 2). There was no daytime
nest attentiveness following egg laying on the first
day. Time spent on the nest increased as laying
proceeded (shown for 2 days later. Fig. 3A) and
females began to exhibit rhythmic on-off bouts
during which time eggs were at times wanned
above PZT. Neglect of eggs still occurred for
considerable periods in daytime. For example.
Day 3 (Fig. 3A) began with the female’s depar-
ture at 0348 hrs (arrow 1 ) and during the

668

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 4. December 2010

Hermit Thrush

0 3 6 9 12 15 18 21 24

Hour of day

American Robin

0 3 6 9 12 15 18 21 24

Hour of day

FIG. I. Representation of nest attentiveness at Tioga Pass, California in four thrushes. All four laid an egg in the
morning hours of Days I through 4. One day following clutch completion (Day 5) is also shown. Black bars indicate when
females were tending their nests at night, white bars when they were absent during daytime, hatched bars when they were
tending their nests in daytime which includes on-off foraging bouts. Time of thermocouple installation occurred on Day 1 at
time X. SR marks time of sunrise, SS time of sunset (PST).

subsequent 170 min, during the coldest part of the
day, Tegg dropped to ambient and followed it
until she returned to the nest (arrow 2) and began
re-warming the eggs. This period of daytime nest

attentiveness continued for 366 min whereupon
the female departed (arrow 3) for 418 min before
coming back on for the night (arrow 4). Thus,
there were two daytime periods off the nest, one

TABLE I . Number and duration of on-off bouts during periods of daytime nest attentiveness by three Hermit Thrushes
and one American Robin. Eggs were laid by both species on Days 1 through 4.

Number of on-off bout.s/day

Duration of on

■off bouts/day (min)

Day

Hermit Thrush

Mean ± SD

American Robin

Total

Hermit Thrush

Mean ± SD

American Robin

Mean ± SD

1

2

F'

3.7 ± 2.9

p

7

23.2 ± 13.7

65.8 ± 52.3

3

7.3 ± 4.0

12

24.6 ± 22.2

37.7 ± 49.6

4

15.3 ± 1.5

17

10.2 ± 5.4

14.1 ± 5.5

5

21.0 ± 1.7

17

10.5 ± 4.2

13.8 ± 4.6

6

19.7 ± 0.6

16

1 1.2 ± 8.3

13.5 ± 5.6

7

21.0 ± 1.7

16

1 1.0 ± 5.3

13.2 ± 2.8

■' Assiimcil from observations of American Robins by Schant/. (iy.f9).

Morton and Percvra • ONSET OF INCUBATION IN THRUSHES

669

Hour of day

Hour of day

FIG. 2. Egg temperatures (Tegg) in two nests (A =
Hermit Thrush; B = American Robin) on the first night
following onset of laying (Day 1). Arrow 1 shows when
night-time attentiveness began, and arrow 2 when it ended.
Concurrent record of air temperatures (Ta) also shown.
Shaded horizontal bars indicate physiological zero
temperature (PZT).

in the morning and one in the afternoon (white
bars, Fig. I ) when the female was, presumably,
mainly engaged in foraging.

An example of this same female’s attentiveness
pattern is shown 4 days later (Fig. 3B) when she
was in full incubation mode. She left the nest in
daytime (beginning at arrow 1 ) only for foraging
bouts and remained on the nest at mid-day for
extended periods when there was exposure to sun.
Eventually, well after sun.set, she began the period
of night-time attentiveness (arrow 2). Tegg
seldom dropped below PZT at this time; embryos
were usually sufficiently warm for growth and
development to occur.

The percentage of total daytime and nighttime
that eggs were heated above PZT increased
gradually, but daytime incubation lagged behind
that of nighttime (Fig 4). Night-time nest occu-
pancy time was 100% from the first egg onward

FIG. 3. Tegg and Ta obtained at a Hermit Thrush nest
on the day the third egg was laid (A) and 4 days later when
regular, full-time incubation was occurring (B). Shaded
horizontal bars indicate PZT. See text for explanation
of arrows.

and daytime occupancy was not (Fig. 1 ). Incuba-
tion onset also occurred more rapidly at night. For
example, during the night that Hermit Thrush A
was on her first egg it was not warmed sufficiently
for development to occur (Fig. 2A). Tegg during
the next night, after the second egg had been laid,
exceeded PZT 70.8% of the time and on the
following night it was 98.7% (Fig. 4A). In
contrast, in the daytime hours during which the
second egg was laid, PZT was exceeded for only
21 ol the 168 min (12.5%) the female was on the
nest, or 2.2% of the total daytime period
(Fig. 4A). PZT was exceeded in the daytime
hours, during which the third egg was laid, for 126
ot the 366 min (34.4%) that a female was on the
ne.st (Fig. 3A), or 13.1% of the total daytime
period (Fig. 4A). Time that eggs were held above
PZT was at or near 100% by the third night for
Hermit Thrushes, but daytime heat tran.sfer to
eggs was not maximized at 92 to 98% until the
day following clutch completion (Fig. 4A, B, C).
The American Robin followed a night-time
pattern similar to that of Hermit Thrushes, but
reached the daytime maximum above PZT (98 to
100%) a day ahead of them (Fig. 4D).

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

670

FIG. 4. Percent of time that Tegg was above PZT during nighttime and during daytime for three Hermit Thrushes (A,
B, C) and an American Robin (D). Egg laying occurred on Days 1 through 4.

Hermit Thrush females were attentive to their
nests about 38% of the time on Day I (1 hr of
daytime for laying and 8 hrs of nighttime in the
first 24 hrs). This increased steadily on each
laying day thereafter as more time became
devoted to the nest during daylight hours
(Fig. 5). In contrast, no incubation occurred on
Day 1, but it began to increase thereafter with
eggs being heated in the beginning mostly at night
(Fig. 4) then more so in daylight, summing
eventually to 80% on Day 4 before reaching a
sustainable maximum of ~95% on Day 5 (Fig. 5).
The American Robin also increa.sed its incubation
time with laying, reaching 98% on Day 4.

Frequency of on-off bouts increa.sed during
laying until Day 5 for Hermit Thrushes and Day 4
for the American Robin (Table I). The slightly
higher percentage of time eggs were held above
PZT by the American Robin occurred because,
unlike Hermit Thrushes (Fig. 3B), its eggs seldom
cooled to below 28° C when it was off feeding
during the cold hours of early morning.

DISCUSSION

Nest-attentive behavior in thrushes differed
greatly with time of day. Nocturnal occupancy
of nests began with the first egg and was fully

FIG. .“i. Percent of time that Hermit Thrush females [n
= 3) were tending their nests (filled circles) and warming
their eggs above PZT (open circles). Egg laying occurred
on Days 1 through 4.

Morton and Pereyra • ONSET OF INCUBATION IN THRUSHES

671

sustained thereafter (Fig. 1). Diurnal attentive-
ness, in eontrast, developed only gradually and
was not complete until after the final egg was laid
(Fig. 1). Daytime episodes of attentiveness were
intemipted by short periods off the nest for
foraging (on-off bouts). These increased in
frequency and decreased in duration as laying
proceeded and became stable at time of clutch
completion, a behavioral marker of entry into full-
time incubation (Table 1 ). Aside from those brief
forays off the nest there were other, much longer
portions of the day, that were presumably devoted
mainly to foraging. These lengthy periods of nest
neglect occupied more than 90% of daytime hours
on the first-egg day (Day 1), decreasing to about
15% on the last-egg day (Day 4) for Hermit
Thrushes and to 0% for the American Robin. We
hypothesize that females required these long
feeding periods to meet the nutritional demands
of ovogenesis and that these demands were not
trivial. For instance, data on egg weights and body
weights of thrushes indicate that clutch mass may
exceed one-third of a non-laying female’s body
mass (Jones and Donovan 1996, Sallabanks and
James 1999). Mountain White-crowned Sparrows
(Zonotrichia leucophrys oriantha) on this same
study area, had a four-egg clutch mass about equal
to 45% of a female’s body mass, and resources for
ovogenesis were accumulated before and during
laying. Body mass increased by about 5% per day
in the 3 days prior to first ovulation and decreased
only slightly each day thereafter with each new
egg. This is about twice the rate of mass gained in
females (and males) engaged in autumnal pre-
migratory fattening in this same population,
indicating that laying females are intensely
hyperphagic (Morton 2002).

A trade-off between daytime nest neglect and
extended periods for foraging by laying thrushes
seems necessary, but what could be the value of
their early nocturnal attentiveness? This question
was examined for Red-winged Blackbirds (Age-
laius phoeniceus) by Clotfelter and Yasukawa
(1999). They concluded early nest attentiveness
did not provide increased egg viability or
protection from predators, but may have reduced
parasitism by Brown-headed Cowbirds (Molo-
thrus ater). Cowbirds were rarely present at Tioga
Pass and none of their eggs was detected in 92
Hermit Thrush and 30 American Robin clutches
(Morton and Pereyra 2004). Our visits to thrush
nests were too infrequent to allow comment on
timing of predation (nocturnal or diurnal), but

exposure to freezing conditions at night or early
morning, as well as .severe storms, was known to
cause mortality in eggs and nestlings of other
species (Morton 2002). Nocturnal attentiveness
might preserve egg viability, at least under high
altitude conditions. We found on our study area,
however, that interspecific variation occurred.
Nocturnal attentiveness by four species (Rock
Wrens [Salpinctes ohsoletus]. Song Sparrows
[Melospiza melodia]. Hermit Thrushes, and
American Robins) began early in the laying
sequence (Oppenheimer and Morton 2000,
Morton 2002, this study), whereas in two others
(Dusky Flycatchers [Empidonax oberholseri].
White-crowned Sparrows) it was delayed until
after the penultimate egg was laid (Zerba and
Morton 1983, Morton and Pereyra 1985). Early
onset of nocturnal attentiveness occurs in a wide
range of passerines, including many cavity nesters
(Wang and Weathers 2009), but its adaptive
significance seems unresolved.

A key finding of our study was that the pattern
of heat transfer to eggs sufficient for embryonic
development differed from that of nest attentive-
ness. For example, all four thrushes began night-
time sitting after the first egg (Fig. 1), but none
supplied sufficient warmth for incubation to begin
until after the second egg (Figs. 2, 4). Hermit
Thrushes were attentive for more than half of the
daylight hours by Day 3 (Fig. 4A, B, C), but PZT
was usually exceeded only briefly during periods
that the female was on the nest (Fig. 3A). Mean
fraction of daytime where Tegg was sufficiently
high for embryonic development on Day 3 was
only 23.7 ± 9.5% for Hermit Thrushes and 38. 1 %
for the American Robin (Fig. 4). Females were
developing the behavioral rhythm of incubation
but full contact of eggs with the incubation patch
in daytime was withheld because, during the
night-time portion of Day 3, PZT was exceeded
by all four thrushes at close to 100% of the time
(Fig. 4). This indicates the Day 3 patch was
functionally developed but incubation behavior
was not with daytime incubation lagging behind
nighttime.

Total nest attentiveness behavior and of incu-
bation (daytime plus nighttime) for the 4 days of
laying and for 3 days thereafter increased
gradually during laying, but with different trajec-
tories (Fig. 5). Hermit Thrushes foraged for at
least 2 hrs in early morning before returning to lay
their final egg (Fig. 1), and then, immediately
thereafter, began full-time incubation. Apparently,

672

THE WILSON JOURNAL OF ORNITHOLOGY • V'oL 122, No. 4, December 2010

the American Robin dealt more easily with
nutritional demands because her prolonged early
morning feeding periods ended on Day 3
(Fig. ID), and incubation behavior was fully
developed by Day 4 (Fig. 4D).

Incubation onset in European Starlings {Stunnis
vulgaris) appeared to be pre-programmed (Meijer
1990), a conclusion supported by our study. For
example, day-night timing of heat transfer to eggs
during the laying period was much the same for
all three Hermit Thrushes and their close relative,
the American Robin (Fig. 4). Incubation devel-
oped gradually during the laying period in all four
birds with full behavioral and physiological
capacities converging at the time of clutch
completion (Fig. 5). These results agree with
those obtained for other species of passerines
inhabiting different environments (Wang and
Weathers 2009), and with early field observations
of Hermit Thrushes (Stanwood 1910) and Amer-
ican Robins (Howell 1942). We suggest this
consistency is due to strongly expressed neuroen-
docrine and endocrine influences (Goldsmith
1982, Sharp 1997, Shaip et al. 1998, Sharp and
Blache 2003, Vleck 1998) that have been tuned
by natural selection to entrain incubation beha-
viors that optimize embryonic development and
survival.

Results quite different from ours have been
published. For example, incubation onset by
European Pied Flycatchers (Ficechda hypoleuca)
occurred at any time within a 4-day span, from the
day of clutch completion to 3 days prior (Potti
1998). Even more plasticity was reported in a
recent, large study of American Robins wherein
females were presumed to begin incubating at any
time within a 7-day span, from 4 days after clutch
completion to 2 days prior (Rowe and Weather-
head 2009). Accurate information on incubation
onset in both of these studies was crucial to the
hypotheses presented, yet this key event was
defined as the date when eggs first felt warm to
the touch. Obviously, according to this criterion,
birds could be classified as “incubating” when
eggs were under partial incubation, or full-time
incubation, or even no incubation if Tegg was
below PZT. We hope that more complete
information, as provided by egg temperatures,
will be obtained before data from studies are
incorporated into current paradigms of timing and
control of incubation onset.

Studies of incubation have frequently suffered
from inconsistencies in terminology and effec-

tiveness of methods, and there has been a failure
of investigators to use knowledge about underly-
ing mechanisms. This lack of physiological
insight is unfortunate because it can be an
impediment to understanding ecological effects
(Denny and Helmuth 2009) and even the proxi-
mate causation of behavioral phenomena (Drick-
amer 1998); avian reproduction has model-system
potential for collaborations between evolutionary
ecologists and endocrinologists (Wingfield et al.
2008). We believe the adaptive value of early
incubation onset probably does not lie in hatching
asynchrony (Mead and Morton 1985, Morton
2002). It has been shown, for example, that partial
incubation before clutch completion can help
maintain egg viability (Veiga 1992, Stoleson and
Beissinger 1999), and reduce the magnitude of
microbial trans-shell infections (Cook et al. 2003).

ACKNOWLEDGMENTS

This research was done with the permission of the
LSDA, Forest Service and the California Department of
Fish and Game.

LITERATURE CITED

Clark, A. B. and D. S. Wilson. 1985. The onset of
incubation in birds. American Naturalist 125:603-61 1.
Clotfelter, E. D. and K. Yasukawa. 1999. The function
of early onset of nocturnal incubation in Red-winged
Blackbirds. Auk 116:417^26.

Cook, M. I.. S. R. Beissinger, G. A. Toranzos, R. A.
Rodriguez, and W. J. Arendt. 2003. Trans-shell
infection by pathogenic micro-organisms reduces the
shelf life of non-incubated bird's eggs: a constraint on
the onset of incubation? Proceedings of the Royal
Society of London, Series B 270:2233-2240.

Dawson, A. 2008. Control of the annual cycle in birds:
endocrine constraints and plasticity in response to
ecological variability. Philosophical Transactions of
the Royal Society of London, Series B 363:1621-
1633.

Denny, M. and B. FIelmuth. 2009. Confronting the
physiological bottleneck: a challenge from ecomecha-
nics. Integrative and Comparative Biology 49:197-
201.

Drickamhr, L. C. 1998. Vertebrate behavior: integration of
proximate and ultimate causation. American Zoologist
38:.39-42.

Goldsmith, A. R. 1982. Plasma concentrations of prolactin
during incubation and parental feeding throughout
repeated breeding cycles in Canaries (Seriniis canar-
ii(.s). journal of Endocrinology 94:51-59.

Haflorn, S. 1978. Egg-laying and regulation of egg
temperature during incubation in the Goldcrest
Regiilus regains. Ornis Scandinavica 9:2-21.
Hafi'orn, S. 1981. Incubation during the egg-laying period
in relation to clutch-size and other aspects of

Morton and Pcrcvra • ONSET OF INCUBATION IN THRUSHES

673

reproduclion in the Great Tit Parns major. Ornis
Scandanavica 12:1 69- 1 85.

Hafforn, S. 1988. Incubating female passerines do not let
the egg temperature fall below the ‘physiological zero
temperature' during their absences from the nest. Ornis
Scandanavica 19:97-110.

Hahn. T. P., J. C. Wingfield, R. Mullen, and P. J.
Deviche. 1995. Endocrine bases of spatial and
temporal opportunism in Arctic-breeding birds. Amer-
ican Zoologist 35:259-273.

Hahn, T. P.. K, W. Sockm.an, C. W. Breuner, and M. L.
Morton. 2004. Facultative altitudinal movements by
Mountain White-crowned Sparrows {Zonotrichia leii-
coplnys orianthu) in the Sierra Nevada. Auk 121:
1269-1281.

Howell, J. C. 1942. Notes on the nesting habits of the
American Robin (Tiirdiis migratioriiis L.). American
Midland Naturalist 28:529-603.

Jones, P. W. and T. M. Donovan. 1996. Hermit Thrush
{Catharu.s guttatus). The birds of North America.
Number 261.

Mead, P. S. and M. L. Morton. 1985. Hatching
asynchrony in the Mountain White-crowned Sparrow
(Zonotrichia leucophrys oriantha): a selected or
incidental trait? Auk 102:781-792.

Meijer, T. 1990. Incubation development and clutch size in
the starling. Ornis Scandanavica 21:163-168.

Morton, M. L. 1978. Snow conditions and the onset of
breeding in the Mountain White-crowned Sparrow.
Condor 80:285-289.

Morton, M. L. 2002. The Mountain White-crowned
Sparrow: migration and reproduction at high altitude.
Studies in Avian Biology 24:1-236.

Morton, M. L. and M. E. Pereyra. 1985. The regulation
of egg temperatures and attentiveness patterns in the
Dusky Flycatcher (Empidona.x oherhol.xeri). Auk 102:
25-37.

Morton, M. L. and M. E. Pereyra. 2004. Clutch sizes and
nesting habits of birds at Tioga Pass. Western Birds
35:62-70.

Oppenheimer. S. D. and M. L. Morton. 2000. Nesting
habits and incubation behavior of the Rock Wren.
Journal of Field Ornithology 71:650-657.

Pom, J. 1998. Variation in the onset of incubation in the
Pied Flycatcher (Ficedida hypolenca): fitness conse-
quences and constraints. Journal of Zoology, London
245:335-344.

Rowe, K. M. C. and P. J. Weatherhead. 2009. A third
incubation tactic: delayed incubation by American
Robins (Tnrdns migratoriii.s). Auk 126:141-146.

Sallabanks. R. and F. C. James. 1999. American Robin

(Tnrdns migratorins). The birds of North America.
Number 462.

SCHANTZ, W. E. 1939. A detailed study of a family of
robins. Wilson Bulletin 51:157-169.

Sharp, P. J. 1997. Neurobiology of the on.set of incubation
behaviour in birds. Pages 193-202 in Frontiers in
environmental and metabolic endocrinology (S. K.
Maitra, Editor). University of Burdwan. Burdwan,
India.

Sharp, P. J. and D. Blache. 2003. A neuroendocrine
model for prolactin as the key mediator of seasonal
breeding in birds under long- and short-day photope-
riods. Canadian Journal of Physiology and Pharma-
cology 81:350-358.

Sharp, P. J., A. Dawson, and R. W. Lea. 1998. Control of
luteinizing hormone and prolactin secretion in birds.
Comparative Biochemistry and Physiology Part C 1 19:
275-282.

Stanwood, C. j. 1910. The Hermit Thrush; the voice of the
northern woods. Bird-Lore 12:100-103.

Stoleson, S. H. and S. R. Beissinger. 1995. Hatching
asynchrony and the onset of incubation in birds,
revisited. CuiTent Ornithology 12:191-270.

Stoleson, S. H. and S. R. Beissinger. 1999. Egg viability
as a constraint on hatching synchrony at high ambient
temperature. Journal of Animal Ecology 68:951-962.

Veiga, j. P. 1992. Hatching asynchrony in the House
Sparrow: a test of the egg-viability hypothesis.
American Naturalist 139:669-675.

Vleck, C. M. 1998. Hormonal control of incubation/
brooding behavior: lessons from wild birds. Proceed-
ings of the WPSA European Poultry Conference. Israel
10:163-169.

Wang, J. M. and W. W. Weathers. 2009. Egg laying, egg
temperature, attentiveness, and incubation in the
Western Bluebird. Wilson Journal of Ornithology
121:512-520.

White, F. N. and J. L. Kinney. 1974. Avian incubation.
Science 186:107-1 15.

Wingfield, J. C. 1984. Environmental and endocrine
control of reproduction in the Song Sparrow. Melos-
piza melodia. General and Comparative Endocrinolo-
gy 56:406^16.

Wingfield, J. C., M. E. Visser, and T. D. Williams.
2008. Introduction. Integration of ecology and endo-
crinology in avian reproduction: a new synthesis.
Philosophical Transactions of the Royal Society of
London, Series B 363:1581-1588.

Zerba, E. and M. L. Morton. 1983. The rhythm of
incubation from egg laying to hatching in Mountain
White-crowned Sparrows. Ornis Scandanavica 14:
188-197.

The Wilson Journal oj Ornithology 122(4):674-680, 2010

FLEXIBILITY IN NEST-SITE CHOICE AND NESTING SUCCESS OF
TURDUS RUFIVENTRIS (TURDIDAE) IN A MONTANE FOREST IN

NORTHWESTERN ARGENTINA

SILVIA B. LOMASCOLO,' " A. CAROLINA MONMANY,--^

AGUSTINA MALIZIA,^ AND THOMAS E. MARTIN^

ABSTRACT. — We studied the consequences of nest-site choice on nesting success under differing disturbance levels for
the Rufous-bellied Thrush (Turdus rufiventris). We compared nest-site choice and nest success between a disturbed site and
an undisturbed site in a montane subtropical forest in northwestern Argentina. We found no overall difference in daily
predation rate (DPR) between the disturbed and undisturbed sites. However, DPR of nests on bromeliads was significantly
lower at the microhabitat level than on other types of subtrates at the disturbed site. T. rufiventris used bromeliads for
nesting more often than expected by chance at the disturbed site. DPR did not differ between substrates at the undisturbed
site and T. rufiventris used all substrates according to their availability. Nests had higher predation at the disturbed site
when DPR on non-bromeliad substrates was compared between disturbed and undisturbed sites. Nest fate was independent
of nest height. Our results suggest T. rufiventris' flexibility in nest-site choice, as reflected by increased use of the safest
sites, i.e., bromeliads, in the disturbed site compared to the undisturbed site, may allow this species to survive in an
otherwise much riskier habitat. Our results illustrate how microhabitat-scale effects can mediate landscape scale effects.
Received 24 October 2009. Accepted 7 April 2010.

Understanding habitat influences on nesting
success of birds may be key to their successful
conservation, given the sensitivity of this life
stage to habitat disturbance (Martin 1992, Easton
and Martin 2002). Nest success is influenced by
nest-site choice in large part because nest-site
characteristics can influence nest predation rates
(Martin and Roper 1988, Martin 1993, Holt and
Martin 1997, Martin 1998, De Santo et al. 2002,
Easton and Martin 2002, Mezquida and Marone
2002, Kellett et al. 2003, Fontaine et al. 2007).
Nest predation may be one of the main agents of
natural selection influencing evolution of life
history traits and nest-site choice (reviewed by
Lima 2009, Martin and Briskie 2009). Flexibility
in choosing a nest site may allow birds to optimize
fitness by exploiting safer substrate types in
disturbed conditions, but studies of nesting
flexibility and their consequences for nest success
are rare.

Nest-site choice, including nesting substrate
and nesting height, may be evolutionarily conser-

' lADIZA-CCT MENDOZA. CONICET, Avenida Ruiz
Leal .s/n, CC 507, 5500 Mendoza, Argentina.

^ Univer.sidad de Puerto Rico, Departamento de Biologi'a,
CN 235, P. O. Box 70377, San Juan, PR 00936, USA.

’LIEY-IER, Univer.sidad Nacional de Tucuman, Argen-
tina, CC 34, Yerba Buena, 4107 Tucuman, Argentina.

‘‘uses, Montana Cooperative Wildlife Research Unit,
Avian Studies Program, 205 Natural Science, University of
Montana. Missoula, MT 59812, USA.

^Corresponding author; e-mail: .slomascolo(§>gmail.com

vative in many species (Martin 1988, Martin and
Roper 1988), which may constrain plasticity of
choices. Some bird species use the same nest sites
in disturbed and undisturbed forest patches, even
though this increases predation in disturbed forest
patches (Holt and Martin 1997, Easton and Martin
2002). Plastic changes in nest-site selection in
response to predation risk have been observed in
some species (e.g., Marzluff 1988, Eggers et al.
2006, Peluc et al. 2008). Variation in predation
risk is often associated with disturbance of
habitats, which also generally reflects differing
habitat structure; both differing predation risk and
habitat structure could influence nest-site choice
(Martin 1992). Few studies have compared nest
site choices under differing predation risk or
differing habitat structure.

Plasticity of nest-site choice is particularly
interesting for tropical and subtropical birds,
because they are often thought to have more
specialized niches (MacArthur 1972). Conse-
quently, their flexibility of nest-site choice to
varying predation risk or habitat structure may be
constrained. No study has examined plasticity of
nest-site choice by a tropical or subtropical bird.
Thus, examination of flexibility of nest-site
choice and consequences of that choice on nesting
success under differing disturbance levels for
tropical or subtropical birds is needed.

Our objective was to test whether nest-site
choice and nesting success of the Rufous-bellied
Thrush {Turdus rufiventris', Turdidae) differed

674

Lomciscolo cl al. • RUFOUS-BELLIED THRUSH IN MONTANE l-ORESI'S

675

between a disturbed site and an undisturbed site in
a montane subtropical forest in northwestern
Argentina. Tiirdiis rufiventris is a common bird
that consumes fruits of many species in this
environment (Malizia 2001 ), potentially having an
important role in regeneration of native plants. It
is widely distributed in central and northern
Argentina (Ridgely and Tudor 1989, de la Pena
and Rumboll 1998), and is one of the few native
forest species that inhabits semi-urban locations
(de la Pena and Rumboll 1998, Ferretti et al.
2005). The ability of members of the genus
Tiirdus, including T. rufiventris (e.g., de la Pena
and Rumboll 1998, Ferretti et al. 2005), to live in
human-disturbed areas makes them good model
species to study how birds adapt to habitat
disturbance. Understanding the adaptability of
species to disturbance is becoming increasingly
important as anthropogenic habitat alterations
become increasingly common and intact habitats
increasingly rare (Pimm et al. 2001). Examining
the effect of habitat characteristics in nest-site
choice flexibility and nesting success is important
in a threatened but understudied environment such
as the subtropics (de la Pena 1979, Mezquida and
Marone 2002). We examined nest-site preferences
and consequences for nesting success of T.
rufiventris between undisturbed and disturbed
forest sites in northwestern Argentina.

METHODS

The study was conducted between October
1997 and January 1998 at El Rey National Park
(hereafter El Rey; 24°42'S, 64°38'W) in Salta
Province, and at Sierra de San Javier Biological
Park (hereafter San Javier; 24°47'S, 65°22'W)
in Tucuman Province. Both sites are in north-
western Argentina and are part of the Yungas
ecosystem represented by a subtropical montane
forest (Brown 1995, Brown et al. 2001). An old
cattle farm, El Rey was declared a national park in
1948 and has been pre.served intact with no human
activity since then. It encompasses 44,162 ha, is
200 km from the nearest city, and no significant
logging has been conducted in the area; it
represents the undisturbed site. The canopy is
dominated by Cinnamonuun porphyrium (Laur-
aceae), Blepharocalix salicifolius (Myrtaceae),
Cedrella lilloi, and C. angustifoUa (Meliaceae).
A second stratum is comprised of species
typically <20 m high, including Allophyllus
edulis (Sapindaceae), Zanthoxylum coco (Ruta-
ceae), and Primus tucunianensis (Rosaceae)

(Blake and Rouges 1997). San Javier was created
by the National University of Tucuman in 1973
and covers 14,100 ha; it is 15 km west of the city
of San Miguel de Tucuman and 2 km northwe.st of
the city of Yerba Buena. The area has been
subject to different land u.ses including .selective
logging focused on the most valuable timber
species (Cedrela lilloi) affecting a large propor-
tion of the sieiTa, modern agriculture (sugar cane,
citrus, horticulture, floriculture), and expanding
urbanization (Grau et al. 2008). Trekking and
biking trails cross the San Javier site. The
secondary forest where we sampled occurs near
agricultural and urban sectors (Grau et al. 1997,
Aragon and Morales 2003), and represents our
disturbed site. Common canopy species at San
Javier are Parapiptadenia exelsa (Fabaceae),
Cinnamomurn porphyrium (Lauracae), Juglans
australis (Juglandaceae), and Myrsine laetevirens
(Myrsinaceae) while the subcanopy is dominated
by Piper tucumanum (Piperaceae), Allophylus
edulis (Sapindaceae), and Psychotria cartagenen-
sis (Rubiaceae) (Grau et al. in press). The native
pioneers Heliocarpus popayanensis (Malvaceae),
Tecoma stans (Bignoniaceae), Solanum ripariwn
(Solanaceae), and exotic colonizers including
Morus spp. (Moraceae), Ligustrum luciduni
(Oleaceae), and Citrus spp. (Rutaceae) (Grau
and Aragon 2000) are also common at San Javier.
Bromeliads are one of the main groups represent-
ed among the abundant epiphytes both at San
Javier and at El Rey (Blake and Rouges 1997).

We located nests following Martin and Geupel
(1993). This method involves detecting and
following birds canying food or nesting material
to the nest, or following female calls. Nest
searching was done daily from 0630 to 1245 hrs
from October to December. Nests were checked
every 2-4 days to record nest stage (e.g., building,
incubating, nestlings), number of eggs or nestlings
when nests were accessible, nesting substrate, and
nest height (Martin and Geupel 1993). The nest
was considered failed or fledged if, after three
checks, no bird activity was recorded at the nest
(the last check had to be at least 30 min),
depending on the stage recorded at the last active
visit. We were unable to assign the cause of
failure of many nests due to inaccessibility, but
fledging was confirmed whenever possible by
looking for parents with food or fledglings near
the nest shortly after the assumed fledge date. We
calculated nest success following Mayfield (1961 )
using the approach outlined by Hensler and

676

THE WILSON JOURNAL OL ORNITHOLOGY • Voi 122. No. 4. December 2010

Nichols (1981). We compared daily predation
rates between sites and between substrates using a
Chi-square test based on program CONTRAST
(Hines and Sauer 1989).

The two most common substrates used at San
Javier, bromeliads and Psychotria ccirtagenensis,
seemed to differ substantially in mean height, and
the differences in daily predation rate found
between the substrates could be attributed to
substrate height and not to type of substrate per
se. Thus, we tested whether the difference in
substrate height was statistically significant using
a Mann-Whitney f/-test, and whether predation
was independent of substrate height in San Javier
using a Chi-square test.

We measured substrate availability based on
the observed substrates used by T. rufiventris to
examine if nests were randomly placed. We
followed the quarter method (Matteuci and Colma
1982), modified to include bromeliad availability.
We established two 200 m-long transects at
random in the study area. The starting point of
the transect was at the end of a random number of
steps chosen by a person not familiar with the
study site. A second random number indicated the
orientation of the transect as measured by a
compass. We established four quadrants every
10 m along each of these transects. We projected
an imaginary line 17-m high (the maximum nest
height recorded during this study) from the center
point, and recorded the plant species closest to the
imaginary line in each quadrant. We recorded
when a bromeliad was the closest substrate, but
not the substrate supporting the bromeliad. Only
plants with >1.5 cm diameter at breast height
(dbh) were recorded because no nest was found on
plants with a smaller dbh. At least 160 plants were
counted along each tran.sect. We analyzed these
data using a Z-test for comparing two proportions
(Zar 1999) to examine if nest site was chosen
according to plant availability.

RESULTS

We found 44 nests of Rufous-bellied Thrushes
on 12 different substrates (Fig. I A) at San Javier
of which only 27 had at least one egg laid. Only
tho.se 27 nests were u.sed to calculate nest success.
All 44 nests at San Javier were included in the
substrate use analyses. Sixty-five nests with at
least one egg laid were found at El Rey on 19
different substrates (Table I). The.se nests were
u.sed for estimating nest success and substrate u.se.

Nests at San Javier were most common on

Psychotria carthagenensis and bromeliads. How-
ever, bromeliads were used in higher proportion
than expected according to their availability at
San Javier while the other substrates (including P.
carthagenensis) were used in lower proportion
than expected (Z = 10.8, df = 2, P < 0.0001;
Table 2). Substrates at El Rey were used as
expected by their availability (Z = 0.66, df = 2, P
= 0.26; Table 2).

Overall daily predation rate (DPR ± SE) at San
Javier (0.0684 ± 0.0095, n = 27, exposure days =
248.5) did not differ from El Rey (0.0592 ±
0.0089, n = 65, exposure days = 709; X- = 0.669,
df = \, P = 0.414). Nest success on P.
carthagenensis, the most common understory
shrub at San Javier, was not different from success
on all substrates other than bromeliads (X“ = 0.5,
df = P = 0.48) and substrates were combined
for further analysis. Nests on bromeliads at San
Javier were more successful (DPR = 0.0308 ±
0.0175 nests/day, n = 7, exposure days = 97.5)
than nests on other substrates (DPR = 0.1126 ±
0.0257 nests/day, n — 20, exposure days = 151)
(X^ — 6.92, df = 1, P = 0.0085). Success of nests
on bromeliads at El Rey (DPR = 0.0522 ± 0.0207
nests/day, n — 9, exposure days = 115) was not
significantly different from success on other
substrates (DPR = 0.0606 ± 0.0098, n = 56,
exposure days = 594) (X" = 0. 17, df = P =
0.68). We believe that lack of difference between
substrates in DPR at El Rey was not affected by
low sample size given the significant difference
found for San Javier, where sample size was even
smaller. Bromeliads were used more often at San
Javier than at El Rey, and the high nest success on
bromeliads may increase overall nest success at
San Javier. Thus, we compared nest success
between San Javier and El Rey without considering
nests on bromeliads. Non-bromeliad nests experi-
enced higher daily predation rates at San Javier
than at El Rey (X" = 4.076; df = 1 ; P = 0.044).

Nests built on P. carthagenensis at San Javier
were significantly lower (.T = 2.1 m) than those
on bomeliads (x = 8.9 m) (Mann-Whitney U =
10, P = 0.0004). However, nest success was
independent of nest height (X" = 0.50, df = 5,' P
= 0.48). Nests occurred between 1 and 17 m
above ground with a bimodal distribution
(Fig. I A). Most nests were between 1 and 2 m
above the ground with a second peak of nests at
10 m. The two most common substrates mostly
explained the distribution of nests at different
heights (Fig. IB).

LomascoU) cl al. • RUFOUS-BELLIED THRUSH IN MONTANE I-ORESTS

677

Nest height (m)

14 n

Lisi Pitu Psca Aled Mosp Lilu Vine BIsa Citrus Myla Brom Cipo

(1,5) (20) (2.1) (2.6) (2.8) (4.3) (5.0) (6.5) (8.0) (8.0) (8.9) (12.5)

Substrate

(average nest height in m)

FIG. 1. (A) Distribution of Tardus rufivenlris nest heights. (B) Number of Tardus rafiventris nests found per substrate

species (n = 42). Numbers in parentheses show mean height at which nests were built on each substrate. Lisi: Ligastram
sinense; Pitu: Piper tucumanuiv, Psca: Psychotria carthagenensis-, Aled: Allophylus edalis\ Mosp: Moras spp., Lilu:
Ligastram lacidanv. Vine: unidentified vine; Blsa: Blepharoccdyx salicifolias\ Citrus: Citrus spp.; Myla: Myrsine
laetevireas', Brom: bromeliads; Cipo: Cinnamonuiai porphyriam.

678

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, December 2010

TABLE 1. Substrates used by Tiirdus rufiventris at El
Rey National Park, Salta, Argentina.

Substrate species

# Nests

Family

Acacia aroma

3

Fabaceae

A. visco

1

Fabaceae

Allophyhis edulis

21

Sapindaceae

Celtis spinosa

1

Celtidaceae

Condalia buxifolia
Enterolobium

3

Rhamnaceae

contortisiliquum

1

Fabaceae

Eugenia uniflora

1

Myrtaceae

Gledit.sia amorphoides

2

Fabaceae

Nectandra pichurim

1

Lauraceae

Pogonopu.^ tuhulosus

1

Rubiaceae

Samhuciis peruviana

2

Adoxaceae

Scutia buxifolia

5

Rhamnaceae

Side roxy Ion ohtusifolium

1

Sapotaceae

Urera caracasana

2

Urticaceae

Vassohia hreviflora

1

Solanaceae

Xylo.sma puhescens

1

Salicaceae

Unidentified bromeliad

9

Bromeliaceae

Unidentified fern

1

Unidentified moss

2

Unidentified Myrtaceae

2

Myrtaceae

Unidentified vine

4

Total

65

DISCUSSION

The high variability of substrate type and nest
height u,sed by Rufous-bellied Thrushes is con-
trary to that expected if nesting was conserved
within a species (Knight and Fitzner 1985;
Dhindsa et al. 1988; Martin 1988, 1993; but .see
Forstmeier and Weiss 2004, Eggers et al. 2006,
Peluc et al. 2008). This species nested on at least
12 substrate types, including understory shrubs,
bromeliad epiphytes, canopy trees, and exotic as
well as native species. Turdiis rufiventris al.so
placed nests at a wide range of heights from 1 .5 to
17 m.

Rufous-bellied Thrushes did not seem to
specialize on one substrate type, but were
selective of nesting substrate, particularly at the
site where microhabitat selectivity was associated
with potential fitness benefits. Turdus rufiventris
favored bromeliads, the substrate associated with
highest nest success at San Javier, in accordance
with the expectation that birds should maximize
nest success (Martin 1998). In contrast, this
species was not selective with respect to substrate
at El Rey, where predation rates did not differ
between nest sites.

Our results on nest-site choice and nest success
in relation to disturbance differed depending on
the scale at which we analyzed our data. Our
results at a large scale did not support the
hypothesis that anthropogenic disturbance in-
creases the threat of nest predation (i.e., overall
DPR was similar for San Javier vs. El Rey) (Holt
and Martin 1997, De Santo et al. 2002, Easton and
Martin 2002, Kellett et al. 2003). However, when
examined at the microhabitat scale, we found that
predation rate was much higher at the disturbed
than at the undisturbed site. The higher predation
rate at the disturbed site on non-bromeliad nest
sites suggests predation rates were higher in the
disturbed habitat but birds compensated through
nest-site choice. These results confirm the impor-
tance of examining ecological patterns at multiple
.scales (Holling 1992, Levin 1992); if we had
ignored microhabitat scale mechanisms, large
scale disturbance effects would have been misin-
terpreted.

What makes bromeliads a safer nesting site at
San Javier, and why is this not the case at El Rey?
We showed that substrate height is not the
determining factor; thus, bromeliads must be
influencing other aspects of nest-site structure
that affect predation risk. We believe bromeliad

TABLE 2. Nest success and number of nests of Turdus rufiventris (Turdidae) on bromeliads and other substrates, and
substrate availability at El Rey National Park, Salta, and Sierra de San Javier Biological Park, Tucuinan, Argentina.

El Rey National Park

.Sierra dc San Javier Biological Ptirk

Overall DPR

0.0592 ± 0.0089

0.0684 ± 0.0095

DPR on bromeliads

0.0522 ± 0.0207

0.0308 ± 0.0175

DPR on other substrates

0.0606 ± 0.0098

0. 1 1 26 ± 0.0257

Number of nests on bromeliads (% of total nests)

9 (14%)

7 (26%)

Number of nests on other substrates (% of total nests)

56 (86%)

20 (74%)

Substrate availability

Bromeliads (%)

7

3

Others (%)

93

97

Lomdscolo Cl al. • RUFOUS-BELLIED THRUSH IN MONTANE E’ORESTS

679

structure may make nests less visible and less
vulnerable to predation by white-eared opossums
(Didelpliis al biventer) and black rats {Rattiis
rottus), which are common around houses in
disturbed areas, but not to predation by House
Wrens (Troglodytes aedon). Plush-crested Jays
(Cyanocorax chrysops), and brown capuchin
monkeys (Cehiis apella), common nest predators
in the undisturbed site (Auer et al. 2007). In
addition, the relevant structural feature may be
one that only influences predation risk in
disturbed areas. Predator communities most likely
differ between disturbed and undisturbed sites
(Auer et al. 2007; J. P. Jayat, pers. comm.); thus,
the relevant structural feature must be one that
conceals nests from predators that occur in
disturbed sites, either exclusively or at least in
higher numbers.

Our results suggest a potential mechanism by
which birds could adapt to the negative effects of
human disturbance on nest predation risk. Turdus
rufiventris apparently adjusts its nest-site choices
to compensate for elevated predation risk in
disturbed habitats. Additional studies of other
species that compare both microhabitat choices
and microhabitat-predation relationships in dis-
turbed versus undisturbed areas are necessary to
establish the generality of this mechanism. Studies
of a range of species that vary in adaptability to
human disturbance would be particularly helpful.
Our ability to make general conclusions from this
study is somewhat limited given the inclusion of
only one disturbed site and one undisturbed site.
Additional studies incorporating more replication
would be valuable.

ACKNOWLEDGMENTS

SL was supported by an undergraduate student fellow-
ship from Secretan'a de Ciencia y Tecnica from Universidad
Nacional de Tucuman. Mitchell Aide made valuable
comments on the manuscript. J. J. Lomascolo, T. A. and
V. A. Monmany, and A. M. Garzia helped with transpor-
tation and general support. I. M. Paz Posse helped with
fieldwork.

LITERATURE CITED

Aragon, R. and J. M. Morales. 2003. Species compo-
sition and invasion in NW Argentina secondary
forests: effects of land use history, environment and
landscape. Journal of Vegetation Science 14:195-204.
Auer, S. K., R. D. Bassar, J. J. Fontaine, and T. E.
Martin. 2007. Breeding biology of passerines in a
subtropical montane forest in northwestern Argentina.
Condor 109:321-333.

Blake, J. G. and M. Rouges. 1997. Variation in capture
rates of understory birds in El Rey National Park,
northwestern Argentina. Ornitologia Neotropical
8:185-193.

Brown, A. D. 1995. Las selvas de montana del noroeste de
Argentina: problemas ambientales e importancia de su
conservacion. Pages 9-18 in Investigacion, conserva-
cion y desarrollo en selvas subtropicales de montana
(A. D. Brown and H. R. Grau, Editors). Universidad
Nacional de Tucuman, Argentina.

Brown, A. D., H. R. Grau, L. R. Malizia, and A. Grau.
2001. Argentina. Pages 623-659 in Bosques nublados
del Neotropico (M. Kappelle and A. D. Brown,
Editors). INBio, San Jose, Costa Rica.

de la Pena, M. R. 1979, Enciclopedia de las aves
argentinas. Volume 7. Libren'a y Editorial Colmegna
S. A., Santa Fe, Argentina.

de la Pena, M. R. and M. Rumboll. 1998. Birds of
southern South America and Antarctica. Princeton
University Press, Princeton, New Jersey, USA.

De Santo, T. L., M. F. Willson, K. E. Sieving, and J. J.
Armesto. 2002. Nesting biology of tapaculos (Rhi-
nocryptidae) in fragmented south-temperate rainforests
of Chile. Condor 104:482-^95.

Dhindsa, M. S., P. E. Komers, and D. A. Boag. 1988.
Nest height of Black-billed Magpies: is it determined
by human disturbance or habitat type? Canadian
Journal of Zoology 67:228-232.

Easton, W. E. and K. Martin. 2002. Effects of thinning
and herbicide treatments on nest-site selection by
songbirds in young, managed forests. Auk 119:685-
694.

Eggers, S., M. Griesser, M. Nystrand, and j. Ekman.
2006. Predation risk induces changes in nest-site
selection and clutch size in the Siberian Jay.
Proceedings of the Royal Society of London, Series
B 273:701-706.

Ferretti, V., P. E. Llambias, and T. E. Martin. 2005.
Life history differences between populations of a
neotropical thrush challenge food limitation theory.
Proceedings of the Royal Society of London. Series B
272:769-773.

Fontaine, J. J., M. Martel, H. M. Markland, A.
Niklison, K. Decker, and T. E. Martin. 2007.
Testing ecological and behavioral correlates of nest
predation. Oikos 116:1887-1894.

Forstmeier, W. and I. Weiss. 2004. Adaptive plasticity in
nest-site selection in respon.se to changing predation
risk. Oikos 104:487-499.

Grau, H. R. and R. AragUn. 2000. Arboles invasores de la
Siena de San Javier, Tucuman, Argentina. Pages 5-20
in Ecologi'a de arboles exoticos en las Yungas (H. R.
Grau and R. Aragon. Editors). LIEY-Universidad
Nacional de Tucuman, Argentina.

Grau, H. R., M, F. Arturi, A. D. Brown, and P. G.
Acenolaza. 1997. Floristic and structural patterns
along a chronosequence of secondary forest succession
in Argentinean subtropical montane forests. Forest
Ecology and Management 95:161-171.

Grau, H. R., L. Paolini, A. Malizia, and J. Carilla. In
press. Distribucion, estructura y dinamica de los

680

THE WILSON JOURNAL OF ORNITHOLOGY • Voi 122, No. 4, December 2010

bosques de la Sierra de San Javier. Pages 000-000 in
E.stado Ambiental de la Sien'a de San Javier (M. E.
Hernandez, Editor). Universidad Nacional de Tucu-
man, Argentina.

Grau, H. R., M. E. Hernandez., J. Gutierrez, N. I.
Gasparri, C. Casaveccia, E. E. Flore, s-1valdi, and
L. Paolini. 2008. A peri-urban neotropical forest
transition and its consequences for environmental
services. Ecology and Society 13:35.

Hensler, G. L. and J. D. Nichols. 1981. The Mayfield
method of estimating nesting success — a model,
estimators and simulation results. Wilson Bulletin
93:42-53.

Hines, J. E. and J. R. Sauer. 1989. Program CONTRAST:
a general program for the analysis of several survival
or recovery rate estimates. Fish and Wildlife Technical
Report 24. USDI, Fish and Wildlife Service, Wash-
ington, D.C., USA.

Holling, C. S. 1992. Cross-scale morphology, geometry
and dynamics of ecosystems. Ecological Monographs
62:447-502.

Holt, R. F. and K. M.artin. 1997. Landscape modification
and patch selection: the demography of two secondary
cavity nesters colonizing clearcuts. Auk 1 14:443^55.

Kellett, D. K., R. T. Alisauskas, and K. R. Mehl. 2003.
Nest-site selection, interspecific associations, and nest
success of King Eiders. Condor 105:373-378.

Knight, R. L. and R. E. Fitzner. 1985. Human
disturbance and nest site placement in Black-billed
Magpies. Journal of Field Ornithology 56:153-157.

Levin, S. A. 1992. The problem of pattern and scale in
ecology. Ecology 73:1943-1967.

Lima, ,S. L. 2009. Predators and the breeding bird:
behavioral and reproductive flexibility under the risk
of predation. Biological Reviews 84:845-513.

MacArthur, R. H. 1972. Geographical ecology: patterns
in the distribution of species. Harper and Row, New
York, USA.

Malizia, L. R. 2001. Seasonal tJuctuations of birds, fruits,
and flowers in a subtropical forest of Argentina.
Condor 103:45-61.

Martin, T. E. 1988. Processes organizing open-nesting
bird assemblages: competition or nest predation?
Evolutionary Ecology 2:37-50.

Martin, T. E. 1992. Breeding productivity considerations:
what are the appropriate habitat features for manage-
ment? Pages 455^73 in Ecology and conservation of

neotropical migrants (J. M. Hagan and D. W. Johnston,
Editors). Smithsonian Institution Press, Washington,
D.C., USA.

Martin, T. E. 1993. Nest predation and nest sites. New
perspectives on old patterns. BioScience 43:523-532.

Martin, T. E. 1998. Are microhabitat preferences of
coexisting species under selection and adaptive?
Ecology 79:656-670.

Martin, T. E. and J. V. Briskie. 2009. Predation on
dependent offspring: a review of the consequences for
mean expression and phenotypic plasticity in avian life
history traits. The year in evolutionary biology 2009.
Annals of the New York Academy of Sciences
1168:201-217.

Martin, T. E. and G. R. Geupel. 1993. Nest monitoring
plots: methods for locating nests and monitoring
success. Journal of Field Ornithology 64:506-519.

Martin, T. E. and J. J. Roper. 1988. Nest predation and
nest-site selection of a western population of the
Hermit Thrush. Condor 90:51-57.

Marzluff, j. M. 1988. Do Pinyon Jays alter nest placement
ba.sed on prior experience? Animal Behaviour 36:1-
10.

Matteuci, S. D. and a. Colma. 1982. Metodologi'a para el
estudio de la vegetacion. Monograph Number 22.
General Secretary of the OEA, Washington, D.C.,
USA.

Mayfield, H. 1961. Nesting success calculated from
exposure. Wilson Bulletin 73:255-261.

Mezquida, E. T. and L. Marone. 2002. Microhabitat
structure and avian nest predation risk in an open
Argentinean woodland: an experimental study. Acta
Oecologica 23:313-320.

Peluc, S., S. T. Sillett, j. T. Rotenberry, and C. K.
Ghalambor. 2008. Adaptive phenotypic plasticity in
an island songbird exposed to novel predation risk.
Behavioral Ecology 19:830-835.

Pimm, S. L., M. Ayres, A. Balmford, G. Branch, K.
Brandon, T. Brooks, R. Bustamante, and R.
Costanza. 2001. Can we defy nature’s end? Science
293: 2207-2208.

Ridgely, R. S. and G. Tudor. 1989. The birds of South
America. Volume 1. University of Texas Press, Austin,
USA.

Zar, j. H. 1999. Biostatistical analysis. Fourth Edition.
Prentice-Hall, Upper Saddle River, New Jersey, USA.

The Wilson Jounuil of Ornithology 1 22(4);68 1-688, 2010

BREEDING BIOLOGY OE THE TAIWAN BARBET
{MEGALAIMA NUCHALIS) IN TAIPEI BOTANICAL GARDEN

SHANG-YAO LIN,'^ FANG-CIAN LU,' FU-HSIEN SHAN,' SHU-PING LIAO,'
JING-LING WENG,' WEI-JEN CHENG,' AND CHAO-NIEN KOH'-'

ABSTRACT.— We studied the breeding biology of the Taiwan Barbet (Megalainia intchalis) in the Taipei Botanical
Garden (TBG) during the 2008 and 2009 breeding seasons. Breeding pairs produced a mean of 1.8 broods per season
and had a mean clutch size of 3.0 eggs. The mean incubation period was 13.8 days, and both parents shared
incubation. The mean nestling period was —27.5 days, considerably shorter than that of other Asian barbets. The
shorter nestling period may be related to the low fledgling success rate, multiple broods per season, and constant
animal matter food in TBG throughout the entire nesting period. Ambient temperature influenced the amount of time
spent by adults inside the nest cavity incubating eggs and brooding nestlings. Adult males provided more food to
nestlings than adult females, while adult females cleaned the nest more often than males. The overall egg-to-fledgling
success rate was low at 32.8%, because of infertile eggs and failure of eggs with embryos to hatch, abandonment,
anthropogenic disturbances, predation, weather disturbances, and other unknown factors. Received 7 January 2010.
■Accepted 15 April 2010.

The Asian barbets (Family Megalaimidae) are
one of the least studied groups compared to
barbets in Africa and Central and South America
(Yahya 1988). The Asian barbets belong predom-
inantly to the genus Megalaima, which includes
28 species and is centered in Southeast Asia
(Short and Horne 2001). Barbets are mostly
frugivores and are important pollinators and seed
dispersers. Some of the more insectivorous
barbets, such as White-cheeked Barbet (M.
viridis), have a vital role in controlling insect
pests in agricultural regions (Yahya 2000).
Barbets are primary tree-cavity nesters that
excavate their own nesting and roosting holes;
thus, losses of dead trunks or branches have
detrimental effects on this group (Short and Horne
2001). Gradual shrinkage of forest cover may
have a far-reaching effect on barbets as is the case
of the Red-crowned Barbet (M. rafflesii), which
has nearly been eradicated from Thailand because
of forest clear-cutting (Wells 1999). Detailed
information on the breeding biology of most
Megcdahna spp. is almost non-existent.

The Taiwan Barbet (M. tniclutli.s) is an endemic
species restricted to Taiwan, and was a subspecies
of the Black-browed Barbet (M. oorli niicluilis)
until recent genetic comparisons (Feinstein et al.
2008), and morphological and behavioral differ-

' Forest Protection Division. Taiwan Forestry Research
Institute. Number 53 Nanhai Road. Zhongzheng District.
Taipei City. Taiwan (R.O.C.).

^8011 Ryan Road, Unit 105. Richmond. BC V7A 2E4,
Canada.

^Corresponding author; e-mail: nien@tfri.gov.tw

ences among the subspecies (Collar 2006) gave it
full species status. The Taiwan Barbet is com-
monly distributed throughout Taiwan at low- to
mid-elevations. They commonly occur in broad-
leaf forests in both natural and man-made
habitats, including parks, botanical gardens,
school yards, and trees along sidewalks. This
sexually monomorphic species begins breeding in
April and continues into August (Fang 2008);
detailed information on its breeding biology and
parental roles is extremely limited. Koh and Lu
(2009) investigated characteristics of nest trees
and nest cavities of Taiwan Barbets in Taipei
Botanical Garden (TBG). They discovered the
species nested mostly in Ciunamomum camphora
trees and in dead standing trees or live trees with
dead branches. The only other research on this
species was of its general biology in Yangming-
shan National Park, Taiwan (Ho 1990). Primary
cavity nesters such as the Taiwan Barbet can be
affected by land management practices, especially
in urbanized environments (James and Kannan
2009) where urban trees may be trimmed or
completely removed during beautification actions
or when the trees po.se a safety hazard for
pedestrians or vehicles in traffic. The Taiwan
Barbet is commonly sold in pet stores and origins
of these captures are questionable, and most likely
illegal. Our objective is to present information on
the breeding biology and behavior of the relative-
ly poorly known Taiwan Barbet. More specifical-
ly, we ascertained the length of the incubation and
brooding periods, parental roles, and hatching and
fledging success rates in Taipei Botanical Garden.

681

682

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, December 2010

METHODS

Study Area. — This study occun'ed in Taipei
Botanical Garden (TBG) (25° 01 ' 53" N, 121° 30'
39" E) from March to September 2008 and 2009.
The botanical garden encompasses 8 ha of urban
green space in central Taipei City, Taiwan and has
>1,500 living plant species. The monthly tem-
perature ranged from 14.2 to 30.1° C, averaging
22.7° C in 2008 and 2009; monthly precipitation
ranged from zero to 1668.5 mm, averaging
209.3 mm (TBG, unpubl. data). This area of high
plant diversity and abundance in the heavily
urbanized city center provides excellent habitat
for urban wildlife, especially avian species. Eorty-
five avian species have been reported in TBG with
30 resident species and the other escapees,
migrants, and transients (Koh 1998). Koh (1998)
noted the year-round presence of Taiwan Barbets
and categorized their status as resident. The data
base from the Wild Bird Society of Ilan in Taiwan
(http://wildbird.e-land.gov.tw) indicates Taiwan
Barbets were observed almost every month from
December 2003 to October 2005, suggesting the
presence of a stable population in TBG.

Nest Searching, Observing, and Monitoring. —
Searches for paired barbets and their nests were
conducted opportunistically on foot within TBG
at least once each day from early March to mid-
September in 2008 and 2009. Searches were based
on visual and auditory cues, and nests were
located by following adults. The gender of the
adults was ascertained by observing their posi-
tions during copulation (male mounting the
female from rear) and was identifiable by learning
the key features between individuals in the pair.
Adults and fledglings were captured in mist nets,
measured, and color-banded during the nestling-
provisioning stage and after fledging, respective-
ly. The condition and status of nest cavities were
inspected using an endo.scope. The endoscope is
custom made with a pinpoint camera at the end of
an extendable pole of up to 6 m in length with a
small monitor attached to the camera via exten-
sion cables. This equipment allowed us to look
inside the nest, and videotape and photograph the
nest’s current conditions, whether it was a nesting
cavity, abandoned, or incomplete. The endoscope
also allowed us to identify the dates eggs were
laid, clutch size, and .stages of nestling develop-
ment. Breeding behaviors, including cavity exca-
vation, copulation, nestling provisioning, nest
cleaning, and time of entrance and exit ot parents

were observed and recorded using spotting scopes
and binoculars from distances that would not
disrupt the birds’ natural behaviors and actions.
Time spent inside the nest cavity during the
incubation and nestling periods by adults was
defined as incubation time and brooding time,
respectively. The brooding time did not include
time spent by adults inside the nest provisioning
nestlings or nest sanitation.

The incubation period was measured from the
day the last egg was laid to the day the first egg
hatched and was subdivided into two stages: early
(1-6 days after incubation) and late (7-15 days
after incubation). The nestling period was from
the day the first egg hatched to the day the first
nestling fledged. The nestling period was subdi-
vided into two stages: early (1-13 days after
hatching) and late (14-27 days after hatching).
The nestling period for nest sanitation and
nestling provisioning data was subdivided into
three stages: early (1-10 days after hatching),
middle (11-20 days after hatching), and final
(21 days and onward after hatching). Nest
sanitation was recorded when adults removed
wood chips mixed with fecal matter from the nest
cavity, and nestling provisioning was recorded
when adults brought food into the nest cavity.
Identification of food type was performed to the
best ability of the observers. Eood types fed to
nestlings were categorized as “plant matter’’,
“animal matter’’, and “unknown’’. All variables
were grouped into temporal periods of 4 hrs each,
from 0600 to 1000 (morning), 1000 to 1400
(midday), and 1400 to 1800 hrs (afternoon), to
evaluate mean differences of the variables during
different periods of the day and between the
parents in each time period (Yahya 1988). Success
rates were calculated by dividing the number of
nestlings or fledglings by the number of eggs or
nestlings in the previous life stage, respectively.
We removed the nest cavity and retrieved three
eggs of ~1 week of age to conduct physical
measurements of the eggs from one abandoned
and Hooded nest. Hourly temperatures were
obtained from the weather stations at TBG. The
associated temperatures for each incubation and
brooding period (hrs and days) were averaged to
obtain the mean hourly and daily temperature for
each incubation and brooding period.

Statistical Analysis. — Statistical analyses were
performed using SPSS for Windows Version 12.0
(SPSS Institute Inc. 2000) and Microsoft Office
Excel 2003. Examination of data normality was

Lin et al. • TAIWAN BARBET BREEDING BIOLOGY

683

TABLE 1 . Incubation and brooding period.s, and success
Garden. Taiwan.

rates of Taiwan Barbets during 2008-2009

in Taipei Botanical

Mean ± SE («)

Range

Number of eggs laid

3.0 ± 0.1 (1 1)

2.()-4.0

Number of eggs hatched

2.0 ± 0.3 (13)

O.O^.O

Egg hatching success rate, %

44.0 ± 12.5 (9)

0.0-100.0

Number of nestlings successfully fledged

LI ± 0.2 (15)

0.0-3.0

Fledging success rate — relative to number of eggs, %

32.8 ± 9.9 (11)

0.0-100.0

Fledging success rate — relative to number of hatchlings, %

45.4 ± 13.7 (9)

0.0-100.0

Length of incubation period, days

13.8 ± 0.3 (6)

13.0-15.0

Length of nestling period, days

27.5 ± 1.0 (6)

23.0-29.0

conducted using the one-sample Kolmogorov-
Smirov test. Means were compared using paired
r-tests or one-way ANOVA for data with normal
distributions. Either Mann-Whitney f/-tests or
Kruskal- Wallis tests were used for data with
non-normal distributions. Tukey’s HSD tests were
used, when appropriate, to examine which means
of groups significantly differed from one another.
Means ± SE are presented.

RESULTS

The breeding season of Taiwan Barbets began
in April when the first observed pair at the study
site laid their first egg and ended in August with
fledging of the last nestling. We located eight and
20 broods in 2008 and 2009, respectively.
Identified pairs in 2008 and 2009 produced a
mean of 1.8 broods/pair (range = 1. 0-3.0, n —

1 3 ) and had a brood success rate of 7 1 .4% (n =
28). The last egg of the .season in 2008 was laid on
6 August. However, due to natural weather
disturbances, this nest was abandoned and the
season terminated on 17 August. The first egg in
2009 was estimated to have been laid on 20 April,
and the last nestling fledged on 6 August.

Courtship Behavior and Copulation. — Court-
ship feeding was observed four times in 2009.
Males fed unidentified berries on three separate
occasions to females and an unidentified insect on
another occasion. Copulation may or may not
occur in conjunction with courtship feeding. The
male mounted the female in one occasion while
still holding three berries in his beak. The female
selected one berry after copulation from the
male’s beak and the male ate the rest. The
average copulation attempt was 10.8 ± 4.0 sec
in = 5). An unique behavior ob.served during the
breeding season was the beak-tapping action
performed by an adult male at the cavity entrance
in = 4).

Cavity E.xcavation. — The adult male signifi-
cantly excavated the nest cavity longer than its
partner. Males {n = 4) excavated for 17.6 ±
1.7 min/hr compared to 6.9 ± 0.8 min/hr for
females (paired /-test, / = 5.447, df = 3, E <
0.05). The time required to complete a nesting
cavity for one particular pair was 22 days.

Egg Laying and Clutch Size. — Eemales laid an
egg per day (n = 6), and the mean clutch size was

3.0 ± 0.1 (range = 2.0-4.0 eggs, n = 11;
Table 1). We opened the nest cavity and mea-
sured three eggs from a flooded and abandoned
nest cavity. The eggs were white in color and
subelliptical-shaped. Mean egg length was 26.9 ±
0.5 mm (range = 26.1-27.7 mm, n = 3) and the
mean width was 19.8 ± 0.2 mm (range = 19.4-

20.0 mm, n = 3); mean egg weight was 5.4 ±
0.4 g (range = 5.0-5. 8 g, n = 2). One other egg
was addled when retrieved and weighed 3.0 g.

Incubation Period. — The incubation period
began when the last egg of the clutch was laid
and lasted 13.8 ± 0.3 days (range = 13.0-

15.0 days, n = 6; Table 1). Time spent by adults
incubating eggs throughout the incubation period
was 34.1 ± 2.3 min/hr (range = 21.6-43.3 min/
hr, n = 9). Males and females incubated equally
in early and late stages. Incubation occurred most
frequently in the early morning and decreased as
the day progressed (range = 29.5^7.4 min/hr;
Fig. 1). Eggs were most frequently incubated by
adults during 0600 to 1000 hrs (ANOVA, E =
3.837, df = 25, P < 0.05) when compared to 1000
to 1400 and 1400 to 1800-hr periods (Tukey test,
P < 0.05). Mean ambient temperature was lowest
at 23.4° C during early mornings and increased
to 29.5° C from 1300 to 1400 hrs, and then
decreased to 27.2° C from 1700 to 1 800 hrs (TBG,
unpubl. data; Fig. 1 ). Males incubated longer than
females during the 0600 to 1000 and 1400 to
1800-hr periods but the differences were not

684

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, December 2010

U

o

<0

Uh

3

<u

C3

TD

C

aj

Time periods

FIG. 1. Hourly incubation time by Taiwan Barbets in different hours of the incubation period, and the mean daytime
temperature for each day during 2008-2009 in Taipei Botanical Garden, Taiwan. The mean temperature was calculated by
averaging the daytime temperatures of each incubation period. Error bars are ± SE.

significant. Females stayed overnight with the
eggs during the 5 days where we continued
observation past 1800 hrs (n = 4). The egg
hatching success rate was 44.0 ± 12.5% at the
end of the incubation period (n = 9; Table 1 ), and
chicks were born naked and with eyes clo.sed.

Nestling Period. — The nestling period of Tai-
wan Barbets la.sted 27.5 ± 1.0 days (range = 23-
29 days, /; = 6; Table 1). Adults spent 9.0 ±
2.9 min/hr (range = 0.8-21.1 min/hr, /; = 8)

throughout the nestling period brooding the
nestlings, and brooding time in the early stage
was significantly longer than in the late stgge
(independent /-test, / = 4.196, df = 10, F <
0.001). Females spent a longer period of time
brooding nestlings than males during the nestling
period, but we did not find any significant
difference between them. Brooding time by
parents was longer in the early morning and
decreased at the end of the observation period in

Liu Cl al. • TAIWAN BARBET BREEDING BIOLOGY

685

<1-!

OJ

OJj

O

P

30.0

o

27.5 3

a

i-

<u

25.0 I

>u

22.5 ■

■o

20.0 S

17.5

15.0

12.5

10.0

7.5

5.0

2.5

0.0

Time periods

FIG. 2. Hourly brooding time by Taiwan Barbets in different hours of the nestling period and the mean daytime
temperature for each day during 2008—2009 in Taipei Botanical Garden, Taiwan. The mean temperature was calculated by
averaging the daytime temperatures of each brooding period. Error bars are ± SE.

late afternoon (Fig. 2). Temperatures were coolest
in the early morning and late afternoon and
warmest in mid-day (Fig. 2). Females continued
to be the primary parent staying overnight inside
the nesting cavity during the nestling period (/; =
4) while a male was ob,served once with the
nestlings overnight. Only 45.4 ± 13.7% of the
nestlings fledged successfully (n = 9; Table 1).

Nestling Provision, Food Types, and Nest
Sanitation. — Frequency of nestling provision and

nest cleaning increased as brooding time de-
creased. The overall food provisioning rate was
1.3 ± 0.2 times/hr and increased during the
nestling period (range = ().3-3.0 times/hr. n = 8).
On average, males delivered food more frequently
but not more significantly than females during the
nestling period. The greatest amount of animal
matter (27.1%) was fed to nestlings during the
middle stage of the nestling period and decreased
to 22.3% in the final stage. There were no

686

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

differences in food provisioning rates between
animal and plant matter during the nestling
period. Plant matter consisted entirely of berries
from many different plant species, while animal
matter consisted of mostly insects, except for four
special small vertebrates, one gecko and three
lizards. The insects provided to nestlings were:
Coleoptera ( 1 1.5%), Hemiptera (7.9%), Hymenop-
tera (1.8%), Mantodea (1.3%), Odonata (<1.0%),
Lepidoptera (<1.0%), Orthoptera (<1.0%), and
Blattaria (<1.0%); the remainder were insects of
unidentified Orders (74.8%) (/? = 8).

Nest sanitation was a frequent and important
task performed throughout the nestling period,
averaging 0.9 ± 0.1 times/hr (range = 0.5-1. 2
times/hr, n = 8) by both parents. Females
maintained a higher nest sanitation rate than
males throughout the nestling period in all three
time periods (paired /-test, / = —4.206, df = 7,
P < 0.01), especially in the early stage (paired
/-test, / = —4.376, df = 4, P < 0.05). Adults
cleaned the nest cavity more frequently and
significantly during the 0600 to 1000-hr period
than in the 1400 to 1800-hr period (Tukey test,
P < 0.05).

DISCUSSION

The breeding season of Taiwan Barbets began
in April, similar to other Megalaima spp. that live
at similar latitudes. The breeding seasons of some
Megalaima spp. may start as early as January and
end as late as October, but commonly occur from
March to July (Short and Horne 2001 ). Ho ( 1990)
reported the breeding season of Taiwan Barbets in
Yangmingshan National Park, Taiwan concluded
in late August. The breeding .season at TBG also
terminated in August. The average breeding
season length of Taiwan Barbet in TBG was
>3 months in length. We hypothesize that length
of the breeding sea, son is affected by weather
disturbances, number of broods per pair, length of
incubation and nestling periods, and the repro-
ductive condition of adult females after each
brood.

Egg Laying and Clutch Size. — The dimensions
of Taiwan Barbet eggs are smaller than eggs of M.
oorti nnchalis documented by Short and Horne
(2001). More egg data are needed to provide a
more preci.se range and mean of egg dimensions.
The mean clutch of 3.1 eggs was greater than for
M. viridis, an Asian barbet of similar body size
(Yahya 1988). However, M. viridis has larger
eggs and higher egg hatching success rate than

Taiwan Barbets. We hypothesize the Taiwan
Barbet’ s large clutch size helps to compensate
for low egg survival.

Incubation Period. — Eggs of the same clutch
are laid on consecutive days and hatch on the
same day, as observed for M. viridis and M.
rubricapilla (Yahya 1988). Time required for
Taiwan Barbet eggs to hatch is —14 days, within
the range of 12 to 15 days for other Asian
barbets (Short and Horne 2001). Adults were
frequently inside the nest incubating eggs in the
early morning and, by late afternoon, their time
inside the cavity diminished. This temporal
difference is most likely influenced by ambient
temperatures. Temperatures were cooler in early
morning and late afternoon and warm during the
mid-day, which correspond with more and less
time by adults incubating the eggs, respectively.
M. viridis and M. rubricapilla parents are also
more attentive during cool mornings and rainy
days than during the afternoon (Yahya 1988).
The effect of ambient temperature fluctuations
on behaviors of breeding adults supports the
hypothesis that optimal temperature has an
important role in affecting the success of
breeding pairs (White and Kinney 1974). Males
incubated more in morning and afternoon
periods than females; we hypothesize males
relieve females after their overnight stay inside
the cavity and become the primary incubator in
the early morning. Males again take on the role
of the primary incubator in late afternoon as the
female forages and prepares for her nocturnal
incubation duty in the evening.

Nestling Period. — The nestling period of
—27.5 days for Taiwan Barbets is short compared
to 35-38 days for five small- to large-sized
Megalaima spp. (Short and Horne 2001). We
hypothesize the shorter nestling period may be
related to the low fledgling success rate, multiple
broods per .season, and steady animal matter food
provided by the parents throughout the entire
nesting period. Martin and Li (1992) commented
that length of the nestling periods was inversely
related to risks of nest failure and number of
broods the parents produced in a breeding season.
The lledgling failure rate in our study was
relatively high compared to other Megcdaima
spp. (Yahya 1988), and parents did produce
multiple broods per breeding season. Taiwan
Barbets supplied nestlings with animal matter food
during all stages of the nesting period, especially
during the middle stage. Adult M. rubricapilla and

Liu et al. • TAIWAN BARBET BREEDING BIOLOGY

687

M. viriclis ted their nestlings more plant matter food
than animal matter food for the majority of the
nesting period (Yahya 1988). We suggest that
Taiwan Barbet nestlings are provided with a
constant source of animal protein and can develop
faster than other Megalaima spp.

Provisioning and Nest Sanitation.— Males at
our study site have a slightly larger role in
provisioning nestlings than females, while fe-
males have a larger role in nest sanitation than
males throughout the day. Overall, the parents
more or less cooperated with each other in
nestling provisioning and nest sanitation. A
female disappeared during the nestling period on
one occasion, leaving the male to do all nest
sanitation and nestling provisioning. The three
nestlings died in the male’s care. This result
suggests that fledging success may be greatly
influenced by the performance of either parent
due to their cooperative roles in nestling provi-
sioning and nest sanitation.

Success and Mortality Rates. — Yahya (1988)
reported the egg hatching success for M. viridis
and M. riibricapilla to be 71.1 and 80.0%,
respectively, which are higher than that of Taiwan
Barbets in our study. Causes for failures of
Taiwan Barbets were infertile eggs and failure
of eggs with embryos to hatch (40.0%), human
influences (25.0%), weather disturbances
(15.0%), and unknown egg disappearances
(20.0%), possibly a result of nest predation. The
fledgling success rate was also lower than for M.
viridis and M. riibricapilla at about 75.7 and
75.0%, respectively (Yahya 1988). The specific
causes of nestling mortality are unknown, but
27.2% of the nestlings died inside the nest and
were removed from the nest cavity by the parents,
and 36.4% of the nestlings starved to death
because parents had not returned to the nests.
Mortalities of the remaining unfledged nestlings
were of unknown disappearances. Potential fac-
tors that reduce fledgling success of cavity nesters
include nest parasites, nest desertion, starvation of
young, predation, and hyperthermia (Ricklefs
1969, Nilsson 1986, LaBranche and Walters
1994). The specific causes that result in low egg
hatching and fledgling success rates remain
unclear and are a priority for future researchers.

The workload differences between parents and
a slightly more important role of females than
males during incubation to nestling periods
provide a new perspective into the relatively
unknown breeding biology of Taiwan Barbets.

Low breeding success rates merit attention and
protection, especially for threats the species has
from city beautification processes and the pet
trade industry.

ACKNOWLEDGMENTS

We thank Hsiao-Wei Yuan and Tzung-Su Ding for
comments on earlier drafts. We also thank C. E. Braun for
proofreading the English of the manuscript, and two
anonymous reviewers for commenting on the manuscript.
We are grateful to C.-W. Chen, Y. S. Ho, W.-C. Yeh, and
Y.-M. Chen for assistance with field work. We thank M.-S.
Huang, H.-W. Chou, S.-Y. Peng, and L.-S. Lung for
support. We are also grateful to K.-K. Kuo for supplying us
with a professional endoscope to investigate the conditions
inside nest cavities. We sincerely thank S.-Y. Chen and the
many volunteers who contributed to the successful
completion of this study. Our study was supported by
grants from the Council of Agriculture and National
Science Council, Taiwan.

LITERATURE CITED

Collar, N. J. 2006. A taxonomic reappraisal of the Black-
browed Barbet Megalaima oorti. Forktail 22:170-173.
Fang, W.-H. 2008. A guide to all birds of Taiwan. Owl
Publishing House, Taipei, Taiwan.

Feinstein, J., X. Yang, and S.-H. Li. 2008. Molecular
systematics and historical biogrography of the Black-
browed Barbet species complex (Megalaima oorti).
Ibis 150:40-49.

Ho, Y.-C. 1990. The biological study of Muller's Barbet
Megalaima oorti imchalis in Yangmingshan National
Park. Thesis. National Taiwan University, Taiwan.
James, D. A. and R. Kannan. 2009. Nesting habitat of the
Great Hornbill (Biiceros hicornis) in the Anaimalai
Hills of .southern India. Wilson Journal of Ornithology
121:485^92.

Koh, C.-N. 1998. Report on the bird fauna in the Taipei
Botanical Garden. Taiwan Journal of Forest Science
13:69-78.

Koh, C.-N. and F.-C. Lu, 2009. Preliminary investigation
on nest-tree and nest-cavity characteristics of the
Taiwan Barbet (Megalaima niichalis) in Taipei
Botanical Garden. Taiwan Journal of Forest Science
24:213-219.

LaBranche, M. S. and J. R. Walters, 1994. Patterns of
mortality in nests of Red-cockaded Woodpeckers in
the sandhills of .southcentral North Carolina. Wilson
Bulletin 106:258-271.

Martin, T. E. and P. Li. 1992. Life history traits of open-
vs. cavity-nesting birds. Ecology 73:579-592.

NilS-SON, S. G. 1986. Evolution of hole-nesting in birds: on
balancing selection pressures. Auk 103:432-435.
Ricklefs, R. E. 1969. An analysis of nesting mortality in
birds. Smithsonian Contributions to Zoology 9:1^8.
Short, L. L. and J. F. M. Horne. 2001. Toucans, barbets,
and honeyguides. Oxford University Press, Oxford,
United Kingdom.

688

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, December 2010

SPSS Institute Inc. 2000. SPSS for Windows. Version
12.0. SPSS Institute Inc., Chicago, Illinois, USA.
Wells, D. R. 1999. The birds of the Thai-Malay Peninsula.
Volume I. Non-passerines. Christopher Helm, London,
United Kingdom.

White, F. N. and J. L. Kinney. 1974. Avian incubation.
Science 1 86: 107-1 15.

Yahya. H. S. a. 1988. Breeding biology of barbets,
Megalaima spp. (Capitonidae: Piciformes) at Periyar
Tiger Reserve, Kerala. Journal of the Bombay Natural
History Society 85:493-511.

Yahya, H. S. A. 2000. Food and feeding habits of Indian
barbets, Megalaima spp. Journal of the Bombay
Natural History Society 97:103-1 16.

The ll 'ilson Journal of Ornithology 1 22(4):689-698. 2010

BREEDING BIOLOGY OF THE GOLDEN-FACED TYRANNULET
{ZIMMERIUS CHRYSOPS) IN VENEZUELA

WILLIAM GOULDING' AND THOMAS E. MARTIN'

ABSTRACT.— We present the first detailed information on the breeding biology of the Golden-faced Tyrannulet
(Ziiumernis chiysops). Information was gathered from 96 nests in Yacambn National Park, Venezuela during the 2002 to
_008 breeding seasons. The enclosed nest was similar to descriptions of nests of other species in the genus. Eggs were laid
on alternate days with mean (± SE) clutch size of 1.98 ± 0.02 (n = 45) and fresh weight of I.6I6 ± 0.020 g (/; = 48). Only
the female incubated and the incubation period averaged 16.9 ± 0.3 days {n = 10). Nest attentiveness {% time on the nest)
averaged 66.0 ± 1.6% {n = 40) and increased from early to mid- and late-incubation. Incubation behavior yielded an
average 24-hi egg tempeiature ot 34.88 ± 0.45° C (n = 7 nests, 43 days). The nestling growth rate constant for body mass
(k — 0.285 ± 0.01 1) was slow even for tropical tyrannids. The nestling period for nests where exact hatch and fledging days
were observed ranged from 17 to 19 days with an average of 18.0 ± 0. 2 days (n = 9). Both females and males fed nestlings
at a rate that increased over the nestling period with a mean of 4.41 ± 0.65 trips/hr {n = 10) during days 1 and 2 after
hatching, and 14.93 ± 2.36 trips/hr {n = 6) at pin-break (days 10-11). Daily predation rates were similar in egg-laying
(0.052 ± 0.025; n = 76.5 exposure days) and incubation periods (0.068 ± 0.010; n = 575.5 exposure days), but were lower
during the nestling period (0.039 ± 0.010; n = 377.0 exposure days). The total daily predation rate (0.057 ± 0.007; n =
989.0 exposure days) indicated only 12% of nests were successful. These breeding biology parameters forZ chiysops differ
substantially from other tyrant-flycatchers and temperate species, further highlighting the diversity within the Tyrannidae.
Received 12 December 2009. Accepted 24 May 2010.

The Golden-faced Tyrannulet (Zimmerius
chrysops) is a member of the complex and large
New World Family Tyrannidae. Tyrannids are
diverse in behavior and occupy a broad range of
habitats and niches, reaching their greatest
diversity in the Neotropics (Ridgely and Tudor
1994, Hilty 2003, Fitzpatrick 2004), where they
comprise >20% of passerine species (McGowan
2004). General trends in avian life histories cannot
be clarified without studies of tropical species to
compensate for the historical bias toward temper-
ate species (Martin 1996, 2004; Stutchbury and
Morton 2001). The paucity of information on
tropical bird species is particularly evident for
tyrannids. Information on the breeding biology of
most tropical tyrannids is sparse despite their
prominence in the neotropical avifauna, and
diversity of form and function.

Zimmerius chry'sops is an example of the sparse
information on tropical tyrannids. Published
breeding biology information on Z. chrysops is
little more than a description of the enclosed,
dome-shaped nest with a side entrance, and
observed breeding dates (Hilty and Brown 1986,
Best et al. 1996). Yet, it is broadly distributed
with a range that extends from Peru, north through

' uses. Montana Cooperative Wildlife Research Unit.
University of Montana. Missoula, MT 59812, USA.

-Current address: Indooroopilly, Brisbane 4068 Queens-
land, Australia.

^ Corresponding author; e-mail: willgoulding@yahoo.com.au

the Andes to Colombia and Venezuela, and occurs
from —300 to 2,400 m asl (Hilty and Brown 1986,
Ridgely and Tudor 1994, Hilty 2003). The species
is associated with the middle to upper forest
layers, secondary growth, forest edges, as well as
with plantation and garden habitats (Fjeldsa and
Krabbe 1990). Our objectives are to: (1) present
detailed reproductive biology information for Z.
chrysops, and (2) compare this information with
available temperate and tropical avian life history
information.

METHODS

We studied Golden-faced Tyrannulets during
the March-June breeding seasons in 2002-2008 in
Yacambu National Park, Lara State, Venezuela
(09° 42' N, 69° 42' W). Yacambu contains
premontane and montane cloud forest of the
northern Andes and encompasses a gradient from
500 to >2,300 m asl (Fierro-Calderon and Martin
2007). Peak precipitation in Yacambu occurs from
late April through July (Fierro-Calderon and
Martin 2007) with clouds and rain usually
enveloping habitats after the morning period,
producing cool wet conditions. We worked with
equal field effort in the same areas across years in
a 1 ,350 to 2,000 m asl re.search area.

Nests were located primarily by observing
parental behavior, but also by systematic search-
ing. Dimensions of newly constmeted nests were
measured before use and weather affected their

689

690

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

shape. Eggs were described following Preston
(1953) as clarified by Palmer (1962). Eggs were
weighed for fresh egg mass between days 0 and 3
of incubation, and opportunistically later. The
incubation period was defined as the period
between the last egg laid (day 0) and the last
chick hatched (Nice 1954, Martin 2002). Nestling
period was defined as the number of days
between the last egg hatching and the last chick
leaving the nest.

Video cameras were placed at a discreet distance
(>5 m) from the nest and camouflaged with
vegetation to minimize any influence on parental
behavior. Nests were filmed from 30 min after
sunrise for 6 to 8 hrs (Martin and Ghalambor 1999,
Martin 2002). Filming occuiTed on days 2-3 in
early incubation, during the middle of the incuba-
tion period, and ~2-3 days before expected hatch
day (early, middle, and late incubation). Nests
found after laying were filmed opportunistically on
other days (Martin 2002). Nest attentiveness at
individual nests was measured as the percent time
females were on the nest based on the number of
minutes spent incubating divided by the total
number of minutes a nest was filmed (Martin 2002).

Incubating egg temperatures (°C) were mea-
sured by inserting a thermistor into the egg center
through a small opening in the shell that was
sealed w'ith adhesive (Weathers and Sullivan
1989, Martin et al. 2007). HOBO Stowaway
XTI dataloggers (Onset Corp., Bourne, MA,
USA) were connected to the thermistor by
discreetly inserting a fine wire through the nest-
wall. Thermistors were inserted within the first
2 days of incubation and dataloggers recorded
temperatures in 12-24 sec intervals for 5-7 days
per nest following Martin et al. (2007).

Nests were videotaped during the early nestling
period (days 1-3), during pin-break when flight
feathers broke their sheaths (days 10-11), and the
late nestling period preceding expected fledging
(day 15 and above). Nests were opportunistically
filmed on other days. Videotapes were analyzed
for parental brooding attentiveness (%) based on
the number of minutes brooding divided by the
total number of minutes a nest was filmed, as well
as the number of provisioning trips/hr (Martin and
Ghalambor 1999, Martin et al. 200()b). Analysis
of variance (ANOVA) was used to test for
changes in incubation and nestling attentiveness
between stages. Changes in brooding and provi-
sioning rates with ne.stling age were analyzed
using correlation coefficients (/■).

Chicks were measured every second day (from
day located) at the same time of day and at pin-
break. Standard measurements taken were mass,
tarsal length, wing chord, and primary feather pin
measurements. These data were used to calculate
the growth rate constant (k) following Ricklefs
(1967) and Remes and Martin (2002). Overall
nesting success and predation rates for egg-laying,
incubation, and nestling periods were calculated
using the Mayfield method (Mayfield 1961, 1975;
Hensler and Nichols 1981; Johnson 2007). Chi-
square tests of independence (X^) were used to
test for differences in daily predation rates among
nest stages (egg laying, incubation, and nestling
periods).

Egg and nestling masses were measured using
an ACCULAB (Elk Grove, IL, USA) portable
electronic scale with a precision of 0.001 g; all
measurements were taken using Mitutoyo Digital
Calipers (Kingsport, TN, USA) with a precision of
0.01 mm. SPSS Version 15.0 (2006) was used for
statistical tests and means are given ± one
standard error (SE).

RESULTS

Breeding Habitat. — Golden-faced Tyrannulets
nested most commonly near 1,400 m asl and were
not encountered above 1,500 m asl. This altitudi-
nal zone contained primary and secondary growth
premontane and montane cloud forest with
patches of regenerating coffee {Cojfea arabica)
plantations. Active nesting pairs were encountered
near forest edges including roads, forest trails,
water bodies, and areas of historical disturbance
inside the forest boundary. Areas where nests
were found contained suitable resources for nest
construction (moss and lichen) on tree trunks.

Nest Construction and Sites. — Pairs were ob-
served exploring potentially suitable nest loca-
tions together prior to the nest-building process.
Only one of the pair, presumably the female,
constructed the nest (// = 88 nest-building
observations), but the mate generally accompa-
nied the nest-building individual. Nests were
typically constructed in a ball of moss on the
side of a tree, under a branch, or in an epiphyte
(Fig. 1). Two di.stinct building approaches were
observed for nest construction. The first and more
common approach involved building into an
existing lump of moss or lichen. The second
involved building the globular nest onto a bare
trunk surface. In the first approach, when a
suitable portion of moss or lichen was identified.

Goiilc/ing and Martin • GOLDEN-FACED TYRANNULET BREEDING BIOLOGY

691

FIG. 1. Typical nest (with eggs and nestling) of the
Golden-faced Tyrannulet with incubating female (Photo-
graphs by W. Goulding).

the adult would open a cavity first with the beak
and head and, when at the mid-body section,
would use its wings to thrust apart the moss and
widen a cavity. The adult would collect materials
to create the inside of the nest once a cavity was
established. The adult used more of the other
nesting materials for construction inside with the
moss in situ. However, in the second approach
where the nest was created on a relatively bare
surface, initial construction involved slinging the
nest material with spider webs from a point on the
bark or from a small piece of moss or lichen.
Females were observed repeatedly descending to
the ground to scour exposed banks and road
cuttings for rootlets and spider webs during
construction (e.g., 6 visits in 10 min to the same
location on a bank). They also loudly called from
beside the nest after building visits with the
nearby male usually responding. Females were
observed re-using nesting material from previous
unsuccessful nests, particularly the lining, but new
material was often used. The small size of the nest
often resulted in destruction of the nest during
predation events.

The mean height of nests above ground was
5.43 ± 0.42 m (n = 55). The external measure-
ments were 1 16.14 ± 4.27 mm (/; = 10) in height

FIG. 2. Seasonal distribution of nest initiation (date the
first egg is laid in a nest) for the Golden-faced Tyrannulet at
weekly (7 day) intervals.

by 70.83 ± 2.50 mm (« = 6) in width. Nest cups
had an external depth of 48.25 ± 2.87 mm (/? =
10) with internal depth and internal diameter
being 32.85 ± 2.55 mm {n = 10) and 40.57 ±
2.71 mm (n = 7), respectively. Analysis of 11
nests showed an outer supporting structure of
moss into which spider webs, black rootlets,
lichen, and spider egg-casings of varying propor-
tions were inserted with internal layers of finer
dark rootlets. The final internal layer was a golden
lining of soft, comose seed-arils fitting descrip-
tions of liana seeds in the genus Odontadenia
(Apocynaceae) (Orlando Vargas Ramirez, pers.
comm.). The golden lining was characteristic of
all Z. chn’sops nests that we observed.

The earliest nest was initiated (i.e., first egg
laid) on 13 March and the latest on 15 May with a
mean initiation date of 11 April ± 2.2 days (n =
61) across years (Fig. 2). Nest-building activity
began in February. Two nests that fledged chicks
in early May 2006 were re-used in February 2007
with new material added.

Eggs and Clutch Size. — Eggs were elliptical in
shape and cream-white with liver-brown spots and
markings mostly concentrated in a ring toward the
obtuse end (Fig. 1). Eggs were 17.17 ± 0.25 mm
in length by 13.42 ± 0.1 1 mm in width (/? =11).
Eggs weighed between 1.286 and 1.861 g with a
mean mass of 1.616 ± 0.020 g (/? = 48). Fresh
egg weight represented 18% of mean adult body
mass of 9.0 ± 0.2 g {n = 25).

Clutch size was usually two with only one of 45
nests found prior to incubation having a single
egg, yielding a mean clutch size of 1.98 ± 0.02
eggs (n = 45). The mean number of days
observed between nest completion and laying of

692

THE WILSON JOURNAL OL ORNITHOLOGY • Vul. 122, No. 4, December 2010

Early Middle Late

Incubation period

■>

CD

C

'c

o

w

>

o

25

20

15

10

5

0

B

• •

Si**

• :

5 10 15

Nestling age (day)

20

FIG. 3. Average (A) nest attentiveness, and (B) on- and
off-bout durations across three periods of incubation: early
(days 1-5), middle (days 6-11), and late (days 12-18).
Sample sizes reflect numbers of nests.

FIG. 4. The change with nestling age in (A) female
brooding behavior (% time spent brooding) and (B) rates
that parents visit the nest to provision nestlings.

the first egg was 4.71 ± 0.47 (n = 1). Eggs were
laid in the morning on alternate days.

Incubation. — Females incubated the eggs with-
out help from males. Males were rarely observed
visiting nests during incubation, and did so at only
two of 40 nests with video observations of 6-8 hrs
each, yielding an average of 0.01 ± 0.01 visit.s/hr
(n = 40).

Females were not observed incubating prior to
laying the last egg and eggs hatched synchro-
nously. Nest attentiveness averaged 66.0 ± 1.6%
{n = 40) and was lower (ANOVA, F2.37 = 6.8,
P = 0.003) in early incubation compared with
middle and late incubation periods (Fig. 3A).
Incubation on-bouts averaged 33.35 ± 1.94 min
(n = 40) and did not change over the incubation
period (ANOVA, F2.37 = 1.1, F = 0.3; Fig. 3B).
Length of off-bouts averaged 16.19 ± 1.59 min (n
= 40) and decreased from early to later incubation
stages (ANOVA. F2.37 = P = 0.055;
Fig. 3B). Incubation behavior yielded an average
24-hr egg temperature of 34.88 ± 0.45° C (/? = 7
ne.sts, 43 days of .sampling). The incubation period
was 16.9 ± 0.3 days (/? = 10) for nests found

prior to beginning of incubation and where exact
hatch was observed.

Nestling Period. — Both adults were observed
provisioning chicks and the male was also
observed passing food to the female when she
was brooding. Adults regularly fed the nestlings
fruit that appeared to be mistletoe berries
(possibly Antidaphne viscoidea or Phoradendron
spp.). which were regularly found stuck near the
beak of chicks. Females regularly regurgitated
seeds when incubating and brooding, taking care
to eject them outside the nest. The percentage of
time females brooded decreased through the
nestling period (r == —0.86, P < 0.001), and
stopped after the eighth primary pin feather broke
its sheath on days 10-11 (Fig. 4A). Provisioning
rates increased over the nestling period (r = 0.80,
P < O.OOl, Fig. 4B). Parents visited the nest an
average of 4.41 ± 0.65 trips/hr (n = 10) during
days I and 2 after hatching, and 14.93 ± 2.36
trips/hr (/; = 6) at pin-break.

Chicks had gray down distributed lightly over
the head and body, and orange skin and beak at
hatching (Fig. I). The nestling period for nests
where exact hatch and fledging days were

Goiilcling and Marlin • GOLDEN-FACED TYRANNULET BREEDING BIOLOGY

693

10 1

8 -

S 6 -

c/)

CO

I 4

2 ^
0

A

.1*

..I'*

.i

• i

I • •

/c= 0.285 + 0.011

E

E

O)

c

CO

CO

CT3

18

15 -

12 -

9 -

6 -

O)

c

0)

■O

i—

o

o

O)

c

3

55

50

45

40

35

30

25

20

15

10

5

B

!•

:ii

M

:i

/f= 0.200 + 0.012

.1

• •

• • • • •

• •
I •

A-= 0.201 + 0.010

0 5 10 15

Nestling age (day)

FIG. 5. Relation.ship.s of (A) mas.s, (B) tarsus length,
and (C) wing chord length plotted against age for Golden-
faced Tyrannulets and their estimated growth rate constants
(A). The dashed lines represent mean adult sizes.

observed ranged from 1 7 to 19 days with an
average of 18.0 ± 0.2 days (n = 9). Nestling mass
on days 10 and 1 1 (i.e., pin-break) was 6.966 ±
0.152 g in = 9). The growth rate constant (A.) for
body mass was low (Fig. 5 A) and nestlings were
close to mean adult weight of 9.0 ± 0.2 g (/; =
25) at fledging (Fig. 5A). Growth rate based on

tarsus length was slower at k = 0.200 ± 0.012 and
tarsus size was essentially the same length as
adults at 16.26 ± 0.17 mm (/; = 22) at fledging
(Fig. 5B). Growth rate based on wing chord was k
= 0.201 ± 0.010, but wing chord at fledging was
less than adult size (Fig. 5C) of 52.03 ± 0.62 mm
in = 20).

Nesting Success and Predation. — Predation
accounted for 91.6% of failures with remaining
failures attributed to nests failing, abandonment,
and poor nest condition that allowed chicks to fall
out (re-used nest). Birds seemed to be a main
predator of nests with punctured egg remains at
times found near the nest. Nest predation often
coincided with presence of army ants iLabidiis
praedator) and large mixed-bird flocks feeding
near the nest. One nest was also filmed being
predated by a capuchin monkey iCehus oliva-
ceus). Snakes were not recorded in video
observations and rarely observed at this elevation
in Yacambti across all years (TEM, pers. obs.).

The total daily predation rate was 0.057 ±
0.007 in = 989 exposure days) with a total daily
mortality rate of 0.064 ± 0.008. An estimated
12% of nests were successful based on a total
nesting period of 37 days. The daily nest
predation rate during egg-laying (0.052 ± 0.025;
n = 76.5 exposure days) did not differ iX- = 0.4,
P = 0.6) from the incubation period (0.068 ±
0.010; n = 575.5 exposure days). However, the
daily predation rate during the nestling period
(0.039 ± 0.010; n = 377 exposure days) was
lower iX~ = 4.2, P = 0.029) than during
incubation.

DISCUSSION

Golden-faced Tyrannulets in Yacambti Nation-
al Park exhibited breeding biology parameters
different from most typical tyrannids (Fitzpatrick
2004). The nest appears globular, as also de-
scribed for congeners (e.g., Z. acer [Beebe et al.
1917], Z. iinprohiis [Hilty 2003 j. and Z. vili ssinnis
jSkutch 1960]), but varied in appearance depend-
ing on location of attachment and amount of
exposed nest. Nests of Z. clirysops, compared with
congeners, were slightly larger than those of Z.
vilissinnis (Skutch 1960), but had a shallower
nest-CLip than ob.served for Z. acer (Beebe et al.
1917; this was a single observation). Nest
construction and materials were virtually the same
as those observed for Z. vilissinnis, including the
soft seed-aril lining (Skutch I960), and similar to
Z. acer (Beebe et al. 1917). The use of epiphytic

694

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

plant resources for nesting by neotropical birds is
well recognised (Nadkarni and Matelson 1989,
Fierro-Calderon and Martin 2007).

Golden-faced Tyrannulets show flexibility in
nesting height with 71% of nests in Yacambil
below the 8-12 m height range observed in
Colombia (Hilty and Brown 1986), and a nest
recorded ~45 m above ground in an emergent
Ceiba tree in Ecuador (Cisneros-Heredia 2006).
Our observations do not seem to reflect biases
towards finding low nests because we found the
majority of nests through parental behavior. We
found nests to the tops of canopies by watching
parental behavior. The lower height of nests at our
site might reflect the lower stature of the disturbed
forest areas and edges in which the species
occurred at our site. Differences between nesting
height in disturbed or regrowth areas and primary
forest have been observed in other tyrannids (e.g.,
Z. vilissinuis [Skutch 1960] and Todiroslnim
chrysocrotaphum [Hilty and Brown 1986]), indi-
cating structural differences in habitat influence
nest height variation.

Nest architecture seems to be an evolutionarily
conservative character that has been used to
indicate phylogenetic relationships (Mobley and
Prum 1995, Sheldon and Winkler 1999, Zys-
kowski and Prum 1999). Enclosed nests are
thought to be a derived character from the more
common open nests reported for the majority of
tyrannids (Collias and Collias 1984, Fitzpatrick
2004). It is also thought to be a character that
evolved independently in different tyrannid sub-
families (Fitzpatrick 2004). The sparse available
information on nests of species of Zimmerius
indicates only subtle differences between conge-
ners. However, nests of the majority of species in
the genus have yet to be investigated. Nest
similarity may support a close relationship be-
tween Camptostoma (Haverschmidt 1954) and
Zimmeriii.s (unless evolved independently; Lanyon
1988). These two genera, together with the closely
related Phyllomyias virescens and others in that
genus that may be discovered to have enclosed
nests (Fitzpatrick 2004, Ohlson et al. 2008,
Rheindt et al. 2008), repre.sent the only enclosed
nest-builders in the elaeniine clade proposed by
Ohl.son et al. (2008). Despite differences in
external materials, they have similarly attached
and shaped immobile nests lined with ‘plant wool’
(not feathers) with a simple side-opening near the
top (Beebe et al. 1917, Haverschmidt 1954, Skutch
1960, this study).

The observed peak of breeding activity of Z.
chtysops is just prior to and overlapping the peak
precipitation period from late April through July
in Yacambu (Fieiro-Calderon and Martin 2007).
This is a pattern in avian breeding biology
observed at other neotropical sites (Skutch 1950,
1960; Marchant 1959; Best et al. 1996; Medeiros
and Marini 2007), including some species in
lowland habitats in Venezuela (Cruz and Andrews
1989). This peak fits within the general breeding
period of January to June described for tyrant-
flycatchers north of Peru (Fitzpatrick 2004). The
season is relatively short compared with other
passerine species in this area (e.g., Biancucci and
Martin 2008, Niklison et al. 2008, Cox and Martin
2009). Skutch (1950) found from his observations
in Central America that, in general, avian
populations at higher altitudes exhibited a more
marked and narrow breeding season than in
lowlands. Z. chrysops is thought to breed in
Colombia from April until November at the lower
elevation of 1,000 m asl (Hilty and Brown 1986).
A record in February from Eoja Province in
Ecuador also reports nesting activity by the
species at a similar altitude to Yacambu sites
(Best et al. 1996). The short season and absence of
any observations of birds re-nesting in the same
season after a successful breeding attempt indi-
cates this species is single-brooded, typical of
tyrannids (Eitzpatrick 2004).

Eggs of Z. chtysops conform to the typical egg
coloration observed in most tyrannids (Von
Ihering 1904, Skutch 1960, Wetmore 1972). They
are slightly larger than an egg of Z. acer measured
by Beebe et al. (1917) and slightly shorter and
wider than eggs of Z. vilissimus observed by
Skutch (1960). The clutch size of two is on the
lower edge of the range of 2-6 observed in the
family and the 2^ of most tropical tyrannids
(Fitzpatrick 2004).

The 17-day incubation period is long relative to
the range of 12 to 16 days known for most
temperate and tropical tyrannids, and approaches
a few unusual tropical species that have suspend-
ed pendulant nests and long incubation periods
(Fitzpatrick 2004). However, the incubation
period was near the average for 33 other non-
tyrannid species studied in Yacambii National
Park (Martin and SchwabI 2008). The 18-day
nestling period akso is in the upper range for
tyrannids, which is from 12 to 24 days with most
being 14 to 17 days (Fitzpatrick 2004). Our
observations support that mistletoe fruit in the

Gouh/ing and Martin • GOLDEN-FACED TYRANNULET BREEDING BIOLOGY

695

area (e.g.. Antidapline viscoidew, Restrepo et al.
2002, Kelly et al. 2004) comprise an important
portion of the diet of nestlings, as with Z.
vilissimus (Skutch 1960), and not only for adults
tor which this has already been observed with
others in the genus (Fjeldsa and Krabbe 1990,
Alonso and Whitney 2001, Fitzpatrick 2004).
Slow growth has been found in other tropical
species that feed fruit to their offspring (Morton
1973). A comparison of the growth rate constant
(A.) for mass with that published and summarized
for other tropical and temperate species (Ricklefs
1976, Willis et al. 1978, Oniki and Ricklefs 1981,
Starck and Ricklefs 1998, Remes and Martin
2002) reveals it is slow even for the slowest
growing of tropical tyrannids {k < 0.3). Tropical
tyrannids from Central America and Brazil had
growth rates (k) extending from a low of 0.248 in
another species {Mionectes macconnelli) that
builds enclosed nests (Willis et al. 1978) to rates
>0.4, similar to those observed for some faster
growing tyrannids breeding in North America
such as Tyrannus tyranmis {k = 0.438) and
Sayornis phoebe (k = 0.425) (Murphy 1981),
Tyrannus verticalis {k = 0.416) and T. forficatiis
(k = 0.394) (Murphy 1988), Empidonax minimus
(k — 0.499-0.505) (Briskie and Sealy 1989), and
E. oberholseri (k = 0.425) (Pereyra and Morton
2001). Other non-tyrannid passerines investigated
in Yacambii National Park also had much higher
growth rates (Niklison et al. 2008, Biancucci and
Martin 2008, Cox and Martin 2009).

Incubation temperature is a reflection of
parental attentiveness with cooler temperatures
(low attentiveness) associated with longer incu-
bation period lengths (Lyon and Montgomerie
1985, Martin 2002, Hepp et al. 2006, Martin et al.
2007). The observed 24-hr incubation temperature
for Z. chrysops is below the optimal temperature
range for embryonic growth (White and Kinney
1974, Webb 1987). However, egg temperature
was close to the expected value for a tropical
species with this length of incubation period and
nest attentiveness (Martin et al. 2007: figure 2).

Skutch (I960) considered Z. vilissimus in Costa
Rica to have high nest attentiveness for a .small
flycatcher. He observed mean on and off-bouts
during incubation of 32.5 and 12.5 min, respectively
with overall estimates of attentiveness of 67 and
72%. This indicated higher attentiveness (and shorter
off-bouts) for this congener than for Z. dvysops in
Yacambti National Park. Attentiveness for Z.
chrysops was in the middle of the range of 39 other

species studied in Yacambu (Martin and SchwabI
2008), but was lower than for most temperate
passerines (Conway and Martin 2000: appendix 1 ;
Martin 2002; Chalfoun and Martin 2007).

Z. vilissimus had higher provisioning rates than
those observed for Z. chrysops (Skutch 1960).
However, both had mean feeding rates at least
two-fold higher than other species studied in
Yacambu National Park (Fierro-Calderon and
Martin 2007, Biancucci and Martin 2008, Nikli-
son et al. 2008, Cox and Martin 2009). This could
reflect compensation for lower protein levels in
fruit fed to nestlings (Morton 1973), although
flycatchers seem to feed at higher rates than other
species (Martin et al. 2000b). Consequently,
predation levels could be influenced through
greater nest detection due to increased parental
visits (Skutch 1949; Martin et al. 2000a, b).
However, predation rates were lower during the
nestling period than during earlier periods when
parents were less active.

Nesting success for Z. chrysops was among the
lowest observed for a member of the Tyrannidae
when compared with other southern hemisphere
and neotropical species. Skutch (1985) summa-
rized data from different species in the humid
Neotropics showing that 39.7% of nests found
were successful with tyrannids exhibiting a range
of 12.1 to 50%. Mionectes oleagineus in Trinidad
had the lowest tyrannid nesting success (12.1%), a
trend observed in other small species in Trinidad
(Snow and Snow 1979). Low reproductive
estimates were observed in Suiriri islerorum
(>13.6%, mean = 20.8%) and S. ajfmis (19%)
in tropical savannah of Brazil (Lopes and Marini
2005, Franca and Marini 2009). In Costa Rica.
36% of nests of Zimmerius vilissimus were
successful and fledged chicks (Skutch 1985).
Other non-tyrannid neotropical species in Panama
were observed to have low nesting success similar
to Z. chtysops (e.g., some formicariids and a
thamnophilid species; Robinson el al. 2000).
However, nesting success was generally higher
(23-74%) and daily predation rates lower for other
tyrannid species in the Neotropics and southern
hemisphere; e.g., Argentina (Mezquida and Marone
2001, Auer et al. 2007). Puerto Rico (Toixes Baez
and Collazo 1992), Brazil (Willis et al. 1978,
Aguilar et al. 2000, Medeiros and Marini 2007), and
Panama (Robinson et al. 2000, Dyrcz 2002).

The extremely poor nesting success of Z.
chrysops reflects high predation pressure, and
may reflect the tendency for this species to nest

696

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, Deceniber 2010

along forest edges where predation levels are high
(Sdderstrom 1999). Multiple re-nest attempts
would be expected to occur to compensate for
high predation (Martin 1995, Fitzpatrick 2004),
but the short breeding season (i.e.. Fig. 2) and our
observations suggest that only 2-3 nesting
attempts are made.

The small clutch size of Z. chrysops with a
relatively long incubation period and slow nest-
ling growth suggest this species is at the ‘slow’
end of the slow-fast life history gradient (Martin
1996, 2004). The solely neotropical distribution,
lack of morphological specialization (Traylor
1977), and indication that only conservative
differences in nest architecture occur within the
genus, place importance on investigating repro-
ductive parameters of the diversity of Zimmeriiis
species. Phenotypic similarity or dissimilarities
would be valuable for understanding the finer
scale processes influencing reproduction and
broader life history variation within tyrannids;
e.g., localized predation rates imposed upon
otherwise phylogenetically close and similar
species (Skutch 1949, 1985; Bosque and Bosque
1995; Martin 1995; Martin et al. 2000a).

ACKNOWLEDGMENTS

We thank L. A. Biancucci, Karolina Fieno-Calderon, A.
M. Niklison. M. J. Foguet. A. A. Majewska, and R. A.
Ruggera for discussions and assistance in the field. We also
thank John Kanowski, Elena Arriero. Marline Maron. and
C. A. McAlpine for help with the manu.script. This study
was made possible in part by support under NSF grants
DEB-()543178 and DEB-0841764 to T. E. Martin. Permit
numbers were DM/0000237 from FONACIT. PA-INP-005-
2004 from INPARQUES, and 01-03-03-1 147 from Mini.s-
terio del Ambiente. Specific equipment identities are
provided to aid specific methods and do not represent an
endorsement of these companies by USGS.

LITERATURE CITED

Aguilar. T. M., M. Maldonado-Coelho, and M. A.
Marini. 2000. Nesting biology of the Gray-hooded
Flycatcher {Mionectes niflveiitri.s). Ornitologia Neo-
tropical 1 1 :223-230.

Alonso. J. A. and B. M. Whitney. 2001. A new
Zimmerin.s tyrannulet (Aves: Tyrannidae) from white
sand forests of northern Amazonian Peru. Wilson
Bulletin I 13:1-9.

Auer. S. K.. R. D. Bassar. J. J. Fontaine, and T. E.
Martin. 2007. Breeding biology of passerines in a
subtropical montane forest in northwestern Argentina.
Condor 109:321-333.

Beebe. W.. G. I. Hartley, and P. G. Howes. 1917.
Tropical wildlife in British Guiana. Volume 1. New
York Zoological Society, New York, USA.

Best, B. J., M. Checker. R. M. Thewlis, A. L. Best, and
W. Duckworth. 1996. New bird breeding data from
southwestern Ecuador. Omithologia Neotropical 7:69-
73.

Biancucci, L. A. and T. E. Martin. 2008. First description
of the breeding biology and natural history of the Ochre-
breasted Brush Finch (Atlapeles seminifus) in Vene-
zuela. Wilson Journal of Ornithology 120:856-862.

Bosque, C. and M. T. Bosque. 1995. Nest predation as a
selective factor in the evolution of development rates
in altricial birds. American Naturalist 145:234-260.

Briskie. j. V. and S. G. Sealy. 1989. Determination of
clutch size in the Least Elycatcher. Auk 106:269-278.

Chalfoun, a. and T. E. Martin. 2007. Latitudinal
variation in avian incubation attentiveness and a test
of the food limitation hypothesis. Animal Behaviour
73:579-585.

Cisneros-Heredia, D. a. 2006. Notes on breeding,
behaviour and distribution of some birds in Ecuador.
Bulletin of the British Ornithologists Club 126:153-
164.

COLLIAS, N. E. AND E. C. COLLIAS. 1984. Nest building and
bird behavior. Princeton University Press, Princeton,
New Jersey, USA.

Conway, C. J. and T. E. Martin. 2000. Evolution of
passerine incubation behavior: inlluence of food,
temperature, and nest predation. Evolution 54:670-
685.

Cox. W. A. AND T. E. Martin. 2009. Breeding biology of
the Three-Striped Warbler in Venezuela: a contrast
between tropical and temperate parulids. Wilson
Journal of Ornithology 121:667-678.

Cruz, A. and R. W. Andrews. 1989. Observations on the
breeding biology of passerines in a seasonally flooded
savanna in Venezuela. Wilson Bulletin 101:62-76.

Dyrcz, a. 2002. Breeding ecology of the Social (Myioze-
tetes .'iiniili.s) and Rusty-margined (M. cayanensis)
flycatchers at Barro Colorado Island, Republic of
Panama. Ornitologia Neotropical 13:143-151.

Fierro-CalderOn, K. and T. E. Martin. 2007. Repro-
ductive biology of the Violet-chested Hummingbird in
Venezuela and comparisons with other tropical and
temperate hummingbirds. Condor 109:680-685.

Fitzpatrick, J. W. 2004. Family Tyrannidae (Tyrant
Flycatchers). Pages 170-257 in Handbook of the birds
of the world. Volume 9. Cotingas to pipits and
wagtails (J. del Hoyo. A. Elliott, and D. A. Christie,
Editors). Lynx Edicions, Barcelona, Spain.

FjeldsA, j. and N. Krabbe. 1990. Birds of the High Andes.
Zoological Museum. University of Copenhagen. Den-
mark.

Franca, L. F. and M. A. Marini. 2009. Low and
variable reproductive success of a neotropical tyrant-
flycatcher. Chapada Flycatcher (Siiiriri islerorinn).
Emu 109:265-269.

Haver.schmidt. F. 1954. The nesting of the Ridgway
Tyrannulet in Surinam. Condor 56:139-141.

Hensler, G. L. and j. D. Nichol.s. 1981. The Mayfield
method of estimating nesting success: a model,
cstimalors and simulation results. Wilson Bulletin
93:42-53.

Gouhliiii' and Marlin • GOLDEN-FACED TYRANNULET BREEDING BIOLOCJY

697

Hepp, G. R.. R. a. Kennamer. and M. H. Johnson. 2006,
Maternal effects in Wood Ducks: incubation temper-
ature influences incubation period and neonate phe-
notype. Functional Ecology 20:307-314.

Hilty, S. J. 2003. Birds of Venezuela. Second Edition,
Princeton University Press, Princeton, New Jersey,
USA.

Hilty, S, J, and W. L. Brown. 1986. A guide to the birds
of Colombia. Princeton University Press, Princeton,
New Jersey, USA.

Johnson, D. H. 2007. Estimating nest success: a guide to
the methods. Studies in Avian Biology 34:65-72.

Kelly, D, L., G. O'Donovan, J. Feehan, S. Murphy, S.
O. Drangeid, and L. Marcano-Berti. 2004. The
epiphyte communities of a montane rain forest in the
Andes of Venezuela: patterns in the distribution of the
flora. Journal of Tropical Ecology 20:643-666.

Lanyon, W. E. 1988. A phylogeny of the thirty-two genera
in the Elaenia assemblage of tyrant flycatchers.
American Museum Novitates 2914:1-57.

Lopes, L. E. and M. A. Marini. 2005. Low reproduc-
tive success of Campo Suiriri {Siiiriri ajfmis) and
Chapada Flycatcher (S. isleronim) in the central
Brazilian Cerrado. Bird Conservation International
15:337-346.

Lyon, B. E. and R. D. Montgomerie. 1985. Incubation
feeding in Snow Buntings: female manipulation or
indirect male parental care? Behavioral Ecology and
Sociobiology 17:279-284.

Marchant, S. 1959. The breeding season in S.W. Ecuador.
Ibis 101:137-152.

Martin, T. E. 1995. Avian life history evolution in relation
to nest sites, nest predation and food. Ecological
Monographs 65:101-127.

Martin, T. E. 1996. Life history evolution in tropical and
south temperate birds: what do we really know?
Journal of Avian Biology 27:263-272.

Martin, T. E. 2002. A new view for avian life history
evolution tested on an incubation paradox. Proceed-
ings of the Royal Society of London, Series B
269:309-316.

M.artin. T. E. 2004. Avian life-history evolution has an
eminent past: does it have a bright future? Auk
121:289-301.

Martin, T. E. and C. K. Ghalambor. 1999. Males feedini^
females during incubation. 1. Required by microcli-
mate or constrained by nest predation? American
Naturalist 153:131-139.

Martin, T. E. and H. Schwabl. 2008. Variation in maternal
effects and embryonic development rates among
passerine species. Philo.sophical Transactions of the
Royal Society of London, Series B 363:1663-1674.

Martin, T. E., J. Scott, and C. Menge. 2000a. Nest
predation increases with parental activity: .separating
nest site and parental activity effects. Proceedings of
the Royal Society of London. Series B 267:2287-
2293.

Martin, T. E., P. R. Martin, C. R. Olson, B. J.
Heidinger, and j. j. Fontaine. 2000b. Parental care
and clutch sizes in North and South American birds.
Science 287:1482-1485.

Martin, T. E., S. K. Auer, R. D. Bassar, A. M. Niklison,
AND P. Lloyd. 2007. Geographic variation in avian
incubation periods and parental influences on embry-
onic temperature. Evolution 61:2558-2569.

Mayfield, H. 1961. Nesting success calculated from
exposure. Wilson Bulletin 73:255-261.

Mayfield, H. F. 1975. Suggestions for calculating nest
success. Wilson Bulletin 87:456^66.

McGowan, K. J. 2004. Introduction: the world of birds.
Pages 1.1-1.114 in Handbook of bird biology (S.
Podulka. R. Rohrbaugh Jr., and R. Bonney. Editors).
Cornell Laboratory of Ornithology, Ithaca, New York.
USA.

Medeiros, R. C. S. and M. A. Marini. 2007. Biologia
reprodutiva de Elaenia chiriqnensis (Lawrence) ( Aves,
Tyrannidae) em Cerrado do Brasil Central. Revista
Brasileira de Zoologia 24:12-20.

Mezquida, E. T. and L. Marone. 2001. Factors affecting
nesting success of a bird assembly in the central Monte
Desert, Argentina. Journal of Avian Biology 32:287-296.

Mobley, J. A. and R. O. Prum. 1995. Phylogenetic
relationships of the Cinnamon Tyrant, Neopipo
cinnamomea, to the tyrant flycatchers. Condor
97:650-662.

Morton, E. S. 1973. On the evolutionary advantages and
disadvantages of fruit eating in tropical birds.
American Naturalist 107:8-22.

Murphy, M. T. 1981. Growth and aging of nestling Eastern
Kingbirds and Eastern Phoebes. Journal of Field
Ornithology 52:309-316.

Murphy, M. T. 1988. Comparative reproductive biology of
kingbirds (Tyrannus spp.) in eastern Kansas. Wilson
Bulletin 100:357-376.

Nadkarni, N. M. and T. j. M.atelson. 1989. Bird use of
epiphyte resources in neotropical trees. Condor
91:891-907.

Nice, M. M. 1954. Problems of incubation periods in North
American birds. Condor 56:173-197.

Niklison, A. M., J. I. Areta, R. A. Ruggera. K. L.
Decker. C. Bosque, and T. E. Martin. 2008. Natural
history and breeding biology of the Rusty-breasted
Antpitta [Grallariciila fernigineipectiis). Wilson Jour-
nal of Ornithology 120:345-352.

Ohlson, j., j. FjeldsA, .and P. G. P. Ericson. 2008. Tyrant
flycatchers coming out in the open: phylogeny and
ecological radiation of Tyrannidae (Aves, Passeri-
formes). Zoologica Scripta 37:315-335.

Oniki, Y. and R. E. Ricklefs. 1981. More growth rates of
birds in the humid New World tropics. Ibis P3-349-
353.

Palmer, R. S. (Editor). 1962. Handbook of North
American birds. Yale University Press. New Haven.
Connecticut. USA.

Pereyra, M. E. and M. L. Morton. 2001. Nestling growth
and thermoregulatory development in subalpine Dusky
Flycatchers. Auk 118:116-136.

Preston, F. W. 1953. The shapes of bird eggs Auk
70:160-182.

RemeS, V. and T. E. Martin. 2002. Environmental
influences on the evolution of growth and develop-
ment rates in pas.serines. Evolution 56:2505-2518.

698

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 4, December 2010

Restrepo, C., S. Sargent, D. J. Levey, and D. M.
Watson. 2002. The role of vertebrates in the
diversification of New World mistletoes. Pages 83-
98 in Seed dispersal and frugivory: ecology, evolution,
and conservation (D. J. Levey, W. R. Silva, and M.
Galetti, Editors). CAB International, Wallingford,
United Kingdom.

Rheindt, F. E., j. a. Norman, and L. Christidis. 2008.
Phylogenetic relationships of tyrant-flycatchers ( Aves:
Tyrannidae), with an emphasis on the elaeniine
assemblage. Molecular Phylogenetics and Evolution
46:88-101.

Ricklefs, R. E. 1967. A graphical method of fitting
equations to growth curves. Ecology 48:978-983.

Ricklefs, R. E. 1976. Growth rates of birds in the humid
New World tropics. Ibis 118:179-207.

Ridgely, R. S. and G. Tudor. 1994. The birds of South
America, Volume 2. The suboscine passerines.
University of Texas Press, Austin, USA.

Robinson, W. D., T. R. Robinson, S. K. Robinson, and J.
D. Brawn. 2000. Nesting success of understory' forest
birds in central Panama. Journal of Avian Biology
31:151-164.

Sheldon. F. H. and D. W. Winkler. 1999. Nest
architecture and avian systeinatics. Auk 1 16:875-877.

Skutch, a. F. 1949. Do tropical birds rear as many young
as they can nourish? Ibis 91:430^55.

Skutch, A. F. 1950. The nesting seasons of Central
American birds in relation to climate and food supply.
Ibis 92:185-222.

Skutch, A. F. 1960. Life histories of Central American
birds. Volume 2. Pacific Coast Avifauna Number 34.

Skutch, A. F. 1985. Clutch size, nesting success, and
predation on nests of neotropical birds, reviewed.
Ornithological Monographs 36:575-594.

Snow, B. K. and D. W. Snow. 1979. The Ochre-bellied
Flycatcher and the evolution of lek behavior. Condor
81:286-292.

SoderstrOm, B. 1999. Artificial nest predation rates in
tropical and temperate forests: a review of the effects
of edge and nest site. Ecography 22:455-463.

SPSS. 2006. Version 15.0.0. SPSS Inc., Chicago, Illinois,
USA.

Starck, j. M. and R. E. Ricklefs. 1998. Data set of avian
growth parameters. Pages 381-423 in Avian growth
and development (J. M. Starck and R. E. Ricklefs,
Editors). Oxford University Press, New York, USA.

Stutchbury, B. j. M. and E. S. Morton. 2001.
Behavioral ecology of tropical birds. Academic Press,
San Diego, California, USA.

Torres Baez, P. and J. A. Collazo. 1992. Biologia
reproductiva y datos preliminares sobre patrones de
crecimiento del Jui de Puerico {Myicircinis antillarum)
en Cabo Rojo, Puerto Rico. Caribbean Journal of
Science 28:30-37.

Traylor Jr., M. A. 1977. A classification of the tyrant
Ilycatchers (Tyrannidae). Bulletin of the Museum of
Comparative Zoology 148:129-184.

Von Ihering, H. 1904. The biology of the Tyrannidae with
respect to their systematic arrangement. Auk 21:31 3-322.

Weathers, W. W. and K. A. Sullivan. 1989. Nest
attentiveness and egg temperature in the Yellow-eyed
Junco. Condor 91:628-633.

Webb, D. R. 1987. Thermal tolerance of avian embryos: a
review. Condor 89:874-898.

Wetmore, a. 1972. Birds of the Republic of Panama. Part 3.
Smithsonian Institution Press, Washington, D.C., USA.

White, F. N. and J. L. Kinney. 1974. Avian incubation.
Science 186:107-1 15.

Willis, E. O.. D. Wechsler, and Y. Oniki. 1978. On
behavior and nesting of McConneH’s Flycatcher
(Pipromorpha macconnelli): does female rejection
lead to male promiscuity? Auk 95:1-8.

Zyskowski, K. and R. O. Prum. 1999. Phylogenetic
analysis of the nest architecture of neotropical
Ovenbirds (Furnariidae). Auk 1 16:891-91 1.

The Wilson Journal of Oniithology 122(4):699-7()5. 2010

HABITAT TYPE IS RELATED TO NEST MASS AND FLEDGING
SUCCESS OF ARCTIC WARBLERS

JULIE C. HAGELIN,'’^ MARGARET L. PERRY,' BEN S. EWEN-CAMPEN,'-^
DEREK S. SIKES,- AND SUSAN SHARBAUGH^

ABSTRACT. Relatively little is known about Arctic Warblers (Phylloscopus borealis) that breed in central Alaska. We
monitored Arctic Warbler populations in two adjacent but distinct habitat types in central Alaska (high elevation, 'open
shrub and lower elevation, ‘dense shrub’). We collected 95 nests over three breeding seasons to learn more about nest-
building behavior, nest mass, composition, fledging success, and nest parasites. Females were the sole builders of ground
nests, which were primarily comprised ot moss, grass, and a lining of moose (Aices alces) hair. Dry weight of nests was
20 g. but differed up to ~3-fold within each habitat type each season. Nests from open shrub habitats were more massive
and contained less moose hair lining than nests in dense shrub. Open shrub nests fledged more young during the most
productive breeding season. We report the first record of the parasitic blowfly Protocalliphora tundrae in Arctic Warbler
nests and for Alaska. Blowfly parasitism (55% of nests with hatchlings) was similar in both habitat types and did not
correlate with fledging success, or nest mass. Nests with greater amounts of moss tended to have lower levels of blowfly
infestation. Received 26 September 2009. Accepted 10 June 2010.

The Arctic Warbler (Phylloscopus borealis) is
an understudied old-world warbler that winters in
southeast Asia and has an expansive subarctic
breeding range that extends from Norway through
Siberia into central Alaska (Tucker 1949, Price
and Beck 1989, Lowther and Sharbaugh 2008).
Little is known about its cryptic ground-nesting
habits throughout its range. Alaska is the only
location in North America where Arctic Warblers
(Fig. 1 A) breed and these populations are consid-
ered a separate subspecies (P. b. kennicotti)
(Lowther and Sharbaugh 2008, Reeves et al.
2008).

Documentation of Arctic Warbler nests in
Alaska prior to our observations included a report
on 1 1 nests near Nome (Price and Beck 1989) and
two nests in Denali National Park by Murie
(1956). Our objectives are to present detailed
information about: ( 1 ) nest construction, (2)
orientation, (3) composition, (4) fledging success,
and (5) nest parasites in two adjacent, but distinct
habitat types (high elevation, ‘open shrub’ and
low elevation, ‘dense shrub’) in a central Alaskan
population of P. b. kennicotti.

' Swarthmore College Department of Biology, Swarth-
more, PA 19081, USA.

"University of Alaska, Museum of the North, Fairbanks,
AK 99775, USA.

■’Alaska Bird Observatory, Fairbanks, AK 99701, USA.
■’Department of Organismic and Evolutionary Biology,
Harvard University, Cambridge, MA 02138, USA.

’Corresponding author; e-mail:
jhageli 1 @swarthmore.edu

METHODS

Study Site. — We studied a breeding population
of Arctic Warblers between 2004 and 2006 in the
Tangle Lakes region of the Denali Highway in
interior Alaska (63° N, 146° W). All data were
collected from four 10-ha plots, each 150 m from
the Denali Highway at mileposts 23, 26, 30, and
33, respectively. Pairs of plots represented two
distinct habitat types of the region. ‘Open shrub’
plots were at higher elevation (1,121-1,154 m)
and included north and east-facing slopes. This
habitat type contained meadow-like clearings
suiTounded by 1-2 m tall willow shrubs (Sali.x
spp.) and dominated by gramminoids (Deschamp-
sia caespitosa and Carex spp.), lupines (Lupinus
arcticus), and burnet (Sanguisorba spp.). Nests
were collected from open shrub plots in all 3 years.
Dense shrub plots were at lower elevations
(910-1,012 m) with flat ground or a gradual, east-
facing slope. This habitat type was dominated by
dwarf birch (Betula nano) and willow. Nests were
collected in dense shrub habitats only in 2004 and
2005, as birds typically nested in these locations
at much lower densities.

Nest Monitoring, Collection, and Composi-
tion.—VJc located most nests by systematically
searching the ground in areas where we observed
singing males. Each nest was monitored at least
once every 4 days through laying, hatching, and
rearing of young. Daily nest checks began 2 days
prior to the expected fledging date to ascertain the
number of young successfully fledged.

Observations of nest building were made each
season and we collected each nest once fledging

699

700

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

KIG. I. (A) Color-banded adult Arctic Warbler (Phylloscopiis borealis keniiecolli). Males and females are
monomorphic. (Photograph by Ron Teel). (B) Opening of Arctic Warbler nest in the side of a tussock in open shrub
habitat. (Photograph by Sue Guers). (C) Ne.st with fledglings associated with willows in dcn.se shrub habitat. (Photograph
courtesy of Alaska Bird Ob.servatory).

Ilagcliii Cl al. • ARCTIC WARBLER NEST MASS

701

was complete (between 20 Jul and I Aug). We
first recorded nest orientation (in 2005 and 2006
only) by obtaining a compass bearing from a stick
inserted straight into the nest opening prior to
removing the nest from the ground. Nests are a
domed mass of material that can be removed by
lifting up the sides and the bottom, once the edge
of the nest is located within the suiTounding
tundra. We carefully removed non-nest material
that clung to the outside of the nest during
extraction and retrieved any bits of nest material
left in the depression. Each nest was placed in a
separate plastic collection container for transport
to the laboratory, then transferred to a paper sack
and dried overnight in an oven at 135° C.

Dry nest mass was measured using an analyt-
ical balance. Two key components of nests were
measured each year: nest mass and mass of moose
{Alces dices) hair, which comprised the lining of
many nests. Our relatively large sample of nests
from open shrub habitat (n = 80) allowed us to
examine differences across years. Twenty-six
nests from open shrub habitat were dissected
further in 2006 to quantify the mass of moss and
grass, two key components of nest structure. We
calculated the percent composition of hair, moss,
or grass as a proportion of nest mass.

A snowstorm on 21 June 2006 caused birds to
abandon the eight nests (and clutches) that we had
located at that point in the season. However,
Arctic Warblers began re-nesting after snow
began to melt (on 23 Jun 2006) and we
subsequently located 22 new nests. We collected
all 30 nests, including the eight that had been
abandoned. This sample enabled us to examine
any differences between nests built before and
after the storm (e.g., With and Webb 1993).

Nest Parasites, Incidental Arthropods, and
Fledging Success. — We removed, identified, and
counted all puparia of an abundant nest parasite
(a blowfly) after nests were dried but before
they were weighed. We also identified incidental
arthropods. Puparia are the rigid outer shells of
developing blowfly larvae. The larval stage
obligately feed on the blood of nestlings (Sab-
rosky et al. 1989). Counting puparia is a
conservative measure of parasitism, because it
avoids double-counting flies that have successful-
ly pupated into adults. All nests that failed prior to
hatching were excluded from analyses (n = 8 that
failed during the Jun 2006 storm, n — 3
depredated or abandoned during other years), as
puparia only occur after blowfly larvae have fed

on nestlings. All arthropod specimens were
deposited in the University of Alaska Museum
Insect Collection.

Statistical Analyses. — The mass of nest com-
ponents and parasite counts exhibited strongly
non-normal distributions, which made parametric
statistics inappropriate. All data are reported as
medians and 25-75% interquartile ranges [IQR].
We used Wilcoxon rank sum tests to examine
differences in nest construction and fledging
success between habitat types. We used Rayleigh
tests to ascertain whether orientation of nest
openings deviated from a random distribution
(Wilkie 1983). We examined differences between
years within each habitat type using Kruskal-
Wallis ANOVA. We used Spearman's rho (p)
tests to assess non-parametric correlation(s),
including relationship(s) between nest mass,
different nest components, fledging success,
fledge date, and number of puparia per nest.

RESULTS

Nesting Behavior. — Our observations are based
on 95 nests (/? = 80 from open shrub habitat, n =

1 5 from dense shrub habitat) collected over 3 years
(n = 24 in 2004, n = 41 in 2005, n = 30 in 2006).
All nests were built on the ground with domed
roofs and circular side entrances (Fig. IB), and at
times incorporated willow stems (Fig. 1C). Each
nest consisted of an outer layer of moss and coarse
grass with an inner layer of finer materials
including grasses and moose guard hairs, which
formed a tightly-compressed, disc-shaped mat
upon which eggs were laid.

We witnessed construction of three nests in
early June 2005. We observed two color-marked
pairs at two nests for ~3 hrs. Males did not caiTy
nesting material or enter the nest during construc-
tion. However, males flew to and from nests in
close proximity to the female that was gathering
nest material. The male perched above the nest
while the female entered and built the structure
itself. Females typically gathered materials within
100 m of the nest; we did not observe travel
between study plots. Chasing and physical
interactions between the resident male and other
males occurred frequently in the air directly above
the nest during construction.

The storm on 21 June 2006 blanketed the study
area with 30 cm of snow which persisted for
5 days; eight nests identified at that stage of the
season failed. Only one individual at each of six
nests was color-marked. Two marked birds

702

THE WILSON JOURNAL OF ORNITHOLOGY • Voi 122, No. 4. December 2010

( 1 male and 1 unknown, each with an unmarked
mate) re-nested in the study area after snowmelt.
One nest was depredated, but the other fledged
young. We only included the initial nests from the
two marked birds in our data set, as re-nests
lacked statistical independence.

Nest Construction. — Nests differed noticeably
between habitats. Those in open shrub plots were
—30% more massive (/? = 68, median [IQR] =
20.7 g [17.3-24.0]) than those in dense shrub {n =
13, median [IQR] = 15.2 g [13.4-20.0]; df = 1,
X' = 6.96, P = 0.0083). Nests in dense shrub
contained nearly four times the amount of moose
hair (n = 13, median [IQR] = 1.1 g [0.4-1. 3])
than nests in open shrub (n = 68, median [IQR] =
0.3 g [0.0-0.8]; df = 1, x' = 6.18, P = 0.013).

We measured differences in nest mass up to
—3-fold every season (range: 2. 0-3. 5 fold) within
each habitat. There was no coirelation between
nest mass and hair mass in either habitat type
(open shrub: n = 68, Spearman’s p = —0.15, P =
0.22; dense shrub: n = 13, Spearman’s p =
— 0.02, P = 0.95). Orientation of nest entrances
did not differ from a random distribution (open
shrub plots: n — 62, mean vector (p) = 48°,
vector length (r) = 0.13, Rayleigh test Z = 0.82,
P = 0.82; dense shrub plots: n = 7, p = 228°, r =
0.40, Z = 2.02, P = 0.13). Entrances of nests built
after the 2006 snowstorm were also randomly
oriented (n = 20, p = 42°, r = 0.26, Z = 1.0, P =
0.37). Mass of nests and moose hair did not differ
between pairs of plots within each habitat type
(df = 1, 0.055 < X" ^ 1-85, 0.17 < P < 0.82).

Dissection of 26 nests from open shrub habitat
in 2006 indicated moss and grass accounted for
most of the nest mass (83.4%), and that each was
present in similar proportions (median [IQR] =
45.5% [31.0-54.4] and 37.9% [35.9-55.1], re-
spectively). Moose hair, which was primarily
incorporated into the nest cup, comprised 2.0%
[0.7^. 2] of nest mass. Any remaining mass
(—15%, or —3.0 g) resulted from small pieces
of leaves, twigs, and feather dander. Nest mass
also correlated positively with moss and grass
contained in the nest structure (moss: n = 24,
Spearman’s p = 0.62, P = 0.0013; grass: n = 24,
Spearman’s p = 0.38, P = 0.065).

Nests from open shrub habitat in 2006 were
significantly lighter (by 3.2 g or 14.5%) than
tho.se from the previous two seasons (median
(IQR I 2006: n = 28, 18.9 g [13.9-21.9], 2005:
n = 33, 22.1 g (17.4-28.21, 2004: n = 7, 22.1 g
[20.8-25.7]; df = 2, x^ = 8.13, P = 0.017). The

analysis pooled pre-storm (n = 8) and post-storm
{n = 20) nests from 2006, as all comparisons (of
nest mass, mass of moose hair, moss, grass) failed
to reach statistical significance (df = 1, 0. 16 < X‘
< 0.44, 0.51 < P < 0.68). Moose hair in nests
from open shmb did not differ over 3 years (df =
2, x“ = 2.23, P = 0.33). Nest mass and moose
hair in dense shrub did not differ over 2 years
(2004, 2005; both analyses: df = 1, x" = 0.18,
P = 0.67).

Fledging Success, Fledge Date, and Nest
Construction. — The median number of young
fledged in open shrub habitat differed each year
(df = 2, x" = 24.33, P < 0.0001). Birds were
most successful in 2005 [n = 29, median [IQR] =
6 young [5-6]), followed by 2004 {n = 15,
median [IQR] = 4 young [4-5], and the 2006
storm (n = 20, median [IQR] = 4 young [3-5];
pairwise comparisons: df = 1, 4.39 < — 22.74,

0.0001 < P < 0.036). Open shrub habitat had
significantly greater fledging success in 2005 than
dense shrub habitat (/? = 7, median [IQR] = 3
young [0-6], df = 1, X' = 5.46, P = 0.019). The
two habitats did not differ in 2004 (dense shrub: n
= 8, median [IQR] = 4 young [3-5], df = 1, X“ ~
0.018, P = 0.89). We detected no yearly
differences in fledging success in dense shrub
habitat (df = 1, X“ = 0.22, P = 0.64). The two
plots within each habitat type also did not differ
(df = 1, open shrub: X“ = 0.23, P = 0.62; dense
shrub: X' = 0.032, P = 0.86).

Median fledge date in open habitat differed
significantly between years {n = 14, 2004: 20
July; n = 32, 2005: 22 July; n = 20, 2006: 27
July; df = 2, x^ = 34.49, P < 0.001). The storm
disrupted breeding in 2006 and resulted in a
median fledge date that was 5-7 days later than in
previous years (df = 1, X“ ~ 13.48, P < 0.001).
Median fledge dates in dense shrub habitat were
identical to those reported for 2004 and 2005.

Nest mass in open shrub habitat coirelated
positively with three measures of reproductive
success when all years were combined (n = 60,
young fledged per nest: Spearman’s p = 0.28,
P = 0.039; young hatched: Spearman’s- p = 0.33,
P = 0.012; clutch size: Spearman’s p = 0.30, P =
0.024). Nest mass al.so correlated negatively with
Hedge date (n = 60, Spearman’s p = —0.26, P =
0.05). The relationships were not detectable,
however, when the disrupted 2006 .season was
excluded (/? = 44, 0.03 < Spearman’s p < 0.09,
0.60 < P < 0.86), or when each year was
analyzed separately (7 < /? < 33, —0.48 <

I/agclin et al. • ARCTIC WARBLER NEST MASS

703

Spearman’s p < 0.15, 0.27 < P < 0.74). Hair
mass in open shrub nests did not correlate with
any measure of reproductive success (/? = 60,

— 0.11 < Spearman’s p < 0.12, 0.35 < f <
0.49). Nest mass, hair mass, reproductive success
or fledge date of nests from dense shrub had no
detectable relationships (7 < // < 14, 0.078 <
Spearman’s p < 0.20, 0.52 < P < 0.80).

Nest Parasites, Incidental Arthropods, and
Fledging Success. — We extracted 1,241 parasitic
blowfly puparia and 187 adults from Arctic
Warbler nests. The blowfly was identified as
Protocalliphora tundrae. Incidental arthropods
included three families of beetles (Order: Cole-
optera) (Leiodidae: Catops alpinns, Carabidae:
Patrobus foveocollis, Staphylinidae: Quediiis
brunnipennis).

Forty-six of 84 nests that successfully hatched
young (55%) contained blowfly puparia, and the
proportion of parasitized nests was similar in both
habitats (open shrub habitat: 39 of 70 [56%],
dense shrub: 7 of 14 [50%]). Most nests contained
few puparia (median [IQR] = 1 [0-22]). The
greatest number of puparia recorded in one nest
was 157 (from open shrub habitat in 2006).
Blowfly parasitism did not differ by habitat type
(open shrub plots: n = 69, median [IQR] = 1 [0-
23]; dense shrub plots: n = 14, median [IQR] = 1
[0-13]; df = l,f = 0.29, P = 0.59) or by study
plot (df = 3, = 3.76, P = 0.29). There was no

correlation between puparia count per nest and
fledging success (n = 81, Spearman’s p =

— 0.056, P = 0.62) or nest mass (n = 73,
Spearman’s p = 0. 12, F = 0.29).

Parasite incidence per nest in open shrub
habitat differed significantly between years with
nests in 2004 containing few puparia (n = 15,
median [IQR] = 0.0 [0. 0-0.0]) compared to those
in 2005 (n = 33, median [IQR] = 4 [0-28] and
2006 (/? = 21, median [IQR] = 12 [0-28]; df = 2,
X" = 15.74, P = 0.0004). The mass of moss
contained in open shrub nests tended to be
negatively associated with puparia count (/? =
17, Spearman’s p = —0.45, P = 0.07). Nests
from dense shrub plots had no correlations
between nest characteristics (mass, hair), puparia
count, or young fledged (12 < /; < 13, —0.07 <
Spearman’s p < 0.26, 0.40 < P < 0.83).

DISCUSSION

Nesting Behavior and Nest Composition. — The
95 nests we examined add new details to our
understanding of Arctic Warblers that breed in

central Alaska. First, consistent with other studies
of Phyllo.scopus warblers, including other popu-
lations of P. borealis (Barlein 2006, Clement
2006), females are the primary, if not sole, nest
builder while males sing nearby and defend the
site. Second, nests weigh ~20 g, and over 83% of
nest mass is attributable to moss and grass. Our
observations of nest location, shape, and lining of
animal hair are consistent with other reports from
Alaska (Murie 1956, Price and Beck 1989), and
other Phylloscopus species (Bi 2004). Third, our
data indicate nest openings were randomly
oriented. Exposure can influence a nest’s micro-
climate (With and Webb 1993), and several
ground-nesting species exhibit clear patterns with
regard to orientation, including birds that build
domed nests or breed at high latitudes (e.g..
Burton 2007, Long et al. 2009). The position of
openings of Arctic Warbler nests may be based on
other parameters that we did not measure,
including shelter near the nest site. It is also
possible orientation is opportunistic given the
uneven topography of tundra hummocks.

Nest Construction. — Nests in open shrub habitat
were more massive and contained less moose hair
lining than those in dense shrub. Two non-
exclusive explanations may account for this result.
We hypothesize that local availability of nest
materials differs between habitat types. For
example, we often saw moose hair snagged on
branches in dense shrub habitat, but moose were
less common and had less opportunity (fewer
shrubs of adequate height) to snag hair in open
shrub habitat. The extensive breeding range and
variety of nesting habitats of Phylloscopus
borealis (Lowther and Sharbaugh 2008) are also
consistent with an opportunistic focus upon local
resources during nest building (Barlein 2006).

An inverse relationship between nest mass and
moose hair is suggestive of a thennal trade-off
between the two habitat types. Pilot studies
conducted on nests in situ, but after fledging, failed
to detect any positive correlation between nest mass
or the mass of moose hair lining and the insulation
quality of nests (/? = 41 nests, unpubl. data). We
believe this warrants additional research, prefeira-
bly during incubation or brooding, as thennal
properties likely result from a combination of
factors, including amount and kind(s) of nest
material, local topography of the breeding site,
and parental behavior (Skowron and Kern 1980).

Fledging Success, Fledge Date, and Nest
Construction.— Htih'xm type was associated with

704

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4, December 20IU

a measurable difference in fledging success. Birds
in open shrub plots during 2005 fledged twice the
number of young as those in dense shrub. Open
shrub plots at our study site also had a greater nest
density (Sharbaugh et al. 2007), which supports
the conclusion that open shrub habitat is generally
more favorable to Arctic Warblers.

The snow storm on 21 June 2006 interrupted
breeding, as the median fledge date in 2006
occurred 5-7 days later than previous years. Nests
were significantly lighter only in 2006, suggesting
birds built somewhat smaller nests due to the
shortened season caused by the storm. We
detected no difference, however, between nests
built before and after the 2006 storm, possibly due
to our small sample of pre-storm nests {n = 8).

It is unclear why Arctic Warblers exhibit such
striking (~3-fold) differences in nest mass within
each habitat type. Nest mass coixelated positively
with fledging success and negatively with fledge
date in open shrub habitat when all years were
combined. These patterns were no longer evident,
however, when the disrupted 2006 season was
removed, or in smaller (yearly) samples. Nest
mass in other avian species can indicate quality of
the female nest builder (e.g., Soler et al. 1998,
Mainwaring et al. 2008), or enhanced parenting
effort (e.g., Szentirmai et al. 2005, Broggi and
Senar 2009).

Nest Parasites, Incidental Arthropods, and
Fledging Success. — The puparia collected repre-
sent the first Alaska record of the avian blowfly
Protocalliphora tundrae. Our observations in-
crease the number of known avian hosts of P.
tundrae from two (Snow Bunting [Plectrophena.x
nivali.x] and Savannah Sparrow [Passerculus
sandwichensi.s]) to three, all of which are
ground-nesting birds (Sabrosky et al. 1989). P.
tundrae is considered a rare tundra inhabitant of
northern Canada and the Yukon (Sabrosky et al.
1989). Incidental arthropods included Catops
alpinus, a dung and carrion scavenging beetle
known to occur in alcid nests (Perreau 1998, Peck
and Cook 2002), and two other beetles that are
generalist predators (Patrobus foveocollis and
Quedius hrunnipennis). Q. hrunnipennis in Alas-
kan tundra is unusual, as it has been reported to be
a mature forest specialist from Alberta (Pohl et al.
2007).

Blowfly parasitism of young varied annually
but did not correlate with fledging success. This
result is consistent with both observational and
experimental investigations of other Protocalli-

phora species and passerine hosts (Roby et al.
1992, Wittmann and Beason 1992). One recent
investigation, however, suggests blowflies con-
tribute to mortality and reduced movement soon
after fledging (Strebyl et al. 2009).

There are at least two different, non-exclusive
explanations for the negative relationship between
blowfly puparia and moss mass that warrant
future study. First, moss may create a cool
microhabitat. Cool environments can cause Pro-
tocalliphora to delay or fail during larval
development (Bennett and Whitworth 1991).
Second, moss may cause nests to retain moisture,
and Protocalliphora larvae, unlike other blow-
flies, survive poorly when humidity is high
(Bennett and Whitworth 1991).

ACKNOWLEDGMENTS

The Alaska Bird Observatory provided all field support
for student interns (MLP, BSE-C). Pete Eisner, Sue Guers.
Anne Ruggles, and David Shaw provided valuable support
and suggestions during this project. We sincerely thank the
following entomologists for identifications they provided:
Terry Whitworth (Protocalliphora), Jan Ruzicka (Catops).
Robert Davidson (Patrobus foveocollis), and Ales Smetana
(Quedius). We thank Xin Lu for comparative details on nest
building in Phylloscopus and Richard Ring for assistance
during nest collection. We acknowledge Melissa and Nina
Sikes for counting puparia. Nicholas Buttino and Sarah
Hunter-Smith assisted with nest dissections. Undergraduate
summer research stipends were provided by the Robert K.
Enders Eield Biology Award (BSE-C) and the Lande
Research Award (MLP) at Swarthmore College.

LITERATURE CITED

Barlein, F. 2006. Family Sylviidae (Old World Warblers).
Pages 492-578 in Handbook of birds of the world.
Volume 1 1 (J. del Hoyo, A. Elliott, and D. A. Christie,
Editors). Lynx Edicions, Barcelona, Spain.

Bennett, G. F. and T. L. Whitworth. 1991. Studies on
the life history of some species of Protocalliphora
(Diptera: Calliphoridae). Canadian Journal of Zoology
69:2048-2058.

Bi, Z. L. 2004. Study of the vocalizations, breeding ecology
and nest evolution of the genus Phylloscopus.
Dissertation. Institute of Zoology, Chinese Academy
of Sciences. Beijing, China.

Broggi, J. and J. C. Senar. 2009. Brighter Great Tit
parents build bigger nests. Ibis 151:588-591.

Biirton. N. H. K. 2007. Intraspecific latitudinal variation in
nest orientation among ground-nesting pas.serines; a
study using published data. Condor 109:441^46.
Clement. P. 2006. Arctic Warbler. Pages 661-662 in
Handbook of birds of the world. Volume 1 1 (J. del
Hoyo, A. Elliott, and D. A. Christie. Editors). Lynx
Edicions. Barcelona. Spain.

Long. A. M., W. E. Jen.sen. and K. A. With. 2009.

Ila^eliii d al. • ARCTIC WARBLER NEST MASS

705

Orienlation ol Grasshopper Sparrow and Eastern
Meadowlark nests in relation to wind direction.
Condor 1 1 1 : 395-399.

Lowther, P. E. and S. Sharbaugh. 2008. Arctic Warbler
{Phylloscopiis borealis). The birds of North America.
Number 590.

MAtNWARiNG, M. C.. C. M. H. Benskin, and I. R.
Hartley. 2008. The weight of female-built nests
correlates with female but not male quality in Blue Tit
Cyanistes caeruleiis. Acta Ornithologica 43:43—48.

Murie, A. 1956. Nesting records of the Arctic Warbler in
Mount McKinley National Park. Alaska. Condor
58:292-293.

Peck, S. B. and J. Cook. 2002. Systematics, distributions,
and bionomics of the small carrion beetles (Coleop-
tera: Leiodidae: Cholevinae: Cholevini) of North
America. Canadian Entomologist 134:723-787.

Perreau. M. 1998. Coleopteres commensaux des oiseaux
Alcidae au nord-est de la Siberie (Coleoptera,
Silphidae, Agyrtidae et Leiodidae Cholevinae). Bulle-
tin de la Societe Entomologique de France 103:83-89.

PoHL. G. R.. D. W. Langor, and J. R. Spence. 2007. Rove
beetles and ground beetles (Coleoptera: Staphylinidae,
Carabidae) as indicators of harvest and regeneration
practices in western Canadian foothills forests. Bio-
logical Conservation 137:294—307.

PRtCE, T. and D. Beck. 1989. Observations on the breeding
of the Arctic Warbler in Alaska. Condor 91:219-221.

Reeves. A. B., S. V. DROVETSKt, and I. V. Fadeev. 2008.
Mitochondrial DNA data imply a stepping-stone
colonization of Berengia by Arctic Warbler Phyllo-
scopiis borealis. Journal of Avian Biology 39:567-575.

Roby, D. D., K. L. Brink, and K. Wlitmann. 1992.
Effects of bird blowfly parasitism on Eastern Bluebird
and Tree Swallow nestlings. Wilson Bulletin 104:630-
643.

Sabrosky, C. W., G. F. Bennett, and T. L. Whitworth.
1989. Bird blow flies (Proiocalliphora) in North

America (Diptera: Calliphoridac) with notes on the
Palearctic species. Smithsonian Institution Press,
Washington, D.C., USA.

Sharbaugh, S., R. Ring, S. Guers, and D. Shaw. 2007.
Breeding ecology and habitat associations of the
Arctic Warbler (Phyllo.scopus borealis) in central
Alaska. Final Report 2004-2006. Alaska Bird Obser-
vatory, Fairbanks, USA. http://www.ala.skabird.org/
?page_id = 249

Skowron, C. and M. Kern. 1980. The insulation of nests
of selected North American songbirds. Auk 97:816-
824.

SoLER, J. J., J. J. Cuervo, and A. P. Moller. 1998. Nest
building is a sexually selected behaviour in the Barn
Swallow. Animal Behaviour 56:1435-1442.

Strebyl. H. M., S. M. Peterson, and P. M. Kapfer. 2009.
Fledging success is a poor indicator of the effects of
bird blow flies on Ovenbird survival. Condor 1 1 1:193-
197.

SZENTIRMAl, 1., J. KOMDEUR, AND T. SZEKLEY. 2005. What
makes a nest-building male successful? Male behavior
and female care in Penduline Tits. Behavioral Ecology
16:994-1000.

Tucker, B. W. 1949. Kennicott’s Willow Warbler
Acanthopneuste borealis kennicotti (Baird) Pages
330-339 in Life histories of North American thrushes,
kinglets, and their allies (A. C, Bent, Editor). U.S.
National Museum Bulletin 196.

Wilkie, D. 1983. Rayleigh test for randomness of circular
data. Applied Statistics 32:31 1-312.

With, K. A. and D. R. Webb. 1993. Microclimate of
ground nests: the relative importance of radiative
cover and windbreaks for three grassland species.
Condor 95:401—413.

WiTTMANN, K, AND R. C. Beason. 1992. The effect of
blowfly parasitism on nestling Eastern Bluebird
development. Journal of Field Ornithology 63:286-
293.

The li 'ilsoii Journal of Oniilhology 122{4):706-715. 2010

HOME RANGE, HABITAT USE, AND NEST SITE CHARACTERISTICS
OF MISSISSIPPI KITES IN THE WHITE RIVER
NATIONAL WILDLIFE REFUGE, ARKANSAS

TROY J. BADER’ AND JAMES C. BEDNARZ'

ABSTRACT. — We located 39 Mississippi Kite {Ictinia mississippiensis) nests during the 2004 and 2005 breeding
seasons in the White River National Wildlife Refuge in the Mississippi Alluvial Valley of Arkansas, USA. Radio
transmitters were placed on seven adult and eight juvenile kites; 649 locations based on ground telemetry and 393 locations
from aircraft were recorded for six adults {n = 5 males, 1 female) and five juveniles (/; = 2 males, 3 females). The mean
90% kernel home range was 3,098 ha for adult kites {n = 6) and 439 ha (n = 5) for juveniles. Radio-marked Mississippi
Kites significantly used mature forest (65.6%), second growth (15.5%), and water (10.0%) relative to availability (59.6,
I 1.5, and 7.1%, respectively). Agriculture fields (6.5%) and wetlands (1.5%) were used significantly less within the home
ranges of kites relative to availability ( 1 7.6 and 3.9%, respectively). Tree height, diameter at breast height (dbh), and height
of nest tree emergence above the surrounding trees were significantly greater for nest sites than randomly-selected
overstory trees. Nest trees were significantly closer to the edge than randomly-selected trees. Most (57%) Mississippi Kite
nests were in crotches of secondary branches in the nest tree. Conservation of super-emergent trees and mature forests is
needed for nesting and foraging areas. Second-growth forest should be allowed to mature (>75 years) and uniform canopies
should be avoided. Received 7 August 2009. Accepted 28 April 2010.

The Mississippi Kite (Ictinia mississippien.sis)
breeds in bottomland and riparian areas through-
out the Mississippi River Valley and southeastern
North America (e.g., Kalla 1979, Evans 1981,
Whitmar 1987, Barber 1995, St. PieiTe 2006).
This raptor also inhabits riparian areas, shelter-
belts, oak (Querciis spp.)-shrub prairies, and
savannas in the western United States (Parker
1988). This kite has adapted to nesting in urban
golf courses, parks, and neighborhoods primarily
in the western portion of its range (Parker 1988).
The ability of the western population to adapt to
encroachment of humans has allowed it to
increase in number and expand into new areas
(Parker and Ogden 1979).

Kite populations in the eastern United States
were affected by severe habitat alteration in the
late 18()()s and early 190()s. Gosselink and Lee
(1989) estimated that, at time of European
settlement, there were ~8() million ha of forested
freshwater wetlands in the United States with the
majority in the Mississippi Alluvial Valley. The
area of bottomland hardwood forests in the
Mississippi Alluvial Valley decreased to 4.8
million ha by 1937, and to only 2.1 million ha
by 1970 (Gosselink et al. 1990).

' Department of Biological Sciences, Arkansas State
University. P. O. Box 599, Jonesboro. ,AR 72467, USA.

’Current address: USDA-ARS, Stuttgart National Aqua-
culture Research Center, P. O. Box 1050, Stuttgart, AR
72160. USA.

'Corresponding author; e-mail: troybader@hotmail.eom

Several researchers have described habitat use
and nest site characteristics of Mississippi Kites,
often based on small sample sizes, in Tennessee
(Kalla 1979), Illinois (Evans 1981), Missouri
(Whitmar 1987, Barber 1995), and Arkansas (St.
Pieire 2006) within the Mississippi Alluvial Valley;
almost all of this work remains unpublished. Allan
and Sime (1943) observed daily activities of 169
Mississippi Kites in the panhandle of Texas and
subjectively estimated home-range size to be
~250 ha. Barber (1995) using telemetry estimated
mean home-range size of kites along the Mis-
sissippi River in Missouri as 1,160 ha (/? = 10),
much larger than estimates from the Texas
panhandle. Barber (1995) compared the number
of telemetry locations within specific vegetation
types to estimates of the proportions of these types
available in the surrounding landscape.

Forested habitats fragmented with agricultural
fields and developments are not ecologically
equivalent to the relatively contiguous-forested
habitats once used by kites in the 180()s and early
19()()s. The vast changes in the landscape probably
affected home-range size and habitat use, and may
have influenced reproductive success of this
species (Bader and Bednarz 2009). A better
understanding of how Mississippi Kites use
landscapes that are present today will be useful
to conserve and enhance habitats for (his species.
Our objectives were to examine: (1) use of
habitats and landscapes, and (2) selection of nest
sites by Mississippi Kites in (he Mississippi
Alluvial Valley.

706

Bader and Bednarz • llOMli RANGE: AND llABi rA'I’ OE' MISSISSIPPI KI TES

707

Wildlife Refuge and vicinity in Arkansas, 2004-2005.

METHODS

Study Area. — White River National Wildlife
Refuge (WRNWR; Fig. I) is in Arkansas, Desha.
Monroe, and Phillips counties in southeastern
Arkansas. It is —64,750 ha in size and is divided
into two units by Arkansas Highway I. The South
Unit, where the majority of our work was
conducted, is the larger of the units (41,440 ha).
This unit extends from St. Charles, Arkansas
south to the confluence of the White River and the
Arkansas River Canal (Fig. I). The WRNWR
consists mostly of bottomland hardwoods open to
the public for recreational use, hunting, and
fishing and is managed for both game and non-
game wildlife species (Bader 2007).

Nest Searching. — We began searching for
Mississippi Kite nests on 9 and 23 April 2004

and 2005, respectively. Once a Mississippi Kite
was located, we watched it for possible breeding
activities such as carrying sticks or copulation,
which indicated kites were nesting nearby. We
recorded the location of each nest with a Global
Positioning System (GPS) receiver and left the
site immediately to minimize disturbance. We
periodically returned to areas where Mississippi
Kites were seen and continued searching until a
nest was lound or kites abandoned the area,
indicating they were no longer attempting to
breed.

Capturing and Radio-marking Kites. — We used
a mist-net system with a live Great Horned Owl
{Bubo virginianus) or Red-shouldered Hawk
(Buteo lineal us) near nests with nestlings (7-
30 days of age) to capture adult Mississippi Kites

708

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122. No. 4. December 2010

( Hamerstrom 1963, Barber et al. 1998). The mist-
net system consisted of two 2.6 X 6 m mist nets
(72 mm mesh) connected to a pulley system on
telescoping metal poles. This allowed the top of
the net to be elevated ~7 m above ground with
the bottom of the lower net —1.5 m off the
ground. The total dimensions of the system using
two mist nets were 5.2 m in height X 6 m wide
(Bader 2007).

We placed a U.S. Geological Survey (USGS)
aluminum band and two or three plastic color
bands on legs of each kite for unique identifica-
tion. Selected kites were also fitted with a radio
transmitter using a modified figure-eight leg
harness (Radley 2002). Mississippi Kites were
fitted with a 6.0-g transmitter (Holohil Systems
Ltd., Carp, ON, Canada). We recorded several
linear measurements, mass, and obtained a blood
sample from each kite, before it was released.
Gender of each kite was assigned based on
plumage characteristics (Parker 1999), and veri-
fied by DNA analysis in the laboratory (Bader
2007). We left the nest site immediately once a
kite was released to minimize any potential stress
on adult and nestling kites.

We climbed to each nest and banded the
nestling(s) when they were 3-4 weeks of age. The
mass of each nestling was measured, and a USGS
aluminum band and two or three plastic color bands
were placed on their legs for identification of
individuals. Each nestling was also fitted with a
radio transmitter using the procedure described for
adults and returned to the nest.

Gender Identification. — A 0.05-cc sample of
blood was drawn from the brachial vein of each
kite using a hypodermic needle and a small
capillary tube (Faaborg et al. 1995) and placed in
Longmire Solution (Longmire et al. 1988). The
protocol outlined by Donohue and Dufty (2006)
was followed to extract DNA from the blood and
i.solate the CHD-Z and CHD-W bands (Griffiths et
al. 1998) to identify the gender of captured kites.

Radiotelenietry of Adult and Juvenile Kites. —
Triangulation was used to obtain locations from
the ground and a fixed-wing aircraft was used to
identify point locations from the air. Receiver
sites for triangulations were established —500 m
apart along roads or all-terrain vehicle trails near
nest sites and compass azimuths were recorded in
the direction of the strongest signal from the
radio-marked kite. Compass azimuths from at
least three receiver sites were taken within a 5-
min period to estimate the error ellipse around

each location. Eight triangulations with a >10-
min interval between each triangulation were
completed 2-3 days each week. The 10-min
period allowed time for the kite to move between
triangulation attempts, minimizing autocorrelation
between locations (Swihart and Slade 1985). The
triangulation azimuth was entered into the OTA
Triangulation Program (Ripper et al. 2007), which
provided coordinates for the estimated location.
We could calculate an eiTor ellipse using this
program that theoretically has a 95% probability
of including the true location of the kite. We
eliminated all estimated locations that had an
error ellipse greater than 10 ha to improve
accuracy. This relatively large error ellipse
criteria was chosen to accommodate the high
mobility of kites (typical home range size
>3,000 ha).

We obtained locations from the air (DeVault et
al. 2003) for each radio-marked kite 2-3 times per
flight depending on the amount of flight time
available. Locations of individual kites were
recorded >20 min apart. Flights were conducted
2-3 days each week using a Cessna 172 or 182
equipped with a pair of “side-looking” 4-element
yagi antennae (Bader 2007).

Habitat Use. — Point locations were entered into
Geographic Information System (GIS) ArcView
software and home-range polygons were devel-
oped for each kite. Two types of home-range
polygons were generated using point locations;
minimum convex polygon (MCP) and 90% kernel
(White and Garrott 1990). We used the 90%
probability of use contours as estimates of home-
range size. The home-range polygon of each kite
was overlaid onto USGS Digital Orthophotograph
Quarter Quadrangles (DOQQs) taken in fall 2004.

ArcMap 8.3 at a scale of 1 :3,000 was used to
classify habitats into seven cover types within
each home range. The mature forest cover type
consisted of tracts of large trees that had relatively
even canopy with few openings and had not been
recently logged (>30 year). Second-growth cover
type consisted of young forest with large openings
in the canopy, abandoned fields, and recently-
logged forest. The water cover type included all
open waterways (e.g., lakes, bayous, and rivers).
The wetland cover type included all forested and
non-forested swamps and low-lying vernal areas.
The agriculture cover type included all types of
agriculture fields (e.g., crop and hay fields). The
urban cover type consisted of noticeably devel-
oped areas (e.g., rural towns). The road cover type

Bcicler ami Bcdmirz • HOME RANGE AND IIARITAT OE' MISSISSIRRI KITES

709

included highways, county roads, and obvious
private roads. Polygons were drawn around each
cover type and the sum of all polygons within a
cover type was used to calculate the number of
hectares available of that cover type within a
given home range.

Bailey’s confidence intervals (Bailey 1980,
Cherry 1996) were used to examine whether kites
used cover types within their home range in
proportion with that available. We used the Chi-
square distribution and a Bonfen'oni correction
based on the number of available cover types to
construct a 95% confidence interval (Cl) around
the proportion of telemetry locations within each
cover type. The confidence interval was compared
to the proportion of habitat that was available
within the kite’s home range to delineate the level
of kite use.

Nest-site Characteristics. — Vegetation data
were collected at every nest site after young
fledged or the nest failed and at an equal number
of sites selected using a random-number table.
The distance (0-250 m) was paced while using a
compass to follow the random azimuth to locate
the reference sample site and the closest overstory
tree was chosen as the plot center. The same
vegetation data were collected within a 0.04-ha
circular plot around nest and random trees. We
followed the BBIRD Protocol (Martin et al. 1997)
and included species of the tree, tree height, tree
diameter at breast height (dbh), distance to forest
edge, dominant and co-dominant trees within the
surrounding area, and species and height of each
tree in the overstory and within the circular plot.
Data collected on nest-site characteristics includ-
ed nest height, orientation of nest in tree, and
position of nest in tree (primary or secondary
fork).

We collected data on perch trees, defined as the
tree(s) in which kites perched in the morning and
frequently during the day in the immediate
vicinity (<200 m) of their nests. Measurements
collected at perch trees were the same as tho.se
taken for nest trees, but also included distance
from nest and azimuth to nest. We recorded
whether the tree was dead or alive and classified it
as a primary or secondary perch tree. The primary
perch tree was the tree that one or both kites used
the majority (>50%) of the time. The secondary
perch tree, if present, was the tree u.sed by one or
both kites less frequently (<50%).

Statistical Analysis. — We used a paired r-test
(SAS Institute 1999) to compare Mississippi Kite

nest sites with sites selected at random (Sokal and
Rohlf 1995), and a /-test to compare characteris-
tics of primary and secondary perch trees.

RESULTS

We located 21 Mississippi Kite nests in 2004
and 18 in 2005. We captured seven adult kites,
each from separate nests (4 in 2004, 3 in 2005), in
22 trapping attempts. We made 1 1 attempts to
capture chicks in 2004 and 2005 and captured
nine (at ages of ~3.5^ weeks) from nine nests.

Analysis of DNA to identify gender of captured
adult and nestling kites was conducted on 14
blood samples (6 adults and 8 nestlings). Five
male and one female adult kites, and four male
and four female nestling kites were identified.
Five of six adults (83%) were identified correctly
to gender based on plumage coloration.

Radio-marked Mississippi Kites were located
using ground receiver sites from 9 July to 28
August 2004 and from 14 July to 20 August 2005.
Locations of each radio-marked kite were record-
ed from a fixed-wing aircraft from 28 July to 19
August 2004 and from 20 July to 26 August 2005.
We recorded 649 locations from the ground (2004
= 449 locations, 2005 = 200 locations) and 393
locations from the air (2004 = 134 locations,
2005 = 259 locations).

MCP and 90% kernel home ranges were
generated for each radio-marked kite (Table 1,
Fig. 2). The mean (± SE) MCP home range size
for adult male radio-marked kites was 3,289.1 ±
1,419.3 ha; n = 5). Home-range data were
collected for only one female and the MCP home
range was 1,899.8 ha. The mean (± SE) MCP
home range for male juvenile kites was 536.2 ±
10.8 ha; n = 2) and 508.4 ± 316.4 ha; n = 3) for
female juvenile kites. The mean (± SE) home
range using the 90% kernel method for adult male
radio-marked kites was 3,567.3 ± 1,393.5 ha; n =
5). The home range size for the one radio-marked
adult female kite using the 90% kernel method
was 752.2 ha. The mean 90% kernel home range
for male juvenile kites was 299.0 ± 23.0 ha (/; =
2) and 533.0 ± 266.7 ha (/? = 3) for female
juvenile kites.

We estimated habitat use within home ranges of
12 Missi.ssippi Kites (6 adults. 6 juveniles).
Mississippi Kites significantly used mature forest
(95% Cl = 0.611-0.698; Table 2) more than its
availability (0.596). Water was also significantly
used (95% Cl = 0.074-0.130; Table 2) more
relative to its availability (0.071). Mississippi

710

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, December 201 U

TABLE 1. Minimum convex polygon (MCP) and 90% kernel home range data for Mississippi Kites radiotracked in the
White River National Wildlife Refuge, Arkansas, 2004-2005.

Band #

Gender

Age-

# of

triangulations

# of aerial
locations

Total

locations

MCP (ha)

90% Kernel (ha)

Year

745-57320

M

AHY

32

20

52

1,398.64

1,594.23

2004

745-57323

L

HY

53

17

70

78.37

55.75

2004

745-57318

L

AHY

103

21

124

1,899.81

752.19

2004

745-57322

M

HY

47

19

66

547.00

321.96

2004

745-57319

M

AHY

1 12

18

130

626.10

431.14

2004

745-5732 1

F

HY

48

19

67

1,125.44

975.75

2004

745-57324

M

AHY

54

20

74

6,225.47

7,770.00

2004

745-57326

M

AHY

95

60

155

7,235.58

5,896.23

2005

745-57327

M

AHY

22

13

35

959.68

2,145.07

2005

745-57329

F

AHY

5

18

23'-'

497.44

1,192.65

2005

745-57328

M

HY

0

1 1

1 U

0

0

2005

745-5733 1

M

HY

35

72

107

525.42

275.94

2005

745-57332

_b

HY

10

26

36'--

548.27

545.45

2005

745-57330

F

HY

33

61

94

321.37

567.61

2005

745-57333

F

HY

0

0

o-^

0

0

2005

AHY = alter hatch year or breeding adult; HY = Hatch year or nestling.

No blood was collected.

■■ Few locations because kite was depredated (not included in calculation of MCP or kernel means).
Locations while chick was still in the nest (not included in calculation of MCP or kernel means).

No data collected before nestling was depredated (not included in calculation of MCP or kernel means).

Kites also significantly used .second growth (95%
Cl = 0.123-0.190; Table 2) more relative to its
availability (0.115). Agriculture fields and wet-
lands were significantly used less relative to
availability (Table 2). Mississippi Kites used both
roads and urban areas in proportion to their
availability (Table 2).

The mean nest tree height (33.26 m) was
significantly taller (/ = 5.37, df = 33, P <
0.0001; Table 3) than paired randomly-selected
trees (26.77 m). Mean dbh of nest trees (75.58 cm)
was significantly greater (t = 8.44, df = 33, P <
O.OOOl; Table 3) than paired randomly-.selected
trees (48.34 cm). Mean height of nest-tree
emergence (5.29 m) above the surrounding
canopy was significantly greater {[ = 3.44, df =
33, P = 0.0016; Table 3) than the mean height of
randomly-selected tree emergence (0.70 m). The
mean (± SE) height of Mississippi Kite nests was
26.89 ± 0.82 m, while the mean surrounding
canopy height was 27.68 ± 0.74 m, indicating
kites tended to build nests almost level with the
height of the surrounding canopy.

The mean (± SE) distance (61.06 ± 15.65 m)
to the forest edge from Mississippi Kite nests was
significantly less (t = 4.16, df = 33, P = 0.()()()2;
Table 3) than the distance (126.19 ± 16.18 m) to
the forest edge from randomly-.selected sites.
Other measured variables did not differ between
nest and random sites (Table 3).

Twenty of 35 nest sites (57%) were in crotches
of .secondary branches. The remaining 15 nests
were on primary branches of nest trees. Nests
were randomly placed in any direction throughout
the nest trees. Mississippi Kite nests were found in
six species of trees with Overcup oak {Quercus
lyrata\ n = II) used most commonly. Sweetgum
{Liquidamber styracifliuv, n = 7), willow oak {Q.
phellos) ill = 7), Nuttall oak (Q. nuttallir, n = 3),
American sycamore (Platcmus occidentals-, n =
2), and white oak {Q. allxv, /? = I ) were used less
frequently as nesting sites.

Mississippi Kites used numerous species of
trees for perching. Black willow {Salix nigra-, n =
9) and water hickory (Carya aquatica-, n = 8)
were the two most common species used for
primary perches. Overcup oak (n = 5) and Nuttall
oak (// = 4) were also commonly used for primary
perches. Sweetgum (/; = 3), slippery elm (Ulinus
rubra-, n = 3), willow oak (// = 2), and water oak
{Q. nigra-, n = 2) were also used as primary perch
trees. Nineteen (52.8%) of 36 primary perch trees
(Table 4) were live with a dead limb used as the
perch, six were live trees with sparsely-leaved
limbs, six were live trees with live limbs used as
perches, and five were dead trees.

Secondary perch trees included water hickory
(/; = 3), overcup oak (// = 2), water oak (/; = 1),
and sLigarberry (Celiis laevigata-, n = 1). One
dead tree, three live trees with sparsely-leaved

Iknkr and Hcdnarz • IlOMli RANGE AND HABITAT Ol' MISSISSlIMM Krn;S

71 i

f,

•7’

Mississippi Kite nest

0

250 500

llMetei s

1,000

Kite locations

I I MCP

90% Kernel home range

FIG. 2, Minimum Convex F^olygon (MCP) and 90% kernel home range of a radio-tracked Mississippi Kile (Band #
745-57331) in the White River National Wildlife Refuge, Arkansas, 2005.

712

THE WILSON JOURNAL OF ORNITHOLOGY • Voi 122. No. 4. December 2010

TABLE 2. The proportion of cover types within home ranges available and used by radio-tracked adult (/? = 6) and
juvenile (n = 6) Mississippi Kites in the White River National Wildlife Refuge, Arkansas, 2004-2005.

Cover type

ha

Locations in
cover type

Proportion of
cover type available

Proportion of
observed use

95% Confidence
Interval of use

Agriculture

3,683.83

57

0.176

0.065*

0.044-0.090

Mature forest

12.497.36

577

0.596

0.656*

0.61 1-0.698

Second growth

2.417.41

136

0.1 15

0.155*

0.123-0.190

Road

64.83

2

0.003

0.002

0.000-0.01 1

Wetland

821.65

13

0.039

0.015*

0.006-0.029

Urban

38.59

6

0.002

0.007

0.001-0.018

Water

1,488.13

88

0.071

0.100*

0.074-0.130

* Significantly different from amount available.

limbs, two live trees with a dead limb, and one
live tree with live limbs were all used as
.secondary perch trees.

There were no significant differences in dbh,
tree height, distance to edge, or distance to nest
between primary and secondary perch trees
(Table 4). Primary and secondary perch trees
were significantly closer {t = 5.6, df = 31, P =
<0.001 ) to an edge and had a significantly larger
(t = 7.5, df = 34, P < 0.0001) dbh than trees
selected at random (Tables 3, 4). There was no
significant difference in height of primary and
secondary trees and trees selected at random.

DISCUSSION

The mean 90% kernel home range size of male
Mississippi Kites (3,567 ha, n = 5) was larger
than the one female monitored (752 ha). We
suggest this difference may be real because males
are more likely to venture long distances from the
nest during their daily foraging activities (Dunstan
et al. 1978, Barber 1995). Males range widely,
probably to gain access to habitat types or food

resources that are not available near the nest site.
Access to a larger foraging area and possibly to
areas with more abundant prey makes the male’s
role as a food provider to the nestlings vital to
success of the nest. Female kites may have smaller
home ranges because of their role in the primary
defense of the nest; they afso tend to the nestlings
more than males, similar to other female raptors
(Newton 1978, Collopy 1984, Bader 2007).

Barber (1995) estimated the mean male MCP
Mississippi Kite home range size in Missouri as
1,357 ha (/7 = 6), notably smaller than the mean
male MCP home ranges that we estimated
(3,289 ha). The mean female kite home range in
Missouri was 865 ha (/; = 4; Barber 1995), also
considerably smaller than the MCP home range of
the one female that we radiotracked ( 1 ,900 ha).
We suggest these difference may be due to either
differences in land.scapes of the respective study
areas (the Missouri study was conducted in a more
fragmented, forest-agriculture landscape) and the
techniques used (no aerial telemetry was used in
the Missouri study), or both factors.

TABLE 3. Habitat characteri.stics al Mi.ssi.ssippi Kite ne.st .sites {n = 39) and sites selected at random (/; = 39) in the
White River National Wildlife Refuge. Arkansas, 2()04-2()()5.

Variable

Nest .site

Random site

p’

Mean ± .SE

Range

Mean ± SE

Range

Tree height (m)

33.26 ± 0.97

24.09^3.07

26.77 ± 0.92

14.30^1.42

■ <0.0001*

DBH (cm)

75.58 ± 3.32

52.70-1 13.80

48.34 ± 2.25

20.30-79.00

<0.0001*

No. of trees in 0.04-ha plot

3.56 ± 0.30

1 .00-7.00

3.53 ± 0.31

1 .00-8.00

0.9378

Mean tree heighP (m)

27.68 ± 0.74

20.77-37.57

26.85 ± 0.72

20.25-37.32

0.3535

Tree emergence"' (m)

5.29 ± 0.93

-8.53-17.85

0.70 ± 0.76

1 1.20-9.39

0.0016*

Distance to forest edge (m)

61.06 ± 6.13

0-142.50

126.19 ± 16.18

0-3 1 1 .00

0.0002*

* Significanlly tlifferenl from silcs selected at random. !’ £ 0.05.

" /’ ^ probability that means are dilTerenl based on a paired t-tesl.

Mean iree height - mean height of trees in a ().l)4-ha plot surrounding the nest or random tree.

^ Tree emergence - the difference between mean height of nest trees above surrounding trees in 0.04 ha plot or mean height of random trees above surrounding
trees in 0.04-h:i plot.

liuilcr and Bednarz • HOME RANGE AND HABITAT OE MISSISSIPPI KITES

713

TABLE 4. Characteristics ot primary (n — 36) and secondary (n = 7) perch trees ol' Mississippi Kites in the White
River National Wildlife Refuge. Arkansas, 2004-2005.

Primary Secondary

Variable

Mean ± SE

Range

Mean ± SE

Range

Tree height (m)

30.41 ± 1.68

14.3-66.5

27.23 ± 3.31

18.6-41.46

DBH (cm)

74,45 ± 2.77

36.1-1 1 1.8

84.14 ± 7.43

55.5-103.8

Distance to nest (m)

82.68 ± 8.66

1 1.3-231.0

78.53 ± 10.02

45.0-120.0

Distance to edge (m)

18.74 ±3.17

0-77.3

29.89 ± 16.49

0-124.5

Our data documented that Mississippi Kites
significantly used open waterways more than
available to kites within their home ranges
(Table 2). Open water likely provided habitat for
prey items such as dragonflies (Odonata) and
frogs (Anura). Barber (1995) did not report this
cover type to be more abundant within kite home
ranges compared to reference areas, but it was a
large part of most home ranges that he sampled.

Agricultural fields were used less within the
kite home ranges in Arkansas than available
(Table 2), but were significantly more abundant
in kite home ranges in Missouri (Barber 1995).
Soybeans, milo, com, cotton, and fallow fields
were included in the agriculture cover type in the
Missouri study. Rice, soybeans, and cotton were
the primary crops farmed in the areas surrounding
the WRNWR.

Wetland areas were significantly underused
(Table 2), but we suggest wetlands indirectly
may be an important area for foraging kites.
Wetlands likely produce prey, but this prey is not
restricted to these areas and probably moves into
other cover types where kites may forage on them.
Wet areas are used as foraging habitats by other
raptors such as the Red-shouldered Hawk (Bed-
narz and Dinsmore 1981, Bloom et al. 1993,
Howell and Chapman 1997).

Mature forests were significantly u.sed more
within the home ranges of Mississippi Kites in
Missouri (Barber 1995) and in Arkansas (Table 2).
In Tennessee, Kalla (1979) reported wooded areas
in the floodplain provided preferred nesting sites for
kites and key habitat for prey species. Mature
forests provide habitat for dragonflies and cicadas
(Homoptera), the two most common prey items
delivered to nests in the WRNWR (Bader 2007).
Red-shouldered Hawks were demonstrated to use
forested habitats in California (Bloom et al. 1993)
and bottomland hardwood forests in Georgia
(Howell and Chapman 1997) more than available.
Lai'ge contiguous tracts of mature forests seem to be

an important cover type for kites providing both
nest sites and prey.

Second growth was used significantly more
within home ranges of kites in Arkansas (Table 2),
but not in Missouri (Barber 1995). This cover type
is common in numerous landscapes, and can
provide foraging habitat and marginal nesting
habitat for kites. Timber stands classified as second
growth can provide nest sites if super-emergent
trees are not removed during timber harvest
operations or are created during timber haiwest.
Fallow fields, which consist of early successional
growth, can provide habitat for aerial insects and
other prey items. The second growth cover type was
not common in kite home range in Missouri and
only accounted for 3.8% of home ranges and 1 .2%
of the sampled suiTOunding areas.

Kites used urban areas and roads in proportion
to availability within their home range. Both
urban areas and roads were near other cover types
u.sed by kites. Kites may have used thermals
produced in these areas and probably did not
actively select these sites for foraging.

Mississippi Kites selected trees taller than the
mean overstory canopy height for their nest tree
(Table 3). U.se of super-emergent trees for nesting
allows this long-winged raptor efficient access to
the ne.st (Barber et al. 1998). Kites in WRNWR
tended to build nests almost level with the top of
the suiTOLinding canopy. Barber (1995) and Whit-
mar ( 1 987) reported nests were higher than the tops
of the suiTOLinding trees. Placing nests at or slightly
above the tops of the surrounding trees may
provide a view of the surrounding area and
potential predators (Whitmar 1987, Barber 1995).
Placing nests slightly below the surrounding
canopy height would give the nest more protection
from the wind and make the nest less conspicuous
to predators. Mississippi Kite nest trees have a
larger dbh than randomly-selected trees (Table 3).
Barber (1995) also reported the dbh of nest trees to
be significantly larger than random trees. The large

714

THE WILSON JOURNAL OF ORNITHOLOGY • Voi 122. No. 4, December 2010

dbh of nests trees may be correlated with age and
size ot the super-emergent trees typically used for
nesting sites.

Mississippi Kites selected nest sites closer to
the forest edge than random sites and all forest
edges comprised of lakes, bayous, or the White
River. This pattern was also documented during
research previously conducted on the refuge (St.
Pierre 2006). These bodies of water provide open
areas for kites to forage for aerial insects. Barber
(1995) reported kite nests were located closer to
field edges more frequently than water edges.
Field edges were uncommon within the interior of
WRNWR, where waterways comprised the ma-
jority of the available edges. Whitmar (1987)
reported kite nests in the Mississippi River Valley
in Missouri had a mean distance to an edge of
49.2 m and a mean distance to water of 96.3 m.
Edge habitats, particularly natural edges, seem to
be an important attribute of Mississippi Kite
nesting habitat.

Kite nests were more commonly built on
secondary than on primary branches. Barber
( 1995) akso documented this pattern for kite nests
in Missouri and suggested this may be an anti-
predator defense. Our data and those reported by
Whitmar (1987) indicated kite nests were seem-
ingly randomly placed throughout the nest tree.

CONSERVATION IMPLICATIONS

Mature tracts of forests should be conserved
because this habitat type provides nesting sites
and a foraging habitat for Mississippi Kites.
Large, super-emergent trees should be con.served
for kite nesting habitat. Management that pro-
motes development of additional large, super-
emergent trees will likely be beneficial to kites.
Timber harvests consistent with kite conservation
should be conducted .so uneven-height canopy
forests are created. We recommend that second-
growth forest should be allowed to develop longer
between harvests (>75 years) to provide foraging
and nesting habitat for kites. Managers should
avoid even-aged regeneration of dear-cuts and
fallow fields. Natural waterways, including wet-
lands, should be con.served for production of prey
and as foraging areas.

ACKNOWLEDGMENTS

This project was t'uncicci primarily by the U.S. Fish anti
Wildlife Service (USFWS) and Arkansas Game and Fish
Commission (AGFC) through a State Wildlife Grant.
Substantial additional funds were provided by Arkansas

State University. We thank K. L. Rowe (AGFC) for support
and assistance on this project. We thank the USFWS,
especially R. E. Hines and staff at the White River National
Wildlife Refuge, for help on the logistics for this project.
We are especially grateful to our research technician. W. D.
Edwards for hard work and dedication to this project. We
greatly appreciate the assistance in the field and in the
laboratory provided by J. D. Brown. T. J. Benson. D. J.
Baxter. T. R. Edwards. E. J. Bader, and A. M. St. Pierre.

LITERATURE CITED

Allan. P. F. and P. R. Sime. 1943. Distribution and
abundance of the Mississippi Kite in the Texas
panhandle. Condor 45:1 10-1 12.

Bader, T. J. 2007. Reproductive success, causes of nesting
failures, and habitat use of Swallow-tailed and
Mississippi kites in the White River National Wildlife
Refuge, Arkansas. Thesis. Arkansas State University,
Jonesboro, USA.

Bader, T. J. and J. C. Bednarz. 2009. Reproductive
success and causes of nesting failures for Mississippi
Kites: a sink population in eastern Arkansas? Wetlands
29:598-606.

Bailey. B. J. R. 1980. Large sample simultaneous
confidence intervals for the multinomial probabilities
based on transformation of the cell frequencies.
Technometrics 22:583-589.

Barber, J. D. 1995. Mississippi Kite habitat use and
reproductive biology in the Mississippi flood plains of
southeast Missouri. Thesis. University of Missouri,
Columbia, USA.

Barber. J. D., E. P. Wiggers, and R. B. Renken. 1998.
Nest-site characterization and reproductive success of
Mississippi Kites in the Mississippi River floodplains.
Journal of Wildlife Management 62:1373-1378.
Bednarz, J. C. and J. J. Dinsmore. 1981. Status, habitat
use, and management of Red-shouldered Hawks in
Iowa. Journal of Wildlife Management 45:236-241.
Bi.oom, P. H., M. D. McCrary, and M. J. Gibson. 1993.
Red-shouldered Hawk home-range and habitat use in
southern California. Journal of Wildlife Management
57:258-265.

Cherry. S. 1996. A comparison of confidence interval
methods for habitat u.se-availability .studies. Journal of
Wildlife Management 60:653-658.

COLLOFY, M. W. 1984. Parental care and feeding ecology
of Golden Eagle nestlings. Auk 101:753-760.
DeVault, T. L.. W. L. Stephens, B. D. Reinhart, O. E.
Rhodes Jr., and I. L. Brisbin Jr. 2003. Aerial
telemetry accuracy in a forested landscape. Journal of
Raptor Research 37:147-151.

Donohue. K. C. and A. M. Dui-ty Jr. 2006. Sex
determination of Red-tailed Hawks (Biiteo jamaicensis
c(iliinis) using DNA analysis and morphometries.
Journal of Field Ornithology 77:74-79
Dunstan, T. C., j. H. Harper, and K. B. Phipps. 1978.
Habitat use and hunting strategics of Prairie Falcons,
Rcd-tailcd Hawks, and Golden Eagles. Final Report
submitted to USDI, Bureau of Land Management,
Denver, Colorado, USA.

Bader and Bednarz • HOME RANGE AND IIARITA T Ol' MISSISSIPPI KITES

715

Evans, S. A. 1981. Ecology and behavior ol the Missi.ssippi
Kite (Icrinia niississippiensis) in .southern Illinois.
Thesis. Southern Illinois University, Carbondale, USA.
Faaborg, J,, P. G. Parker, L. Delay, T. De Vries, J. C.
Bednarz, S. M. Paz, J. Naranjo, and T. A. Waite.
1995. Conlirmation of cooperative polyandry in the
Galapagpos Hawk (Biiieo galapagoensis). Behavioral
Ecology and Sociobiology 36:83-90.

Gosselink, j. G. and L. C. Lee. 1989. Cumulative impact
assessment in bottomland hardwood forests. Wetlands
9:89-174.

Gosselink, J. G., G. P. Shaffer, L. C. Lee, D. M.
Burdick, D. L. Childers, N. C. Leibowitz, S. C.
Hamilton, R. Boumans, D. Cushman, S. Fields, M.
Koch, and J. M. Visser. 1990. Landscape conserva-
tion in a forested wetland watershed. Can we manage
cumulative impacts? Bio.science 40:588-600.
Griffiths, R., M. C. Double, K. Orr, and R. J. G.
Dawson. 1998. A DNA test to sex most birds.
Molecular Ecology 7:1071-1075.

Hamerstrom, F. 1963. The use of Great Horned Owls in
catching Marsh Hawks. Proceedings of the Interna-
tional Ornithological Congress 13:866-869.

Howell, D. L. and B. R. Chapman. 1997. Home range and
habitat use of Red-shouldered Hawks in Georgia.
Wilson Bulletin 109:131-144.

Kalla, P. I. 1979. The distribution, habitat preference, and
status of the Mississippi Kite (Ictinia mississippiensis) in
Tennessee, with reference to other populations. Thesis.
East Tennessee State University, Johnson City, USA.
Longmire, j. L., a. K. Lewis, N. C. Brown, J. M.
Buckingham, L. M. Clark, M. D. Jones, L. J.
Meincke, j. Meyne, R. L. Ratliff, F. A. Ray, R. P.
Wagner, and R. K. Moyzis. 1988. Isolation and
molecular characterization of a highly polymorphic
centromeric tandem repeat in the family Falconidae.
Genomics 2:14-24.

Martin, T. E., C. R. Paine, C. J. Conway, W. M.
Hochachka, P. Allen, and W. Jenkins. 1997.
BBIRD field protocol. USGS, Montana Cooperative

Wildlife Research Unit, University of Montana,
Mis.soula, USA.

Newton, I. 1978. Feeding and development of Sparrow-
hawk Accipiler nisus nestlings. Journal of Zoology
184:465-487.

Parker, J. W. 1988. Mississippi Kite Ictinia mississip-
piensis. Pages 166-186 in Handbook of North
American birds (R. S. Palmer, Editor). Volume 4.
Yale University Press, New Haven, Connecticut, USA.

Parker, J. W. 1999. Mississippi Kite (Ictinia mississip-
piensis). The birds of North America. Number 402.

Parker, J. W. and J. C. Ogden. 1979. The recent history
and status of the Mississippi Kite. American Birds
33:1 19-129.

Radley, P. H. 2002. Artificial nest structure use, post-
fledging habitat u,se, and dispersal of Barn Owls (Tyto
alba pratincola) in the Delta Region of Arkansas.
Thesis. Arkansas State University, Jonesboro, USA.

Ripper, D., J. C. Bednarz, and D. E. Varland. 2007.
Landscape use by Hairy Woodpeckers in managed
forests of northwestern Washington. Journal of
Wildlife Management 71:2612-2623.

SAS Institute. 1999. SAS/STAT user's guide. Version
8.2. SAS Institute Inc., Cary, North Carolina, USA.

SoKAL, R. R. AND F. J. Rohlf. 1995. Biometry: the
principles and practices of statistics in biological
research. W. H. Freeman and Company, New York,
USA.

St. Pierre, A. M. 2006. Reproductive ecology of Swallow-
tailed and Mississippi kites in southeastern Arkansas.
Thesis. Arkansas State University, Jonesboro, USA.

Swihart, R. K. and N. a. Slade. 1985. Testing for
independence of observations in animal movements.
Ecology 66:1 176-1 184.

White, G. C. and R. A. Garrott. 1990. Analysis of
wildlife radio-tracking data. Academic Press, New
York. USA.

Whitmar II, W. 1987. Quantification of the nesting site
components in the Mississippi Kite (Ictinia mississip-
piensis). Thesis. Southeast Missouri State University,
Cape Girardeau, USA.

The li ilson Journal of Ornithology 1 22(4);7 16-724, 2010

BIOLOGICAL, GEOGRAPHICAL, AND CULTURAL ORIGINS OF THE
LOON HUNTING TRADITION IN CARTERET COUNTY,

NORTH CAROLINA

STORRS L. OLSON, HORACE LOFTIN,- AND STEVE GOODWIN'

ABSTRACT. — A tradition of shooting Common Loons (Gavici immer) for food and for bone fishing lures was
established on Shackleford Banks, North Carolina, by the mid- 19th century. This strongly ingrained tradition continued to
be maintained, primarily by residents ot nearby Markers Island, when inhabitants of the banks moved inland about 1899.
The practice probably arose because, on the east/west-tending Shackleford Banks, loons migrating northward in spring flew
sufficiently low over land to be within shotgun range. Spring loon shooting, although illegal since 1918, grew to the point
that dozens of hunters might be present on the banks on a given day. A strict law enforcement crackdown on this activity
began in 1950, and the banks were effectively shut down for loon shooting. Loons continued to be shot opportunistically
nearby, but a growing cultural intolerance of this practice brought the loon hunting tradition to an end. We document the
existing memories and the few, scattered written sources concerning this unique local interaction between humans and

birds. Received 30 March 2010. Accepted 23 June 2010.

On 6 May 1950, eleven federal and state
wildlife enforcement officers staged a long
planned, coordinated raid on Shackleford Banks,
Carteret County, North Carolina, apprehending
nearly 100 hunters, of whom 78 were formally
charged with illegally shooting loons (Anony-
mous 1950) in violation of the Migratory Bird
Treaty Act of 1918. This locally notorious event
marked both the apogee and the rapid demise of a
demographically narrow but strongly ingrained
regional tradition, dating to the previous century,
in which loons were hunted both for food and for
use of their bones in making fishing lures. Loons
do not usually factor so directly in human
economy, nor does a similar loon hunting tradition
.seem to have evolved eksewhere, so we investi-
gated whether some unique combination of
biological and physiographical conditions may
have given rise to the loon hunting tradition in
Carteret County. We also attempted to document
through interviews and published and unpublished
local records, how the values and attitudes of one
small community were molded by the perception
of loons as an exploitable natural resource. The
Carteret County loon hunting tradition has
es.sentially been eradicated because of changes
in the economic, legal, and ethical environment.
Knowledge of this aspect of regional wildlife

' Division of Birds, National Mu.seiim of Natural History,
.Smithsonian Institution. P. O. Box 37012, Washington,
D.C. 20013. USA.

’Carteret County Historical Society, 1008 Arendell
Street, Morehead City, NC 28557, USA.

'P. O. Box 2027, Beaufort, NC 28516, USA.

■* Corresponding author; e-mail: olsons@si.edu

heritage is in danger of vanishing, hence the
importance of its historical documentation.

METHODS

We interviewed residents of Carteret County,
North Carolina, who had participated in loon
hunting, or who had firsthand knowledge of some
aspect of the circumstances under which loons
were obtained or were directly employed for
human use. We gathered numerous other sources
that, although published, are of a very restricted or
evanescent nature and mostly unavailable in
major research libraries. These publications are
held in the libraries of the Carteret County
Historical Society, Morehead City, NC, and the
Core Sound Waterfowl Museum and Heritage
Center, Harkers Island, NC; photocopies of
pertinent portions, as well as all of our interview
notes, have been filed in the Smithsonian
Institution Archives, Washington, D.C.

GEOGRAPHICAL CONSIDERATIONS

The key component of this story is a physio-
graphical feature, Shackleford Banks, a long,
13.8 km (8.6 mi), narrow, barrier island with its
northern shore separated from Harkers Island by
Back Sound and whose southern shore is washed
by the open Atlantic (Fig. 1). Unlike most
Atlantic barrier islands that tend to run north to
south, Shackleford Banks is oriented nearly east to
west. The eastern end of the island is sheltered by
Cape Lookout and lies somewhat farther south
than the western end. This east/west orientation is
of great significance when considering the spring
migratory pathway of loons.

716

Olson et al. • LOON HUN LING IN NOR'LI I C'AROIJNA

717

FIG. 1. Outline map ot Carteret County, North Carolina, USA, showing localities mentioned in the text. Black
emphasizes islands of particular importance in the history of loon hunting. Inset: Carteret County (in black) in relation to
other counties of North Carolina.

Shackleford Banks was settled by the early
1700s and later came to have a considerable
human population concentrated in two communi-
ties, Diamond City and Wade’s Shore (or Wade’s
Hammock), whose inhabitants were mainly shore
whalers who supplemented their livelihoods
taking mullet (striped mullet Miigil cephalus)
and porpoises (bottle-nosed dolphin Tursiops
tnincatiis). A growing tendency for people to
move off the banks to areas closer to commerce
and other amenities was greatly accelerated in the
late 19th century when the settlements were
devastated, and Diamond City’s high protective
sand dune eroded away, by the great San Ciriaco
hurricane that hit the outer banks on 17 August
1899. With the realization that the banks no
longer provided a secure place to live, the
Shacklefordians migrated inland, settling as far
west as Salter Path on Bogue Banks, in the
“Promise’ Land’’ section of Morehead City, and
to the northeast as far as Cedar Island. The
majority, however, retreated to Markers Island.
Shackleford Banks was essentially uninhabited by
1902 and was used mainly as pasturage for
hooved stock for most of the 20th century. Much
of the preceding was summarized from Stick
(1958). Shackleford Banks became part of the

Cape Lookout National Seashore in 1966 (Public
Law 89-366).

LOON BIOLOGY AND BEHAVIOR AS
PERTAINS TO HUNTING

The Common Loon (Gavia immer) is a laige,
fish-eating waterbird that breeds on fresh-water
lakes of boreal forests across North America and
winters at sea along both coasts as far south as
Mexico. It is a heavy bird with males reaching a
maximum of 7.6 kg (16.8 lbs) and averaging almost
6 kg (13 lbs), with females averaging 4.7 kg
( 10.3 lbs) (Evers et al. 2010). There is considerable
variation in weight (Evers et al. 2010), but spring
migrants would probably seldom weigh much less
than 4.5 kg (10 lbs). Loons probably were taken
opportunistically for food in any subsistence
culture, but they were seldom included among
game birds desired by North American commercial
waterfowlers or sportsmen. Not only are loons
difficult to hunt, but because they are fish-eaters,
there is a perception that they would taste “fishy’’
and not be particularly desirable for the table, much
as fish-eating mergansers (Mergus) are generally
disdained.

Loons are foot-propelled diving birds that
spend their entire lives at the surface of or

718

THE WILSON JOURNAL OL ORNITHOLOGY • VuL 122, No. 4, December 2010

underneath the water, except when flying or
incubating eggs. The Common Loon has a very
high wing-loading and requires a long run across
water to get airborne (Savile 1957, Evers et al.
2010). Common Loons do not experience an
energetically costly wing molt on the lakes of
their northern breeding areas, which freeze over
soon after the young have fledged. Thus, molt is
delayed until they arrive in coastal wintering areas
where all flight feathers are molted simultane-
ously and the birds are flightless until the new
feathers grow in (Woolfenden 1967). Loons are
not only reluctant to fly at .sea in winter because it
is energetically costly, but for considerable
periods of time they are incapable of flight. Loons
are solitary or thinly distributed in winter and do
not respond to decoys as do waterfowl. They also
ride low on the surface of the water, making them
a difficult target for hunters. Pursuit by boat, even
rapid motorized craft, may be ineffectual because
once a loon dives it is difficult to predict when
and where it may briefly surface.

The shooting journal of George Henry Mackay
( 1929) exemplifies the difficulty of shooting loons
under normal circumstances and their greater
vulnerability during their northward flight in
spring. Mackay was an avid hunter of waterfowl
and shorebirds as well as an amateur ornithologist
and conservationist who kept a detailed Journal,
including numbers of all species shot, of his
hunting trips in New England from 1865 to 1922.
Of the thousands of birds recorded in his diary, he
shot only 22 loons, of which 19 were taken at
West Island, Rhode Island, on dates spanning 13
to 25 April, when birds were evidently passing
over land shortly after departing on their spring
flight north.

In spring migration, loons lly northward either
along the coast or overland (Evers et al. 2010), but
once having left the water they gain altitude and
fly far above shotgun range. Birds along most of
the north/south-trending eastern Atlantic Coast
reach such high altitudes over water, perhaps
offshore out of sight of land, and are thus not
su.sceptible to shooting.

Along .Shackleford Banks, however, loons accu-
mulate off the east/west running ocean shoreline in
spring and take off directly over land where they
can be intercepted by gunners. The original
Shacklefordians took advantage of this Haw in the
loons’ usual immunity to put meat on their table.
Markers Island later became the focal point for the
continuation of loon hunting because of its

proximity to the banks and the greater number of
former Shackleford residents there (Eig. 1).

Many of the same conditions apply to Bogue
Banks. Residents of the settlement of Salter Path
in particular were known for shooting and
relishing loons (Stephens 1984, Dudley 1993).
Eikewise, banksmen who settled on the mainland
at Broad Creek crossed the sound to Bogue Banks
for morning loon shoots near what is now known
as Emerald Isle (Stephens 1984). We were told,
however, that more loons cross at Shackleford
Banks than farther west and this may come from a
preference of loons to continue their northward
flight over the waters of Core Sound, rather than
over land.

LOON HUNTING IN CARTERET COUNTY,
NORTH CAROLINA

We constructed the following history from
interviews with local residents conducted in
2008 through 2010, augmented by a few scattered,
generally recondite published sourees. Most
respondents’ personal reeol lections date to the
1940s and early 1950s. Exactly when the loon
hunting tradition arose is unrecorded, but one
respondent’s father told him that it was estab-
lished by at least the mid- 1800s. That seems
reasonable considering that the practiee must have
been in place well before the Shaekleford diaspora
following the hurricane of 1899, as also suggested
by Guthrie (1950). On Markers Island, according
to Paul and Paul (1996:13), the “sport... has been
a part of their lives since the earliest days on the
Island.’’ Guthrie (1993:31) considered it “a
favorite sport... since the conception of ‘Diamond
City’ over a hundred years ago.’’

Loon hunting oecurred in spring, beginning in
March but mostly in April through mid-May. The
usual procedure was for hunters to depart Markers
Island before daylight and leave their boats on the
protected inner shore of Shackleford Banks. They
dispersed on foot across the dunes to points along
the Atlantic beach (Fig. 2) to await the flight of
loons coming off the ocean that usually began at
first light. Hunts occurred spontaneously as there
was no deliberate organization. The recollection
of one resident of Markers Island was that
shooting began so regularly at daybreak that “it
was as good an alarm cloek as you could ask for’’
and that at one point during World War II the
shooting sounded like an invasion. The loons flew
at their lowest altitude early in the morning and
attained greater heights as the day progressed.

Olson Cl al. • LOON IIUN TING IN NOR 11 1 CAROLINA

719

FIG. 2. Markers Island resident Otis C. Willis (b. November 1918, d. June 1996) with the results of a loon hunt in 1943.
Barely visible in the background in the photograph on the left is the Cape Lookout lighthouse and two boats proceeding
northward through “The Drain’’ or Barden Inlet, which was created during the hurricane of September 1933. Thus, the
photograph was taken at the easternmost end of Shackleford Banks. This photograph was incorporated as a patch entitled
“The Loon Hunt” (illustrated in Amspacher 2009:94) in a community quilt on display in the Core Sound Waterfowl
Museum and Heritage Center, Markers Island, North Carolina.

Wind direction had an effect on flights of loons. A
northeast wind was favored for shooting because a
head wind made it easier for loons to take off from
the water and fly directly over the banks. A
southwest tailwind would keep birds flying for a
greater part of the day and was more advanta-
geous for birds once airborne, but those birds
would likely have had to take off farther out to sea
and turn before crossing the banks.

We were told that certain hunters once sank
barrels in the sand for concealment (or shelter?),
but no blind was usually deemed necessary
beyond perhaps crouching behind any convenient
large object that might have washed ashore. Anti-
submarine buoys that had broken loo.se were
mentioned as one such object used during and
immediately after WW 11. At the height of the
loon shooting, hunters spaced themselves so

regularly along the beach that there was hardly a
place a loon could cross the banks without being
in range of .someone's shotgun. At times hunters
were forced to stand behind one another. Loons
are large and tough, and heavy shot, usually #2 or
#4, was used. Even experienced gunners could be
foiled at times, as one former shooter remembered
expending a box of 25 shotgun shells without
bringing down a single loon, although he then
bagged five with his next five shots.

The object was to shoot the birds so they would
fall on the beach or in the dunes behind. Wounded
birds that fell in the water might escape either by
diving and swimming away, or, when dead, might
float off if winds or cunents did not bring them
back to the beach. Birds were not field dressed
and were taken back intact, which meant an
arduous trek back over the island with a heavy

720

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 4. December 2010

load. One respondent recalled a day. when he was
too small to carry a gun, that he had accompanied
his father and uncle to the banks on a morning
when they shot and retrieved 18 loons. That was
probably near the practical upper limit for two
people to carry because 18 loons together would
have weighed well over 90 kg (200 lbs). That
day’s take was transported to the boat dangling
from a pole carried over the shoulders of the
hunters, a common practice, save for one loon that
the youngster dragged behind him. Later, on
Markers Island, the boy sold the birds for 50 cents
apiece or whatever anyone would give him.

There was no fall loon shoot because birds
would be flying high and sometimes at night when
arriving from the north and crossing land. But
once the local appetite for loons had been
established, birds might be taken opportunistical-
ly, other than in spring, on the banks. One
respondent told of becoming adept at shooting
loons on the water with a pistol. Stephens
(1984:124) reported that Bogue Islanders “used
to go to the [Bogue] sound and firelight them in
November,” which would also involve shooting
birds on the surface of the water, but in that case
at night with aid of a lantern.

USES OF LOONS

Food. — The primary impetus for hunting loons
was for food. For 19th century residents of
Shackleford Banks who had survived a winter
on salt fish, root crops, and such grits and flour as
they may have received in trade for mullet and
whale oil in the previous months, a loon
represented a sizeable chunk of fresh meat as
well as a welcome change in diet. Loons do not
migrate north until well after most ducks and
geese have left for breeding areas and the “hardy
group of fisher-folk” on Markers Island “would
rather have a ‘loon-in-the-pot’ than the more
popular goose, canvasback, or red-head duck”
(Paul and Paul 1996:13-14). We made a point of
asking whether cormorants (Plialcicrocorca) were
ever shot for food. It was admitted that a
cormorant would probably be just as good to eat
as a loon but they were not taken.

One of Stephens’ (1984:124) sources said that
“1 love the skin the best in the world” and Dudley
(1993:137), whose .source may have been the
same, mentions that the birds were sometimes
“picked.” The prospect of plucking a loon is
daunting and would require a great amount of
work, so that a person would really have to desire

the skin to go to the effort. Fat is what gives any
kind of meat most of its flavor and most of a
loon’s fat is either attached to the skin or
sequestered among the viscera. A loon with the
skin on would likely be much more fishy-tasting
than one that had the skin and feathers removed,
although Dudley (1993:137) condescendingly
avers that “fishy taste [is] barely discernable to
the natives [of Markers Island and Salter Path].”
By far the greater number of loons prepared for
cooking were skinned and usually only the breast
and thighs were consumed. We have been assured
that the end result did not taste fishy.

We inquired about loons of an acquaintance
who came originally from Cedar Island and who
had returned to Carteret County in retirement. Me
related that he had never shot a loon but would
know exactly what to do if someone gave him
one, and he proceeded to describe the entire
process from carcass to stew bowl. Jocularly
known as “Markers Island turkey” (Amspacher
1987:217), additional sources for loon recipes
(e.g., Stephens 1984, Amspacher 1987) are
basically similar. The meat is soaked in water
with salt or baking soda, browned in fried-out salt
pork, and stewed to a thick gravy with some
combination of onions, potatoes, rutabagas, and
cornmeal dumplings. The process, which was at
times done outside in large iron pots, would
usually take most of a day.

Fishing Lures. — A secondary, but once impor-
tant use of loons, was for fishing lures made from
the bones, which were used for trolling with
handlines, mainly for bluefish (Pomatomus salt-
atrix) and Spanish mackerel {Scomberomorus
nuiculatus). M. Loftin remembers that in the
193()s loon bone lures were almost the only ones
u.sed for trolling in the Beaufort area. In addition
to bones from birds shot for food, during WWII
when tankers along the east coast were being sunk
by German U-boats, bones for lures also came
from loons that died from spilled oil and were cast
up on the beach.

Almost all of our sources, both oral and
published (e.g., Dudley 1993:137, Guthrie 1993,
Paul and Paul 1996) insist that leg bones were
used for the fishing lures and it seems from the
descriptions that the tibiotarsus (drumstick) was
meant. The femur of a loon is extremely short and
curved, and the tarsometatarsus (foot) is flat and
solid, so that neither of those bones would have
been useful. That wing bones were also used is
evident from the report that: “Watermen also

Olson cl al. • LOON HUNTING IN NORT H CAROLINA

721

FIG. 3. Bones (A, B) of Common Loon (Gavia inimer)
and trolling lures (C-E) made from loon bones. A:
complete left tibiotarsus and fibula; B: complete right
humerus; C, D: from Dudley (1993:136); E: from collection
of Roy H. Lewis. Stacy. North Carolina (photograph by J.
R. Humphrey). The notch in D is clearly from hook wear.

want the wing and leg bones of the loons. They
bleach them in the sun and cut them into 2 'A -inch
[63 mm] lengths for fishing lures. The hollow
bones are sometimes the only lures that will catch
bluefish or Spanish mackerel” (Dudley 2001:
160). Furthermore, Dudley (2001:165), relates
that ‘‘in olden days the hunters sold the leg and
wing bones to Cheek’s Hardware in Morehead
City for 10 cents a piece.” The only loon bone
lures that we have been able to trace (Fig. 3 C-E)
were clearly not made from any leg bone. They
are all simple, completely terete cylinders that can
only have been made from the humerus, the upper
wing bone (Fig. 3B). The bone was beveled at one
or both ends and the only one we examined
personally (Fig. 3E) appeared to have been pared
down to a smaller diameter. These bone cylinders
were wired to a hook and the wire was said to
have been twisted at the leading end of the bone in
such a way as to give it a movement under water
that was attractive to fish. An elderly resident of
Markers Island who agreed to demonstrate how to
make a lure from a loon bone protested that the
leg bones presented to him were too small for the

purpose, lending further support to the available
evidence that it was the humerus that was
fashioned into lures. It is possible that the
tibiotarsus was also used, as its flattened proximal
surfaces (Fig. 3A) would probably have imparted
an erratic motion under water that might have
been attractive to fish. Such a lure would have had
thinner walls and been more subject to splitting
than one made from a humerus, which may
account for why none seems to have survived if
they ever existed. The walls of the humerus of a
loon are much heavier and denser than in any
bone of a duck or goose, which, had they been
used for lures, would probably not have lasted
long before being broken.

Following the demise of loon hunting, the
supply of loon bones was greatly diminished and
following WW II, loon bone lures were replaced
by metal spoons and other trolling devices. Loon
bone lures have now nearly completely disap-
peared from tackle boxes and tool sheds of
Markers Islanders.

LORE AND LEGENDS OE LOONS IN
CARTERET COUNTY

It is natural that an activity that was such an
integral part of local traditions as loon hunting
would be incoiporated into local storytelling and
literature. More or less apocryphal stories about
loon hunting on Shackleford Banks are still in
circulation and may be recognized in several
variants. One involves a newcomer to the sport
who was taken to the banks one morning and
witnessed a loon flying parallel to the beach and
having each hunter along the line shoot feathers
out of it until it kept on flying off into the distance
nearly nude, provoking the hope that it would
freeze to death. A version of this is told by Paul
and Paul (1996). As loons were normally shot
Hying over the island, not parallel to it, and loons
are not prone to losing many feathers when shot,
this is likely a carryover from some dove hunter's
tale. A similar anecdote that may be related
involves a champion skeet shooter being invited
to try his prowess on the banks who utterly failed
to bring down a loon. In contrast with those
failures, another story concerns a hunter returning
from the banks to Markers Island with a large
haul. He repaired to the local movie theater,
where he knew a large portion of the population
would be, and announced that he had a skiff full
of loons if anybody wanted one. Whereupon, the

722

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4, December 2010

theater emptied as people made haste for the skiff
to get their loon.

Shooting and eating loons has also been
incorporated into local humor, such as the joke
about the man who was arrested and confessed to
having shot a “seagull.” Before sentencing, the
judge asked the defendant why he wanted to shoot
a gull in the first place, and the answer was that if
it was cooked right it was “nigh bout goods a
loon” (Williamson 2009:44).

Even poetic inspiration has been invoked in
loon lore. We were told by several residents of
Markers Island that they recalled a high school
boy in the 1950s who had written what we were
first told was a story but were later assured was a
poem about loon hunting, but from the perspective
of the loon. We have not been able to trace a copy
of said poem but it must have been an evocative
one for its existence to be recalled more than half
a century later. A more concrete literary contri-
bution to the subject is a mystery novel. Shooting
at Loons (Maron 1994), set in Carteret County, in
which the prediliction of certain Markers Island
fishermen to poach loons illegally and make a
stew of them is mentioned more than once.

TME END OF TME LOON
MUNTING TRADITION

It became illegal to take or possess loons in
1918 with the adoption of the Federal Migratory
Bird Treaty Act. Knowledge and enforcement of
the law would doubtless have been scant and
sporadic at first and the hardship years of the
Great Depression, followed by World War II,
would have made a limited local penchant for
loon meat seem like a paltry Federal offense. But
afterwards the Shackleford loon hunt became a
victim of its own success. Too many people were
participating in the hunt and were being joined to
.some extent by “sportsmen” from “off,” whose
main interest in loons would have been in
extending their season to be afield with a gun.
This created an additional problem of birds left
dead on the beach by shooters who did not share
the locals’ appetite for stewed loon. Such
conspicuous illegal activity could no longer be
ignored.

In the late 1940s an undercover wildlife agent
settled on Markers Island with his family, where
he came to be trusted by local residents, who
conversed with him about fishing and hunting,
including loon shooting. It was later perceived
that his main purpose on Markers Island, however.

had been to set up the inevitable raid on the
Shackleford Banks loon shoot (e.g., Guthrie
1993), a perception that was reinforced by the
tact that he left the island before or immediately
after the raid. The raid occurred on Saturday, 6
May 1950, and was led by Robert Malstead, a
Federal agent of the U.S. Fish and Wildlife
Service, who was joined by “Andrew Jones,
chief state game protector” and nine other state
agents. Some 100 loon hunters were “rounded
up,” but only 78 were actually charged, of whom
“approximately 70” were from Markers Island.
Fifty loons were seized as evidence and Jones
“estimated that 200 more were killed.” We were
told that some 25 or 30 other hunters escaped,
which does not seem unlikely given the low ratio
of agents to hunters. One of our sources, a
teenager at the time, was among the escapees. The
agents estimated that hunters were scattered
“150 feet [45.7 m] apart for a distance of almost
two miles [3.2 km]” and fired from “2000 to
2500 shots” in the first 15 min after the first shot
at 0430 hrs. The math does not add up but it is
clear there was a lot of shooting by a lot of people
on that day. Decades later, Malstead provided his
somewhat dramatized recollection of that day,
complete with a photograph of 10 of the agents
(the 1 1 th was presumably the photographer) with
some of the confiscated loons (Dudley 2001:159-
163) but we have taken all portions in quotes
above from a more immediate source (Anony-
mous 1950).

Those hunters who were convicted in federal
court were fined $25, which is equivalent to about
$225 in 2009 dollars, an amount that would have
made an impression on those who had to earn that
$25 by their labors on the waters. One may
imagine that $1,750 of the entire 1950 economy
of Markers Island had a much more severe impact
than would be conveyed simply by converting that
figure into its modern spending power.

In retrospect, the May 1950 raid was a great
deterrent success for law enforcement. There may
have been no particular stigma (in fact perhaps the
opposite) for most of those apprehended as loon
shooters in Carteret County, but the public could
no longer be unaware that such activity was
egregiously illegal. The enforcers kept up their
vigilance on Shackleford Banks for the next 2-
3 years, after which any .semblance of the loon
shoots of old was abandoned.

Loon shooting persisted and probably still goes
on to a limited extent. Munting on Shackleford

Olsun et al. • LOON HIINTINC] IN NOR I II CAROLINA

723

Bunks required leaving one’s boat on the sound
side to hunt on the ocean side, and, as a federally
protected reserve, shooting there became too
risky. But the occasional loon would still be taken
along the “island shore,’’ meaning the south
shore of Harkers Island. One such incident
resulted in a falling loon hitting a power line on
the way down and knocking out electricity for part
of the island, which “did not go over’’ well
among some of those in positions of authority at
the time. Even after the law had come down hard
in 1950, residents of Salter Path and elsewhere on
Bogue Banks continued to “sneak to the beach
hills and elsewhere in an attempt to get a loon....
Some of them didn’t kill many the whole spring.
But if they killed two or three, they divided them
with their family connections. It was a big thing.
They called in their sister or father or mother or
something like that’’ and still do so “if they can
get a hold of them’’ (Stephens 1984:124). Dudley
(1993:137) reported that at “Salter Path, the
hunters sat on high knolls or dunes anticipating
loon flights over low-lying areas or swashes. With
the pressure of development and high-rise condo-
miniums, this practice has nearly ceased. Today,
loon hunters position themselves around their
homes, or wherever they can obtain visibility of
loon flights, and the shooting continues. Gun shots
are still heard, and game wardens still abound.’’
Given the pace of development on Bogue Banks
since that passage was written, the incidence of
surreptitious loon shooting there has by now
probably all but vanished.

EPILOGUE

An impassioned, even lyrical, paean evoking the
emotions and near ritual importance of the loon
hunt was printed in the Carteret County News-
Times shortly after the great raid of May 1950
(Guthrie 1950). Guthrie contended that the laws
that overregulated hunting had come into effect
because of “fancy” “upstate” sportsmen, where-
as the Carteret County loon shooters participated
in the “only exciting sport left untouched by the
law. Their father had done it, their grandfather
before, even before their forefather had moved
from the outer banks. You weren’t a man until you
had shot a loon.” Hours before daylight, fathers
would rouse sons too small to carry a gun from
their beds to cross the sound to the banks and
participate in the Harkers Island rites of spring. In
the later days of shooting on the banks, “the
excitement of the forbidden hunt [would add] zest

to the sport” (Paul and Paul 1996:13). One of our
respondents, now a pillar of the Harkers Island
community, stirring up recollections over 60 years
old, admitted that although he did not really know
what it was like to be addicted to anything, he
thought that he had been a loon hunting addict.
The motivation was not the thought of a meal, it
was the uniqueness of “the sport.”

Old traditions die hard, but shooting and eating
loons as part of the economy and culture of
coastal Carteret County is now a nearly forgotten
part of the past. The Common Loon has become
the object of such romantic and mythical
sentimentality in most of North America (Evers
et al. 2010) that there could now be no tolerance
for even a small unsanctioned harvest of this
species. The day is not far away when no one on
Harkers Island will recall the taste of stewed loon,
just as there is almost no one alive who
remembers how to fashion a trolling lure from a
loon bone. This interesting and unique local
tradition between humans and birds now exists
almost only as a historical phenomenon that
deserves to go on record.

ACKNOWLEDGMENTS

We are deeply indebted to Ira Lewis, James Lewis, James
Lewis Jr., Richard Lewis, Frank Moore, and James Rose for
sharing their recollections with us. We are most grateful to
Barbara Willis Yeomans for supplying the photographs of
her father, Otis Willis. Roy H. Willis allowed us to study
and photograph a loon bone lure in his collection and Jack
Dudley allowed us to reproduce figures of lures from his
book. We thank Jack Spencer Goodwin, Johanna R.
Humphrey, and Rodney Kemp for assistance with the
manuscript. We are indebted to the staff of the Carteret
County Historical Society. Morehead City, NC. and the
Core Sound Waterfowl Museum and Heritage Center,
Harkers Island, NC, for numerous considerations. We thank
Brian Schmidt and Christina Gebhard for their creative
work with the figures. We also thank G. R. Graves for his
comments on an earlier manuscript and H. M. Reeves for
his most insightful remarks and for providing an important
reference.

LITERATURE CITED

Amspacher, K. W. (Editor). 1987. Island born and bred.
Tenth Printing. Harkers Island United Methodist
Women, Harkers Island, North Carolina, USA.
AM.SPACHER, K. W. (Editor). 2009. Our Down East
memories. Down East community quilt. Quilt Number
1. Core Sound Waterfowl & Heritage Mu,seum,
Harkers Island. North Carolina, USA.

Anonymous. 1950. Game protectors .seize 78 hunters at
Cape Lookout. Carteret County News-Times, Tues-
day, 9 May 1950: page 1.

724

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, December 2010

Dudley, J. 1993. Carteret waterfowl heritage. Decoy
Magazine. Burtonsville. Maryland, USA.

Dudley, J. 2001. Wings; North Carolina waterfowling
traditions. Coastal Heritage Series, Morehead City,
North Carolina, USA.

Evers, D. C., J. D. Paruk. J. W. McIntyre, and J. L.
Barr. 2010. Common Loon Gavia iminer. The birds
of North America. Number 313.

Guthrie, J. L. 1993. The last loon hunt. Mailboat 3^:31-
33. [A short-lived newsletter published on Markers
Island.)

Guthrie, S. 1950. Is it a crime to kill a loon for food?
Carteret County News-Times, Friday, 12 May 1950.

Mackay, G. H. 1929. Shooting journal of George Henry
Mackay. John C. Phillips, Cambridge, Massachusetts,
USA (Undated reprint by Kessinger Publishing [no
city of publication] with 3 intercalated, unnumbered
pages of "Introduction” by H. M. Reeves dated 1 July
2008).

Maron, M. 1994. Shooting at loons. Mysterious Press,
New York, USA.

Paul, M. and G. Paul. 1996. Carteret County, NC:
folklore, facts, and fiction. Beaufort Historical Asso-
ciation. Beaufort, North Carolina, USA.

Savile, D. B. O. 1957. Adaptive evolution in the avian
wing. Evolution 11:212-224.

Stephens, K. H. R. 1984. Judgment land: the story of Salter
Path. Published by the author, Havelock, North
Carolina, USA.

Stick, D. 1958. The Outer Banks of North Carolina, 1584-
1959. University of North Carolina Press. Chapel Hill,
USA.

Williamson, Sonny [C. L). 2009. Salt spots for breakfast.
Published by the author, Marshallberg, North Carolina,
USA.

Woolfenden, G. E. 1967. Selection for a delayed
simultaneous wing molt in loons (Gaviidae). Wilson
Bulletin 79:416-420.

The Wilson Journal of Oniitlioloi'y 122(4):725-737, 2010

MIGRANT SONGBIRD SPECIES DISTRIBUTION AND HABITAT USE
DURING STOPOVER ON TWO ISLANDS IN THE GULE OE MAINE

REBECCA W. SUOMALA,' SARA R. MORRIS, ^

KIMBERLY J. BABBITT,' AND THOMAS D. LEE'

ABSTRACT. — We compared the distribution of migrant bird species between two islands in the Gulf of Maine to
e.xamine it ditterences in habitat resulted in differences in avian species composition and relative abundance during
stopover. Ninety-one species were captured on both islands and those species captured on only one island were either
breeding species or rare visitors to the islands. Ditterences in bird species distribution between islands were species-specific
and consistent among sampling periods for nearly all species. Twelve species were captured more frequently on Star Island
and 1 1 species more frequently on Appledore Island. Stopover species distribution appeared to be related to habitat
structure, vegetation, diet, and habitat area. Scrub-shrub/open habitat breeding species and forest breeding species were not
evenly distributed between islands. Island use was most clo.sely associated with breeding habitat. All but two of the eight
species that breed in scrub-shrub or open habitat were captured more frequently on Star Island. Ten of the species more
common on Appledore Island breed in forested habitat. Nine of the 1 1 species more common on Appledore Island are area-
sensitive in breeding areas, suggesting potential area sensitivity during migration. Differential habitat use indicates a large
number of stopover sites in a wide variety of habitats are necessary to meet migration needs of passerine species. Received
17 November 2006. Accepted 22 April 2010.

Suitable stopover habitat is critical to a
successful migration (Moore et al. 1995) and loss
of stopover habitat may be a factor in long-term
population declines (Sherry and Holmes 1995,
Finch and Wang 2000, Hutto 2000). Migrants
have increased risk of mortality during migration
(Bairlein 1992; Moore et al. 1992, 1995; Hutto
1998; Sillett and Holmes 2002.) without habitat
that can satisfy their needs to replenish critical
energy supplies, reorient (Baird and Nisbet 1960),
escape adverse winds and weather, recover from
muscle fatigue or injury, and avoid the dehydra-
tion of daytime flight. More information is needed
on habitat requirements of passerines during
migration (Moore et al. 1992, 1995; Sherry and
Holmes 1995; Hutto 2000; Petit 2000) to
understand what constitutes valuable stopover
habitat and influences habitat selection (Moore
et al. 1995, Dunn 2000).

Migrants show differential habitat use including
species-specific di.stribution among stopover hab-
itats (Hutto 1985, Moore et al. 1990, Winker
1995), but the basis for habitat selection is not
clear. Finch and Wang (2000) reported bird
densities during migration were affected by

' Department of Natural Resources and the Environment,
206 Nesmith Hall, University of New Hampshire, Durham,
NH 03824, USA.

’Canisius College, Department of Biology, 2001 Main
Street, Buffalo, NY 14208, USA.

Current address: New Hampshire Audubon Society, 84
Silk Farm Road, Concord, NH 03301, USA.

■* Corresponding author; e-mail: bsuomala@nhaudubon.org

vegetation structure and habitat type, but few
studies have examined fine-scale habitat relation-
ships during migration (Rodewald and Britting-
ham 2004). There is some evidence that songbirds
may select stopover habitat with vegetative
characteristics similar to their breeding habitat
(Parnell 1969, Petit 2000). Other factors affecting
stopover habitat use may include suitability to a
species’ foraging method (Bairlein 1992), food
availability (Hutto 1985, Moore et al. 1995),
dietary shifts (Parrish 2000), predation and
competition (Moore et al. 1992), and site
characteristics including area (Moore et al.
1995), temperature, and humidity (Winker 1995).

Star and Appledore islands are part of the Isles
of Shoals on the New England coast, USA. They
provide a unique opportunity to examine the
relationship between habitat and migrant stopover
patterns because large numbers of songbirds stop
at the archipelago during spring and fall migra-
tions (Morris et al. 1994, 1996), and the islands
are <1 km apart. Weather and body mass can
influence selection of stopover sites (Moore and
Simons 1992) and the potential confounding
effect that different weather conditions have on
migration behavior can be eliminated by compar-
ing two adjacent sites. Both islands are catego-
rized as maritime shrub thicket (Sperduto and
Nichols 2004), which encompasses a range from
low-growing shrubs to tall thickets; comparison of
these islands presents an opportunity to examine
the relationship of birds and stopover habitat at a
fine scale. Species richness and abundance in

725

726

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4. December 2010

migration may be correlated with patch size
{Martin 1980, Blake 1986, Somershoe and
Chandler 2004), as has been shown during the
breeding season when number of species and
individuals increases with area (Freemark and
Merriam 1986, Blake and Karr 1987, Faaborg et
al. 1995). Star Island is smaller than Appledore
Island allowing consideration of area effects on
species composition during stopovers.

Our objectives were to examine: (1) whether
there were differences in species richness, com-
position, and distribution between the islands; and
(2) whether a species’ breeding habitat or food
habits were related to stopover habitat use. We
predicted more species would be captured on
Appledore Island and that habitat differences
between islands would result in differences in
species composition and frequency of capture.

METHODS

Study Site. — The Isles of Shoals is a group of
nine small islands and several ledges in the Gulf
of Maine, 14.5 km southeast of Portsmouth, New
Hampshire (42° 58' N, 70° 36' W), and 9.7 km
from the nearest point of the mainland. Appledore
Island, Maine is the largest of the islands at 33.6 ha
and Star Island, New Hampshire is 13.4 ha. The
minimum shore-to-shore distance between Star
and Appledore islands is ~0.7 km and the study
sites were 1 .6 km apart.

Bird Surveys. — Bird surveys were conducted
simultaneously on both islands in 1999 and 2000
using mist nets during spring and fall migration
following the protocol established for the Apple-
dore Island Migration Banding Station (Morris et
al. 1994, 1996). There were four sampling
periods: 11 May-8 June 1999, 10 May-8 June
2000, and 16 August-30 September 1999 and
2000. Weather permitting, mist nets (6 or 12 X
2.6 m, 4 shelves, 30 mm mesh) were operated
during daylight hours with the nets opened just
before sunrise, closed at sunset, and checked
every 30 min throughout the day. Up to five nets
were operated on Star Island and up to 10 nets
were operated on Appledore Island. Captured
birds were brought to a central location on each
island and banded with USGS aluminum bands.

Habitat Surveys. — We collected habitat data al
net sites rather than at random locations distrib-
uted over each entire island. There is extreme
landscape variability on both islands with habitat
ranging from rocky splash zones or mowed lawns
to thick scrub-shrub relatively close to the nets.

We characterized the natural vegetation that
surrounded the mist nets to ensure data were
representative of the habitat used by the species
captured. These data should not be considered
representative of the habitat on the entire island.

Vegetative structure and composition at each
site were measured using a modified version of
James and Shugart (1970) and Winker et al.
(1992). We established eight circular survey plots,
each centered on a pair of mist nets (or a single
net, if there was only one in a given area). There
were three plots on Star Island and five on
Appledore Island covering all net locations. Each
plot was 0.04 ha (radius = 1 1 .35 m) and was
divided into two semicircles, separated by the net
lane (1-2 m) to remove the non-vegetated lane
from analysis.

We established six 1.5 m-diameter circular
subplots in each plot using a stratified random
design so each subplot was entirely within the
larger plot. Subplots were located by extending a
random distance into alternating halves of the plot
from six equidistant points (3.94 m apart) along
the diameter of the plot, excluding the outer 1 .5 m.
We established a 1 m-wide transect in each half of
the plot to measure shrub density. Transects were
parallel to the net lane and 0.3 m from the edge of
the net lane to reduce the influence of trimming
along net lanes.

We defined trees as woody stems with a
diameter at breast height (dbh) >3.0 cm and
shrubs as any woody vegetation <3.0 cm dbh and
over 0.75 m in height. We recorded live tree
species, height, and dbh (0.1 -cm diameter tape) in
each plot. We recorded only dbh for dead trees.
We recorded species and height for each live
shrub stem in the transect, and counted stems
separately if they branched within 0. 1-0.2 m
above the ground. We counted the number of
stems for dead shrubs. Heights for live trees and
shrubs were measured to the nearest 0.1 m.
Separate measurements were taken for each
species in a cluster.

We measured herbaceous cover, canopy cover,
and vegetation density in each subplot. We
visually estimated herbaceous cover in each
subplot to the nearest 5%. We visually estimated
canopy cover from the center of each subplot and
recorded cover to the nearest 5% at six l-m height
classes beginning with 0-1.0 m (to account for
low woody plant cover that functions as canopy
for ground-feeding birds). We used a density
board (Nudds 1977) placed at the center of a

Sinmuila cl ciL • MIGRATION STOPOVER DISTRIBUTION AND HABITAT USE

727

subplot to measure vegetation density and esti-
mated percent cover from 2 m distant, rather than
the recommended \5 m (Nudds 1977), because of
the extreme density of the shrub vegetation. We
chose a random direction using eight possible
points of the compass and rejected points that
were outside the full plot after walking 2 m from
the board. We recorded vegetation cover to the
nearest 5% in each of the four sections of the
density board (0-0.5, 0.5- 1.0, 1.0-1. 5, 1.5-2.0m).

All habitat data were collected between 6 May
and 10 June 2000 by RWS to eliminate variability
among observers. Subplot measurements were
obtained between 5 and 10 June to ensure that
vegetation density and canopy cover measure-
ments were at the same time after leaf out.

Data Analysis. — We used Systat Version 10
(SPSS Inc. 2000) for statistical analyses. The
analyses, except bird species richness, include
only those species that do not breed on the Isles of
Shoals (i.e., only stopover migrants). We used raw
counts of birds captured for all Chi-square tests.

We used a heterogeneity Chi-square analysis
(Zar 1999) to test for differences in species
distribution between islands and analyzed each
sampling period separately. We used Freeman-
Tukey deviates in Systat to examine which
species were responsible for significant Chi-
square tests. We removed those species from the
analysis until there was no difference {P > 0.05)
between islands. Expected values for individual
species were calculated using the total number of
migrant birds captured on each island. The
heterogeneity test was the prime analysis used in
examining differences between islands, even
though expected values for some .species were
below the desired criteria for Chi-square tests. We
u.sed a Chi-square goodness-of-fit test for each
species as a .secondary analysis to compare
number of individuals captured between islands
without violating the expected value criteria.
Expected values were >5 in all tests except
Purple Pinch [Carpodacus purpureu.s) and Savan-
nah Sparrow (Passerciilus sandwichensi.s) in
spring 2000.

We followed Jones et al. (2002) and the
recommendations of Gotelli and Ellison (2004),
and did not use Bonferroni corrections in the
analyses to reduce the likelihood that real
differences or relevant patterns would be missed.
Bonferroni coiTections are not as practical for
diverse communities requiring multiple statistical
tests (Moran 2003) such as the number of migrant

birds in this study. It is unlikely the results from
all tests will be spurious (Moran 2003).

Species were included in the distribution analysis
if there were > 1 0 individuals banded on at least one
island in any two sampling periods, and were
analyzed in only those sampling periods when their
numbers met this minimum criterion. The only
exception was Purple Pinch, which is an irruptive
migrant present in large numbers during fall 1999
only. We included it in the spring 2000 analysis,
although only eight individuals were banded, to
examine if the return migration in spring had the
same distributional pattern. Recaptures were not
included in the distribution analysis.

We used a Mann-Whitney f/-test to compare
mean number of tree and shrub stems per plot
between islands for each species with more than
20 stems on one of the islands. We computed the
mean per plot for each island and compared
means between islands using a Mann-Whitney
f/-test for the habitat data; live tree and shrub
density (all species combined), dead tree and
shrub density, tree and shrub height, and tree dbh.
A mean for each individual plot was calculated
first for the latter three measures, which were used
to calculate the mean per plot for the entire island.
We divided tree heights into five 1-m height
classes beginning with 2.0 m and shrub heights
into five 0.5-m height classes beginning with
0.5 m. We compared the distribution of trees and
shrubs by height class between islands with a
heterogeneity Chi-square analysis, and used a
Mann-Whitney (/-test to examine differences
among mean number of stems per plot in each
height class. We divided tree dbh into seven size
classes: 3-3.9, 4-4.9, 5-5.9, 6-6.9, 7-10.9, II-
15.9, and >16.0 cm. The larger size classes with
few trees were combined to create suitably sized
groupings. We compared the distribution of trees by
dbh class between islands with a heterogeneity Chi-
square analysis. We used ANOVA to compare
percent herbaceous cover between islands and
M ANOVA with associated univariate tests, when
the M ANOVA was significant, to compare mean
vegetation density at four height classes and canopy
cover at six height cla.s.ses between islands.

We assigned habitat types to each bird .species
for breeding habitat to compare species differenc-
es between islands with habitat associations in
breeding areas. Habitat classifications were: early
successional, forest, forest edge, low forests at
high elevations and in bogs, scrub-shrub, and open
areas (including grasslands and emergent wet-

728

THE WILSON JOL'RNAL OF ORNITHOLOGY • Vol. 122, No. 4. December 2010

lands). We also examined vegetation type and
area sensitivity in breeding areas. Area sensitivity
information was obtained from Robbins et al.
(1989) and Birds of North America species
accounts (Poole and Gill 1993-1997, 1998-
2002); breeding habitat information was primarily
from DeGraaf and Yamasaki (2001) supplement-
ed by Ehrlich et al. (1988). We classified fall
trugivory into four categories using the mean
percent of fruit in fecal samples during fall
migration on Block Island, Rhode Island (Pairish
1997); low 0-29, medium 30-59, medium-high
60-89, and high 90-100%.

RESULTS

Bird Species Richness. — We banded 5,443 birds
(3,833 migrants) on Star Island and 8,372 (6,604
migrants) on Appledore Island during the four
sampling periods. Mean (± SD) captures for Star
Island were 1,234 ± 28.3 in spring and 1,487 ±
255.3 in fall, and for Appledore Island were 2,140
± 302.6 in spring and 2,046 ± 233.3 in fall.
Ninety-nine species were captured on Star Island
and 102 on Appledore Island; 91 of the species
were captured on both islands. Mean (± SD)
species richness for all sampling periods com-
bined was 68 ± 4.65 for Star Island and 74 ± 3. 1 6
for Appledore Island. Mean (± SD) species
richness during spring was 64 ± 0.71 on Star
Island and 71 ± 0.71 on Appledore Island; during
fall it was 72 ± 0.71 on Star Island and 76 ± 2.12
on Appledore Island.

Eight species were captured only on Star Island
and I I species only on Appledore Island. Some
species were captured infrequently and were rare
visitors to the Isles of Shoals or unusual captures
in songbird mist nets. More of these “accidental”
species would be expected on Appledore Island
due to probability (more net hours). Species
richness on the two islands was identical at 70
species with 100% overlap in species composition
between the islands when we removed breeding
and accidental species from the analysis.

Bird Species Disirihiilion. — The distribution of
migrant species between islands differed in each
sampling period (spring 1999: = 242.27, df =

28, P < 0.001 ; .spring 2000: f = 206. 19, df - 33,
P < O.OOl; fall 1999: yj = 325.92, df = 31. R <
O.OOl; fall 2000: f = 400.69, df = 34, P <
O.OOl). The difference between islands was
attributed to a fairly consistent group of species
(Table I). Twelve species were captured more
frequently than expected on Star Island (scientific

names in Table 1): Traill’s Flycatcher, Ruby-
crowned Kinglet, Cedar Waxwing, Magnolia
Warbler, Yellow-rumped Warbler, Wilson’s War-
bler, Yellow-breasted Chat (fall, not tested in
spring). Savannah Sparrow (spring, not tested in
fall), Lincoln’s Sparrow (fall but not spring).
Swamp Sparrow, White-throated Sparrow (fall
but not spring), and Purple Finch. Eleven species
were captured more frequently than expected on
Appledore Island (Table I ): Red-eyed Vireo,
Brown Creeper (fall, not tested in spring), Veery
(fall but not spring), Swainson’s Thrush (spring
but not fall). Black- throated Blue Warbler, Black-
and-white Warbler, American Redstart (fall but
not spring), Ovenbird, Northern Waterthrush,
Mourning Warbler (spring but not fall), and
Canada Warbler (spring but not fall).

Habitat. — Thirteen species of shrubs and eight
species of trees were recorded on Star Island and
15 species of shrubs and nine species of trees on
Appledore Island (Table 2). The most common
shrub species in the netting area on Star Island
were chokeberry, raspberry, and rose; the most
common trees were winterberry, sumac, and
bayberry (scientific names in Table 2). The most
common shrubs in the netting area on Appledore
Island were chokecheri'y, winterberry, and aiTow-
wood; the most common trees were chokecheiry,
serviceberry, and arrowwood. Only raspberry and
winterberry were in the top five species on both
islands; the other dominant species were typically
scarce or absent on the other island. Baybeiry and
huckleberry were recorded only on Star Island,
and apple, arrowwood, black cherry, and pin
cherry were recorded only on Appledore Island.

The mean number of trees on Appledore Island
was higher than on Star Island, but this difference
was not significant for all heights combined
(Table 3). One Star Island plot had 228 trees,
but the mean of the two remaining plots was only
65 trees. No plot on Appledore Island had <142
trees. The distribution of trees over five height
classes differed (y~ — 375.5, df = 4, R < 0.001)
between islands (Fig. 1). Appledore Island had
more trees per plot in the highest three classes and
Star Island had more in the lowest class. The
difference between islands was not significant for
the 3-3.9 m height class. There were also fewer
dead trees per plot on Star Island than Appledore
Island (Table 3).

Trees on Star Island had a smaller mean dbh per
plot (range = 3.0-10.7 cm) than Appledore Island
(range = 3.0-33.2 cm; Table 3). The distribution

Snoniala cl at. • MIGRAI ION STOPOVER DISTRIBUTION AND HABITAT USE

729

ot trees over seven dbh classes dit't'ered (x“ =
50.6, dt = 6, P < O.OOl) between islands with
fewer trees on Star Island in all dbh classes, and
fewer than 10 trees in the four classes greater than

6.0 cm. Star Island had shorter mean tree height
per plot than Appledore Island (Table 3). There
were only six trees taller than 4.0 m on Star Island
(3 plots), whereas Appledore Island had 529 that
were 4.0 m or taller (5 plots).

Shrub density was higher on Star Island than
Appledore Island, primarily because of a differ-
ence in the number of small-diameter shrubs
(Table 3). Star Island had more dead shrubs per
plot than Appledore Island but the difference was
not significant (Table 3). Overall mean shrub
height did not differ between islands (Table 3).
However, distribution of shrubs over five height
classes (Fig. 1) differed (yj = 729.5, df = 4, P <
0.001) between islands. Star Island had signifi-
cantly more shrubs per plot in the lowest three
classes and Appledore had significantly more in
the highest height class. There was no difference
(P = 0.881) between islands in the 2-2.4 m class.

There was a slight difference between islands in
vegetation density for all four heights combined
(P4.43 = 2.48, P = 0.058). Star Island had
significantly higher mean percent cover than
Appledore Island in the lowest two height classes
(0-0.5 and 0. 5-1.0 m) but there was no significant
difference in the two upper heights.

Percent canopy cover differed (p6.4i = 4.25,
P = 0.002) between islands. Star Island had
higher percent cover in the lowest two height
classes from 0.0 to 2.0 m, and Appledore Island
had higher percent cover in the highest three
classes (3.1 m and higher); one height class (2.1-

3.0 m) did not differ between islands. Star Island
had mean (± SD) herbaceous cover of 52.5 ±
34.2% and Appledore Island had a mean cover of
35.8 ± 35.0%, but the difference was not
significant (Pi,46 = 2.69, P = 0.1 12).

Comparison of Bird Distribution with Habi-
tat.— There were fewer trees on Star Island and
they were smaller in height and dbh than
Appledore Island which had only six trees taller
than 4.0 m (3 plots), whereas Appledore Island
had 529 that were 4.0 m or taller (5 plots). Star
Island had more shrubs in all height classes except
the tallest (>2.5 m). and had the highest
vegetation density in all height classes except
the highest (1. 5-2.0 m). Appledore Island had
more canopy cover except in the 0-2.0 m range.
Thus, Star Island more closely resembled a typical

scrub-shrub habitat in structure while taller
thickets with higher canopy cover characterized
Appledore Island. This difference allowed us to
compare differences in bird distribution between
islands in relation to habitat (Table 4).

There were significant differences in distribu-
tion between islands iy' = 4.5, df = 1, P =
0.033) when bird species were grouped based on
breeding habitat as ( 1 ) scrub-shrub or open, or (2)
forest. Six of the eight species that breed in
scrub-shrub or open habitat were captured
significantly more frequently on Star Island
(Table 4); one of the other two showed no
difference between islands (Chestnut-sided War-
bler) and the other was more numerous on
Appledore Island (Mourning Warbler). Six of
32 forest-breeding species were more common
on Star Island, 10 were more common on
Appledore Island, and 16 showed no difference
between islands (Table 4). Nine of 12 forest-
breeding species that are area-sensitive in
breeding areas were more common on Appledore
Island, and the other three species showed no
difference between islands (Table 4).

DISCUSSION

Star Island more closely resembled a typical
scrub-shrub habitat in structure with denser shrub
vegetation, more trees, and greater canopy cover
<3.0 m. Appledore Island had greater tree
density, canopy cover, and shrub density >3.0 m.
Bird species richness and composition varied
minimally between islands; however, relative
abundance of some species varied greatly proba-
bly because of vegetation differences. These
results are consistent with Rodewald and Britting-
ham (2004) who reported relative abundance
during migration varied among habitats despite
broad habitat use by all species.

Differences in species distribution between Star
and Appledore islands were species-specific and
consistent between sampling periods. Bairlein
(1983) and Moore et al. (1990) also reported high
consistency of species-specific habitat distribu-
tions and that migrants show selective use of
certain habitats. There was no consistent relation-
ship to winter habitat but there was to breeding
habitat.

Most migrant species that breed in scrub-shrub
and open habitat were more numerous on Star
Island, except for Chestnut-sided and Mourning
warblers. The shorter vegetation on Star Island

TABLE 1. Bird species distribution between Star Island (SI) and Appledore Island (AP) based on banding data from 1999 and 2000. H/g = Chi-square heterogeneity and
goodness-ot-tit analyses; H = species responsible for the difference between islands. A — indicates no difference between the islands and blank = capture numbers insufficient for
analysis in that sampling period. Island = location where relatively more of that species were captured.

730

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4, Decenib^f 2010

00

c/5 O
CQ

W)

I I I

*

* I

W) I

X T

o (N

— rs CM

^ 1

0.

c/5 1

<

*

*

* 1

*

OX) 1

OX)

X T

O CN

cu

<

Dh

c/0 <;

* * I

W) OX) I

c/5

*

*

Ot)

I I

^ T ^

— ro IT)

— m — — — I

+ + * I I I

OX) OX) OX) I I I

T ^ T T T T

* * *

* OX) * I
OX) OX) 1

^ T ^ T

I I I

I

I I

sO O ^ >/5 O
CN — (N (N CN

mONcncNicN-^'— —

(N

OS‘/5’OOOsD(N<7N'OOsCNU^i'*:^'aNfN'0^00
— — — ^ —

1

Oh

j

1

Oh

cu

Oh

0-

1

1 ,

1

c/5

<

c/5

<

<

<

<

c/5

+

*

*

+

*

*

*

*

*

*

*

*

*

*

*

*

-S?

OX)

OX)

OX)

OX)

OX)

OX)

OX)

1

OX)

1

T

T

K X

^ ^ ^ ^

T

T

X

cn

o

r--

m

r--

m

00

iTj

oo

00

r-

—

X

lo

(N

—

(N

ON

o

r^.

(N

rn

fO

(N

(N

00

r--

__

t/5

(N

r^,

On

C^i

>/5

r-

O

ro

Y*

rs»

00

o

to

*

*

*

*

*

*

1

*

*

*

OX)

OX)

OX)

OX)

OX)

'S?

X

1

T

1

T

T

(0^

Y-

m

r-

oo

tT)

X

o

O

<N

i/'j

ON

(N

OX

r<'i

(ON

(N

IT)

rsi

X

o

00

(N

(ON

r^<

X

C/O

*

I *0X) I

Q-

<

*

OX)

I I

rx 00 00

O lO ON

Oh

<

*

OX) I

^ 7

-H I a, -H

c/5 I < C>0

^ 0- -H

C/5 < C/5

*

*

OX)

I I

*

+ *
OX) OX)

1^ ^
7 T

*

00

I I

4- *

Ot) 00

I I I

7 T T

D-

<

■x-

*

00 ,

0. Oh a. >-H CL,
<<<£/)<

* * **
6XJ OXJ M OJj au

^ ^ X I ^

T T

o

o

o

00

X

O

o

in

o

O

00

ro

O'

in

•Y

<N

X

l-H

~H

in

ox

o

o

X

(N

Y"

o

,

X

in

ro

X

O'

ON

in

CN

X

(N

in

*

*

*

*

*

*

1

*

*

OX)

*

*

*

OX)

1

OX)

OX)

1

OX)

1

OX)

1

OX)

T

T

T

T

T

X

in

ON

1

m

o

X

r--

o

o

00

O)

ro

r--

ox

m

ro

00

(N

(N

in

(N

— H

m

— H

o

ro

in

ox

O)

in

o

X

o

X

in

,

o

r-

ox

o

O'

X

O'

rO

(N

ox

—

m

‘ncNrNkoooNDmoNONONooo
CN— — OOiOr^i/5rn(N — 'O

r--OroND — or^ooosooooN

I I I

I I OX) OX) I

I I I I I

*

* * ,
OX) OX) I

*

OX)

I I X I

ONrOsOC^XOOCNr-
— ^ O rn — IT)

(N — Tt

C C3

J-

G

to

^ ..5

•S

S ^
s ^

^ 5

U kj

-C)

-s:

Ck

-5:

■S

5

C3

kj

C:

9

ki c^

*«; “C:

to

2

■33

C3

s:

Q

Cj

s:

C

s

s:

to

Q k ^ =5 -S

^ ^
^ S ^

s:

H*

c

^ cS -
k. g C

G ?=

G

s

G

G

Ck G

O (ON CN (N

G

§■

G

G 'k
00 k<

Jh.

2 ’

So

G
G

§1

Sli^li-'t^Uc^QcGUcSi^^QQCiCiCiQQ^

G

g

G

G

G

G

to

Q

2

.G

c

G

-G

G

C)0
G

§■ s 2

Hh .—. *-

LO Lo Lo

-2 -2
CL 2

CL

2 ,2

g g
<0 23
CL L-

o ^

I o

^ I

E p

C «
o ^

z w

Urn

G

JZ

G

3

G

u.

E

G

X

"O

G

G

3

G

G

X

E

c/)

cc

"g

E

Urn

>-

H

o o

V (L)

-C

o

-C

OX) aj
c TTr

■s ^ s

(D

E p

> >

"O C3

3

Z

CL '

“O

(U

“O *

c

5 =
S ^
o

5

JZ

H

-i- ^
o ^

ra 2

p -C

"2 p (j cj

OJ ^ Si - C G ^

prOTD-O o-TT-C^ L) P
Ucd-— j^ggl-Odg^

_iujQacLCc:DiQaOoi>oo

2^

s:i

c3

3

03 ^

G

C u.

E 3

^ S

^ JZ

"O c/5 ^ G

p ca o .x:

u z z u

_G

JZ

u.

3

G

^ QQ

_G

-D

u.

C3

C

J=;

3 (U

^ 3

_ O

C ^
OX) G
C3 ^

S 5

_G
X)

^ p -P

^ o

"p -o
CL G

E ^

3

V X _
^ Y X
o ^

— G G

G _E J2
> CO CQ

_G

X

3

^ c

3 2

o

E

3

G

X B
c3 X

— "O

a 3

G G
3 3

t:

3

c/i

*o

> G

j- G u
C X ^
c3 "£

5> ^

> >

G

3

OX)

c

CO

E 5 -a

3 t/) E

O ■" §

CQCD<OZS^U

TABLE 1. Continued.

Siiofiiahi e! al.

MIGRATION STOPOVER DISTRIBU I ION AND HABITAT USE

731

00

*

*

OiJ

'sj-

(N

on

* *

* * -h
ojj b/j Oij

^ ^ T

OrO'^vOt^jOO
(N m

00 oor^oo'^oso

(N — (N —

* *

* I I *

W) I I W)

^ 56 T ^

O 04 CN On (N lO
— (N

— C'! — r-vosor^cN

m — ON

on I on

X T

bX)

X

i ^ i
T T T

*

bc

— 00 r- m u-j (N cN
(N so (N — '

O — NO ON On OO 00
— (N CN

'S?

X

X T T

O NO ON O

(N »0 On m Tf O
— r~<

c

13

g-

5

C3

S Ci.

s

?i0

CN

3

Do

O

fc

rt

Cl

C/5

JC

r3

C

C

>

cs

on

> ^
O s

t X)

C3

O- P

O > ■

^ o

^ §
00 g_

t/> c/5

!s Q-
o e

O .
■- & :
J OO ;

O I
"S o

o -g

(U E
£i S ti.
•g E u
“ B-

O C3 3
Qi PO Q-

with its greater density at lower heights and
searcity of taller trees may be more suitable for
scrub-shrub species than Appledore Island. Bair-
lein (1983) also reported differences in species
along a gradient of vegetation density, and
Rodewald and Brittingham (2004) reported
shrub-sapling breeding species were most abun-
dant in early successional and edge habitat during
fall stopover.

Ten of the 1 1 migrant species captured
significantly more frequently on Appledore Island
breed in forested habitat with the exception being
Mourning Warbler. However, not all forest-
breeding species were more numerous on Apple-
dore Island. Nine of the 10 forest-breeding species
that were more numerous on Appledore Island are
area-sensitive in breeding areas. No area-sensitive
species was more numerous on Star Island. Other
studies have reported evidence of area-sensitivity
during migration (Martin 1980, Cox 1988,
Somershoe and Chandler 2004). The influence
of patch size on stopover habitat use is also
influenced by microhabitat and landscape (Petit
2000). The influence of habitat area in our study
was not reliably separated from habitat structure.
For example, area-sensitive species may be less
flexible in migration stopover habitat than other
forest breeding species and, thus, more common
on Appledore Island because of its more forest-
like habitat.

Microhabitat factors likely contributed to the
distribution patterns of forest-breeding species.
American Redstart breed in habitat with a dense
sapling (2. 5-9. 9 cm dbh) understory (DeGraaf and
Yamasaki 2001). Appledore Island had far more
trees in this size class than Star Island. Appledore
Island also had more and larger trees than Star
Island. Black-and-white Warblers were recorded
only on large limbs of larger trees during spring
migration in Kentucky (Mason 1979) and were
more common on Appledore Island. Brown
Creeper was the only forest-breeding species
more common on Appledore Island that is not
area-sensitive. It forages primarily on large-
diameter trees with deeply furrowed bark (Hejl
et al. 2002). This feeding substrate is almost
totally lacking on Star Island but is present on
Appledore Island.

Fruit availability also seemed to affect the
distribution of species. Three of the six species
more numerous on Star Island that were not scrub-
shrub or open habitat breeders appeared to be
influenced by presence of fruit-bearing shrubs.

732

THE WILSON JOURNAL OF ORNITHOLOGY • Vul. 122, No. 4, December 2010

TABLE 2. Tree and shrub species composition on Star and Appledore islands sampling plots (0.04 ha) in 2000.
Differences in means based on Mann-Whitney CZ-tests for between islands. A — indicates insufficient numbers for analysis
(20 or fewer stems). Species with <10 stems (total trees and shrubs combined) on both islands were not analyzed*^.

Common name

Scientific name

Star Island

Shrubs Trees

Appledore Island

Shrubs Trees

Difference

shrubs trees

Mean
per plot

Mean

per plot

Mean

per plot

Mean
per plot

Apple

Pyrus malus

0.6

4.0

Arrowwood

Viburnum recognitum

—

—

56.0

23.6

+

+

Northern bayberry

Myrica pensyl van ica

35.3

12.0

—

—

+

+

Blackberry

Rubus allegheniensis

12.3

—

9.4

—

—

Black cherry

Primus serotina

—

—

—

10.0

*

Chokeberry”

Pyrus spp.”

430.7

—

0.8

—

Chokecherry

Primus virginiana

3.3

9.0

87.8

55.0

*

*

Huckleberry

Gaylussacia baccata

11.3

—

—

—

—

Pin cherry

Pyrus pensylvanica

—

—

2.8

11.8

—

Poison ivy

Toxicodend ran radicans

19.0

1.0

—

0.4

-h

Raspberry

Rubus ideaus

296.0

—

30.2

—

—

Ro.se'’

Rosa spp.'’

158.0

—

13.4

—

*

Serviceberry

Amelanchier spp.

1.0

6.7

2.0

46.6

*

Staghorn sumac

Rhus typhina

1.0

19.3

5.4

9.6

—

-

Winterberry

Ilex verticillata

76.3

68.3

57.2

36.2

-

-

+ = 0.05 < P < 0.10. * P < 0.05, ** P < 0.01, and — = differences not significant.

“ Red and purple chokeberry (P. arbutifolia. P. florihimda) and possible hybrids.

Virginia rose {R. virginiana) on both islands, pa.sture rose [R. Carolina) only on Appledore Island.

*■ Species with fewer than 10 stems: carrion-flower (Smilax herbacea). common elder iSambiiciis canadcn.sis). bittersweet nightshade iSalaniim dulcamara), red
cedar Uuniperus virginiana). red maple (Acer rubrum), Virginia creeper (Parthenoci.'isus quinquefolia).

The Yellow-rumped Warbler feeds on bayberry
during fall migration (Parrish 1997); this shiub
was moderately common on Star Island but absent
from the Appledore Island plots. The other two
species, Cedar Waxwing and Purple Finch, were
two of the four highly frugivorous species present
in the fall (Table 4). Additionally, the Yellow-

breasted Chat was both highly frugivorous and a
scrub-shrub breeder, and was also more numerous
on Star Island.

Both vegetation structure and species compo-
sition can affect habitat suitability and use (Rice
et al. 1984, Moore et al. 1995, Finch and Wang
2000). There were differences in plant species and

TABLE 3. Differences in vegetation physiognomy between Star Island and Appledore Island based on a Mann-
Whitney U-test of means in each category; df = I for all tests. There were three plots (0.04 ha) on Star Island and five plots
on Appledore Island. Data collected in 2000.

Category

Star Lsland

Appledore Island

u

p

# stems

Mean/plot ± SD

# stem.s

Mean/plot ± SD

Live trees

Density

358

1 19.3 ± 95.30

986

197.2 ± 36.69

lO.OO

0.456

Height' (m)

3. 1 ± 0.48

4.3 ± 0.44

15.00

0.025

Dbh” (cm)

4.4 ± 0.44

5.4 ± 0.71

14.00

0.053

Dead trees

13

4.3 ± 2.08

120

24.0 ± 14.05

15.00

0.025

Live shrubs

Density

3,139

1,046.3 ± 80.71

1,341

268.2 ± 132.14

O.OO

0.025

Height' (in)

1.3 ± 0.27

1.5 ± 0.23

1 1 .00

0.297

Dead shrubs

919

306.3 ± 231.65

437

87.4 ± 65.89

2.00

0.101

“ Mean per plot for each island was calculated from means of each individual plot on that island.

Suonia/a e/ al. • MIGRATION STOPOVER DISTRIBUTION AND HABITAT USE

733

(A)

o

Q.

<D

CL

W

0)

0)

L.

c

ro

<u

■ Star
□ Appledore

2 0-2,9 3.0-3.9 4.0^.9 5. 0-5.9

Height class (m)

> 6.0

(B)

900.0

800.0

o

700.0 -

Q.

(1)

600.0 -

CL

CO

n

500.0 -

3

CO

400.0 -

c

(ti

(D

300,0 -

200.0 -

100.0 -

0.0 -

0.5-0.9 1.0-1. 4 15-1,9 2.0-2. 4

Height class (m)

I 1

1 T

r

LI

dii_

■ Star
□ Appledore

>2.5

FIG. I , Distribution of (A) trees and (B) shrubs by height class on Star Island. New Hampshire, and Appledore Island,
Maine, in 2000. Bars are means ± SD; * = P < 0.05, and + = 0.05 < P < 0. 10.

vegetation .structure between islands; the relative
importance of these two factors is difficult to assess
(Rice et al. 1984). Such was the case with Yellow-
rumped Warbler, which feeds on bayberry during
fall migration (Parrish 1997) and was more
common on Star Island, the only island where
bayben-y occuiTed. However, the effect of habitat
structure cannot be distinguished from the presence
of bayberry (Kwit et al. 2004). The primary factor
is likely structural differences for most species

because migrants must adapt to a wide variety of
plant species along their migration route, and
species differences between islands were primarily
based on breeding habitat staicture.

Seasonal shifts in habitat have been ob.served at
other sites (Winker et al. 1992, Weisbrod et al.
1993, Wang et al. 1998), but only Blue-headed
Vireo in our study had a seasonal switch between
islands. Thus, most species in this study were
captured more frequently on the same island

734

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122. No. 4. December 2010

TABLE 4. Ecological classifications for each avian species. SI = Star Island, AP = Appledore Island, - = no
ditterence between islands, and blank = species not present in sufficient numbers to analyze. Area sensitivity without
parentheses is from Robbins et al. (1989); in parentheses is from individual species accounts in The Birds of North America.

Species

More

captures

spr/fall

Area

sensitivity

(breeding)

General

category*

Breeding habitat

Veg Special

type'’ characteristics

Frugivory

(fall)*

Northern Flicker

/-

No

F-FE

M

Eastern Wood-Pewee

No

F-FE

D-M

Yellow-bellied Flycatcher

-/-

(No)

F-SF

C

Forests and bogs

Medium

Traill’s Flycatcher

SI/SI

(No)

SS-FE

D

Med-hi

Least Flycatcher

-/-

(Yes?)

ES-F-FE

D-M

Medium

Eastern Phoebe

/-

(No)

FE

D-M

Med-hi

Blue-headed Vireo

AP/Sl

No

F

C-M

Medium

Philadelphia Vireo

/-

(No)

F-FE-ES

D-C-M

Med-hi

Red-eyed Vireo

AP/AP

Yes

F-FE

D-M

Med-hi

Red-breasted Nuthatch

-/-

(No)

F

C-M

Mature forests

Brown Creeper

ZAP

No

F

C-D-M

Mature forests

Low

Golden-crowned Kinglet

/-

(No)

F

C

Low

Ruby-crowned Kinglet

Sl/Sl

(No)

F

C-M

Low

Veery

-ZAP

Yes

F

D

Dense understory

Med-hi

Swain.son’s Thrush

APZ-

(Yes?)

F-FE-ES

C-M

Med-hi

Cedar Waxwing

SI/Sl

No

FE

D

FruitZberry producers

High

Nashville Warbler

-Z-

(No)

ES-F-FE

C-D-M

Medium

Northern Parula

-Z-

Yes

F-FE

C-D-M

Mature forests

Medium

Chestnut-sided Warbler

-z-

No

SS-ES-FE

D

Dense low growth

Magnolia Warbler

SIZSI

No

F-ES

C-M

Dense understory

Med-hi

Black-throated Blue Warbler

APZAP

Yes

F

D-M

Dense saplingZshrub

Medium

understory

Yellow-rumped Warbler

SIZSI

(No)

F

C-M

Med-hi

Black-throated Green Warbler

-Z-

No

F

C-M

Mature forests

Medium

Blackburnian Warbler

-Z

(No)

F

C-M

Mature forests

Blackpoll Warbler

-z-

(No)

F-SF

C

Low northern spruce

Medium

Black-and-white Warbler

APZAP

Yes

F

D-M

Forests with saplings

Low

American Redstart

-ZAP

No (Yes?)

ES-F-FE

D

Thick sapling understory

Low

Ovenbird

APZAP

Yes

E

D-M

Mature forests

Medium

Northern Waterthrush

APZAP

Yes

F-FE

D

WetlandZpond edges

Med-hi

Mourning Warbler

APZ-

(No)

SS-ES

D-M

Dense saplingsZshrubs

Med-hi

Wilson's Warbler

SIZSI

(No)

SS-SF

D

Low

Canada Warbler

APZ-

Yes

F

D-M

Den.se understory

Low

Yellow-brea.sted Chat

ZSI

No

SS-FE

D

No overstory

High

Savannah Sparrow

SIZ

(No)

0

G

Lincoln’s Sparrow

-ZSI

(No)

SS-EE-SF

D-M

Med-hi

Swamp Sparrow

SIZSI

(No)

SS

D-G

Wetlands

Medium

White-throated Sparrow

-ZSI

(No)

FE-ES-SF

D-C-M

Brushy understory

Med-hi

Ro.se-breasted Grosbeak

-Z

Yes

FE-F

D

High

Baltimore Oriole

-Z-

No

FE

D

Purple Finch

SIZSI

(No)

F-FE

C-M

High

“ Habitat categories: ES = early succe.ssional. F = forest. FE = forest edge, SF = short forests at high elevation and in bogs. SS = scrub/shrub, and O = open
areas, grasslands, emergent wetlands.

Vegetation type (veg type), C = coniferous. D = deciduous, G = grassland/sedge/reed, and M = mixed coniferous-deciduous.

' Frugivory ba.sed on mean percentage of fruit in fecal samples (Parrish 1997): low 0-29. medium .70-.S9. medium-high (med-hi) 60-89. high 90-100%. and blank
= not in analysis.

during both .spring and fall. The consi.stency of
habitat .selection by a wide range of migrant
species between spring and fall supports the
hypothesis that habitat .selection in passerines is
endogenous (Bairlein 1983, Morton 1990, Winker

1995). Migrant distribution might be expected to
change between spring and fall, after the large
proportion of young migrants have had more
exposure to a variety of habitats on both their
southward and return migrations, if habitat

Siiomala cl al. • MIGRATION STOPOVER DISTRIBUTION AND HABITAT USE

735

selection was learned. Additionally, the correla-
tion ot stopover habitat with breeding habitat
indicates innate preferences as suggested by
Moore et al. { 1995).

The apparent association with breeding habitat
by migrants during stopover is consistent with
Petit’s (2000) analysis of habitat selection in five
studies. Smaller physiognomic and plant species
differences between islands were associated with
differences in species distribution between is-
lands. Our study suggests stopover habitat use was
related to habitat structure, plant species, diet
(extent of frugivory), and habitat area, but the
predominant factor(s) may differ among species.

Habitat differences between these two islands
were smaller in structure and species type than
described in other stopover habitat studies (e.g.,
Bairlein 1983, Moore et al. 1990, Rodewald and
Brittingham 2004); but differences between is-
lands were associated with differences in bird
species distribution. Nocturnal migrants typically
land when it is still dark and are unlikely to make
stopover decisions based on these small habitat
differences. Migrants may conduct habitat explo-
ration flights during the day (Moore and Simons
1992, Wiedner et al. 1992), but few birds banded
on one island were recaptured on the other island
(Suomala 2005) indicating little movement after
initial landing. We suggest .several mechanisms to
explain the detection of small habitat differences:
(1) migrants may be landing at the Isles of Shoals
later in the morning than migrants at other locations
and are able to see detailed habitat features; (2)
migrants may be able to see certain features in the
dark; and/or (3) there is a non-visual component of
stopover habitat selection, similar to the vocal
social cuing that Betts et al. (2008) reported was
more important than vegetation structure in habitat
choice for breeding site .selection.

CONSERVATION IMPLICATIONS

Priority stopover habitat for the bird communi-
ty may be difficult to define due to differing
habitat preferences of individual species (Hutto
2000, this study). Protection of stopover sites with
a wide variety of habitats, both forest and scrub-
shrub, is important to meet the needs of migrant
passerines (Weisbrod et al. 1993, Rodewald and
Brittingham 2004, this study).

The specie.s-specific differences in habitat use
in this study suggest migrants are not using the
Isles of Shoals as simply emergency stops,
landing at random out of desperation. Scrub-shrub

habitat is naturally occurring on the coastal plain
and a good source of berry-producing shrubs,
which may be one of the best areas for migrants to
deposit fat (Parrish 1997). Other studies have
demonstrated the importance of edge and shrub-
land habitats with fruit to fall migrants (Parrish
1997; Suthers et al. 2000; Rodewald and Britting-
ham 2002, 2004), but the management of scrub-
shrub habitat for migrants may be overlooked as
an important priority.

Land use changes along the coast are occurring
rapidly with resulting deterioration of stopover
habitat (Mabey and Watts 2000). Coastal devel-
opment creates a landscape that is often inhospi-
table to migrating birds. Conservation efforts for
migrant songbirds should encompass the remain-
ing small patches of natural habitat that may be
critical to migrant survival (Blake 1986). Educa-
tion and local land use policies may be effective
tools for maintaining small habitat patches on
private lands (Mabey and Watts 2000).

ACKNOWLEDGMENTS

This project was made possible by support from the Star
Island Corporation and New Hampshire Audubon, grants
from the Eastern Bird Banding Association and the
University of New Hampshire Graduate School, and the
generous donations of many individuals. This effort would
not have been possible without help of volunteers at New
Hampshire Audubon, the Star Island banding station, and
the Appledore Island Migration Banding Station, supported
in part by the Shoals Marine Laboratory. This paper is
contribution # 14 of the Appledore Island Migration
Banding Station and contribution # 153 of the Shoals
Marine Laboratory.

LITERATURE CITED

Baird, J. and I. C. T. Nisbet. 1960. Northward fall
migration on the Atlantic Coast and its relation to
offshore drift. Auk 77:1 19-149.

Bairlein, F. 1983. Habitat .selection and a.ssociation of species
in European pas.serine birds during southward, post-
breeding migrations. Omis Scandinavica 14:239-245.
Bairlein, F. 1992. Morphology-habitat relationships in
migrating songbirds. Pages 356-369 in Ecology and
con.servation of neotropical migrant landbirds (J. M.
Hagan and D. W. Johnston. Editors). Smithsonian
Institution Press, Washington, D.C., USA.

BETT.S, M. G., A. S. Hadley, N. Rodenhouse, and J. J.
Nocera. 2008. Social information trumps vegetation
structure in breeding-site selection by a migrant
songbird. Proceedings of the Royal Society of London.
Series B 275:2257-2263.

Blake, J. G. 1986. Species-area relationship of migrants in
i.solated woodlots in east-central Illinois. Wilson
Bulletin 98:291-296.

736

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

Blake, J. G. and J. R. Karr. 1987. Breeding birds of
isolated woodlots: area and habitat relationships.
Ecology 68:1724-1734.

Cox, J. 1988. The influence of forest size on transient and
resident bird species occupying maritime hammocks of
northeastern Florida. Florida Field Naturalist 16:25-34.

DeGraaf, R. M. and M. Yamasaki. 2001. New England
wildlife: habitat, natural history, and distribution.
University Press of New England, Hanover, New
Hampshire, USA.

Dunn, E. H. 2000. Temporal and spatial patterns in daily
mass gain of Magnolia Warblers during migratory
stopovers. Auk 117:12-21.

Ehrlich, P. R., D. S. Dobkin, and D. Wheye. 1988. The
birders handbook: a field guide to the natural history of
North American birds. Simon and Schuster Inc., New
York, USA.

Faaborg, j., M. Brittingham, T. Donovan, and J.
Blake. 1995. Habitat fragmentation in the temperate
zone. Pages 357-380 in Ecology and management of
neotropical migratory birds: a synthesis and review of
critical issues (T. E. Martin and D. M. Finch. Editors).
Oxford University Press, New York, USA.

Finch, D. M. and Y. Wang. 2000. Landbird migration in
riparian habitats of the middle Rio Grande: a case
study. Studies in Avian Biology 20:88-98.

Freemark, K. E. and H. G. Merriam. 1986. Importance of
area and habitat heterogeneity to bird assemblages in
temperate forest fragments. Biological Conservation
36:1 15-141.

Gotelli, N. j. and a. M. Ellison. 2004. A primer of
ecological statistics. Sinauer Associates Inc., Sunder-
land, Massachusetts, USA.

Hejl, S. j., K. R. Newlon. M. E. McFadzen, J. S. Young,
AND C. K. Ghalambor. 2002. Brown Creeper (Certhia
ainericana). The birds of North America. Number 669.

Hutto, R. L. 1985. Seasonal changes in the habitat
distribution of transient insectivorous birds in south-
eastern Arizona: competition meditated? Auk 102:
120-132.

Hutto, R. L. 1998. On the importance of stopover sites to
migrating birds. Auk 115:823-825.

Hutto, R. L. 2000. On the importance of en route periods
to the conservation of migratory landbirds. Studies in
Avian Biology 20:109-1 14.

James, F. C. and H, H. Shugart Jr. 1970. A quantitative
method of habitat description. Audubon Field Notes
24:727-736.

Jones, J., C. M. Francis, M. Drew, S. Fuller, and M. W.
S. Ng. 2002. Age-related differences in body ma.ss and
rales of mass gain of passerines during autumn
migratory stopover. Condor 104:49-58.

Kwn, C., D. J. Levey, C. H. Greenberg, S. F. Pearson, J.
P. McCarty, S. Sargent, and R. L. Mumme. 2004.
Fruit abundance and local distribution of winter
Hermit Thrushes (Cutharus I’ullatu.s) and Yellow-
rurnpcd Warblers (Dendroiea eoronata) in South
Carolina. Auk 121:46-57.

Mabey, S. E. and B. D. Watts. 2000. Conservation of
landbird migrants: addressing local policy. Studies in
Avian Biology 20:99-108.

Martin, T. E. 1980. Diversity and abundance of spring
migratory birds using habitat islands on the Great
Plains. Condor 82:430-439.

Mason, W. 1979. Habitat selection by the Parulidae during
spring migration along the South Fork Creek in
Glasgow, KY. The Kentucky Warbler 55:39^2.

Moore, F. R. and T. R. Simons. 1992. Habitat suitability
and stopover ecology of neotropical landbird migrants.
Pages 345-355 in Ecology and conservation of
neotropical migrant landbirds (J. M. Hagan and D.
W. Johnston, Editors). Smithsonian Institution Press,
Washington, D.C., USA.

Moore, F. R., P. Kerlinger, and T. R, Simons. 1990.
Stopover on a gulf coast barrier island by spring trans-
gulf migrants. Wilson Bulletin 102:487-500.

Moore, F. R., S. A. Gauthreaux Jr., P. Kerlinger, and
T. R. Simons. 1992. Stopover habitat: management
implications and guidelines. Pages 58-69 in Status and
management of neotropical migratory birds (D. M.
Finch and P. W. Stangel, Editors). USDA, Forest
Service, General Technical Report RM-229. Rocky
Mountain Station, Fort Collins, Colorado, USA.

Moore, F. R., S. A. Gauthreaux Jr., P. Kerlinger, and
T. R. Simons. 1995. Habitat requirements during
migration: important link in conservation. Pages 121-
144 in Ecology and management of neotropical
migratoi7 birds: a synthesis and review of critical
issues (T. E. Martin and D. M. Finch. Editors). Oxford
University Press, New York, USA.

Moran, M. D. 2003. Arguments for rejecting the sequential
Bonferi'oni in ecological studies. Oikos 100:2.

Morris, S. R., M. E. Richmond, and D. W. Holmes. 1994.
Patterns of stopover by warblers during spring and fall
migration on Appledore Island, Maine. Wilson Bulle-
tin 106:713-718.

Morris, S. R., D. W. Holmes, and M. E. Richmond. 1996.
A ten-year study of the stopover patterns of migratory
passerines during fall migration on Appledore Island.
Maine. Condor 98:395^09.

Morton, E. S. 1990. Habitat segregation by sex in the
Hooded Warbler: experiments on proximate causation
and discussion of its evolution. American Naturalist
135:319-333.

Nudds, T. D. 1977. Quantifying the vegetative structure of
wildlife cover. Wildlife Society Bulletin 5:1 13-117.

Parneix, j. F. 1969. Habitat relations of the Parulidae
during spring migration. Auk 86:505-521.

Parrish, J. D. 1997. Patterns of frugivory and energetic
condition in Nearctic landbirds during autumn migra-
tion. Condor 99:681-697.

Parrish, J. D. 2000. Behavioral, energetic, and conserva-
tion implications of foraging plasticity during migra-
tion. Studies in Avian Biology 20:53-70.

Pinrr, D. R. 2000. Habitat use by landbirds along Nearctic-
neotropical migration routes: implications for con.ser-
vation of stopover habitat. Studies in Avian Biology
20:1.5-33.

Poole, A. and F. Gii.i, (Editors). 1993-1997. The birds of
North America. The Academy of Natural Sciences,
Philadelphia, Pennsylvania and The American Orni-
thologists' Union, Washington, D.C., USA.

Siioniala el al. • MIGRATION STOPOVER DISTRIBU TION AND HABITAT USE

737

Poole, A. and F. Gill (Editors). 1998-2002. The birds of
North America. The Birds of North America Inc.,
Philadelphia, Pennsylvania, USA.

Rice, J., B. W. Anderson, and R. D. Oiimart. 1984.
Comparison of the importance of different habitat
attributes to avian community organization. Journal of
Wildlife Management 48:895-91 1.

Robbins, C. S., D, K. D.awson, and B. A. Dowell. 1989.
Habitat area requirements of breeding forest birds of
the Middle Atlantic States. Wildlife Monographs 103.

Rodewald, P. G. and M. C. Brittingham. 2002. Habitat
use and behavior of mixed species landbird flocks
during fall migration. Wilson Bulletin 114:87-98.

Rodewald, P. G. and M. C. Brittingham. 2004. Stopover
habitats of landbirds during fall: use of edge-
dominated and early-successional forests. Auk 121:
1040-1055.

Sherry, T. W. and R. T. Holmes. 1995. Summer versus
winter limitation of populations: what are the issues
and what is the evidence. Pages 85-120 in Ecology
and management of neotropical migratoi'y birds: a
synthesis and review of critical issues (T. E. Martin
and D. M. Finch. Editors). Oxford University Press,
New York, USA.

SiLLETT, T. S. and R. T. Holmes. 2002. Variation in
survivorship of a migratory songbird throughout its
annual cycle. Journal of Animal Ecology 71:296-308.

SOMERSHOE. S. C. AND C. R. CHANDLER. 2004. Use of oak
hammocks by neotropical migrant songbirds: the role
of area and habitat. Wilson Bulletin 1 16:56-53.

Sperduto, D. D. AND W. F. Nichols. 2004. Natural
communities of New Hampshire. New Hampshire

Natural Heritage Bureau and The Nature Con.servancy,
Concord, New Hampshire, USA.

SPSS Institute Inc. 2000. Sysiat Version 10.0. SPSS
Institute Inc.. Chicago, Illinois, USA.

SuOMALA, R. W. 2005. Migrant songbird species distribu-
tion, behavior, and habitat use during stopover: a
comparison of two islands in the Gulf of Maine.
Thesis. University of New Hampshire, Durham, USA.
SUTHERS, H. B., J. M. Bickal, AND P. G. Rodewald. 2000.
Use of successional habitat and fruit resources by
songbirds during autumn migration in central New
Jersey. Wilson Bulletin 1 12:249-260.

Wang, Y., D, M. Finch, F. R, Moore, and J. F. Kelly.
1998. Stopover ecology and habitat use of migratory
Wilson's Warblers. Auk 1 15:829-842.

Weisbrod, A. R., C. J. Burnett, J. G. Turner, and D. W.
Warner. 1993. Migratory birds at a stopover site in
the Saint Croix River Valley. Wilson Bulletin
105:265-284.

Wiedner, D. S., P. Kerlinger, D. A. Sibley, P. Holt, J.
Hough, and R. Crossley. 1992. Visible morning
flight of neotropical landbird migrants at Cape May.
New Jersey. Auk 109:500-510.

Winker, K. 1995. Habitat selection in woodland Nearctic-
neotropic migrants on the Isthmus of Tehuantepec. I.
Autumn migration. Wilson Bulletin 107:26-39.
Winker, K., D. W. Warner, and A. R. Weisbrod. 1992.
Migration of woodland birds at a fragmented inland
stopover site. Wilson Bulletin 104:580-598.

Zar, j. H. 1999. Biostatistical analysis. Fourth Edition.
Prentice-Hall Inc., Upper Saddle River, New Jersey,
USA.

The Wilson Journal of Ornithology 122(4):738-743, 2010

CONCENTRATED MIGRATORY MORNING ELIGHT AT LAKE
PONTCHARTRAIN, LOUISIANA, USA

PETER H. YAUKEY'

ABSTRACT. — I describe large volume morning flights of migrant songbirds in autumn along the south shore of Lake
Pontchartrain, Louisiana, USA. These flights have at times exceeded 10,000 birds/hr since their discovery in 2002, ranking
the site among the continent’s highest volume regular migratory songbird concentration locations. Migrants move in
"reverse” direction northeastward up a peninsula in Lake Pontchartrain, and fly across a 9-km water gap to reach the north
shore, usually struggling into a headwind. Flights are dominated by Yellow and Yellow-rumped warblers (Dendroica
petechia and D. coronata). Eastern Kingbird {Tyrannus tyranniis). Indigo Bunting {Passerina cyanea), Cedar Waxwing
(Bomhycilla cedronim), and American Robin (Tiirdus migratorius) during their respective migratory peaks. These species
have exceeded 800 birds/hr, and 70 other small landbird species have been recorded. Flights of >100 birds/hr were more
frequent after nights with more hours of northerly winds. Received 24 August 2009. Accepted 30 April 2010.

The Nearctic-neotropical migratory system
involves the movement of billions of birds
southward within the Western Hemisphere, some
staying within temperate latitudes and others
traveling to the Neotropics. The major barrier to
the large contingent of migrants headed from
eastern North America to tropical America is the
Gulf of Mexico, which many species cross in a
single extended flight. The dynamics of this trans-
Gulf movement have been most extensively
studied in spring. Northbound migrants normally
arrive on the central Gulf Coast during daylight,
often after having flown >6 hrs longer than in a
normal overland nocturnal flight. Arriving trans-
Gulf migrants normally over-fly coastal regions
and disperse into habitats >100 km inland,
causing near-coastal areas to be labeled the
“coastal hiatus” (Lowery 1945). Their reasons
for passing far inland before alighting are
unknown, but the paucity of forest cover in the
wetland landscape adjacent to the coast is a likely
factor. Fall movements are even more poorly
understood. Able (1972) noted that weather
conducive to forward migration is more limited
in fall than spring when prevailing southeasterly
winds impede crossing the Gulf except when
broken temporarily by the passage of cold fronts
with their accompanying northerly tailwinds.

The objectives of this paper are to: (1) report
the existence of an unusual migratory movement
phenomenon at Lake Pontchartrain, Louisiana,
and (2) discuss the insights it provides into the
dynamics of autumnal migration on the Gulf
Coast. The volume of bird movement discovered
to occur here ranks this newly discovered site as

' Department of Geography, University of New Orleans,
New Orleans, LA, 70148, USA; e-mail: pyaukey@uno.edii

one of the major migratory songbird concentration
locations in the Western Hemisphere.

METHODS

Study Site. — Lake Pontchartrain is ~100 km
north of the Gulf of Mexico, from which it is
separated by urban New Orleans and expansive
wetlands (Fig. 1). The city extends 10-20 km
southward from the central portion of the lake,
while wetlands border the lake’s southwestern
and southeastern flanks, the latter on the Bayou
Sauvage National Wildlife Refuge. The landscape
outside the southern huiricane-protection levee of
the city changes abruptly to bottomland hardwood
and swamp forests, which transition southward
through shrub-dominated wetlands into open fresh
and brackish marshes, and finally to salt marsh
that is flanked in some places by bamer islands
and headlands. Several large semi-tidal lakes and
inlets penetrate nearly as far northward as the city.
The corridor of the Mississippi River winds
through the city before heading southeast to the
Gulf, its <3 km fringe of flanking field and
woodlots forming the largest contiguous area of
dry land south of the city.

Lake Pontchartrain is roughly oval in shape,
39 km north-south and 65 km east-west at its
widest, but is pinched by the extension of a
peninsula northward near its eastern end that
reduces the ovewater distance to 9 km. This
peninsula ends at two adjacent points of land. South
Point and Pointe aux Herbes (Fig. 1C, D), from
which railroad and highway bridges depail for the
north shore. Migratory songbird movements are
greatest on this peninsula, especially progressing
northward up its western Hank and departing from
either of these two points to cross the lake.

738

Ycmkey • MORNING FLIGHT OF MIGRANTS

739

FIG. 1. Study area showing locations where major morning movements have been recorded: (A) University of New
Orleans; (B) Bayou Sauvage National Wildlife Refuge bicycle path; (C) South Point; (D) Pointe aux Herbes. Map by Steve
Stevens and John Adams.

Northward curvature of the Gulf shoreline wraps
open wetland and coastal lakes north to the east of
Lake Pontchartrain, and essentially no forest cover
lies east or southeast of its eastern end. Pine {Pinus
spp.) forest is absent south of Lake Pontchartrain,
but is common on its north shore.

Field Procedures. — Preliminary scouting of
morning flight was conducted by visiting several
coastlines and land-marsh interfaces near Lake
Pontchartrain in fall 2002. Systematic collection
of data began in fall 2003, when volunteer
observers were stationed at Pointe aux Herbes
(Fig. ID) to document movement of birds across
the water gap and find correlations with weather
patterns recorded at Lakefront Airport (Fig. 1
between sites A and B). Additional visits were
made to sites C and D through 2009, mostly on
days with northerly winds with the frequency of
visitation declining after Hurricane Katrina in
2005. Observers counted birds passing their
observation point in 15-min intervals starting near
dawn, and typically lasting 1-3 hrs.

I used surface wind data beeause of their hourly
resolution, despite their differing at times from
winds aloft at the altitude of nocturnal migration.
The number of nocturnal hours of northerly
(between WNW and ENE) winds was tabulated
preceding each flight count because, in the case of
the cold front passages that usually preceded the
diurnal movements noted, the number of hours of
post-frontal conditions before daybreak may
affect the opportunity for nocturnal migrants to
be wind-drifted south of the lake. Bird counts
conducted over >30 min were used to calculate
hourly passage rates.

RESULTS

The presence of significant diurnal song bird
movements along the Lake Pontchartrain penin-
sula was not discovered until 22 November 2002,
although local observers had infrequently visited
this peninsula in previous years and occasionally
noted migratory birds moving in anomalous
directions. A continuous heavy movement of

740

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

TABLE 1. Peak rates of hourly and daily passage for selected species at Lake Pontchartrain, Louisiana, USA.

Species

Peak hourly passage (date)

Peak daily passage'' (date)

Eastern Kingbird

American Robin

Cedar Waxwing

Yellow Warbler

Yellow-rumped Warbler

Indigo Bunting

25.000 (5 Sep 2003)

12.000 (28 Nov 2004)

3.500 (22 Nov 2002)

800*' (31 Aug 2004)

4.500 (22 Nov 2002)

880 (15 Oct 2002)

35.000 (5 Sep 2003)

20.000 (28 Nov 2004)

7.000 (22 Nov 2002)

1.200 (23 Aug 2009)

12,500 (23 Nov 2002)

1.000 (15 Oct 2002)

■' All counts were obtained during <3 hrs of counting.
E.xtrapolated from 41 min observation.

songbirds northeastward along the western flank
of the peninsula (Fig. IB) was witnessed on the
morning of 22 November 2002 following a cold
front passage. Frequent timed counts over 2 hrs
produced an estimated total of 25,000 birds
dominated by American Robins (Tiirdus migra-
torins). Cedar Waxwings (Bombycilla cedrorum),
and Yellow-rumped Warblers (Dendroica coro-
nata). These birds were flying low into the NNE
wind, using a narrow tree row that separated the
lake levee from the landward marshes as a
windbreak. The next morning another 18,500
birds were estimated to pass up the shore in 3 hrs
at this site, dominated by the same species. Over
the next few weeks, as the fall migration
dwindled, similar but smaller bird movements
were tracked up the peninsula and found to be
crossing the lake at South Point and Pointe aux
Herbes (Fig. 1C, D). The lakeshore along this
western flank of the peninsula is not readily
accessible, being behind gated roads within the
Bayou Sauvage National Wildlife Refuge.

A few morning songbird flights detected earlier
in the same fall at a site 0.5 km south of the
lakefront and near the western base of the peninsula
.seemed in retrospect to have represented similar
patterns of movement, but for neotropical migrant
species (Fig. lA). On 5 September 2002, —1,400
Ea.stern Kingbirds (Tyninnus tyrannus) were count-
ed in 1 hr flying ENE parallel to the lakefront, and
on 15 October, —1,350 small songbirds, predom-
inantly Indigo Buntings (Passerina cyanea), had
been seen passing northward toward the nearby
lakeshore in 68 min; their behavior upon reaching
the shore was not identified.

Flights exceeding 1 00 birds/hr were observed on
37 days each fall during 2()02-2()()4, and occurred
as early as 8 August and as late as 28 January.
Sub.sequently, flights have been detected as early as
13 July 2008 (proceeding up the Missi.ssippi River

south of the lake). The largest flights have
occurred during peak passages of Yellow Warblers
(Dendroica petechia) and Eastern Kingbirds in
August-September, Indigo Buntings in October,
and American Robins, Cedar Waxwings, and
Yellow-rumped Warblers in November. Counts
during northeast winds have repeatedly exceeded
1,000 birds/hr, and six individual species have
peaks of >500/hr with robin and kingbird counts
suipassing 10,000/hr (Table 1). Species identifica-
tion of many of the less frequent species was
difficult until 2004, when more visits were made to
site C, which has the best viewing conditions.
Seventy-two species of small migratory land birds,
mostly nocturnal migrants, were identified as they
crossed or attempted to cross the lake at sites C and
D from 2003 to 2009.

All significant movements recorded at the
lakefront have been northeastward, except for
seven oriented at a bearing west of north in
departing across the water from the western point
(South Point, site C). These directions result in a
longer overwater crossing and have only occuiTed
when morning winds were from west of north.

Wind directions were analyzed for flights from
2002 through 2004, the most consistent period of
field observation (Eig. 2). Elights peaking at
>100 birds/hr occuired with differing frequency
when different numbers of hours of northerly
winds had occurred the night before (yy = 23.0,
n = 108, P = 0.0017). They occurred on a
majority of mornings following >9 hrs of
northerly winds, but less than a tenth of those
with <2 hrs. Northerly winds are unu.sual on the
Gulf Coast except after frontal passages, which
were infrequent in August but frequent each
November during the study. Thus, the northerly
winds conducive to mass movements at the lake are
more common during late fall periods dominated
by temperate zone migrants. Another source of

Yaukcv • MORNING FLIGHT OF MIGRANTS

741

70

0-2 3-5 6-8 9-11

Nocturnal hours of north wind

FIG. 2. Percentage of days with morning movements
>100 birds/hr that followed nights with different hourly
frequencies of northerly winds.

northeast winds in early fall has been the occasional
tropical cyclone that has entered the eastern Gulf of
Mexico; Tropical Storm Henry produced northeast
winds during the massive kingbird movement on 5
September 2003 (Table 1), while 550 km to the
ESE near the Florida coast.

DISCUSSION

Why are there regular northward movements of
migrant birds across Lake Pontchartrain on
mornings after fall cold fronts? There are three
potential explanations. (1) The movements repre-
sent forward migration of birds attempting to head
eastward, but detouring northward to avoid the
open Gulf and marshes ahead to the east. (2) The
movements represent migrants not attempting
forward progress, but searching locally for
suitable stopover habitat in which to rest or feed.
(3) The birds are making a corrective movement
to compensate for wind drift or navigation eiTors.

The first hypothesis of detoured forward
migration is appealingly simple, but contradicts
the basic widespread recognition that circum-Gulf
migration in Louisiana is westbound (e.g., Lowery
1974). This has been established by radar (Able
1972), visual studies of morning flight (Duncan
and Weber 1985), and cage orientation studies
(Sandberg and Moore 1996). Other weaknesses
include: (1) birds on some dates move primarily
NW, (2) it seems unlikely that large numbers of
birds would migrate forward into headwinds (yet
most heavy Lights are during NE winds), and (3)
birds seldom approach the overwater crossing up

the east side of the peninsula as would be
expected if birds were detouring from the south
as proposed.

The second hypothesis, that birds are .seeking
foraging and resting habitat is consistent with the
patchiness of forest and scrub habitats south of the
lake. The overwhelming dominance of north-
bound birds at the peninsula could be a conse-
quence of shoreline configuration — only north-
bound birds are able to approach the peninsula
overland, and the lakeshore concentrates birds
approaching from that direction. At least one
species. Tree Swallow (Tachycineta bicolor),
appears to commute across the lake in its daily
search for food. This winter resident regularly
forms nocturnal roosts in late fall in sugar cane
{Saccharum ojficinaruin) fields 80 km to the
southwest that may include over a million birds
(Seymour 2009). I attributed large Tree Swallow
flights (e.g., 18,000 in 1.5 hrs on 6 November
2004) to roost dispersal and omitted them from
this analysis, but it is difficult to imagine most
other songbirds crossing the lake as commuters.
The capacity of swallows for strong flight as they
depart across the lake contrasts with the greater
struggle of many other songbirds. Some species
have so much difficulty that they appear to make
repeated failed departures in short order; for
instance 100 departures by Golden-crowned and
Ruby-crowned kinglets (Regulus satrapa and R.
calendula) in 7.5 min on 2 November 2005
appeared to involve many repeat attempts to cross
the lake. It is difficult to imagine that searching
behavior would have such a strong northeastward
impulse (or into-wind impulse) that birds would
be willing to cross 9 km of water into a headwind,
expending energy and potentially exposing them-
selves to accipiters and falcons that are common
on the peninsula in autumn. This would also fail
to explain why other resident songbirds do not
engage in such flights (only Rock Pigeons
\Coiumba livia\ do so regularly, and usually
southbound), and why Lights of Yellow-rumped
Warblers and American Robins dwindle precipi-
tously in mid-winter despite remaining numerous
in the region. Searching for food may contribute
to cross-lake movements by some species, but
does not appear to be a sufficient explanation for
most.

The third hypothesis, corrective movement, is
favored in part because it has been cited as the
explanation for coastal retreat movements alon^
the Atlantic shoreline of northeastern North

742

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4, December 2010

America including at Cape May, New Jersey
(Baird and Nisbet 1960) and in Sweden at
Falsterbo, a site where diurnal reverse-directional
movements are also massive in fall (Alerstam and
Ulfstrand 1975, Alerstam 1978, Akesson et al.
1996). Swainson’s Thrushes {Cathanis ustulatus)
on the northern Gulf Coast have also been
observed to orient inland in flight cages in the
early morning when they are in lean body
condition in fall, as did lean Red-eyed Vireos
(Vireo olivaceus) released at night (Sandberg and
Moore 1996, Sandberg et al. 2002). That migrants
would exhibit a directional response to remove
themselves from the coastal 100 ± km of
Louisiana is also consistent with the knowledge
they appear to find these areas unsuitable during
spring, normally flying over them when arriving
from a spring Gulf crossing (Lowery 1945,
Gauthreaux 1971). Fall migrants intending to
winter in the region could also backtrack if they
find the coastal wetlands unsuitable wintering
habitat. The northward direction of both neotrop-
ical and temperate migrants could be either a
reflection of attempting to backtrack to an earlier
position on the migratory route before habitat
became inhospitable, or of attempting to compen-
sate for wind drift by retreating into the wind,
which is generally northerly after nights of heavy
migration in fall. Other authors have reported
changes in diurnal flight orientation to compen-
sate for nocturnal wind drift (Baird and Nisbet
1960, Able 1977, Gauthreaux 1978, Moore 1990),
and apparent reversal of migratory orientation
over inhospitable areas (DeSante 1973, Thorup
and Rabpl 2007). Migrants might retreat because
wind drift has carried them farther south than
intended on that date in their migration in addition
to avoiding inhospitable habitat. Some large
flights at South Point have involved species that
did not arrive in numbers elsewhere in coastal
Louisiana until weeks later (e.g., 7,000 Cedar
Waxwings on 22 November 2002).

The occurrence of large movements of Amer-
ican Robins and Cedar Waxwings along the Gulf
Coast is interesting because the Gulf Coast forms
the southern edge of their wintering ranges at this
longitude. The large (lights of Yellow-rumped
Warblers, occurring mainly in late November,
may also represent temperate winterers rather than
trans-Gulf migrants, as the bulk of other trans-
Gulf migrant species have departed the northern
Gulf Coast by I November (Lowery 1974). The
other major concentration sites of migrants in

North America usually involve birds well north of
their southern range limits, at latitudes traversed
by a large portion of the population during their
migration. Large flights near these species’
southern range limits are surprising, because
relatively small proportions of their populations
migrate this far south.

The evidence suggests the movements observed
along Lake Pontchartrain are predominantly birds
retreating from coastal wetlands. If correct, this
adds an important detail to the dynamics of
migration near the Gulf Coast. It indicates the
“coastal hiatus” phenomenon for neotropical
migrants long documented for spring migration
has a counterpart in fall. In the case of temperate
zone migration, it indicates that overshoot of
wintering areas by a short distance (—100 km)
along the Gulf Coast is a common phenomenon.
In either case, the species involved must incur a
stressful upwind flight to terra firma that may
take several hours and deplete energy reserves. It
also implies, in the case of neotropical migrants,
that habitats lying Just north of the coastal
wetlands might be especially important as stop-
over sites forming the last stops before the
demanding trans-Gulf flight.

The numbers of birds involved in the Lake
Pontchartrain flights make it one of only a handful
of places in North America where diurnal
movements of migrant song birds (excluding
swallows) are documented to regularly exceed
1,000 birds/hr. These include sites along the
Atlantic shoreline from Long Island Sound to
Delaware Bay (e.g.. Lighthouse Point, Connecti-
cut and Cape May, New Jersey), on the shorelines
of Lake Erie and Lake Ontario (e.g.. Holiday
Beach, Ontario, and Braddock Bay, New York), at
Tadoussac on the north shore of the St. Lawrence
River in Quebec, and along the Gulf of Mexico
shoreline near Veracruz, Mexico. It is unusual in
its inclusion of a 9 km elective overwater flight
(as opposed to return to land from the ocean),
although southbound Blue-gray Gnatcatchers
{Polioptila caerulea) regularly make an overwater
crossing at Smith Point on the Texas coast (John
Arvin, pers. comm.). The predominant species are
those known to be at least partially diurnally
migratory, as is the case elsewhere. Most have
large regular diurnal movements at one or more of
these other localities, except apparently the Indigo
Bunting. Large flights are characteristically ac-
companied by other species, including a wide
variety of nocturnal migrants. However, many

Yaukcv MORNING FLIGHT OF MIGRANTS

743

understory species apparently do not cross the
lake as no attempts have been observed by
catbirds, thrashers, Catharus and Hylocichla
thrushes, and wrens; the only sparrows have been
those ot the most open habitats (e.g.. Savannah
Sparrow [Passerciilus scmdwichensis]).

If flights represent northward coiTective move-
ments, they might be less common along the
eastern half of the northern Gulf Coast, which
lacks the wide expanse of coastal wetlands that lie
south ot Lake Pontchartrain. Broad wetlands
border the coast for 500+ km west from the
mouth of the Mississippi River throughout
Louisiana and the upper coast of Texas, and these
would seem good candidates for reverse flights.
However, without a funneling geographic feature,
these flights might not be concentrated as they are
at Lake Pontchartrain. The 40-km shoreline of
Lake Pontchartrain apparently turns an otherwise
diffuse northward movement into a spectacle. If
one assumes that birds arrive at South Point or
Pointe aux Herbes after accumulating along the
lakeshore as they arrive from the south and detour
eastward, then arrival of only five northbound
birds/hr at each 0.1 km of lakefront (estimating
the detection range of an urban observer on the
ground) would accumulate into a flight of 2,000
birds/hr at the peninsula. These concentrating
factors are probably lacking elsewhere in the
coastal wetlands, but study of movement patterns
of migrants throughout the area would be
worthwhile, and may provide insights regarding
which areas are most critical to their preparations
to cross the Gulf of Mexico.

ACKNOWLEDGMENTS

I thank Laura Alexander, Jessica Campbell, Oscar Camp-
bell, Laura Dancer, Pete Dunne, David Mizrahi, Walter
Ellison, Duncan Evered, David Juran, David Little, Craig
Matthews, David Muth, B. Mac Myers, R. D. Purrington,
Melissa Ryce, Denise Spindell, Steve Stevens, Paul Sykes,
and Phillip Wallace for helpful discussions and/or assistance
in data collection. David Muth. J. Van Rem.sen Jr., Ken Able,
and an anonymous referee provided helpful reviews of the
manuscript. The Bayou Sauvage National Wildlife Refuge
was helpful in allowing access to the study site.

LITERATURE CITED

Able, K. P. 1972. Fall migration in coastal Louisiana and
the evolution of migration patterns in the Gulf region.
Wilson Bulletin 34:231-241.

Able, K. P. 1977. The orientation of passerine nocturnal
migrants following offshore drift. Auk 94:320-330.

Akesson, S., L. Kaki.s.son, G. Walinuer, and
T. Alerstam. 1996. Bimodal orientation and the
occurrence of temporary reverse bird migration during
autumn in southern Scandinavia. Behavioral Ecology
and Sociobiology 38:293-302.

Alerstam, T. 1978. Reoriented bird migration in coastal
areas: dispersal to suitable resting grounds? Oikos
30:405-408.

Alerstam, T. and S. Ulfstrand. 1975. Diurnal migration
of passerine birds over south Sweden in relation to
wind direction and topography. Ornis Scandinavica
6:135-149.

Baird, J. and I. C. T. Nisbet. 1960. Northward fall
migration on the Atlantic Coast and its relation to off-
shore drift. Auk 77:1 19-149.

DeSante, D. F. 1973. Analysis of the fall occurrences and
nocturnal orientation of vagrant wood warblers
(Parulidae) in California. Dissertation. Stanford Uni-
versity, Palo Alto, California, USA.

Duncan, R. A. and W. C. Weber. 1985. The Yellow
Warbler: a diurnal circum-Gulf fall migrant. Florida
Field Naturalist 13:20-22.

Gauthreaux Jr., S. A. 1971. A radar and direct visual
study of passerine .spring migration in southern
Louisiana. Auk 88:343-365.

Gauthreaux Jr., S. A. 1978. Importance of daytime flights
of nocturnal migrants: redetermined migration follow-
ing displacement. Pages 219-227 in Animal migration,
navigation, and homing (K. Schmidt-Koenig and W,
T. Keeton, Editors). Symposium held at the University
of Tubingen, 17-20 August 1977. Proceedings in Life
Sciences. Springer-Verlag, Berlin, Germany.

Lowery Jr., G. H. 1945. Trans-Gulf spring migration of
birds and the coastal hiatus. Wil.son Bulletin 57:92-121.

Lowery Jr., G. H. 1974. Louisiana birds. Louisiana State
University Press, Baton Rouge. USA.

Moore, F. R. 1990. Evidence for redetermination of
migratory direction following wind displacement.
Auk 107:425-^28.

Sandberg, R. and F. R. Moore. 1996. Migratory
orientation of Red-Eyed Vireos, Vireo olivacens. in
relation to energetic condition and ecological context.
Behavioral Ecology and Sociobiology 39:1-10.

Sandberg, R., F. R. Moore, J. Bi.ackman. and
M. Lohmu.s. 2002. Orientation of nocturnally migrat-
ing Swainson’s Thrush at dawn and dusk: importance
of energetic condition and geomagnetic cues. Auk
1 19:201-209.

Seymour, M. 2(K)9. Chasing the storm. Louisiana Department
of Wildlife and Fisherie.s, Baton Rouge. USA. http://

www.wlf.louisiana.gov/onlinestore/laconscrvationist/

pcekinside

Thorup, K. and j. Rab0l. 2007. Compensatory behaviour
after displacement in migratory birds: a meta-analysis
of cage experiments. Behavioral Ecology and Socio-
biology 61 :825-841.

The Wilson Journal of Ornithology 122(4):744-754, 2010

NIGHT MIGRANT FATALITIES AND OBSTRUCTION LIGHTING AT
WIND TURBINES IN NORTH AMERICA

PAUL KERLINGER,' " JOELLE L. GEHRING," WALLACE P. ERICKSON,'
RICHARD CURRY,^ AAETAB JAIN,' AND JOHN GUARNACCIA^

ABSTRACT. — Avian collision latality data from studies conducted at 30 wind farms across North America were
examined to estimate how many night migrants collide with turbines and towers, and how aviation obstruction lighting
relates to collision fatalities. Fatality rates, adjusted for scavenging and searcher efficiency, of night migrants at turbines 54
to 125 m in height ranged from <1 bird/turbine/year to ~7 birds/turbine/year with higher rates recorded in eastern North
America and lowest rates in the west. Multi-bird fatality events (defined as >3 birds killed in 1 night at 1 turbine) were rare,
recorded at <0.02% (n = 4) of ~25.000 turbine searches. Lighting and weather conditions may have been causative factors
in the tour documented multi-bird fatality events, but flashing red lights (L-864, recommended by the Federal Aviation
Administration [FAA]) were not involved, which is the most common obstruction lighting used at wind farms. A Wilcoxon
signed-rank analysis of unadjusted fatality rates revealed no significant differences between fatality rates at turbines with
FAA lights as opposed to turbines without lighting at the same wind farm. Received 30 May 2006. Accepted 29 June 2010.

Songbirds often collide with communication
towers, lighthouses, skyscrapers, and other struc-
tures during nocturnal migration with fatalities, at
times, numbering in the hundreds or even
thousands of birds in a single night (Banks
1979, Avery et al. 1980, Trapp 1998, Kerlinger
2000). Research suggests lighting has a primary
role in attracting or disorienting night-migrating
songbirds at those structures, especially during
overcast, foggy, or rainy conditions (Cochran and
Graber 1958, Caldwell and Wallace 1966, Avery
et al. 1976). Many studies, attempting to assess
how lights influence bird behavior, have focused
on communication towers. Larkin and Erase
(1988) used tracking radar to show that, during
fog and low cloud ceiling, night migrants circled a
>3()5-m tall communication tower with Federal
Aviation Administration (FAA)-approved ob-
struction lighting, but departed the tower when
lights were extinguished. Gauthreaux and Belser
(2006) used marine surveillance radar and infra-
red scopes to compare night migrant activity at
two tall (>305 m), lighted communication towers
with guy wires and at a .separate control plot. One

' Curry & Kerlinger L.L.C., P. O. Box 453. Cape May
Point, NJ 08212. USA.

^ Michigan Natural Features Inventory. Stevens T. Mason
Building. P. O. Box 30444, Lansing, Ml 48909, USA.

'Western EcoSystems Technologies Inc.. 2003 Central
Avenue. Cheyenne. WY 82001. USA.

^ Curry & Kerlinger L.L.C., 1734 Susquehannock Drive,
McLean. V A 22101, USA.

^302 Bryn Mawr Drive SE. Albuquerque. NM 87106,
USA.

1407 Finniown Road. Waldoboro, ME 04572. USA.
''Corresponding author; e-mail: pkerlinger@comca.st.net

of the towers had flashing and steady-burning red
lights, the other had only flashing white strobe
lights. They found that birds flew in straight flight
paths over the control plot but, at the lit towers,
flight paths were curvilinear and birds concen-
trated near the towers. More birds concentrated at
the tower with flashing and steady-burning red
lights than at the tower with flashing white strobe
lights. This led Gauthreaux and Belser (2006) to
suggest that research was needed to improve
understanding of the attraction of different
obstruction lighting systems to night-migrating
.songbirds.

A recent study in Michigan by Gehring et al.
(2009) of 24 communication towers with different
heights, support systems, and lighting found that
towers lit at night with only tJa.shing red or white
lights had significantly fewer avian fatalities than
towers lit with a combination of steady-burning and
fla.shing lights. These results suggest avian fatalities
can be reduced, perhaps by 50-71%, at communi-
cation towers supported by guy wires by replacing
steady-burning lights with flashing lights.

Avian collision fatalities are well documented
at wind turbines (turbine rotor or tower), but the
estimated total number of birds killed annually is
small relative to communication towers and other
structures (Erickson et al. 2005, National Re-
search Council 2007). Wind turbines >60 m in
height are often equipped with FAA-approved
obstruction lighting, but not all turbines need to be
lit, as long as gaps between lighted structures do
not exceed 0.8 km (FAA 2()()()). Thus —25-33%
of turbines at most wind farms should have
obstruction lighting on the nacelle (where the

744

Ker/iiigcr et al. • NIGHT MIGRANT FATALITIES AT WIND TURBINES

745

I

FIG. 1 . Wind turbine showing location of Federal
Aviation Administration obstruction lights on the nacelle.

blades are attached to the rotor hub; Fig. 1)
adjacent to where the rotors are attached. The
FAA (2000) recommends flashing red lights be
placed on the turbine nacelle but flashing white
and steady-burning red lights have also been used
on a limited number of turbines. The flashing red
lights are model L-864, a red strobe, LED (light-
emitting diode), or pulsating incandescent light
that flashes 20-40 times/min with an intensity of
2,000 candela. Steady-burning red lights are
classified as model L-810, an incandescent red
lighting that has a minimum intensity of 32.5
candela. L-810 lights are red and occasionally are
modified to flash. Flashing white lights are
classified as model L-865 and are a white strobe
typically set at 40-60 flashes/min with an
intensity of up to 2,000 candela. These same
types of obstruction lights are used on communi-
cation towers (FAA 2000, Gehring et al. 2009).

Concerns about avian mortality have prompted
fatality studies at wind farms across the United
States and Canada. Most studies are available as

reports to meet permit requirements and are often
reviewed by state and federal wildlife agencies.
This paper reviews existing information on night
migrant mortality at wind farms derived from 31
studies at 30 wind farms. Our objectives were to:
(1) examine the incidence of multi-bird fatality
events at individual wind turbines and their
relationship with lighting, and (2) examine
whether disproportionately greater numbers of
fatalities occur at turbines equipped with FAA-
approved lights as opposed to turbines without
approved lights.

METHODS

Data were extracted from post-construction,
avian fatality studies conducted at 30 wind farms
across the United States and Canada (Table 1).
Studies prior to 1995 were not included because
turbines lacked obstruction lighting and were
<50 m in height (to maximum blade tip height).
Excluded studies in this category were conducted
at Altamont Pass Wind Resource Area, San
Gorgonio Pass, and Tehachapi Mountains, all in
California. The results of studies at several small
wind farms conducted after 1995 were also not
included because lighting or search methodolo-
gies could not be verified. Other wind farms in the
United States and Canada were not included
because they have not been studied or, if they
have been studied, reports were not available.

Data extracted included geographic region;
turbine nameplate production in megawatts
(MW); turbine height (to the maximum blade tip
height); total number of turbines in the wind farm
and the subset of turbines studied; lighting type,
including the total number of turbines with
lighting and the number of lit turbines studied;
study duration and search interval during migra-
tion seasons; unadjusted number of nocturnal
migrant carcasses found; and estimated fatality
rate (birds/turbine/year). Carcasses were assigned
as night-migrating songbirds or similar species
(i.e., cuckoos, etc.) based on migration tendency
(nocturnal vs. diurnal) and date recorded.

Methodologies for studying collision mortality
at wind farms varied, but basically followed
accepted practices (e.g., Anderson et al. 1999).
Carcass searches were usually conducted every
week to 1 month in the western United States,
whereas they were usually conducted every 1-
2 days to 1 month in the eastern United States and
Canada. Some studies used different search
intervals at different subsets of turbines. Searchers

TABLE I. Estimated avian migrant mortality per year (spring and fall) from 31 studies at 30 wind farms (Buffalo Mountain studies were done at each of the 2 types of
turbines on site).

746

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. !22, No. 4. December 2010

V- -

C c ”

I

>>-a

>» ^

O

(N

OJ

W)

-O 0-
3 O

= >

O c 00
O tU O
O r , o
CN U (N

ac

H

1£ IS

'mm

3 3

V V

c

o

^ I

c/3

-§ ^

— r-

tS rs
1) u
>>

sC —
ri

o —

O'! C^l

■a "o

u u

OX) OXJ

c c
13 13

CA c/3

-3

u- u-

o^

^ :>
ON

ON

:> r--

0

c

15

u.

3

1

c^i

I

so

C

O

£

o

P

"O

1)

OXJ

c

—

u.

4

4

O

o

ON

OC \0

— : o

o

o

o

(N

o

(N

2003

D

X

k.

o

r--

o

o

04

*3

*3

3

04

o

"3

D

D

D

”3

m

D

in

o

Ui

C

C

(—

D

o

o

D

O

O

5

OX)

o

C4

OX)

c/3

c/:

c/3

04

c

d

D

D

D

3

r-

3

X

o

o

‘C

O

3

D

m

m

>

CD

D

C

0)

c

-D ^
u. u. u.
3 3 3

^ >v

£ £
15 15

U. Urn

3 3

V V V V V

^ m o

3

X

JO

V3

3

X

C

00

c

r—

D

00

o

(N

3

D

04

£

4

in

1

—

d

04

S

>> m

s >

lO

— — (N

5 —
^ —

•o "O “O

ODD

“ " " o

ojj bij 00 ^

c c c —

13 13 13 ?

c/3 c/3 ;/3

"O

D

Oil

c

3 c3

u. u. E

r- r-

-C D
c/3 C
3 O

E Z

^ 00
vO ro
^ 00
so r^i

> ^ ^

I s s

so vO
lO sO sC

— d d

4

(N

^ ON

r- so
o^

sO

r-
ro

r^i \C

— d

<1,

u

o

Cj

S

0^: /

C/5

D

CA

X

c

c

O'

d.

3

o

o

D

-d

O C

_D

D

D

■£

X

-

u

-C

3

c

c/3

D

0/j

3

X

C ''

P -

c/3

D

X

w

a.

3

5

> c

J .i

. QC

5 ^
U ^

D

C O

z £

'd

c

E

u

3

V

vn

(N

D

C

V

o

o

o

(N

3

D

C

O

c/3

C

s:

o

D

C

E

3

I

r-

o

o

CN

"3

D

>v

r-

o

o

"3

D

r-

o

o

r4

>.

D

C

D o .ii

Q c/5 H

U- u
D ^ ^

.£ "d Ij

JD C C
3 Ip Ip

CN — —

I V V

(N

c

o

(N

"3

D

D

O

X

D

C

E

Urn

3

V

"O

>>

•o

3

D

C

o

WJ

•o

D

*- D

II

3 ^

Ic

— ^ c/3

D ^ SO

«- fNj

>v -a ^ ^

“O C 00 ^

3 3 w -

D

D

D

(N

3

D

>»

OJ)

c

c

o

D D C

I

3
"O

c

T3 d

D X

-3

3

3

*3

D

<N

■3' m
CN (N
^ 00
(N m

D

D

d

•a

D

OJD

3

■O XJ 3

D D C

SO

CN

CN

04

“O

D

3 u
D

C/5

00

ON

sO

ri

o^

00

o

ON

c/5

o

(N

-S'

3 ^

D

(J5

sO

r-

I

O 'Ti o

— <N 04

On

00

I I

ro

d

^ s s s

lO) 00 O

— — r'i

so

SO

C/5

D

c

D

U

3

O

a,

o

H

D

OX)

X

c

D

D

c/3

D

k.

u

z

s

D

00

x

5

o

3

O

UJ

z

o

c/3

c

<

c

c O

U .2

ON

o

00

o

o

(N

D

D

X D 03
On ON ON
COO
O O O
04 CN CN

(J5 Z

>N

*5

3

E

u

3

D

.£ "o

_0 D D
3

— CC3 3

.§ .§

“X ^ )’ c/3 c/3

D ( D D

' I

I

(N

cN O

^ Z V

o o

z z

00 On 04 ON r-

>>>>>%

3 3 3

^ c/3 X

>> >. c

X

c

3

-2 3
X

rp fs)

c/3 c/3 c/3 w

c 1=;

-C

3

c §

c/3

c/3

(*S

U

!- C

c

c

C_/

^ E
•2 o

3

D

D ?
>v C

o

£

o

£

P

in

t-' —

04

04 r--

00

so

sd

2 ^

:::2, ^ r- — in

2 so d d

w'lncNcn —

"S "O X X X

^ D D D D

Um u, u u i-

OX)

OX)

1 OX)

OX)

OX)

OX)

OX)

OX)

c

p

X)

^ c

c

c

c

3

_c

E

E

3

04 E

E

E

E

E

E

c/3

c/3

•w' CA

CA

CA

CA

CA

CA

E

3

O

3

E

E

E

3

3

UL.

X

(X

X

X

X

X

X

X

ON SO On On On

o

(n

00

ro

^ Cl

in

04

04

ro

S E ^

m

00

^ E

00

so

04

cn m oo

(N CN —

. ^
CN in so so in

^ ^ ^ ^ J

s s s s s

in in in in m

W P

E 3

ca CO

D

D

C

3

3

D

cn

£

E

D

to

3

X

O

U

X .2 ^

s ^

53 ^

S < oa

>-

Z

61)

Lh

3

C

D

U UJ

TABLE 1. Continued.

Keiiingcr ct al. • NIGHT MIGRANT FATALITIES AT WIND TURBINES

747

^ 3

00 ^ •

00 C

^

o.

>.TD

-O ^
(N 2^

o O

o ^

o

(N

a>

wj

c

^ •;;;

(U (U

E E
5 15

u u,

5 E

<N

I I

lO <N

>%

cd

■o

O

S §

oi)

.E o

fe .E

^ Z

<u o

.E E
15 15

L. U

E E

—

V I

o vS

o

o

CN

"E

5

"O

OJ

15

(U

c

15

u

3

CN

I

<U

oxj

(N

O

c c/1
C

u. u.

O <U

u.

UJ

o

04

■o

CS

o

”E

00

u

"O

o

*o

c

o

c

C3

04)

C

D

O

>-

u-

^ >-»

0) <u

E ^
15 15

U Urn

3 3

^ rn

V I

o o

dj

c

15

u.

3

<N

I

SO

m

year

year;

year

O

year

3

O

E

c

c

g

3

O

—

(N

—

—

00

SO

—

oo

00

A

1/^ in n,
^ vC

A

Z

>- (U

Z ^

C/2 CJ

o

00

r-i

os

OO

1

in

00

00

— *

s

—

>

>

2

00

00

s

— ‘

00

s

nO

m

sO

in

d

d

d

I

r--

04

00

q

(^i

< ^
CL S

C

E

c

o

5

o

o

CN

<U <D

c c

15 15

U, Urn

E E

7 f

—

I I

O CN

CN

I

ON

— ON

SO

so

so

s

— d

^ ^ ^ 3 Z "E z

■O Q.
n c3

3 3

C CQ

H

QQ

o

c

15

Fn

3

f

m

c/2

^ c/2

(1)

"O 0)

c/2

on

« -g

c/2

>%

CS

c/2 tZ)

CS QJ

T3 O

c/2

>»

C3

>,

C3

o ^

3

-Q <D
^ 0)

■a

•o

*U

7

"O

7 ^

r--

1

7 ^

r^

1

r- o

r--

m —

tn d

(n

c/2

c/: (U

is

^ C3

(U > 0)

I

(N — —

r-

oo

m

(n

X

3

cd

OS
' —
OS

(N

CN

CN

' —

r-

00

00

o

1)

*0

C4

X)

X

7

X

15

fU

u

X

"O

X

~o

0)

u.

o

t

«u

L_

f

04)

c

0)

L.

(D

u-

X

o

u

V

Urn

04)

04)

o

04)

04)

04)

04)

OJD

o

u

04)

04)

g

3

so

3

3

3

“S

3

E

>>

3

3

-3

IE

1

15

15

15

cd

CJ

15

15

X

15

-5

C/l

(Zl

c/2

c/2

c/2

3

c/2

c/2

cd

c/2

c/2

-E

cd

3

C3

3

o

_cd

Cd

E

Cd

cd

ll

U.

PL

PL

PL

z

PL

PL

PL

PL

00

H

>

>

z

dj

u

uT

<D

3

X

o

c/T

ol

z

o

DO

U-

1)

3

E

U

3

3

Cd

D

b.

o

_5

00

3

-D

c/2

'cd

3

c

(55

U

JZ

C/0

•E

U

ID

u,

cd

<D

C/0

3

O

S

1

E

(U

C4

.E

'u

UP

15

X

PP

U

K

C3

w

TABLE 2. Night migrant mortality comparison at lit and unlit turbines; wind farms with flashing red lights only. Lit-unlit column represents the difference between numbers of
actual carcasses found at lit vs. unlit turbines, on a per turbine basis.

748

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4, December 2010

U S C

5 c ^

c -S

3 ^

3. Z

^ —

.3 C

o o

A

S

Q. OJ

O r"

5

c.

A

a,

tS

aj

O ^

OJ ^

S-

— ON

-a

2

o

Q-

(U

o

C3

3

3 3

CJ O

cr w 'zz

y: :a 'V.

U

3 a. 3

Cl

c/3 ly;

w S '■S

C3 > C3

O (TD

T3 -O

O O
CL CL
O O

2 ^

y: y.
B w

o o

ir c/3

■o -o

qj CD

u

O O
CL CL
U (U

3 3

O CJ

3 2
^ U

c

A

cx.

A

■a

o

u

o

CL

(D

c

C

3

0

’E

3

3

3 ■

w''

00

y

y

y

y

y

y

:/) ,

0

B

B

B

C/l

B

a

B

■

(U

0

0

0

3

0

0)

0

L.

k.

u.

u

3

L.

L.

u. -

3

3

3

3

. _<

3

3

3

3

3

3

3

00

3

3

3

3“

P"

cr

O’

3

O’

O'

O' ■-

^ u u

o o

O' On

W)
c c

U

-C -C

U U

U

CL CL
CL CL

-2 -2

U. L.

(U

>

o

ro

O

o

CN

o c

C “

00

ii

(N (N 04

^ -
00 ?;

O OJ (U
^ >

U U c/:i O O

o

o

(N

o o
o c

(N (N

<u

^ y y y

^ ^ ^ r- .

= CJ o o ^

■§ ‘C 'C 'u 0^

^ W LU m H

r--

0

(N

(3

U

CJ

"O

3

X)

00

3

OS

OS

Os

r-

OS

OS

04

(N

0

0

0

0

0

0

0

CJ

Os

os

0

0

0

0

0

0

0

V

0

0

04

04

04

04

04

04

04

A

3

*3

0

-0

0

0

04

CJ

04

0

3

*3

3

*3

"3

3

*3

3

0

0

0)

0

0

CJ

CJ

c/:

-C

u-

3

3

3

3

c

c

c

3

3

_3

3

u

0

Q

on

00

'3

'3

3

3

3

3

’3

CJ

00 00

c c

3 3

O O
>- >-

sO

r-

'3-

ro

so

os

0-

O'

ro

so

0-

0

ro

'3-

0

0

Os

*3-

0

00

0

0

04

0

0

—

0

m

04

04

00

ro

VO

0

04

0

r--

m

d

d

d

d

d

d

d

d

d

d

d

d

d

ro

d

d

d

d

d

d

7 =" =

u u •—

oc

00

sO

ro

m

3"

ro

0

0-

0-

ro

04

0

0

0

c^t

r--

0-

—

ro

ro

ro

r-

so

3*

ro

0

so

VO

00

os

m

0

0

^ 3

0“i

10

o|

ro

—

04

ro

04

so

3;

ro

q

—

oc

r-

q

10

04

>0

Hi

d

d

d

d

d

3"

d

d

oi

d

carcasses
found at
lit turbine

■3’

ro

3"

ro

CO

'3*

fO

04

00

04

3-

00

00

VO

04

VO

04

ro

ro

04

04

-i- vY r-* r^i

so 00 or^ —

VO o — r-- o
GO O 00 O ^
«oOriO04 —

o d d o o o

0^ ^ ro —

sO On a^ sO VO
n fo —

00 00 m

r- o
vO o
— 04

o o
o o
o r-

SO

d d

o

o

(N

O ro —

SO r-

o o

103 O

— 06

C<', Tt

m

(N

so

00

(N

vo r- —
m

103 o

Os Tj- o o V03

00(NiO3Or--
m 00 103 00

d d d — d

^ Os so

^ o

r-j (^1

3

*0

o

o

o

0)

O)

0^1

ro

00

t—4

y

0

>v

sz

”

CJ

’3

Ui

3

“O

CJ

y

so

00

0

0

0

0

0

0

0

3

04

y

04

00

o

o

>-

>- > >

<u

-C

o-

u,

3

y

3 ^
T3 >

u

S

ME,

OJ

y

01

>-'

y

04

CJ

00

y

CJ

04)

Un

3

CJ

y

CJ

OX)

CJ

OX)

uT

CJ

Cj'

CJ

3

3

00

0

s

X

z

d

CJ

0

u

z

d

CJ

JT

CJ

L.

z

72

5

CJ

-C

CJ

u

;o

c5

2

CJ

;o

5

72

5

CJ

• E

3

0

0

3

a

u-

(/;

u

y

L.

0

3

3

CJ

y

0

3

3

CJ

y

y

y

CL

3

CJ

LO

CJ

CL

CJ

CJ

CL

jj

CL

3

3

(y5

CJ

CJ

ca

3

3

3

3

3

0

.4_)

to

3

s

S

U

U

s

S

S

S

S

s

S

4;

O

Ui

3

<U

U-}

Kerliiii^er ct al. • NIGHT MIGRANT FATALITIES AT WIND TURBINES

749

were olten site for 2-5 days/week because it
required several days for turbines at a site to be
searched. On-site technicians and maintenance
staff at most wind farms were instructed to report
avian fatalities, increasing the likelihood of
recording large scale collision events at towers
that were not part of the study.

We report both the raw numbers of night-
migrant carcasses found during each study and
estimates of fatalities/turbine/year. Most studies
calculated estimates of fatalities/turbine/year from
raw numbers of carcasses found, searcher effi-
ciency tests or estimates, carcass removal (scav-
enging) tests or estimates, and area adjustments if
the entire area beneath a turbine could not be
searched. We used rates from studies in similar
habitats to estimate fatalities likely/turbine/year
for studies where searcher efficiency and carcass
removal rates were not measured empirically. We
did not estimate fatality rates at three sites where
large portions of migration seasons were not
covered or where searcher efficiency and carcass
removal rates could not be estimated. Our
rationale for including raw, unadjusted fatality
numbers was to provide a comparison of actual
carcass finds to fatality estimates, and to permit
comparison with communication tower studies,
where coiTections for carcass removal and search-
er efficiency have rarely been used.

We compared the unadjusted number of night
migrant fatalities/turbine/year at lit (flashing red)
and unlit turbines from 12 of 30 wind farms
(Table 2) using a non-parametric Wilcoxon
signed-rank test (Wilcoxon 1945). We wished to
ascertain whether a significantly greater number
of wind farm studies reported more carcasses at lit
turbines versus unlit turbines.

RESULTS

Eight wind farms were examined in the western
United States, seven in the central region, 12 in
the eastern region, and three in Canada (Table 1).
Two studies were reported for Buffalo Mountain,
Tennessee because the type of lighting changed
when the wind farm was expanded and larger
turbines were installed. The number of turbines at
these sites ranged between one and 454 and the
number of turbines studied ranged from one to
153. Turbine nameplate production ranged from
0.4 to 2.0 MW, and turbine height ranged from 54
to ~125 m. Obstruction lighting was present at 28
of 30 wind farms studied. The percentage of
turbines equipped with lighting ranged from

~25-33% (n = 22 studies) to 100% in = 7
studies). Lighting on most illuminated turbines
was flashing red lights (mainly L-864), but a few
sites had steady-burning red (L-81()), and two
sites had flashing white lights (L-865). More than
one type of lighting was used at four sites. Study
durations ranged from >3 months to 5 years.
There was variation in search interval among
studies. We estimate the total number of individ-
ual turbine searches for all studies combined to be
~25,000 during spring and fall migration seasons.

Incidence of Multi-bird Fatality Events. — The
number of cai'casses found in the studies ranged
from zero to 36/year, but when the number of
turbines studied was considered, the average at
turbines ranged from zero to ~5/turbine/year. Only
four studies reported multi-bird fatality events
(defined as >3 birds killed in one night at one
stiTicture), but they were at turbines with lighting
other than flashing red or at turbines with ancillary,
non-EAA-type lighting. Fourteen freshly-killed
birds ( 1 1 warblers, 2 flycatchers, and 1 vireo) were
recovered on 19 May 1999 at two adjacent turbines
in a section of the wind fai'm at Buffalo Ridge in
southwestern Minnesota (Johnson et al. 2000)
where eveiy other turbine was lit with steady-
buming red lights. This event may have been
related to a severe thunderstorm that occuired the
night before the carcasses were discovered. This
was the only multi-bird fatality event reported at the
Minnesota site during 4 years of study.

Two multi-bird fatality events (3 fatalities at
turbine # 2 on 10 Oct 2000, 7 fatalities at turbine
# 3 on 31 Oct 2002) were recorded at Buffalo
Mountain I in Tennessee (Nicholson et al. 2005)
where each of three turbines had a pair of flashing
white lights. The 10 October 2000 mortality event
occuiTed during clear, cold, windy conditions
following passage of a strong cold front, and the
31 October 2002 mortality event occurred during
mild, rainy weather.

About 27 night-migrating passerines were
found dead on the morning of 23 May 2003 at
Mountaineer in West Virginia (Kerns and Ker-
linger 2004) in the vicinity of three turbines and
an electrical substation that was brightly lit at
night by at least four sodium-vapor lamps (steady-
burning white flood lights). Heavy fog occuired
the night before this fatality event was di.scovered
by maintenance workers, who alerted searchers.
Seventeen of the fatalities were discovered at
turbine # 23, which was —50 m from the
.substation, five at the substation (collisions with

750

THE WILSON JOURNAL OF ORNITHOLOGY • Voi 122, No. 4, December 2010

fencing), and three and two respectively at the
flanking turbines, # 22 and 24, which were ~250
and 125 m, respectively from the substation. No
lurther multi-bird fatality events occurred at the
substation and adjacent turbines, including foggy
nights, after the sodium-vapor lamps were extin-
guished. Few fatalities were found at searched
turbines throughout the study, including those that
had flashing red lights.

Comparison of Lit and Unlit Turbines at
Same Site. — The FAA does not recommend that
all turbines at a wind farm be lit, but we were able
to test whether disproportionate numbers of
fatalities occurred at turbines with obstruction
lighting. There were two sites at which none of
the turbines was lit and seven at which all of
turbines were lit (Table 1). The two completely
unlit sites had 0.55-0.60 MW turbines in the 59-
61 m height range, and estimated fatality rates of
<1 night migrant/turbine/year. Sites where all
turbines were lit had 0.66-1.65 MW turbines in
the 79-1 17 m height range, and estimated fatality
rates of <1 to ~7 night migrants/turbine/year.
Sites in eastern North America reported slightly
greater estimated fatality rates (Table 1). This
appeared to be the case for turbines of roughly
equal height. Wind farms (n = 22) with both lit
and unlit turbines had 0.36-2.00 MW turbines in
the 54-125 m height range. Estimated fatalities
ranged from <1 to ~4 birds/turbine/year with
greater rates in eastern North America. Absolute
numbers of fatalities were sufficiently large at 13
of 22 sites to make quantitative comparisons of
fatality rates at lit and unlit turbines (Table 2).
The rates were roughly the same at sites with
flashing red lights, but may have been slightly
greater at turbines lit with steady-burning red
lights. We did not find that a significantly greater
proportion of wind farm studies (Table 2) report-
ed greater numbers of fatalities at lit turbines
versus unlit turbines than expected by chance
(Wilcoxon sign-rank test, n — 20, Z = 0.18, P —
0.43). This lack of evidence of a consistent trend
with respect to fatality rates at lit versus unlit
turbines across North America indicates it is
highly unlikely flashing red lights are associated
with greater fatalities. All but one of the studies
reviewed that tested for differences in fatality
rates at lit and unlit turbines reported no
significant differences. The one exception was at
Maple Ridge in New York, where a marginally
significant difference was reported (0.10 > P >
0.05) in one of two statistical te.sts.

No significant differences in collision mortality
were reported at wind farms with lit and unlit
turbines where some or all lit turbines had other
than flashing red lights. These included Crescent
Ridge, Illinois (Kerlinger et al. 2007), Buffalo
Ridge, Minnesota (Johnson et al. 2000, 2002), and
Buffalo Mountain, Tennessee (Fiedler et al.
2007). Unadjusted mortality at Erie Shores,
Ontario (James 2008), where all lit turbines had
steady-burning red lights, was four times greater
at lit than at unlit turbines (0.4/turbine vs. 0.1/
turbine). The unadjusted rates, when turbines
along the shore of Lake Ontario were removed
from the calculations, were similar (0.35/turbine
at lit vs. 0.28/turbine at unlit).

DISCUSSION

Collision fatality data from 30 wind farms in
North America demonstrate that fatality rates of
night migrating birds at these structures are
relatively low, ranging between ~1 to 7/turbine/
year. A comparison of estimated fatalities by
geographic region showed a gradient with fatal-
ities increasing from western to eastern North
America. This mirrors continent-wide studies of
bird migration patterns (Lowrey and Newman
1966, Gauthreaux et al. 2003) that show the
density of migration in central and eastern states is
greater than recorded in western states.

What is striking about the data from wind farms
is the relative absence of large-scale fatality events,
similar to those recorded at tall communication
towers supported by guy wires, where collisions of
hundreds of birds at times occur in a single night
(Avery et al. 1980, Kerlinger 2000). Only four
incidents (<0.02% of searches) were reported of
multi-bird fatality events at wind turbines (/? = >3
carcasses at a single turbine on a single night)
during ~25,000 turbine searches in all studies
combined. That so many studies and so many
searches have been conducted at wind turbines
without recording large-scale fatality events strong-
ly suggests the probability of large-scale fatality
events occurring is extremely low. Ear more
systematic research has been conducted at wind
turbines than at communication towers or other
structures, which supports the finding that large-
scale fatality events rarely occur at wind turbines.

We believe there are three reasons why large-
scale fatality events apparently do not occur at
wind turbines and why fatality rates at wind
turbines are much lower than has been reported
for communication towers. First, communication

Kerlingcr el al. • NIGI 11' MIGRANT FATALITIES AT WIND TURBINES

751

towers at which lai'ge-scale fatality events and lai'ge
numbers ot fatalities have been reported are taller
than wind turbines. Wind turbines have rarely
exceeded 125 m in height (this study), while
communication towers for which large-scale fatal-
ities have been found largely exceed 150 m and
often exceed 305 m (Avery et al. 1980, Shire et al.
2000). Thus, communication towers extend into
altitudes where more night migrants fly (Kerlinger
and Moore 1989), than do wind turbines. Second,
all communication towers for which lai'ge numbers
of night migrant fatalities and large-scale events
have been noted have guy wires (Shire et al. 2000),
whereas wind turbines do not.

The third reason for higher fatality rates and
large-scale fatality events at communication towers
opposed to wind turbines may be related to the
types of lights that are placed on communication
towers and wind turbines. A majority of commu-
nication towers equipped with aviation obstruction
lights have both steady-burning red (L-810) lights
and flashing red (L-864) lights (FAA 2000). Wind
turbines are most often equipped with only flashing
red (L-864) lights (FAA 2000). The four multi-bird
fatalities at single (or adjacent) turbines during a
single night occurred at turbines with flashing white
lights, steady-burning red lights, or ancillai7 facility
lighting (sodium vapor lamps). Flashing red lights
were not implicated in these multi-bird fatality
events at turbines.

Gehring et al. (2009) recently demonstrated for
communication towers that steady-burning red
lights attract night migrants, but flashing red
lights do not. Gehring et al. (2009) also demon-
strated towers at heights ranging from 116 to
146 m, supported by guy wires and equipped with
only flashing red lights, experienced 50-70%
fewer fatalities than towers with both flashing and
steady burning red lights. The average fatality rate
was 17.5 carcasses found per 40 days of peak
migration (20 days in spring and 20 days in fall)
per communication tower with both flashing and
steady-burning red lights. The estimated fatality
rate at these towers, when adjustments for
searcher efficiency and carcass removal by
scavengers is included (equal to about a 2-fold
increase; Gehring et al. 2009), is likely to be as
high as 70 night migrants/tower/year when
adjustments were made to include the entire
spring and fall migration seasons. Gehring et al.
(2009) reported 74 carcasses/40 days of searching/
year (spring and fall seasons) without carcass
removal or searcher efficiency adjustments at

towers >305 m in height with guy wires.
Adjustments for carcass removal and searcher
efficiency, as well as adjustments that include the
entire spring and fall migration seasons, increased
these estimates to —300 birds/tower/year. Thus,
communication towers equipped with guy wires
and a combination of flashing and steady-burning
red lights have fatality rates that are one to two
orders of magnitude greater than wind turbines.

We strongly suggest that reported fatalities of
night-migrating birds are minimal at wind tur-
bines based on results reported for communication
towers (Gehring et al. 2009), especially when
compared to tall, communication towers with guy
wires. Note that Smallwood et al. (2010) estimat-
ed mortality at wind turbines was greater when
novel scavenger removal methods were used. We
did not find evidence that large-scale fatality
events occur at wind turbines or that the flashing
red lights normally used on wind turbines cause
large numbers of fatalities of night migrants. Our
results, combined with those reported by Gehring
et al. (2009), strongly suggest that wind turbines
be equipped only with flashing red lights (strobe
or LED) and that steady burning red lights not be
used on turbines. It is prudent that post-construc-
tion fatality studies continue at wind turbines,
especially those taller than those we report, and at
turbines erected in geographic areas where studies
have not been conducted and greater numbers of
birds may migrate.

ACKNOWLEDGMENTS

We thank the diligent researchers, who spent thousands
of hours in the field collecting data, and the wind-power
industry, which funded the studies analyzed.

LITERATURE CITED

Anderson, R. L., M. Morrison, K. Sinclair, and M. D.
Strickland. 1999. Studying wind energy/bird inter-
actions: a guidance document. Metrics and methods
for determining or monitoring potential impacts on
birds at existing and proposed wind energy sites.
National Wind Coordinating Committee. Washington,
D.C., USA.

Avery, M., P. Springer, and .1. Cassel. 1976. The effects
of a tall tower on nocturnal bird migration — a portable
ceilometer study. Auk 93:281-291.

Avery, M. L., P. F. Springer, and N. S. Dailey. 1980.
Avian mortality at man-made structures: an annotated
bibliography. USDI, Fish and Wildlife Service. FWS/
OBS-80/54. Washington. D.C., USA.

Banks, R. C. 1979. Fluman related mortality of birds in the
United States. USDI, Fish and Wildlife Service,
Scientific Report Wildlife 215. Washington, D.C., USA.

752

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 4. December 2010

Caldwell, L. and G. Wallace. 1966. Collections of
migrating birds at Michigan television towers. Jack-
Pine Warbler 44:117-123.

Cochran, W. and R. Gr.aber. 1958. Attraction of
nocturnal migrants by lights on a television tower.
Wilson Bulletin 70:378-380.

Derby, C., A. Dahl, W. Erickson, K. Bay, and J. Hoban.
2007. Post-construction monitoring report for avian
and bat mortality at the NPPD Ainsworth Wind Farm.
Report to the Nebraska Public Power District,
Columbus, Nebraska, USA. Western EcoSystems
Technology Inc., Cheyenne, Wyoming, USA.

Erick-SON, W., K. Kronner, and B. Gritski. 2003. Nine
Canyon Wind Power Project avian and bat monitoring
report: September 2002-August 2003. Report to the
Nine Canyon Technical Advisory Committee and
Energy Northwest, Benton County, Washington,
USA. Western EcoSystems Technology Inc., Chey-
enne, Wyoming, USA.

Erickson, W. P., G. D. Johnson, and D. P. Young. 2005.
A summary and comparison of bird mortality from
anthropogenic causes with an emphasis on collisions.
Pages 1029-1042 in USDA, Forest Service, General
Technical Report PSW-GTR-191. Albany, California,
USA.

Erickson, W. P., G. D. Johnson, M. D. Strickland, and
K. Kronner. 2000. Avian and bat mortality associated
with the Vansycle Wind Project. Umatilla County,
Oregon: 1999 study year. Report to Umatilla County
Department of Resource Services and Development,
Pendleton, Oregon, USA. Western EcoSystems Tech-
nology Inc., Cheyenne, Wyoming, USA.

Erickson, W. P., J. Jeffrey, K. Kronner, and K. Bay.
2004. Stateline Wind Project Wildlife Monitoring
Final Report: July 2001-December 2003. Technical
report peer-reviewed by and submitted to FPL Energy,
the Oregon Energy Facility Siting Council, and the
Stateline Technical Advisory Committee, Umatilla
County. Oregon and Walla Walla County. Washing-
ton. USA. Western EcoSystems Technology Inc.,
Cheyenne, Wyoming, USA.

Federal Aviation Adminlstration (FAA). 2000. Ob-
struction marking and lighting. AC 70/7460- IK. U.S.
Department of Transportation. Washington. D.C.,
USA.

Fiedler, J. K.. T. H. Henry. R. D. Tankersley, and C. P.
Nicholson. 2007. Results ol' bat and bird mortality
monitoring at the expanded Buffalo Mountain Wind-
farm. 2005. Tennessee Valley Authority, Knoxville.
USA.

Gauthreaux Jr., S. and C. Belser. 2006. Effects of
artificial night lighting on migrating birds. Pages 67-
93 in Ecological consequences of artificial night
lighting (C. Rich and T. Longcore. Editors). Island
Press. Washington. D.C., USA.

Gauthreaux Jr.. S. A., C. G. Belser. and D. A. van
Bi.aricom. 2003. Using a network of WSR-88D weather
surveillance radars to define patterns o( bird migration at
large spatial scales. Pages 335-346 in Avian migration
(P. Bcrthold. E. Gwinner. and E. Sonnenschein. Editors).
Springer-Verlag. Berlin. Germany.

Gehring, j. L., P. Kerlinger, and a. M. Manville II.
2009. Communication towers, lights, and birds:
successful methods of reducing the frequency of avian
collisions. Ecological Applications 19: 505-514.

Howe, R. W., W. Evans, and A. T. Wolf. 2002. Effects of
wind turbines on birds and bats in northeastern
Wisconsin. Report to Wisconsin Public Service
Corporation and Madison Gas and Electric Company,
Madison, Wisconsin, USA. University of Wisconsin-
Green Bay, USA.

Jain, A. A. 2005. Bird and bat behavior and mortality at a
northern Iowa windfarm. Thesis. Iowa State Univer-
sity, Ames, USA.

Jain, A., P. Kerlinger, R. Curry, and L. Slobodnik.
2007. Annual report for the Maple Ridge Wind Power
Project: post-construction bird and bat fatality study-

2006. Report to PPM Energy and Horizon Energy and
Technical Advisory Committee for the Maple Ridge
Project Study, McLean, Virginia, USA. Curry &
Kerlinger LLC, McLean, Virginia, USA.

Jain, A., P. Kerlinger, R. Curry, and L. Slobodnik.
2009a. Annual report for the Maple Ridge Wind Power
Project: post-construction bird and bat fatality study -

2007. Report to PPM Energy and Horizon Energy and
Technical Advisory Committee for the Maple Ridge
Project Study. Houston, Texas, USA. Curry &
Kerlinger LLC. McLean, Virginia, USA.

Jain, A., P. Kerlinger, R. Curry, and L. Slobodnik.
2009b. Annual report for the Maple Ridge Wind
Power Project: post-construction bird and bat fatality
study-2008. Report to PPM Energy and Horizon
Energy and Technical Advisory Committee for the
Maple Ridge Project Study, Houston, Texas, USA.
Cuny & Kerlinger LLC. McLean, Virginia. USA.

Jain, A., P. Kerlinger, R. Curry, L. Slobodnik, J.
Histed, and J. Meacham. 2009c. Annual report for
the Noble Clinton Windpark LLC: postconstruction
bird and bat fatality study-2008. Report to Noble
Environmental Power LLC, Bliss, New York. USA.
Curry & Kerlinger LLC, McLean, Virginia, USA.

Jain, A., P. Kerlinger, R. Curry, L. Slobodnik, J.
Quant, and D. Pursell. 2009d. Annual report for the
Noble Bliss Windpark LLC: postconstruction bird and
bat fatality study-2008. Report to Noble Environmen-
tal Power LLC, Bliss, New York, USA. Curry &
Kerlinger LLC, McLean, Virginia, USA.

Jain. A., P. Kerlinger. R. Curry, L. Slobodnik, A.
FUER.ST. and C. Hansen. 2009e. Annual report for the
Noble Ellenburg Windpark LLC: postconstruction bird
and bat fatality study-2008. Report to Noble Environ-
mental Power LLC. Bliss, New York, USA. Curry &
Kerlinger LLC, McLean, Virginia, USA.

James, R. D. 2004. Bird observations at the Pickering wind
turbine. Ontario Birds 21:84-97.

James. R. D. 2008. Erie Shores Wind Farm. Port
Burwell. Ontario: fieldwork report for 2006 and
2007, during the first two years of operation. Report
to Environment Canada. Ontario Ministry of Natural
Resources, Erie Shores Wind Farm LP-McQuarrie
North American, and AIM PowerGen Corporation,
Ontario, Canada.

Kerlini^cr cl al. • NIGHT MIGRANT I’ATALITIES AT WIND TURBINES

753

James, R. D. and G. Coady. 2003. Bird monitoring at
Toronto’.s Exhibition Place wind turbine. Ontario
Birds 22:79-88.

Johnson, G., W. Erickson, J. White, and R. McKinney.
2003. Avian and bat mortality during the first year of
operation at the Klondike Phase I Wind Project,
Sherman County, Oregon. Draft report to Northwest-
ern Wind Power. Goldendale, Washington, USA.
Western EcoSystems Technology Inc., Cheyenne,
Wyoming, USA.

Johnson, G. D., W. P. Erickson, M. D. Strickland, M. F.
Shepherd, and D. A. Shepherd. 2000. Avian
monitoring studies at the Buffalo Ridge. Minnesota
Wind Resource Area: results of a 4-year study. Report
to Northern States Power Company. Minneapolis,
Minnesota. USA. Western EcoSystems Technology
Inc., Cheyenne, Wyoming, USA.

Johnson, G. D., W. P. Erickson, M. D. Strickland, M. F.
Shepherd, D. A. Shepherd, and S. A. Sarappo.
2002. Collision mortality of local and migrant birds at
the large-scale wind power development on Buffalo
Ridge, Minnesota. Wildlife Society Bulletin 30:879-
887.

Kerlinger, P. 2000. Avian mortality at communications
towers: a review of recent literature, research, and
methodology. Report to the USDI, Fish and Wildlife
Service, Washington, D.C., USA. Curry & Kerlinger
LLC, McLean, Virginia, USA.

Kerlinger. P. 2001. Avian mortality study at the Green
Mountain Windpower Project, Garrett, Somerset
County, Pennsylvania: 2000-2001. Report to National
Wind Power, Oakland, California, USA. Curry &
Kerlinger LLC, McLean, Virginia, USA.

Kerlinger, P. 2002a. An assessment of the impacts of
Green Mountain Power Corporation's wind power
facility on breeding and migrating birds in Searsburg,
Vermont. Report NREL/SR-500-28591 to National
Renewable Energy Laboratory, U.S. Department of
Energy, Golden, Colorado. USA. Cape May, New
Jersey, USA.

Kerlinger, P. 2002b. Avian fatality study at the Madison
Wind Power Project, Madison, New York. Report to
PG&E Generating, Athens, New York, USA. Curry &
Kerlinger LLC, McLean, Virginia, USA.

Kerlinger, P. and R. Curry. 2000. Impacts of a small
wind power facility in Weld County, Colorado, on
breeding, migrating, and wintering birds: preliminary
results and conclusions. Pages 64-69 in Proceedings of
the National Avian-Wind Power Planning Meeting III.
National Wind Coordinating Committee/RESOLVE,
Washington, D.C., USA.

Kerlinger. P. and F. R. Moore. 1989. Atmospheric
structure and avian migration. Current Ornithology
6:109-142.

Kerlinger. P., R. Curry, A. FIasch, and J. Guarnaccia.
2007. Migratory bird and bat monitoring study at the
Crescent Ridge Wind Power Project. Bureau County,
Illinois: September 2005-August 2006. Report to
Orrick, Herrington, and Sutcliffe LLP, Washington,
D.C, USA. Curry & Kerlinger LLC, McLean, Virginia,
USA.

Kerlinger, P., R. Curry, L. Culr, A. Jain, C. Wilk-
erson, B. Fischer, and A. Ha.sch. 2006. Post-
construction avian and bat I'alality monitoring study
for the High Winds wind power project, Solano
County, California: two year report. Report to High
Winds LLC and FPL Energy, Livermore. California,
USA. Curry & Kerlinger LLC, McLean, Virginia,
USA.

Kerns, J. and P. Kerlinger. 2004. A study of bird and bat
collision fatalities at the Mountaineer Wind Energy
Center, Tucker County, West Virginia: annual report
for 2003. Report to FPL Energy and the MWEC
Technical Review Committee, Vero Beach, Florida,
USA. Curry & Kerlinger LLC, McLean, Virginia,
USA.

Koford, R., a. Jain, G. Zenner, and A. Hancock. 2005.
Avian mortality associated with the Top of Iowa Wind
Farm. Report to the Iowa Department of Natural
Resources. Ames, USA. Iowa State University, Ames.
USA.

Larkin, R. and B. Frase. 1988. Circular paths of birds
flying near a broadcasting tower in cloud. Journal of
Comparative Psychology 102:90-93.

Lowrey Jr, G. H. and R. J. Newman. 1966. A continent-
wide view of bird migration on four nights in October.
Auk 83:547-586.

National Research Council. 2007. Environmental
impacts of wind-energy projects. NRC, Committee
on Environmental Impacts of Wind Energy Projects.
The National Academies Press, Washington, D.C.,
USA.

New Jersey Audubon Society. 2008. Post-construction
wildlife monitoring at the Atlantic County Utilities
Authority-Jersey Atlantic Wind Power Facility: peri-
odic report covering work conducted between 1
January and 30 September 2008. Report to the New
Jersey Board of Public Utilities and New Jersey Clean
Energy Program, Newark, New Jersey, USA. New
Jersey Audubon Society, Cape May Court House. New
Jersey, USA.

Nicholson, C. P., R. G. Tankersley Jr., J. K, Fiedler,
and N. S. Nicholas. 2005. Assessment and prediction
of bird and bat mortality at wind energy facilities in
the southeastern United States: final report. 2005.
Tennessee Valley Authority, Knoxville, USA.

ScHNELL. G. D.. E. A. Mosteller, and j. Grzybowski.
2007. Post-construction avian/bat risk assessment and
fatality study for the Blue Canyon II Wind Power
Project. Oklahoma: summary of first-year findings.
Report to Horizon Wind Energy. Houston. Texas.
USA. Sam Noble Oklahoma Museum of Natural
History, University of Oklahoma. Norman, USA

Shire, G. G., K. Brown, and G. Winegrad. 2000.
Communication towers: a deadly hazard to birds.
American Bird Con.servancy, Washington. D.C.. USA.

Smallwood, K. S., D. A, Bell, S. A, Snyder, and J. E.
DiDonato. 2010. Novel scavenger removal trials
increase wind turbine-caused avian fatality estimates.
Journal of Wildlife Management 74:1089-1097.

Stantec CONSUI.TING. 2008. 2007 spring, summer, and fall
post-construction bird and bat mortality study at the

754

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4. December 2010

Mars Hill wind farm, Maine. Report to UPC Wind
Management LLC, Cumberland, Maine, USA. Stantec
Consulting, Topsham, Maine, USA.

Stantec Consulting. 2009. Post-construction monitoring
at the Mars Hill wind farm, Maine-year 2, 2008.
Report to First Wind Management LLC, Portland,
Maine, USA. Stantec Consulting, Topsham, Maine,
USA.

Tierney, R. 2007. Buffalo Gap I Wind Farni avian
mortality study, February 2006-January 2007, final
survey report. Report to AES West Inc., Albuquerque,
New Mexico, USA. TRC, Albuquerque, New Mexico,
USA.

Trapp, J. L. 1998. Bird kills at towers and other man-made
structures: an annotated partial bibliography (1960-
1998). USDI, Fish and Wildlife Service, Washington,
D.C., USA.

TRC Environmental Corporation. 2008. Post-construc-
tion avian and bat fatality monitoring and grassland
bird displacement surveys at the Judith Gap Wind
Energy Project, Wheatland County, Montana. Report

to Judith Gap Energy LLC, Chicago, Illinois, USA.
TRC Environmental Corporation, Laramie, Wyoming,
USA.

WiLCOXON, F. 1945. Individual comparisons by ranking
methods. Biometrics 1:80-83.

Young Jr., D. P., W. P. Erickson, R. E. Good, M. D.
Strickland, and G. D. Johnson. 2003. Avian and bat
mortality associated with the initial phase of the Eoote
Creek Rim Windpower Project, Carbon County,
Wyoming: November 1998-June 2002. Report to
Pacificorp Inc., Portland, Oregon, SeaWest Wind-
power Inc., San Diego, California, and Bureau of Land
Management, Rawlins District Office, Rawlins,
Wyoming, USA. Western EcoSystems Technology
Inc., Cheyenne, Wyoming, USA.

Young Jr., D. P., W. P. Erickson, K. Bay, S. Nomani, and
W. Tidhar. 2009. Mount Storm Wind Energy Facility,
phase 1 post-construction avian and bat monitoring,
July-October 2008. Report to NedPower Mount Storm
LLC, Houston, Texas, USA. Western EcoSystems
Technology Inc., Cheyenne, Wyoming, USA.

The H'ilsoii Jouriuil oJOniilhology 122(4):755-761, 20 10

DEVELOPMENTAL CHANGES IN SERUM ANDROGEN LEVELS OE
EASTERN SCREECH-OWLS {MEGASCOPS ASIO)

CORINNE P. KOZLOWSKI' -^ AND D. CALDWELL HAHN^

ABSTRACT. — We studied androgen production during development in nestling Eastern Screech-Owls (Megascops asio)
and hypothesized that gender and hatch order might inOuence serum levels of testosterone and androstenedione.
Testosterone levels were highest immediately after hatching and declined significantly in the 4 weeks leading to fledging.
The average level of testosterone for 1-7 day-old owls was 3.99 ± 0.68 ng/ml. At 22-28 days of age, the average
testosterone level for nestling owls was 0.83 ± 0. 1 8 ng/ml. Testosterone levels did not differ between males or females. The
average testosterone level for male nestlings was 2.23 ± 0.29 ng/ml and 2.39 ± 0.56 ng/ml for female nestlings. The
average level of androstenedione tor nestling owls was 1.92 ±0.11 ng/ml and levels remained constant throughout
development. Levels were significantly higher in males than females. The average androstenedione level was 1.77 ±
0.16 ng/ml tor male nestlings and 1.05 ± 0.24 ng/ml tor female nestlings. Hatching order did not affect levels of either
androgen. Our results provide a foundation for future studies of androgen production by nestling owls. Received 19 January
2010. Accepted 30 June 2010.

Androgens influence many aspects of adult bird
physiology and behavior, including metabolic rate
(Hannsler and Prinzinger 1979), lipid storage
(Wingfield 1984), timing of molt (Runfeldt and
Wingfield 1985), song production (Hunt et al.
1997), and aggression (Hau et al. 2000). Andro-
gens also influence the behavior of nestling birds,
as testosterone levels in earlier-hatching White
Stork nestlings {Ciconia ciconia) and Tawny Owl
nestlings {Strix aluco) are correlated with elevated
aggression towards siblings (Sasvari et al. 1999,
2006). Testosterone enhances the probability of
aggressive behavior in young Domestic Chickens
(Callus gallus domesticus) (Young and Rogers
1978) and facilitates aggression in Black-headed
Gull chicks (Chroicoceplialus ridihundus)
(Groothuis and Meeuwissen 1992, Ros et al.
2002). Testosterone also facilitates begging be-
havior in nestling Atlantic Canaries (Serinus
Canaria) (Buchanan et al. 2007) and European
Pied Flycatchers (Ficedida hypoleuca) (Goodship
and Buchanan 2006).

Sex steroids are present early in avian devel-
opment and are important in sexual differentia-
tion. Testicular hormone synthesis in Domestic
Chicken embryos has first been detected on day 6
of incubation, and peaks in plasma levels of
testosterone occur on day 14 (Woods and Weeks

' St. Louis Zoo, Research Department, 1 Government
Drive, St. Louis, MO 63110, USA.

^University of Missouri-St. Louis, Biology Department,
8001 Natural Bridge Road, St. Louis, MO 63121, USA.

^USGS, Patuxent Wildlife Researeh Center, 12100
Beeeh Forest Road, Laurel, MD 20708, USA.

“Corresponding author; e-mail:
corinnekozIowski@umsl.edu

1969). Similarly, androgen production begins on
day 8 of incubation in embryos of male Japanese
Quail (Coturnix japonica), and androgen concen-
trations in plasma reach a plateau before hatching
(Ottinger and Bakst 1981, Schumacher et al.
1988). Less is known concerning the maturation
of the hypothalamic-pituitary-gonadal (HPG) axis
in altricial birds. Sex steroids are present in the
plasma of altricial hatchlings of several species
(Hutchison et al. 1984, Adkins-Regan et al. 1990,
Schlinger and Arnold 1992, Silverin and Sharp
1996). However, the typical levels of androgens
produced by nestling birds, whether androgens are
produced primarily in the gonads or the adrenal
glands, or how these levels change through
development have not been studied in detail.
Many field studies sample individuals only once
(e.g., Fargallo et al. 2007, Gil et al. 2008).
However, a number of studies suggest that
circulating androgen levels can vary during
development. Testosterone levels of ne.stling
Great Tits (Panis major) decline significantly
during development (Silverin and Sharp 1996),
and circulating androgen levels initially increase,
then decrea.se post-hatching in nestling Zebra
Finches (Taeniopygia guttata) (Adkins-Regan et
al. 1990).

Evidence concerning differences between male
and female nestlings varies among species. Male
and female Common Kestrel (Falco tinnunculus)
(Fargallo et al. 2007), European Pied Flycatchers
(Goodship and Buchanan 2006), Zebra Finches
(Naguib et al. 2004), and nestling Lesser Black-
backed Gulls (Lams fuscus) (Verboven et al.
2003) show no difference in circulating androgen
levels. However, nestling male Black Coucals

755

756

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 4. December 2010

{Centropus grillii) have higher levels of circulat-
ing androgens than females (Goymann et al.

2005) , and nestling female Atlantic Canaries
(Weichel et al. 1986) and Japanese Quail
(Ottinger et al. 2001) have higher levels of
circulating androgens than males.

Hormonal differences have also been reported
among nestlings in regards to hatching order and
may be due to social interactions (Wingfield et al.
1990). Second-hatching chicks were shown to
have higher levels of circulating testosterone 1 day
after hatching, as well as during aggressive
encounters with older siblings in obligately
siblicidal Nazca Boobies (Siila grcinti) (Ferree et
al. 2004). First-hatched White Stork nestlings are
more aggressive, receive more food, and have
higher plasma levels of testosterone than their
siblings (Sasvari et al. 1999). Similarly, plasma
testosterone levels decreased with hatching order
in 3-day-old Tawny Owls raised by females in
poor condition. Owlets raised by females in good
condition, however, had similar levels of testos-
terone, regardless of hatching order (Sasvari et al.

2006) . Hatching order in other species is not
related to plasma androgen levels (Naguib et al.
2004).

We describe developmental patterns of andro-
gen production by Eastern Screech-Owls (Mega-
scops asio) during the nestling phase. Androgens,
including testosterone and androstenedione, are
produced primarily by the gonads. The adrenal
glands are also a source of androgens, specifically
androstenedione and dehydroepiandrosterone
(DHEA), but production is minimal compared to
the gonads. Most androgenic effects are mediated
by testosterone and dihydrotestosterone. Andro-
stenedione also circulates in the bloodstream at
relatively high levels, but it has a significantly
lower affinity for the androgen receptor compared
to testosterone and dihydrotestosterone, and a
relatively low biological activity in vivo. Instead,
androstenedione serves primarily as a substrate
for the production of testosterone and estrogens
(Hadley and Levine 2007). Several studies have
characterized levels of testosterone in nestling
birds (Fargallo et al. 2007. Gil et al. 2008), but
little is known about typical serum levels or
developmental patterns of androstenedione.

We measured .serum testosterone and andro-
stenedione levels weekly in captive nestling
Eastern Screech-Owls from hatching until Hedg-
ing. Our objectives were to: ( I ) characterize levels
of testosterone and androstenedione produced by

nestling owls, (2) compare how these levels
changed during the developmental period between
hatching and fledging, (3) ascertain whether levels
differ between male and female nestlings, and (4)
whether levels differ among nestlings in regard to
hatching order.

METHODS

Species and Study Area. — The Eastern Screech-
Owl is a small owl (males are typically 160 g,
females 200 g) with a distribution that extends
east of the Rocky Mountains and south from the
Canadian boreal forest to the Tropic of Cancer in
Mexico (Gehlbach 1994). Clutches normally
consist of 4-6 eggs; the first 2-3 eggs are laid
1 day apart with increasing intervals afterwards
(Gehlbach 1994). Onset of incubation varies
among females and depends on environmental
conditions, but may begin with the first, second,
or third egg. Incubation typically lasts 28 days;
eggs hatch asynchronously over 2-3 days, and
sibling feeding hierarchies are established. The
young fledge between 28 and 30 days of age
(Gehlbach 1994).

The owls used in this study are part of a captive
colony of breeding birds maintained at Patuxent
Wildlife Research Center (PWRC) in Laurel,
Maryland, USA. Owls are fed two mice per bird
daily and Nebraska Bird of Prey Diet fortified
with Vionate. They are housed in 12 X 3 m
outdoor flight cages that contain nest-boxes, and
pairs are kept together year-round. Egg-laying
begins in March. Females solely incubate, and
males provide food to incubating females and
nestlings after hatching.

Serum Collection and Androgen Assays. — Fif-
ty-two blood samples were collected weekly during
spring 2007 from 26 nestlings in nine clutches.
Samples were collected only from nestlings
hatching from first-laid clutches, and nestlings
hatched between 28 April and 13 May 2007.
Individual owls ranged from 2 to 27 days of age,
and clutches contained 2-5 nestlings. Seven owl
nestlings were sampled once, 12 were sampled
twice, and 7 were sampled three times. No birds
were sampled more that three times during the
4 weeks of blood collection. Serum samples were
obtained by venipuncture from the brachial vein
using a 27-gauge needle until the last week before
Hedging, when they were taken from the jugular
vein using a 25-gauge needle. A separate blood
sample was also taken for identification of males
and females Just before Hedging. Seium samples

Kozlowski imd Hahn • SERUM ANDROCJEN LEVELS IN NESTEINCJ OWLS

757

were frozen until shipped to the St. Louis Zoo for
hormone analysis.

Serum samples were transferred to 2-ml
eppendorf tubes and diluted 1:2 with phospho-
buftered saline (PBS). Samples were mixed with a
vortex and an equal volume of 100% ethanol was
added to precipitate proteins and lipids following
Kozlowski and Ricklefs (2010). Samples were
immediately homogenized upon adding ethanol
and allowed to incubate at room temperate for
10 min. Samples were then spun at 12,282 g in a
centrifuge for 10 min. The supernatant was poured
into a sterile cryule tube and frozen at —70° C
until an assay was performed.

All samples were analyzed using radioimmu-
noassay (RIA) in the Endocrinology Laboratory at
the St. Louis Zoo. Ethanol extracts were thawed
and spun in a centrifuge at 12,282 g for 10 min to
remove any remaining lipids in preparation for
assay. Testosterone and androstenedione concen-
trations were measured using commercially avail-
able coated-tube radioimmunoassay kits (Coat-A-
Count © Total Testosterone 1251 Kit, and Coat-A-
Count © Direct Androstenedione 1251 Kit, Diag-
nostic Products Corporation, Los Angeles, CA,
USA). The lowest detection limit was 0.05 ng/ml
and upper limit was 40 ng/ml in the testosterone
assay. The lowest detection limit of the andro-
stenedione assay was 0.04 ng/ml and the upper
detection limit was 8.7 ng/ml. Both kits used in
this study have highly specific antibodies. The
testosterone antibody cross-reacts as follows: 5(3-
androstan-3a, 17(3-diol: 0.4%; androstenedione:
0.5%; 5(3-androstan-3p, 17p-diol: 0.2%; 5a-dihy-
drotestosterone: 3.3%; 5( 10)-estren-17a-ethinyl-
17P-ol-3-one: 0.2%; 4-estren-17a-methyl-17P-ol-
3-one: 1.1%; 4-estren- 1 7-ol-3-one: 20%; 19-
nortestosterone: 20%; ethisterone: 0.7%; 19-
hydroxyandrostenedione: 2.0%; 1 1-ketotestoster-
one: 16%; methyltestosterone: 1.7%; norethin-
drone: 0.1%; 1 1 P-hydroxyyestosterone 0.8%; and
triamincinolone: 0.2%. The androstenedione an-
tibody cross-reacts as follows: androsterone:
0.14%; DHEA: 0.16%; progesterone: 0.16%;
spironolactone: 0.11%; 5a-dihydrotestosterone:
0.21%; and testosterone: 1.49%. Cross-reactivi-
ties for all other compounds are below 0.1%.

Assays were conducted following kit directions
with the exception that kit standards, which are
supplied in human serum, were replaced by
standards diluted in 10% steroid-free calf serum.
Standard diluent was added to extracted serum
samples, and steroid-free owl serum extract was

added to standards and quality controls to equalize
the matrices of standards and samples. Owl .serum
extract and calf serum were stripped of steroids
using dextran-coated charcoal (DCC# 6241,
Sigma Chemical, St. Louis, MO, USA) prior to
use. All samples were analyzed in the same assay
in duplicate for both testosterone and androstene-
dione. Mean ± SE intra-assay variation of
duplicate samples was 8.96 ± 1.21 for testoster-
one and 9.66 ± 2.42 for androstenedione.

Assay Validation. — We added a known amount
of radioactively labeled hormone to a serum
sample before extraction to measure extraction
efficiency of both testosterone and androstenedi-
one, and measured the amount of radioactivity
after the extraction process. Serum was pooled
from the experimental samples and divided into
10 samples, each containing 15 pL of serum. In
five samples each for testosterone and andro-
stenedione, 15 pL of 1-125 labeled hormone and
15 pL of PBS were added to the 15 pL of serum.
Control samples consisted of 15 pL of labeled
hormone and 30 pL of PBS. Two samples of
15 pL of 1-125 hormone were set aside to
measure the total radioactivity present in the
sample. Serum and control samples were extract-
ed as described, and the supernatant was
transferred from each sample to an individual
12 X 75 plastic test tube. The total amount of
radioactivity in each sample was measured in
both serum and control samples, and compared to
the total count tubes to calculate the recovery
percentage in each sample. Percent recovery was
91-97% for both hormones and did not differ
between control and serum samples (testosterone:
t = 1 .22, df = 8, P = 0.26; androstenedione: t =
0.73, df = 8, P = 0.49).

Pour samples that contained high levels of
hormone were diluted by 1/2, 1/4, and 1/8 with
steroid-free owl serum extract for both testoster-
one and androstenedione. Serial dilutions of
extracted .serum samples were parallel to the
standard curve (test of equal slopes, P > 0.34 for
testosterone; P > 0.29 for androstenedione) (Zar
1999), supporting that no additional substances in
the extract were cross reacting with the antibody.
We assessed the accuracy of both assays by
adding a known amount of either testosterone or
androstenedione to four serum extracts containing
low values of hormone. Addition of known
amounts of the androgens at two dosage levels
resulted in recovery of 101 ± 3.6% of added
testosterone and 96 ± 2.9% for androstenedione.

758

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, December 2010

O)

c

c

o

c

0)

o

c

o

o

c

0)

O)

o

•D

c

05

£

0)

CO

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0

• Male
O Female

\

\

Testosterone Androstenedione

FIG. I. Concentration {x ± SE) of testosterone and
androstenedione in the serum of male and female Eastern
Screech-Owl nestlings (n = 19 males, 7 females).

Identification of Males and Females. — We
followed Griffiths et al. (1998) to identify each
individual nestling Screech-Owl as male or
female by amplifying introns of the genes CHD-
W and CHD-Z through PCR.

Statistical Procedures. — All statistical analysis
was conducted using JMP 7.0.2 © (SAS Institute
2008). Blood samples were collected weekly, and
nestlings were not sampled on the same day
during each week; thus, nestlings were divided
into 7-day age classes. These classes ranged from
1-7 to 22-28 days of age. General Linear Models
were used to assess the relationship between
testosterone and androstenedione in relation to
age-class, hatch-order, brood size, and male/
female. A separate analysis was conducted for
each androgen. Nest of origin and individual owl
were treated as random effects. Hatch-order,
male/female, brood size, and all two-way inter-
actions were included as fixed effects. Non-
significant fixed effects were dropped from the
model. A Tukey HSD test was used to separate
means when results were significant.

RESULTS

Both testosterone and androstenedione were
detected in the serum of nestling Screech-Owls.
The average concentration of testosterone in
nestling owls was 2.30 ± 0.26 ng/ml and levels
ranged from 0.17 to 7.59 ng/ml. The average
concentration of androstenedione was 1 .92 ±
0.1 1 ng/ml and ranged from 0.98 to 3.51 ng/ml.

Testosterone and androstenedione levels were
unaffected by both brood size and hatch-order, and
these effects were dropped from both statistical
models. Seven of the nestlings in our study were

5.0

O)

c

c

o

4.0

03

L_

O 3.0

o

c

o

o

c

(D

O)

O

k_

C

03

E

D

w

2.0

1.0

0.0

• Testosterone
O Androstenedione

14

21

28

Age (days)

FIG. 2. Weekly concentration (± SE) of testosterone
and androstenedione in the serum of Eastern Screech-Owl
nestlings: n = \2 nestlings in age-class 1; n = 18 nestlings
in age-class 2; /t = 17 nestlings in age-class 3; /t = 5
nestlings in age-class 4.

females, and 19 were males. No differences in
serum testosterone were observed between male
and female nestlings (7^1,46 = 0.68, P = 0.41).
Average testosterone levels were 2.23 ± 0.29 ng/ml
for male nestlings and 2.39 ± 0.56 ng/ml for female
nestlings. In contrast, serum androstenedione levels
differed significantly between male and female
nestlings. Male owls had higher levels of andro-
stenedione than females (F| 4^ = 5.173, P = 0.030).
Average androstenedione levels were 1.77 ±
0.16 ng/ml for male nestlings and 1.05 ± 0.24 ng/
ml for female nestlings (Fig. 1).

Testosterone levels did not differ between male
and female nestlings, and the data were pooled for
analysis. Nestling testosterone concentrations
declined with age (P3.46 = 11.489, P < 0.001)
(Fig. 2). In contrast, androstenedione levels did
not vary by age for either male (^3,46 = 1.58, P =
0.21 ) or female (^3,46 = 0.37, P = 0.78) nestlings.
Testosterone levels were higher than androstene-
dione levels during the 2 weeks following
hatching. However, by weeks 3 and 4, testoster-
one and androstenedione were present at similar
levels. Average testosterone levels in chicks 1-
7 days of age measured 3.99 ± 0.68 ng/ml. By
week 4 (days 22-28), levels declined to less that
one-quarter of their initial level, to 0.83 ±0.18 ng/
ml. Androstenedione concentrations averaged
1.72 ± 0.38 ng/ml for 1-7 day old male nestlings
and 1.01 ± 0.65 ng/ml for 1-7 day old female
nestlings. Average androstenedione levels by 22-
28 days of age were 1.39 ± 0.15 ng/ml for males
and 1 .25 ± 0.58 for females.

Kozlonski and Halm • SERUM ANDROGEN LEVELS IN NESTLING OWLS

759

DISCUSSION

Testosterone and androstenedione are both
measurable in nestling Screech-Owl serum from
week 1 post-hatching until fledging. Our results
demonstrate both a significant change in the level of
testosterone during development and a difference
in the pattern of testosterone and androstenedione
during development. Testosterone is highest during
the first week following hatching, and then
declines. Androstenedione levels remain relatively
constant from hatching through fledging.

The HPG axis has been studied in detail in
precocial species, but less is known about honnone
production by altricial nestlings. The ovary and
testes actively produce steroid hormones, including
testosterone, during early embryonic development
through post-embryonic development in precocial
Domestic Chickens (Tanabe et al. 1979) and
Mallards {Anas platyrhynchos) (Tanabe et al.
1983). The adrenal glands also appear to be
important for androgen synthesis in precocial
species. Testosterone is found at high concentra-
tions in the adrenal gland shortly before hatching in
both Domestic Chickens (Tanabe et al. 1979) and
Mallards (Tanabe et al. 1983).

Developmental studies of the HPG axis in
altricial species are restricted to passerines. Limited
work suggests the HPG axis of these species
matures later in development. Silverin and Sharp
(1996) demonstrated the hypothalamus and pitui-
tary gland in Great Tits become functional 9 days
after hatching, approximately halfway through their
nestling phase. Injections of GnRH at this age elicit
increases in male testosterone or female estradiol.
The adrenal gland may also be an important source
of androgens for altricial nestlings. Castration of
nestling Zebra Finches did not significantly reduce
plasma androgen levels, suggesting these androgens
may arise primarily from the adrenal gland
(Adkins-Regan et al. 1990). Our study suggests
nestling Screech-Owls are capable of producing
measureable levels of both testosterone and andro-
stenedione. However, whether these androgens are
gonadal or adrenal is unknown. Further study is
needed to ascertain if development of the HPG axis
in other altricial species, including owls, is similar
to passerine species.

We found a decrease in nestling testosterone
levels during post-hatching development similar
to other studies (Adkins-Regan et al. 1990,
Silverin and Sharp 1996). Testosterone levels in
Great Tit nestlings are highest in newly hatched

birds, and reach basal levels by 3 days of age
(Silverin and Sharp 1996). Male Zebra Finches
have higher levels of plasma testosterone during
days 2-14 of development than in older nestlings
(Adkins-Regan et al. 1990). Similarly, Schlinger
and Arnold (1992) found male Zebra Finches had
higher levels of plasma androgens at 7-9 days of
age compared to 4-6-day-old birds or 10- 13-day-
old birds. Elevated testosterone levels are also
characteristic of the period surrounding hatching
in several precocial species, including Japanese
Quail (Ottinger and Bakst 1981, Schumacher et al.
1988) and Domestic Chickens (Tanabe et al. 1979).
This suggests testosterone may have a role in
hatching. In contrast, androstenedione levels of
nestling owls did not vary during post-hatching
development. Androstenedione levels of precocial
species change during development, but little is
known about developmental patterns in altricial
nestlings. Androstenedione levels in Domestic
Chickens are elevated in pre-pubescent males
(Culbert et al. 1977), and then decrease between 9
and 16 weeks of age when testicular maturation
occurs (Sharp et al. 1977). Whether similar changes
occur during nestling owl development is unknown.

We found no differences in testosterone levels
between male and female nestling Screech-Owls at
any stage of development. This is similar to the
findings for several other species (Groothuis and
Meeuwissen 1992, Verboven et al. 2003) including
Common Kestrels (Fargallo et al. 2007). In
contrast, we detected a difference in androstenedi-
one level, as male Screech-Owls had higher levels
of androstenedione than females from hatching
through fledging. There is limited evidence that
male nestlings of other species may also maintain
higher levels of androstenedione than females.
Male European Starling {Sturnus vulgaris) nest-
lings have been shown to produce higher levels of
DHEA (a precursor of androstenedione) than
female nestlings (Chin et al. 2008). Elevated levels
of testosterone in nestling birds have been corre-
lated with behavioral changes, including enhanced
aggression (Sasvari et al. 1999, 2006), and begging
intensity (Goodship and Buchanan 2006). Howev-
er, it is unknown whether higher levels of
androstenedione in male nestlings would produce
similar effects. Androstenedione has a much lower
affinity for the androgen receptor compared to
testosterone and a relatively low biological activity
(Hadley and Levine 2007). Whether higher levels
of androstenedione in male nestlings have any
physiological effects is unknown.

760

THE WILSON JOURNAL OF ORNITHOLOGY • Vul. 122, No. 4. December 2010

Hatching order was not associated with differ-
ences in androgen levels in nestling Screech-
Owls. The correlation between hatching order and
serum androgens can be associated with the
presence of sibling feeding hierarchies. First-
hatched nestlings in several species have higher
plasma levels of testosterone, are more aggres-
sive, and receive more food than younger siblings
(Sasvari et al. 1999, 2006). Sibling feeding
hierarchies are often established in wild Screech-
Owl broods, and nestlings vigorously compete
for food resources (Gehlbach 1994). However,
the owls in this study were from a captive colony
where they have an abundant food supply. High
androgen levels are known to be immunosuppres-
sive and reduce growth for nestlings, in addition to
increasing their aggressive behavior (Ros 1999,
Fargallo et al. 2007 ). We have observed that captive
owl nestlings show little sibling competition, and
lack of a correlation between androgens and
hatching order may be associated with reduced
competition. Future work will characterize the
androgen levels of Screech-Owl nestlings when
food resources are restricted. This will allow us to
ascertain if androgen levels are affected by hatching
order when there is strong sibling competition
among siblings for food.

ACKNOWLEDGMENTS

We thank USGS Patuxent Wildlife Research Center and
G. W. Smith for use of the owl colony and research support
to DCH, and the Maryland Ornithological Society,
American Mu.seum of Natural History Frank W. Chapman
Fund, and Sigma Xi for research support to CPK. We thank
J. E. Bauman and the St. Louis Zoo Research Department
for access to endocrinology laboratory facilities and R. E.
Ricklefs for access to genetic laboratory facilities. We
thank W. C. Bauer and M. M. Maxie for care of the owls,
and Kelly Amy and Nathan Rolls for assistance in
collecting blood samples. Helpful reviews of the manuscript
were provided by B. A. Rattner. D. J. Hoffman, J. .1.
Atwood, and two anonymous reviewers. Use of trade,
product, or firm names does not imply endorsement by the
U.S. Government.

LITERATURE CITED

Adkin.s-Rf.gan, E.. M. Abdelnabi. M. Mobarak, and
M. A. Ottinger. 1990. Sex .steroid levels in developing
and adult male and female Zebra Finches (Poephila
fpittatci). General and Comparative Endocrinology
78:93-109.

Buchanan, K. L., A. R. Goi.dsmith. C. A. Hinde, S. C.
Grifmth, and R. M. Kii.nf.r. 2007. Does testosterone
mediate the trade-off between nestling begging and
growth in the Canary (Seriinis canarin)! Hormones
and Behavior .32:664-671.

CuLBERT, J., P. J. Sharp, and J. W. Wells. 1977.
Concentrations of androstenedione. testosterone and
LH in the blood before and after the onset of
spermatogenesis in the cockerel. Reproduction and
Fertility 51:153-154.

Chin, E. H., A. H. Shah, K. L. Schmidt, L. D. Sheldon,
O. P. Love, and K. K. Soma. 2008. Sex differences
in DHEA and estradiol during development in a wild
songbird: jugular versus brachial plasma. Hormones
and Behavior 54:194-202.

Fargallo, J. A., J. Marti'nez-Padilla, A. Toledano-
Diaz. j. Santiago-Moreno, and j. a. DAvila. 2007.
Sex and testosterone effects on growth, immunity, and
melanin coloration in nestling Eurasian Kestrels.
Journal of Animal Ecology 76:201-209.

Ferree, E. D., M, C. Wikelski, and D. J. Anderson.

2004. Hormonal correlates of siblicide in Nazca
Boobies: support for the challenge hypothesis. Animal
Behavior 46:655-662.

Gehlbach, F. R. 1994. The Eastern Screech Owl: life
history, ecology and behavior in the suburbs and
countryside. Texas A & M University Press, College
Station, USA.

Gil, D., E. Bulmer, P. Celis, and M. Puerta. 2008.
Increased sibling competition does not increase
testosterone or corticosterone levels in nestlings of
the Spotless Starling (Stiirnus iinicolor). Hormones
and Behavior 54:238-243.

Goodship, N. M. and K. L. Buchanan. 2006. Nestling
testosterone is associated with begging behaviour and
fledging success in the Pied Flycatcher, Ficedula
hypoleitca. Proceedings of the Royal Society of
London, Series B 273:71-76.

Goymann, W., B. Kempenaers, and j. C. Wingfield.

2005. Breeding biology, sexually dimoiphic develop-
ment and nestling testosterone concentrations of the
classically polyandrous African Black Coucal, Cen-
tropus grillii. Journal of Ornithology 146:314-324.

Griffiths, R., M. C. Double, K. Orr, and R. J. Dawson.
1998. A DNA test to sex most birds. Molecular
Ecology 7:1071-1075.

Groothuis, T. G. G. and G. Meeuwissen. 1992. The
influence of testosterone on the development and
fixation of form of displays in two age classes of young
Black-Headed Gulls. Animal Behaviour 43:189-208.

Hadley, M. E. and J. E. Levine. 2007. Endocrinology.
Sixth Edition. Pearson Prentice-Hall, Upper Saddle
River, New Jersey, USA.

Hannsler, I. AND R. Prinzinger. 1979. The influence of the
.sex hormone testosterone on body temperature and
metabolism of Japanese Quail. Experientia 35:509-5 10,

Hau, M., M. Wikelski, K. K. Soma, and J. C. Wingfield.
2000. Testosterone and year-round territorial aggres-
sion in a tropical bird. General and Comparative
Endocrinology 117:20-33.

Hunt, K. E., T. P. Hahn, and J. C. Wingfield. 1997.
Testosterone implants increase song but not aggression
in male Lapland Longspurs. Animal Behaviour
.54:1 177-1 192.

Hutchison, J. B., J. C. Wingfiei.d, and R. E. Hutchison.
1984. Sex steroid differences in plasma concentrations

Koz/ow.ski ami Hahn • SERUM ANDROCJEN LEVEES IN NESTLINCi OWLS

761

ot steroids during the sensitive period tor brain
ditterentiation in the Zebra Finch. Journal of Endocri-
nology 103:363-369.

Kozlowski. C. P. and R. E. Ricklefs. 2010. Egg size and
yolk steroids vary across the laying order in cockatiel
clutches: a strategy for reinforcing brood size
hierarchies? General and Comparative Endocrinology
168:460-465.

Naguib. M., K. Riebel, A. Marzal, and D. Gil. 2004.
Nestling immunocompetence and testosterone covary
with brood size in a song bird. Proceedings of the
Royal Society of London, Series B 271:833-838.

Ottinger, M. a. and M. R. Bakst. 1981. Peripheral
androgen concentrations and testicular morphology in
embryonic and young male Japanese Quail. General
and Comparative Endocrinology 43:170-177.

OiTiNGER, M. A., S. Pitts, and M. A. Abdelnabi. 2001.
Steroid hormones during embryonic development in
Japanese Quail: plasma, gonadal, and adrenal levels.
Poultry Science 80:795-799.

Ros, a. F. H. 1999. Effects of testosterone on growth,
plumage pigmentation, and mortality in Black-headed
Gull chicks. Ibis 141:451-459.

Ros, A. F. H., S. J. Dieleman, and T. G. G. Groothuis.
2002. Social stimuli, testosterone, and aggression in
gull chicks: support for the challenge hypothesis.
Hormones and Behavior 41:334—342.

Runfeldt, S. and J. C. WtNGFiELD. 1985. Experimentally
prolonged sexual activity in female sparrows delays
termination of reproductive activity in their untreated
mates. Animal Behavior 33:403^1 1.

SAS Institute. 2008. JMP 7.0.2. SAS Institute Inc., Cary,
North Carolina, USA.

SasvAri, L., Z. Hegyi, and P. Peczely. 1999. Brood
reduction in White Storks mediated through asymme-
tries in plasma testosterone concentrations in chicks.
Ethology 105:569-582.

SasvAri, L., Z. Hegyi, and P. Peczely. 2006. Maternal
condition affects the offspring testosterone concentra-
tion in early nestling period: an experimental test on
Tawny Owls. Ethology 1 12:355-361.

SCHLINGER, B. A. and A. P. ARNOLD. 1992. Plasma sex
steroids and tissue aromatization in hatchling Zebra
Finch: implications for sexual differentiation of
singing behavior. Endocrinology 130:289-299.

Schumacher, M., J. Sulon, and J. Balthazart. 1988.
Changes in serum concentrations of steroids during

embryonic and post-hatching development of male and
female Japanese Quail, Cotuniix coturnix japonica.
Journal of Endocrinology 1 18:127-134.

Sharp, P. J., J. Culbert, ANt) J. W. Wells. 1977.
Variation in stored and plasma concentrations of
androgens during sexual development in the cockerel.
Journal of Endocrinology 74:467-476.

Silverin, B. AND P. J. Sharp. 1996. The development of
the hypothalamic-pituitary-gonadal axis in juvenile
Great Tits. General and Comparative Endocrinology
103:150-166.

Tanabe, Y., T. Yano, and T. Nakamura. 1983. Steroid
hormone synthesis by the testes, ovary, and adrenals of
embryonic and post-embryonic ducks. General and
Comparative Endocrinology 49:144-153.

Tanabe, Y., T. Nakamura, K. Fujioka, and O. Doi. 1979.
Production and secretion of sex steroid hormones by
the testes, the ovary, and the adrenal glands of
embryonic and young chickens (Callus domesticus).
General and Comparative Endocrinology 39:26-33.

Verboven, N., P. Monaghan, D. M. Evans, H. Schwabl,
N. Evans, and C. Whitelaw. 2003. Maternal
condition, yolk androgens and offspring performance:
a supplemental feeding experiment in the Lesser Black-
Backed Gull (Larus fiiscus). Proceedings of the Royal
Society of London, Series B 270:2223-2232.

Weichel, K., G. Schwager, P. Heid, H. R. GOttinger,
AND A. Pesch. 1986. Sex differences in plasma steroid
concentrations and singing behaviour during ontogeny
in canaries (Serinus canaria). Ethology 73:281-294.

Wingfield, J. C. 1984. Androgens and mating systems:
testosterone-induced polygyny in normally monoga-
mous birds. Auk 101:665-671.

Wingfield, J. C., R. E. Hegner, A. M. Dufty, and G. F.
Ball. 1990. The challenge hypothesis: theoretical
implications for patterns of testosterone secretion,
mating systems, and breeding strategies. American
Naturalist 136:829-846.

Woods, J. E. and R. C. Weeks. 1969. Ontogenesis of the
pituitary-gonadal axis in the chick embryo. General
and Comparative Endocrinology 13:242-254.

Young, C. E. and L. J. Rogers. 1978. Effects of steroidal
hormones on sexual, attack, and search behavior in the
i,solated male chick. Hormones and Behavior 10:107-
1 17.

Zar, j. H. 1999. Biostatistical analysis. Fourth Edition.
Prentice-Hall. Upper Saddle River, New Jersey. USA.

The Wilson Journal of Ornithology 122(4):762-766, 2010

STRESS RESPONSIVENESS DECREASES WITH AGE IN PRECOCIAL,

JUVENILE CHUKAR

MOLLY J. DICKENS’ - AND L. MICHAEL ROMERO'

ABSTRACT. — Development of the hypothalamic-pituitary-adrenal (HPA) axis and subsequent corticosterone (CORT)
release in newly hatched birds is a balance between limiting exposure to the detrimental effects of CORT on growth and
development, and the necessity of mounting an acute stress response. We measured the stress responsiveness of juvenile Chukar
(Alectoris chukar) 20 to 60 days post-hatch prior to molting into full adult plumage. The integrated CORT response during
60 min of restraint in these individuals decreased with age. Comparisons to mean adult integrated CORT values imply the
youngest juveniles have greater responses than adults while the oldest juveniles have reduced CORT responses. This pattern is
currently an anomaly within the altricial-precocial range and suggests specific tradeoffs, such as molting into adult plumage,
may affect timing of HPA suppression during development. Received 23 November 2009. Accepted 20 April 2010.

The acute stress response in adult animals is an
adaptive physiological reaction to noxious stimuli
and is hallmarked by release of glucocorticoids
following a hormone cascade along the hypotha-
lamic-pituitary-adrenal (HPA) axis (Sapolsky et
al. 2000). Corticosterone (CORT) is the predom-
inant glucocorticoid secreted in avian species and,
after CORT is released into the system, negative
feedback mechanisms act to quickly shut down
CORT secretion (Dallman and Bhatnagar 2001).
Excessive CORT release, either due to a disrupted
HPA axis or altered negative feedback sensitivity,
can lead to physiological and behavioral pathol-
ogies (McEwen 2005). Glucocorticoids, in terms
of development, have severe effects on growth as
they promote muscle breakdown; elevated con-
centrations of glucocorticoids lead to muscle
wasting and decreased growth rates (Wingfield
1994, Morici et al. 1997, Sapolsky et al. 2000). In
addition, glucocorticoid concentrations can affect
key aspects of neuronal development in the central
nervous system (Sapolsky and Meaney 1986).
Elevated CORT can also affect feather develop-
ment (Romero et al. 2005, DesRochers et al. 2009).

Development of the HPA axis and the stress
response system in young birds post-hatch is
considered to be a delicate balance between
benefiting from the ability to mount an acute
stress response and limiting the damaging effects
of glucocorticoids. This balance requires tight
regulation due to the impact of glucocorticoids on
growth and feather development (Walker et al.
2005, Wada et al. 2007). Altricial laboratory
rodents display a clear and easily defined “stress
hyporesponsive period” or SHRP, beginning

' Department of Biology, Tufts University, Medford, MA
021.55, U.SA.

^Corresponding author; e-mail: molly.dickens@tufts.edu

2 days postnatal (reviewed by Sapolsky and
Meaney 1986). A hyporesponsive period has also
been noted in many altricial and semi-altricial
avian species including Magellanic Penguins
{Spheniscus magellanicus) (Walker et al. 2005),
White-crowned Sparrows {Zonotrichia leucophrys
nuttalli) (Wada et al. 2007), Northern Mocking-
birds {Mimus polyglottos) (Sims and Holberton
2000), American Kestrels (Falco sparverius)
(Love et al. 2003), Slender-billed Prions {Pachyp-
tila belcheri) (Quillfeldt et al. 2009), and Atlantic
Canaries (Serinus canaria) (Schwabl 1999).
Hatchlings were shown to be fully capable of
mounting an adult-like response within days of
hatching by precocial species including Mallards
{Anas platyrhynchos) (Holmes et al. 1990),
chickens {Gallus gallus domesticus) (Freeman
1982), and Domestic Turkeys (Meleagris gallo-
pavo) (Wentworth and Hussein 1985), and in
semi-precocial Great-winged Petrels (Pterodroma
macroptera) (Adams et al. 2008) and the semi-
altricial Snowy Owl {Bubo scandiacus) (Romero
et al. 2006),

The HPA response, in most studies of altricial
and semi-altricial avian species, slowly increases
with age and fledglings about to leave the nest are
close to adult-like responses (Schwabl 1999, Love
et al. 2003, Walker et al. 2005, Bias et al. 2006,
Wada et al. 2007). Chickens, a precocial species,
show a clear but short hyporesponsive period
immediately following hatching (Freeman 1982).
However, there are limited data on development
of the stress response in precocial species that
match studies tracking development of the stress-
respon.se system in altricial species.

Our objective was to measure the stress
responsiveness of juvenile Chukar {Alectoris
chukar), within an age range in which they were

762

Dickens and Romero • STRESS RES[T)NSI VENESS OF WIED CIIUKAR

763

post-hatch but had yet to develop full adult
plumage, to track the pattern of their HPA
development.

METHODS

Study Species. — Chukar are non-migratory de-
sert-adapted ground birds. They were introduced
into North America from the steep, arid mountains
of southwestern Eurasia as game birds, and can
now be found extensively throughout the arid
mountains of the American west (Christensen
1996). We captured wild juvenile and adult
Chukar from a remote region of the Mojave
Desert within the U.S. Navy China Lake Naval
Air Weapons Station (CLNAWS) near Ridge-
crest, California (117°37'W, 36° 3' N). Limited
human access into CLNAWS assured minimal,
non-experimental human interaction with our
population. Juvenile Chukar in desert environ-
ments can be relatively easy to capture due to
limited dispersal during summer months and
faithfulness to a specific water source (Larsen
et al. 2007).

Chukar are precocial and leave the nest with
their mother immediately after hatching, and are
fully capable of flight within 2 weeks of hatching
(Christensen 1996). We trapped Chukar at human-
made rainwater catchments known as “guzzlers”
and one spring-fed pool from mid-July to August
in 2006 and 2008. The guzzler system and
trapping technique at CLNAWS have been
described (Delehanty et al. 2004).

Field Procedures. — We weighed Chukar im-
mediately upon capture and u,sed the absence of a
full black collar and eye band coloration as an
indication of their juvenile status (Christensen
1996). We captured 31 juvenile Chukar with body
mass ranging from 115 to 390 g. Juveniles
representing the range of masses were captured
evenly across the sampling period. These masses,
according to the growth chart in Pis (2003),
indicate juveniles were between 20 and 60 days of
age. Adults in this region are, on average, 500 g
(based on raw data collected for Dickens et al.
2009a, b; Pig. I). Juveniles typically reach adult
size at ~12 weeks of age (Christensen 1996). Our
study was limited by method of capture and time
period in which it was conducted. We did not
obtain samples from chicks immediately post-
hatch as we were not following nests, and trapping
began within a period in which most chicks were
at least a few weeks of age.

Capture-Stress Protocol. — Birds captured were
provided with shade, ad libitum water, and were
removed from the trap within 3 hrs of entering.
Time spent in the trap was not ideal and we
previously justified this experimental methodology
(Dickens et al. 2009a). All individuals for which we
were able to obtain baseline samples (samples taken
within 3 min of approaching the trap, n = 27)
demonstrated baseline samples well below their
respective peak restraint-induced CORT concentra-
tions, and all baseline concentrations were within
the range expected for this species.

We sampled CORT responsiveness of all
individuals by placing them into an opaque bag
and taking blood samples at 15, 30, and 60 min to
measure the capacity for an endocrine stress
response. All animal use complied with AALAC
guidelines and was approved by the Tufts
Institutional Animal Care and Use Committee.
CORT was measured using radioimmunoassay
(RIA) (Wingfield et al. 1992) as previously used
for this species (Dickens et al. 2009a, b).

Statistical Analysis. — We calculated the relative
total CORT release over the 60-min restraint
period as an integrated CORT value using
measurements at 15, 30, and 60 min (Romero
2004). This measurement allowed us to compare
the overall capacity of the HPA axis to release
CORT over the 60-min restraint period. We
performed a regression comparing each individu-
al’s integrated CORT versus their mass (Graph-
Pad, Prizm 4b). We also analyzed a group of
adults sampled upon first capture from the wild to
compare adult mass and the adult stress respon-
siveness. We present these results as means ± SE
(subset of data reanalyzed from that in Dickens
et al. 2009a, b).

RESULTS

We found a correlation between body mass and
integrated CORT concentrations that was signifi-
cantly different from zero (Eig. I; r = 0.32, P <
0.001) for the 31 juveniles captured and sampled.
Adult Chukar (/? = 19), captured during the same
years as the juveniles, had a mean integrated CORT
response of 2,188.0 ± 142.5 ng/mL/min and a
mean body mass of 453.3 ± 12.2 g. There was no
con-elation between adult mass and integrated
CORT concentrations (r = 0.002, P = 0.904).

DISCUSSION

CORT response decreased with age in the
precocial Chukar for the age range investigated

764

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 4, December 2010

FIG. 1. Individual points (n = 31) represent a single juvenile’s integrated CORT measurement versus body mass. The
regression line is significantly different from zero (r~ = 0.32, P < 0.001). The open circle represents the adult mean values
± SE for each parameter {n = 19).

and the heaviest, oldest, birds had the lowest
response. The average adult response was in the
middle of the juvenile response range. The
youngest birds had relatively higher stress re-
sponses than adults while the oldest juveniles,
closest to full maturity, had a relatively lower
response than that of adults.

This pattern contrasts to that observed in
altricial and semi-altricial species for which the
HPA response becomes more similar to adult-like
responses as the nestlings reach Hedging (e.g..
Love et al. 2003, Walker et al. 2005, Wada et al.
2007). The slow development of the HPA
respon.se in these altricial species fits well with
the developmental hypothesis summarized by
Bias et al. (2006). Altricial species are either
completely dependent or substantially dependent
on parents for all aspects of their lives (Ricklefs
1979). They are also completely constrained to
the nest and their capacity to e.scape a stressor,
such as a predator, is extremely limited. Having
an HPA response to an ine.scapable stimulus
would lead to elevated CORT concentrations and

do more harm to their growth and development
than good for their recovery from the stressor
(Bias et al. 2006).

Precocial young, in contrast, are out of the nest
and exposed to the same stressors as adults.
Precocial juveniles of a variety of species are as
fully capable of a quick escape from a predator as
are adults (Herrel and Gibb 2004) and, for
Chukar, all major requirements for quick flight
e.scape occur within 8 days post-hatch (Jackson et
al. 2009). Juveniles can physically escape as well
as adults, and it also seems likely they initiate an
adult-like CORT respon.se. Obtaining an adult-
like response early in life is the pattern ob.served
in the limited studies that investigate precocial
young. There is a hyporesponsive period -in
chickens immediately following hatch but chicks
are able to mount a full CORT respon.se within the
first day (Freeman 1982). Full responsiveness by
the semi-precocial Great-winged Petrel was mea-
sured after hatch and chicks remained responsive
throughout the juvenile period (Adams et al.
2008). However, Mallard duckling respon.ses.

Dickens am! Romero • STRESS RESPONSIVENESS OE WILD CIIUKAR

765

immedialely alter hatch, were more robust and
actually decreased to match the adult response by
day 21 (Holmes et al. 1990). We also observed a
more robust response by younger than older
Juvenile Chukar. We were not able to sample
individuals earlier than 20 days after hatch and do
not know whether Chukar hatchlings show some
hyporesponsive period immediately after hatch
similar to altricial species. We also cannot discuss
the potential HPA response profile of the birds
during this earlier developmental stage. However,
the striking difference that our data confirm is that
precocial young can have an adult or an even
greater CORT response within a few weeks of
hatching.

The capacity of the juvenile Chukar response
appears to decrease beyond the adult response as
they become older. These data imply a hidden
tradeoff in elevating CORT concentrations during
the late part of the Juvenile stage. Preventing
excessive CORT exposure outweighs the benefits
of having the capacity for an adult CORT
response. The tradeoff may reflect continued
growth during this time, although the rate of
growth is less than in younger Juveniles (Pis
2003). Alternatively, successful molting of Juve-
nile feathers to grow adult feathers during this
time may be sufficiently important to trump full
HPA responsiveness. Chukar begin prebasic molt
at 5 weeks of age, and the molt is nearly complete
by 16 weeks (Christensen 1996). CORT concen-
trations, both baseline and stress-induced concen-
trations, are low during seasonal molt in adults
(Romero and Remage-Healey 2000). Elevated
CORT concentrations can lead to decreased
protein deposition in feathers (Romero et al.
2005) and, perhaps as a result, reduced feather
quality (DesRochers et al. 2009). Juveniles are
~250 g in mass at 5 weeks of age (the time of
Prebasic molt according to Christensen 1996), if
we extrapolate age from mass in the graph
provided by Pis (2003), this age relates to the
point at which the line (Fig. I) decreases below
the adult average. Older Juveniles are growing
adult flight feathers and replacing other body
feathers to form the black band and collar typical
of this species. Elevated CORT concentrations in
Barn Owl {Tyto alba) chicks can lead to decreased
deposition of pigment in growing feathers (Roulin
et al. 2008). Maintaining low CORT during this
period of feather replacement may be important
for not only building strong adult flight feathers
but also for acquiring appropriate adult coloration.

More research is necessary to ascertain whether
this pattern of HPA responsiveness is unique to
Chukar, present in precocial species in general, or
present only in precocial species that have distinct
adult coloration patterns. This pattern is currently
an anomaly in the growing literature investigating
how the HPA develops within altricial-precocial
species and may help understand the specific
tradeoffs that affect when birds should suppress
HPA responsiveness or express full adult-like
responses.

ACKNOWLEDGMENTS

We thank the China Lake Naval Air Weapons Station for
the opportunity to conduct these studies, T. G. Campbell
for his support on the base, and D. J. Delehanty, K. A.
Earle, Christine Lattin, J. L. Brubaker, and Joshua
Leisenring for help in the field. Funding was provided by
NSF Grant # IOB-0542099 to LMR and Sigma Xi Grants-
in-Aid of Research to MJD.

LITERATURE CITED

Adams, N. J., J. F. Cockrem, E. J. Candy, and G. A.
Taylor. 2008. Non-precocial Grey-faced Petrel
chicks (Pterodroma macroptera gouldi) show no
age-related variation in corticosterone responses to
capture and handling. General and Comparative
Endocrinology 157:86-90.

Blas, j., R. Baos, G. R. Bortolotti, T. A. Marchant,
AND F. Hiraldo. 2006. Age-related variation in the
adrenocortical response to stress in nestling White
Storks {Ciconia eiconia) supports the developmental
hypothesis. General and Comparative Endocrinology
148:172-180.

Christensen, G. C. 1996. Chukar (Alectoris chukar). The
birds of North America. Number 258.

Dallman, M. F. AND S. Bhatnagar. 2001. Chronic stress
and energy balance: role of the hypothalamo-pitui-
tary-adrenal axis. Pages 179-210 in Handbook of
physiology; Section 7: the endocrine system;
Volume IV: coping with the environment: neural
and endocrine mechanisms (B. S. McEwen and H. M.
Goodman, Editors). Oxford University Press. New
York, USA.

Delehanty, D. J., S. S. Eaton, and T. G. Campbell.
2004. From the field: Mountain Quail fidelity to
guzzlers in the Mojave Desert. Wildlife Society
Bulletin 32:588-593.

Desrochers. D. W., j. M. Reed. J. Awerman. J. A.
Kluge, J. Wilkinson, L. 1. van Griethuijsen. J.
Aman, and L. M. Romero. 2009. Exogenous and
endogenous corticosterone alter feather quality. Com-
parative Biochemistry and Physiology — Part A: Mo-
lecular & Integrative Physiology 152:46-52.

Dickens, M. J., D. .1. Delehanty, and L. M. Romero.
2009a. Stress and translocation: alterations in the stress
physiology of translocated birds. Proceedings of the
Royal Society. Series B 276:2051-2056.

766

THE WILSON JOURNAL OF ORNITHOLOGY • VoL 122, No. 4, December 2010

Dickens, M. J., K. A. Earle, and L. M. Romero. 2009b.
Initial transference of wild birds to captivity alters
stress physiology. General and Comparative Endocri-
nology 160:76-83.

Freeman, B. M. 1982. Stress non-responsiveness in the
newly-hatched fowl. Comparative Biochemistry and
Physiology — Part A: Physiology 72:251-253.

Herrel, a. and a. C. Gibb. 2004. Ontogeny of perfomiance
in vertebrates. Pages 1-6 in Symposium on the ontogeny
of performance in vertebrates held at the 7th Interna-
tional Congress of Vertebrate Morphology. University
of Chicago Press, Boca Raton, Florida, USA.

Holmes, W. N., J. Cronshaw, and K. E. Rohde. 1990.
Stress-induced adrenal steroidogenic responsiveness of
the Mallard duck {Anas platyrhynchos) during early
post-natal development. Comparative Biochemistry
and Physiology — Part A: Physiology 92:403^08.

Jackson, B. E., P. Segre, and K. P. Dial. 2009. Precocial
development of locomotor performance in a ground-
dwelling bird (Alectoris chukar): negotiating a three-
dimensional terrestrial environment. Proceedings of
the Royal Society, Series B 276:3457-3466.

Larsen, R. T., J. T. Flinders, D. L. Mitchell, E. R.
Perkins, and D. G. Whiting. 2007. Chukar watering
patterns and water site selection. Rangeland Ecology
and Management 60:559-565.

Love, O. P., D. M. Bird, and L. J. Shutt. 2003. Plasma
corticosterone in American Kestrel siblings: Effects of
age, hatching order, and hatching asynchrony. Hor-
mones and Behavior 43:480-488.

McEwen, B. S. 2005. Stressed or stressed out: what is the
difference? Journal of Psychiatry & Neuroscience
30:315-318.

Morici, L. a., R. M. Elsey, and V. A. Lance. 1997.
Effects of long-term corticosterone implants on growth
and immune function in juvenile alligators. Alligator
mississippiensis. Journal of Experimental Zoology
279:156-162.

Pis, T. 2003. Energy metabolism and thermoregulation in
hand-reared Chukars (Alectoris chukar). Comparative
Biochemistry and Physiology — Part A: Molecular &
Integrative Physiology 136:757-770.

Quillfeldt, P., M. Poisbleau, O. Chastel, and j. E.
Masello. 2009. Acute stress hyporesponsive period
in nestling Thin-billed Prions Pachyptila helcheri.
Journal of Comparative Physiology A: Neuroethol-
ogy, Sensory, Neural, and Behavioral Physiology
195:91-98.

Rk’KLEFS, R. E. 1979. Adaptation, constraint, and compro-
mi.se in avian postnatal development. Biological
Reviews of the Cambridge Philosophical Society
54:269-290.

Romero, L. M. 2004. Physiological stress in ecology:
lessons from biomedical re.search. Trends in Ecology
and Evolution 19:249-255.

Romero. L. M.. D. W. Holt, M. Maple.s, and J. C.
Wingfield. 2006. Corticosterone is not correlated
with nest departure in Snowy Owl chicks (Nyctea

scandiaca). General and Comparative Endocrinology
149:119-123.

Romero, L. M. and L. Remage-Healey. 2000. Daily and
seasonal variation in response to stress in captive
starlings (Sturnus vulgaris): corticosterone. General
and Comparative Endocrinology 1 19:52-59.

Romero, L. M., D. Strochlic, and J. C. Wingfield. 2005.
Corticosterone inhibits feather growth: potential
mechanism explaining seasonal down regulation of
corticosterone during molt. Comparative Biochemistry
and Physiology — Part A: Molecular & Integrative
Physiology 142:65-73.

Roulin, a., B. Almasi, a. Rossi-Pedruzzi, a. L.
Ducrest, K. Wakamatsu, I. Miksik, j. D. Blount,
S. Jenni-Eiermann, and L. Jenni. 2008. Corticoste-
rone mediates the condition-dependent component of
melanin-based coloration. Animal Behaviour 75:
1351-1358.

Sapolsky, R. M. and M. j. Meaney. 1986. Maturation of
the adrenocortical stress response: neuroendocrine
control mechanisms and the stress hyporesponsive
period. Brain Research Reviews 1 1:65-76.

Sapolsky, R. M., L. M. Romero, and A. U. Munck. 2000.
How do glucocorticoids influence stress responses?
Integrating permissive, suppressive, stimulatory, and
preparative actions. Endocrine Reviews 21:55-89.

Schwabl, H. 1999. Developmental changes and among-
sibling variation of corticosterone levels in an altricial
avian species. General and Comparative Endocrinol-
ogy 116:403^08.

Sims, C. G. and R. L. Holberton. 2000. Development of
the corticosterone stress response in young Northern
Mockingbirds (Mimus polyglottos). General and Com-
parative Endocrinology 1 19:193-201.

Wada, H., T. P. Hahn, and C. W. Breuner. 2007.
Development of stress reactivity in White-crowned
Sparrow nestlings: total corticosterone response in-
creases with age, while free corticosterone response
remains low. General and Comparative Endocrinology
150:405-413.

Walker, B. G., J. C. Wingfield, and P. D. Boersma.
2005. Age and food deprivation affects expression of
the glucocorticosteroid stress response in Magellanic
Penguin (Spheniscus magellanicus) chicks. Physiolog-
ical and Biochemical Zoology 78:78-89.

Wentworth, B. C. and M. O. Hussein. 1985. Serum
corticosterone levels in embryos, newly hatched, and
young turkey poults. Poultry Science 64:2195-2201.

Wingfield, J. C. 1994. Modulation of the adrenocortical
response to stress in birds. Pages 520-528 in Perspec-
tives in comparative endocrinology (K. G. Davey, R. E.
Peter, and S. S. Tobe, Editors). National Research
Council of Canada, Ottawa, Ontario, Canada.

Wingfield, J. C., C. M. Vleck, and M. C. Moore. 1992.
Seasonal changes of the adrenocortical response to
stress in birds of the Sonoran Desert. Journal of
Experimental Biology 264:419-^28.

Short Communications

The Wilson Journal of Ornithology 122(4):767-77l, 2010

Composition Dynamics of an Avian Assemblage

Fritz L. Knopf

ABSTRACT. — I summarize the 10-year composition
dynamics of a riparian avifauna on the Arapaho
National Wildlife Refuge, Colorado, USA. Only 15 of
the 89 breeding bird species (19,393 individuals) were
present each of the 10 years whereas 25 species were
only recorded in a single year. Number of species in any
given year ranged from 30 to 57 (34 to 64% of the 10-
year cumulative species total). The number of species
detected >1 and <10 years represented over half of all
species with no pattern of either increasing or
decreasing across years of study. Short-term definitions
of species richness preclude the ability to define the
major dynamic within an avian assemblage — species
that use the site an intermediate number of years.
Understanding the ecological prerequisites for these
species should provide greater insight into current
physical habitat and climate changes occurring on-site.
Received 16 February 2010. Accepted 24 June 2010.

Ornithologists have historically inventoried a
local avifauna to define the biological significance
of an area: the more species present, the presumed
greater ecological diversity and conservation
significance of the locale. The number of species
present is often referred to as species richness, or
alpha diversity when viewed from broader spatial
scales (Knopf and Samson 1994). Ecological
studies have used species richness to interpret
avian responses to changes in landscapes resulting
from either natural processes or management
practices (Conroy and Noon 1996, Brand et al.
2008, Johnson et al. 2009).

Annual variation in species richness and
relative species composition define conservation
significance of an area (Fleishman et al. 2006).
However, among-year variation in species com-
position is generally ignored and the number of
species recorded across the 2-4 years of most
studies is often presented as a single number to
define conservation significance. Species immi-
gration and emigration are assumed to be equal in

‘ U.S. Fish and Wildlife Service, National Ecology
Research Center, 1300 Blue Spruce Drive, Fort Collins,
CO 80521, USA.

^Current address: 713 Boulder Circle, Fort Collins, CO
80524, USA; e-mail: FLKnopf@Yahoo.com

such studies, although data addressing this
assumption are notably lacking.

I noted that species composition within an
avian assemblage varied annually while species
richness remained comparatively stable during a
10-year study of the breeding-bird assemblage in
a shrub willow {Salix spp.) association along the
Illinois River in the Arapaho National Wildlife
Refuge, 1980-1989. The patterning of composi-
tion change was much more dynamic than the
standard assumption that individuals of some
species colonize an area as individuals represent-
ing other species leave. I briefly summarize the
10-year composition dynamics of avifaunal
change with the intent to discourage use of
species richness derived from short-term studies
in assessments of conservation significance or
efficacy of management actions at local sites.

METHODS

My coworkers and I studied the breeding bird
assemblage along the Illinois River floodplain
within the Arapaho National Wildlife Refuge
(ANWR), Jackson County, Colorado, USA, from
1980 through 1989. The refuge is at an elevation
of 2,485-2,516 m, within an intermountain glacial
basin of upland vegetation dominated by sage-
brush (Artemisia spp.). The riparian shrub assem-
blage comprises species of willows (Salix bebbi-
ana, S. caudata, S. exigua, S. geyeriana, S.
monticola, S. planifolia, S. pseudocordata, and
5. wolfii) that grow to heights of 6 m (Cannon and
Knopf 1984). Photographs of the study area are in
Knopf et al. (1988: 281) and Sedgwick and Knopf
(1992:724).

We inventoried the breeding bird assemblage
annually in the latter half of June, well after spring
migrants had left ANWR. Each year, two
observers each surveyed bird species and individ-
uals at 200 (185 in 1980) points spaced at 100-m
intervals (to reduce visual overlap between survey
points) along each side and at random distances
perpendicular to the river bank. The survey was
conducted from the stream bank where the willow
stand was less than 20 m wide. The radius of the

767

768

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

TABLE 1. Number of bird species recorded on 400,
10-min survey plots along the Illinois River floodplain on
the Arapaho National Wildlife Refuge, Colorado, 1980-
1989.

Year

No. species

No. individuals

1980“

30

1,079

1981

35

1,231

1982

50

1,867

1983

45

1,841

1984

57

2,309

1985

45

1,866

1986

45

1,955

1987

43

2,154

1988

42

2,225

1989

57

2,866

1980-1989

89

19,393

‘‘ Sampling effort was 370 plots in 1980.

survey circle was not restricted to allow detection
of larger, wary species.

Avian surveys were conducted from one-half
hour before sunrise until 1000 hrs MDT each
morning except in periods of inclement weather.
An observer moved to each survey point, waited
1 min and then recorded the species of each bird
detected within the following 10-min interval.
Each observer participated in multiple years of the
surveys.

RESULTS

We detected 1,079-2,866 individuals represent-
ing 30-57 avian species during surveys of the

Illinois River floodplain in any given year, 1980
through 1989 (Table 1). Cumulative totals of
19,393 individuals representing 89 species were
recorded across the 10-year period. The number of
species recorded in any specific year ranged from
34 to 64% of the collective number of species
recorded across the 10-year period.

Detection of a given species in the floodplain
also varied among years (Table 2). Only 15 (17%)
of the cumulative 89 species were seen in all
10 years. Many more, 25 (28%), species were
seen in only one of the 10 years. The number of
detections within a species, ranged from >4,000
Yellow Warblers {Dendroica petechia) (Buckland
et al. 1993) to a single record for many species
that were only recorded in 1 year.

Over half (49 of 89, 55%) of the species
recorded on ANWR were present > 1 but
<10 years. The number of species detected in
an intermediate number of years showed neither a
clear pattern of increasing nor decreasing with
increasing years of study.

DISCUSSION

Ten of the 89 species present within the
floodplain were waterfowl reflecting the manage-
ment objective of ANWR to enhance breeding
habitats for waterfowl by promoting taller herba-
ceous vegetation to encourage nesting. Six raptor
species foraged within the floodplain although
most nested off the study area. Most cavity-
nesting species did not have on-site nest sub-
strates, but did forage as a part of the breeding

TABLE 2. Number of years a species was recorded on 400. 10-min survey plots along the Illinois River floodplain on
the Arapaho National Wildlife Refuge, Colorado. 1980-1989.

No. years-

No. species

Specie.s'’

10

15

AMRO, BBMA, BCCH, BHCO, BRBL, GWTE, LISP, RWBL, SASP, SOSP,
WeSP, WEME, WIFE, WISN, YEWA

9

7

AMGO, GADW, HOWR, KILL, MALL, YHBL, WIPH

8

3

AMWI, BCNH, MODO

7

6

CITE, COGR, CONI, NOEL, SPSA, SWHA

6

8

BWTE, AMCR, COYE, FOSP, LESC, NOSH. TRSW, VEER

5

4

BASW, BRSP, NOPI, SORA

4

6

AMCO, GBHE, GTTO, NOHA, SEOW. WIWA

3

9

AMKE, BEKI, DUEL, GHOW, PISI, REDH, SWTH, VESP, WAVI

2

6

CAGO, CLSW, GRCA, MAWR. RCKI, GRSG

1

25

AMBI, BTHU, BUOR, CANV, CHSP, COFL, CORA, DOWO, EAKI,
EUST, FOTE, GRHE, GRYE, HOLA, INBU, LABU, LEOW, OCWA,
RNPH, RTHA, SATH, SSHA, VIRA, WEKl, YBCU

Sampling etTurl was 370 plots in 1980.
Species codes in the Appendix.

SHORT COMMUNICATIONS

769

bird assemblage, as did Green-tailed Towhee
{Pipilo chloruriis) and Sage Thrasher (Oreos-
coptes momanits) that nested in the sagebrush
uplands. Greater Sage-Grouse [Centrocerciis uro-
phasianiis) were not known to nest within the
floodplain, but occasionally moved broods into
the riparian vegetation. Collectively, these species
contributed a meaningful component to the
species richness and conservation significance of
the site, although the floodplain only provided one
component of the habitat required for breeding.

Earlier studies on ANWR included both basic
studies of single species, i.e., Yellow Warbler
(Knopf and Sedgwick 1992, Howe and Knopf
2000) and Willow Flycatcher (Enipidonax traillii)
(Sedgwick and Knopf 1988, 1989, 1992), and
applied studies of grazing-response guilds (Knopf
et al. 1988, Stanley and Knopf 2002). Those
studies focused on species that occurred at
densities adequate to insure sample sizes to
support statistically rigorous analyses — species
that were present all 10 years. Their certain
presence revealed these species were more
tolerant of ecologic or climatic change occumng
locally within the study area across the decade.
Extending avian surveys on ANWR another
10 years would result in no increase in the
number of species with >95% probability of
appearance each year.

Twenty-five species were present only one of
10 years. These species were likely non-breeders
as spring migrants had left the refuge before the
late June surveys. Some individuals may have
been post-breeding dispersers from other locales,
whereas others were stragglers as they were well
outside documented breeding ranges. Extending
avian surveys on the ANWR another 10 years
would increase the cumulative list of avian
species for the site to 100 species or more across
a 20-year period. This latter component of species
richness is not ‘rich’ at all, but peripheral to any
meaningful definition of the breeding bird assem-
blage or management actions on the ground.

Interpreting or monitoring the significance of
an area for avian conservation using a list of
species recorded during a short-term study is
problematic in that one component of the
assemblage (species present every year) is eco-
logically tolerant of subtle environmental changes
and another component is only present inciden-
tally. Declines or increases in numbers of both
common and rare/accidental species would only
be expected following relatively catastrophic

changes in the landscape. The larger problem,
however, is that short-term studies are unable to
distinguish “incidental” species from those that
periodically use the site during the breeding
season. Research investigations focusing on the
ecological prerequisites for .species that use an
area intermittently should provide greater insight
into more subtle physical and climate changes that
drive species habitats within a local avian
assemblage. The ability to identify these species
already exists where standardized protocols to
monitor breeding birds have been in place for
multiple years.

ACKNOWLEDGMENTS

1 am especially grateful to R. W. Cannon, J. F. Ellis, W.
H. Howe, T. E. Olson, and J. A. Sedgwick for conducting
avian surveys. E. C. Patten. Project Leader/Manager,
graciously provided open access to Arapaho National
Wildlife Refuge. This work was completed when I was a
research wildlife biologist with the U. S. Fish and Wildlife
Service, and I am grateful for the intellectual support of my
administrators and academic peers at the National Ecology
Research Center during the 1980s.

LITERATURE CITED

Brand, L. A., G. C. White, and B. R. Noon. 2008. Eactors
influencing species richness and community composi-
tion of breeding birds in a desert riparian corridor.
Condor 1 10:199-210.

Buckland, S. T., D. R. Anderson, K. P. Burnham, and
J. L. Laake. 1993. Distance sampling. Chapman and
Hall, London, United Kingdom.

Cannon, R. W. and F. L. Knopf. 1984. Species
composition of a willow community relative to
seasonal grazing histories in Colorado. Southwestern
Naturalist 29:234-237.

Conroy, M. J. and B, R. Noon. 1996. Mapping of species
richness for conservation of biological diversity:
conceptual and methodological issues. Ecological
Applications 6:763-773.

Fleishman, E., R. F. Noss, and B. R. Noon. 2006. Utility
and limitations of species richness metrics for conser-
vation planning. Ecological Indicators 6:54.3-553.
Howe, W. H. and F. L. Knopf. 2000. The role of
vegetation in cowbird parasitism of Yellow Warblers.
Pages 200-203 in Ecology and management of
cowbirds and their hosts (J. N. M, Smith. T. L. Cook.
S. 1, Rothstein. S. K. Robinson, and S. G. Sealy,
Editors). University of Texas Press. Austin. USA.
Johnson, T. N., R. D. Applegate. D. E. Hoover. P. S.
Gipson, and B, K. Sandercock. 2009. Evaluating
avian community dynamics in restored riparian
habitats with mark-recapture models. Wilson Journal
of Ornithology 121:22-40.

Knopf, F. L. and F. B. Samson. 1994. Conservation of

770

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4, December 2010

avian diversity in riparian corridors. Conservation
Biology 8:669-676.

Knopf, F. L. and J. A. Sedgwick. 1992. An experimental
study of nest site selection by Yellow Warblers.
Condor 94:734-742.

Knopf, F. L., J. A. Sedgwick, and R. W. Cannon. 1988.
Guild structure of a riparian avifauna relative to
seasonal cattle grazing. Journal of Wildlife Manage-
ment 52:280-290.

Sedgwick, J. A. and F. L. Knopf. 1988. A high incidence

of Brown-headed Cowbird parasitism of Willow
Flycatchers in Colorado. Condor 90:253-256.

Sedgwick, J. A. and F. L. Knopf. 1989. Region-wide
polygyny in Willow Flycatchers. Condor 91:473^75.

Sedgwick, J. A. and F. L. Knopf. 1992. Describing
Willow Flycatcher habitats: scale and gender perspec-
tives. Condor 94:720-733.

Stanley, T. R. and F. L. Knopf. 2002. Vegetation and
avian responses to grazing in a shrub-willow flood-
plain. Conservation Biology 16:225-231.

APPENDIX. Alphabetical list of species codes (Table 2).

Code

Species

AMBI

American Bittern (Botaurus lentiginosus)

AMCO

American Coot (Fulica americana)

AMCR

American Crow {Corvus brachyrhynchos)

MAKE

American Kestrel {Falco sparverius)

AMGO

American Goldfinch {Spinus tristis)

AMRO

American Robin (Turdus migratorius)

AMWl

American Wigeon (Anas americana)

BASW

Bam Swallow (Hirundo rustica)

BBMA

Black-billed Magpie (Pica hudsonia)

BCCH

Black-capped Chickadee (Poecile atricapillus)

BCNH

Black-crowned Night-Heron (Nycticorax nycticorax)

BEKl

Belted Kingfisher (Megaceryle alcyon)

BHCO

Brown-headed Cowbird (Molothrus ater)

BRBL

Brewer’s Blackbird (Euphagus cyanocephalus)

BRSP

Brewer’s Sparrow (Spizella breweri)

BTHU

Broad-tailed Hummingbird (Selasphorus platycercus)

BUOR

Bullock’s Oriole (Icterus bullockii)

BWTE

Blue-winged Teal (Anas discors)

CAGO

Canada Goose (Branta canadensis)

CANV

Canvasback (Aythya valisineria)

CHSP

Chipping Sparrow (Spizella passerina)

CITE

Cinnamon Teal (Anas cyanoptera)

CLSW

Cliff Swallow (Petrochelidon pyrrhonota)

COFL

Cordilleran Flycatcher (Empidonax occidentalis)

COGR

Common Crackle (Quiscalus quiscula)

CONI

Common Nighthawk (Chordeiles minor)

CORA

Common Raven (Corvus corax)

COYE

Common Yellowthroat (Geothlypis trichas)

DOWO

Downy Woodpecker (Picoides pubescens)

DUFL

Dusky Flycatcher (Empidonax oberholseri)

EAKl

Eastern Kingbird (Tyrannus tyrannus)

EUST

European Starling (Sturnus vulgaris)

FOSP

Fox Sparrow (Passerella iliaca)

FOTE

Forster’s Tern (Sterna forsteri)

GADW

Gadwall (Anas strepera)

GBHE

Great Blue Heron (Ardea herodias)

GHOW

Great Homed Owl (Bubo virginianus)

GRCA

Gray Catbird (Dumetella carolinensis)

GRHE

Green Heron (Butorides virescens)

GRYE

Greater Yellowlegs (Tringa melanoleuca)

GTTO

Green-tailed Towhee (Pipilo chlorurus)

GWTE

Green-winged Teal (Anas crecca)

HOLA

Horned Lark (Eremophila alpestris)

HOWR

House Wren (Troglodytes aedon)

SHORT COMMUNICATIONS

771

APPENDIX. Continued.

Code

Species

INBU

Indigo Bunting {Passerimi cyanea)

KILL

Killdeer (Charadrius vocifents)

LABU

Lazuli Bunting (Passerimi amoena)

LEOW

Long-eared Owl (Asia otus)

LESC

Lesser Scaup (Aythya ajfinis)

LISP

Lincoln’s Sparrow (Melospiza lincolnii)

MALL

Mallard (Anas platyrhynchos)

MAWA

MacGillivray’s Warbler (Oporornis tolmiei)

MODO

Mourning Dove (Zenaida macroura)

NOEL

Northern Flicker (Colaptes auratus)

NOHA

Northern Harrier (Circus cyaneus)

NOPI

Northern Pintail (Anas acuta)

NOSH

Northern Shoveler (A. clypeata)

OCWA

Orange-crowned Warbler (Vermivora celata)

PISI

Pine Siskin (Spinus pinus)

RCKI

Ruby-crowned Kinglet (Regulus calendula)

REDH

Redhead (Aythya americana)

RNPH

Red-necked Phalarope (Phalaropus lobatus)

RTHA

Red-tailed Hawk (Buteo jamaicensis)

RWBL

Red-winged Blackbird (Agelaius phoeniceus)

GRSG

Greater Sage-Grouse (Centrocercus urophasianus)

SASP

Savannah Sparrow (Passerculus sandwichensis)

SATH

Sage Thrasher (Oreoscoptes montanus)

SEOW

Short-eared Owl (Asia flammeus)

SORA

Sora (Porzana Carolina)

SOSP

Song Sparrow (Melospiza melodia)

SPSA

Spotted Sandpiper (Actitis macularius)

SSHA

Sharp-shinned Hawk (Accipiter striatus)

SWHA

Swainson’s Hawk (Buteo swainsoni)

SWTH

Swainson’s Thrush (Catharus ustulatus)

TRSW

Tree Swallow (Tachycineta bicolor)

VEER

Veery (Catharus fuscescens)

VESP

Vesper Sparrow (Pooecetes gramineus)

VIRA

Virginia Rail (Rallus limicola)

WAVI

Warbling Vireo (Vireo gilvus)

WCSP

White-crowned Sparrow (Zonotrichia leucophiys)

WEKI

Western Kingbird (Tyrannus verticalis)

V/EME

Western Meadowlark (Sturnella neglecta)

WIFE

Willow Flycatcher (Empidonax traillii)

WIPH

Wilson’s Phalarope (Phalaropus tricolor)

WISN

Wilson’s Snipe (Gallinago delicata)

WIWA

Wilson’s Warbler (Wilsonia pusilla)

YBCU

Yellow-billed Cuckoo (Coccyz.us americanus)

YEWA

Yellow Warbler (Dendroica petechia)

YHBL

Yellow-headed Blackbird (Xanocephalus xanocephalus)

Ill

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 4. December 2010

The Wilson Journal of Ornithology 122(4):772-776, 2010

A Digital Spot-mapping Method for Avian Field Studies

Kevin E. Jablonski,' "^ Stacy A. McNulty,^ and Matthew D. Schlesinger^

ABSTRACT. — Avian mapping, also known as spot
mapping or territory mapping, is a breeding season bird-
survey technique that traditionally uses paper maps on
which locations of birds are recorded. This method is
often considered the most accurate in yielding a density,
but has been criticized as being inefficient, time
consuming, and inexact. We describe a novel digital-
mapping method, incorporating a hand-held computer
and high-accuracy global positioning system receiver
(GPS), used in an ongoing boreal birds study. Digital
mapping surpassed our expectations as to efficiency,
flexibility, and work flow. We expect this method will
become increasingly useful in many types of field
studies, especially as costs decrease (currently ~$2,100
for the field receiver used). Received 4 January 2010.
Accepted 23 June 2010.

Avian mapping, also known as spot mapping or
temtory mapping, is a widely used bird-survey
technique in which an observer attempts to record
the exact locations of individual birds within a
defined study plot. The method emerged in the
early 20th century as part of a naturalistic,
observational approach to field biology, and was
first described by Williams (1936:376) who wrote
that “by the use of weekly maps ... it is possible
to build up a series . . . covering the nesting season,
which will show concentrations of records about
certain localities.” Subsequent re.searchers ex-
panded on Williams’s idea and offered methods of
statistical Interpretation of mapping data (Ken-
deigh 1944, Pough 1947, Hall 1964). However,
field methods for mapping in ornithology have
remained essentially static since Robbins (1970)
offered an international standard (Verner 1985,
Manuwal and Carey 1991, Bibby et al. 2000).

The mapping method most commonly used
involves establishment of a physical grid, 8-20 ha
in size, using flagging or stakes to mark gridline

' SUNY ESF-244 Illick Hall. I Fore.stry Drive, Syracu.se,
NY 13210. USA.

^Adirondack Ecological Center, SUNY ESF, 6312 State
Route 28N, Newcomb, NY 128.32, USA.

’New York Natural Heritage Program. 62.3 Broadway,
3th Floor, Albany. NY 12233, USA.

’Corresponding author;
e-mail: kevinJablonski@gmail.com

intersections. An observer walks slowly along the
gridlines noting the location of observed birds on
a paper map using a set of symbols to indicate
species, M/F, and behavior. The observer com-
piles the maps from each visit after 5-10 visits,
creating one map per species. This enables
delineation of individual birds’ teiritories and
estimation of breeding-bird density and diversity,
as well as habitat structure and composition at
bird locations or territories (Robbins 1970,
Manuwal and Carey 1991, Ralph et al. 1993,
Bibby et al. 2000).

The main limitations of the paper mapping field
method as currently applied concern two areas:
(1) the large time investment required for
collecting and compiling data, and (2) the chances
of making errors during data recording and
interpretation. It can be time-consuming to create
a physical grid using flagging and stakes over 8-
20 ha, especially in thick vegetation or rough
terrain. Compiling several maps for each bird
species requires hours of data preparation and
invites human eiTor in interpreting and transfer-
ring data from one map to another. Recording
detailed bird locations, behavior, and movements
on maps scaled for practical field use can be
inaccurate on paper maps (Verner and Milne
1990).

We studied habitat associations of lowland
boreal birds in Adirondack Park, New York, USA
with 10-min variable circular plot point counts in
2008. However, we decided that a more location-
explicit approach was wairanted for subsequent
data collection after the initial field season. Our
study was focused on a group of rare, cryptic birds
in wet and often densely vegetated terrain. Thus,
we sought a method that would provide more
llexibility in detecting rare birds and allow us to
spend more time at high-diversity sites than
allowed by fixed-point counts. These require-
ments led us to use mapping as a technique for the
2009 field sea.son. We suspected that paper-based
mapping was not sufficiently efficient, accurate,
or llexible for our needs and would be laborious to
incorporate into our existing Geographic Infor-
mation System (GIS). We (1) de.scribe a novel

SHORT COMMUNICATIONS

773

TABLE 1. Target species of lowland boreal bird study.
Spring Pond Bog, Adirondack Park, New York.

Species

Common name

Picoides circticiis

Black-backed Woodpecker

Contopus cooperi

Olive-sided Flycatcher

Empidomix flaviventris

Yellow-bellied Flycatcher

Pehsoreus canadensis

Gray Jay

Poecile luidsoniciis

Boreal Chickadee

Dendroica tigrina

Cape May Warbler

D. palmanim

Palm Warbler

D. castanea

Bay-breasted Warbler

Melospiza lincolnii

Lincoln’s Sparrow

Eitphagits carol inLis

Rusty Blackbird

digital-mapping method that incoiporates a high-
accuracy global positioning system receiver
(GPS), a hand-held computer, and a GIS-derived
grid, and (2) discuss the advantages over tradi-
tional paper mapping as well as potential
drawbacks.

METHODS

Our study area was the Spring Pond Bog
Preserve, owned by The Nature Conservancy, in
northern Adirondack Park, New York (44° 22'
23" N, 74° 30' 21" W). We applied the digital-
mapping method to our study of habitat associa-
tions of lowland boreal birds, a suite of 10 species
that occur in the scattered lowland boreal habitats
of the Adirondacks (Table 1 ). We established five
9-ha plots and visited each plot five times from 28
May to 8 July 2009, starting 15 min after sunrise.

We created a GIS prior to entering the field
using ArcGis Desktop 9.2 (ESRI 2007) that
contained orthographic aerial photographs, wet-
lands, roads, trails, and other geospatial data
layers. We created five 300- X 300-m digital plots
centered on point-count locations where we had
recorded high target bird diversity in 2008. We
divided each of these plots into a grid of 900 10-
X 10-m .squares (Fig. 1 ), and created a point layer
for recording bird observations. We created drop-
down menus for attributes such as date, species,
seen/unseen, M/F, and behavior to facilitate
data entry in the field (Fig. 2). We loaded the.se
data layers into ArcPad 7.1 on a Trimble
Nomad handheld computer. The Nomad was
equipped with a Wide-Area Augmentation System
(WAAS)-enabled Garmin lOX Bluetooth GPS
receiver.

We conducted plot visits in much the same way
as paper mapping. However, we used the Nomad

along with a compass for navigation and data
recording. We selected a random starting point
and path through each plot by spinning a compass,
and then proceeded at a calm, quiet pace through
the plot using the Nomad’s “Tracklog” feature to
en.sure that we followed grid lines so that we came
within 20 m of every grid point. A full 9-ha plot
visit lasted 90-120 min. No flagging or .stakes
were necessary to mark the plot or grid.

When we detected an individual of one of the
target species, we ascertained its exact location
and noted behavioral data related to territoriality
or nesting activity. We then walked to the bird’s
location and, after ensuring the chosen GPS
accuracy measure. Positional Dilution of Preci-
sion (PDOP), was below the pre-established
threshold of 3.5, created a new data point using
the Nomad touch screen. We decided on a PDOP
threshold of 3.5 based on recommendations in the
literature (i.e., Kennedy 2002) and expected
accuracy under variable field conditions (from
open, boggy peatlands with low relief to dense
conifer Vv'etlands). We used the drop-down menus
in creating this point (Fig. 2) to record attributes
of the detection. We changed the map view
according to navigational or data-recording needs
throughout the visit. For example, when locating
an individual bird, we zoomed to a scale of
~ 1:400 (in which a grid square is 2.54 cm wide),
whereas while navigating we used a scale of
1:1,000-2,000. For comparison, a typical paper
map with a 9-ha grid would be scaled to 1:1,600
(in which a grid square is 0.64 cm wide). Each
week we backed up the data set to our desktop
GIS.

RESUFTS

We recorded 153 individual detections of nine
of the 10 target lowland boreal bird species in 25
plot visits. The number of target individuals
detected per visit ranged from 3.8 to 10.2 (.v =
6.12, a = 2.44) while total target species recorded
per grid ranged from 2 to 7 (.v = 5.0. a = 1.87).
We compiled species detection maps for all sites
after uploading the Nomad data to a desktop GIS
(Fig. I).

DISCUSSION

The digital spot-mapping method was easy to
use and work flow efficiency exceeded our
expectations for spatial and attribute accuracy,
and data acquisition and storage speed. Drop-
down menus made field data entry fast and

774

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4. December 2010

f±llSL^\

Legend
A RUBL
YPWA

FIG. I. Example of 300- X 300-m (9-ha) digital grid with accumulated detections of Rusty Blackbird (RUBL) and
Palm Warbler (YPWA) at site SPB08, Spring Pond Bog, Adirondacks, New York.

unambiguous, unlike the complex system of paper
map symbols an observer must correctly interpret
to record breeding bird attributes. Dense or
diverse bird populations can render paper map-
ping time-consuming and confusing, if not
unusable. Digital mapping has no functional upper
limit, as existing data points do not preclude
recording additional observations at the same
location. The map can be scaled in the field to
show more or less detail, and dense bird
populations can be recorded without clutter.
Errors of translation can be substantially reduced,
as there is no need to interpret and compile
multiple paper maps, especially il that compila-
tion occurs after the details of field observations
fade from memory.

The Bluetooth GPS quickly and consistently
obtained accurate locations, keeping us on the
path of travel with no need to mark the plot in
advance. The Nomad straps safely to one hand,
enabling the observer to enter data, u.se binocu-
lars, and navigate over obstacles and through

wetlands while maintaining the travel path.
Battery power lasted for —20 hrs of field use.
We kept the electronic devices charged with an
AC adapter for our vehicle.

We did not directly compare the two methods
but, in our experience, digital-mapping data can
mirror data acquired by typical paper mapping
studies. A review of recent studies that used the
traditional mapping method (Fink et al. 2006,
Toms et al. 2006, Villard et al. 2007, Romdal and
Rahbek 2008, Gale et al. 2009) indicates that each
could have incorporated digital mapping with
little or no change to other aspects of the study
while achieving a probable gain in efficiency in
data acquisition and processing. Digital mapping
can be also adapted to record unlimited additional
data point attributes, whereas paper mapping
would require additional data sheets for any
additional data.

We encountered some problems with the
digital-mapping method related to the specific
combination of device-software interface. ArcPad

SHORT COMMUNICATIONS

775

^ ArcPad

Terr detect

♦•I

Date

7/2/09 ▼

Site

SPB04

Grid_no

549

Species

YBFL ▼

Seen

q

N

Y

ArcPad

Terr detect

Sex

Terri

Terr2

Terr3

Nestl

Nest2

M

▼

Singing

▼

'W

J

Carrying Food
Carrying Matter
Copulation

Flush

Nest Found

Page 1

Page 2

^ Page 3

^ Page 1

Page 2

^ Page 3

1 i ►

FIG. 2. Screenshot of Trimble Nomad running ArcPad 7.1 with drop-down menus used to enter data in lowland boreal
bird mapping study. Spring Pond Bog, Adirondacks, New York.

7.1 is cumbersome to use in the field without
access to an ArcGIS-equipped computer because
changes to menus, data layers, or data sets must be
made using ArcGIS. The operating system of the
Nomad (Windows Mobile 6) hindered data
processing, especially in customization and add-
ing functionality. The interface between ArcGIS
and ArcPad was limited, requiring us to download
GIS shapefiles using Windows Sync rather than
ArcGIS.

A second limitation of digital mapping is cost
of implementation. The Trimble Nomad-Garmin
GPS combination costs ~$2,100 U.S., which does
not include the cost of the ESRI software license
or desktop computer. However, as recent devel-
opments with GPS-enabled personal devices such
as the Apple iPhone indicate, a combination of
improved and cheaper technology, as well as free,
customizable, open-source GIS software, is likely
to substantially lower this cost in the near future.

The digital spot-mapping method is useful for
scientific studies requiring accurate bird locations
and may also be suitable for citizen science
applications (Silvertown 2009). Many birding
enthusiasts, just as they have long been equipped
with binoculars and paper bird guides, now carry

smart mobile phones with electronic bird guides
from groups such as National Audubon Soci-
ety. For example, the Apple iPhone application
BirdsEye (http://ebird.org/content/ebird/news/
BirdsEye) provides real-time access to the Cornell
Laboratory of Ornithology’s eBird data base,
which contains millions of citizen-reported rec-
ords showing where and when birds were
observed in North America. The capacity to
record information about birds with a digital
spot-mapping application combined with a GPS,
camera, and recorder to capture bird photographs
and songs for later identification may enhance
information gathering and minimize interpretation
errors.

The potential of digital spot-mapping should
improve as hand-held computing also improves.
The mapping device of the near future will likely
be a small, affordable, hand-held, touch-screen
computer, equipped with a GPS unit capable of
sub-meter accuracy, simple open-source GIS
software loaded with (or accessing wirelessly)
maps and high-resolution aerial photographs, and
capability for syncing with a central storage
server/computer for data transfer. The potential
applications of such a device are myriad, and

776

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, December 2010

digital mapping can be expected to revolutionize
aspects of field ornithology.

ACKNOWLEDGMENTS

We thank The Nature Conservancy for property access,
and the Adirondack Park Agency and NYS GIS Clearing-
house for data layers. The manuscript was improved with
comments from C. J. Ralph and two anonymous reviewers.
Funding was provided by SUNY ESF. the Edna Bailey
Sussman Foundation, the Northern New York Audubon
Society, and the Wildlife Conservation Society.

LITERATURE CITED

Bibby, C. J., N. D. Burgess, D. A. Hill, and S. Mustoe.
2000. Bird census techniques. Academic Press,
London, United Kingdom.

Environmental Systems Research Institute (ESRI).
2007. ArcGIS Desktop, Version 9.2. Environmental
Systems Research Institute, Redlands, California,
USA.

Fink, A. D., F. R. Thompson III, and A. A. Tudor. 2006.
Songbird use of regenerating forest, glade, and edge
habitat types. Journal of Wildlife Management
70:180-188.

Gale, G. A., P. D, Round, A. J. Pierce, S. Nimnuan, A.
Pattanavibool, and W. Y. Brockelman. 2009. A
field test of distance sampling methods for a tropical
forest bird community. Auk 126:439-448.

Hall, G. A. 1964. Breeding bird census — why and how.

Audubon Field Notes 18:413^16.

Kendeigh, S. C. 1944. Measurement of bird populations.

Ecological Monographs 14:67-106.

Kennedy, M. 2002. The Global Positioning System and
GIS. Taylor and Erancis Inc., New York, USA.

Manuwal, D. a. and a. B. Carey. 1991. Methods for
measuring populations of small, diurnal forest birds.
USDA, Eorest Service, General Technical Report
PNW-GTR-278. Portland, Oregon, USA.

PouGH, R. H. 1947. How to take a breeding bird census.

Audubon Magazine 49:290-297.

Ralph, C. J., G. R. Geupel, P. Pyle, T. E. Martin, and D.
F. Desante. 1993. Handbook of field methods for
monitoring landbirds. USDA, Forest Service, General
Technical Report PSW-GTR-144. Albany, California,
USA.

Robbins, C. S. 1970. Recommendations for an international
standard for a mapping method on bird census work.
Audubon Field Notes 24:723-726.

Romdal, T. S. and C. Rahbek. 2008. Elevational zonation
of afrotropical forest bird communities along a
homogeneous forest gradient. Journal of Biogeography
36:327-336.

SiLVERTOWN, J. 2009. A new dawn for citizen science.

Trends in Ecology and Evolution 24:467^71.

Toms, J. D., E. K. A. Schmiegelow, S. J. Hannon, and M.
ViLLARD. 2006. Are point counts of boreal songbirds
reliable proxies for more intensive abundance estima-
tors? Auk 123:438-154.

Verner, j. 1985. Assessment of counting techniques.

Current Ornithology 2:247-302.

Verner, J. and K. A. Milne. 1990. Analyst and observer
variability in density estimates from spot mapping.
Condor 92:313-325.

Villard, M., E. K. a. Schmiegelow, and M. K.
Trzcinski. 2007. Short-term response of forest birds
to experimental clearcut edges. Auk 124:828-840.
Williams, A. B. 1936. The composition and dynamics of a
beech-maple climax community. Ecological Mono-
graphs 6:317—108.

SNORT COMMUNICATIONS

111

The Wilson Joiinuil of Ornithology 122(4):777-780, 2010

Variation of Type B Song in the Endangered Golden-Cheeked Warbler

( Denclroica chrysoparia )

Wendy J. Leonard,' Jayne Neal,' and Rama Ratnarn^'^

ABSTRACT. — The Golden-cheeked Warbler (Den-
droica chrysoparia), a member of the Parulidae, uses a
two-category song system referred to as Type A and
Type B. However, songs recorded from adult males in
Bexar County, Texas, USA, did not resemble the typical
A or B songs reported for this species. We report an
analysis of these atypical songs from six adult males.
Quantitative features of the measured songs and syllable
structure suggest this atypical song is more comparable
to a B song than an A song with some modifications.
The atypical song has dropped a terminal syllable of the
known B song, and selectively modified one of the
syllables to include frequency modulations (up and
down slurs). Further, the known B song was not heard in
the investigated sites. We conclude the atypical song is
a variant of the B song. Received 10 December 2009.
Accepted 30 June 2010.

The Golden-cheeked Warbler (Dendroica chry-
sopctria) (GCWA), a member of the Parulidae, has
a two-category song system, referred to as “first
category” and “second category” (Spector
1992). “First category” songs, used by unmated
males during daylight, are used in disputes with
other males and use declines after mating (Spector
1992). These songs, which can be simple and used
near females, are usually sung at low rates
(Bolsinger 2000). The “second category” song
is used during the dawn chorus or before sunrise,
and is more common later in the breeding season
(Spector 1992). This song is more complex and
faster than the first category song and is preceded
with short call notes (chips) (Pulich 1976, Spector
1992, Bolsinger 2000). It can also be more
variable than the first category song (Bolsinger
2000). Bolsinger (1997, 2000) uses the more
common terminology ‘A song’ (for first category
songs) and ‘B song’ (for second category songs).

One of the authors (JN) repeatedly heard a
GCWA song in spring and summer of 2008 and
2009 that was not a typical A or B song. These

' San Antonio Parks and Recreation Natural Areas, 21395
Milsa Road, San Antonio, TX 78256, USA.

"Depailment of Biology, University of Texas at San
Antonio, One UTSA Circle, San Antonio, TX 78249, USA.
^Corresponding author; e-mail: rama.ratnam@utsa.edu

birds were visually identified and confirmed to
be GCWA. Males sang these songs throughout
the season, although they were heard more
frequently later in the breeding season. It was
not evident whether the song was Type A or B.
We recorded song sequences in 2009 from six
individuals at three sites in northern Bexar
County, Texas, USA. Our goal was to quantita-
tively compare the new song with reported songs
for this species to correctly classify this song as
an A or B song.

METHODS

Songs were recorded from adult male GCWAs
on City of San Antonio, Texas properties. These
sites included Grey Forest (29°37'39"N, 98°42'
4"W), a City Park (29° 38' 9" N, 98° 38' 24" W),
and a site near Camp Bullis (29° 37' 12"N, 98°
35'24"W). We recorded over a distance of
— 10 km in areas of deciduous woodlands and
live oak/juniper {Quercus/Juniperus). Recordings
were made using omnidirectional microphones
(MKE-2, Sennheiser Electronic Corporation, Old
Lyme, CT, USA) connected to a high resolution
digital audio recorder (Model 722T, Sound
Devices EEC, Reedsburg, WI, USA) that ampli-
fied and digitized signals at 44.1 kSamples/sec,
and stored them as WAV files in the built-in hard
drive. Songs were analyzed using Audition
(Adobe Systems Inc., San Jose, CA. USA) and
MATLAB (The MathWorks Inc., Natick, MA,
USA). A band pass filter (sixth order. Type II
Chebychev, 2.5 to 10.5 kHz) was used to
eliminate energy outside the frequency band of
GCWA songs. Spectrograms were generated for
each segment using a Fast Fourier Transform
(lOO-msec segments, Hann windowed, 95%
overlap). Individual songs were extracted from
the sequences by visually scanning the spectro-
grams.

Songs were divided into syllables, and each
.syllable was classified and given a lower-case
letter designation. Quantitative analysis of songs
followed Bolsinger (1997, 2000). Histograms

778

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, December 2010

TABLE 1. Components of known GCWA and atypical GCWA songs. Control data (A song, B song) are from
recordings deposited at Macaulay Library. Atypical songs are from 6 males in northern Bexar County, Texas recorded in
May and June 2009. Values in parentheses are ranges.

Song complexity Song rate Pause duration Number of chips Frequency of

Duration (sec) (elements/song) (songs/min) (sec) preceding song modal intensity (Hz)

A song' L9U1.6-L9) 3" 5.T (3.9-6.0) 10.2“ (1-3-13.9) 0.7“ (0-1) 5,354“ (4,580-5,980)

B song' 2.0“(L8-2.3) 5'’ (4-6) 6.8“ (3. 1-9.1) 8.1“ (4.5-17.1) 5.1“ (0-14) 4,863“ (4,280-5,420)

Atypical song (2009) 2.1 (L3-2.3) 5.8 (4-7) 5.8 (3.4-11.2) 9.0 (2.5-15.7) 3.5 (0-12) 4,675 (4,200-5,080)

Data from recordings deposited at Macaulay Library, Cornell Laboratory of Ornithology. Audio record numbers: J. S. Bolsinger (A song birds = 4, B song birds
= 3; circa 1994) 109301, 109338, 109340, 109734, 109736, 109749-109752, 109754, 109760, 109761; W. H. Gunn (B song bird = 1; c. 1977) 68956, G, A. Keller
(A song bird = 1; circa 1986) 40560, 40561, T. A. Parker III (A song bird = 1; circa 1986) 38337, 38344, 38347, R. C. Stein {B song bird = 1; circa 1961) 9190-
9195.

*’ Data from Bolsinger (1997:38, table 1.1).

were constructed for the following parameters: (1)
song complexity (defined as the number of
syllables per song), (2) number of chips preceding
a song, (3) song duration, (4) pause duration
(between songs in a bout), (5) song rate, and (6)
frequency of modal intensity (defined as the
frequency component with the greatest energy).
Pause durations exceeding the 90th percentile
were excluded from histograms as they indicated
gaps between bouts. A and B song recordings
made between 1961 and 1994 were obtained from
the Macaulay Library (Table 1). The recordings
served as control data and were analyzed similarly
for number of chips, song duration, pause
duration, song rate, and frequency of modal
intensity. The frequency of modal intensity was
measured from the spectrum of a syllable rather
than the entire song (as adopted by Bolsinger
1997). The syllable was selected so it had
comparable duration and frequency bandwidth
across the set of atypical and control songs.

Statistical comparisons of parameters were
performed among three groups: (1) the atypical
2009 recordings, (2) control A songs, and (3)
control B songs. Song parameters were averaged
over all songs for a given bird, and then averaged
over the number of birds. Thus, sample sizes
reflect the number of individual birds recorded,
and not the total number of recorded songs. The
goal was to ascertain if the atypical 2009
recordings were similar to the A song or the B
song by comparing for differences in means of the
measured parameters. All comparisons were
performed using the Kruskal-Wallis test (Sokal
and Rohlf 1995) to look for significant differences
between means of the groups. A pair-wise
multiple comparison test was used to examine
which pairs of groups differed in their means. The
multiple comparison procedures (with a Bonfer-

roni correction) reported confidence intervals at
a-level of 0.01.

RESULTS

Two hundred and fifty-nine songs were record-
ed from six adult GCWA males from 17 May to
13 June 2009. Numbers of songs produced by
each bird (in decreasing order) were: 102, 63, 37,
26, 17, and 14. A total of 161 control A songs was
obtained from recordings of six birds (numbers of
songs were 67, 24, 21, 19, 16, 14), and 1 15 control
B songs were obtained from recordings of five
birds (numbers of songs were 35, 33, 19, 18, 10).

Representative control A and B songs (Fig. lA,
B, respectively) were compared with atypical
songs (2 examples shown in Fig. 1C, D). A visual
comparison of the spectrograms suggests the
atypical song more closely resembles the known
B song than the A song. However, there are
differences in the number of syllables and the
structure of the song. The known B song has five
syllables labeled a through e (Fig. IB) whereas
the atypical song has only four syllables and lacks
the terminal e syllable. Syllable a is markedly less
intense in the atypical song and lacks the buzzy
nature in the known B song. In some instances it
was so faint as to be barely audible above the
background noise. Syllable b is well-preserved
and has been used to align the songs (Fig. IB, C,
D), and syllable d has been modified in duration
but is otherwise unchanged. The greatest modifi-
cation occurs in syllable c which in the known B
song is a complex of two tonal elements centered
between 6 and 4 kHz (Fig. IB). This syllable
(marked as c' in Fig. 1C, D) in the atypical song
has been lengthened and includes complex
frequency modulations (up and down slurs); it is
more variable across individuals than other
syllables (Fig. 1C, D).

SHORT COMMUNICATIONS

779

A

7,6

6

2,6

B

7S

6 -

26

Song A (Bobinger 1097)
d

Song B (Botsinger 1997)

* t) j e

c

7.5

6

26

5

26

0.5 s«c

FIG. 1. Song spectrograms of (A) known A song, and
(B) known B song. These songs were recorded between
1961 and 1994 and served as control data. C and D are
atypical songs reported in this study. The known B and
atypical songs are aligned on the offset of syllable b.

The similarities and differences between the
known A, B, and atypical songs were quantified
by analyzing song parameters (Table 1). Song
complexity was similar for the atypical song and
the known B song with the A song demonstrating
less complexity. The A song is not normally
preceded by chips and a comparison of the mean
number of chips was possible only between the B
song and the atypical song. Mean number of chips
was significantly higher in the B song than the
atypical song, but the number of chips produced
had a similar range. Frequency of modal intensity
was estimated from the d syllable for all songs.
This syllable in the control A song was identified
as the pre-terminal syllable (Fig. lA) as it most
closely resembles the d syllable in the atypical
and control B songs. The mean frequency of
modal intensity for the known B song (4,803 Hz)
and atypical song (4,675 Hz) were closer to one
another than to the A song (5,354 Hz), but they all
significantly differed from one another, as did
mean song durations for all pairs. There were no
significant differences between song rates of the
atypical and A or B songs, and the pause duration

between the atypical and B songs. All other paired
comparisons were significantly dii'ferent.

DISCUSSION

The atypical GCWA songs recorded in 2009 in
northern Bexar County are different from the B
song reported for this species in a number of
respects, but the spectrogram analysis, syllable
similarity, and structure suggest this atypical song
is likely to be a variation of the known B song
(Fig. 1 ). Further, A songs are not usually preceded
by chips or calls notes whereas the known B song
and the atypical song were frequently preceded
by chips. These findings suggest the atypical song
is a B or “second category” song (Spector 1992)
with a markedly modified c syllable, and lacking
the terminal e syllable found in the control B
song.

The typical A song was heard infrequently
during the field investigations, but was not
recorded. The atypical B song was heard more
frequently during the later part of the GCWA
breeding season. This agrees with the description
of A and B song use by Spector (1992). A or
“first category” songs dominate early in the
breeding season and then decline after pairing
(Spector 1992). B or “second category” songs are
more common later in the breeding season
(Spector 1992). We note the typical B song was
not heard from these six birds during field
investigations.

We propose the atypical song is the B song
which has undergone variation over time. This has
important implications for conservation. Accurate
population estimates are essential for assessing
the status of the federally listed endangered
GCWA; these assessments often rely on auditory
detections. Our study sites are in a region of Bexar
County that is experiencing rapid urban develop-
ment with clearing of oak and Ashe juniper
iJuniperus ashei) woodlands necessary for breed-
ing. Further investigations over several seasons
are needed to identify the ecological significance
and permanence of the song change. Are the
changes in song structure temporary resulting
from a loss of GCWA breeding habitat, a change
in vegetation, or increased habitat fragmentation
(Laiolo and Telia 2005)? Or, are the changes
permanent resulting from changes in morphology
(Derryberry 2009)7 Our work is preliminary and
was conducted late in the season. A more detailed
study will be conducted next season to fully
characterize this B song variant, but there is

780

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

sufticient evidence to warrant larger and detailed
surveys of the species over its breeding range,
particularly in isolated and fragmented habitats.

ACKNOWLEDGMENTS

The authors thank J. S. Bolsinger for helpful discussions,
Macaulay Library for audio recordings of Dendroica songs,
L. B. Hawkins for making an audiovisual recording of a
GCWA singing the new song, and M. D. Valero for
comments on the manuscript. The research was funded by
the Norman Hackerman Advanced Research Program
(NHARP) (Grant Number 010115-0101-2007) from" the
Texas Higher Education Coordinating Board (to RR) and
UTSA STARS (to RR).

LITERATURE CITED

Bolsinger, J. S. 1997. Patterns of use and variations in the
songs of the Golden-cheeked Warbler (Dendroica

chrysoparia). Thesis. University of Massachusetts,
Amherst, USA.

Bolsinger, J. S. 2000. Use of two song categories by
Golden-cheeked Warblers. Condor 102:539-552.

Derryberry, E. P. 2009. Ecology shapes birdsong
evolution; variation in morphology and habitat ex-
plains variation in White-crowned Sparrow song.
American Naturalist 174:24-33.

Laiolo, P. and j. L. Tella. 2005. Habitat fragmentation
affects culture transmission: patterns of song matching
in Dupont’s Lark. Journal of Applied Ecology
42:1183-1 193.

PULICH, W. M. 1976. The Golden-Cheeked Warbler, a
bioecological study. Texas Parks and Wildlife Depart-
ment, Austin, USA.

SoKAL, R. R. AND F. J. Rohlf. 1995. Biometry. W. H.
Freeman and Co., New York, USA.

Spector, D. a. 1992. Wood warbler song systems: a
review of paruline singing behaviors. Current Orni-
thology 9:199-238.

The Wilson Journal of Ornithology 122(4):780-783, 2010

Nocturnal Social Cues Attract Migrating Yellow-breasted Chats

Mark G. Alessi,'-^ Thomas J. Benson,' and Michael P. Ward'

ABSTRACT. — We tested the hypothesis that migrat-
ing birds use nocturnal conspecific song when selecting
stopover habitat using data from the Yellow-breasted
Chat (Icteria virens). We broadcast nocturnal song in
unsuitable habitat (i.e., a manicured orchard) and
alternated broadcast nights with nights where no song
was broadcast. We caught significantly more individu-
als (8 males, 7 females) on mornings following
treatments relative to control nights when no songs
were broadcast (2.5 vs. 0 bird.s/morning, respectively).
Eleven of 15 (73%) chats were removed from the nets
after sunrise (mean = 32 min after sunrise, range = 5-
60 min), and birds were captured on overcast, cloudy,
and clear mornings. Only one bird was recaptured at the
site, and only one male was detected singing at the site,
suggesting individuals quickly left the area. Conspecific
nocturnal songs for chats appear to be an important cue
for selecting stopover habitat. Received H March 2010.
Accepted 9 July 2010.

Reliable cues of high-quality habitat are
important for migratory birds becau.se energetic
and fitness costs of not locating suitable habitat

' Illinois Natural History .Survey, Institute of Natural
Resource Sustainability, University of Illinois, Champaign,
IF 61820, USA.

^Corresponding author; e-mail: mgalessKG'gmail.com

are likely high. Migratory birds use a suite of
navigational cues to migrate between wintering
and breeding areas (Able and Bingman 1987);
however, the specific cues used to select stopover
habitat remain largely unknown. Studies suggest
that both visual (Moore et al. 1995) and acoustic
cues (Ward and Schlossberg 2004, Mukhin et al.
2008) may be involved. However, many noctur-
nally migrating passerines select stopover loca-
tions prior to sunrise (Cochran et al. 1967, Moore
and Aborn 2000) when visual cues may be absent
or difficult to evaluate. Cues that facilitate
locating suitable habitat are likely highly valuable
for migrating birds, and one may be nocturnal
song.

Nocturnal song by diurnally active birds
appears to be uncommon; however, recent
research suggests nocturnal song may be an
important habitat location cue (Herremans 1990,
.Schaub et al. 1999, Mukhin et al. 2008). Several
diurnally active species that sing at night have
specific nocturnal songs. Eor example, research
on Common Nightingales {Luscinia megarhyu-
chos) suggests males sing a whistle song at night
to attract females (Glutz von Blotzheim 1988,
Amrhein et al. 2002, Roth et al. 2009). The

SHORT COMMUNICAl IONS

781

nocturnal song of Yellow-breasted Chats {Icteria
vireiis) is also given at a frequency and amplitude
that may allow detection farther than their diurnal
song (Canterbury 2007). Female and male chats
may use this song as an indicator of suitable
habitat and mates. However, the relative impor-
tance of conspecific vocalization versus habitat
cues is unknown.

We broadcast the nocturnal song of male
Yellow-breasted Chats, a song that is consistently
sung at night during spring migration (Canterbury
2007), in unsuitable habitat during spring migra-
tion. The shrubland habitat in which chats
stopover is patchy, and the probability of a
migrating individual quickly locating suitable
habitat at dawn is unlikely; thus, birds would
benefit by using pre-dawn cues to locate appro-
priate areas. We hypothesized that both male and
female chats use nocturnal song as an indicator of
suitable habitat, and that both males and females
would select areas with conspecific nocturnal
song regardless of habitat suitability. We predict-
ed that chats attracted to unsuitable habitat would
quickly leave the site once they found it was
unsuitable.

METHODS

This experiment was conducted on Univer-
sity of Illinois property in southeast Urbana,
Champaign County, Illinois, USA during spring
migration between 29 April and 15 May 2009.
The site was a 210 X 30 m linear strip of six
rows of apple (Mains spp.) trees spaced 6.5-m
apart with mowed strips of grass between each
row of trees. The habitat north and east of the
site was residential and commercial property,
while the habitat to the we.st and south was a
mixture of agricultural land with additional
patches of apple trees and mature forest. The area
was selected because it offers woody vegetation,
but lacked the size and dense shrub layer that
chats require for both breeding (Thompson and
Nolan 1973) and stopover habitat (Parnell 1969).
Yellow-breasted Chats are not known to nest
within the Champaign city limits (Kleen et al.
2004), and the nearest shrubland with breeding
chats is ~16 km away. The site is routinely
monitored during statewide bird surveys and other
research projects, and chats have yet to be
recorded at this location.

We alternated between control nights, when no
songs were broadcast, and treatment nights with
song playbacks when the weather allowed (i.e..

not raining). We used a mixture of commercially
available chat songs (Elliott et al. 2000) and chat
songs we recorded from individuals in east-central
Illinois. The recording included a male whistle
song, the song only used at night (Canterbury
2007). The entire recording was 100 sec in length
and was repeated continuously using a Eox Pro
FX3 playback system. The whistle song was
imbedded in two locations during the 100-sec
playback: 15 sec and 80 sec, and was 0.5 sec in
duration. We placed the system in the center of
the site in a weatherproof plastic container and
directed the device at a 45° angle southward
toward the sky. Songs were broadcast on treat-
ment nights from sunset to 1.5 hrs after sunrise.
We placed mist nets approximately every 30 m on
the site to increase the probability of capturing all
chats moving through the vegetation; nets were
numbered 1 (South net) to 6 (North net) and were
opened for 1.5 hrs before and after sunrise each
morning for both treatment and control nights.
Nets were checked every 20 min for birds. We
banded all captured chats with a USGS band,
recorded gender, and the net number in which
each individual was captured prior to releasing
them at the site. We conducted 5-min, unlimited-
radius point counts each morning in two locations
at the site after closing the nets to quantify the
number of singing chats. One point was at the
southernmost end of the strip while the other was
at the northernmost end. We conducted a Chi-
square test to examine if chats were captured more
often than expected on mornings following
treatments.

RESULTS

We captured 15 chats (8 males, 7 females)
during four of six mornings that we broadcast chat
song at night and captured no new individuals
during any of the five control mornings (x' =
12.5, P < 0.001). Eight of 15 chats (53%) were
captured in nets closest to the speaker (chat song),
while seven (47%) were captured in the south-
ernmost nets. Eleven of 15 (73%) chats were
removed from the nets after sunrise (mean =
32 min, range = 5-60 min), and birds were
captured on overcast, cloudy, and clear mornings.
Only one male chat was heard singing during
point counts and it was during the morning he was
captured. No male chats were recorded singing on
control days. Only one individual was recaptured;
a female 3 days after initial capture.

782

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4. December 2010

DISCUSSION

Migrating chats were captured only on morn-
ings following nights when conspecific vocaliza-
tions were broadcast; no birds were captured
during mornings following nights with no song.
Both migrating male and female chats responded
to experimental addition of nocturnal male song,
suggesting male chat song functions as a cue for
selection of stopover habitat. Male chats did not
establish territories and vacated the area. The lack
of appropriate habitat did not preclude chats from
briefly stopping, as long as conspecific songs
were present.

Nocturnal song of diurnally active birds has
been documented, primarily in species that breed
in patchy or ephemeral habitat (Barclay et al.
1985, Walk et al. 2000). For example, male
Common Nightingales breed in ephemeral habitat
(Hewson et al. 2005) and also sing nocturnally for
mate attraction (Amrhein et al. 2002). Male chats
also appear to primarily use long-range whistle
songs only at night (Canterbury 2007). Male chats
will sing continuously at night for up to 6 days
after arrival in breeding areas, potentially provid-
ing abundant social cues early in the season when
most females are returning from wintering
locations (Canterbury 2007).

We broadcast nocturnal song between sunset
and 1 .5 hrs after sunrise, and chats were attracted
either during the night or shortly after sunrise.
Selecting stopover habitats at sunrise may be
adaptive if increased visibility allows for better
habitat as.sessment (Bolshakov and Bulyuk 1999);
these results suggest that, at least for chats,
individuals are selecting habitat based on social
cues. Visibility can also have a role in affecting
the importance of social cues (Herremans 1990);
birds in our study were captured during cloudy,
overcast, and clear mornings, but whether visibil-
ity had a role in why birds were captured is
unknown.

Conspecific attraction has received increasing
attention in recent years, and is important for both
habitat selection and planning conservation and
management actions (Ahlering and Faaborg
2006). Several studies have demonstrated that
birds use conspecific attraction to assess and
select habitat (Mukhin 2004, Serrano et al. 2004,
Ward and Schlossberg 2004, Ahlering 2005, Betts
et al. 2008, Ktitorov et al. 2010). Most conspecific
attraction studies have not attempted to attract
birds to unsuitable habitat, nor have studies

focused exclusively on nocturnal song (but see
Mukhin et al. 2008, Ktitorov et al. 2010).
Attracting birds to unsuitable rather than suitable
habitat is a more reliable method of assessing the
importance of acoustic social cues (Chemetsov
2006, Betts et al. 2008).

Heterospecifics may also use these cues while
selecting habitat (Mbnkkonen et al. 1999). We did
capture other shrubland bird species on mornings
following treatments, including Orchard Orioles
{Icterus spurius). Brown Thrashers {Toxostoma
rufum), and Field Sparrows (Spizella pusilla), but
either did not capture or caught fewer individuals
following control nights. These species do not
commonly breed in developed areas. A separate
playback experiment is necessary to examine if
these species use chat song in addition to
vegetation structure to select suitable habitat.
Many of the bird species that breed in patchily
distributed habitats in North America are known
to sing at night (Barclay et al. 1985, Walk et al.
2000) and may be using nocturnal heterospecific
song as a cue during migration; future research is
needed to address conspecific and heterospecific
attraction in shrubland birds.

ACKNOWLEDGMENTS

We thank D. G. and M. N. Barron, M. R. Heimbuch, and
D. C. Elbert for assistance in the field. We are grateful for
comments from J. D. Brawn, D. A. Enstrom, and P. J.
Weatherhead. We also thank C. E. Braun and three
anonymous reviewers for comments on the manuscript.
This research was conducted under Institutional Animal
Care and Use Committee protocol #06248 and banded
under Federal permit #06507.

LITERATURE CITED

Able, K. P. and V. P. Bingman. 1987. The development of
orientation and navigation behavior in birds. Quarterly
Review of Biology 62:1-29.

Ahlering, M. A. 2005. Settlement cues and resource use
by Grasshopper and Baird’s sparrows in the upper
Great Plains. Dissertation. University of Missouri,
Columbia, USA.

Ahlering, M. A. and J. Faaborg. 2006. Avian habitat
management meets conspecific attraction: if you build
it, will they come? Auk 123:301-312.

Amrhein, V., P. Korner, and M. Naguib. 2002. Nocturnal
and diurnal singing activity in the nightingale:
correlations with mating status and breeding cycle.
Animal Behaviour 64:939-944.

Barclay, R. M. R., M. L. Leonard, and G. Friesen.
1985. Nocturnal singing by Marsh Wrens. Condor
87:418-422.

Bett.s, M. G., a. S. Hadley, N. Rodenhouse, and J. J.
Nocera. 2008. Social information trumps vegetation

SHORT COMMUNICATIONS

783

structure in breeding-site selection by a migrant
songbird. Proceedings of the Royal Society of London,
Series B 275:2257-2263.

Bolshakov, C. V. and V. N. Bulyuk. 1999. Time of
nocturnal flight initiation (take-off activity) in the
European Robin Erithacus rubecula during spring
migration: direct observations between sunset and
sunrise. Avian Ecology and Behaviour 2:51-74.

Canterbury, J. L. 2007. Songs of the wild: temporal and
geographical distinctions in the acoustic properties of
the songs of the Yellow-breasted Chat. Dissertation.
University of Nebraska, Lincoln, USA.

Chernetsov, N. 2006. Habitat selection by nocturnal
passerine migrants en route', mechanisms and results.
Journal of Ornithology 147:185-191.

Cochran, W. W,, G. G. Montgomery, and R. R. Graber.
1967. Migratory flights of Hylocichla thrushes in
spring: a radiotelemetry study. Living Bird 6:213-225.

Elliott, L., L. Stokes, and D. Stokes. 2000. Stokes field
guide to bird songs: Eastern Region. Time Warner
Audio Books, New York, USA.

Glutz von Blotzheim, U. N. 1988. Handbuch der vogel
mitteleuropas. Band 11/1. AULA — Verlag, Wiesba-
den, Germany.

Herremans, M. 1990. Body-moult and migration overlap
in Reed Warblers (Acrocephalus scirpaceus) trapped
during nocturnal migration. Gerfaut 80:149-158.

Hewson, C. M., R. J. Fuller, and C. Day. 2005. An
investigation of habitat occupancy by the nightingale
Luscinia megarhynchos with respect to population
change at the edge of its range in England. Journal of
Ornithology 146:244-248.

KLEEN, V. M., L. CORDLE, AND R. A. MONTGOMERY. 2004.
The Illinois breeding bird atlas. Special Publication
Number 26. Illinois Natural History Survey, Urbana,
USA.

Ktitorov, P., A. Tsvey, and A. Mukhin. 2010. The good
and the bad stopover: behaviours of migrant Reed
Warblers at two contrasting sites. Behavioral Ecology
and Sociobiology 64:1135-1143.

Monkkonen, M., R. Hardling, j. T. Forsman, and j.
Tuomi. 1999. Evolution of heterospecifics attraction:
using other species as cues in habitat selection.
Evolutionary Ecology 13:91-104.

Moore, F. R. and D. A. Aborn. 2000. Mechanisms of en
route habitat selection: how do migrants make habitat
decisions during stopover? Studies in Avian Biology
20:34-42.

Moore F. R., S. A. Gauthreaux, P. Kerlinger, and T. R.
Simons. 1995. Habitat requirements during migration:
important link in conservation of neotropical landbird
migration. Pages 121-144 in Ecology and manage-
ment of neotropical migratory birds (T. E. Martin and
D. M. Finch, Editors). Oxford University Press, New
York, USA.

Mukhin, A. L. 2004. Night movements of young
Reed Warblers (Acrocephalus scirpaceus) in sum-
mer: is it postfledging dispersal? Auk 121:203-
209.

Mukhin, A., N. Chernetsov, and D. Kishkinev. 2008.
Acoustic information as a distance cue for habitat
recognition by noctumally migrating passerines during
landfall. Behavioral Ecology 19:716-723.

Parnell, J. F. 1969. Habitat relations of the Parulidae
during spring migration. Auk 86:505-521.

Roth, T., S. Sprau, R. Schmidt, M. Naguib, and V.
Amrhein. 2009. Sex-specific timing of mate searching
and territory prospecting in the nightingale: nocturnal
life of females. Proceedings of the Royal Society of
London, Series B 276:2045-2050.

SCHAUB, M., R. Schwilch, and L. Jenni. 1999. Does tape-
luring of migrating Eurasian Reed Warblers increase
number of recruits or capture probability? Auk
116:1047-1053.

Serrano, D., M. G. Forero, J. A. DonAzar, and J. L.
Tella. 2004. Dispersal and social attraction affect
colony selection and dynamics of Lesser Kestrels.
Ecology 85:3438-3447.

Thompson, C. F. and V. Nolan Jr. 1973. Population biology
of the Yellow-breasted Chat (Icteria virens L.) in
southern Indiana. Ecological Monographs 43:145-171.

Walk, J. W., E. L. Kershner, and R. E. Warner. 2000.
Nocturnal singing in grassland birds. Wilson Bulletin
112:289-292.

Ward, M. P. and S. Schlossberg. 2004. Conspecific
attraction and the conservation of territorial songbirds.
Conservation Biology 18:519-52.

784

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

The Wilson Journal of Ornithology 122(4):784-788, 2010

Range Expansion of the White-breasted Waterhen {Amaiirornis phoenicurus)

into Micronesia

Donald W. Buden’-'’ and Stanley Retograh

ABSTRACT. — We report the first occurrence of the
White-breasted Waterhen {Amaiirornis phoenicurus) in
Oceania based on discovery of a breeding population on
Woleai Atoll, Yap, Micronesia. Islanders believe it first
arrived on Woleai in the early 1970s, at a time when the
species was beginning a period of range expansion that
may be ongoing. Yap Outer Islanders report the
waterhen is present also on nearby Eauripik, Ifaluk,
and Laraulep atolls, where it arrived after it colonized
Woleai. We summarize extralimital, nonbreeding dis-
persal of A. phoenicurus and report a recent first record
for Palau. Received 15 January 2010. Accepted 28 April
2010.

The White-breasted Waterhen, Amaiirornis
phoenicurus is known to breed from Pakistan,
India, Sri Lanka, and the Maidive Islands east
through southeast Asia to Japan and the Philip-
pines, and south to Christmas and Cocos (Keel-
ing) islands (Australian Indian Ocean Territory),
and the Lesser Sundas (Taylor 1996). Steadman
(2006) reported that it is not indigenous to tropical
Oceania now or prehistorically, and makes no
mention of any vagrant sightings of it. We report
the first known occurrence of this species in
Oceania based upon a breeding population on
Woleai Atoll, Yap, Micronesia, hitherto known
only to island residents.

LOCALITY

Woleai Atoll (n7°23'N, 143° 55' E) is within
Yap State, one of four states (along with Chuuk,
Pohnpei, and Kosrae) comprising the Federated
States of Micronesia (FSM) (Fig. 1). The FSM
and the Republic of Palau, immediately to the
west, are in the Caroline Islands, which span
—3,000 km of the west-central Pacific Ocean
between the Philippines and Indonesia to the west
and the Marshall Islands to the east. Woleai Atoll

' Division of Natural Sciences and Mathematics. College
of Micronesia-FSM. P. O. Box 159, Kolonia. Pohnpei.
Federated States of Micronesia 96941.

"Yap State Department of Education. P. O. Box 220,
Colonia. Yap. Federated States of Micronesia 96943.
’Corresponding author; e-mail: don_buden@conifsm.lm

comprises 22 low (4 m maximum elevation asl)
coralline islands distributed along a barrier reef
around a 23.4 km' lagoon. The total land area is
—4.5 km'; Falalop is the largest island ( 1 .53 kni').
The nearest land area is Ifaluk Atoll, 50 km to the
east, Eauripik Atoll is 115 km to the southwest,
and Faraulep (= Feshaiulep) Atoll is 150 km to
the northeast. Sorol Atoll and the high islands of
Yap proper are —390 and 660 km to the
northwest, respectively. The vegetation on Fala-
lop, Woleai, consists largely of coconut {Cocos
micifera) agroforest, secondary woodland (shrubs
and small trees), and marsh vegetation in swampy
areas where people often cultivate taro (Cyrto-
spernia chaniissonis). Archaeological data for
Woleai are lacking, but the earliest known dates
of occupation for Fais Island and Ngulu, Ulithi,
and Lamotrek atolls among Yap Outer islands
range from 1,200-1,900 BP (Intoh 1997, Rainbird
2004). The national census in 2000 recorded 975
residents on Woleai (Yap Branch Statistics Office
2002).

OBSERVATIONS

On 18 June 2009, while collecting insects on
Falalop Island, Woleai Atoll, DWB heard several
A. phoenicurus calling at different locations in
low, wet areas along the airstrip and in the vicinity
of the settlement. He saw none of the birds during
—6 hrs on the island, and identified the calls in
retrospect. He queried several local residents and
learned only that the bird was called Kuwakuwa,
an onomatopoeic rendering of its call, and that it
probably arrived on the island about 30 years ago.
The residents’ descriptions of the bird were
surprisingly vague, considering its distinctive
appearance.

In response to our request to teachers on Woleai
for a photograph or a specimen, Ellisa Siyaneral
and Julian Yangerely captured a bird, which died
shortly thereafter, in mid-July 2009. The .speci-
men was frozen and subsequently delivered to SR
on 19 August via the government field trip ship.
Eddie Haleyalig delivered photographs of the

SHORT COMMUNICATIONS

785

FIG. 1 . The Federated States of Micronesia and islands where Amaurornis phoeniciirus has been reported.

specimen, taken by his daughter Jennifer, to
DWB, who identified it as A. phoeniciirus.
DWB prepared the bird as a study skin and depo-
sited it in the collection of the Natural History
Museum, Tring, UK (BMNH # 2010.13.1).

We distributed a questionnaire among Woleai
residents, and 22 of 30 respondents reported
seeing nests of the Kuwakuwa, 20 of 20 indicated
nests were in or near taro patches, 23 of 31
reported seeing eggs of this species, and 21 of 26
observed young, all on Falalop Island. Angelburt
Igemera, a student at College of Micronesia,
reported having seen a nest with eggs on Seliap
Island, and student Dickson Tiwelfil reported
hearing and observing the Kuwakuwa in taro
patches on Wottegai Island. These and other
responses indicate the White-breasted Waterhen is
present on all of the inhabited islands of Woleai
Atoll. We are confident these reports all pertain to
A. phoeniciirus as there are no other rallids on the
atoll and no other species with which it is likely to
be confused.

The extent of this species’ distribution else-
where among the Yap Outer Islands is uncertain.
Interviews with islanders indicate this species is
present at least on Eauripik, Ifaluk, and Faraulep
atolls, and that it probably arrived on these islands

from Woleai. Julius Sapelalut (pers. comm.).
Chief Public Defender, FSM Public Defenders
Office, and former resident of Ifaluk, reported that
A. phoeniciirus was not seen on Ifaluk until the
early-to-mid- 1980s, by which time it was already
well established on Woleai. Breeding has not been
confirmed on Ifaluk, but is probable because the
birds are heard calling year-round (J. Sapelalut,
pers. comm.). John Haglelgam (pers. comm.), a
professor at COM, reported his sister recalled
flushing a Kuwakuwa from a nest containing one
egg in a taro patch on Eauripik Atoll in 2008, and
Mickael Rapiy (pers. comm.) reported encounter-
ing this species on Faraulep Atoll a number of
years ago, after first having seen it on Woleai
~2 years earlier. The distribution and abundance
of A. phoeniciirus in central Yap State atolls
requires further study to supplement these anec-
dotal accounts.

Additionally, Mark Vereen, a resident of
Ngermetengel, on the west side of Babeldaob
Island, Palau, observed a White-breasted Water-
hen in his yard daily from 30 October to 3
November 2009. The bird was initially identified
by Doug Pratt from photographs taken by M.
Vereen, documenting the first record of A.
phoeniciirus for the Republic of Palau. It was

786

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4. December 2010

usually seen at the edge of dense vegetation, and
not tar from an extensive wetland area. It often
mingled with domestic chickens and was occa-
sionally observed feeding on earthworms in a
perpetually wet area of the yard. The disappear-
ance of the bird was coincidental with the
acquisition of two young dogs that roamed the
yard freely.

RANGE EXPANSION AND DISPERSAL

The appearance of the White-breasted Water-
hen in Micronesia apparently is coincident with a
broad range expansion by this species. Amaur-
ornis phoeniciinis is resident year-round through-
out most of its breeding range, but northern birds
are migratory and winter as far south as Sumatra,
and as far west as the Arabian Peninsula (Taylor
1996). Quantitative data on migration and dis-
persal are largely lacking, but Desmond Allen and
P. D. Round (correspondence, April 2010)
recorded 106 trapped during banding surveys at
Dalton Pass, Luzon Island, Philippines, during
November and December 2009. It was the most
numerous of 12 species of rallids encountered
during the survey. To what extent these were birds
originating from outside the Philippines, is
unknown.

The waterhen was rarely seen in Japan until the
1970s, but was then observed with increasing
frequency, and first recorded breeding in Kyushu
in 1982 (Nakamura 1987). It is now a year-round
resident in Japan’s Ryukyu Islands and Haha-jima
in the Ogasawara (= Bonin) Islands and continues
to advance northward into the Japanese main
islands (Brazil 1991, Taylor 1996). The waterhen
is now so common on Iriomote Island that it may
be displacing the Slaty-legged Crake (Ralliiia
ei(riz.(moides) there (Taylor 1996). Sakaguchi and
Ishihira (1998) attributed an increase in the
number encountered on Iriomote Island during
March-May to an influx of breeding adults, and
Nakamura (1987) indicated —73% of the birds
recorded in Japan were from March to August
with the highest frequency in May.

The White-breasted Waterhen was first record-
ed in Korea in 1970 and a failed breeding attempt
was recorded there in 2001 (Park 2002, Moores

2007) . The first recorded successful nesting in
Korea was in 2008 and two young were observed
at Namyang Bay on 30 August (Edelsten et al.

2008) . The waterhen was first recorded on Seram
in the Moluccas in 1987 (Bowler and Taylor
1989), and may now be resident there (Coates and

Bishop 1997). Colonization of Australian Indian
Ocean Territory islands may have been at
approximately the same time. Two young pro-
duced by a breeding pair in 1993, and again in
1994, and incidental sightings of other adults were
the first records of A. phoenicurus on Christmas
Island (Carter 1994), and the waterhen probably
became established as a resident breeder on Cocos
(Keeling) Islands sometime between the late
1990s and 2004 (Johnstone and Darnell 2004,
Dooley 2005, Reid and Hill 2005, Hadden 2006).
The White-breasted Waterhen was first recorded
on the Arabian Peninsula in Oman in 1977
(Gallagher and Rogers 1980), and Pedersen
(2008) considered it a rare to uncommon winter
visitor in Dubai (United Arab Emirates). One
photographed on 30 October 2009 is the first
record for Saudi Arabia and the northernmost
record for the Arabian Peninsula (Lobley and
Roberts 2010), but the birds have not yet bred
anywhere on the peninsula (M. C. Jennings, pers.
comm.).

Examples of long distance dispersal or vagran-
cy in this species include: one found alive in the
Seychelles on 24 December 1991 (Skeirett 1994);
one found dead in central Mongolia and another in
southern Mongolia in May 2004 and June 2004,
respectively (Gombobaatar et al. 2005); several
records from the Russian Far East, including one
found dead on Bolshoy Pelis Island on 20 June
and another in the Olginskiy District on 15 July
1987 (Nazarov and Kazykhanova 1988); one
observed in the Sikhote-Alin Nature Reserve on
the Pacific Coast of Sikhote Mountains in 1996
(Elsukov 1999) and another on Sakahlin Island on
20 September 1994 (Kozin 1995); one was
captured, banded, and released on the Avacha
River near Elizovo, northwest from Petropavlosk-
Kamchatskiy, Kamchatka Peninsula on 1 Novem-
ber 1996 (Gerasimov 1996).

DISCUSSION

We do not know exactly when White-breasted
Waterhens first arrived on and colonized Woleai
Atoll. The con.sensus among islanders we inter-
viewed suggests the species became established- in
the early 1970s, which would roughly coincide
with the early stages of this species’ ongoing
range expansion. None of the elders we inter-
viewed recalled seeing it any earlier, and none
was recorded anywhere in Micronesia by Baker
(1951), who included records of other species
encountered on Woleai by Japanese ornithologists

SHORT COMIVIUNICAl IONS

787

during the early 19()()s. That a resident population
has gone unreported from Woleai for over 30 years
is readily explained by the absence of knowl-
edgeable observers. Commercial flights to the
atoll have been, and continue to be, infrequent and
iiTegular. The FSM and Yap State government
ships also visit Woleai irregularly, but usually
spend only a brief part of a day before continuing
on.

Comparisons of photographs of the bird from
Woleai with specimens at the American Museum
of Natural History by Mary LeCroy, and at the
Smithsonian Institution by StoiTS Okson, indicate
the Woleai specimen is within the range of
variation of the nominate subspecies. The other
subspecies tend to have either less (A. p.
leiicoinelanus, Sulawesi, Moluccas, and Lesser
Sundas) or more (A. p. insularis and A. p.
midnicohariciis, Andaman and Nicobar Islands)
white in the forehead (Taylor (1996). The bird
photographed at Palau also appears to be an
example of the nominate subspecies (photographs
examined by DWB and Doug Pratt). The water-
hen population on Woleai could have originated
from the nearest sedentary population of A. p.
phoenicurus in the Philippines, but more likely
from migratory northeast Asian birds that strayed
east from their usual migratory path. The White-
breasted Waterhen appears to have bypassed
Palau, Yap, and the Marianas, high islands with
seemingly suitable habitat much nearer to poten-
tial source populations, in establishing a foothold
in Micronesia. The waterhen’s furtive and skulk-
ing habits may limit sightings, but its loud and
frequent vocalizations make it hard to overlook.
Possibly the recent Palau sighting is a harbinger of
wider colonization of Micronesia.

ACKNOWLEDGMENTS

We thank Carlo Cu.stodio. Guy Kirwan, Ali.son Pirie,
Elena Ilyashenko. Robert Meese, Doug Pratt, Robert Pry.s-
Jone,s, Bruce Robert, and Philip Round for a.ssistancc with
the literature search. Information on recent colonizations
and extralimital records was provided by Walter Boles
(for Australian Indian Ocean Territory islands), Adrian
Skerrett and Robert Prys-Jones (Seychelles Islands). Mike
Jennings (the Arabian Peninsula), Sundev Gombobaatar
(Mongolia), Kiyoaki Ozaki (Japan), and Vladimir Loskot,
Yuri Gerasimov, and Valentin Ilyashenko (English
translations of records from Ru.ssia). We thank Mary
LeCroy and Storrs Olson for comparing photos of the
Woleai bird with examples of different subspecies of A.
phoenicurus at the American Museum of Natural History
and the National Museum of Natural History, respective-
ly; Ellisa Siyaneral and Julian Yangerelyaro for obtaining

the first specimen of A. phoenicurus from Woleai, John
Talugmai for delivering it to Yap, and Eddie Haleyalig
and daughter Jennifer for the photographs, which made
the initial identification possible. We thank Julius
Sapelalut for information on the status of this species on
Ilaluk Atoll and are grateful also to the many other
residents and former residents of Yap Outer Islands who
shared their local knowledge of the Kuwakuwa. Addi-
tionally, we thank Doug Pratt for calling our attention to
the recent sighting of A. phoenicurus from Palau and
Mark Vereen for contributing this record, and Desmond
Allen and P. D. Round for contributing bird banding data
from the Philippines. We also appreciate the helpful
comments by D. Pratt, R. Prys-Jones, and an anonymous
reviewer on an earlier draft of the manuscript.

LITERATURE CITED

Baker, R. H. 1951. The avifauna of Micronesia, its origin,
evolution, and distribution. University of Kansas
Publications, Museum of Natural History 3:1-359.
Bowler, J. and J. Taylor. 1989. An annotated checklist of
the birds of Manusela National Park, Seram: birds
recorded on the Operation Raleigh Expedition. Kukila
4:3-29.

Brazil, M. 1991. The birds of Japan. Smithsonian
Institution Press, Washington, D.C., USA.

Carter, M. 1994. Birds of Australia’s Christmas Island.
Wingspan 13:18-21.

Coates, B. J. and K. D. Bishop. 1997. A guide to the birds
of Wallacea: Sulawesi, the Moluccas and Lesser Sunda
Islands, Indonesia. Dove Publications. Alderly,
Queensland, Australia.

Dooley, S. 2005. Twitchers' corner. Wingspan 15:41.
Edelsten, T, S. Kyu-Sik, and R. Newlin. 2008. Birds
Korea bird news (August), http://www.birdskorea.org/
Birds/Birdnews/BK-BN-birdnews-2008-08.shtml
Elsukov, S. V. 1999. Birds. Terrestrial animals' cadastre
of the Sikhote-Alin Nature Reserve and northern
Primorye. Vladivostok:29-75.

Gallagher, M. D. and T. D. Rogers. 1980. On some
birds of Dhofar and other parts of Oman. Journal of
Oman Studies, Special Report 2:347-385.

Gerasimov, Y. N. 1996. Accidental flight of Awaurornis
phoenicurus to Kamchatka. The Russian Journal of
Ornithology, Express-issue 1996 (5):6.

Gombobaatar, S., D, Sumiya. C. W. Leahy, and E.
Potapov. 2005. New records of bird species in
Mongolia. Amphibians, reptiles and birds. Scientific
Journal for Ornithology 2:121-128.

Hadden, D. 2006. Cocos (Keeling) Island birds. Wimzspan
16:34-37.

Intoh, M. 1997. Human dispersals into Micronesia.

Anthropological Science 105:15-28.

.IOHN.STONE, R. E. and J. C. Darnell. 2004. Appendix B.
Annotated checklist of birds from Cocos-Keelins
Islands. Pages 477-496 in Handbook of Western
Australia birds. Volume 2. Passerines (Blue-winged
Pitta to Goldfinch) (R. E. Johnstone and G. M. Storr,
Editors). Western Australian Museum. Canberra.

Kozin, A. N. 1995. Record of the new rallen in Sakhalin

788

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

White-breasted Crake. Bulletin of the Sakahlin
Museum 2. Yuzno-Sakhalinsk:289.

Lobley, G. R. and P. Roberts. 2010. First record of
White-breasted Waterhen Amaurornis phoenicurus in
Saudi Arabia. Sandgrouse 32:55-56.

Moores, N. 2007. Selected records from Socheong Island,
South Korea. Forktail 23:102-124.

Nakamura, K. 1987. Range expansion of White-breasted
Waterhen, Amaurornis phoenicurus, into Japan and its
colonization. Bulletin of the Kanagawa Prefectural
Museum (Natural Science) 17:1-11.

Nazarov, Y. N. and M. G. Kazykhanova. 1988. Another
finding in the USSR, the White-breasted Crake
Amaurornis phoenicurus Pennant. Page 141 in Rare
birds of the Far East and their protection (N. M.
Litvenenko, Editor). Vladivostok, Russia.

Park, Jin-YOUNG. 2002. Current status and distribution of
birds in Korea. Dissertation. Kyung Hee University,
Seoul, South Korea.

Pedersen, T. 2008. Birds of the United Arab Emirates — an
annotated checklist, http://www.tommypedersen.com/
UAEchecklist-annotated.htm

Rainbird, P. 2004. The archaeology of Micronesia.
Cambridge University Press, Cambridge, United
Kingdom.

Reid, J. R. W. and B. M. Hill. 2005. Recent surveys of the
Cocos Buff-banded Rail (Gallirallus philippensis
andrewsi). Report to the Australian Government
Department of the Environment and Heritage. Center
for Resource and Environmental Studies, Australian
National University, Canberra.

Sakaguchi, N. and ISHiHiRA. 1998. Distribution and
population dynamics of the White-breasted Waterhen
Amaurornis phoenicurus in Iriomote Island, Okinawa,
southern Japan. Island Studies in Okinawa 6:25^0.

Skerrett, a. 1994. Seychelles Bird Records Committee.
Birdwatch (Seychelles) 9:15-19.

Steadman, D. W. 2006. Extinction and biogeography of
tropical Pacific birds. University of Chicago Press,
Chicago, Illinois, USA.

Taylor, P. B. 1996. Eamily Rallidae (Rails, gallinules, and
coots). Pages 108-209 in Handbook of birds of the
world. Volume 3 (Hoatzin to Auks) (J. del Hoyo, A.
Elliott, and J. Sargatal, Editors). Lynx Edicions,
Barcelona, Spain.

Yap Branch Statistics Office. 2002. Yap state census
report, 2000 FSM census of population and housing.
Division of Statistics, Department of Economic
Affairs, National Government, Colonia, Yap, Federat-
ed States of Micronesia.

The Wilson Journal of Ornithology 122(4):788-791, 2010

First Reproductive Record of Wilson’s Plover in Baia de Todos os Santos,

Northeastern Brazil

Vitor O. Lunardi'"^ '^ and Regina H. Macedo“

ABSTRACT. — The Wilson’s Plover (Charadrius
wilsonia) is widely distributed along the coast of the
Americas. We present the first reproductive record in
Bai'a de Todos os Santos, Brazil, broadening the
southernmost limit of its breeding area along the
Atlantic Coast to 12 44' S, 38° 45' W. We recorded
a family with a subadult in 2007, and a family with
chicks and a nest in 2008. The female invested more
time in incubation than the male in 96 hrs of diurnal
nest observations. There were 102 interruptions during
incubation due to the approximation of domestic
animals (cattle and horses). The nest was abandoned

' Programa de Pos-Graduay'ao em Ecologia, IB, Campus
Universitario Darcy Ribeiro. Universidade de Brasilia,
Brasilia. DF. Brazil. 70910-900.

^ Departamento de Zoologia, IB. Campus Universitario
Darcy Ribeiro. Universidade de Brasilia, Brasilia, DF,
Brazil. 709 1 0-900.

’Current address: Departamento de Ciencias Animals,
Universidade Federal Rural do .Semi-Arido, Mossoro, RN,
Brazil. 59625-900.

■'Corresponding author; e-mail: lunardi.vitor@gmail.com

9 days after egg laying. An experiment with artificial
nests suggests that 30% of Wilson’s Plover nests may
be destroyed by free-ranging domestic animals in this
Bala. Received II Januaiy 2010. Accepted 5 June 2010.

The Wilson’s Plover {Charadriu.s wilsonia) is a
widely distributed Charadriidae associated with
coastal habitats in the Americas with resident
populations in numerous sites. It breeds on the
Atlantic Coast, between the eastern USA and
northeastern Brazil, in the Caribbean Islands, and
on the Pacific Coast, between Mexico and
northwestern Peru (Corbat and Bergstrom 2000).
The species exhibits plumage differences through-
out its range, which attracted the attention of
systematists, leading to subdivision of the species
into five subspecies (Wiersma 1996, Corbat and
Bergstrom 2000, Grantsau and Lima 2008).
Wilson’s Plover is seemingly monogamous (Cor-
bat and Bergstrom 2000), and both males and

SIIOR r COMMUNIC A HONS

789

females are eapable of performing ineubalion
(Bergstrom 1981).

Historically, the breeding range of Wilson’s
Plover has been contracting and reproductive
populations in the current northernmost limit are
threatened, mainly due to destruction of breeding
areas and interference during reproduction and
while resting. However, little is known about
threats outside this region (Corbat and Bergstrom
2000).

Records of reproduction and information about
populations of Wilson's Plover in the Southern
Hemisphere are scarce and restricted to a few sites
(Tovar 1968, Rodrigues et al. 1996, Wiersma
1996, Corbat and Bergstrom 2000, Grantsau and
Lima 2008). We describe: (1) the first breeding
record of Wilson’s Plover in the southernmost
limit of its breeding range on the Atlantic Coast
(Bata de Todos os Santos, Brazil), and (2) features
of the species’ reproductive behavior and poten-
tial threats to the study population.

METHODS

Monitoring of Nests and Families. — We
recorded breeding of Wilson’s Plover in the
coastal region of Saubara, Bahia, western coast
of Baia de Todos os Santos, Brazil (12° 44' S,
38° 45' W). This region is characterized by a
complex of salt marshes (restingas), mangroves
(Rhizophora, Laguncularia, Avicennia), suprati-
dal salt flats (apicum), and intertidal areas. The
climate is humid tropical with distinct dry (Sep-
Feb) and rainy (Mar-Aug) seasons. We com-
pared photos of individuals from the study site
for species identification with specimens depos-
ited in the Museu de Zoologia da Universidade
de Sao Paulo-MZUSP (specimens: 41395,
41396, 42745, 8(3034, and 80035). We based
our comparisons on the Wilson’s Plover sub-
species identification key to the Atlantic Coast
developed by R. Grantsau (Grantsau and Lima
2008).

OBSERVATIONS

The first evidence of breeding of Wilson’s
Plover in the region was the record of a family,
comprising a pair with a subadult, foraging in the
intertidal zone on 17 December 2007. The
following year, we found a Wilson’s Plover nest
with two eggs on 13 August 2008 in a restinga
area. The next morning, taking advantage of the
absence of the pair from the nest, we took nest and
egg measurements and recorded a third egg

FIG. 1. Ne.sl and chick of Wilson's Plover at Bai'a de
Todos os Santos. Brazil. Photographs by Vitor O. Lunardi.

(Fig. I). We performed 96 hrs of focal observa-
tion of the nest to examine parental investment in
incubation, between 0600 and 1800 hrs. during
the first 8 days of incubation ( 14 to 21 Aug 2008).
Observations were conducted from a fixed point
inside of Ihe mangrove, 26 m from the nest, to
avoid investigator disturbance. The nest with the
eggs was abandoned on 22 August 2008, probably
due to the intense disturbance by domestic
animals (cattle and horses). The nest with eggs
remained intact until 30 August 2008, when they
were found destroyed.

On 27 August 2008 we found another family of
Wilson’s Plover with three chicks in an apicum
area. Each chick (Fig. 1 ) was measured, weighed,
and marked with a metal band (CEMAVE/Brazil).
This family remained for the next 7 days at the

790

THE WILSON JOURNAL OL ORNITHOLOGY • Vot. 122, No. 4, December 2010

Days

FIG. 2. Number of artificial nests destroyed, predated,
and intact in a restinga area at Bai'a de Todos os Santos,
Brazil, during 25 days.

same site with the aggregated chicks always
accompanied by the adults.

Experiment to Verify Predation Intensity. — We
observed Wilson’s Plover pairs defending territo-
ries in a restinga area between May and October
2007 and 2008, also used as pasture by free-
ranging cattle and horses from the local commu-
nity. We conducted an experiment with artificial
nests to estimate nest loss considering these
animals may destroy nests. We built 30 artificial
nests (cavities in the soil similar in shape and size
to Wikson’s Plover nests) in January 2009, placed
in a series, 30 m apart, along a strip of restinga.
We added three Common Quail (Coturnix cotur-
nix) eggs to each artificial nest. We made three
daily visits to each nest {0600, 1200, and 1800
hrs) over 25 days (approximate incubation time
for the species) to record nests that were: ( 1 )
intact, (2) destroyed (with at least 1 destroyed egg
and tracks of domestic animals), and (3) predated
(absence or destruction of at least 1 egg; no signs
of domestic animals).

Nine of 30 artificial nests (Fig. 2) were
destroyed with 77.7% of occurrences during the
day (()6()()-180() hrs). Fifteen nests were predated
with 66.7% of the.se events at night (18()()-()6()()
hrs).

Monitoring of Nests and Families. — The only
nest found consisted of a 25-mm cavity in sandy
soil with a diameter of 102 mm and fragments of
dry vegetation around the eggs. It was next to
horse feces and a shrubby false buttonweed
(Spermaeoce verticillala L., Rubiaceae; height
17.5 cm) surrounded by grasses (Fig. 1). The

three eggs measured: length = 34.9, 34.8, and
34.2 mm; width = 25.2, 25.0, and 24.4 mm; and
mass = 14.0, 14.0, and 13.5 g. The eggshells were
colored light cream with dark brown stains over
the entire surface, but with a higher concentration
on the blunt end.

The female invested 53.3 hrs (93.3%) in
incubation during the observation period while
the male invested 3.8 hrs (6.7%). The female was
observed incubating between 0600 and 1733 hrs,
and the male after 1714 hrs. The female
interrupted incubation 125 times (T ± SD =
1.29 ± 1.41 /hr, n = 96) of which 102 (81.6%)
were caused by approach of domestic animals,
mainly cattle. After leaving the nest, the female
often exhibited alarm and distraction behaviors:
squatting, head-up (followed by vocalization),
mock-brooding and/or broken-wing. In 54.3% of
these interruptions the male approached the nest,
exhibiting head-up and crouch-run display (be-
havioral descriptions in Bergstrom 1988b). The
female was reluctant to return to the nest after
disturbances.

The measures obtained for the three chicks
were: right tarsus = 23.5, 23.3, and 22.1 mm;
CLilmen = 17.3, 16.8, and 16.1 mm; and mass =
17.0, 16.5, and 15.5 g. The beaks were black; legs
and feet grayish; plumage upperparts were
brownish and black in a .scalloped pattern, and
underparts and neck were white. The age and
gender of the chicks could not be assigned.

DISCUSSION

We present the first breeding record of Wilson’s
Plover in Baia de Todos os Santos, Brazil. This is
the southernmo.st record on the Atlantic Coast for
the species, representing an extension of their
breeding range (Corbat and Bergstrom 2000,
Grantsau and Lima 2008). Grantsau and Lima
(2008) recorded breeding for Wilson’s Plover (C.
wilsonia brasiliensis) in Mangue Seco, Brazil,
~23() km north of Bala de Todos os Santos. The
plumage of the males and females in our study area
was similar to that of C. vv. brasiliensis deposited at
MZUSP, and suggests our records correspond to
this subspecies.

Bergstrom (1988a) reported Wilson’s Plover
nests mainly near vegetation and occasionally near
objects (e.g., cow manure, stones), emphasizing
that both may function as a protection barrier
against the wind. Objects and vegetation provided
protection for the nest that we found againsl

SHORT COMMUNICATIONS

791

easterly and southeasterly winds, which are
predominant in August, when we found the nest.

The number, dimensions, and color of the eggs
and size of the chicks we report are similar to
those previously published for the species (Ro-
drigues et al. 1996, Corbat and Bergstrom 2000,
Grantsau and Lima 2008). The female invested
more time in diurnal incubation than the male.
However, the male may have invested in noctur-
nal incubation, since they initiated this activity at
the end of the day. Previous observations for the
species indicate males incubate mainly at night
(Thibault and McNeil 1995).

The experiment with artificial nests revealed
that, in addition to natural threats (predators), use
of coastal areas for grazing is another hazard to
the reproductive success of Wilson’s Plover in
Baia de Todos os Santos. Trampling by roaming
animals was previously considered a potential
threat for Wilson’s Plover eggs and recent hatched
nestlings at Ilha do Curupu, Maranhao, Brazil
(Rodrigues et al. 1996). Plovers leave their nests
when disturbed and are extremely reluctant to
return when intruders are present near nest sites.
The constant interruption of incubation by inter-
ference may lead to inadequate egg temperature
regulation, and exposing eggs to predation and
overheating (Corbat and Bergstrom 2000). The
female often left its nest when disturbed by free-
ranging cattle and horses. Both females and males
invested actively in predator avoidance behaviors.
We assume the intense disturbance was the
primary cause of nest and egg desertion.

Our observations confirm Baia de Todos os
Santos as a breeding area of Wilson’s Plover,
extending the limit of its breeding range. Use of
coastal areas for grazing is critical to consider in
any project directed at conserving this population.

ACKNOWLEDGMENTS

This research was supported by fellowships from
Conselho Nacional de Desenvolvimeiilo Cienti'fico e
Tecnologico and Coordena^'ao de Aperfeiij'oamento de
Pessoal de Ni'vel Superior. We are grateful to CEMAVE/
ICMBio for authorization to capture and band the birds. We
thank L. F. Silveira for authorization and access to the
collection of MZUSP, J. G. Jardirn for taxonomic
identification of S. verticillata. and D. G. Lunardi for
revision of an earlier version of the manuscript. We thank
two anonymous referees for helpful and constructive
comments on the manuscript.

LITERATURE CITED

Bergstrom, P. W. 1981. Male incubation in Wilson's
Plover {Cha rad fills wilsonia). Auk 98:835-838.
Bergstrom, P. W. 1988a. Breeding biology of Wilson's
Plovers. Wilson Bulletin 100:25-35.

Bergstrom, P. W. 1988b. Breeding displays and vocaliza-
tions of Wilson’s Plovers. Wilson Bulletin 100:36-49.
Corbat, C. A. and P. W. Bergstrom. 2000. Wilson’s
Plover (Charadriits wilsonia). The birds of North
America. Number 516.

Grantsau, R. and P. C. Lima. 2008. Uma nova sub-
especie de Charadriits wilsonia (Aves, Charadrii-
formes) para o Brasil. Atualidades Ornitoloaicas
142:4-5.

Rodrigues, A. A. F., D. C. Oren, and A. T. L. Lopes.
1996. New data on breeding Wilson's Plovers
Charadrius wilsonia in Brazil. Wader Study Group
Bulletin 81:80-81.

Thibault, M. and R. McNeii.. 1995. Day- and night-time
parental investment by incubating Wilson's Plovers in
a tropical environment. Canadian Journal of Zoology
73:879-886.

Tovar. H. 1968. Areas de reproduccion y distribucion de
las aves marinas en el litoral peruano. Boleti'n del
Instituto del Mar del Peru 1 :523-526.

WiERSMA, P. 1996. Species accounts. Family Charadriidae
(plovers). Pages 410^42 in Handbook of the birds of
the world. Volume 3 (Hoatzins to Auks) (.1. del Hoyo,
A. Elliott, and J. Sargatal. Editors). Lynx Edicions.
Barcelona, Spain.

792

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, December 20 10

The Wilson Journal of Ornithology 1 22(4):792-795, 2010

Hatching Synchrony, Green Branch Collecting, and Prey Use by Nesting

Harpy Eagles {Hcirpia harpyja)

Adrian S. Seymour, Graham Hatherley," Francisco Javier Contreras,-^

James Aldred,"* and Fergus Beeley”^

ABSTRACT. — We observed an occupied Harpy
Eagle (Harpia harpyja) nest over three separate
periods in eastern Venezuela. Both eggs in the clutch
hatched on the same day, and two nestlings competed
in the nest for 14 days before one succumbed. The
female adult collected green branches 45 times over
60 days of observation. Green branch deliveries were
positively associated with prey deliveries and our
observations best support the ‘nest sanitation by
covering prey’ hypothesis for the adaptive significance
of green branch collecting. Prey delivery rate to the
nest averaged one delivery every 2.4, 2.1, and 3.7 days
in the three observation periods. Three-toed sloths
(Bradypus tridactylus) and wedge-capped capuchins
(Cehus olivaceiis) were the most common prey items
brought to the nest. Received 14 April 2010. Accepted
23 July 2010.

The Harpy Eagle (Harpia harpyja) is one of the
largest and most emblematic eagles in the world,
and is a species of considerable conservation
concern throughout its range in Central and South
America (Vargas et al. 2006). However, remark-
ably little is known of its ecology and behavior.
The main reasons for this lack of knowledge are
that Harpy Eagles are hard to locate and observe
in the rainforest habitats in which they live. Much
of what is known about their behavior comes from
observations during nesting and chick rearing
when their movements become centered on a nest
(summarized by Schulenberg 2009). However,
becau.se of the difficulty in ob.serving nests (they
are usually positioned 20 m or more up in the
canopy), only a few studies have been made and
behavior at the nest is still poorly understood. For

' 13a Northumberland Road. Bristol. B.S6 7AU. UK.
’Emirates Natural History Unit. 59 Cotham Hill, BS6
6JR. UK.

'Motel Taguapire. Via Rio Grande. El Palmar, Estado
Bolivar, Venezuela.

■‘Canopy Acce.ss Ltd.. 81a Hill Road, Clevedon, BS2I
7F’L. UK.

’’98 Coombe Dale. Bristol, B.S9 2JE, UK.
‘’Corresponding author;
e-mail: aseymour_uk@yahoo.co.uk

example, Harpy Eagles have been observed to
collect green branches (i.e., branches with green
foliage) during the nesting period (Rettig 1978), a
behavior seen in many species of raptors. The
adaptive signifance of this behavior remains
unknown. Several hypotheses have been put
forward including: (1) nest sanitation by repelling
iivsects with the volatile components of leaf
chemistry, (2) nest sanitation by covering prey
remains, (3) advertisement of nest occupancy, (4)
maintenance of humidity in the nest, and (5)
provision of shade for the nestlings (Wimberger
1984).

Little is known about factors affecting prey use
by Harpy Eagles during nesting, and only limited
data are available (summarized by Schulenberg
2009). Harpy Eagles are distributed over a large
geographic area, and the composition of prey
species varies considerably in different parts of its
range (Emmons and Feer 1997). It is probable that
regional variations in relative abundance and ease
of catching prey also occur.

Our objectives were to: ( I ) use quantitative and
qualitative ob.servations of Harpy Eagle nesting
behavior to explore the adaptive significance of
green branch collecting, (2) report prey use by a
pair of nesting Haipy Eagles in a fragmented
forest habitat in Venezuela, and (3) report an
unusual case of hatching synchrony and extended
sibling competition.

OBSERVATIONS

We observed an active Harpy Eagle nest for
60 days over three separate periods (2 May-2 Jun
2009, 5 Aug-3 Sep 2009, 8 Jan- 10 Feb 2010) in the
Sierra Imataca of Bolivar State, ea.stern Venezuela
(07° 56.926' N, 61 43.244' W). We monitored the
nest from a blind on an arboreal platform 25 m
from the ground ~6() m from the nest tree, a tall
(>35 m) silk-cotton or ceiba tree (Ceiha pentaii-
dra). The nest was also monitored using a small
remote camera fixed to a branch ~3 m above the
nest platform.

SI lORT COMMLINICA'I IONS

793

Green Branch Collecting. — Harpy Eagles col-
lecled branches with green leaves on 46 separate
occasions in the 60 days the nest was observed.
The lemale collected the green branches in all but
one occasion. Most often the female would
anange the green branches over prey remains.
She did not cover the chick(s) with green branches
and hypothesis # 5 is not supported by our
observations. The ambient humidity of the rain-
forest habitat is high, suggesting that hypothesis #
4 is unlikely to explain the main function of green
branches in the nest. One or more green branch
deliveries appeared more likely to occur on days
when prey was also delivered, but this association
was not statistically significant (2X2 contingen-
cy table: y- = 2.39, df = 1, P = 0.12). However,
there was a period of 3 consecutive days during an
exceptionally long fasting period when the
female’s green branch collecting behavior was
atypical in that ( 1 ) it was much more frequent
than usual (branch collecting observed 13 times in
3 days compared to the expected frequency of 5.7
based on the average green branch delivery rate
on ‘normal’ delivery days), and (2) the female
jettisoned 38% of these green branches before
reaching the nest, which was not normally done. If
these 3 days of unusual behavior are omitted from
the statistical analysis, the association between
green branch collecting and prey delivery be-
comes significant (2X2 contingency table: x~ —
3.94, df = 1, P < 0.047). These observations
support hypothesis # 2. We observed flies and
other winged insects frequently flying around the
nest. The female and the remaining chick
responded to flying insects by vigorously shaking
their heads or snapping with their beaks. The
female or chick would react in this way to flying
insects at least once on 9 of the 21 observation
days (i.e., 43% of days) in the first observation
period (chick 0-21 days of age). This contrasts
with the second observation period (chick 95-
112 days of age) when they reacted to flying
insects only once in 15 observation days (i.e., 7%
of days). We saw the female peck at crawling
insects in the nest at least once in 10 of the 21
observation days (i.e., 48% of days) during the
first observation period, while in the second
period, we observed this behavior on only 3 of
15 ob.servation days (20% of days). Responses to
arthropods and green branch collecting were more
frequent in the first period when the chick was
more vulnerable (average 1.3 branch deliverie.s/
day in the first period vs. 0.9 deliveries/day in the

second observation period). The.se observations
support both (#’s 1 and 2) nest sanitation
hypotheses for green branch collection by Harpy
Eagles. Most of the green branches were from the
C. pentandra tree which supported the nest, but
some were also taken from a nearby fig (Ficus
spp.) tree. Branches were taken from another tree
species on only one occasion.

Prey Delivery. — We recorded 23 prey deliver-
ies in the 60 days the nest was monitored. The
male Harpy Eagle delivered 12 prey items to the
nest in 29 days (average of 1 delivery every
2.4 days) in the first observation period. He
delivered five pale-throated three-toed sloths
(Bradypus tridcicty’lus), four wedge-capped capu-
chins (Cehiis oiivaceus), and three unidentifiable
prey items. Seven prey deliveries were recorded in
15 days (average 1 delivery every 2.1 days) in the
second observation period, four by the male and
three by the female. Three of the seven prey
delivered were capuchin monkeys, two were
sloths, and two could not be identified. Eour prey
deliveries were observed in 15 days in the third
observation period (average 1 delivery every
3.7 days). Two deliveries were made by the male,
and the other two were made by the female. It was
not possible to identify these prey items as the
only video recorded of these deliveries was from
positions that did not allow us to see into the nest.

Synchronous Hatching. — Both eggs in the
Harpy Eagle nest hatched on the same day (12
May 2009). Two chicks were seen in the nest up
to 14 days after hatching, at times seen struggling
to climb on top of each other. Only one of the
chicks was ever observed being fed by the female
at a time, although we could not tell if it was
always the same chick.

DISCUSSION

Our observations of Harpy Eagles at the nest
suggest that green branch collection is important
for nest sanitation. Harpy Eagle behavior in
response to arthropods (head shaking and peck-
ing) was much more frequent in the first period of
observation than subsequent observation periods
suggesting that arthropods were more problematic
when the chick(s) were newly hatched. This
observation can be explained by a change in
arthropod abundance/activity around the nest
between different observation periods or, alterna-
tively, the female Harpy Eagle may have been
more diligent in removing potentially dangerous
arthropods when the chick was very small and

794

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 4, December 2010

presumably more vulnerable to parasitism or
being bitten and stung.

Little is known about the effect of parasites and
diseases on Harpy Eagle nestlings. Arthropod
ectoparasites such as botflies in the genus
Philonus can have a highly detrimental effect
on survival ot nestlings of many bird species in
the Neotropics (Dudaniec and Kleindorfer 2006)
including small Falconiformes (Delannoy and
Cruz 1991). Stinging and biting arthropods such
as ants and wasps may also pose a threat to a
harpy chick when it is small. Rettig (1978)
observed a nesting female Harpy Eagle to snap
at and kill insects, particularly bees and wasps that
got close to the nest. The female Harpy Eagle in
Rettig’s (1978) study preferred collecting branch-
es of Mora trees rather than those of the C.
pentandra tree in which it was nesting. These
observations suggest high selectivity for particular
kinds of green branches, but we could not tell
whether this selectivity was based on arthropod-
repelling quality or some other factor such as the
ea.se at which the branches could be broken. The
‘nest sanitation by repelling insects with volatile
components of leaf chemistry’ hypothesis cannot
be ruled out. However, there was no consistency
in branch choice between studies and in.sects were
observed on the nest even when green branches
were present (i.e., insect repelling capability of
the green branches did not seem to be strong).

The most common prey delivered to the nest in
Rettig's (1978) study were sloths and capuchin
monkeys. Sloths have been shown to be important
components of Harpy Eagle diets in widely
different parts of its range including Panama
(Touchton et al. 2002) and the Brazilian Amazon
(Galetti and de Carvalho 2000). Red-howler
monkeys (Alnaatta seniculus) were common in
our study area with .several groups living within a
5()0-m radius of the nest. Red-howler monkeys
can be taken by Harpy Eagles (Rettig 1978, Peres
1990, Sherman 1991), and the similar-sized
mantled howler (A. palliata) is the preferred
primate prey of Harpy Eagles on Barro Colorado
Island in Panama (Touchton et al. 2002). Howev-
er, we did not observe any howler remains in the
nest.

Rettig (1978) ob.served a prey delivery rate of
one prey every 3.5 days in the first part of the
nesting period when the male was doing all the
hunting and one prey every 2.5 days later in the
nesting period when both adults were hunting. We
observed the male Harpy Eagle deliver prey on

average once every 2.4 days in the first period,
and both parents delivered at an average rate of
once every 2.1 days in our second observation
period. There are many possible explanations for
this apparently higher delivery rate at our study
site, including differences in prey abundance or
even ease of catching prey. One reason for
increased prey delivery could be that the forest
fragmentation characterizing our study area re-
sulted in large amounts of forest edge habitats
which may be favored by Harpy Eagles hunting
monkeys (Eason 1989).

Many eagle chicks show siblicidal behavior,
but hatching is usually asynchronous (Simmons
1986) making competition between chicks much
more asymmetric. Our observations of synchro-
nous hatching differ from those of Rettig (1978)
who only observed one of two Harpy Eagle eggs in
the clutch hatch. Our obseiwation is unexpected as
more equally-sized competing chicks have a similar
potential to damage each other, which could result
in the reduced survival of both competitors.

More data are needed to quantify Harpy Eagle
behavior and evaluate hypotheses for the adaptive
significances of these behaviors. However, unless
methods for finding and monitoring Harpy Eagle
nests are greatly improved so that large sample
sizes can be obtained for reasonable effort, our
understanding of Harpy Eagle behavior is likely to
remain obscure.

ACKNOWLEDGMENTS

We are grateful for perniis.sion to film the nest event
granted by Centro Nacional Autonomo de la Cinemato-
grafia (CNAC) of Venezuela's Ministry of Culture
(No. CNAC/ PCIA No. 095-2009). We thank Senor Rafael
Garcia and his son Larry for permission to conduct the
study on their land, and for their great hospitality and
kindness. We also thank Alejandro Lamus and Jose Hurtado
for their highly competent, professional assistance and good
companionship, and Emilio Perez for arranging permits.
Finally we thank C. E. Braun and two anonymous referees
for helpful suggestions on early versions of the manuscript.

LITERATURE CITED

Df.i.annoy, C. a. and a. Cruz. 1991. Philomis parasitisjii
and nestling survival of the Puerto Rican Sharp-
shinned Hawk. Pages 93- 1 (H in Bird-parasite inter-
actions: ecology, evolution, and behavior (.1. E. Love
and M. Zuk, Editors). Oxford University Press,
Oxford, United Kingdom,

OtiDANiEC, R. Y. AND S. Kleindorkbr. 2006. Effects of the
parasitic Dies of the genus Philorni.s (DipteraiMusci-
dac) on birds. Emu 106:13-20.

SHORT COMMUNICATIONS

795

Eason, P. OH*-). Harpy Eagle allenipts predation oti adult
howler monkey. Condor 9 1 :469-47().

ENtMONS. L. H. AND F. Feer. 1997. Neotropieal rainforest
mammals: a field guide. Second Edition. University of
Chicago Press, Chicago, Illinois, USA.

G.ALEtTi, M. AND 0. DE Carvai.ho, 2000. Slotlis in the diet
ot a Harpy Eagle nestling in eastern Amazon. Wilson
Bulletin 112:535-536.

Peres, C. A. 1990. A Harpy Eagle successfully captures an
adult male red howler monkey. Wilson Bulletin
102:560-561.

Rettig, N. L. 1978. Breeding behavior of the Harpy Eagle
(HarpUi harpyja). Auk 95:629-643.

Schulenberg, T. S. 2009. Harpy Eagle (Harpia harpyja).
Neotropical birds online (T. S. Schulenberg, Editor).
Cornell Laboratory of Ornithology. Ithaca, New York,
USA. http://neotropical.birds.cornell.edu/portal/specie,s/
over\'iew?p_p_spp=206 1 3

StiERMAN, P. T. 1991. Harpy predation on a red-howler
monkey. Folia Primatologica 56:53-56.

Simmons, R. 1986. Offspring quality and the evolution of
cainism. Ibis 130:339-357.

Touciiton, J. M., Y. HstJ, and A. PAtj.ERONi. 2002.
Foraging ecology of reintroduced captive-bred sub-
adult Harpy Eagles {Harpia harpyja) on Barro
Colorado Island. Panama. Ornitologia Neotropical
13:365-379.

Vargas, J. D., D. Whitacre, R. Mosquera, J. Albu-
querque, R. Piana, j. M. Thiollay, C. Marquez, J.
E. Sancmez, M. Lezama-Lopez, S. Midence, S.
Matola, S. Aguilar, N. Rettig, and T. Sanaioiti.
2006. Status and current distribution of the Harpy
Eagle {Harpia harpyja) in Central and South America.
Orintologia Neotropical 17:39-55.

Wimberger, P. H. 1984. The use of green plant materials in
bird nests to avoid ectoparasites. Auk 101:615-618.

The Wilson Journal of Ornithology 122(4):795-799, 2010

Screaming Cowbird Parasitism of Nests of Solitary Caciques and

Cattle Tyrants

Alejandro G. Di Giacomo,' Bettina Mahler,^ and Juan C. Reboreda^-^

ABSTRACT. — The Screaming Cowbird (Molothrus
rufoaxillaris) is one of the most specialized brood
parasites with only three known hosts: Baywing
(Agelaioicles hadiiis), the main host throughout most
of its range, and two alternative hosts in some areas of
its distribution, Chopi Blackbird (Gnorimopsar chopi)
and Brown-and-yellow Marshbird (Pseiuloleisies vir-
escens). We studied Screaming Cowbird parasitism in
northeast Argentina where this parasite uses Baywings
and Chopi Blackbirds as hosts. We monitored 69 nests
of Baywings, 251 of Chopi Blackbirds, 31 of Solitary
Caciques (Cacicus soli lari us), and 30 of Cattle Tyrants
{Macheiornis rixosa). The frequency of Screaming
Cowbird parasitism on Baywing nests was 80% and
was 46% for Chopi Blackbirds. We recorded one event
of Screaming Cowbird parasitism on one nest of
Solitary Caciques and three events of Screaming
Cowbird parasitism on one nest of Cattle Tyrants. The
identities of parasitic eggs in both hosts were confirmed
by sequencing the mtDNA control region. We propose

' Departamento de Conservacion. Aves Argentinas-Aso-
ciacion Ornitologica del Plata. Matheu 1246, C1249AAB
Buenos Aires, Argentina.

"Departamento de Ecologi'a, Genetica y Evolucion,
Facultad de Ciencias Exactas y Naturales. Universidad
de Buenos Aires, Pabelldn II Ciudad Universitaria,
CI428EGA Buenos Aires, Argentina,

•’Corresponding author;
e-mail: reboreda@ege.fcen.uba.ar

these events of parasitism resulted from recognition
errors by Screaming Cowbird females that regularly
parasitize Baywings and Chopi Blackbirds. The nest of
Solitary Caciques had been frequently visited by a pair
of Baywings before Screaming Cowbird parasitism
occuiTed, and the nest of Cattle Tyrants was near an
active Chopi Blackbird nest that had been previously
parasitized by Screaming Cowbirds. Received 5 Janu-
ary 2010. Accepted 7 April 2010.

The Screaming Cowbird (Molothru.s ntfoo.xil-
lari.s) i.s a specialized brood parasite, which
exclusively u.ses the Baywing (Agelaioides ha-
diiis) as a host throughout most of its range
(Ma.son 1980, Fraga 1998). It also parasitizes the
Brown-and-yellow Marshbird (P.seudoletstes vir-
e.scens) in .some areas (Mermoz and Fernandez
2003) and the Chopi Blackbird (Gttornuojisctr
chopi) (Sick 1985, Di Giacomo 2005). Other
species have been reported as hosts of the
Screaming Cowbird, but there is general agree-
ment that most of these reports were based on
mis-identified Shiny Cowbird (M. honarien.sis)
eggs (Friedmann 1963, Mason 1980, Fraga 1986).

One hypothesis posited to explain host .speci-
ficity in brood parasites is that females imprint on

796

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122. No. 4. December 2010

their foster parents and, once mature, search for
nests of the same species in which to lay their
eggs (Nicolai 1964, Slagsvold and Hansen 2001).
This hypothesis has been directly supported by
e.xperiments with captive brood-parasitic Village
Indigobirds {Vidua chalyheata) bred in captivity
and foster-reared by their normal host or by an
experimental foster species and tested as adults
for host choice (Payne et al. 2000). Indirect
evidence in support of this hypothesis include: ( 1 )
the host specificity of Common Cuckoo (Cucidus
canorus) females, but not males (Marchetti et al.
1998), and (2) the association between host
species and parasite’s mitochondrial, but not
nuclear DNA (Gibbs et al. 2000).

Host imprinting and occurrence of recognition
errors when parasitic females search for host nests
(i.e., to lay eggs in nests of hosts other than the
foster parents) may result in use of new hosts.
This may lead to the formation of host specific
races, if only females imprint on their hosts as by
Common Cuckoos (Gibbs et al. 2000) or in
speciation, if both males and females imprint on
their hosts (i.e., indigobirds \ Vidua spp.]; Payne et
al. 2002, Sorenson et al. 2003).

Friedmann and Kiff (1985) suggested host
imprinting may also occur in cowbirds and
individual females may be host specialists. The
evidence, however, is controversial for Brown-
headed Cowbirds (M. ater). One study showed no
genetic differentiation in mtDNA between cow-
bird chicks raised by two different hosts (Gibbs et
al. 1997), whereas results from another study at
the same site, showed that half of the females laid
in nests of a single host (Alderson et al. 1999).
MtDNA haplotype frequency distributions in
Shiny Cowbirds differed among parasitic chicks
from nests of four hosts (Mahler et al. 2007),
which indicate the occurrence of a non-random
laying behavior by females of this parasite. A
recent study (Mahler et al. 2009) of the host
specialist Screaming Cowbird, showed that fre-
quency distributions of mtDNA haplotypes dif-
fered between cowbird chicks from nests of
Bay wings and Chopi Blackbirds. This result
indicates host choice by females is not random
and female Screaming Cowbirds may preferen-
tially parasitize nests of the foster parents. The
number of mutations between directly related
haplotypes in the latter study was small, which
indicates that haplotype divergence in the popu-
lation is relatively recent in evolutionary terms.
Mahler et al. (2009) suggested a few Screaming

Cowbirds may have switched recently from
Baywings to Chopi Blackbirds, giving rise to a
new race that only uses the latter host.

Successful host switches require females to lay
eggs in nests of a host other than the foster parent
(recognition eiTors), that the new host successfully
rears parasitic females, and these females subse-
quently use the new foster parent as a host. We
present direct evidence through genotyped eggs of
Screaming Cowbird parasitism of two new hosts:
Solitary Cacique {Cacicus solitarius) and Cattle
Tyrant (Machetornis rixoso). Our study was
conducted in an area where Screaming Cowbirds
parasitize Baywings and Chopi Blackbirds. We
propose parasitism of the new hosts resulted from
recognition errors by Screaming Cowbird females
that regularly parasitize Baywings and Chopi
Blackbirds.

METHODS

Study Area. — The study was conducted at
‘Reserva Ecologica El BaguaF in the Province
of Eormosa, Argentina (26° 18' 17.5" S; 58° 49'
51.1" W). This reserve is an open savanna of
3,300 ha in the eastern or humid Chaco region.
Average annual rainfall at the study site is
1,350 mm and mean monthly temperatures vary
from 16.9° C in July to 26.7° C in January.

Data Collection and Analy.si.s. — We monitored
nests of potential hosts of Shiny and Screaming
cowbirds during the 1997-1998 to 2008-2009
breeding seasons. We found 69 nests of Bay-
wings, 251 of Chopi Blackbirds, 31 of Solitary
Caciques, and 30 of Cattle Tyrants. Most nests
were found during construction, laying, and
incubation, and were visited every 2-3 days until
chicks fledged or the nest failed. Individual eggs
were marked with waterproof ink and assigned to
the host or to Shiny or Screaming cowbirds on the
basis of background color, spotting pattern, and
shape (Fraga 1983). We also genetically identified
27 and 31 Screaming Cowbird eggs in nests of
Baywings and Chopi Blackbirds, respectively, by
sequencing the control region of the mtDNA
(Mahler et al. 2009). Screaming Cowbird eggs
from nests of Solitary Caciques and Cattle Tyrants
were also genetically identified. The initial
assignment of Screaming Cowbird eggs on the
basis of color and spotting pattern was confirmed
by genetic analysis in all cases. Frequency of
parasitism was calculated as number of nests with
parasite eggs or chicks divided by total number of

SHORT COMMUNICAl IONS

797

nests toiind, and intensity of parasitism as number
ot parasitic eggs per parasitized nest.

Generic Identificcition of Screoining Cowhird
— We sequenced a 600 base pair fragment of
the mtDNA control region using primers MBO-LI
and MBO-H2 (Mahler et al. 2007). We collected
Screaming Cowbird eggs and artificially incubat-
ed them tor 48 hrs to obtain some embryonic
development. We kept eggs at -20° C until they
were processed. Embryonic tissue was obtained
from the eggs and stored in DMSO buffer for
posterior DNA extraction following a standard
ethanol protocol (Miller et al. 1988). PCR
reactions were performed as described by Mahler
et al. (2009). We sequenced the amplified
products on an Applied Biosystems Model 3100
Genetic Analyzer using ABI Big Dye™ Termi-
nator Chemistry. We compiled the sequences in
Bioedit Version 7.0.5. 3 software (Hall 1999) and
compared them with those deposited in the
EMBL, GenBank, under accession numbers
EU199785-EUI99795 (Mahler et al. 2009).

RESULTS

Screaming Cowbirds regularly parasitized
Baywings and Chopi Blackbirds at our study site.
The frequency of parasitism of Baywing nests was
80% (n = 69 nests) with an intensity of parasitism
of 2.4 ± 0.2 eggs per parasitized nest (x ± SE, n
= 51 nests). Erequency and intensity of parasitism
of Chopi Blackbird nests were 46% (/? = 251
nests) and 3.0 ± 0.2 eggs (/? = 116 nests),
respectively.

Shiny Cowbirds parasitized only one of 31
nests of the Solitary Cacique (3%). We recorded
one event of Screaming Cowbird parasitism in
one nest of Solitary Caciques in December 2007.
Baywings had visited this nest several times
during the previous days and had tried, unsuc-
cessfully, to usurp the nest. Screaming Cowbird
parasitism occurred before host laying and, after
parasitism, the caciques deserted the nest. The egg
was collected and incubated, DNA extracted, and
the control region sequenced. The .sequence of the
mtDNA control region corresponded to Scream-
ing Cowbirds’ haplotype H5 (Mahler et al. 2009).

We also monitored 30 nests of Cattle Tyrants
but Shiny Cowbirds parasitized none of them. We
recorded three events of Screaming Cowbird
parasitism of one Cattle Tyrant nest during laying
in November 2008. This nest was <1 m from an
active Chopi Blackbird nest that was also multiple
parasitized by Screaming Cowbirds. Cattle Ty-

rants incubated the Screaming Cowbird’s eggs of
which one hatched but the chick disappeared from
the nest 48-72 hrs after hatching. We collected
the other two eggs, one of which was rotten and
we could not extract DNA from it. The .sequence
of the mtDNA control region of the other egg
corresponded to Screaming Cowbirds’ haplotype
HI (Mahler et al. 2009).

DISCUSSION

We confirmed Screaming Cowbird parasitism
of two new hosts. Solitary Cacique and Cattle
Tyrant, in an area where this parasite regularly
uses Baywings and Chopi Blackbirds. Solitary
Caciques and Cattle Tyrants were previously
reported as parasitized by Shiny Cowbirds, but
until now they had not been mentioned as
parasitized by Screaming Cowbirds (Ortega
1998). We propose these parasitism events
resulted from recognition errors by female
Screaming Cowbirds that regularly parasitize
Baywings and Chopi Blackbirds. A pair of
Baywings had frequently visited the nest of
Solitary Caciques before Screaming Cowbird
parasitism occuiTed, and the nest of the Cattle
Tyrants was near an active nest of Chopi
Blackbirds that had been previously parasitized
by Screaming Cowbirds.

Baywings seldom build nests but rather exploit
a wide variety of covered nesting sites, including
old nests built by other species and holes in trees
(Eraga 1988, De Marsico et al. 2010). Baywings at
our study site usually use old nests of Greater
Thornbirds {Pliacellodomns ruber) and Little
Thornbirds (P. sihilatrix) but, at times they usurp
and use nests of Solitary Caciques and becards
{Pachyromphus spp.) (Di Giacomo 2005). This
may promote recognition eiTors by Screaming
Cowbird females searching for nests occupied by
its main host. Baywings also u.se holes in trees,
which may have favored colonization of hole-
nesting Chopi Blackbirds.

The first step in the process of colonization of a
new host would be occurrence of recognition
errors when parasitic females search for host nests
(i.e., to lay eggs in nests of a host species other
than their foster parents). We show this type of
error may occur at low frequencies in Screaming
Cowbirds. However, the low frequency of para-
sitism of Solitary Caciques and Cattle Tyrants
also suggests the other steps necessary for
parasitism of a new host (i.e., rearing parasitic
females that subsequently use the same species as

798

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

a host; Payne et al. 2000, 2002; Sorenson et al.
2003) have not yet occurred in these hosts. The
three known hosts of Screaming Cowbirds
{Baywings, Chopi Blackbirds, and Brown-and-
yellow Marshbirds) are cooperative breeders
(Orians et al. 1977, Fraga 1991, Di Giacomo
2005), which suggests competition for food with
nest mates is critical for Screaming Cowbirds.
Thus, this parasite has only been successful in
hosts’ nests where competition is less intense, as
breeding pairs have one or more helpers at the
nest that contribute to chick provisioning (De
Marsico and Reboreda 2008). This hypothesis
requires further investigation.

ACKNOWLEDGMENTS

We thank Alparamis SA and Aves Argentinas/
Asociacion Omitologica del Plata for allowing us to
conduct this study at Reserva El Bagual. We thank Peter
Lowther and Spencer Sealy for helpful comments on a
previous version of this manuscript. BM and JCR are
Research Fellows of CONICET. This study was supported
by Agencia Nacional de Promocion Cientifica y Tecnolo-
gica (Grant 06-00215) and the University of Buenos Aires
(Grant XI 84).

LITERATURE CITED

Alderson, G. W., H. L. Gibbs, and S. G. Sealy. 1999.
Determining the reproductive behaviour of individual
Brown-headed Cowbirds using microsatellite DNA
markers. Animal Behaviour 58:895-905.

De Marsico, M. C. and J. C. Reboreda. 2008. Dif-
ferential reproductive success favours strong host
preferences in a highly specialized brood parasite.
Proceedings of the Royal Society, Series B 275:2499-
2506.

De Marsico, M. C., B. Mahler, and J. C. Reboreda.
2010. Reproductive success and nestling growth of
Baywings parasitized by Screaming and Shiny cow-
birds. Wifson Journal of Ornithology 122:417-431.

Di Giacomo, A. G. 2005. Aves de la Reserva El Bagual.
Pages 201-465 in Historia natural y paisaje de la
Reserva El Bagual, Provincia de Formosa, Argentina
(A. G. Di Giacomo and S. F. Krapovickas, Editors).
Temas de Naturaleza y Conservacion 4. Aves
Argentina.s/Asociacion Omitologica del Plata, Buenos
Aires.

Fraoa, R. M. 1983. The eggs of the parasitic Screaming
Cowbird (Molothrus nifoa.xilloris) and its host, the
Bay-winged Cowbird (M. hculiii.s): is there evidence
for mimicry? Journal fiir Ornitologie 124:187-193.
Fraga. R. M. 1986. The Bay-winged Cowbird (Molothru.s
hcuHns) and its brood parasites: interactions, coevolu-
tion and comparative efficiency. Thesis. University of
California at Santa Barbara. California, USA.

Fraga, R. M. 1988. Nest sites and breeding success of Bay-
winged Cowbirds (Molothrus hadius). Journal liir
Ornithologie 129:175-183.

Fraga, R. M. 1991. The social system of a communal
breeder, the Bay-winged Cowbird .Molothrus hadius.
Ethology 89:195-210.

Fraga, R. M, 1998. Interactions of the parasitic Screaming
and Shiny cowbirds (Molothrus rufoa.xillaris and M.
bonariensis) with a shared host, the Bay-winged
Cowbird (Molothrus hadius). Pages 173-193 in
Parasitic cowbirds and their hosts. Studies in coevo-
lution (S. I. Rothstein and S. K. Robinson, Editors).
Oxford University Press, New York, USA.

Friedmann, H. 1963. Host relations of the parasitic
cowbirds. U.S. National Museum Bulletin 233.

Friedmann, H. and L. F. Kief. 1985. The parasitic
cowbirds and their hosts. Proceedings of the Western
Foundation of Vertebrate Zoology 2:225-304.

Gibbs, H. L, P. Miller, G. Alderson, and S. G. Sealy.
1997. Genetic analysis of Brown-headed Cowbirds
Molothrus ater raised by different hosts: data from
mtDNA and microsatellite DNA markers. Molecular
Ecology 6:189-193.

Gibbs, H. L., K. Marchetti, M. D. Sorenson, M. de L.
Brooke. N. B. Davies, and H. Nakamura. 2000.
Genetic evidence for female host-specific races of the
Common Cuckoo. Nature 407:183-186.

Hall, T. A. 1999. BioEdit: a user-friendly biological
sequence alignment editor and analysis program for
Windows 95/98/NT. Nucleic Acids Symposium Series
41:95-98.

Mahler, B., V. A. Confalonieri, I. J. Lovette, and J. C.
Reboreda. 2007. Partial host fidelity in nest selection
by the Shiny Cowbird (Molothrus bonariensis), a
highly generalist avian brood parasite. Journal of
Evolutionary Biology 20:1918-1923.

Mahler. B., Y. Sarquis Adamson, A. G. Di Giacomo, V.
A. Confalonieri, and J. C. Reboreda. 2009.
Utilization of a new host in the host-specialist brood
parasite Molothrus rufoaxillaris: host switch or host
acquisition? Behavioral Ecology and Sociobiology
63:1603-1608.

Marchetti, K., H. L. Gibbs, and H. Nakamura. 1998.
Host-race formation in the Common Cuckoo. Science
282:471^72.

Mason, P. 1980. Ecological and evolutionary aspects of
host selection in cowbirds. Thesis. University of
Texas, Austin, USA.

Mermoz, M. E. and G. j. Eernandez. 2003. Breeding success
of a specialist brood parasite, the Screaming Cowbird,
parasitizing an alternative host. Condor 105:63-72.

Miller, S. A., D. D. Dykes, and H. F. Polesky. 1988. A
simple salting out procedure for extracting DNA from
human nucleated cells. Nucleic Acids Research
16:1215.

Nicolai, J. 1964. Der brutparasitismus dcr Viduinae als
ethologisches problem. Pragungsphanomene als fak-
toren der rassen- und artbildung. Zeitschrift fiir
Ticrpsychologie 21:1 29-204.

Orians, G. H., C. E. Orians, and K. J. Orians. 1977.
Helpers al the nest in some Argentine blackbirds.
Pages 137-151 in Evolutionary ecology (B. Stone-
hou.se and C. Perrins, Editors). Macmillan Press,
London, United Kingdom.

SHORT COMMUNICATIONS

799

Ortega, C. P. 1998. Cowbirds and other brood parasites.

University of Arizona Press, Tucson, USA.

Payne, R. B., L. L. Payne, J. L. Woods, and M. D.
Sorenson. 2000. Imprinting and the origin of para-
site-host species associations in brood-parasitic
indigobirds. Vidua chalxbeata. Animal Behaviour
59:69-81.

Payne, R., B. K. Hustler, R. Stjernstedt, K. M. Sefc,
AND M. D. Sorenson. 2002. Behavioural and genetic
evidence of a recent population switch to a novel host

species in brood-parasitic indigobirds Vidua chaly-
heaia. Ibis 144:373-38.5.

Sick, H. 1985. Ornitologia brasileira: uma introduc9ao.
University of Brasilia, Brazil.

Slagsvold, T. and B. T. Hansen. 2001. Sexual imprinting
and the origin of obligate brood parasitism in birds.
American Naturalist 158:354-367.

Sorenson, M. D., K. M. Sefc, and R. B. Payne. 2003.
Speciation by host switch in brood parasitic indigo-
birds. Nature 424:928-931.

The Wilson Journal of Ornithology 1 22(4):799-803, 2010

Snakes are Important Nest Predators of Dickcissels in an

Agricultural Landscape

Page E. Klug,''“"^ L. LaReesa Wolfenbarger,‘ and John P. McCarty'

ABSTRACT. — We used video cameras to monitor 33
Dickcissel (Spiza americana) nests during 2003-2004
in the highly-fragmented, agricultural ecosystem of
eastern Nebraska and western Iowa. Nine nests fledged
young, 20 were completely depredated, three were
partially depredated, and one was abandoned due to
ants. Nine snakes, six small mammals, six common
raccoons (Procyon lotor), two Brown-headed Cowbirds
(Molothrus ater), and one American mink (Neovison
vison) were documented as nest predators. These results
suggest a diversity of predators is responsible for
depredation of Dickcissel nests with snake predation
being an important cause of nest failure. Received 2]
December 2009. Accepted 24 May 2010.

Nest predation is the leading cause of repro-
ductive failure for many bird species across
numerous regions and habitats (Martin 1993),
and may be an important factor in avian
population declines (Heske et al. 2001 ). Grassland
birds are of special concern becau.se they have
undergone declines greater than any group of
birds in North America (Knopf 1994) and have
shown decreasing reproductive rates with in-
creased habitat fragmentation (Herkert et al.
2003). Nearly 80% of grasslands in the Great
Plains of North America have been converted to
agricultural u.se, and <4% of the tallgrass prairie

' Department of Biology, University of Nebraska at
Omaha, Omaha, NE 68182, USA.

"Current address: Department of Biology. University of
Notre Dame, Notre Dame, IN 46556, USA.

■’ Corresponding author; e-mail: pklug@nd.edu

remains (Samson and Knopf 1994). Grassland
birds nesting in the remaining grasslands often
experience low reproductive success due to
landscape features in agroecosystems that attract
a diversity of nest predators (Bergin et al. 2000).

Previous studies have identified predators of
grassland bird nests, but more studies are
necessary to understand how predator communi-
ties differ among regions and what factors may
contribute to these differences (Weatherhead and
Blouin-Demers 2004). At least two video studies
of nesting birds that have occun'ed in North
American grasslands were in areas that could be
considered tallgrass prairie, but were conducted at
the edge of the historical tallgrass prairie range.
Thompson and Burhans (2003) documented
predators of shrub-nesting birds in old fields in
Missouri, and Renfrew and Ribic (2003) docu-
mented predators of five grassland species in
grazed pastures in the Driftless Area of Wiscon-
sin; both studies were conducted along the
ecotone with eastern deciduous forests. Pietz and
Granfors (2000) studied nest predation on 10
species of grassland pas,serines and Grant et al.
(2006) studied two passerine species in mixed-
grass prairie in North Dakota. To date, no studies
have been conducted in the highly fragmented,
agricultural ecosystem in the core of the former
tallgrass prairie biome. Our objective was to
identify predators of Dickcis.sel (Spiz.a americano)
nests within small grassland patches interspersed
within intensive row crop agriculture. Dickcissels
are of con.servation concern and have shown

800

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 4. December 2010

TABLE 1. Nest predators recorded on video with corresponding
and nestling stages.

number of Dickcissel nests depredated at incubation

Predator

Incubation

Nestling

# Events

Snakes

Fox snake (Elaphe viilpinci)

3

2

5

Plains garter snake (Thanmophis rcidi.x)

0

2

2

Eastern yellowbelly racer {Coluber constrictor flavivenrris)

0

1

1

Bullsnake (Pituophis catenifer)

1

0

1

Small mammals

Thirteen-lined ground squirrel (Spermophilus tridecemlineatiis)

2

1

3

Franklin’s ground squirrel {S. frtmklinii)

0

1

1

Cricetid rodent (Peromyscus spp.)

2

0

2

Mid-sized mammals

Common raccoon {Procyon lotor)

4

2

6

Other

Brown-headed cowbird (Molothrus ater)

1

1

2

American mink (Neorison vison)

0

1

1

Totals

13

1 1

24

population declines similar to other grassland
passerines, but their predator community has not
been documented on video. Dickcissels are an
ideal model for understanding how nest predator
communities vary among regions, landscapes, or
habitat types because they inhabit marginal
grasslands in extremely fragmented landscapes
as well as relatively unfragmented landscapes.

METHODS

We monitored Dickcissel nests at DeSoto
National Wildlife Refuge, Boyer Chute National
Wildlife Refuge, Allwine Prairie Preserve, Cum-
ing City Cemetery and Nature Preserve, and a
privately owned Conservation Reserve Program
(CRP) parcel in eastern Nebraska and western
Iowa (King et al. 2009). The grasslands were
managed with prescribed burning but no livestock
grazing. Vegetation in the grasslands was either
dominated by native warm-sea.son grasses or by
exotic cool-season grasses with variation in forb
density (Klug 2005). We video-documented
predators of Dickcissel nests in 15 grassland
fragments ranging from 4 to 45 ha (.v ± SE = 20.3
± 3.1). The landscape within 1,600 m of
grassland fragments included six major land cover
types based on the evaluation of orthophotographs
in ArcMap CIS (scale of 1:1500) by Klug et al.
(2009). We classified the landscape as an
agroecosystem becau.se 20-64% (.v ± SE = 52.5
± 3.9) of the surrounding land was in crop

production (e.g., corn or soybeans). Grasslands
covered 2-43% (.v ± SE = 17.8 ± 2.5) and trees
6^2% (.V ± SE = 15.7 ± 2.5) of the landscape.
Human development covered 0.5-22% (.v ± SE =
4.0 ± 1 .7), and roads <2% (T ± SE = 1 .2 ± 0. 1 ),
whereas wetlands and water bodies covered 1-
14% (T ± SE = 5.4 ± 1.0) of the landscape.

We used 24-hr time-lapse video systems to
monitor nests and identify predators removing or
consuming eggs and nestlings. We constructed six
battery-powered video systems, which included a
camera (V-1214-IR Weatherproof Bullet Camera
with infrared illumination, Marshall Electronics,
El Segundo, CA, USA) attached to a time-lapse
video recorder (SSC-960 Time Lapse VCR,
Samsung Electronics America Inc., Ridgefield
Park, NJ, USA) by 8.5 m of cable (buried under
vegetation and litter) and powered by a sealed
lead acid battery connected by a power converter
(375w PowerVerter, Tripp-Lite World Headquar-
ters, Chicago, IE, USA). The camera was attached
to a wood dowel and placed within 10 cm of the
nest so the camera was completely obscured by
vegetation, but vegetation did not completely
obscure the field of view. We minimized
disturbance at the nest by setting up the system
in <20 min between lOOO and 1600 CST, in dry,
moderate temperatures. We set up systems during
both incubation and nestling stages, but only after
females had incubated for 3-5 days to minimize
abandonment. Nest contents were checked re-

SHORT COMMUNICA I IONS

801

TABLE 2. Nest predators ot grassland passerines documented on videt)
and Wisconsin.

in Nebraska/Iowa, North Dakota, Missouri,

Nebraska

Predator

North

Dakota^

North

Dakota'

Missouri*^ Wisconsin'

Snake (Elaphe rulpina, E. ohsoleia, Thaninophis spp., Coluber constrictor.

Lanipropeltis getnla, L. calligaster, Pitnophis catenifer)

9

2

33 3

Common raccoon (Procyon lotor)

6

1

4 8

Squirrel (Sperniophiliis tridecemUneatus. S. franklinii, Scinrns niger)

4

13

1 1

1 4

Brown-headed Cowbird (Molotlirns ater)

2

2

5

1 1

Mouse or vole (Peroniyscns spp., Zapns spp., Microtns spp.)

2

2

3

2

Long-tailed weasel/American mink (Mnstela frenata, Neovison vison)

1

1

1

American badger (Taxidea taxiis)

2

3

7

Bird ot prey (Biiteo spp., Circus cyaneiis, Tyto alba)

2

1

3 2

Red fox/coyote/domestic dog (Vnlpes vitlpes, Canis latrans. C. lupus fatniiiaris)

2

1

Domestic cat (Pells catiis)

1

Striped skunk (Mephitis mephitis)

1

Virginia opossum (Didelphis virginiana)

]

White-tailed deer (Odocoileus virginianus)

2

7

Unidentified

1

7

1

Totals

24

27

30

46 24

Pietz and Grantors (2000).
Grant et al. (2006).

Thompson and Burhans (200.t).
Renfrew and Ribic (2003).

motely by attaching a portable monitor to the
system every 24 hrs given that the tape ran out
after 24 hrs. We archived the tape to identify the
predator using a high-resolution monitor and slow
motion function if the contents differed from the
previous day. We recorded time, nest stage, nest
contents, and contents removed for each depre-
dation event.

We estimated daily survival rates (DSR) for
nests using Program MARK with the design
matrix tools and the logit-link function (Dinsmore
et al. 2002), and calculated variance using the
Delta method (Powell 2007). We also estimated
DSR tor nests without video systems to better
understand the influence of the equipment on nest
survival. Means ± SE are presented.

RESULTS

We video-monitored 33 nests over 2 years
(2003 = 15 nests, 2004 = 18 nests) for 287
observation days during incubation (// = 148 days)
and nestling (// = 139 days) stages. Nests were
monitored an average of 12.9 ± 1.0 days. We
placed cameras at 19 nests that recorded both
incubation and nestling stages, 1 1 nests during the
incubation stage only, and three nests during the
nestling stage only. The DSR of nests monitored
with video was 0.946 ± 0.01 for an overall

survival of 0.333 ± 0.07 if we extrapolated to a
20-day nesting cycle (i.e., the average length of a
Dickcissel nesting cycle). The probability of
predation for nests with cameras was similar to
nests not monitored with video systems (DSR =
0.934 ± 0.005, overall survival rate = 0.255 ±
0.03). Twenty of the nests monitored completely
failed due to depredation, three nests were
partially depredated before eventual Hedging of
young, nine nests fledged young with no depre-
dation, and one nest was abandoned after an ant
infestation.

Nine snakes, six small mammals, six common
raccoons {Procyon Infor), two Brown-headed
Cowbirds (Molotlirus ater), and one American
mink (Neovison vison) were documented as nest
predators (Table I ). We documented one instance
of a multiple predation at one nest (i.e., a Brown-
headed Cowbird removed three Dickcissel eggs
prior to a thirteen-lined ground squirrel \Spernw-
philiis tridecewUneatus] consuming the remaining
2 Brown-headed Cowbird and 1 Dickcissel egg)".
We documented partial predation by a fox snake
(Elaphe vulpiiia) late in the nesting cycle at two
nests, which caused the surviving nestlings to
fledge prematurely. A fox snake at another nest
was documented consuming one egg and leaving
three, which later hatched and Hedged. We

802

THE WILSON JOURNAL OL ORNITHOLOGY • Voi 122, No. 4. December 2010

observed one female Brown-headed Cowbird
removing Dickcissel nestlings without laying
eggs. We recorded 13 and 1 1 depredations at the
incubation and nestling stages, respectively (Ta-
ble 1 ). Depredation by cricetid rodents and
raccoons were nocturnal, whereas depredation
by ground squirrels, mink, and cowbirds were
diurnal. All depredations by snakes were diurnal
except two by fox snakes, which were nocturnal.

DISCUSSION

The potential predator community in agricul-
tural systems is diverse; thus, the composition of
the community, and identity of the dominant
predator, will ultimately dictate nest predation
risk (Klug et al. 2009). We found at least 10
species responsible for predation of Dickcissel
nests, including both obligate grassland species
(e.g., ground squirrels and snakes) and more wide-
ranging habitat generalists (e.g., common raccoon
and Brown-headed Cowbird; Table 1). Our sam-
ple size of video-monitored nests (/; = 33) is
consistent with other studies in the grasslands of
North America (Table 2), but it is likely that
additional predator species would be detected
with a larger sample size. For example, striped
skunk (Mephitis mephitis), Virginia opossum
(Didelphis virginiana), American Crow (Corvus
brachyrhynchos), and coyote (Canis latrans) have
been documented as nest predators (Renfrew and
Ribic 2003, Peterson et al. 2004, Staller et al.
2005), but were not documented at our study sites
regardless of their presence on predator surveys
(Klug et al. 2009). However, given the consisten-
cy of snake events across years, it is unlikely a
larger sample would change our conclusion that
snakes are among the most important nest
predators for grassland birds in this region.

Information about the predator community is
important for understanding variation in nest
predation and to increase productivity of nesting
birds (Lahti 2009). However, recognizing the
importance of the predator community also brings
challenges. Studies that document predation of
grassland bird nests consistently discover wide-
ranging generalist predators including Brown-
headed Cowbird and common raccoon. Thus,
efforts directed at increasing reproductive success
of birds have often focused on decreasing the
population size of species that have increased due
to agricultural land use (Heske et al. 1999,
Kosciuch and Sandercock 2008). Our study, and
other nest predation studies in grassland habitats.

has revealed considerable nest predation by
snakes and ground squirrels, which are essential
members of prairie communities (Table 2). Rec-
ommendations directed at decreasing the influ-
ence of resident predators in grasslands where
birds are nesting are not straightforward given that
many of these species are also targets of
conservation. For example, fox snakes and
Franklin’s ground squirrels (Spermophihis frank-
linii) are native members of the tallgrass prairie
(Martin et al. 2003) and efforts to control these
predators to increase bird populations must
proceed with caution.

ACKNOWLEDGMENTS

We thank the staff of Boyer Chute and DeSoto National
Wildlife refuges for their support. We also thank the
American Museum of Natural History Frank M. Chapman
Memorial Fund, Sigma Xi Grants-In-Aid of Research.
Center for Great Plains Studies Grant-In-Aid for Graduate
Students, University of Nebraska at Omaha, USDA
Biotechnology Risk Assessment Research Grants Program,
and the U.S. Fish and Wildlife Service for financial support.

LITERATURE CITED

Bergin, T. M., L. B. Best, K. E. Freemark, and K. J.
Koehler. 2000. Effects of landscape structure on nest
predation in roadsides of a midwestern agroecosystem:
a multiscale analysis. Landscape Ecology 15:131-143.
Dinsmore, S. J., G. C. White, and F. L. Knopf. 2002.
Advanced techniques for modeling avian nest survival.
Ecology 83:3476-3488.

Grant, T. A., E. M. Madden, T. L. Shaffer, P. J. Pietz,
G. B, Berkey, and N. j. Kadrmas. 2006. Nest
survival of Clay-Colored and Vesper sparrows in
relation to woodland edge in mixed-grass prairies.
Journal of Wildlife Management 70:691-701.
Herkert, j. R., D. L. Reinking, D. A. Wiedenfeld, M.
Winter, J. L. Zimmerman, W. E. Jensen, E. J. Finck,
R. R. Koford. D. H. Wolfe. S. K. Sherrod. M. A.
Jenkins, J. Faaborg, and S. K. Robinson. 2003.
Effects of prairie fragmentation on the nest success of
breeding birds in the midcontinental United States.
Conservation Biology 17:587-594.

Heske, E. J., S. K. Robinson, and J. D. Brawn. 1999.
Predator activity and predation on songbird nests on
forest-field edges in east-central Illinoi.s. Landscape
Ecology 14:345-354.

Heske, E. J.. S. K. Robinson, and J. D. Brawn. 2(X)1. Nest
predation and neotropical migrant songbirds: piecing
together the fragments. Wildlife Society Bulletin 29:52-6 1 .
Klug, P. E. 2005. The effects of local grassland habitat and
surrounding landscape composition on the predators of
grassland bird nests. Thesis. University of Nebraska,
Omaha, USA.

Klug, P., L. L, Wolfenbarger, and J. P. McCarty.
2009. The nest predator community of grassland birds

SHORT COMMUNICATIONS

803

responds to agroecosysteni habitat at multiple scales.
Ecography 32:973-982.

Knopf. F. L. 1994. Avian assemblages on altered
grasslands. Studies in Avian Biology 15:247-257.

Kosciuch, K. L. and B. K. Sandercock. 2008. Cowbird
removals unexpectedly increase productivity of a
brood parasite and the songbird host. Ecological
Applications 18:537-548.

Lahti. D. 2009. Why we have been unable to generalize
about bird nest predation. Animal Conservation
12:279-281.

Martin. J. M.. E. J. Heske, and J. E. Hofmann. 2003.
Franklin s ground squirrel (Spennophiliis franklinii) in
Illinois: a declining prairie mammal? American
Midland Naturalist 150:130-138.

Martin. T. E. 1993. Nest predation and nest sites.
Bioscience 43:523-533.

PlETZ. P. J. and D. a. Granfors. 2000. Identifying
predators and fates of grassland passerine nests using
miniature video cameras. Journal of Wildlife Manage-
ment 64:71-87.

Peterson. B. L., B. E. Kus. and D. H. Deutschman. 2004.
Determining nest predators of the Least Bell's Vireo

through point counts, tracking stations, and video-
photography. Journal of Field Ornithology 75:89-
95.

Powell, L. 2007. Approximating variance of demographic
parameters using the della method: a reference for
avian biologists. Condor 109:949-954.

Renfrew. R. B. and C. A. Ribic. 2003. Grassland passerine
nest predators near pasture edges identified on video-
tape. Auk 120:371-383.

Samson, F. and F. Knopf. 1994. Prairie conservation in
North America. Bio.science 44:418^23.

Staller, E., W. Palmer, J. Carroll, R. Thornton, and
D. Sisson. 2005. Identifying predators at Northern
Bobwhite nests. Journal of Wildlife Management
69:124-132.

Thompson III, F. R. and D. E. Burhans. 2003. Predation
of songbird nests differs by predator and between field
and forest habitats. Journal of Wildlife Management
67:408-416.

Weatherhead, P. j. and G. Blouin-Demers. 2004.
Understanding avian nest predation: why ornitholo-
gists should study snakes. Journal of Avian Biology
35:185-190.

The Wilson Journal of Ornithology 122(4):803-806, 2010

Interspecific Cavity-sharing Between a Helmeted Woodpecker {Dryocopus
galeatiis) and Two White-eyed Parakeets {Aratinga leucophthalma)

Kristina L. Cockle’-^-'*

ABSTRACT. — Cavity-nesting birds may frequently
compete for a limited supply of nest and roost cavities
in trees, but interspecific sharing of these cavities has
rarely been reported. The globally vulnerable Helmeted
Woodpecker (Dryocopus galeatus), a little-known
Atlantic Forest endemic, is believed to be threatened
by nest-site competition; however, little is known about
its ecology or natural history. 1 report an ob.servation of
a female Helmeted Woodpecker roosting with two
White-eyed Parakeets (Aratinga leucophthalma) in their
non-excavated (natural) nest cavity at Cruce Caballero
Provincial Park. Argentina, and discuss possible impli-
cations for ecology and con.servation of this rare

' Center for Applied Conservation Research. Department
of Forest Science, University of British Columbia, 2424
Main Mall, Vancouver, BC. V6T IZ4, Canada;
e-mail: kri.stinacockle@gmail.com
-Proyecto Selva de Pino Parana, Velez Sarsfield y San
Jurjo S/N, San Pedro, Misiones, (3352), Argentina.

-’Fundacion de Historia Natural Felix de Azara. Departa-
mento de Ciencias Naturales y Antropologi'a. Universidad
Maimonides. Valentm Virasoro 732 (CI405 BDB), Buenos
Aires, Argentina.

woodpecker. Received 22 Januarv 2010. Accepted 20
April 2010.

Tree cavities are u.sed by many bird species for
nesting and roosting, and supply of cavities may
be a key re.source that limits populations (Newton
1994) and structures communities (Martin et al.
2004, Aitken and Martin 2008). Competition
within species for a limited supply of cavities
may lead to physical conflict resulting in injuries
to adults and death of their young (Snyder et al.
1987, Heinsohn and Legge 2003). Cavity limita-
tion may also create conditions that favor co-
operative breeding (Heinsohn and Legge 2003).
Cavity usurpation is common among competing
.species (Snyder et al. 1987, Renton and Brighu
smith 2009. Strubbe and Matthysen 2009), but
cavities could also be shared between species
(Robinson et al. 2006). Interspecific cavity-
shaiing would allow individuals to avoid costs
as.sociated with aggressive usurpation of cavities.

804

THE WILSON JOURNAL OL ORNITHOLOGY • Vul. 122. No. 4. December 2010

Sharing could also provide benefits including
predator vigilance.

The Helmeted Woodpecker (Dtyocopits galeatiis)
is rare, little-known, globally vulnerable, and
endemic to the Atlantic Forest of Paraguay,
Argentina, and Brazil (BirdLife International
2008). It is considered a bird of tall forests (Winkler
et al. 1995) and may be most common in mature
forest in the Province of Misiones, Argentina
(Bodrati and Cockle 2006). Little else is known
about its natural history or ecology, but competition
for nest sites with the larger and more abundant
Lineated Woodpecker (D. lineatiis) has been pro-
posed as a key thieat (BirdLife International 2008).

OBSERVATIONS

1 observed a female Helmeted Woodpecker
roosting overnight with a pair of White-eyed
Parakeets (Aratinga leiicophthalma), while the
latter were incubating in a naturally occurring,
non-excavated cavity in a large live branch of
grapia (Apuleia leiocarpa). The site was in Cruce
Caballero Provincial Park (26° 3 L S, 53° 58' W),
Province of Misiones, Argentina, in primary
mixed Atlantic Forest with laurel (Nectandra
and Ocotea spp.), guatambti {Balfourodendron
riedelianum), and Parana pine {Araucaria an-
giistifolia) (Cabrera 1976). The Helmeted Wood-
pecker is scarce in the park, and the White-eyed
Parakeet is abundant (Bodrati et al. 2010). A pair
of White-eyed Parakeets laid at least two eggs in
the cavity in December 2008 (Cockle et al.
2010), but their nest failed. The cavity was empty
when checked for contents on 21 October and 2
November 2009.

The cavity was shared for at least two
consecutive nights on 3 and 4 December 2009
by a Helmeted Woodpecker and a pair of White-
eyed Parakeets. The cavity contained one white
egg on 3 December 2009. The following day, 1
arrived at the cavity at 0520 hrs, 20 min before
dawn. The sky was dark, cloudy, and threatening
to rain. A female Helmeted Woodpecker appeared
from inside the cavity at 0525 hrs. She remained
at the cavity entrance until 0537 hrs when she
tlew out. clung briefly to the roots of an epiphytic
giiembe {Philodendron spp.) about 2 m distant,
then returned to the cavity. She remained in or
near the cavity entrance until 0752 hrs and 1 could
clearly distinguish her brown face from the chin to
above the eye. She left the cavity three times to
peck at the giiembe roots or move up the branch
within I m of the cavity entrance. Three times she

seemed to glean insects just inside the cavity
entrance. She did not descend to the floor of the
cavity (egg chamber) and remained visible Just
inside the entrance. It rained from 0616 to 0625 hrs
but the sun shone directly on the cavity from
0712 hrs onwai'ds. At 0752 hrs the woodpecker
became agitated, jumping forward and backward
inside the cavity. She flew out of the cavity and
disappeared into the forest. Two White-eyed
Parakeets immediately emerged from the bottom
of the cavity and flew away in silence. Neither the
woodpecker nor the parakeets vocalized that
morning, and there was no sign of aggression.

1 returned to the cavity at 0510 hrs the
following day accompanied by A. A. Bodrati.
The sky was clear and the nest tree was
illuminated by bright moonlight. We could
immediately see the Helmeted Woodpecker in
the cavity, apparently awake. She flew from the
nest cavity 10 min before sunrise at 0530 hrs. The
two White-eyed Parakeets emerged immediately
from the bottom of the nest. They stayed 6 min at
the entrance and then flew away in silence.

The cavity was 12.4 m above ground in a live
healthy grapia that measured 30 m in total height
and 87 cm in diameter at breast height. It had one
lateral entrance 6 cm wide X 38 cm tall. The
cavity was 90 cm deep vertically and 14 cm deep
horizontally. An epiphytic giiembe higher on the
branch was rooted inside the cavity. 1 believe the
parakeets roosted in the egg chamber and
incubated the egg, and the woodpecker roosted
above them near the cavity entrance.

J. M. Klavins showed me another roost cavity
at Cruce Caballero Provincial Park used by a
Helmeted Woodpecker in May 2005. That cavity
seemed to have been formed where a branch
broke off the tree, and either a woodpecker
excavated the knot-hole or the cavity formed
through decay processes. It was about 12 m high
in the main trunk of a live healthy tree. 59 cm in
diameter, 24 m tall, at the edge of the park. It was
later occupied by a nest of bees in 2006-2007, a
roosting male Lineated Woodpecker in 2008 and
2009, and a successful nest of Scaly-headed
Parrots {Pionu.s niaxiniiliani) in 2009 (KLC and
J. M. Lammertink, unpubl. data).

DISCUSSION

Reports of interspecific cavity-sharing are
uncommon. Skutch (1969) reports a cavity in
Central America to which a pair of Streak-headed
Woodcreepers {Lepidocolaptes souleyetii) and a

SHORT COMMUNICATIONS

805

Tawny- winged Woodcreeper (Dein/rocincla ana-
hotiiia) brought nest material alternately; the
Tawny-winged Woodcreeper incubated the two
eggs and raised the single nestling, a Streak-
headed Woodcreeper. Robinson et al. (2006)
report a mixed brood of Red-breasted Nuthatches
(Sitta canadensis) and Mountain Chickadees
(Poecile gambeli) raised in the same cavity by
parents of both species in North America. I
observed several incidences of probable interspe-
cific nest-usurpation during tour breeding seasons
studying cavity-nesting birds in the Atlantic
Forest, but this is the first time I observed
simultaneous interspecific cavity-sharing among
birds, and it appears to be the first such report for
South America.

Much remains to study about the role of tree
cavities in the ecology and conservation of the
Helmeted Woodpecker (Winkler et al. 1995).
Apparent nests were reported in a cavity 7.8 cm in
diameter, 2.3 m high in an unidentified tree at
Iguazu National Park in Misiones in spring 1985
(adults seen entering and leaving cavity without
food) and by the side of a road in adjacent Igua9u
National Park in Brazil in November 1988, but no
further descriptions are available (Collar et al.
1992; Chebez 1994; Winkler et al. 1995; Hernan
Casahas, pers. comm.). A photograph of the first
cavity by Martin Adamovsky (published in
Chebez 1994) shows it to have a circular entrance,
apparently excavated by a woodpecker. Juan
Mazar Barnett (pers. comm.) reported a male
quietly excavating a cavity in a dead branch of a
live laurel (Nectandra spp.) tree at Puerto
Peninsula, Misiones, on 31 October 1994. A
female was nearby at times, always passive, and
did not excavate. Madrono Nieto and Esquivel
(1995) reported two pairs nesting in spring 1994
at Reserva Natural del Bosque Mbaracayu in
Paraguay without further details.

The roost hole shared with the White-eyed
Parakeets was not excavated by a woodpecker.
The Helmeted Woodpecker has a narrow-based,
weak-tipped bill and forages mainly by scaling
and pecking (Brooks et al. 1993). Further studies
may find that Helmeted Woodpeckers frequently
reuse old cavities, entering into competition with
secondary cavity-nesting birds such as parrots.
Cavities may be in short supply in the Atlantic
Forest: high-grade logging has already removed
the largest trees with the greatest potential for
cavity formation from nearly all remaining forest
(Cockle et al. 2010).

ACKNOWLEDGMENTS

I thank A. A. Bodrati for finding the White-eyed Parakeet
nest in 2008 and sharing in my observations in 2009; Kathy
Martin for di.scussions and supervision; D. W. Cockle for
building the cameras used to inspect cavities; J. M. Klavins,
J. M. Segovia, and E. A. Jordan for sharing their
observations or helping in the field; and Kathy Martin, A.

R. Norris, Hernan Casanas, Juan Mazar Barnett and J. M.
Lammertink for unpublished data or comments on the
manuscript. This work was financed by a Re.search Grant
from the British Ornithologists’ Union, a Conservar La
Argentina grant from Aves Argentinas and BirdLife
International, a Namkoong Family Fellowship, a Donald

S. McPhee Fellowship, Rufford Foundation Booster Grant,
Columbus Zoo and Aquarium Conservation Fund, National
Science and Engineering Research Council of Canada
(NSERC) Canada Graduate Scholarship, Killam Pre-
doctoral Fellowship, Neotropical Bird Club Conservation
Award to A. A. Bodrati, and research support from NSERC
and Environment Canada to Kathy Martin. Equipment was
loaned or donated by Environment Canada, Idea Wild, and
the Area de Manejo Integral de la Reserva de Biosfera
Yaboty. Research was authorized by the Misiones Minis-
terio de Ecologia, RNR y Turismo.

LITERATURE CITED

Aitken, K. E. H. and K. Martin. 2008. Resource selection
plasticity and community responses to experimental
reduction of a critical resource. Ecology 89:971-980.
BirdLife International. 2008. Threatened birds of the
world 2008. CD-ROM. BirdLife International, Cam-
bridge, United Kingdom.

Bodrati, A. and K. Cockle. 2006. Habitat, distribution,
and conservation of Atlantic Forest birds in Argentina;
notes on nine rare or threatened species. Omitologia
Neotropical 17:243-258.

Bodrati, A., K. Cockle, J. M. Segovia, I. Roesler, J. I.
Areta, and E. Jordan. 2010. La avifauna del Parque
Provincial Cruce Caballero, Provincia de Misiones,
Argentina. Cotinga 32:41-64.

Brooks, T. M., R. Barnes, L, Bartrina, S. H. M.
Butchart, R. P. Clay, E. Z, Esquivel, N. I.
Etcheverry, J. C. Lowen, and j. Vincent. 1993.
Bird surveys and conservation in the Paraguayan
Atlantic Forest: Project CANOPY '92 final report.
Study Report Number 57. BirdLife International.
Cambridge. United Kingdom.

Cabrera, A. L. 1976. Enciclopedia Argentina de agricul-
tura y jardinen'a. Second Edition. Tomo II. Fa.sciculo I.
Regiones fitogeograficas Argentinas. Editorial Acme
S. A. C. 1., Buenos Aires, Argentina.

Chebez, J. C. 1994. Los que se van: especies Argentinas en
peligro. Editorial Albatros, Buenos Aires, Argentina.
Cockle, K., K. Martin, and K. Wiebe. 2010. Selection of
nest trees by cavity-nesting birds in the neotropical
Atlantic Forest. Biotropica: On-line.

Collar, N. J., L, P. Gonzaga, N. Krabbe, A. Madrono
Nieto, L. G. Naranjo, T. A. Parker III. and D. C.
Wege. 1992. Threatened birds of the Americas, the
ICBP/IUCN Red Data Book, 2. Third Edition.

806

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122. No. 4. December 2010

International Council for Bird Preservation, Cam-
bridge. United Kingdom.

Heinsohn, R. and S. Legge. 2003. Breeding biology of the
reverse-dichromatic, co-operative parrot Eclectus ror-
atiis. Journal of Zoology, London 259:197-208.

Madrono Nieto, A. and E. Z. Esquivel. 1995. Reserva
Natural del Bosque Mbaracayu: su importancia en la
conservacion de aves amenazadas, cuasi amenazadas y
endemicas del Bosque Atlantico del Interior. Cotinga
4:52-57.

Martin, K., K. E. H. Aitken, and K. L. Wiebe. 2004. Nest
sites and nest webs for cavity-nesting communities in
interior British Columbia, Canada: nest characteristics
and niche partitioning. Condor 106:5-19.

Newton, 1. 1994. The role of nest sites in limiting the
numbers of hole-nesting birds: a review. Biological
Conservation 70:265-276.

Renton, K. and D. Brightsmith. 2009. Cavity use and
reproductive success of nesting macaws in lowland
forest of southeast Peru. Journal of Field Ornithology
80:1-8.

Robinson, P. A., A. R. Norris, and K. Martin. 2006.
Interspecific nest-sharing by Red-breasted Nuthatch and
Mountain Chickadee. Wilson Bulletin 1 17:400^02.

Skutch, a. F. 1969. Life histories of Central American
birds. III. Families Cotingidae, Pipridae, Formicarii-
dae, Furnariidae, Dendrocolaptidae, and Picidae.
Pacific Coast Avifauna 35.

Snyder. N. F. R. J. W. Wiley, and C. B. Kepler.
1987. The parrots of Luquillo: natural history and
conservation of the Puerto Rican Parrot. Western
Foundation of Vertebrate Zoology, Los Angeles,
California, USA.

Strubbe. D. and E. Matthysen. 2009. Experimental
evidence for nest-site competition between invasive
Ring-necked Parakeets {Psittacula krameri) and native
nuthatches {Sitta europaed). Biological Conservation
142:1588-1594.

Winkler, H., D. A. Christie, and D. Nurney. 1995.
Woodpeckers: a guide to the woodpeckers of the
world. Houghton Mifflin Company, Boston, Massa-
chusetts, USA.

The Wilson Journal of Ornithology 122(4):806-809. 2010

Response to Nestling Throat Ligatures by Three Songbirds

Gabrielle L. Robinson,' Courtney J. Conway,' ""''

Chris Kirkpatrick,' and Dominic D. LaRoche'

ABSTRACT. — We attempted to collect diet samples
using throat ligatures from nestlings of three songbird
species in a riparian woodland in .southeastern Arizona
from May to August 2009. We had success with Song
Sparrows (Melospiza melodiu). observed adult Yellow-
breasted Chats {Icteria virens) reclaim food from
nestlings, and discontinued the use of throat ligatures
when we observed an adult Abert’s Towhee iPipilo
aherfi) remove two, 3^-day-old ligatured nestlings
from its nest. Previous studies have reported problems
(e.g., aggression toward nestlings by adults) with throat
ligatures, but we are the first to document removal (and
subsequent nestling mortality) in response to this
technique. We urge investigators to exercise caution
when using throat ligatures on species for which
evidence of the safety and efficacy of this method are
lacking, especially when nestlings are small in size
relative to adults. Received 18 February 2010. Accepted
29 April 2010.

' School of Natural Re.sources and the Environment.
University of Arizona, Tuc.son, AZ 85721, USA.

’ U.S. Geological Survey, Arizona Cooperative Fish and
Wildlife Research Unit. University of Arizona, Tucson. AZ
85721. USA.

'Corresponding author; e-mail: cconway(®usgs.gov

Documenting the diet of birds is an important
component of many studies designed to test
hypotheses related to ecology and evolution of
birds. Documenting the diet of nestling birds in an
unbiased manner is particularly challenging,
especially when investigators want to identify
prey items to taxonomic levels beyond Order or
want precise estimates of biomass. Investigators
have used throat ligatures to collect diet samples
from nestling birds since the 1930s (Kluijver
1933, Owen 1956, Mellott and Woods 1993).
Fecal analysis, the use of artificial nestlings to
collect food, analysis of stomach contents, and
visual observation (Evans 1964) are often less
effective or more invasive than throat ligatures.
However, researchers have experienced several
problems with throat ligatures related to both
quality of the sample obtained and safety of the
nestlings sampled. For example, food may slip
past the ligature if the ligature is too loose (Owen
1956, Orians 1966), or food is disgorged if the
sampling period is too long or the rate of
provisioning too high (Orians 1966, Johnson et
al. 1980). Moreover, parents will, at times.

SHORT COMMUNICATIONS

807

remove and eat the food from the mouth of the
ligatured nestling (Robertson 1973). Ligatures
have also been tound to affect normal gaping
(Johnson et al. 1980) and begging behavior
(Orians and Horn 1969), and to occasionally
cause death by strangulation (Moore 1986,
Mellott and Woods 1993).

Several authors have described accounts of
aggressive behavior by adults toward nestlings
with throat ligatures. Robertson (1973) mentions
that Red-winged Blackbirds {AgeUiiiis plweni-
ceiis) in a few cases attempted to remove ligatures
from nestlings. Little et al. (2009) reported that
adult Bobolinks (Dolichony.x oryzivorus) pecked
and pulled at nestlings’ throat ligatures at most
nests sampled. Furthermore, aggressive behavior
by adults such as grasping the ligature and
forcibly pulling the nestlings’ heads upwards or
sideways occuired at half of the nests. Despite this
aggression. Little et al. (2009) did not observe any
nestling mortality resulting from the behavior of
the adults. We provide the first documentation of
nestling mortality resulting from removal by
adults of nestlings outfitted with throat ligatures.

METHODS

We monitored nests of Abert’s Towhees (Pipilo
aberti), Yellow-breasted Chats {Icteria virens),
and Song Sparrows (Melospiza melodia) from
May to August 2009 in a riparian woodland along
the Santa Cruz River at Tumacacori National
Historic Park (31° 34' 03" N, 111° 03' 03" W),
Santa Cruz County, Arizona, USA. The elevation
is ~ 1,000 m and the plant communities are
cottonwood-willow iPopuhis-SaUx) riparian for-
est along the river and mesquite-hackberry
{Prosopis-Celtis) woodland in the adjacent up-
land. We attempted to collect diet samples from
nestlings of the focal species by attaching throat
ligatures via the collar method (Kluijver 1933,
Johnson et al. 1980). We constructed ligatures
from 20-22 gauge, solid-core, black or blue
plastic-coated copper wire (RadioShack Corp.,
Forth Worth, TX, USA) and bent the wire into a
U-shape which we fitted around the nestlings’
necks sufficiently tight to prevent swallowing but
sufficiently loose to allow normal breathing and
gaping behavior. We weighed (g) each nestling
before applying the ligature and returned nestlings
to their nest within 10 min. We used binoculars to
observe each sampled nest from a distance of
— 15 m until we saw adults return with food. We
approached the nest after each feeding visit to

remove food from the nestlings’ mouths with
tweezers. We tried to collect samples for 2-4
consecutive feedings at each nest before removing
the ligatures to maximize the size of the food
sample, and to prevent handling the nestlings and
disturbing the nest on more than one occasion. We
compensated for the missed feedings during
sampling by feeding each nestling a comparable
portion of wax worms (pyralid larvae) after we
removed the ligatures.

RESULTS

All three nests with ligatured Song Sparrow
nestlings (3-8 days of age) yielded diet samples
with no apparent problems. We observed adults at
four of five Yellow-breasted Chat nests (nestlings
of 2-5 days of age) remove food from ligatured
nestlings’ mouths before leaving the nest when
the food was not swallowed (i.e., diet samples
were biased; sensii Robertson 1973). We sampled
(with black ligatures) our first Abert’s Towhee
nest containing two 3-4-day-old nestlings on 29
July 2009 at 1052 hrs (MST). Nestling “A”
weighed 7.0 g and nestling “B” weighed 16.7 g.
Both adults amved at the nest with food —25 min
after we applied the ligatures, and attempted to
feed the nestlings for 2-3 min. We observed at
least one of the adults consume food that was
originally placed in a nestling’s mouth, similar to
that observed for Yellow-breasted Chats. We
approached the nest to collect diet samples after
the adults left and found that only nestling "B”
remained in the nest. We searched under the nest
but could not find nestling “A.” We took a diet
sample from nestling “B” and retreated to
observe the nest from a distance. An adult
returned to the nest -10 min later with food and
spent — I min trying to feed the nestling. We then
watched the adult remove the nestling from the
nest and depart from the nest before we lost sight
of the adult. We immediately approached the nest,
began searching for nestling “B,” and found
nestling “B” 2 m from the nest on the ground.
The ligature was missing, and we concluded the
adult had carried the nestling by grasping the
ligature thereby forcing the ligature to open and
slide oft the nestling s neck. We examined
nestling "B,” found no sign of injury, returned
it to the nest, and moved away for observation.
The adult returned about 5 min later to feed the
nestling and then shaded the nestling for about
5 min. The adult appeared to resume typical
behavior as soon as the ligature was removed

808

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

from the remaining nestling, suggesting the
ligature elicited the adult’s response and not the
nestling itself. We did not find nestling “A” and
assumed that it died after being removed from the
nest.

DISCUSSION

Nest sanitation is the process by which parent
birds rid their nests of foreign debris (Welty
1982). Several authors have described parent
birds attempting to sanitize (remove) banded
nestlings when parents perceived leg bands as
foreign debris in the nest (Lovell 1945, Berger
1953, Brackbill 1954). No published account
exists of a throat ligature eliciting nestling
removal by adults despite the superficial resem-
blance of throat ligatures to leg bands on nestling
birds. This behavior has been observed with
Florida Scrub-Jays (Aphelocoma coenilescens)
(Reed Bowman, Archbold Biological Station,
pers. comm.). Ours is the first study to report an
instance of nestling removal by adults (and
subsequent nestling mortality) in response to
use of throat ligatures.

We believe other researchers that have reported
aggressive behavior of adults toward ligatured
nestlings did not observe removal of young
because adults in these studies were physically
incapable of removing nestlings due to the relative
mass of the nestlings. For example. Little et al.
(2009) described aggressive behavior (but not
removal) of ligatured, 6-day-old Bobolink nest-
lings that would have weighed 53% of adult body
mass (Martin and Gavin 1995). Towhee nestlings
in our study were 16 and 37% of adult body mass
based on average adult body mass of 44.7 g
(Tweit and Finch 1994). Nest substrate may also
affect an adult’s ability to successfully remove
nestlings from a nest and may explain interspe-
cific variation in this behavior.

We encourage future investigators to observe
parental behavior at several nests before they use
throat ligatures widely on a species for which
evidence of the safety and efficacy of this method
is lacking, e.specially on younger ne.stlings that
weigh <50% of adult body mass. Some species
accept throat ligatures without problems (e.g..
Song Sparrows in our study) and other species
experience problems that may bias diet samples
(e.g.. Yellow-breasted Chats in our study). The
color or type of material used to make ligatures
may affect the probability of removal and
deserves further study. We used both black and

blue ligatures, but birds may respond differently
to other colors similar to the way that some birds
respond differently to different band colors
(Burley 1988). We are concerned that removal
of young by adults in response to use of throat
ligatures on nestlings may be relatively wide-
spread in birds based on recent observations of
this behavior by Florida Scrub-Jays (Reed Bow-
man, pers. comm.), the observations described
here of Abert’s Towhees, and the recent report of
aggression toward ligatured nestlings by Bobo-
links (Little et al. 2009).

ACKNOWLEDGMENTS

We thank J. M. Moss and Lisa Carrico (U.S. Park
Service, Tumacacori National Historical Park) for logistical
support. Funding was provided by the Arizona Game and
Fish Department from a Wildlife Conservation Fund grant
and U.S. Geological Survey from a Park Oriented
Biological Support grant. Any use of trade names is for
descriptive purposes only and does not imply endorsement
by the U.S. Government.

LITERATURE CITED

Berger, A. J. 1953. Reaction of female Horned Larks to
banded young. Bird-Banding 24:19-20.

Brackbill. H. 1954. Red-eyed Vireo throws banded young
out of nest. Bird-Banding 25:61.

Burley, N. 1988. Wild Zebra Finches have band-color
preferences. Animal Behaviour 36:1235-1237.

Evans, F. C. 1964. The food of Vesper, Field, and
Chipping sparrows nesting in an abandoned field in
southeastern Michigan. American Midland Naturalist
72:57-75.

Johnson, E. J., L. B. Best, and P. A. Heagy. 1980. Food
sampling biases associated with the “ligature meth-
od.” Condor 82:186-192.

Kluijver, H. N. 1933. Bijdrage tot de biologic en ecologie
van den Spreeuw (Sturniis vulgaris vulgaris L.)
gedurende zijn voortplantingstijd. Wageningen, The
Netherlands.

Little, L. P., A. M. Stong, and N. G. Perlut. 2009.
Aggressive response of adult Bobolinks to neck
ligatures on nestlings. Wilson Journal of Ornithology
121:441^44.

Lovell, H. B. 1945. Banded Song Sparrow nestlings
removed by parent. Bird-Banding 16:144-145.
Martin, S. G. and T. A. Gavin. 1995. Bobolink
(Dolichonyx oryzivorus). The birds of North America.
Number 176.

Mellott, R. S. and P. E. Woods. 1993. An improved
ligature technique for dietary sampling in nestling
birds. Journal of Field Ornithology 64:205-210.
Moore, J. 1986. Dietary variation among nestling Starlings.
Condor 88:181-189.

Orians, G. H. 1966. Food of nestling Yellow-headed
Blackbirds, Cariboo Parklands, British Columbia.
Condor 68:321-337.

SHORT COMMUNICATIONS

809

Orians, G. H. and H. S. Horn. 1969. Overlap in food and
foraging of tour species of blackbirds in the potholes
of central Washington. Ecology 50:930-938.

Owen, D. F. 1956. The food of nestling jays and magpies.
Bird Study 3:257-265.

Robertson, R. J. 1973. Optimal niche space of the Red-
winged Blackbird. 111. Growth rate and food of

nestlings in marsh and upland habitat. Wilson Bulletin
85:209-222.

Tweit, R. C. and D. M. Finch. 1994. Abert’s Towhee
(Pipila aberti). The birds of North America. Number
1 11.

Welty, j. C. 1982. The life of birds. Saunders College
Publishing, Philadelphia, Pennsylvania, USA.

EDITORS of The Wilson Bulletin (1888-2005)
and

The Wilson Journal of Ornithology (2006-2010)

Lynds Jones

1888-1900

Frank L. Burns

1901

Lynds Jones

1902-1924

T. C. Stephens

1925-1938

Josselyn Van Tyne

1939-1948

David E. Davis

1949-1950

George M. Sutton

1950-1951

Harrison B. Tordoff

1952-1954

Keith L. Dixon

1955-1958

H. Lewis Batts Jr.

1959-1963

George A. Hall

1964-1973

.John P. Hubbard

1974

Jerome A. Jackson

1975-1978

Jon C. Barlow

1979-1984

Keith L. Bildstein

1985-1987

Charles R. Blem

1988-1997

Robert C. Reason

1998-2000

John A. Smallwood

2001-2003

James A. Sedgwick

2004-2006

Clait E. Braun

2007-

The Wilson Journal oj Ornithology 122(4):8 10-81 1, 2010

William and Nancy Klamm Service Award for 2010:

Jerome A. Jackson

In 1971, a young ornithologist presented his
first paper at a Wilson Ornithological Society
meeting on Dauphin Island, Alabama. This first
paper compared the breeding biology of Red-
bellied (Melanerpes caroliniis) and Red-headed
(M. erythrocephalus) woodpeckers; thus his
passion for woodpeckers and his enthusiasm for
the Wilson Ornithological Society have been
intertwined from early in his ornithological
career.

Jerry Jackson’s work for the Society has been
as extensive as it was varied. In 1973 Jerry was
elected Treasurer, succeeding Bill Klamm in the
position. Bill mentored Jerry during that year and
Jerry's continuing friendship with the Klamms

was largely responsible for their bequest to the
Society’s endowment. He served 1 year as
Treasurer before he was elected Editor of the
Wilson Bulletin, a position he held for 4 years. In
1979 he was elected Second Vice-President, and
he rose through the ranks to serve as President
from 1983 to 1985. He has hosted not one, but two
Wilson Society meetings: one at Mississippi State
University in 1977 and another at Florida Gulf
Coast University in 2002.

Another of JeiTy’s passions is hi.story. In 1988,
Harold Mayfield, George Hall, and Jen'y wrote a
history of the Wilson Ornithological Society to
mark its first century. Jeny’s account of the last
third of the first century not only summarized the

810

William and Nancy Klamm Service Award

81 1

growth ot the Wilson Bulletin and the Society’s
grants and prizes, but also included anecdotes that
provide a llavor of the Society and the people who
shaped it during that time. In that paper, Jerry
wrote about his tenure as Treasurer, saying
"Quite honestly, I was out of my element!" As
well as writing about the history of the Society,
Jerry himself has shaped it through his long tenure
on the nominating committee, his more than three
decades of active participation on the Wilson

Council, and his work on the conservation, history
and archives, and Klamm committees.

With sincere gratitude and great affection, we
are proud to present the 2010 William and Nancy
Klamm Service Award to Jerome A. Jackson for
his four decades of service to the Wilson
Ornithological Society. — Sara R. Morris, Richard
C. Banks, Robert C. Bea.son, Robert L. Curry, and
E. Dale Kennedy (Klamm Service Award Com-
mittee).

The Wilson Journal of Ornithology 122(4):812-819, 2010

Ornithological Literature

Robert B. Payne, Book Review Editor

MOMENTS OF DISCOVERY. NATURAL
HISTORY NARRATIVES FROM MEXICO
AND CENTRAL AMERICA. Edited by Kevin
Winker. University Press of Florida, Gainesville,
Florida, USA. 2010: 384 pages, 50 black and
white illustrations. ISBN: 978-0-8130-3417-1.
$ 75.00 (hardcover). — This book, edited by Kevin
Winker and authored by some of the most
outstanding and prolific ornithologists and mam-
malogists of the 20'*’ century in the northern
Neotropics, covers a set of memories, experienc-
es, and points of views of field work in Central
America. It also contains what might be the last
writings of some charismatic explorers including
Miguel Alvarez del Toro, the Coffeys (Ben and
Lula), Charles Sibley, John Emlen, and Dwain
Warner. Twenty contributions that narrate field
adventures that roughly cover from the late 1930s
to the 1990s are presented almost unedited. Most
of the chapters refer to field collecting experienc-
es in Mexico, understandable in terms of close-
ness and access for USA-based researchers during
the 20'*’ century. However, a handful depicts
experiences in Belize, Guatemala, Costa Rica,
Panama, and El Salvador.

Field work constitutes possibly the most
charming part of an ornithologist’s life and, for
the vast majority of researchers, the first touch of
a foreign fauna becomes an enchanting and
breath-taking experience. Therefore, after the
confrontation to a new culture, new food, new
friends and colleagues, and the addictive combi-
nation of fear, excitement, and especially an
awesome tropical fauna, everything together and
surrounded by an unknown language, most get
their hearts trapped in the tropics; sometimes
literally because some of the authors even married
Latin American women!

The book might seem a kind of anthology, as I
am sure it was intended. However, it is much
more than the transcription of the stories that
otherwi.se you would only hear around the
fireplace in a camp, and Joined by a series of
tequila shots. After a careful reading it is clear the
collection of papers repre.sents very important
moments in natural sciences, becau.se among the
lines one discovers the intricate system of

interrelationships among ornithological research
groups, and individuals that started in the post-
second world war period and flourished well
beyond the 1970s.

For a person passionate about the history of
ornithological exploration, like me, the book
helps us understand diverse aspects of the
sometimes cryptic development of ornithology,
that usually has to be reconstructed and tracked
through dates and places depicted on the labels
of specimens in ornithological collections. For
instance, the ~35 years of collecting expertise of
Bob Dickerman in Mexico and Central America
have yielded more than 8,000 specimens,
deposited in at least 16 museums in Canada,
USA, and Mexico. Several have represented new
taxa and, as a whole, they are one of the most
complete avifaunal surveys developed by any
bird collector in the region. Another interesting
aspect is nostalgia. Anyone, anywhere in the
world that is confronted with a series of
memories of field workers, arrives at the point
of environmental deterioration. Through the
pages we get to places that are now long gone
in a preserved state, and also learn about the
evolution of road systems in Mexico and Central
America, and the problematic security for
developing field work in many regions at
present. One also realizes that something has
not changed at all: the hospitality of locals even
in the most remote localities.

It is, however, not a simple book to read. After
the first few chapters, and understanding that the
Los Tuxtlas and Tamaulipas regions are most
favored by researchers (given the close proximity
to Mexico and the fact that some of the authors
participated in the same expeditions), the nan-a-
tions get to common places, and listings of taxa
collected or observed become boring. Neverthe-
less, 1 really enjoyed the different ways the
authors perceived the culture and the environ-
ment, and clear differences in perception arise
between those that participated in one or a few
collecting trips, and those that spent long periods
in or moved to the region.

The major bias of this book is that it largely
represents the “gringo visitor’’ point of view

812

ORNITHOLOGICAL LITERATURE

813

(only one ot the authors is Mexican while the
others are North American or European), and
therefore presents a situation in which U.S.
researchers come to the tropics as leaders,
teachers, or trainers, and local roles are those
of field assistants and guides, leaving little room
for true scientific collaboration among locals.
This issue was true for several decades until the
late 1970s and the book is really pitiful for not
having writings by the late Bernardo Villa,
Mario Ramos, and Allan Phillips (as noted by
the Editor’s Afterword). It should have included
more representatives of recent generations of
zoologists, especially locals, for which true
scientific collaboration is the way to proceed
in field work across a broad spectrum of
activities that include collecting, measuring,
recording, and censusing populations. However,
this is not Kevin Winker’s fault, given that
several authors failed to submit collaborations.
Also, common allusions to Dwain Warner’s
influence along the chapters is understandable,
given that Winker belonged to that research
group and authors associated with it were more
prone to respond than others.

Some details might have improved the book.
For example, photographs are scarce or absent in
most of the chapters. Scientific names (appar-
ently added by Winker) are at times misspelled
and, at other times, erroneously assigned (e.g.,
Stiirnus for Meadowlarks [Sturnella], or Quisca-
lus major assigned to Boat-tailed Crackles from
Chiapas [indeed Q. mexicamis]). In spite of these
minor issues, the book is worthwhile to read,
and a must in natural history libraries, because
it also represents a nice addition to the growing
literature on the history of ornithology and
biological exploration. Many readers will read
it with nostalgia, others trying to recall unknown
aspects of the Central American natural environ-
ment. The main audience must be the younger
generation of field naturalists in the Neotropics,
because among its pages, they will find experi-
ence, scholarship, and passion through the very
words of those field workers that contributed
(and some still do) to the development of Latin
American science. I thank C. E. Braun for
comments and help in improving the English. —
ADOLFO G. NAVARRO-SIGUENZA, Mu-
seo de Zoologia, Facultad de Ciencias, Uni-
versidad Nacional Autonoma de Mexico,
Apartado Postal 70-399, Mexico D. F.
04510, Mexico; e-mail: fcvgOKgiunam.mx

WYTHAM WOODS: OXFORD’S ECOLOGI-
CAL LABORATORY. Edited by Peter S. Savill,
Chistopher M. Peirins, Keith J. Kirby, and Nigel
Fisher. Oxford University Press, Oxford, UK.
2010: xvii + 263 pages, 80 illustrations, 8 pages of
color plates. ISBN: 978-0-19-954320-5. $99.00
(cloth). — There isn’t, or at least there shouldn’t
be, an ornithologist anywhere who would fail to
recognize Wytham Woods as the site of possibly
the single most productive study of avian
populations ever conducted, that of Great {Parus
major) and Eurasian Blue (Cyanistes caeruleus)
tits, started in 1946 by John Gibb shortly after
David Lack, the new Director of the nearby
Edward Grey Institute (EGI), decided to give up
studying European Robins (Erithicus rubecula) in
favor of nest-box studies. However, this is but one
of the many distinguished studies conducted in
what is, as described by Chris Pemns in the
Introduction to this volume, a “very ordinary... -
piece of ash-maple-hazel woodland typical of
central England’’ that’s not even particularly
large, being only about 1,250 ha (2,750 acres) in
size compared to, say, Hubbard Brook in New
Hampshire — perhaps the closest analogue to
Wytham Woods in the U.S.— which is 3,160 ha
(6,952 acres). Yet, as Montagu Bertie, the fifth
Earl of Abingdon, might have commented after
the Enclosure Act of 1814 allowed him to buy out
the various small freeholders and consolidate
most of what is now the Wytham Woods estate,
“mighty oaks from little acorns grow’’ — and
scientifically, at least, the area ranks among the
mightiest plots of land on earth.

This fascinating book provides a history of
Wytham Woods accessible to professionals and
general readers alike, from its glaciated begin-
nings to recent controversies over land-use
practices that eventually wrestled control of the
estate from the Oxford University Forestry
Department, whose management focused entirely
on improving the quantity of timber production, to
a committee representing the interests of the
broader university, thus opening up the estate to a
wide range of ecological investigations. And an
impressive set of studies they are, many of which
are succinctly summarized here in chapters
written by key figures focusing on the trees^
flowers, invertebrates, birds, and mammals as
well as overall land use and conservation
management. (I’m not sure why reptiles and
amphibians were left out unless it’s a reflection of
a relative paucity of herpetologists in the U.K.).

814

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122, No. 4. December 2010

Among the many eminent ecologists and evolu-
tionary biologists who have done important work
in Wytham Woods, besides Lack and Gibb, are
Dennis Chitty, Nick Davies, Charles Elton, Paul
Feeny, Mike Hassell, Alex Kacelnik, Bernard
Kettlewell, John Krebs, Hans Kruuk, David
Macdonald, Ian Newton, Chris Perrins, Mick
Southern, Richard Southwood, and George Var-
ley, an incomplete list that offers incontrovertible
proof that one needn’t necessarily conduct a study
in some exotic locale to be internationally
successful.

One of the reoccurring themes of this book is
the immense value of long-term studies as a
means of generating data that can be used to
address new questions such as climate change as
well as provide a fertile venue for applying new
technologies like radio-tracking, passive integrat-
ed transponders, and molecular genetics, to name
a few. In most cases, such studies are not
conducted by a single individual, but rather by a
series of students and researchers. Nowhere has
this process been conducted more successfully
than in Wytham Woods, where some 40 students
have completed theses involving some aspect of
the tit study alone, each contributing his or her
piece to a study that has successfully defined the
cutting edge of avian ecology for decades. Of
course, facilitating this is the U.K. system of
higher education, which gives a Ph.D. student
3 years, and only 3 years, to conduct original
research and write a thesis — half the length of
time it takes to get an ecology Ph.D. in the U.S.
Yet the system clearly works, and even if the
seemingly frenetic time schedule followed by
Ph.D.s in the U.K. is unrealistic for most
American students, the potential advantages of
plugging into an established study, as demonstrat-
ed by the dozens of students who have success-
fully worked on the Wytham Wood tits, deserve
to be more widely recognized by U.S. academics.

And still there remain much to be done.
Interestingly, although the original idea of putting
out nest boxes for birds came from foresters who
were hoping to decrea.se los.ses to defoliating
caterpillars by increasing lit populations, it’s still
unclear whether increasing tit densities reduces
caterpillars or not. This and several other
examples di.scussed in the book illustrate that
the precise consequences of many, it not most,
ecological interactions are at best uncertain.
Apparently neither the American nor the British
models of higher education are likely to put

ecologists out of business any time soon!
However, if we are ever to understand complex
ecological interactions, surely key insights will
come from places like Wytham Woods where
there has been intense long-term study of multiple
components of the ecosystem.

All ecologists will find much of interest in this
book, whether they have spent time at Oxford and
the EGI or not. For those of us who are farther
along in our careers, the historical perspective
provided by the book is fascinating. For those
starting out, the book offers numerous examples
of how workers at Wytham Woods have launched
their own successful careers while taking advan-
tage of the wealth of prior information provided
by earlier generations of students. Particularly
standing to benefit is anyone in a position to
initiate or continue long-term studies, whether
they be of trees or birds. Doing so can be difficult,
both because of the problems of obtaining
continuous funding and the difficulties of envi-
sioning the value of work that may not be fully
realized for 25, 50, or even 100 years — long after
one has (hopefully) obtained whatever degrees
one is going to get. But then, this is also long after
most, if not all, of our published works will have
outlived their usefulness, and what better legacy
could any of us leave than data that can be
followed up on by future generations of students,
allowing them the means to answer questions
which we have yet to even formulate? — WAL-
TER D. KOENIG, Senior Scientist, Cornell
Laboratory of Ornithology, 159 Sapsucker
Woods Road, Ithaca, NY 14850, USA; e-
mail; wdk4@cornell.edu

ALL ABOUT BIRDS. A SHORT ILLUS-
TRATED HISTORY OF ORNITHOLOGY. By
Valerie Chansigaud. Princeton University Press,
Princeton, New Jersey, USA. 2010: 239 pages,
many color and black-and-white illustra-
tions. ISBN: 978-0-691-14519-8. $29.95 (hard-
cover).— This book was previously published in
2009 as '"The Hi.stoiy of Ornithology" in the
United Kingdom by New Holland Publishers,
when it was translated from French into English
from its 2007 "Histoire de Tornithologie" ,
published in France by Dclachaux & Niestle.

Chapters are arranged by century with a focus
on ornithology in Europe since the 1500s, and
the text consists mainly of biographical sketches

ORNITHOLOGICAL LITERATURE

815

ot historically important ornithologists. The
index lists 769 people, mainly ornithologists,
but also illustrators and administrators. The
bibliography cites nothing published by the
ornithologists who are sketched in the book;
instead, it cites only secondary sources, 14 for
the history of ornithology, 16 for the history of
illustration, and 1 1 for the history of natural
history. A colored timeline at the end of the
book summarizes the historical points of refer-
ences, voyages and discoveries, the major events
in ornithology by country, and the institutional-
ization ot ornithology in museums and societies
with an emphasis on France.

In comparison, a recent book by Michael
Walters (2003) has an extensive bibliography,
mainly of primary sources with 334 entries. The
two books are about the same length (255 pages
and more text per page in Walters). Both books
draw on the comprehensive book by Stresemann
(1951, 1975), which developed the history of
ornithology from classical times through the first
half of the past century. Both the Chansigaud
and Walters books are arranged first by century
and within century by country of the ornithol-
ogists. Chansigaud includes photographs of early
ornithologists, whereas the corresponding photo-
graphs in Walters are larger, sharper, and with
more contrast; apparently both books used the
same image sources. Chansigaud includes imag-
es of period paintings and photographs of bird
specimens as they were painted for the early
monographs. The bibliographic sketches in
Walters are more detailed and interesting to
read, the deposition of their specimens in
museums is reported, and the historical and
social links between ornithologists are explored
in Walters (2003) but generally not in Chansi-
gaud (2010). For example, Walters describes the
positive intellectual and institutional influence of
Buffon on the early evolutionary ornithologist
Lamarck and the brilliant and “most odious of
men”, Cuvier, who publicly humiliated the older
Lamarck in Lamarck’s own lecture room.
Lamarck subsequently lost his students, who
were his last source of financial support; his
daughters then had to beg for food. When he
died he “was buried in a pauper’s grave”, and
Cuvier’s obituary of Lamarck was so vicious
that the Academie Fran9aise in publishing a
memorial volume to Lamarck refused to publish
Cuvier’s contribution. Chansigaud does not
develop the theme of influences and the give-

and-take between the early ornithologists. In
another example, Chansigaud’s account on
Hume and his work in India has more on
Hume’s substantial political and educational
accomplishments than on his ornithological
contributions. In contrast, Walters notes that
Hume accumulated a collection of 63,000 bird
skins and 19,000 eggs (now in the British
Museum [Natural History]) and the largest
ornithological library in Asia; Hume’s manu-
script book on the ornithology of India was
never published because a servant sold the
manuscript in the market as waste paper.
Walters (2003) includes appendices of the
systematic aiTangements of birds from Charleton
(1668) through Linnaeus, Brunnich, Latham,
Pallas, Gloger, and others including Vieillot,
Temminck, Bonaparte, Reichenow, and Fiirbin-
ger. These lists make fascinating material on the
history of systematic ornithology, but this topic
is not part of Chansigaud’s book. The book ends
with a mention of the important early molecular
systematics work of Charles Sibley up to 1990.

Chansigaud includes many photographs of
birds, both early black-and-white illustrations
and more recent photographs of color plates and
mounted specimens, each in its historical
context; these illustrations are an important
feature of the book. Another positive feature is
the index of historically important ornithologists,
and the biographic details are generally accurate.
Both Chansigaud’s and Walters’s books include
sections on the beginnings of American orni-
thology (Bartram, Wilson, Audubon, Bonaparte.
Cones, Baird, and others in North America;
Marcgraf, Don Felix d’Azara, von Spix, von
Ihering, and Schomburgk through Hudson in
South America); Chansigaud gives more atten-
tion to the development of birding. The history
of ornithology in America also has been
developed recently by Barrow (1998) and
Weiden.saul (2007) with their accounts limited
to the ABA area of North America. For a look
at early ornithology and ornithologists, the facts
in the book appear generally accurate, but as is
evident in the titles of the United Kingdom and
French editions, AU About Birds is a set of
historical biographies and, as in the subtitle of
this edition, is not all about birds. — ROBERT
B. PAYNE. Professor Emeritus. University
of Michigan. 1306 Granger Avenue, Ann
Aibor, MI 48104, USA; e-mail: rbpayne(®
umich.edu

816

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

LITERATURE CITED

Barrow, M. V. 1998. A passion for birds. American
ornithology after Audubon. Princeton University
Press, Princeton, New Jersey, USA.

Stresemann, E. 1951. Der entwicklung der ornitholo-
gie. F. W. Peters, Berlin, Germany.

Stresemann, E. 1975. Ornithology from Aristotle to
the present. Translation of Stresemann (1951) by
Hans J. and Kathleen Epstein, with a Foreword
and an Epilogue on American Ornithology by
Ernst Mayr. Harvard University Press, Cambridge,
Massachusetts, USA.

Walters, M. 2003. A concise history of ornithology.
Yale University Press, New Haven, Connecticut,
USA.

Weidensaul, S. 2007. Of a feather: a brief history of
American birding. Harcourt, Orlando, Florida,
USA.

INVISIBLE CONNECTIONS: WHY MI-
GRATING SHOREBIRDS NEED THE YEL-
LOW SEA. By Peter Battley, Brian McCaffery,
Danny Rogers, Jae-Sang Hong, Nial Moores, Ju
Yung-Ki, Jan Lewis, and Theunis Piersma.
Photographs by Jan van de Kam. CSIRO Pub-
lishing, Collingwood, Victoria, Australia. 2010:
160 pages, 240 photographs. ISBN: 978-0-643-
09659-2. $49.95 AU (paper). — No other group of
migratory birds is as perilously wedded to a
relative handful of critical stopover sites as the
shorebirds. Crisscrossing the globe, undertaking
some of the longest nonstop migrations ever
documented (including, in the case of the Bar-
tailed Godwit [Liniosa Icipponica], the longest),
shorebirds depend on abundant, concentrated food
resources to fuel their hemispheric odysseys.

Few such stopover sites are as important as the
Yellow Sea, through which most of the East
Asian-Australasian flyway funnels every year.
Millions of shorebirds commuting between
Alaska, Siberia, central and .south Asia, Indonesia,
New Zealand, and Australia congregate on the
sea’s vast and fecund tidal flats. A third of all the
Byway’s shorebirds gather along the Yellow Sea
each spring; for the godwits. among others, the
entire Byway population concentrates there during
migration. The importance of the Yellow Sea for
such imperiled species as Nordmann’s Green-
shank {Tringa guttifer), or the Spoon-billed Sand-
piper (Eurynorhynchus pygnieiis)-\he world’s
rarest shorebird-cannot be overstated.

The Yellow Sea is also home to 600 million
people, many of whom, like the shorebirds.

depend upon healthy shellfish populations for
their livelihoods. But as this book lays out in
words and images, both fishermen and birds face
an uncertain future. South Korea in particular has
embarked on a campaign of tidal “reclamation,”
resulting in the loss of hundreds of square
kilometers of formerly rich foraging habitat. The
infamous Saemangeum Project alone has impact-
ed 400 km" of formerly prime shorebird habitat.

Invisible Connections is a beautifully illustrated
plea on behalf of birds whose lives are linked to
the health of the Yellow Sea ecosystem, and
whose migrations encompass more than half the
globe. Written by seven noted shorebird biologists
from the U.S., South Korea, Australia, New
Zealand, and the Netherlands, the meaty text is
almost secondary to the stunning images of Dutch
photographer Jan van de Kam, whose obsession
with shorebirds pays glorious dividends here. The
combination is an eloquent argument for preserv-
ing this migratory phenomenon.

Van de Kam’s photographs are arrestingly
beautiful. Following the migrants from their
tundra breeding areas in Siberia and Alaska, to
such starkly lovely sites as Roebuck Bay in
northwest Australia, where he captures green-
shanks and kangaroos in the same frame, van de
Kam shows the full sweep of the East Asian-
Australasian Byway. His images are rich in
shorebird behavior, but most striking are those-
especially photographs taken on the Yellow Sea-
that show the profound impact of humans on
shorebird habitat, and the linked fortunes of
people, especially shell fishermen, and birds.
(Among the more fascinating are a series of
photos of former shorebird hunters near Shanghai
who now use traditional tools like decoys and
bamboo whistles to trap migrant shorebirds in
clap nets for scientists, instead of for the pot.).

This is an English-only version of an originally
trilingual (English, Chine.se, and Korean) edition
first published in 2008 by Wetlands International
to mark the Ramsar Conference in South Korea.
Like its predecessor, this book is meant to capture
the eye and sympathies of the general public, a
task for which it is well-suited. The photographs
are the obvious hook for non-specialists, but the
text is engagingly written and does an excellent
job of conveying both the immense drama of the
trans/circLim-Pacific migration, and the critical
nature of the Yellow Sea to its migrants.

The opening chapter, by Alaskan biologist
Brian McCaffery, sums up the volume’s theme:

ORNITHOLOGICAL LITERATURE

817

“Time is Running Out.” Subsequent chapters
include “Shorebird Lifestyles,” “Flyways,”
“The Tundra,” “A Southern Holiday?,” “Tidal
Flat Specialists,” and “International Partner-
ships.” Nial Moores, founder of Birds Korea,
and South Korean conservationist Ju Yung-Ki,
end the book with “The Heart of the Fly way,”
directly discussing the threats and opportunities in
the Yellow Sea region.

I have just one criticism. For a book at whose
heart lies geography, the general absence of maps
(except for a few small ones that show overall
flyway patterns), is puzzling and unfortunate. Most
strangely, there is no map of the Yellow Sea itself,
leaving those unfamiliar with the region at a loss to
easily locate the many sites mentioned in the text
and captions. This single failing aside. Invisible
Connections is a gem-and conservationists every-
where can only hope it has the desired effect. —
SCOTT WEIDENSAUL, 778 Schwartz Valley
Road, Schuylkill Haven, PA 17972, USA; e-
mail: scottweidensauK® verizon.net

A PHOTOGRAPHIC GUIDE TO THE BIRDS
OF JAMAICA. By Ann Haynes-Sutton, Audrey
Downer, Robert Sutton, and Yves-Jacques Rey-
Millet. Princeton University Press, Princeton,
New Jersey, USA. 2010: 304 pages, 6 figures,
650 color photographs, and 220 maps. ISBN: 978-
0-691-14391-0. $29.95 (paper). — Peter Vogel and
Catherine Levy asserted that “Conservation of
habitat is the primary measure that will save the
birds of Jamaica... Most non-government organi-
zations are lacking the financial and human
resources to carry out the work that is necessary,
such as advocacy addressed to the government
and decision makers” (Raffaele et al. 1998:30, A
Guide to the Birds of the West Indies, Princeton
University Press, Princeton, NJ, USA). These
words seem particularly prescient today, as the
Jamaican government is poised to sell the Font
Hill Nature Preserve, an extraordinary stand of
old-growth mangroves (especially black man-
grove, Avicennia genninans), ponds, and beaches,
for an 800-unit resort hotel (full disclosure: I’ve
worked there since 1986). Field guides can’t by
themselves protect threatened habitats, wide-
spread in Jamaica, but do comprise an important
tool to educate people able to influence decision
makers.

The new Photographic Guide to the Birds of
Jamaica is the first book to occupy the niche of

comprehensive bird guide for Jamaica. It’s
beautifully produced, compact (127 X 190 mm
[5" X 7.5"|), and durably bound, using semi-
gloss paper in an easy-to-use format with
typically a page per species, or two pages in
the case of most endemics. Rey-Millet treats us
to gorgeous photographs — up to five per species,
some filling the page — typically single-flash,
and perfectly exposed. The multiple photographs
accompanying many accounts highlight variation
in posture, vantages on plumage, and age/male
or female.

A comprehensive, stand-alone Jamaican field
guide will benefit Jamaicans in particular, and is
long overdue considering the island’s species
richness and endemism: 307 bird species — 127
breeding and 180 regular migrants. Forty-eight
species (> 15% of the avifauna) are either
endemic species or subspecies, or Caribbean
endemics. Jamaica has more endemic bird spe-
cies— vividly described in the first chapter of
Roland H. Wauer’s A Birder’s West Indies: an
Island-by-Island Tour (1996, University of Texas
Press, Austin, USA) — than any other Caribbean
island including much larger Cuba and, for
reasons not entirely understood, but probably
linked to the Caribbean region’s complex geolog-
ical history. Jamaica has its own tody (Todus
todus), and four endemic genera: Trochilus (two
streamertail hummingbird species), Loxipasser
anoxanthus (Yellow-shouldered Grassquit), Eu-
neornis campestris (Orangequit), and Nesopsar
nigerrimus (Jamaican Blackbird). Several taxa
challenge birdwatchers, including 23 sandpipers,
25 wood warblers, 12 pigeons/doves, and eight
tyrannid flycatchers.

This new photographic field guide is organized
into an Introduction (19 pages), and .sections on
“Birding in Jamaica” (8 pages), “How to use this
book” (3 pages), and species accounts (236
pages). The introduction covers Jamaica’s loca-
tion, climate, geology and geomorphology, to-
pography, and habitats (12 pages); composition
and origins of the avifauna; bird movements
including winter, austral, and elevational migra-
tions; conservation issues and threats, reinforcing
the importance of habitat and invasive species;
and a history of Jamaican ornithology including
bibliography. The birding section briefly de-
scribes and maps 16 locations with distinctive
habitats (including photographs) and species. This
birding section also provides useful information
for birders, such as trip tips, issues of security.

818

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. 4. December 2010

hazards, and — usefully — the lead author’s e-mail
address.

The species accounts are standardized for quick
reference to all species names, taxonomy (e.g.,
monotypic, polytypic), body length, plumage
descriptions, similar species, voice, and habitat
and behavior — foraging, diet, and nesting. Each
account also contains a boxed description of
geographic range and status in Jamaica (e.g.,
endemic, resident, winter visitor, transient; and
abundance category); standard country map,
color-coded and contoured for up to two levels
of abundance/reliability of occurrence; and rele-
vant lUCN redlist category (critically endangered,
endangered, vulnerable, near threatened).

Seven helpful appendices include “Probably
extinct species” (up to 6 in total, 3 illustrated, but
not the 2 macaws and a parrot), “Vagrants” (77
of the total 307 species listed with plumage
descriptions), “Species endemic to Jamaica” (30
species), “Subspecies endemic to Jamaica” (19
species), “Caribbean endemic species and sub-
species recorded in Jamaica” (18 species), and
a list of species with commercially available
vocalizations. The index includes just Latin
binomial and English common species names,
and a quick index inside the back cover includes
65 of the primary bird groups, mostly families. I
found Just a handful of errors, mostly typographic,
but Colombia was misspelled.

Haynes-Sutton and Rey-Millet intended and
accomplished, in collaboration with Downer and
Sutton (neither of whom, sadly, lived to enjoy the
finished product), a major improvement on
Downer’s and Sutton’s 1990 edition (Birds of
Jamaica — A Photographic Field Guide, Cam-
bridge University Press, Cambridge, UK). The
new guide most importantly includes photos of all
Jamaican species except the vagrants: 627 bird
photos of 219 species, compared to just 64 bird
photos of 47 species, emphasizing endemics, in
the Downer and Sutton edition. The new book
also includes or expands “habitats and behavior”
sections for all photographed species.

Several Caribbean field guides benefit from
regional comprehensiveness, but at the expen.se of
local detail. Most notable for decades was James
Bond's Birds of the West Indies (on my bookshelf
the 5th Edition, 1993, Houghton Mifflin Company,
Boston, MA, USA). A guide of the same title was
published by Helm in 1998, and by Harper Collins
in a new 2010 edition, with standard-format
illustrations of 632 specie.s — including visitors

and vagrants, and all residents except for Trinidad
and Tobago — but lacking illustrations of racial
variation among islands, and with short species
accounts. Two linked guides are also valuable:
Raffaele et al. (1998) presents excellent accounts
of all species including migrants, at the costs of
bulkiness for a field guide and separation of
species plates from accounts. Raffaele et al.
(2003, Princeton Field Guides: Birds of the West
Indies, Princeton University Press, Princeton, NJ,
USA) subsequently produced an efficient field
guide by shortening species accounts, located on
opposite pages to illustrations, in a smaller,
durable package. The latter two guides facilitate
comparisons by grouping similar species on plates
(basically the same plates, except for some larger
format, single-species illustrations in the 1998
guide). The new Jamaican guide best captures the
birds’ vivid colors, and lavishes more attention on
Jamaica’s birds by focusing geographically,
including mapped species distributions within
the island and more detailed species accounts in
some cases. Eor example, more information is
available to distinguish Sad from Stolid flycatch-
ers (Myiarchus barhirostris, and M. stolidus,
respectively), and 1 learned that the Orangequit
ranges to sea level only on Jamaica’s wetter
northeast coast. No electronic app (e.g., for
iPhone) is yet available for Jamaican birds as
for Birds of Dominican Republic and Haiti (http://
press.princeton.edu/blog/20 1 0/06/02/the-birds-of-
the-dominican-republic-and-haiti-iphone-app-
available-all-proceeds-go-to-disaster-relief/).

This new photographic guide will facilitate
identifying and appreciating all regularly encoun-
tered Jamaican birds. The more we, and especially
Jamaicans, come to know the island’s avifauna the
better, since environmental threats are here to
stay.— THOMAS W. SHERRY, Professor,
Department of Ecology and Evolutionary
Biology, 400 Boggs Hall, Tulane University,
New Orleans, LA 70118, USA; e-mail:
tsherry@lulane.edu

DO HUMMINGBIRDS HUM? EASCINAT-
ING ANSWERS TO QUESTIONS ABOUT
HUMMINGBIRDS. By George C. West and
Carol A. Butler. Rutgers University Press, Piscat-
away. New Jersey, USA. 2010: 185 pages, 32
figures, and 8 color plates. ISBN: 978-0-8135-
4738-1. $21.95 (paper). — This book contains 87
questions about hummingbirds organized into

ORNITHOLOGICAL LITERATURE

819

nine chapters, plus four appendices and an
extensive reference section. Hummingbirds are
the favorites of many birders, casual observers,
gardeners, homeowners, and ornithologists piqu-
ing curiosity with their extreme lifestyles and
biology, tiny size, bright colors, and engaging
behaviors.

George West and Carol Butler have teamed up
to produce this fourth in the Rutgers Animal Q&A
Series. Butler authored the three previous titles in
this series. West is a Professor Emeritus of
Zoophysiology at the University of Alaska, Fair-
banks, and a researcher on the physiology and
ecology of birds. He bas publisbed more than one
hundred scientific papers, provided hundreds of
illustrations for others, and recently published the
revised and updated A Birder's Guide to Alaska.

The book is organized into nine main chapters:
Hummingbird Basics, Systems and Senses, Feath-
ers and Bones, Reproduction, Flight and Migra-
tion, Dangers and Defenses, Attracting and
Feeding, Identifying and Photographing, and
Research and Conservation. There are four
appendices covering hummingbird garden plants,
places to see hummingbirds, hummingbird orga-
nizations, and recommended reading and web
sites. A useful feature is tbe References section,
which follows the organization of the book and
provides the literature support for many of the
answers. This particular section might have been
made easier to use by boldfacing the chapter titles,
as the questions in each of the nine chapters restart
numbering again from 1. Thus, there are as many
as nine questions with the same number in this
section.

As a reviewer, I read the book sequentially, but
the format allows, and is intended for, readers to
simply look up a topic for which they seek an
answer. The answers are thorough and well
researched, but they are also self-contained for
the most part even when a topic is complex, using
repetition of material to facilitate as clear an
understanding as possible by reading only the
answer to the relevant question. There are internal
references to other answers for more complex

topics. As a hummingbird researcher who gives
frequent public programs, 1 get asked a lot of
questions, and it appears that virtually every
possible question about hummingbirds has been
included in this book. Obvious questions like
"How many species of hummingbirds are there in
the world?" and “Where is the best place to bang
my hummingbird feeder?" to less typical ones
like "When do hummingbirds molt?" and, of
cour.se, the question that titles the book, are
covered with equal depth and comprehensiveness.

As an Eastern birder, I do notice a western
flavor to many of the answers, but given that the
majority of U.S. species are western this makes
perfect sense. In a few instances, exceptions that
the eastern Ruby-throated Hummingbird {Archi-
lochus colubris) presents are overlooked, includ-
ing for example the role that insects may have in
their diet for fat deposition prior to migration. One
minor quibble is that my personal favorite bird
name. Rainbow Starfrontlet {Coeligena iris), did
not make the list of "Descriptive Hummingbird
Names".

Few actual eiTors are evident. The most imme-
diately apparent is the top photograph on color plate
F, which is captioned as a Chestnut-breasted
Coronet {Boissonneaua matthewsii) but is clearly
not that species, but rather is a Shining Sunbeam
{Aglaeactis cupripemus). Another eiTor appears in
Figure 23, where breeding and winter ranges, and
migration routes, are shown for both Rufous
Hummingbird (Selasphoriis rufus) and Ruby-
throated Hummingbird. There is a migration aiTow
for Ruby-throated from the Yucatan of Mexico to
the Florida Peninsula. Robinson et al. (1996, The
Birds of North America. Number 204.) do not depict
this migration route, and the species is extremely
scarce anywhere in the Caribbean. These few minor
issues and errors could easily be remedied in a
second printing but, regardless. I highly recom-
mend this book for anyone who has an interest in
hummingbirds, or interacts with the public about
birds in any way. — ALLEN T. CHARTIER,
1442 West River Park Drive. Inkster. Ml 48141,
USA; e-mail: amazilial@comcast.net

The Wilson Journal of Ornithology 122(4):820-832, 2010

PROCEEDINGS OE THE NINETY-FIRST ANNUAL MEETING

JOHN A. SMALLWOOD, SECRETARY

The Ninety-first Annual Meeting of the Wilson
Ornithological Society was held from Thursday,
20 May, through Sunday, 23 May 2010, on the
campus of Hobart and William Smith Colleges in
Geneva, New York. Mark E. Deutschlander, Chair
of the Biology Department at Hobart and William
Smith (HWS) Colleges, chaired the Local Com-
mittee, which also included Kristi Hannam of
SUNY Geneseo, Sara Morris of Canisius College,
Margaret Voss of Behrend College at Penn State
Erie, John Waud of the Rochester Institute of
Technology, Erica Cooney-Connor and Nancy
Wilde of HWS Conference and Events, and Pat
Heieck of HWS Dining Services.

The Council met from 0900 to 1600 hrs on
Thursday, 20 May, at the Finger Lakes Institute
on the HWS campus. That evening there was an
ice-breaker social for conferees and guests at the
HWS Scandling Center.

The opening session on Friday convened in the
Albright Auditorium at 0841 hrs with welcoming
remarks from Mark E. Deutschlander, WOS
President E. Dale Kennedy, and WOS Second
Vice-President and Chair of the Scientific Pro-
gram Committee, Robert L. Curry. After present-
ing several items of information for the conferees,
Vice-President Curry concluded the opening
ceremony by introducing Edward H. Burtt Jr.,
who in turn introduced the Margaret Morse Nice
Lecture and the plenary speakers, Robert B. Payne
and Laura Payne, who delivered an informative
presentation, “Brood parasitism in cuckoos,
cowbirds, and African finches.”

The scientific program included 44 papers
organized into 10 sessions, 30 posters, and two
symposia consisting of six papers on migratory
physiology and energetics, and six papers on the
effects of energy development on birds. At the
conclusion of oral presentations on Friday after-
noon, the conferees were treated to a screening of
a remarkable documentary film on the search for
Ivory-billed Woodpeckers {Canipephilus princi-
palis) near the town of Brinkley, Arkansas. The
evening concluded with a reception in the
Vandervort Room of the Scandling Center, which
coincided with the poster session.

The Local Committee hosted birding forays in
the vicinity of the conference site each day from

Thursday through Sunday. In addition, longer
trips were scheduled for Thursday to the Cornell
Laboratory of Ornithology; for Friday and Satur-
day, on a tour of the Finger Lakes Region known
as the Ontario Pathways; and on Sunday, to the
Montezuma Wetlands Complex.

The conferees enjoyed a social hour prior to the
annual banquet, which also was held in the
Vandervort Room. The evening events included
an enjoyable dinner, and afterwards WOS Presi-
dent E. Dale Kennedy joined those assembled in
thanks to the many persons whose hard work had
resulted in a successful conference. President
Kennedy also thanked the three elected members
of Council who had completed their terms of
office, Robert S. Mulvihill, Timothy J. O’Connell,
and Mia R. Revels, and welcomed the three newly
elected members of Council, Mark E. Deutsch-
lander, Paul G. Rodewald, and Rebecca J. Safran.
The following WOS awards and commendations
also were presented:

MARGARET MORSE NICE MEDAL

(for the WOS plenary lecture)

Robert B. Payne and Laura Payne, “Brood
parasitism in cuckoos, cowbirds, and African
finches.”

EDWARD’S PRIZE

(for the best major article in volume 121 of The
Wilson Journal of Ornithology)

W. Andrew Cox and Thomas E. Martin,
“Breeding biology of the Three-striped Warbler
in Venezuela: a contrast between tropical and
temperate parulids.”

STORRS L. OLSON PRIZE

(for the best book review in volume 121 of The
Wilson Journal of Ornithology.)

Richard C. Banks, “The white-cheeked geese:
Branta canadensis, B. maxima, B. "fawrensis",
B. hutchinsii, B. leucopareia, and B. minima.
Ecophysiographic relationships, biogeography,
and evolutionary considerations. Volume 2. West-
ern taxa, biogeography, and evolutionary consid-
erations.”

820

ANNUAL REPORT

821

WILLIAM AND NANCY KLAMM
SERVICE AWARD

(for clislingLiished service to the Wilson Ornitho-
logical Society)

Jerome A. Jackson.

LOUIS AGASSIZ EUERTES AWARDS

Jessica Oswald, University of Florida, “The
biogeography of birds in Peruvian tropical dry
forests.”

Christopher Tonra, University of Maine, ”De-
temiining the relative influence of environmental
and endogenous factors on the transition from non-
breeding to breeding in migratory songbirds.”

GEORGE A. HALL/HAROLD E.

MAYFIELD AWARD

Nanette Mickle, of Woodbridge, Virginia,
“Tracking long-distant songbird migration (Pur-
ple Martins) using geo-locators.”

PAUL A. STEWART AWARDS

Allison Cox, University of Missouri, “Natal
dispersal of Red-bellied Woodpeckers in a
fragmented landscape.”

Kira Delmore, University of British Columbia,
“Divergent migratory behaviors as post-zygotic
barriers to interbreeding in hybrid zones.”

Luke Powell, Louisiana State University, “Us-
ing home range estimates and 30 years of mist-net
captures to determine the effect of land use history
on dispersal of Amazonian birds.”

Barry Robinson, University of Alberta, “Arctic
Peregrine Falcons as indicators of trophic dynam-
ics in coastal ecosystems.”

Taza Schaming, Cornell University, “The
impact of whitebark pine mortality on Clark’s
Nutcracker demography and habitat use.”

Daphna Shaw, University of Florida, “Male
parental investment, relatedness, population densi-
ty, and territorial proximity as factors promoting
extra-pair paternity in the Northern Mockingbird.”

Deborah Visco, Tulane University, “Ontoge-
netic mortality rates in a neotropical understory
pas.serine.”

Kara-Anne Ward, University of Windsor,
“Female mate search tactics in Long-tailed
Manakins: a novel tracking system for monitoring
social interactions.”

ALEXANDER WILSON PRIZE

(for best student paper)

Stephanie G. Wright, Villanova University,
“Hybrid chickadee vocalizations change as the

hybrid zone moves northward in southeastern
Pennsylvania.”

LYNDS JONES PRIZE

(for best student poster presentation)

Jason M. Townsend, SUNY-College of Envi-
ronmental Science and Forestry, “Mercury bio-
accumulation in Calhanis thrushes along an
elevational gradient.”

NANCY KLAMM BEST UNDERGRADUATE
STUDENT ORAL PAPER AWARD

Christina Masco, Cornell University, “Individ-
uality and recognition in the Great Black-backed
Gull, Lams marinus."

NANCY KLAMM BEST UNDERGRADUATE
STUDENT POSTER AWARD

(Their posters judged a tie, two students share this
year’s award.)

Anne Lugg, Kutztown University, “Aging
House Wren nestlings based on feather tract
development, wing chord, and head length.”

Jordan S. Kalish, Ohio Wesleyan University,
“Bacteria and fungi in the plumage of birds of
prey.”

WILSON ORNITHOLOGICAL SOCIETY
TRAVEL AWARDS

Aubrey Alamshah, Ohio Wesleyan University,
“Seasonal changes in maintenance behavior of
the House Sparrow {Passer doniesticiis)."

Katie Glower, University of Minnesota, “The
effects of rotational grazing on relative abundance
of grassland songbirds.”

James Garabedian, Frostburg State University,
“Habitat associations of over-wintering birds in
restored hayfields in the Manassas National
Battlefield Park, Virginia.”

Jessie Hogue-Morgenstern, Denison University,
“The relationship of migrant avian frugivory and
honeysuckle (Lonicera) management in the fall.”

Gustavo Londoho, University of Florida,
“Breeding biology of five species of high eleva-
tion Andean tanagers.”

Lawrence Long, The Ohio State University.
Suivey ol Lake Erie island passerine nest
predation.”

Anne Lugg. Kutztown University, “Aging
House Wren nestlings ba.sed on feather ti'ac't
development, wing chord, and head length.”
Adrian Monroe, Oklahoma State University,

822

THE WILSON JOURNAL OL ORNITHOLOGY • Vul. 122, No. 4. December 2010

"Winter bird response to heterogeneity-based
management in a tallgrass prairie."

Erica Mueller, Montclair State University, “A
spectrographic analysis of American Kestrel
(Fcilco sparve fills) vocalizations. Do broodmates
sound more like each other than non-related
broods?"

Benjamin Padilla, Gordon College, "The
effects of natural and human induced edges on
foraging flocks of wintering songbirds in southern
New England."

Monika Parsons, University of Maine, "Quan-
tifying incubation behavior of Common Eiders
nesting on Jordan’s Delight Island, Maine."

Ashley Rathman, Kutztown University, "The
effect of Wood Thrush hosts on the survival of
Brown-headed Cowbird eggs and nestlings.”

Sonya Richmond, University of Toronto, "In-
Huence of adult age on Rose-breasted Grosbeak
nestling provisioning rates in stands harvested by
single-tree selection."

Marla Steele, Oklahoma State University,
"Effects of cold front passage on migrant raptors
at Shirakaba Pass, Nagano Prefecture, Japan,
autumn 2000-2009."

Brynne Stumpe, Canisius College, "The effects
of a residential wind turbine on bird behavior.”

Emily Thomas, Penn State University, “Effects
of oil and gas development on songbirds."

Judit Ungvari-Martin, University of Florida,
"Variation in understory forest bird communities
of Amazonian forests on different soils."

Gretchen Wagner, CUNY Hunter College,
"Patterns of Brown-headed Cowbird parasitism
on Eastern Phoebes: a ten-year comparison."

Whitney Wiest, University of Delaware, “De-
velopment of a bird community integrity index to
monitor salt marsh condition at national wildlife
refuges.”

Marie Wilson, University of the South, "The
effects of exurbanization on the food and habitat
of Pileated Woodpeckers (Dryocopiis pileatiis)."

Stephanie Wright, Villanova University, "Hy-
brid chickadee vocalizations change as the hybrid
zone moves northward in southeastern Pennsyl-
vania."

Selection committee for the Nice Medal: James
D. Rising (Chair), Robert C. Beason, Robert L.
Curry, Jerome A. Jackson, E. Dale Kennedy, Sara
R. Morris; for the Edwards Prize: Clait E. Braun
(C’hair), James A. Cox, and Kevin Winker; for the
Olson ITizc: Clait E. Braun (Chair), and Slorrs L.
Olson; for the Klamm Service Award: Sara R.

Morris (Chair), Richard C. Banks, Robert C.
Beason, Robert L. Cumy, and E. Dale Kennedy;
for the Fuertes, Hall/Mayfield, and Stewart
Awards: Carla J. Dove (Chair), John Bates, Terry
Chesser, Greg Farley, Rob Faucett, Russ Green-
berg, James Hare, Rebecca Holberton, Tom
Jensen, E. Dale Kennedy, Sarah Kingston, Kevin
McCracken, Chris Milensky, Patricia Escalante
Pliego, Shane Pruett, Nate Rice, Scott Robinson,

B. A. Schreiber, Sarah Sonsthagen, John Swaddle,
Keith Tarvin, Pepper Trail, Jeff Walters, James
Whatton, Chris Witt, Margaret A. Voss, and
Douglas W. White; for the Alexander Wilson
Prize, the Lynds Jones Prize, and the Nancy
Klamm undergraduate presentation awards: Ro-
bert C. Beason (Chair), Edward H. Burtt Jr., John

C. Kricher, Sara R. Morris, Mia R. Revels, John
A. Smallwood, Margaret A. Voss, Doris J. Watt,
and Douglas W. White; and for the WOS Travel
Awards: Timothy J. O’Connell (Chair), Greg
Keller, and Mia R. Revels.

COMMENDATION

WHEREAS THE WILSON ORNITHOLOGI-
CAL SOCIETY held its annual meeting in
Geneva, New York, on the campus of Hobart
and William Smith Colleges, overlooking the
stunningly beautiful Seneca Lake, and

RECOGNIZING that the Committee on the
Scientific Program, under the adroit direction of
Robert L. Cuiry, assisted by Robert C. Beason,
Susan B. Smith, and Scott H. Stoleson, arranged
and executed an exemplary exposition of oral and
poster presentations, which included symposia
exploring the physiology and energetics of
migratory birds, and the effects of energy
development on birds, and

RECOGNIZING that the Committee on Local
Arrangements, through the efforts of Kristi
Hannam, Sara Morris, Margaret Voss, John
Waud, Erica Cooney-Conner, Nancy Wilde, and
Pat Heieck, and particularly Chairperson Mark
Deutschlander, who answered every request with
a "yes” and whose many able assistants made
tho.se events come to pass, provided ah excellent
conference venue in a comfortable and welcoming
campus environment in which dining experiences
exceeded expectations, and

WHEREAS the conferees found this meeting
both informative and enjoyable,

THEREFORE BE IT RESOLVED that the
Wilson Ornithological Society commends the
Committee on the Scientific Program, the Com-

ANNUM. REPOR I

823

mittee on Local Arrangements, and Hobart and
William Smith Colleges tor a most suceessfiil and
rewarding meeting in Geneva.

BUSINESS MEETING

President E. Dale Kennedy called the annual
business meeting to order at 1 136 hrs, Friday, 21
May 2010 in the Geneva Room of the Warren
Hunting Smith Library on the campus of Hobart
and William Smith Colleges, Geneva, New York.
Noting that a quorum was present, she introduced
Secretary John A. Smallwood, who presented a
synopsis of the Council meeting, which had taken
place the previous day. As of 30 April 2010, the
society’s membership stood at 1,579 individuals,
including 209 students, 216 international mem-
bers, and 128 new members of all membership
categories. While the renewal cycle was still
underway, at that time the retention rate was 89%.
In addition, 250 libraries and institutions sub-
scribed to The Wilson Journal of Ornithology.
Secretary Smallwood then asked those assembled
to stand in recognition of the following members
who were recently deceased: William A. Burnham
and Carl D. Marti, both of Boise, ID.

Secretary Smallwood reported the society is in
very good financial state with a healthy endowment
of —$2,500,000, and noted the value of these
investments was increasing as the economy im-
proves. On behalf of Council, the Secretary offered
sincere gratitude to the WOS Finance, Audit, and
Investment Committee, chaired by Allan Keith, for
its astute management of societal assets.

Secretary Smallwood further reported that the
Membership Committee, chaired by Timothy J.
O’Connell, had revised and updated the WOS
membership recruitment poster. Noting that the
poster was currently on display, the Secretary
invited those assembled to “check it out,’’ to join
the society and, if already a member, to encourage
nonmember colleagues to Join.

It was the Secretary’s pleasure to report that
Council had unanimously re-elected Clait E.
Braun as Editor of The Wilson Journal of
Ornithology, and he congratulated Editor Braun
for doing a superb Job, noting the Journal had
grown to nearly a thousand pages, the quality is
first rate, and his record for collecting page
charges from authors has been outstanding.

Finally Secretary Smallwood announced that
Council had proposed two changes to the Bylaws.
First, under Article III, Section and Section 6,
Council proposed to delete the sentences that

require the President and the Treasurer to hold
specific positions relative to OSNA. Second,
Council proposed that the references to the
society’s Journal be changed from the former
name. The Wilson Bulletin, to the current name.
The Wilson Journal of Ornithology. These pro-
posed changes to the Bylaws will be voted upon
by the membership at the next annual meeting, in
Kearney, Nebraska, March 2011.

Secretary Smallwood then gave the Boor back
to President Kennedy, who summarized the
Treasurer’s report in the absence of the Treasurer
(Melinda M. Clark), and introduced Editor Braun,
who summarized the Editor’s report. The written
reports of the Treasurer and the Editor are
included below.

Sara R. Moms, Chair, presented the report of the
Nominating Committee, which also included
Robert C. Beason, Mary Bomberger Brown, and
Jameson Chase: President, E. Dale Kennedy; First
Vice-President, Robert C. Beason; Second Vice-
President, Robert L. Curry; Secretary, John A.
Smallwood; Treasurer, Melinda M. Clark; Mem-
bers of Council for 2010-2013 (three nominees for
three positions), Mark E. Deutschlander, Paul G.
Rodewald, and Rebecca J. Safran. Having asked for
additional nominations from the floor and hearing
none. President Kennedy closed the nominations as
a result of a motion by Robert L. Curry, seconded
by Jerome A. Jackson, which was unanimously
passed by voice vote. Further, Timothy J. O’Con-
nell moved and Jerome A. Jackson seconded that
the Secretary cast a single unanimous vote for the
slate of nominees. By acclamation it became so.

Second Vice-President Curry updated tho.se
assembled on the 201 1 meeting, which will be a
Joint meeting with the Association of Field
Ornithologists and the Cooper Ornithok:»gical
Society, and will occur 9 to 13 March, in Kearney.
Nebraska. Mary Bomberger Brown is serving as
the local host. Vice-President Curry also an-
nounced that preparations were underway for the
2012 meeting: the WOS will be participating in
the North American Ornithological Conference in
Vancouver. British Columbia, 14-18 August. This
meeting will take place on the campus of the
University of British Columbia, and Kathy Martin
will serve as the local chair.

Having completed the business at hand. Pres-
ident Kennedy inquired if any one present had
additional items ol business. Because no one did,
Jed Burtt moved and Ted Davis seconded that we
adjourn. This came to pass at 1204 hrs.

824

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, December 2010

REPORT OF THE TREASURER

Statements of Revenues and Expenses for the Year Ending 31 December 2009

2009 2009 2010

12 Months Actual Annual Budget Annual Budget

Revenues

Contributions $

Student Travel Research Fund

Van Tyne Library Book Fund

Sales - back issues

Sales - books - Van Tyne Library

Subscriptions

Page charges

Royalties

BioOne Electronic Licensing
Mailing list rental income
Memberships
Other income

Total revenues from operations $

Expenses

Journal publication costs:

Editorial honorarium $

Editor travel

Editor supplies/Mailings

Editorial assistance

Copyright expense

Printing - Journal

Artwork

Printing color plates

S

Operating expenses:

Postage & mailing - back issues S

Storage back issues

Van Tyne Library - expenses

OSNA management services

Credit card fees

Travel expenses-OSNA rep

Travel expenses - general and Treasurer

Travel expenses-Ornith Council

Meeting expenses

Membership expenses

Accounting fees

Insurance - D&O

Oftlce supplies

Postage - general

Other expenses

Filing fees

Discretionary expenses

$

Awards

Student banquet $

Membership awards

Hall/Mayfield

Stewart

Fuertcs

Klamm Service Award
Wilson. Lynds Jones. Klamm
Student travel grants

1.240

$

1.800

$

1,800

720

300

500

212

100

110

282

240

200

593

600

750

13.065

15,100

12,000

40,045

18,400

34,000

7,837

4,230

1,000

23,557

17,000

22,500

85

222

270

56,277

37,000

26,000

3,894

3,500

147,807

$

98,492

$

99,130

4,000

$

4.000

$

6,000

982

1,700

1,350

2,768

1.700

2,620

21,123

24,000

18,500

-

100

-

88,637

100.000

90,000

750

1.000

750

3,428

3,600

3,500

121,688

$

136,100

$

122,720

0

$

440

$

0

-

3.50

240

726

1 .000

1.200

14,193

24.300

16,000

3,730

2.200

2.100

-

1 .500

500

370

800

1.000

-

300

250

100

3,750

3,000

-

2.000

2,000

4.165

4,500

4,000

1 .500

1 ,650

1 .475

-

500

305

-

200

200

-

160

100

10

5

10

-

3.000

. 3,000

24,794

$

46.655

$

35,380

2.415

$

1 .200

$

1.200

-

600

600

1 .000

1 ,000

1 ,000

2.000

3.000

4.000

2.500

2,500

5,000

2.000

1 .000

700

700

700

4,700

9,000

5,000

ANNUAL REPOR I

825

Assets

Cash Investments:

Merrill Lynch - cash
Coamerica - Van Tyne checking
Van Tyne Univ. Michigan account
Sutton Fund - cash equivalents
Howland Mgmt - cash equivalent
Total cash and cash equivalents

Other Investments:

Merrill Lynch - common stocks
Merrill Lynch - corp bonds
Merrill Lynch - mutual funds
Sutton Fund - equities
Sutton Fund - corp bonds
Howland Mgmt - equities
Howland Mgmt - fixed income
Total Other Investments
Total Assets

FUND BALANCES
Fund Balances

Restricted funds - Sutton Fund
Unrestricted funds
Net Income

Fund balance - Klamm
Total Fund Balances

STATEMENT OF FINANCIAL POSITION

3 1 December 2009

$ 150,452
452
1.050
28,164
193,342
$ 373,460

ii 595.227
33.043
37.155
I 12,950
9.676
1 . 1 80.604
1 14.858
2.083.513
2.456.973

$

$

i 150.790
491,069
326.310
1.488.804
2,456.973

Award expenses

1 .739

3,()()()

2,553

$

15,054

$

2 1 ,800

$

19,853

Contrihutions

Support - Ornith Council

$

9,300

$

9,300

$

15,000

Support - Ornith Council, restricted to revision

_

_

costs

Am Bird Conservancy Dues

AAZN dues

250

250

$

9,300

$

9,550

$

15,250

Total Expenses

$

170,836

$

214,105

$

193,203

Expenses in excess of revenues before

$

(23,029)

$

(115,613)

$

(94,073)

investment income

Investment activity

Revenues

Investment earnings - budgeted

$

0

$

0

$

0

Realized gain/loss - ML

(14,008)

35,000

27,081

Realized gains/losses-Howland

(47,404)

40,000

(20,000)

Realized gains/losses - Sutton

(3,033)

5,300

(2,000)

Unrealized gain/loss - ML

216,075

21,000

(150.000)

Unrealized gain/loss - Howland

143,664

5,000

(200,000)

Unrealized gain/loss - Sutton

15,416

9,800

(40,000)

Investment earnings - ML

23,411

27,000

41,760

Investment earnings - Howland

30,241

63,000

32,695

Investment earnings - Sutton

4,217

3,900

3,673

Total revenues from investment activity

$

368,579

$

210,000

$

(306,791)

Investment fees

$

19,240

$

28,500

$

14,366

Investment revenues in excess of expenses

$

349,339

$

181,500

$

(321,157)

Total revenues in excess of expenses

$

326,310

$

65,887

$

(415,230)

Melinda M. Clark, Treasurer

826

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

EDITOR’S REPORT

We received 206 new manuscripts in 2009 and
published 124 dating from 2004 (1), 2005 (3),
2007 (21), 2008 (85), and 2009 (14) with a total
published page count of 914. This included 28
Book Reviews, 1 Memorial, 1 Service Award,
Proceedings of the Ninetieth Annual Meeting, List
of Reviewers, Index, and the Table of Contents
for Volume 121. Most papers published were
from North America north of Mexico but the
published material included 33 studies from 16
counties from outside of this area. Fifteen were
from South America (Argentina = 4, Bolivia = 1,
Brazil = 3, Chile = 1, Colombia = 1, Peru = 4,
Venezuela = 1), 12 were from Central America
(Costa Rica = 4, Ecuador = 4, Mexico = 3,
Panama = 1 ), and the remainder were from other
countries (Canary Islands and Spain = 2, China =
1, India = 2, Israel = 1). Over half of all
manuscripts received in a specific year are
rejected, withdrawn, or encouraged elsewhere in
the initial year of receipt. Others may linger and
eventually be withdrawn by the authors or the
Editor for failure of authors to respond.

Electronic submission of all manuscripts is now
encouraged effective as of 1 March 2010. We
earlier encouraged electronic submission from
authors outside of North America. All reviewers
are contacted electronically and a few still request
and return paper copy. Presently, we return paper
copy to authors of first reviews and marked
manuscripts, as it is still easier to protect reviewer
confidentiality and to provide personal edits of
manuscripts by paper copy. All subsequent
correspondence and edits are done electronically.
This process has some lingering issues (extremely
slow delivery of postal mail to some countries in
Central and South America, limited use of English
as a first language) resulting in .some delays in
receipt of materials and understanding what is
needed.

We identified submission of .several LPU’s
(Least Publishable Units) in 2009 and tried to
discourage this practice. We al.so strongly dis-
couraged manu.scripts for which the ‘important'
findings were published el.sewhere. The.se issues
are not easily di.scerned but reviewers can be
helpful in identifying questionable manu.scripts.
We receive manuscripts (as do most journals) that
were originally submitted elsewhere. We treat
the.se manu.scripts fairly but try to learn why they
were discouraged by other Journals. We afso were

threatened with legal action concerning author
sequence on a paper after it was published. Signed
copyright forms (showing author sequence) were
available for all authors and the action was
terminated. Thus, there are valid reasons why
we insist on having signed copyright forms on file
before any manuscript is published.

The editing and publication process has become
more time consuming each year, as there are
many details to consider. I understand why other
Journals have changed proces.se.s and procedures
complete with large numbers of Associate or
Assistant Editors and a variety of editorial staff.
Everything is more complex and one must be alert
to what is best for a particular situation.

Allen Press continues to be very helpful and to
do outstanding work preparing and printing The
Wilson Journal of Ornithology. Their work is
deeply appreciated, as it makes us look compe-
tent. The support of WOS Council and the
Publications Committee is also essential to
making our life easier.

Clait E. Braun, Editor

The reports of the standing committees are as
follows:

REPORT OE THE JOSSELYN VAN TYNE
MEMORIAL LIBRARY COMMITTEE

Janet Hinshaw, WOS Librarian, has provided
the following statistics regarding use of the Van
Tyne Library and the status of the library.

Loan.s

Number of loans: three; number of members to
whom loans were .sent: three; number of photo-
copies or scans provided: six.

Acquisitions

E.xchanges: A total of 120 publications was
received by exchange from 198 organizations or
individuals. We have lost several exchanges in the
past few years and have had to subscribe to keep
those Journals current.

Gifts: We received 24 publications from 20
organizations.

Subscriptions: We al.so received 33 publica-
tions from 23 sub.scriptions. We spent a total of
.$660.84 on sub.scriptions in 2009.

Donations: A total of 13 items was received
from members and friends. The.se included 12
Journal issues and one translation. Donors includ-
ed P. Ames and J. Hinshaw.

ANNUAL REPORT

827

Purchases: one journal issue and three books,
tor $144.95 in credit from Buteo Books.

Dispersal

Back Issues. Due to lack of demand in recent
years and having to clear out of the store room, we
have disposed of the bulk of our stock of back
issues of The Wilson Bulletin.

Gifts Sent Out: one book was donated to Dan
Hanley of the University of Windsor.

Duplicates: 26 books were sold for $2,600.00
credit with Buteo Books.

We remind members that we can lend books or
provide paper or electronic copies of articles from
items in the library. The Google scanning project
has not yet begun for the bird library, but is
tentatively scheduled for fall/winter 2010. I thank
our student worker, Theresa Gorman, for day-to-
day operations, and our secretary, Bev Dole, for
assistance with photocopying.

Jerome A. Jackson, Chair

REPORT OF THE
PUBLICATIONS COMMITTEE

The Publications Committee of the Wilson
Ornithological Society consists of Mary Bomber-
ger Brown (chair), Richard Banks, Michael
Hamas, Fred Lohrer, and Clait Braun (as editor
of The Wilson Journal of Ornithology). During the
past year, we focused our attention on three areas,
all with the goal of continuing to improve the
quality and stature of The Wilson Journal of
Ornithology.

We acquired free access to The Wilson Journal
of Ornithology on BioOne (www.bioone.org) for
Wilson Ornithological Society members. We
worked with Christina Berger at Allen Press and
Christopher Schneider at the Schneider Group/
OSNA to transfer WOS member information to
BioOne; there is now a link to BioOne on the
Wilson Ornithological Society web page. This
free society member access to BioOne offers us
several benefits: ( 1 ) it offers a tangible benefit to
membership; (2) it brings more attention to the
society, potentially increasing membership; and
(3) it increases journal usage. There is no cost to
the Wilson Ornithological Society for this service.

The Table of Contents of The Wilson Journal of
Ornithology is now distributed via e-mail to all
OSNA members before the journal is published.
Early release of the Table of Contents offers us
several benefits: (I) it advertises The Wilson

Journal of Ornithology and The Wifson Ornitho-
logical Society to OSNA members; (2) it helps
members stay abreast of the journal contents; (3)
it reminds members, including lapsed members, to
renew their society membership; and (4) it
encourages authors to consider submitting their
manuscripts to the Journal for publication. There
is no cost to The Wilson Ornithological Society
for this .service.

The guidelines for authors submitting manu-
scripts to The Wil.son Journal of Ornithology and
links to the guidelines on the web page were
updated. The guidelines now clarify the use of the
American Ornithologists’ Union checklist for
names of birds that occur in Mexico and the rest
of North America, and the International Ornitho-
logical Congress checklist for names of birds that
occur outside of North America. The prefeired
font for manuscript submissions has been clarified
as have details of figure/illustration preparation.

We continue to work with the Schneider Group/
OSNA to address the issue of the size of the print
run for The Wilson Journal of Ornithology. We
have not pursued adding a Twitter feed to The
Wilson Ornithological Society web page, al-
though the service is available through OSNA at
no cost to the society.

Mary Bomberger Brown, Chair

REPORT OF THE SCIENTIFIC
PROGAM COMMITTEE

The Committee on the Scientific Program
consisted of WOS Second Vice-President Robert
L. Curry (Chair), Robert C. Reason, Susan B.
Smith, and Scott H. Stoleson. Paper sessions were
moderated by Greg Farley, Elise Ferree, Mary
Garvin, Jerome A. Jackson, Rachel Muheim,
Timothy J. O’Connell, Daniel Shustack, Charles
Smith, Lindsay A. Walters, and Douglas W.
White. The symposium on new perspectives in
migratory physiology and energetics was moder-
ated by Susan B. Smith and Mark E. Deutsch-
lander, and the symposium on the effects of
energy development on birds was moderated by
Scott H. Stoleson.

PAPER SESSIONS

Symposium: New Perspectives in Migratory
Physiology and Energetics

Ulf Bauchinger and Scott McWilliams, Uni-
versity of Rhode Island, “Tissue turnover rate

828

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, December 2010

determines extent of phenotypic flexibility in
organ size of migrating birds.”

Lauren S. MacDade and Paul G. Rodewald,
The Ohio State University, and Kent A. Hatch,
Long Island University, “Contributions of emer-
gent aquatic insects to refueling by spring migrant
songbirds.”

Scott R. McWilliams, University of Rhode
Island, and Barbara Pierce, Sacred Heart Univer-
sity, “The fat of the matter: how dietary fatty acid
composition affects energy expenditure of birds
during migration.”

Jen Owen, Michigan State University, “Links
between energetic condition, immune function,
and disease susceptibility in migrant birds.”

Edwin Price, University of Western Ontario,
“Towards a theoretical framework for avian fat
storage.”

Robert J. Smith, University of Scranton,
“Arrival and transition into the breeding period:
the fitness consequences of timing and condition
for a landbird migrant.”

Symposium: The Effects of Energy
Development on Birds

Gregory A. George, James Sheehan, and Petra
Bohall Wood, West Virginia Cooperative Fish
and Wildlife Research Unit, and Harry Edenborn,
National Energy Technology Laboratory, “Avian
responses to gas well development in West
Virginia.”

Nels Johnson, The Nature Conservancy, “Can
natural habitats survive the energy revolution?”

Todd Katzner, National Aviary, Trish Miller
and Michael Lanzone, Powdermill Nature Re-
serve, David Brandes, Lafayette College, and
Robert Brooks, Pennsylvania State University,
“Threats to migrating Golden Eagles from
development of wind energy.”

Jeff Larkin, Joseph Grata, and Joe Duchamp,
Indiana University of Pennsylvania, and Laura
Patton, Kentucky Department of Fish and Wildlife
Resources. “Developing guidelines for the crea-
tion of Golden-winged Warbler breeding habitat
on reclaimed surface mines in southeastern
Kentucky.”

Emily H. Thomas, Pennsylvania State Univer-
sity, “Effects of oil and gas development on
songbirds.”

Kim Van Fleet, Audubon Pennsylvania, “En-
ergy development and fragmentation effects.”

General Sessions

William H. Barnard, Norwich University,
“Blood parasites in overwintering Rusty Black-
birds.”

Sean R. Beckett, Glenn A. Proudfoot, and Mary
Ann Cunningham, Vassar College, “Large-scale
movement and migration of Northern Saw-whet
Owls.”

J. Alan Clark, Fordham University, “Tracking
birds migrating at night through an urban-rural
coiridor and quantifying the effects of light and
noise pollution.”

Katie Clower, University Minnesota, “The
effects of rotational grazing on relative abundance
of grassland songbirds.”

Robert L. Curry, Villanova University, “Var-
iation in chickadee morphology through time in a
moving hybrid zone.”

Christine Eldredge, Brian Monaco, and Mark E.
Deutschlander, Hobart and William Smith Col-
leges, “Seasonal energetic differences between
migratory, non-migratory, and irruptive song-
birds.”

Harold Eyster, Chelsea, Michigan, “Tolmie’s
MacGillivray’s Warbler: the story of a name.”

Jeanne M. Fair, Los Alamos National Labora-
tory, and Ellen Paul, The Ornithological Council,
“Guidelines to the use of wild birds in research.”

Greg H. Farley, Andree Brisson, Alex Galt, and
Stephanie Kane, Fort Hays State University,
“Patterns in migratory songbird capture between
1966-1980 and 1995-2009 at an autumn banding
station in the High Plains.”

George L. Farnsworth, Dorothy B. Engle, Brett
E. Schrand, and Christopher C. Stobart, Xavier
University, and Rebecca B. Desjardins, North
Carolina State Museum of Natural Sciences,
“Seasonal and geographic variation in nestling
sex ratio in the Northern Mockingbird.”

Elise Ferree and Janis Dickinson, Cornell
University, “Lifetime survival and reproductive
success of extra-pair and within-pair Western
Bluebirds {Sicilia mexicana)."

James E. Garabedian and Frank Ammer,
Frostburg State University, and J. Edward Gates,
University of Maryland, “Habitat associations .of
over-wintering birds in restored hayfields in the
Manassas National Battlefield Park, Virginia.”

Douglas A. Gross, Pennsylvania Game Com-
mission, “Pennsylvania boreal conifer forests and
their bird communities: past, present, and future
potential.”

Jessie Hogue-Morgenstern and James Marshall,

ANNUAL REPORT

829

Denison University, "The relationship of migrant
avian Irugivory and honeysuckle (Lonicera)
management in the fall."

Jerome A. Jackson and Bette J. S. Jackson,
Florida Gulf Coast University, "Winter associa-
tion of birds with the invasive exotic Brazilian
pepper (Schinus terebinthifolius)."

Gustavo Adolfo Londoho, University of Flor-
ida, "Breeding biology of five species of high
elevation Andean tanagers."

Lawrence C. Long and James Marshall, The
Ohio State University, "Survey of Lake Erie
island passerine nest predation.”

Christina Masco and Thomas D. Seeley,
Cornell University, "Individuality and recogni-
tion in the Great Black-backed Gull, Larus
marinus."

Adrian P. Monroe and Timothy J. O’Connell,
Oklahoma State University, “Winter bird re-
sponse to heterogeneity-based management in a
tallgrass prairie."

Erica G. Mueller and John A. Smallwood,
Montclair State University, “A spectrographic
analysis of American Kestrel (Falco sparvehus)
vocalizations. Do broodmates sound more like
each other than non-related broods?"

Rachel Muheim, Lund University, John B.
Phillips, Virginia Polytechnic Institute and State
University, Deiring Hall, Blacksburg, Virginia,
and Mark E. Deutschlander, Hobart and William
Smith Colleges, “Sunrise and sunset polarized
light cues calibrate the magnetic compass in
migratory songbirds.”

Christopher Norment, SUNY College at Brock-
port, “Eifteen years of research on grassland birds
in New York State: habitat relations and manage-
ment implications.”

Timothy J. O’Connell and Andrew D. George,
Oklahoma State University, “Influence of seeded,
exotic grasslands on wintering birds in mixed-
grass prairie.”

Sarah E. Pabian and Margaret C. Brittingham,
Penn,sylvania State University, “Complex rela-
tionships between forest songbirds and soil
conditions.”

Monika Parsons and Frederick A. Servello,
University of Maine, and Cynthia S. Loftin,
USGS Maine Cooperative Fish and Wildlife
Research Unit, “Quantifying incubation behavior
of Common Eiders nesting on Jordan’s Delight
Island, Maine.”

Joel Ralston, University of Albany, “Integrat-
ing species distribution models and genetic data to

lorecasi the effects of climate change on genetic
diversity in boreal birds.”

Sonya Richmond and J. Malcolm, University of
Toronto, E. Nol, Trent University, and D. Burke,
Ontario Ministry of Natural Resources, “InlJu-
ence of adult age on Rose-breasted Grosbeak
nestling provisioning rates in stands harvested by
single-tree selection.”

Sarah T. Saalfeld and Warren C. Conway,
Stephen E. Austin State University, David A.
Haukos, Texas Tech University, and William P.
Johnson, Texas Parks and Wildlife Department,
“Nest suceess of Snowy Plovers in the Southern
High Plains of Texas.”

Matthew S. Savoca, David N. Bonter, Benja-
min Zuckerberg, and Janis L. Dickinson,
Cornell University, and Julie C. Ellis, Tufts
Cummings School of Veterinary Medicine,
“Nesting density an important factor affecting
chick growth and survival in the Herring Gull
{Larus argentatus)."

Richard R. Schaefer and D. Craig Rudolph,
USES Southern Research Station, “Southeastern
American Kestrel nesting habitat and nestling
provisioning in the West Gulf Coastal Plain.”

Chad L. Seewagen and Christopher G. Gu-
glielmo. University of Western Ontario, “Oven-
bird spatial behavior at an urban stopover site:
movement patterns, stopover durations, and the
influence of airival condition.”

Daniel P. Shustack, Massachusetts College of
Liberal Arts, and Amanda D. Rodewald, The Ohio
State University, “Nest predation reduces benefits
to early clutch initiation in an urbanizing
landscape.”

Dennis G. Siegfried, Daniel S. Linville, and
David Hide, Southern Nazarene University,
“Analysis of nest sites of the Resplendent Quetzal
(Paromachrus mociiuw): a relationship between
nest and snag heights.”

Charles R. Smith, Cornell University, "The
elements of science-based, adaptive bird conser-
vation and the public trust doctrine.”

Marla L. Steele and Timothy J. O’Connell,
Oklahoma State University, “Effects of cold front
passage on migrant raptors at Shirakaba Pass,
Nagano Prefecture, Japan, autumn 2000-2009.”

Scott H. Stoleson, USFS Northern Research
Station, “Does habitat choice in the post-breeding
.sea.son affect physiological condition of forest-
interior songbirds?”

Judit Ungvari-Martin, Ari Martinez, and Scott
K. Robinson, University of Florida, “Variation in

830

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4, December 2010

understory forest bird communities of Amazonian
forests on different soils.”

Gretchen F. Wagner and Mark E. Hauber,
CUNY Hunter College, “Patterns of Brown-
headed Cowbird parasitism on Eastern Phoebes:
a ten-year comparison.”

Lindsey A. Walters, Canisius College, and
Natalie Dubois and Thomas Getty, Michigan
State University, “Influences of mate quality
and population growth on House Wren clutch
size.”

Rebecca J. Whelan and Tera C. Levin, Oberlin
College, Jennifer C. Owen, Michigan State
University, and Mary C. Garvin, Oberlin College,
“Short-chain carboxylic acids from Gray Catbirds
(Dumetellci carolinensis) uropygial secretions
vary with testosterone levels and photoperiod.”

Douglas White and E. Dale Kennedy, Albion
College, “Gaps in nocturnal incubation in House
Wrens {Troglodytes aedon)."

Whitney A. Wiest and W. Gregory Shriver,
University of Delaware, and Hal Laskowski, U.S.
Eish and Wildlife Service, “Development of a
bird community integrity index to monitor salt
marsh condition at national wildlife refuges.”

Marie E. Wilson, Jordan M. Casey, David G.
Haskell, and Nicholas Hollingshead, University of
the South, “The effects of exurbanization on the
food and habitat of Pileated Woodpeckers (Dryo-
copus pdeatus)."

Stephanie G. Wright and Robert L. Curry,
Villanova University, “Hybrid chickadee vocali-
zations change as the hybrid zone moves north-
ward in southeastern Pennsylvania.”

POSTERS

Aubrey Alamshah and Edward H. Burtt Jr.,
Ohio Wesleyan University, “Seasonal changes in
maintenance behavior of the House Sparrow
( Passer domesticus)."

Andrew C. Alba and Gregory B. Cunningham,
St. John Eisher College, “The fox in the
henhouse; responses of chickens {Callus domes-
ticus) to the scent of a predator.”

Catherine C. AIsford and Dorothy 1. Fatunmbi,
Canisius College, Brenda S. Keith and Richard S.
Keith, Kalamazoo Nature Center, and Sara R.
Morris, Canisius College, “Does age matter? A
study of whether age affects stopover ecology of
migrant warblers.”

Robert C. Reason, Accipiter Radar Corpora-
tion, “Remote radar monitoring of bird activity
via the network.”

Martyna Boruta and Mark E. Deutschlander,
Hobart and William Smith Colleges, “Ultraviolet
reflectance and age/sex variation in Yellow
Warblers {Dendroica petechia)."

W. P. Brown and M. E. Zuefle, Kutztown
University, “Are bird images returned from
search engine results accurately identified?”

Jameson E. Chace, Salve Regina University,
and Steven D. Eaccio, Vermont Center for
Ecostudies, “Evaluation of breeding habitat
quality for a population in decline: the Canada
Warbler in northeastern Vermont.”

Kailey Chidester, Bethany Bashaw, and Mark
E. Deutschlander, Hobart and William Smith
Colleges, “Determining land quality for migra-
tory birds near the Lake Ontario shoreline: a study
of daily mass gain during fall and spring
migration.”

Kelly Eox, Alyssa Caffarelli, and Lindsey
Walters, Canisius College, and Brian S. Maitner,
Rice University, “Nest defense and parental
investment in the House Wren.”

Casey Eranklin, Jordan Youngman, Mianna
Molinari, and Mark E. Deutschlander, Hobart
and William Smith Colleges, “Energetic condi-
tion of spring migrating North American warblers
at a northern stopover site: a study of Breeding
and Insurance hypotheses.”

Samuel Georgian, Paul Overdorf, and Mark E.
Deutschlander, Hobart and William Smith Col-
leges, “Determining the role of the Insurance
Hypothesis in the stopover ecology of two
kinglets and two flycatchers.”

Erin Greenlee and Ian Hamilton, The Ohio
State University, “Distributional shifts of breed-
ing birds in Ontario, Canada.”

Nathan Grosse, Christopher Norment, and
Mark Norris, SUNY Brockport, and Heidi Ken-
nedy, NY Department of Environmental Conser-
vation, “Area effects removed: vegetation char-
acteristics and grassland bird abundance in a
western New York field.”

Margret I. Hatch, Penn Slate Worthington
Scranton, and Robert J. Smith, University of
Scranton, “Annual variation in arrival of long-
distant migrants and a comparison of first capture
dates and condition between verified breeders and
presumed migrants.”

Jordan S. Kalish and Edward H. Burtt Jr., Ohio
Wesleyan University, “Bacteria and fungi in the
plumage of birds of prey.”

Rachel Kushner, Christy Conley, and James S.
Marshall, Denison University, “Patterns of inter-

ANNUAL REPOR I

831

terence competition in overwintering central Ohio
birds.”

Anne Liigg, W. Brown, D. Alexander, M.
Zuefle, and T. Underwood, Kutztown University,
“Aging House Wren nestlings based on feather
tract development, wing chord, and head length.”

Mianna Molinari and Mark E Deutschlander,
Hobart and William Smith Colleges, “The
influence of temperature on energetic condition
in spring migrating warblers.”

Brad Mudrzynski and Christopher Norment,
SUNY College at Brockport, “Early successional
habitat characteristics and fruit abundance influ-
ence stopover use of fall migrating songbirds.”

Laura M. Niczyporowicz and Todd J. Under-
wood, Kutztown University, and Roland R. Roth,
University of Delaware, “The effectiveness of
constant effort mist-netting in estimating abun-
dance and reproductive success of a Wood Thrush
population.”

Nathan Olszewski, Kevin Sobol, and Lindsey
Walters, Canisius College, “Male response to egg
color in the House Wren.”

Paul Overdorf and Mark E. Deutschlander,
Hobart and William Smith Colleges, “Using wing
morphology and mass to model the potential flight
distances of passerines at a stopover site near
Lake Ontario.”

Benjamin J. Padilla, Ashley Ballou, and
Gregory S. Keller, Gordon College, “The effects
of natural and human induced edges on foraging
flocks of wintering songbirds in southern New
England.”

Ashley N. Rathman and William P. Brown,
Kutztown University, and Roland R. Roth,
University of Delaware, “The effect of Wood
Thrush hosts on the survival of Brown-headed
Cowbird eggs and nestlings.”

Susan B. Smith, Villanova University, “An
assessment of refueling rates and diet of songbirds
during migratory stopover at the Braddock Bay
Bird Observatory.”

Kevin Sobol, Nathan Olszewski, and Lindsey
Walters, Canisius College, “Egg color and female
quality in the House Wren.”

Brynne A. Stumpe and Sara R. Morris, Canisius
College, “The effects of a residential wind
turbine on bird behavior.”

Jason M. Townsend, SUNY College of Envi-
ronmental Science and Eorestry, and Charles T.
Driscoll, Syracuse University, “Mercury bioac-
cumulation in Catharus thrushes along an eleva-
tional gradient.”

Mitch Walters, Cornell University, Marcela
Liljesthrom, Centro Austral de Investigaciones
Cientificas, CADICONICET, Argentina, and Ca-
ren Cooper, Cornell University, “Comparing
female characteristics to egg and yolk size in the
Chilean Swallow (Tachycineta meyeni)."

ATTENDANCE

Alabama: Tuscaloosa, Marie E. Wilson.

Arizona: Tucson, Clait E. Braun, Nancy J. K.
Braun.

California: Concord, Naho Kermeen.

District of Columbia: Carla Dove.

Delaware: Newark, Whitney Wiest.

Elorida: Gainesville, Gustavo Londono, Judith
Ungvari-Martin; Jacksonville, Bynne Stumpe;
Naples, Bette Jackson, Jerome Jackson.

Indiana: Bristol, Doris Watt.

Kansas: Hays, Andree Brisson, Greg Earley,
Cate Healy.

Maine: Orono, Monika Parsons.

Maryland: Bethesda, Ellen Paul; Brandywine,
Rachael Kushner; Columbia, James Smith; Frost-
burg, James Garabedian.

Massachusetts: Chilmark, Allan Keith; East
Falmouth, Ted Davis; Haverhill, Benjamin Pa-
dilla; North Adams, Daniel Shustack; Pocasset,
John C. Kricher.

Michigan: Albion, Dale Kennedy, Doug
White; Ann Arbor, Bob Payne, Laura Payne;
Chelsea, Harold Eyster; East Lansing, Jen Owen.

Minnesota: Minneapolis, Katie Glower.

Missouri: Columbia, Andrew Cox; St. Louis,
William Condit.

New Hampshire: Bow, Christie Eldredge;
Manchester, Jay Pitocchelli.

New Jersey: Lincoln Park, Erica Mueller;
Lyndhurst, Michael Newhouse; Pennington, Ste-
phanie Wright; Randolph, John Smallwood, Mary
Anne Smallwood.

New Mexico: Albuquerque, Erin Greenlee.

New York: Albion, Brad Mudrzynski; Armonk,

J. Alan Clark; Brockport, Nathan Grosse, Chris-
topher Norment; Bron.x, Oriana Chan, Kyle
Wright; Brooklyn, Jesse Ross; Buffalo, Catherine
Alsford, Lindsey Walters; Geneseo, Kristina
Hannam; Geneva, Bethany Bashaw, Mark
Deutschlander, Casey Franklin, Samuel Georgian,
Becky Jones, Mianna Molinari, Brian Monaco,
Paul Overdorf, Bob Taylor, Jordan Youngmann;
Grand Island, Sara Morris, Kevin Sobol; Ham-
burg, Kelly Fox; Hurley, Kailey Chidester;

832

THE WILSON JOURNAL OL ORNITHOLOGY • VoL 122. No. 4. December 2010

Ithcica, Caren Cooper, Elise Ferree, Jane Graves,
Heidi Lovette, Christina Masco, Colleen McLinn,
Matthew Savoca, Charlie Smith, Scott Sutcliffe,
Jason Townsend, Mitch Walters; Menands, Joel
Ralston; New York, Chad Seewagen; Newburgh,
Gretchen Wagner; Niagara Falls, Dorothy Fa-
tunmbi; Poughkeepsie, Glenn Proudfoot; Roches-
ter, Andrew Alba; Rye, Christy Conley; Utica,
Judith McIntyre; West Seneca, Alyssa Caffarelli,
Nathan Olszewski.

Ohio: Cincinnati, Dottie Engle, George Farns-
worth; Columbus, Lawrence Long, Paul Rode-
wald; Delaware, Aubrey Alamshah, Edward Burtt
Jr., Jordan Kalish; Granville, Jessie Hogue-
Morgenstern; Medina, Amanda Fawcett; Oberlin,
Mary Garvin; Sandusky, Robert Season; Wester-
ville, James Marshall.

Oklahoma: Edmond, Dennis G. Siegfried;
Norman, Tamaki Yuri; Stillwater, Adrian Mon-
roe, Tim O’Connell, Marla Steele; Tahlecjuah,
Mia Revels.

Pennsylvania: Barto, George Gregory; Car-
isle, Kim Vanfleet; Dunmore, Margaret Hatch;
Erie, Margaret Voss; Harrisburg, Nels Johnson;

Indiana, Jeff Larkin; Kutztown, Bill Brown, Todd
Underwood; Levine, Scott Stoleson; Markleys-
biirg, Frank Ammer; Mertztown, Ann Lugg;
Orangeville, Douglas Gross; Philadelphia, Laura
Niczyporoxicz; Pittsburgh, Todd Katzner, Fred
McCullough, Michael Sobkowiak; Reading, Ash-
ley Rathman; Scranton, Rob Smith; Sugar Grove,
Pam Stoleson; University Park, Sarah Pabian;
Villanova, Bob Curry, Susan Smith; Warren,
Emily Thomas; Williamsport, Nathan Frank;

Rhode Island: Kingston, Ulf Bauchinger,
Scott McWilliams; Newport, Jameson Chace.

Texas: Nacogdoches, Sarah Saalfeld, Rick
Schaefer.

Virginia: Alexandria, Dick Banks; Shipman,
Allen Hale

Vermont: Northfield, William Barnard; Will-
iston, Sean Beckett;

Canada, Ontario: Ajax, Sonya Richmond;
London, Eddy Price; Mississauga, Silu Wang;
Newmarket, Mike Van den Tillaart; Simcoe, Erica
Dunn, David Hussell; Toronto, James Rising,
Trudy Rising.

Sweden: Lund, Rachel Muheim.

The Wilson Jounuil oj Oniilhology 1 22(4):833-834, 2010

REVIEWERS FOR VOLUME 122

Referees are the life blood ol a journal as editors depend on them to help identify manuscripts with merit and offer
suggestions to improve the data analysis, overall science, and writing. These individuals receive little recognition but are
extremely important in the process of improving the science and quality of what is published. We thank all of those listed
below who served as referees for manuscripts processed (accepted and published, withdrawn, rejected) after I July 2009
through completion ot the December 2010 issue of Volume 122. (Those shown in boldface reviewed more than one
manuscript m this period.) The Wilson Ornithological Society and the editorial staff are indebted to and thank each person
who served as a reviewer. — Clait E. Braun, Editor

K. P. Able, N. J. Adams, J. Aguilar-Amat, J. A.
Ahumada, A. D. Anders, R. G. Anthony, D. P.
Armstrong, E. Arriero, L. Aubry, S. K. Auer, J.
K. Augustine, D. Badzinski, V. Baglione, M. C.
Baker, G. Ballard, W. H. Baltosser, G. Barrantes,
J. Bart, J. Bates, R. C. Beason, G. Beauchamp,

M. Bekoff, J. R. Belthof, M. Benowitz-Freder-
icks, K. S. Berg, T. M. Bergin, M. Bidwell, R. O.
Bierregaard, K. Bildstein, T. R. Birkhead, C. A.
Bishop, J. G. Blake, J. A. Blakesley, L. A. Blanc,

C. R. Blem, P. H. Bloom, C. W. Boal, C. E. Bock,
E. K. Bollinger, J. S. Bolsinger, D. N. Bonter, T.

A. Bookhout, C. A. Botero, W. S. Boyd, J. E.
Bradley, L. A. Brennan, D. Brightsmith, D. W.
Buden, D. E. Burhans, D. B. Burt, E. H. Burtt Jr.,
P. H. Butchko, C. J. Butler, B. S. Cade, T. Cade,

C. Caffrey, J. W. Cain, S. Calme, B. R. Chapman,
J. Chaves-Campos, N. Chernetsov, R. T. Chesser,

R. Cintra, R. B. Clapp, W. Clark, C. E. Clarkson,

N. J. Collar, C. T. Collins, J. C. Connelly, L. M.
Conner, R. N. Conner, C. J. Conway, M. I. Cook,

D. W. Coulton, R. Covas, J. Cox, S. Craig, E. R.

A. Cramer, H. Q. P. Crick, A. Cmz, J. F. Cully, F.

J. Cuthbert, L. D’Alba, W. E. Davis, D. K.
Dawson, D. C. Dearborn, K. L. Decker, C. A.
Delgado, M. M. Delgado, R. DeLotelle, E. P.
Derryberry, A. Desrochers, T. L. DeVault, A. A.
Dhondt, J. Diamond, A. S. Dolby, L. Dos Angos,

C. J. Dove, W. Duckworth, K. M. Dugger, P. O.
Dunn, J. B. Dunning Jr., W. R. Eddleman, B. A.
Eichhorst, J. C. Eitniear, K. S. Elli.son, L.
Ellison, C. S. Elphick, A. Engilis Jr., R. M.
Erwin, R. H. M. Espie, D. L. Euler, J. Faaborg, A.
Farnsworth, T. W. Fawcett, E. Fernandez-Juricic,

K. Fierro-Calderon, J. Ejeldsa, C. R. Foster, M.

S. Foster, R. M. Fraga, C. Francis, M. R.
Francisco, P. C. Frederick, M. Frederiksen, R.
W. Furness, G. A. Gale, V. Garcia, S. A.
Gauthreaux, J. L. Gehring, D. D. Gibson, A.
Giese, J. D. Gilardi, F. B. Gill, M. D. Giovanni, C.

B. Goguen, H. B. Gomes, O. Gordo, J. B. Grand,

D. J. Green, M. C. Green, R. Greenberg, P. D.

Greger, M. Gregg, R. J. Gutierrez, M. J. Guzy, J.
H. Haffer, B. A. Hahn, P. B. Hamel, B. T. Hansen,

A. Harmata, R. E. Harness, J. B. C. Harris, B. J.
Hatchwell, W. M. Healy, S. G. Herford, S. K,
Herzog, E. J. Heske, B. Heinrich, K. Higgins, J.

E. Hines, K. A. Hobson, P. A. R. Hockey, R. W.
Hoffman, C. S. Holling, R. T. Holmes, S. B.
Holmes, G. Holroyd, J. P. Hoover, S. N. G.
Howell, R. L. Hutto, L. D. Igl, D. Ingold, N.
Itonaga, F. C. James, J. M. Jawor, J. R. Jehl, B.
Jobin, D. H. Johnson, L. S. Johnson, O. W.
Johnson, C. D. Jones, S. L. Jones, R. Kannan, P.
Kappes, A. C. Kasner, J. L. Kellermann, C.
Kellner, R. S. Kennedy, D. T. King, C. K.
Kirkpatrick, G. Kirwan, N. A. Klaus, D. Klem
Jr., S. T. Knick, G. Kobriger, W. D. Koenig, E. M.
Kofoed, R. R. Koford, N. K. Krabbe, D. E.
Kroodsma, J. A. Kushlan, H. Kylin, Q. Latif, S. C.
Latta, J. J. Lawler, A. T. Lee, J. D. Ligon, J.
Liguori, C. K. Lim, B. C. Livezey, P. E. Llambias,
H. Lloyd, J. P. Loegering, J. T. Lokemoen, G. A.
Londono, S. C. Lougheed, 1. J. Lovette, J. Lowen.
P. E. Lowther, J. E. Loye, R. S. Lutz, B. E. Lyon,

E. A. MacDougal-Shackleton, R. H. F. Macedo, J.
N. Mager III, D. L. Maney, M. A. Marini, W. R.
Marion, P. P. Marra, L. B. Martin, P. R. Martin, T.
E. Martin, T. L. Master, P. Matyjasiak, J. E.
McCormack, D. McCracken, S. McGehee. K. J.
McGowan, A. Mee, D. J. Mennill, J. M. Meyers.

B. Mila, K. E. Miller, B. A. Millsap, C. M.
Miskelly, S. Mitra, A. P. Mpller, C. E. Moorman,
M. L. Morrison, R. I. G. Morrison, D. H. Morse,
E. S. Morton. R. Mo.s.s, R. L. Mumme, T. G.
Murphy, L. D. Murray. L. N. Naka, Z. Nemeth. S.
A. Nesbitt, P. Newell, D. Newstead, N. Nielsen-
PincLis, G. Niemi, I. C. T. Nisbet, R. Noske, K. P.
Oh, G. H. Orians, L. W. Oring. S. J. Ormerod, J-
M. Paillisson, J. D. Paruk, P. W. C. Paton. R. B.
Payne, A. T. Pearse, N. G. Perlut, B. P. Peterjohn,
L. J. Petit, W. Post, R. Poulin, B. Pranty, H. D.
Pratt, T. Price, H. Prior, M. S. Pruett. B. Pugesek,

C. M. Raley, C. J. Ralph, J. H. Rappole, J. T.

833

834

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4, December 2010

Ratti, L. J. Redmond, K. P. Reese, H. M. Reeves,

J. M. Reid, J. L. Reidy, R. Renfrew, K. Renton,
M. Restani, R. E. Ricklefs, C. D. Rittenhouse, S.

K. Robinson, E. Roche, A. D. Rodewald, R. D.
Rodgers, J. J. Roper, R. N. Rosenfield, J.
Rotenberry, R. R. Roth, D. C. Rudolph, T. J.
Sachtleben, L. Sandoval, V. Santharam, J. C.
Scarl, G. D. Schnell, E. A. Schreiber, K-L.
Schuchmann, T. S. Schulenberg, P. Scott, S. G,
Sealy, T. W. Seamans, P. Seddon, J. Sedinger, J.
A. Shaffer, S. R. Sheffield, L. M. Siefferman, L.
E. Silveira, J. A. Smallwood, C. R. Smith, G. W.
Smith, T. Smulders, D. A. Spector, D. Sperry, M.
T. Stanback, D. F, Stauffer, D. W. Steadman, S.
J. Stedman, K. Steenhof, L. Stein, J. L. Stephens,

I. Stewart, R. Stiehl, D. F. Stotz, P. C. Stouffer, B.
M. Strausberger, B. N. Strobel, J. D. Styrsky, K.
S. Summerville, D. L. Swanson, D. A. Swanson,
P. W. Sykes, M. K. Tarburton, C. F. Thompson,
M. B. Tingley, J. A. Tobias, P. W. Trail, E. N.
Vanderhoff, D. E. Varland, C. S. Vaughan, C. M.
Vleck, S. von Oettingen, H. Wada, J. S. Wakeley,

J. W. Walk, B. Walker, H. S. Walter, J. R.
Walters, D. Watt, P. J. Weatherhead, W. W.
Weathers, H. P. Weeks, W. Wehtje, J. T. Weir, R.
Weiss, C. White, L. A. Whittingham, J. W. Wiley,
S. K. Willson, M. D. Wilson, R. E. Wilson, J. C.
Wingfield, J. E. Woodford, K. Yasukawa, P. D.
Yaukey, S. Yezerinac, R. Yosef, K. J. Zimmer, R.
M. Zink, K. Zyskowski

Editor’s Comments

The rate of submissions has increased in 2010
and it appears we will again receive >200 new
manuscripts in the calendar year. This increases our
need for reviewers. All manuscripts (very few are
initially discouraged as inappropriate prior to the
review process) are sent to a minimum of two
subject area referees. We select potential reviewers
from perusal of the Literature Cited, our file of
subject area reviewers, and those offered by the
authors of specific manuscripts. In a perfect world,
we would expect to receive positive responses from
about two of every three referees contacted to
review a specific manuscript. This only rarely
happens and we may have to contact up to 12 (or
more) potential referees before we find someone to
review a particular manuscript. This problem is not
unique to our Journal as other editors have similar
problems. The most recent Issue of Science (2010,
329: 1466) has a commentary (Bcittling the Paper
Glut) on publishing and the responsibility of authors
of published papers to be willing to review
manuscripts. Their proposed solution would be to
have authors of submitted manuscripts provide
three reviews of other manu.scripts.

Our goal is to provide reviews and editorial
recommendations on suitability of manuscripts in
a timely manner. Searching for someone to agree
to review manuscripts slows the process. We ask
those contacted about reviewing a manuscript to
consider the need for competent and timely
reviews. At the least, reviewers contacted are
asked to respond either yes or no instead of not
responding at all. The areas of delay in the process
of publishing for The Wilson Journal of Ornithol-
ogy are: reviewer availability, timeliness of
reviews, and timeliness in author revision. Some
manuscripts that are encouraged fail to be
returned in a timely manner. Some never are
returned and become de facto rejections or
withdrawals.

We presently do not have a backlog of accepted
manuscripts waiting for publication. Those being
accepted in September and October into early
November this year will be published in the
March Issue of the next year with few exceptions.

Clait E. Braun
Editor, 2010

The ITilson Journal of Ornilholoi'y 1 22(4):835-858, 2010

Index to Volume 122, 2010

Compiled by Leslie A. Robb

This index includes leterences to genera, species, authors, and key words or terms. References are made to scientific
names ot all vertebrates mentioned within the volume and other taxa mentioned prominently in the text, in addition to avian
species. Nomenclature follows the AOU Check-list of North American Birds (Seventh Edition) and F. Gill and M. Wright
and supplements thiough 2010 (Birds of the World, Recommended English Names, Princeton University Press, Princeton,
New Jersey, USA and Oxford, United Kingdom). Reference is made to books reviewed and announcements as they appear
in the volume.

A

Abundance

abundance and distribution of waterbirds in the Llanos of
Venezuela, (577), 102-115
of Griis canadensis nesiotes, 556—562
Accipiter cooperi, 168-173, 290, 470, 475
striatus, 290, 518-531, 768, 771
Actitis macidarius [macidaria, I 15], 768, 771
Aegithalos caudatus, 2 1 2
Aerodramiis brevirostris, 259

fuciphagus inexpectatus, 259-272
hirundinaceus, 270
maximus, 259
salanganiis, 259
spodiopydius assimilis, 158
spp., 153
unicolor, 259
vanikorensis, 270

Agelaioides badius, 417^31 (Frontispiece), 795, 797
Agelaius phoeniceus, 187, 291, 307-317, 402^05, 435,
532-544, 671, 768, 771, 807

Aguilera, Xiomara Galvez and Felipe Chavez-Ramirez,
Distribution, abundance, and status of Cuban
Sandhill Cranes (Grus canadensis nesiotes), 556-
562

Aimophila aestivalis. 518
Aix spon.ta. 1 87
AJaia ajaja, 1 14
Alces alces, 699, 701-703

Aldred, James, .see Seymour, Adrian S., Graham Hatherley,

Francisco Javier Contreras, , and Fergus

Beeley

Alectoris chukar, 762-766

Alessi, Mark G., Thomas J. Benson, and Michael P. Ward,
Nocturnal social cues attract migrating Yellow-
breasted Chats, 780-783
Alopex lagopus, 2
Alophoxiiis pallidiis, 1 76
Alouatta palliata. 794
seniculus, 794

Alvey, Erin C., see Confer, John L., Kevin W. Barnes, and

Amaitrornis phoenicurus. 784-788
phoenicunis insidaris, 787
phoenicurus leiicomelanus, 787
phoenicurus midnicobaricus, 787
Amazilia Candida, 592
cyanocephala, 592-597

cyanocephala chlorostephcma, 592
cyanocephala cyanocephala. 592-597
cyanocephala guatemalensis, 592
rutila, 496
violiceps, 592

Amazon, Lilac-crowned, see Amazona finschi
Orange-winged, see Amazona amazonica
Puerto Rican, see Amazona vittata
Amazona amazonica, 88-94
finschi, 516
vittata, 515

Amazonetta brasiliensis, 108
Ammodramus bairdii, 144, 346-353, 455^64
caudacutus, 340-345
henslowii, 290, 635-655, (Frontispiece)
lecontii, 139-145
maritimus, 343, 532-544
nelsoni, 340-345

savannarum, 144, 290, 455^64, 635-654

Amos, Anthony F., see Foster, Charles R., , and Lee

A. Fuiman
Anas acuta, 768, 771

americana, 376, 613, 768, 770
clypeata, 569, 612-614, 768, 771
crecca, 376, 768, 770
cyanoptera. 613, 768, 770
discors, 108, 111,612-614, 768, 770
laysanensis, 9

platyrhynchos, 19, 302, 565, 569, 613, 759, 762, 768
771

strepera. 768, 770
Anhima cornuta, 105, 114
Anhinga, see Anhinga anhinga
Anhinga anhinga, 105, 108, 111, 114
spp., 380

Ani, Smooth-billed, see Crotophaga ani
Anich, Nicholas M. and Bryan M. Reiley, Effects of a flood
on foraging ecology and population dynamics of
Swain.son’s Warblers, 165-168
Anser albifrons, 49 1

ant, army, see Eciton biirchellii. Labidus praedator.
Labidus spinninodis
fire Red lire], see Solenopsis invicta
Antbird, Immaculate, see Myrmeciza immaculata
Anthus rubescens. 663

spragueii, 346-353, 455-464
Antpitta, Chestnut-crowned, see Grallaria ruficapilla
Chestnut-naped, see Grallaria nuchalis

835

836

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4, December 2010

Great, see Grallaria excelsa
Jocotoco. see Grallaria ridgehi
Moustached, see Grallaria alleni
Pale-billed, see Grallaria carrikeri
Plain-backed, see Grallaria haplonota
Rufous, see Grallaria riifula
Scaled, see Grallaria giiatinialensis
Variegated, see Grallaria varia
Watkins’s, see Grallaria watkinsi
White-bellied, see Grallaria hypoleiica
Yellow-breasted, see Grallaria jlavotincta
Anumhius anniimbi, 42 1 , 422
Aphelocoma coerulescens, 808
Apiis apus, 436
spp., 153

Acjuila adalberti, 364
chrysaetos, 354
Aranudes cajanea. 105, 115
Aramiis giiarauna. 105, 1 15
A rat Inga leucophthahna, 803-806
Archilochus alexandri, 494-502
colubris, 302, 496, 5 1 8-544
Ardea alba. 96, 102, 106, 1 14, 532-544
cocoi, 108, 114

herodias, 96, 232, 532-544, 768, 770
Arenaria interpres, 532-544
Arremon brunneinucha. 507
Arundinicola leucocephala, 105, 115
Asia jlammeiis, 457, 768, 77 1
otus, 768, 771
Asthenes baeri, 602
berlepschi. 602
cactoriim, 602
dorbignyi, 602
humicola, 602
liiizae, 600-603
patagonica, 602
pudibitnda. 602
steinbachi, 602
Athene ciinicuaria, 51-59
cimicuaria hypugaea, 51-59

Atterbcrry-Jones, Megan R. and Brian D. Peer, Cooperative
breeding by Red-headed Woodpeckers, 160-162
Aidacorhynchus prasinits. 507, 509

Ayers, Andrea J. and James W. Armacost Jr., Marsh Wren
eats small fish, 623-624
Aythya affinis. 569, 768, 771
ainericana, 613, 768, 771
inarila. 376
valisineria. 768, 770

B

Babhitt, Kimberly J., .see Suomala. Rebecca W., Sara R.

Morris. . and Thomas D. Lee

Babbler. Rusty-faced, see Robsoniiis rahori
Bader, Troy J. and James C. Bednarz. Home range, habitat
use, and nest site characteristics of Mississippi
Kites in the White River National Wildlife Refuge.
Arkansas. 706-715
badger, American, sec Taxidea taxus
liaeolophus bicolor. 1 26- 134, 171.51 8-544

Baldassarre. Guy A., see Vilella, Francisco, J., Mark S.
Gregory, and

Baldwin. Heather Q., Clinton W. Jeske, Melissa A. Powell,
Paul C. Chadwick, and Wylie C. Barrow Jr., Home-
range size and site tenacity of overwintering Le
Conte’s Sparrow in a fire managed prairie, 139-145

Bao-Zhong, Lu, see Xiao-Ping, Yu, Xi Yong-Mei, ,

Li Xia. Gong Ming-Hao, Shi Liang, and Dong
Rong

Barbet, Black-browed, see Megalaima oorti nuchalis
Red-crowned, see Megalaima rafflesii
Taiwan, see Megalaima nuchalis
White-cheeked, see Megalaima viridis

Barclay, M., see Quinn, J. S., A. Samuelsen, , G.

Schmaltz, and H. Kahn

Barnes. Kevin W., see Confer, John L., , and Erin C.

Alvey

Barrantes, Gilbert, see Chavarn'a-Pizarro, Tania, Gustavo

Gutierrez-Espeleta, Eric J. Euchs, and

Barrow Jr., Wylie C., see Baldwin, Heather Q., Clinton W.
Jeske, Melissa A. Powell. Paul C. Chadwick, and

Bartramia longicauda, 1 1 5
Basileuterus culicivorus, 507
rufifrons. 507, 509
tristriatus, 45 1 , 507
Baywing, see Agelaioides hadiiis
beaver, American, see Castor canadensis
Beckett, Matthew D. and Gary Ritchison. Effects of
breeding stage and behavioral context on singing
behavior of male Indigo Buntings, 655-665

Bednarz. James C., see Bader, Troy J. and

Bednarz. James C.. see Pappas, Sara, Thomas J. Benson,
and

Bee-eater. European, see Merops apiaster
White-fronted, see Merops bullockoides
Beeley, Fergus, see Seymour. Adrian S., Graham Hatherley,
Francisco Javier Contreras, James Aldred. and

behavior

army ant raid attendance and bivouac-checking behavior
by neotropical montane forest birds, 503-512
arthropod foraging by a southeastern Arizona humming-
bird guild. 494-502

cavity-sharing between a Dryocopus galeaius and two
Aratinga leucophthalma, 803-806
Coccyzus americanus hatched and fed by an Agelaius
phoeniceus, 402-405

cooperative breeding of Todiramphiis veneratus, 46-50
cooperative polyandry in Sula siila, 361-365
costs and benefits of foraging alone or in mixed-species
aggregations for Sterna forsteri. 95-101
foraging habits by Aerodramus fuciphagus inexpectatus
and Collocalia esculenta affinis in the Andaman
Islands, India, 259-272
hanging behavior of Corvus cornix. 183-185
hatching synchrony, green branch collecting, and prey
use by nesting Harpia harpyja. 792-795
incubation temperature and behavior in thrushes nesting
at high altitude, 666-673
infanticide by Crotophaga ani. 369-374

INDEX TO VOLUME 122

837

infanticide by Sayoniis phoehe, 620-622
nociin nal social cues attract niigratin^ icti^rici viri^ns^
780-783

ohseivations of a possible foraging tool by Corviis corciw
181-182

Picoides borealis inale/temale foraging differences in
young forest stands, 244-258
provisioning rates of Myiohorus miniutus in Monteverde,
Costa Rica, 29-38

response to nestling throat ligatures by three songbirds,
806-809

site fidelity of Fatco span’eriits wintering in southwest-
ern Florida, 475-483

thermoregulatory behavior in migratory Merops tipi-
aster, 378-380

Benson, Thomas J., see Alessi, Mark G., , and

Michael P. Ward

Benson, Thomas J., see Pappas, Sara, , and James C,

Bednarz

Berardelli, Daniele, Martha J. Desmond, and Leigh Murray,
Reproductive success of Bunowing Owls in urban
and grassland habitats in southern New Mexico,
51-59

Biamonte, Esteban, A new bird species for Costa Rica:
Sapphire-throated Hummingbird (Lepidopyga coer-
ideogiilaris), 194-195

Bildstein, Keith L., see Hinnebusch, Daniel M., Jean-
Fran9ois Therrien, Marc-Andre Valiquette, Bob

Robertson, Sue Robertson, and

Binford, Laurence C. and Joseph A. Youngman, Flight
speeds of migrating Red-necked and Horned
grebes, 374-378

Bird, David M., see Frei, Barbara, , and Rodger D.

Titman

Bittern, American, see Botaurus lentigiiiosiis
Least, see Ixobiychiis exilis
Pinnated, see Botaurus pinnatus
Blackbird, Brewer's, see Euphagus cyanocephalus
Chopi, see Gnorimopsar chopi
Common, see Turdus merula
Red- winged, see Agelaius pboeniceus
Rusty, see Euphagus carolinus
Yellow-headed, see Xanocephalus xanocephalus
Blackcap, European, see Sylvia atricapilla
blowfly, see Protocalliphora tundrae
Bluebird, Eastern, see Sialia sialis
Western, see Sialia mexicana
bluefish, see Pomatomus saltatrix

Boal, Clint W, see Strobe], Brad N. and

bobcat, see Lynx rufus
Bobolink, see Dolichonyx orxzivorus
Bobwhite, Northern, see Colinus virginianus
Boiga spp., 175

Bombycilla cedrorum. 290, 326, 329, 518-531, 725-743
Bonasa umbelliis. 2, 9, 290, 518-531

Bonter. David N., see Brooks, Elizabeth W. and

Booby, Nazea, see Sula granti
Red-footed, see Sula sula
Botaurus lentiginosus, 768, 770
pinnatus. 108, I 14
botfly, see Philornis spp.

Boves, Than J., David A. Buehicr, and Phillip C. Massey,
Interspecific song imitation by a Cerulean Warbler,
583-587

Bracey, Elwood D., see Hayes, William K., . Melissa

R. Price, Valerie Robinette, Eric Gren, and
Caroline Stahala
Bradypus tridactylus, 792, 793
Brant, Pacific Black, see Brania bernicia nigricans
Branta bernicia nigricans, 484^93
canadensis. 491, 768, 770
hutcliinsii, 491

Initchinsii leucopareia, 2, 10

Braun, Clait E., review. Grouse of plains and mountains:
the South Dakota story, by Lester D. Flake, John
W. Connelly, Thomas R. Kirschenmann, and
Andrew J. Lindbloom, 628-629
Braun, Clait E., see Kaler, Robb S. A., Steve E. Ebberl,

, and Brett K. Sandercock

breeding biology

age of first breeding of Nipponia nippon in the Qinling
Mountains, China, 228-235

annual precipitation affects reproduction of Lanins
meridionalis, 334—339

breeding phenology and nesting success of Campvlo-
rhynebus yucataniciis in the Yucatan Peninsula,
Mexico, 439-446

cooperative breeding by Melanerpes erthrocephalus.
160-162

cooperative breeding by Todiramphus veneratus. 46-50
first reported case of cooperative polyandry in Sula sula.
361-365

first reproductive record of Charadrius wilsonia in Bata
de Todos os Santos, northeastern Brazil, 788-
791

nest box use by Pants major in semi-arid rural
residential gardens, 604-608
of Athene cunicuaria in urban and grassland habitats in
southern New Mexico, 51-59
of Garrulax maximus at Lianhuashan, southern Gansu.
China, 388-391

of Larus dominicanus on the Brazilian coast, 39—45
of Megalaiina nuchalis in Taipei Botanical Garden, 681-
688

of Melanopareia torquata. 1 62- 1 65
of Myiohorus miniatus in Monteverde, Costa Rica, 29-
38

of Myiohorus miniatus in Venezuela, 447-454
o{ Zimmerius cinysops in Venezuela, 689-698
parental care, food provisioning, and nest defense by
Carpococcyx renauldi in northeastern Thailand
17.3-177

Passer domesticus associated with reduced Petrocheli-
don pyrrhonota nesting success. 135-138
Brittain. Ross A., Vicky J. Merclsky, and Chris B. Craft.
Avian communities of the Altamaha River estuary
in Georgia, USA. 532-544

Brooks. Elizabeth W. and David N. Bonter. Long-term
changes in avian community structure in a
successional. forested, and managed plot in a
reforesting landscape. 288-295
Bubalus huhalis. 559. 560

838

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4, December 2010

Bubo scandiacus [Nyctea scandiaca, 3, 10], 762
virginianus, 707, 768, 770
Bubulcus ibis, 102, 106, 108, 114, 212
Bucephala clangula, 377, 569
Buceros hydrocorax, 48

Buden, Donald W. and Stanley Retogral, Range expansion
of the White-breasted Waterhen {Amaurornis
phoenicurus) into Micronesia, 784-788

Buehler, David A., see Boves, Than J., , and Phillip

C. Massey

buffalo, water, see Bubalus bubalis

Bulbul, Puff-throated, see Alophoxius pallidus

bullsnake, see Pituophis catenifer [melanoleucus]

Bunting, Conrmon Reed, see Emberiza schoeniclus
Indigo, see Passerina cyanea
Lark, see Calamospiza melanocorys
Lazuli, see Passerina amoena
Painted, see Passerina ciris
Snow, see Plectrophenax nivalis
Burgess, Nathanael J., see Butler, Christopher J., Lisa H.
Pham, Jill N. Stinedurf, Christopher L. Roy, Eric L.

Judd, , and Gloria M. Caddell

Burhans, Dirk E., Brian G. Root, Terry L. Shaffer, and
Daniel C. Dey, Songbird nest survival is invariant
to early-successional restoration treatments in a
large river floodplain, 307-317
Burhinus bistriatus, 105, 106, 108, 115
Bushtit, Long-tailed, see Aegithalos caudatus
Buteo galapagoensis, 46, 364
jamaecensis, 68, 370, 518-544, 768, 771
lineatus, 68-74, 290, 532-544, 707, 713
lineatus alleni, 70
lineatus lineatus, 70
lineatus texanus, 70
platypterus, 171, 290, 518-531,
spp., 801

swainsoni, 768, 771

Butler, Christopher J., Lisa H. Pham, Jill N. Stinedurf,
Christopher L. Roy, Eric L. Judd, Nathanael J.
Burgess, and Gloria M. Caddell, Yellow Rails
wintering in Oklahoma, 385-387
Butorides striatus, 114

virescens, 108, 114, 768, 770

C

Cacicus solitarius, 795-799
Caciques, Solitary, see Cacicus solitarius
Caddell, Gloria M., see Butler, Christopher J., Lisa H.
Pham, Jill N. Stinedurf, Christopher L. Roy, Eric L.

Judd, Nathanael J. Burgess, and

Cairina moschata, 1 1 4
Calamospiza melanocorys, 457
Calcarius mccownii, 460
ornatus, 346-353. 455-464
Calidris himantopus, 569
mauri, 15-22
melanotos, 569

pusilla [semipalmatus, 19], 19, 569
spp., 109, 114, 115
Callosciurus finlaysonii, 175
Calypte anna, 496

Campylorhynchus brunneicapillus, 443, 444, 624
nuchalis, 443
rufinucha, 443
yucatanicus, 439-446
Canary, Atlantic, see Serinus canaria
Canastero, Berlepsch’s, see Asthenes berlepschi
Cactus, see Asthenes cactorum
Canyon, see Asthenes pudibunda
Cip6, see Asthenes luizae
Dusky-tailed, see Asthenes humicola
Patagonian, see Asthenes patagonica
Rusty-vented, see Asthenes dorbignyi
Short-billed, see Asthenes baeri
Steinbach’s, see Asthenes steinbachi
Canis latrans, 457, 801, 802
lupus, 181, 182

lupus familiaris, 556, 559-561, 604-608, 615, 616, 801
Canvasback, see Aythya valisineria
Cao, Lei, Guixia Zhao, Shan Tang, and Hanzhao Guo, The
first reported case of cooperative polyandry in the
Red-footed Booby: trio relationships and benefits,
361-365

Caprimulgus carolinensis, 381-384
capuchin, wedge-capped, see Cebus olivaceus
Caracara plancus, 44

Caracara, Southern Crested, see Caracara plancus
Cardellina rubrifrons, 35, 614-617
spp., 29, 36

Cardinal, Northern, see Cardinalis cardinalis
Red-capped, see Paroaria gularis
Cardinalis cardinalis, 168-173, 291, 326-333, 518-544
Carduelis citrinella, 490
Carpococcyx radiceus, 173, 175, 176
renauldi, 173-117
viridis, 173, 175
Carpodacus mexicanus, 291
purpureus, 291, 725-737
Castor canadensis, 274, 276
cat, domestic, see Felis catus
Catbird, Gray, see Dumetella carolinensis
caterpillar, eastern tent, see Malacosoma americana
Cathartes aura, 354-360, 518-531
Catharus aurantiirostris, 503-512
frantzii, 507
fuscater, 507, 509

fuscescens, 290, 549-552, 725-737, 768, 771
guttatus, 82-87, 290, 302, 666-673
spp., 743

ustulatus, 82-87, 302, 507, 510, 545, 725-737, 742, 768,
771

Catops alpinus, 703, 704
Cebus apella, 679

olivaceus, 693, 792, 793
Centrocercus urophasianus, 2, 768, 769, 771
Centropus grillii, 755
spp., 173

Cercibis oxycerca, 108, 114
Certhia americana, 290, 725-737
Ceryle rudis, 48

Chadwick, Paul C., see Baldwin, Heather Q., Clinton W.

INDEX TO VOLUME 122

839

Jeske, Melissa A. Powell, , and Wylie C.

BaiTow Jr.

Chaetura chapmani, 157
pelagica, 157, 518-531
spp., 153

Chandler, Robert M., Peter Pyle, Maureen E. Flannery,
Douglas J. Long, and Steve N. G. Howell, Flight
feather molt of Turkey Vultures, 354-360

Charadrius alexandrinus, 23, 25
alexandrinus alexandrinus, 25
alexandrinus nivosus, 25
collaris, 105, 115
melodus, 23, 25, 578-582
morinellus, 578
semipalmatus, 23-28
vociferus, 62, 768, 771
wilsonia, 532-544, 788-791
wilsonia brasiliensis, 790

Charter, Motti, Yossi Leshem, Shay Halevi, and Ido Izhaki,
Nest box use by Great Tits in semi-arid rural
residential gardens, 604-608

Chartier, Allen T., review. Do hummingbirds hum?
Fascinating answers to questions about humming-
birds, by George C. West and Carol A. Butler, 818-
819

Chat, Yellow-breasted, see Icteria virens

Chavama-Pizarro, Tania, Gustavo Gutierrez-Espeleta, Eric
J. Fuchs, and Gilbert Barrantes, Genetic and
morphological variation of the Sooty-capped Bush
Tanager (Chlorospingus pileatus), a highland
endemic species from Costa Rica and western
Panama, 279-287

Chavez-Ramirez, Felipe, see Aguilera, Xiomara Galvez and

Chen [Anser, 19] caerulescens caerulescens, 491

Chen, Yang, see lie, Wang, , Lu Nan, Fang Yun, and

Sun Yue-Hua

Cheng, Wei-Jen, see Lin, Shang-Yao, Fang-Cian Lu, Fu-
Hsien Shan, Shu-Ping Liao, Jing-Ling Weng,

, and Chao-Nien Koh

Chickadee. Black-capped, see Poecile atricapillus
Boreal, see Poecile hudsonicus
Carolina, see Poecile carolinensis
Mountain, see Poecile gambeli
Chicken, Domestic, see Callus gallus domesticus
chimpanzee, see Pan spp.
chipmunk, cliff, see Tamias dorsalis
eastern, see Tamias striatus
Chloroceryle aenea, 106, 109, 115
amazona, 115
americana, 109, 115
inda, 106, 109, 115
Chlorospingus pileatus, 279-287
zeldoni, 279

Chlorospingus, Zeledon’s, see Chlorospingus zeldoni
Chlorostilbon assimilis, 597-599
canivetti, 496
Chordeiles gundlachii, 381
minor, 768, 770

Chroicocephalus ridibundus, 755

Chuck- will’s- widow, see Caprimulgus carolinensis

Chukar, see Alectoris chukar
cicada, annual, see Tibicen spp.

periodic, see Magicicada spp.

Ciconia ciconia, 233, 755, 756
Cinclus cinclus, 563-568
leucocephalus, 563-568
mexicanus, 563-568
pallassi, 563-568
schulzi, 563-568

Circus cyaneus, 456, 768, 771, 801
Cistothorus palustris, 62, 532-544, 623-624
Clangula hyemalis, 377

Clarke, Julia A., review, Paleogene birds, by Gerald Mayr,
409^10
climate change

phenology of migration in relation to climate change,
116-125

Coccyzus americanus, 402M05, 518-544, 768, 771
erythropthalmus, 290, 402

Cockle, Knstina L., Interspecific cavity-sharing between a
Helmeted Woodpecker {Dryocopus galeatus) and
two White-eyed Parakeets {Aratinga leucophthalma),
803-806

Cohen, Jonathan B. and James D. Fraser, Piping Plover
foraging distribution and prey abundance in the pre-
laying period, 578-582

Colaptes auratus, 128, 131, 475, 518-531, 608, 725-737,
768, 771

Colibri thalassinus, 496
Colinus virginianus, 518
Collocalia esculenta qffinis, 259-272
Coluber constrictor, 801
constrictor flaviventris, 800
Columba livia, 62, 741
Columbina passerina, 532-544
Condor, California, see Gymnogyps califomianus
Confer, John L., Kevin W. Barnes, and Erin C. Alvey,
Golden- and Blue-winged warblers: distiibution,
nesting success, and genetic differences in two
habitats, 273-278
conservation status

of Crus canadensis nesiotes, 556-562
Contopus cooperi, 773
virens, 518-544, 725-737

Contreras, Francisco Javier, see Seymour, Adrian S.,

Graham Hatherley, , James Aldred, and

Fergus Beeley

Conway, Courtney J., see Kirkpatrick, Chris and

Conway, Courtney J., see Robinson, Gabrielle L., ,

Chris Kirkpatrick, and Dominic D. LaRoche
Cooper, Robert J., see Klaus, Nathan A., Scott A. Rush,

Tim S. Keyes, John Petrick, and

Coot, American, see Fulica americana
Coragyps atratus, 41, 44, 518-531
Cormorant, Double-crested, see Phalacrocorax auritus
Neotropic, see Phalacrocorax brasilianus
Reed, see Phalacrocorax africanus
Corvus brachyrhynchos, 290, 296, 301, 302, 470, 518-544
768, 770, 802

corax, 3, 181-185, 302, 616, 768, 770

840

THE WILSON JOURNAL OL ORNITHOLOGY • IW. 122. No. 4. December 2010

coroiie. 184
o.^sifragiis. 532-544
Coiurnicops noveboracensis, 385-387
Cotiirni.x cotiirnix, 790
japonica. 615, 755, 756, 759
Coucal, Black, see Centropus grillii
Cowbird, Brown-headed, see Mololhni.<i oter
Screaming, see Molothni.s rufoa.xillaris
Shiny, see Malothrus bomiriensis
coyote, see Cants latrans

Craft, Chris B., see Brittain, Ross A., Vicky J. Meretsky,
and

Crake, Slaty-legged, see Raltina eitrizonoicles
Crane, Cuban Sandhill, see Grits canailensis nesiotes
Florida Sandhill, see Grits canadensis pratensis
Mississippi Sandhill, see Grits canadensis piilla
Sandhill, see Grits canadensis
Craves. Julie A., review. Bird banding in North America:
the first hundred years, edited by Jerome A.
Jackson. William E. Davis Jr., and John Tautin,
197-198

Creeper. Brown, see Certhia ainericana
Cre.scentchest, Collared, see Melanopareia torqiiata
Olive-crowned, see Melanopareia ma.xintiliani
Crossarchiis spp.. 560, 561
C ratal us cerberiis. 616
price i. 616
viridis. 457

Crotophaga ani, 369-374
siilciro.stris. 46

Crow, American, see Conus hrachyrhynchos
Carrion, see Con'us corone
Fish, .see Conuis ossifragus
Cuckoo, Black-billed, .see Coccyztis erythropthalnius
Bornean Ground, see Carpococcy.x radiceus
Common, see Ciiculus canorus
Coral-billed Ground, see Carpococcy.x renauldi
Sumatran Ground, see Carpococcy.x viridis
Yellow-billed, see Coccyztis americaniis
Ciiculus canorus. 346, 351, 352, 796
Curnutt, John, review. Ingestion of lead from spent
ammunition: implications for wildlife and humans,
by Richard T. Watson. Mark Fuller. Mark Pokras,
and W. Grainger Hunt. 198-199
Curry. Richard, see Kerlinger. Paul, Joelle L. Gehring.

Wallace P. Erick.son. , Aaftab Jain, and John

Guarnaccia

Cvanocilta cristata. 290. 302, 5 1 8-544,
slelleri. 615, 616
Cyanocora.x chiysops. 679
morio. 507. 509
Cyanolyca cucullata. 507
Cxanoplila spp.. 590
Cynantiuis lalirostris, 496
Cxnoinvs liidovicianus, 51-59
Cvornis herioti. 587-591
Cxphorhinus phaeocephalus. 444
Cyprinodon variegaliis. 96
Cypseloides niger. 158
spp.. 153

D

da Silva. Maria Luisa, see de Moura, l.eiliany Negrao,

Jacques M. E. Vielliard, and

Dacelo novaeguineae, 48

Davis, Craig A., David M. Leslie Jr., W. David Walter, and
Allen E. Graber, Mountain biking trail use affects
reproductive success of nesting Golden-cheeked
Warblers. 465^74

De Marsico, Man'a C.. Bettina Mahler, and Juan C.
Reboreda. Reproductive success and nestling
growth of the Baywing parasitized by Screaming
and Shiny cowbirds. 417^31
de Mendon9a Dantes, Gisele Pires and Joao Stenghel
Morgante, Breeding biology of Kelp Gulls on the
Brazilian coast, 39-45

de Moura, Leiliany Negrao, Jacques M. E. Vielliard, and
Maria Luisa da Silva, Seasonal fluctuation of the
Orange-winged Amazon at a roosting site in
Amazonia, 88-94

deer, white-tailed, see Odocoileiis virginianus
Demongin, Laurent, Maud Poisbleau, Georgina Strange,
and Ian J. Strange. Second and third records of
Snares Penguins (Eiidyptes robustu.x) in the Falk-
land Islands, 190-193
Dendrocincia anahatina. 805
homochroa, 507-509
Dendrocopos .syriacus. 604-608
Dendrocygna autumnalis. 102, 106, 114
hicolor. 1 14

viduala, 102. 106, 108, 1 14
Dendroica adelaidae, 663

caerule.xcens. 26. 518-531, 725-737
castanea. 773
cerulea. 518-531. 583-587
chrysoparia, 465^74, 777-780
coronata. 290. 302, 725-737
discolor, 290, 518-531
doininica, 5 1 8-544

fu.xca. 290, 301, 302, 518-531, 585, 725-737
magnolia. 290, 302. 725-737
palmarum. 773. 774

pensylvanica. 290, 302, 507. 518-531. 549. 550, 552,
725-737

petechia. 738-743, 767-771
petechia hryanti. 443
pinus. 302, 518-544,
striata. 725-737
tigrina. 302, 773

Virens. 290, 296, 301. 302, 507, 518-531, 585, 725-737
Derksen, Dirk V., see Lewis. Tyler L., Paul L. Flint, Joel A.
Schmutz. and

Desmond, Martha .1., see Berardelli, Daniele, . -and

Leigh Murray

Dcy, Daniel C.. see Burhans, Dirk E., Brian G. Root, Terry
L. Shaffer, and

Di Giacomo. Alejandro G.. Bettina Mahler, and Juan C.
Reboreda, Screaming Cowbird parasitism of nests
of Solitary Caciques and Cattle Tyrants, 795-799
Dickcissel, see Spiza ainericana

Dickens, Molly .1. and Michael Romero. Stress responsive-

INDEX TO VOLUME 122

841

ness decreases with age in precocial juvenile
Chukar, 762-766
Diclelphis iilhiventer. 679
vir^iniana. 646, 801, 802

Dieni, J, Scott, see Jones, Stephanie L., , and Paula J.

Gouse

diet

arthropod foraging by a southeastern Arizona humming-
bird guild, 494-502

Cistotlioriis palustris eats small fish, 623-624
hatching synchrony, green branch collecting, and prey
use by nesting Harpia harpyja, 792-795
nestling diet of Carpococcy.x renaiildi in northeastern
Thailand, 173-177

of Tetrastes sewerzowi during preincubation, 177-180
prey items of Tyto alba in an agricultural ecosystem in
northern Utah, 60-67

regional variation in diets of breeding Buteo lineatus,
68-74

Dipper, American, see Cinclus mexicaniis
Brown, see Cinclits pallassi
Rufous-throated, see Cinclus sclmlzi
White-capped, see Cinclus leucocephalus
White-throated, see Cinclus cinclus
dispersal

postfledging and natal dispersal of Nipponia nippon in
the Qinling Mountains, China, 228-235
distribution

abundance and distribution of waterbirds in the Llanos of
Venezuela, (577), 102-1 15

evidence of breeding by Molothrus bonariensis in North
America, 365-369

first reproductive record of Charadrius wilsonia in Bai'a
de Todos os Santos, northeastern Brazil, 788-
791

new bird species tor Costa Rica: Lepidopyga coeruleo-
gularis. 194-195

of Grus canadensis nesiotes, 556-562

phenotypic distribution of Vennivora chrysoplera and V.

pinus in two habitats, 273-278
range expansion of Amaurornis phoenicunis into Micro-
nesia, 784-788

records of Eudyptes robustus in the Falkland Islands,
190-193

dog, domestic, see Canis lupus faniiliaris
Dolichony.x oryzivorus, 291, 432^38, 643, 646-654, 807
dolphin, bottle-nosed, see Tursiops Iruncalus
Donacobius atricapillus\ 105, 115
Donacobius, Black-capped, see Donacobius atricapillus
Dornak, L. Lynnette, Breeding patterns of llenslow's
Sparrow and sympatric grassland sparrow species,
635-645

Dotterel, Eurasian, see Charadrius morinellus
Dove, Mourning, see Zenaida niacroura
Dryocopus galeatus. 803-806
lineatus. 804

pileatus. 128, 131, 290, 518-544
Duck, Black-bellied Whistling, see Dendrocygna autuinnalis
Fulvous Whistling, see Dendrocygna bicolor
Laysan, see Anas laysanensis
Long-tailed, see Clangula hyeinalis

Muscovy, see Cairina niosctiala

White-faced Whistling, see Dendrocygna viduala

Wood, see Aix sponsa

Dunietella carolinensis, 290, 351, 549, 550, 552, 768, 770

E

Eagle, Golden, see Aquila chrysaetos
Harpy, see Harpia harpyja
Spanish Imperial, see Aquila adalberti
Ebbert, Steve E., see Kaler, Robb S. A., , Clait E.

Braun, and Brett K. Sandercock
Eciton burchellii, 503, 504
ecology

annual precipitation affects reproduction of Lanius
ineridionalis, 334-339

avian communities of the Altamaha River estuary in
Georgia, USA, 532-544

breeding patterns of Animod ramus henslowii and sym-
patric grassland sparrow species, 635-645
carotenoid-based male plumage predicts parental invest-
ment in Setophaga ruticilla, 318-325
Charadrius melodus foraging distribution and prey
abundance in the pre-laying period, 578-582
composition dynamics of an avian assemblage, 767-771
costs and benefits of foraging alone or in mixed-species
aggregations for Sterna forsteri. 95-101
long-term changes in avian community structure in a
successional, forested, and managed plot in a
reforesting landscape, 288-295
migrant songbird species distribution and habitat use
during stopover on two islands in the Gulf of
Maine, 725-737

nesting ecology ot Lanius tephronotus in south Tibet,
395-398

nest-site limitation of secondary cavity-nesting birds in
even-age southern pine forests, 126-134
of Lagopus niuta evermanni in the Aleutian Islands, 1-14
paternal .song complexity and offspring ,sex ratios in
Melospiza melodia. 146-152
population trends of Falco sparverius wintering in
southwestern Florida, 475^83
reproductive biology ot a grassland songbird community
in northcentral Montana, 455-464
scale-dependent response by breeding songbirds to
residential development, 296-306
short-term effects of fire on breeding birds in southern
Appalachian upland forests, 518-531
editor's comments, 834

editors. The Wilson Bulletin (1888-2005) and The Wil.xon
Journal of Ornithology (2006-20 10), 809

eggs

description tor Garrulax niaxiinus at Lianhuashan,
southern Gansu, China, 388-391
description for Uinius tephronotus in south Tibet, 395-398
description for Melanopareia torquata. 162-165
description for Zimnierius chry.xops in Venezuela, 689-698
description of nests, nest placement, and eggs of three
Philippine endemic birds, 587-591
eftects ot a Hood on foraging ecology and population
dynamics of Lininothlypis swainsonii. 165-168
egg mass variability of Dolichonyx otyz.ivorus. 432-438

842

THE WILSON JOURNAL OF ORNITHOLOGY • Vol 122, No. 4. December 2010

tirst reproductive record of Charadrius wilsonia in Bala
de Todos os Santos, northeastern Brazil, 788-
791

mimicry of eggs of Molothrus ater and grassland
passerines, 346-353

nests, eggs, and young of Amazilia cyanocephala, 592-
597

Egret, Cattle, see Bubulcus ibis
Great, see Ardea alba
Snowy, see Egretta thula
Egretta caerulea, 108, 1 14

thula, 95-101, 108, 114, 232, 532-544
Eira barbara, 450

Elaphe obsoleta, 275, 465, 470, 472, 801
vulpina, 800-802
Emberiza citronella, 171
schoeniclus, 236, 281

Emerald, Garden, see Chlorostilbon assimilis
White-bellied, see Amazilia Candida
Empidonax alnorum, 290
flavescens, 507
flaviventris, 725-737, 773
minimus, 288, 695, 725-737
oberholseri, 671, 695, 768, 770
occidentalis, 768, 770
traillii, 768, 769, 771
traillii/alnorum, 725-737
virescens, 532-544

Enkerlin-Hoeflich, Ernesto, see Ortiz-Maciel, Sonia Gab-

riela, Consuelo Hori-Ochoa, and

Eptesicus fuscus, 62

Eremophila alpestris, 290, 457, 460, 768, 770
Ergaticus spp., 29

Erickson, Wallace P., see Kerlinger, Paul, Joelle L.

Gehring, , Richard Curry, Aaftab Jain, and

John Guamaccia

Escalante, Ignacio, see Sandoval, Luis and

Eubalaena australis, 39
Eudocimus albus, 109, 114
ruber, 108, 114, 233
Eudyptes chrysocome, 190-193
chrysolophus, 190
robustus, 190-193
sclateri, 191

Eugenes fulgens, 494-502
Eumyias spp., 590
Euphagus carolinus, 773, 774
cyanocephalus, 768, 770
Euphonia hirundinacea, 35

Euphonia, Yellow-throated, see Euphonia hirundinacea
Eurypyga helias, 105, 111, 115
Euxenura maguari, 108, 1 14

Ewen-Campen, Ben S., see Hagelin, Julie C., Margaret L.
Perry, , Derek S. Sikes, and Susan Sharbaugh

F

Falco columbarius, 476, 479
mexicanus, 476
peregrinus, 3, 10, 476
rusticolus, 10

sparverius, 131, 475—483, 608, 762, 768, 770

tinnunculus, 71, 480, 755
Falcon, Peregrine, see Falco peregrinus
Prairie, see Falco mexicanus

Fang, Yun, see Wang, Jie, Chen-Xi Jia, Song-Hua Tang,
, and Yue-Hua Sun

Farias, Izeni Pires, see Jahn, Alex E., Douglas J. Levey,

, Ana Maria Mamani, Julidn Quilldn Vidoz,

and Ben Freeman
Felis catus, 57, 801

Ficedula hypoleuca, 126, 663, 672, 755
Finch, Chestnut-capped Brush, see Arremon brunneinucha
Citril, see Carduelis citrinella
House, see Carpodacus mexicanus
Purple, see Carpodacus purpureus
Zebra, see Taeniopygia guttata
Firewood-gatherer, see Anumbius annumbi
Flannery, Maureen E., see Chandler, Robert M., Peter Pyle,

, Douglas J. Long, and Steve N. G. Howell

Flaspohler, David J., see Ford, Michelle T. and

Flicker, Northern, see Colaptes auratus

Flint, Paul L., see Lewis, Tyler L., , Joel A. Schmutz,

and Dirk V. Derksen
Fluvicola pica, 105, 109, 115
Flycatcher, Acadian, see Empidonax virescens
Alder, see Empidonax alnorum
Ash-throated, see Myiarchus cinerascens
Blue-breasted Blue, see Cyornis herioti
Cordilleran, see Empidonax occidentalis
Dusky, see Empidonax oberholseri
European Pied, see Ficedula hypoleuca
Great Crested, see Myiarchus crinitus
Least, see Empidonax minimus
Olive-sided, see Contopus cooperi
Social, see Myiozetetes similis
Swainson’s, see Myiarchus swainsoni
Traill’s, see Empidonax traillii/alnorum
Willow, see Empidonax traillii
Yellow-bellied, see Empidonax flaviventris
Yellowish, see Empidonax flavescens
Ford, Michelle T. and David J. Flaspohler, Scale-dependent
response by breeding songbirds to residential
development along Lake Superior, 296-306
Foster, Charles R., Anthony F. Amos, and Lee A. Fuiman,
Phenology of six migratory coastal birds in relation
to climate change, 116-125
fox, arctic, see Alopex lagopus

gray, see Urocyon cinereoargenteus
red, see Vulpes vulpes

Frame, Paul F., Observations of a possible foraging tool by
Common Ravens, 181-182

Franzreb, Kathleen E., Red-cockaded Woodpecker male/
female foraging differences in young forest stands,
244-258

Freed, Leonard A., review. Conservation biology of
Hawaiian forest birds, edited by Thane K. Pratt,
Carter T. Atkinson, Paul C. Banko, James D.
Jacobi, and Bethany L. Woodworth, 632-633
Freeman, Ben, see Jahn, Alex E., Douglas J. Levey, Izeni
Pires Farias, Ana Maria Mamani, Julian Quill6n
Vidoz, and

Frei, Barbara, David M. Bird, and Rodger D. Titman,

INDEX TO VOLUME 122

843

Bobolink egg mass variability and nestling growth
patterns, 432-438

Fuchs, Eric J., see Chavam'a-Pizarro, Tania, Gustavo

Gutidrrez-Espeleta, , and Gilbert Barrantes

Fuiman, Lee A., see Foster, Charles R., Anthony F. Amos,
and

Fulica americana, 569, 768, 770
Fundulus heteroclitus, 95
Furnarius rufus, 421

G

Gadwall, see Anas strepera
Galeocerdo cuvier, 378-380
Gallinago delicata, 768, 771
gallinago, 62
paraguaiae, 108, 115

Gallinule, Azure, see Porphyria flavirostris
Purple, see Porphyria martinica
Gallus gallus domesticus [Gallus domesticus, 762, 764,
Gallus gallus, 615], 755, 759
Gambusia affinis, 623

Garrettson, Pamela R., see Lewis, Thomas E. and

Garrulax elliotii, 391
henrici, 391
maximus, 388-391
ocellatus, 388, 391
sukatschewi, 391
Gavia immer, 376, 716-724

Geese, Aleutian Cackling, see Branta hutchinsii leucopar-
eia

Gehring, Joelle L., see Kerlinger, Paul, , Wallace P.

Erickson, Richard Curry, Aaftab Jain, and John
Guamaccia

Gelochelidon [Sterna, 116-125] nilotica. 111, 115
genetics

differences between Vermivora chrysoptera and V. pinus
in two habitats, 273-278

reponed hybridization of Pipilo aberti and P. fuscus,
399-402

variation between migratory and non-migratory Tyran-
nus melancholicus during spring migration in
central South America, 236-243
variation of Chlorospingus pileatus from Costa Rica and
western Panama, 279-287

Geothlypis trichas, 290, 518-544, 549-551, 646, 768, 770
Geotrygon chiriquensis, 509

Germain, Ryan R., Matthew W. Reudink, Peter P. Marra,
and Laurene M. Ratcliffe, Carotenoid-based male
plumage predicts parental investment in the Amer-
ican Redstart, 318-325

Ghestemme, Thomas, see Kesler, Dylan C., ,

Emmanuelle Portier, and Anne Gouni
Gibbs, James P., see Shriver, W. Gregory, Thomas P.

Hodgman, , and Peter D. Vickery

Glaucomys volans, 128, 129, 132, 244
Gnatcatcher, Blue-gray, see Polioptila caerulea
Gnorimopsar [Gnorimpsar, 417] chopi, 795-799
Goldeneye, Common, see Bucephala clangula
Goldfinch, American, see Spinus tristis
Gomes, Henrique Belfort and Marcos Rodrigues, The nest
of the Cipd Canastero (Asthenes luizae), an

endemic furnariid from the Espinhafo Range,
southeastern Brazil, 600-603

Goodwin, Steve, see Olson, Storrs L., Horace Loftin, and

Goose, Cackling, see Branta hutchinsii
Canada, see Branta canadensis
Greater White-fronted, see Anser albifrons
Lesser Snow, see Chen caerulescens caerulescens
Orinoco, see Neochen jubata
gopher, northern pocket, see Thomomys talpoides
Goulding, William and Thomas E. Martin, Breeding
biology of the Golden-faced Tyrannulet (Zimmerius
chrysops) in Venezuela, 689-698
Gouni, Anne, see Kesler, Dylan C., Thomas Ghestemme,

Emmanuelle Portier, and

Gouse, Paula J., see Jones, Stephanie L., J. Scott Dieni, and

Graber, Allen E., see Davis, Craig A., David M. Leslie Jr.,

W. David Walter, and

Crackle, Boat-tailed, see Quiscalus major
Common, see Quiscalus quiscula
Great-tailed, see Quiscalus mexicanus
Grallaria alleni, 394
carrikeri, 394
excelsa, 394
flavotincta, 394
guatimalensis, 394
haplonota, 394
hypoleuca, 394
nuchalis, 394
ridgelyi, 392-395
ruficapilla, 394
rufula, 393
varia, 394
watkinsi, 394

Graves, Bret M., Amanda D. Rodewald, and Scott D. Hull,
Influence of woody vegetation on grassland birds
within reclaimed surface mines, 646-654
Grebe, Eared, see Podiceps nigricollis
Homed, see Podiceps auritus
Least, see Tachybaptus dominicus
Red-necked, see Podiceps grisegena
Greeney, Harold F. and Mery E. Juina J, First description of
the nest of Jocotoco Antpitta (Grallaria ridgelyi),
392-395

Gregory, Mark S., see Vilella, Francisco, J., , and

Guy A. Baldassarre

Gren, Eric, see Hayes, William K., Elwood D. Bracey,

Melissa R. Price, Valerie Robinette, , and

Caroline Stahala

Grosbeak, Black-headed, see Pheucticus melanocephalus
Blue, see Passerina caerulea
Rose-breasted, see Pheucticus ludovicianus
Ground-Dove, Common, see Columbina passerina
Grouse, Prairie, see Tympanuchus spp.

Chinese (Severtzov’s), see Tetrastes sewerzowi
Red, see Lagopus lagopus scoticus
Ruffed, see Bonasa umbellus
Grus canadensis, 570

canadensis nesiotes, 556-562
canadensis pratensis, 561

844

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

canadensis piilla. 561

Guarnaccia. John, see Kerlinger. Paul, Joelle L. Gehring,
Wallace P. Erickson. Richard CuiTy, Aaftab Jain,
and

Guglielmo. Christopher G., see Seewagen. Chad L. and
Gull, see Laras spp.

Black-headed, see Chroicocephalas ridibandas

Glaucous-winged, see Laras glaacescens

Herring, see Laras argentatas

Kelp, see Diras dominicanas

Laughing, see Laras atricilla

Les.ser Black-backed, see Laras fascas

Olrog’s, see Laras atlanticas

Guo. Hanzhao. see Cao, Lei, Giiixia Zhao, Shan Tang, and

Gutierrez-Espeleta, Gustavo, see Chavarn'a-Pizarro, Tania,

, Eric J. Fuchs, and Gilbert Barrantes

Gymnagyps californianas, 354
Gypsophila spp.. 589
Gyrfalcon, see Falco ra.sticolas

H

habitat

Cotarnicops noveboracensis wintering in Oklahoma,
385-387

habitat type is related to nest mass and fledging success
of Phylloscopas borealis, 699-705
home range, habitat use. and nest site characteristics of
Ictinia mississippiensis in the White River
National Wildlife Refuge, Arkansas, 706-715
influence of cover and food resource variation on post-
breeding bird use of timber harvests with
residual canopy trees, 54,5-555
influence of woody vegetation on grassland birds within
reclaimed surface mines, 646-654
migrant songbird species distribution and habitat use
during stopover on two islands in the Gulf of
Maine, 725-737

Fieoides borealis male/female foraging differences in
young forest stands, 244-258
Rhynchopsitta terrisi breeding home range and habitat
selection in the northern Siena Madre Oriental.
Mexico. 513-517

short-term effects of fire on breeding birds in southern
Appalachian upland forests. 518-531
temporal and spatial shifts in habitat use by Branla
bernicla nigricans immediately following flight-
less molt. 484-493

use by Aerodramas faciphagas ine.xpectatas and Collo-
calia escalenia affinis in the Andaman Islands,
India. 259-272

use by Anvnodramas nelsoni and A. caadacalas. 340-
.345

llagclin. Julie C.. Margaret L. FArry. Ben S. Ewen-Campen,
Derek S. Sikes, and Susan Sharbaugh. Habitat type
is related to nest mass and fledging success of
Arctic Warblers, 699-705

Hahn. D. Caldwell, see Kozlowski, Corinne P. and

Haig, Susan M., see Johnson. Matthew, Daniel R.

Ruthrauff, Brian J. McCaffery. , and Jeffrey

R. Walters

Haight, Lois T., see John.son, R. Roy and

Halevi, Shay, see Charter, Motti, Yossi Leshem, , and

Ido Izhaki

Harpactes spp., 176

Harpia harpyja, 792-795

Harrier. Northern, see Circas cyaneas

Hatherley, Graham, see Seymour, Adrian S., ,

Francisco Javier Contreras. James Aldred, and
Fergus Beeley

Hawk. Broad-winged, see Bateo platypteras
Cooper's, see Accipiter cooperi
Galapagos, see Bateo galapagoensis
Hams’, see Parabateo anicinctas
Red-shouldered, see Bateo lineatas
Red-tailed, see Bateo jainaecensis
Sharp-shinned, see Accipiter striatas
Swainson's, see Bateo swainsoni
Hayes, William K., Elwood D. Bracey, Melissa R. Price,
Valerie Robinette. Eric Gren. and Caroline Stahala,
Population status of Chuck-will's-widow (Capri-
nialgas caroiinensis) in the Bahamas, 381-384
Heinrich, Bernd, review. The Eastern Screech Owl: life
history, ecology, and behavior in the suburbs and
countryside, by Frederick R. Gehibach, 202-203
Heliornis falica, 105, 106, 108, 109, 115
Helmitheros vermivonan. 518-531

Henry, Annette E., see Jehl Jr., Jo.seph and

Heron, Black-crowned Night, see Nycticora.x nycticorax
Capped, see Pilherodias pileatas
Cocoi, see A i dea cocoi
Great Blue, see Ardea herodias
Green, see Batorides virescens
Little Blue, see Egretta caeralea
Striated, see Batorides striatas
Whistling, see Syrignia sihilatri.x
Yellow-crowned Night, see Nyctanas.xa violacea
Hilic, David, see Siegfried, Dennis G., Daniel S. Linville,
and

Himantopas me.xicanas, 105, 108. 115
Hinnebusch. Daniel M.. Jean-Fran^ois Therrien, Marc-
Andre Valiquette, Bob Robertson, Sue Robertson,
and Keith L. Bildstein, Survival, site fidelity, and
population trends of American Kestrels wintering
in southwestern Florida, 475^83
Hirando rustica, 138, 622, 663, 768. 770
Hoatzin. see Opisthoconuis hoazin

Hodgman, Thomas, P., see Shriver, W. Gregory, ,

James P. Gibbs, and Peter D. Vickery
hog, wild, see Sas scrofa
home range

home range, habitat use, and nest site characteristics of
Ictinia mississippiensis in the White River
National Wildlife Refuge, Arkansas, 706-715
Rhynchopsitta terrisi breeding home range and habitat
selection in the northern Sierra Madre Oriental,
Mexico, 5 1 .3-5 1 7

size and site tenacity of overwintering Anunodrainas
lecontii in a fire managed prairie. 139-145
size i'or Anunodrainas nelsoni and A. caadacatas, 340-345

INDEX TO VOLUME 122

845

Hoploxyptenis cayanits. I 15

Hopp, Steven L., see Johnson. R. Roy and

Hori-Ochoa, Consuelo, see Orliz-Maciel. Sonia Gabriela,

, and Ernesto Enkerlin-Hoeflich

Hornbill, Rufous, see Biiceros hydrocorax

Homero, Rufous, see Fiinuiriiis nif'us

Howell, Steve N. G., see Chandler, Robert M.. Peter Pyle,

Maureen E. Elannery, Douglas J. Long, and

Hull, Scott D.. see Graves, Bret M., Amanda D. Rodewald,
and

Hummingbird. Azure-crowned, see AmazUia cycinocephala
Black-chinned, see Archilochus alexandri
Blue-throated, see Lampornis clemenciae
Broad-tailed, see Selasphorus platycercus
Fiery-throated, see Panterpe insignis
Magnificent, see Eugenes fidgens
Ruby-throated, see Archilochus colubris
Sapphire-throated, see Lepidopyga coeruleogularis
Violet-crowned, see AmazUia violiceps
Volcano, see Selasphorus flammula
hunting

biological, geographical, and cultural origins of the
Gavia immer hunting tradition in Carteret
County, North Carolina, 716-724
Hylocharis leucofis, 496

Hylocichla mustelina, 290, 507, 509, 510, 518-531, 545
spp. 743

Hypothymis azurea, 591
helenae. 587-591
helenae helenae. 591
helenae personata, 591

I

Ibis. American White, see Eudocimus alhus
Bare-faced, see Phimosus infuscatus
Buff-necked, see Theristicus caudatus
Crested, see Nipponia nippon
Glossy, see Plegadis falcinellus
Green, see Mesemhrinihis cayennensis
Scarlet, see Eudocimus ruber
Sharp-tailed, see Cere ibis oxycerca
Icteria virens [vireos. 524|, 532-544, 725-737, 780-783,
806-809

Icterus bullockii, 768, 770
galbula. 725-737
spurius, 532-544, 782
Ictinia mississippiensis. 706-7 1 5
Index, 835-858

Indigobird, Village, see Vidua chalybeata
Ixobrychus exilis. 106. 108, 114

Izhaki. Ido. see Charter. Motti. Yossi Leshem. Shay Halevi.

J. Mery E. Juifia, see Greeney, Harold F. and

Jabiru. see Jabiru mycieria
Jabiru mycteria. 108. 1 14

Jablonski, Kevin E.. Stacy A. McNulty, and Matthew D.
Schlesinger, A digital spot-mapping method for
avian field studies, 772-776
Jacana jacana. 102, 105. 106. 108. 109. 115
Jacana. Wattled, see Jacana Jacana

Jahn. Alex E., Douglas J. Levey, Izeni Pires Farias, Ana
Maria Mamani, Julian Quillen Vidoz, and Ben
Freeman, Morphological and genetic variation
between migratory and non-migratory Tropical
Kingbirds during spring migration in central South
America, 236-243

Jain. Aaftab, .see Kerlinger. Paul, Joelle L. Gehring.

Wallace P. Erickson. Richard Cuiry, , and

John Guarnaccia

James, Douglas A., .see Leasure, Douglas R., Ragupathy
Kannan, and

Jay, Azure-hooded, see Cyanolyca cucullata
Blue, see Cyanocitta cristata
Brown, see Cyanocorax morio
Gray, see Perisoreus canadensis
Plush-crested, see Cyanocorax chrvsops
Steller’s, see Cyanocitta stelleri
Jehl Jr., Joseph and Annette E. Henry, The postbreeding
migration of Eared Grebes, 217-227

Jeske, Clinton W., see Baldwin, Heather Q., , Melissa

A, Powell, Paul C. Chadwick, and Wylie C. Barrow
Jr.

Jesser, William A., see Rijke, Arie M. and

Jia, Chen-Xi, see Wang, Jie, , Song-Hua Tang, Yun

Fang, and Yue-Hua Sun

Jie, Wang, Yang Chen, Lu Nan, Fang Yun, and Sun Yue-
Hua, Diet of Chinese Grouse (Tetrastes sewerzowi)
during preincubation, 177-180
Johnson, Matthew. Daniel R. Ruthrauff. Brian J. McCaff-
ery, Susan M. Haig, and Jeffrey R. Walters.
Apparent survival of breeding Western Sandpipers
on the Yukon-Kuskokwim River Delta. Alaska. 15-
22

Johnson, R. Roy and Lois T. Haight, Occasional mimicry
and night-time singing by the western Curve-billed
Thrasher (Toxostoma cuixirostre palmeri). 625-
626

Johnson. R. Roy and Steven L. Hopp, The first reported
hybridization of Abert's and Canyon towhees
(Pipilo spp.), 399-402

Jones. Stephanie L., J. Scott Dieni, and Paula J. Gousc.
Reproductive biology of a grassland songbird
community in northcentral Montana. 455-464
Jones. Todd M., Amanda D. Rodewald, and Daniel P.
Shustack. Variation in plumage coloration of
Northern Cardinals in urbanizing landscapes. 326-
333

Judd, Eric L., .sec Butler, Christopher .1.. Lisa H. Pham. Jill

N. Stinedurf. Christopher L. Roy, . Nathanael

J. Burgess, and Gloria M. Caddell
Junco. Dark -eyed, sec Junco hyemalis
Yellow-eyed, see Junco phaeonotus
Junco hyemalis. 291. 518-531
phaeonotus. 6 1 4-6 1 7

K

Kahn. H., .see Quinn. J. S.. A. Samuefsen, M. Barclay. G.
Schmaltz, and

Kaler. Robb S. A.. Steve E. Ebbert. Clait E. Braun, and
Brett K. Sandercock. Demography of a reintro-

846

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4. December 2010

duced population of Evermann’s Rock Ptarmigan in
the Aleutian Islands, 1-14

Kanegae, Mieko F., Marina Telles, Severino A. Lucena,
and Jose Carlos Motta-Junior, Observations on the
breeding biology of the Collared Crescentchest
{Melanopareia torquata), 162-165

Kannan, Ragupathy, see Leasure, Douglas R., , and

Douglas A. James
Kenopia spp., 589

Kerlinger, Paul, Joelle L. Gehring, Wallace P. Erickson,
Richard Curry, Aaftab Jain, and John Guamaccia,
Night migrant fatalities and obstruction lighting at
wind turbines in North America, 744-754
Kesler, Dylan C., Thomas Ghestemme, Emmanuelle
Portier, and Anne Gouni, Cooperative breeding of
the Society Kingfisher {Todiramphus veneratus),
46-50

Kestrel, American, see Falco sparverius
Common, see Falco tinnunculus

Keyes, Tim S., see Klaus, Nathan A., Scott A. Rush, ,

John Petrick, and Robert J. Cooper
Keynan, Oded and Reuven Yosef, Annual precipitation
affects reproduction of the Southern Grey Shrike
(Lanius meridionalis), 334-339
Kielb, Michael A., review. The migration of birds: seasons
on the wing, by Janice M. Hughes, 201
PCilldeer, see Charadrius vociferus

King, Ben F., review, Birds of east Asia: China, Taiwan,
Korea, Japan, and Russia, by Mark Brazil, 203-205
Kingbird, Eastern, see Tyrannus tyrannus
Tropical, see Tyrannus melancholicus
Western, see Tyrannus verticalis
Kingfisher, Amazon, see Chloroceryle amazona
American Pygmy, see Chloroceryle aenea
Belted, see Megaceryle alcyon
Collared, see Todiramphus chloris
Green, see Chloroceryle americana
Green-and-rufous, see Chloroceryle inda
Micronesian, see Todiramphus cinnamominus
Pied, see Ceryle rudis
Ringed, see Megaceryle torquata
Society, see Todiramphus veneratus
Tuamotu, see Todiramphus gambieri
Kinglet, Golden-crowned, see Regulus satrapa
Ruby-crowned, see Regulus calendula
kingsnake, Sonoran mountain, see Lampropeltis pyrome-
lana

Kirkpatrick, Chris and Courtney J. Conway, Nest predators
of ground-nesting birds in montane forest of the
Santa Catalina Mountains, Arizona, 614-617
Kirkpatrick, Chris, see Robinson, Gabrielle L., Courtney J.

Conway, , and Dominic D. LaRoche

Kiskadee, Great, see Pitangus sulphuratus
Kite, Mississippi, see Ictinia mississippiensis
Klamm Service Award, 2010: Jerome A. Jackson, 810-81 1
Klaus, Nathan A., Scott A. Rush, Tim S. Keyes, John
Petrick, and Robert J. Cooper, Short-term effects of
fire on breeding birds in southern Appalachian
upland forests, 518-531

Klippenstine, Dwight R. and Spencer G. Sealy, Assessing
generalized egg mimicry: a quantitative compari-

son of eggs of Brown-headed Cowbirds and
grassland passerines, 346-353
Klug, Page E., L. LaReesa Wolfenbarger, and John P.
McCarty, Snakes are important nest predators of
Dickcissels in an agricultural landscape, 799-803
Knopf, Fritz L., Composition dynamics of an avian
assemblage, 767-771

Koenig, Walter D., review, Wytham Woods: Oxford’s
ecological laboratory, edited by Peter S. Savill,
Chistopher M. Perrins, Keith J. Kirby, and Nigel
Fisher, 813-814

Koh, Chao-Nien, see Lin, Shang-Yao, Fang-Cian Lu, Fu-
Hsien Shan, Shu-Ping Liao, Jing-Ling Weng, Wei-
Jen Cheng, and

Kookaburra, Laughing, see Dacelo novaeguineae
Kozlowski, Corinne P. and D. Caldwell Hahn, Develop-
mental changes in serum androgen levels of Eastern
Screech-Owls {Megascops asio), 755-761
Kratter, Andrew W., see Reetz, Matthew J., Jacob M.
Musser, and

Kumar, Anjali, see O’Donnell, Sean, , and Corina

Logan

L

La Sala, Luciano and Sergio Martorelli, First report of
Olrog’s Gull depredation by sympatric Kelp Gulls,
188-189

Labidus praedator, 503, 504, 693
spinninodis, 504
Lagopus lagopus, 4, 10
lagopus scoticus, 179
leucura, 2

muta evermanni, 1-14 (Frontispiece)
muta gabrielsoni, 2
Lampomis clemenciae, 494-502
Lampropeltis calligaster, 801
getula, 801
pyromelana, 616
Lanius bucephalus, 436
collurio, 398
excuhitor, 334. 398
ludovicianus, 210, 398, 457
meridionalis, 334-339
schach, 398
senator, 398
tephronotus, 395-398

Lapwing, Pied, see Hoploxypterus cayanus

Lark, Horned, see Eremophila alpestris

LaRoche, Dominic D., see Robinson, Gabrielle L.,

Courtney J. Conway, Chris Kirkpatrick, and

Larus argentatus, 116-125
atlanticus, 188-189
atricilla, 95-101
dominicanus, 39-45, 188-189
fuscus, 755
glaucescens, 3, 209
occidentalis, 44
spp., 457

Laughingthrush, Brown-cheeked, see Garrulax henrici
Elliot’s, see Garrulax elliotii
Giant, see Garrulax maximus

INDEX TO VOLUME 122

847

Snowy-cheeked, see Garrulax sukatschewi
Spotted, see Garrulax ocellatus
Leasure, Douglas R,, Ragupathy Kannan, and Douglas A.
James, House Spairows associated with reduced
Cliff Swallow nesting success, 135-138
Lee, Thomas D., see Suomala, Rebecca W., Sara R. Morris,

Kimberly J. Babbitt, and

Leiser, John K., see Schreffler, Lisa, , and Terry L.

Master

Leonard, Wendy J., Jayne Neal, and Rama Ratnam,
Variation of Type B song in the endangered
Golden-cheeked Warbler (Dendroica chrysoparia),
777-780

Lepidocolaptes affinis, 609
souleyetii, 805

Lepidopyga coeruleogularis, 194-195
Leptoptilos crumeniferus, 233

Leshem, Yossi, see Charter, Motti, , Shay Halevi, and

Ido Izhaki

Leslie Jr., David M., see Davis, Craig A., , W. David

Walter, and Allen E. Graber
Leucocarbo atriceps, 39

Levey, Douglas J., see Jahn, Alex E., , Izeni Fires

Farias, Ana Maria Mamani, Julian Quillen Vidoz,
and Ben Freeman

Lewis, Thomas E. and Pamela R. Garrettson, Parasitism of
a Blue-winged Teal nest by a Northern Shoveler in
South Dakota, 612-614

Lewis, Tyler L., Paul L. Flint, Joel A. Schmutz, and Dirk V.
Derksen, Temporal and spatial shifts in habitat use
by Black Brant immediately following flightless
molt, 484-493

Liang, Shi, see Xiao-Ping, Yu, Xi Yong-Mei, Lu Bao-

Zhong, Li Xia, Gong Ming-Hao, , and Dong

Rong

Liao, Shu-Ping, see Lin, Shang-Yao, Fang-Cian Lu, Fu-

Hsien Shan, , Jing-Ling Weng, Wei-Jen

Cheng, and Chao-Nien Koh
Limnothlypis swainsonii, 75-81, 165-168
Limpkin, see A ramus guarauna

Lin, Shang-Yao, Fang-Cian Lu, Fu-Hsien Shan, Shu-Ping
Liao, Jing-Ling Weng, Wei-Jen Cheng, and Chao-
Nien Koh, Breeding biology of the Taiwan Barbet
(Megalaima nuchalis) in Taipei Botanical Garden,
681-688

Linville, Daniel S., see Siegfried, Dennis G., , and

David Hille
lion, see Panthera leo

Loftin, Horace, see Olson, Storrs L., , and Steve

Goodwin

Logan, Corina, see O’Donnell, Sean, Anjali Kumar, and

Lomdscolo, Silvia B., Carolina Monmany, Agustina Mal-
izia, and Thomas E. Martin, Flexibility in nest-site
choice and nesting success of Turdus rufiventris
(Turdidae) in a montane forest in northwestern
Argentina, 674-680

Long, Douglas J., see Chandler, Robert M., Peter Pyle,

Maureen E. Flannery, , and Steve N. G.

Howell

Longspur, Chestnut-collared, see Calcarius ornatus

McCown’s, see Calcarius mccownii
Loon, Common, see Gavia immer

Lu, Fang-Cian, see Lin, Shang-Yao, , Fu-Hsien Shan,

Shu-Ping Liao, Jing-Ling Weng, Wei-Jen Cheng,
and Chao-Nien Koh

Lu, Xin, Chen Wang, and Tonglei Yu, Nesting ecology of
the Grey-backed Shrike (Lanius tephronotus) in
south Tibet, 395-398

Lucena, Severino A., see Kanegae, Mieko F., Marina

Telles, , and Jose Carlos Motta-Junior

Lunardi, Vitor 0. and Regina H. Macedo, First reproductive
record of Wilson’s Plover in Baia de Todos os
Santos, northeastern Brazil, 788-791
Luscinia megarhynchos, 780, 782
Lynx rufus, 615, 616

M

Macaca nemistrina leonina, 173-177
macaques, pig-tailed, see Macaca nemistrina leonina
MacDougall-Shackleton, Elizabeth A., see Potvin, Domi-
nique A. and

Macedo, Regina H., see Lunardi, Vitor O. and

Machetornis rixosa, 795-799

mackerel, Spanish, see Scomberomorus maculatus

Macronectes giganteus, 39

Magicicada spp., 404

Magpie, Black-billed, see Pica hudsonia

Mahler, Bettina, see De Marsico, Maria C., , and Juan

C. Reboreda

Mahler, Bettina, see Di Giacomo, Alejandro G., , and

Juan C. Reboreda
Malacosoma americana, 404

Malizia, Agustina, see Lomascolo, Silvia B., Carolina

Monmany, , and Thomas E. Martin

Mall2Lrd, see Anas platyrhynchos

Mamani, Ana Maria, see Jahn, Alex E., Douglas J. Levey,

Izeni Pires Farias, , Julian Quilldn Vidoz, and

Ben Freeman
Manacus vitellinus, 321

Manakin, Golden-collared, see Manacus vitellinus
Manchi, Shirish S. and Ravi Sankaran, Foraging habits and
habitat use by Edible-nest and Glossy swiftlets in
the Andaman Islands, India, 259-272
Marra, Peter P., see Germain, Ryan R., Matthew W.

Reudink, , and Laurene M. Ratcliffe

Marshbird, Brown-and-yellow, see Pseudoleistes virescens
Marti, Carl D., Dietary trends of Bam Owls in an
agricultural ecosystem in northern Utah. 60-67

Martin, Thomas E.. see Goulding, William and

Martin, Thomas E., see Lomascolo, Silvia B., Carolina

Monmany, Agustina Malizia, and

Martin, Thomas E., see Ruggera, Romin and

Martorelli, Sergio, see La Sala, Luciano and

Massey, Phillip C., see Boves, Than J., David A. Buehler,
and

Master, Terry L., see Schreffler, Lisa, John K. Leiser, and

McCaffery, Brian J., see Johnson, Matthew, Daniel R.

Ruthrauff, , Susan M. Haig, and Jeffrey R.

Walters

848

THE WILSON JOURNAL OL ORNITHOLOGY • Uo/. 122. No. 4. December 2010

McCarty. John P., see King, Page E., L. LaReesa
Wolfenbarger, and

McDermott. Molly E. and Petra Bohall Wood. Influence of
cover and food resource variation on post-breeding
bird use of timber harvests with residual canopy
trees. 545-555

McNulty. Stacy A., see Jablonski. Kevin E., , and

Matthew D. Schlesinger

McWhorter. Todd J., see Powers. Donald R., Jessamyn A.

Van Hook. Elizabeth A. Sandlin, and

Meadowlark, Eastern, see Sturnella magna
Western, see Sturnella neglecta
Megacery'le alcyon. 768, 770
tarquata, 109, 111, 115
Megakiima nuchalis, 681-688
oorti nuchalis, 681
rajflesii, 681
riihricapilla, 686, 687
riridi.'i, 681

Megascops asio, 128, 129, 131, 608, 755-761
Melanerpes caroliniis, 128, 129, 132,518-544
erythrocephaliis, 128, 160-162, 532-544, 608
fonnicivorous, 46, 160, 370, 608-61 1
striatiis, 160
Melaniita fusca, 376
nigra, 377
perspicillata, 376
Melanopareia niaximiliani, 164
torquata. 162-165

Meleagris gallopavo, 290, 518-531, 762
Melletti, Mario and Marzia Mirabile, Hanging behavior of
the Hooded Crow (Con'us corni.x), 183-185
Melospiza georgiana, 725-737
lincolnii, 725-737, 768, 771, 773
melodia. 146-152, 207, 290, 302, 307-317, 518-531,
671, 768, 771, 806-809
Meiozone ieucotis, 503-5 1 2
Mephitis mephitis, 646, 801, 802

Meretsky, Vicky J., see Brittain. Ross A., . and Chris

B. Craft

Merganser. Red-breasted, .see Mergiis serrator
Mergiis serrator, 377
spp.. 717

Merlin, see Falco coluniharius
Merops apiaster, 378-380
huUockoides, 46. 48
Mesemhrinihis cayennensis. I 14
methods

a digital spot-mapping method for avian field studies.
111-llb

breeding patterns of Ammodramus henslowii and sym-
patric grassland sparrow species. 635-645
Microtus niontanus. 61. 62
pennsxlvanicits. 61. 62
spp.. 60. 61. 64. 65. 457, 801
migration

concentrated migratory morning llight at Lake Pont-
chartrain. Louisiana. U.SA, 738-743
Coturnicops novehoracensis wintering in Oklahoma,
38.5-387

effects of fat and lean body mass on migratory landbird
stopover duration, 82-87

flight speeds of migrating Podiceps grisegena and P.
aiiritus, 374-378

migrant songbird species distribution and habitat use
during stopover on two islands in the Gulf of
Maine. 725-737

morphological and genetic variation between migratory
and non-migratory Tyrannus melancholicus in
central South America, 236-243
nocturnal social cues attract migrating Icteria virens,
780-783

phenology of migration in relation to climate change,
1 16-125

postbreeding migration of Podiceps nigricollis, 217-
227

Miller, Karl E., Nest-site limitation of secondary cavity-
nesting birds in even-age southern pine forests,
126-134

Miller, Matthew J., review. The birds of the Republic
of Panama. Part 5. Gazetteer and bibliography,
by Deborah C. Siegel and Storrs L. Olson, 199-
200

Mimus poiygiottos, 532-544, 625-626, 762
satiirninus, 420, 421, 423. 425
Ming-Hao, Gong, see Xiao-Ping, Yu. Xi Yong-Mei, Lu

Bao-Zhong, Li Xia, , Shi Liang, and Dong

Rong

mink, American, see Neovison vison
minnow, sheepshead, see Cyprinodon variegatus
Mionectes macconnelli, 695
oleagineus, 695

Mirabile, Marzia, see Melletti. Mario and

Mniotilta varia, 302. 507. 518-531, 72.5-737
Mock. Douglas W. and P. L. Schwagmeyer. Not the Nice
sparrow. The 2007 Margaret Morse Nice Lecture,
207-216

Mockingbird, Chalk-browed, see Mimus .saturninus
Northern, see Mimus poiygiottos
Molothrus ater, 75-81, 150, 186, 291, 307-317, 346-353,
366. 367, 404, 459, 460, 462, 463, 467, 532-544.
652, 671, 768, 770, 796. 799-803
honariensis, 365-369, 417-431. 600-603, 795-799
rufoa.xil laris, 417^31, 795-799
Momotus momota, 503-5 1 2
Monarch, Short-crested, see Hypothymis helenae
mongoose, see Crossarchus spp.
monkey, brown capuchin, see Cehus apella
capuchin, see Cehus olivaceus
mantled howler, see Aloiiatia palliata
red-howler, see Alouatia seniculus

Monmany. Carolina, see Loma.scolo, Silvia B., ,

Agustina Malizia, and Thomas E. Martin
moose, .see Alces alces

Morgante, Jo5o Stenghel, see de Mendon^a Dantes, Gi.sele

Pi res and

morphology

inlluence of age and season on measurements of
Streptoprocne hiscutata, 153-159
night migrant fatalities and obstruction lighting at wind
turbines in North America, 744-754

INDEX TO VOLUME 122

849

variation between migratory and non-migratory Txnin-
niis nielancholiciis during spring migration in
central South America, 236-243
variation of Clilorosiviigiis pileatus from Costa Rica and
western Panama, 279-287

Morris, Sara R., see Suomala, Rebecca W., ,

Kimberly J. Babbitt, and Thomas D. Lee
mortality or injury

annual bird mortality in the bitumen tailings ponds in
northeastern Alberta, Canada, 569-576
night migrant fatalities and obstruction lighting at wind
turbines in North America, 744-754
response to nestling throat ligatures by three songbirds,
806-809

Morton, Martin L. and Maria E. Pereyra, Development of
incubation temperature and behavior in thrushes
nesting at high altitude, 666-673
mosquitofish, see Gamhiisia affinis
Motmot, Blue-crowned, see Mumotiis momota
Motta-Junior, Jose Carlos, see Kanegae, Mieko F., Marina

Telles, Severino A. Lucena, and

mouse, deer, see Peromyscits spp.
see Mus musculus

Moyle, Robert G„ see Sanchez-Gonzalez, Luis A., Carl

Oliveros, Nevong Puna, and

Miigil cephalus. 111, 720
mullet, striped, see Miigil cephalus

Mumme, Ronald L., Breeding biology and nesting
success of the Slate-throated Whitestart (Myio-
borus miniatus) in Monteverde, Costa Rica,
29-38

mummichog, see Fimditlus heteroclitus
Murray, Leigh, see Berardelli, Daniele, Martha J. Desmond,
and

Mus musculus. 61, 62, 65, 476
spp., 64

muskrat, see Ondatra zihethicus

Musser, Jacob M., see Reetz, Matthew J., , and

Andrew W. Kratter
Mustela altaica. 397
frenata, 62, 457, 801
Mycteria americana. 108, 1 14
Myiarchus cinerascens, 1 3 1

crinitus. 126-134, 290, 518-544,
swainsoni, 242

Myioborus bnmniceps. 33, 35, 451
flavivertex, 45 1
melanocephalus. 33, 451
miniatus. 29-38, 447-454, 507
miniatus aurantiacus. 45 1
miniatus ballux. 447
miniatus hellmayri. 45 1
miniatus pallidiventris. 45 I
ornatus. 451
pictus, 34, 45 1
torqiiatus. 33, 35, 45 1
Myiozetetes similis. 596
My Otis spp., 62

Mynneciz.a immaculata. 507, 509

N

Nan, Lu, see Jie, Wang, Yang C'hen, , Fang Yun, and

Sun Yue-Hua
Napothera spp., 589
natural history

of Myioborus miniatus in Venezuela, 447^54
Navarro-Sigiienza, Adolfo G., review. Moments of discov-
ery. Natural history narratives from Mexico and
Central America, edited by Kevin Winker. 812-813

Neal, Jayne, see Leonard. Wendy J., . and Rama

Ratnam

Neochen jubatu. 108, 114
Neotoma mexicana. 615, 616
Neovison vison, 799-803
nest

breeding biology of Zimmerius chrvsops in Venezuela.
689-698

desciption of Asthenes luizae nest from the Espinha90
Range, southeastern Brazil, 600-603
description of Chlorostiibon assimilis nest from Costa
Rica, 597-599

description of Myioborus miniatus nest in Venezuela,
447^54

description of nests and eggs of MeUmopareia torquata.
162-165

description of nests, nest placement, and eggs of three
Philippine endemic birds, 587-591
first description of the nest of Grallaria ridgelvi. 392-
395

habitat type is related to nest mass and fledging success
of Phylloscopus borealis. 699-705
nests, eggs, and young of Atnazilia cyanocephala. 592-597
observations and predation of a Carpococcyx renauldi
nest in northeastern Thailand, 173-177
of Garrulax maximus at Lianhuashan. southern Gansu,
China, 388-391

of Myioborus miniatus in Monteverde. Costa Rica, 29-
38

nesting

home range, habitat use, and nest site characteristics of
Ictiiua mississippiensis in the White River
National Wildlife Refuge, Arkansas, 706-715
nest box use by Pants major in semi-arid rural
residential gardens, 604-608
nest sites of Pharomachrus mocinno: relationship
between nest and snag heights, 608-61 1
nest-site choice and nesting success of Turdus ruflventris
in a montane forest in northwestern Argentina,
674-680

nest-site limitation of secondary cavity-nesting birds in
even-age southern pine forests. 126-134
of Myioborus miniatus in Monteverde. Costa Rica, 29-
38

Passer domesticus associated with reduced Petrocheli-
don pyrrhonota nesting success. 135-138

nestling

effect of nest parasitism by Molothrus rufoaxillaris and
M. bonariensis on nestling growth of Age-
laioides budius. 4 1 7—13 1

first reproductive record of Charadrius wilsonia in Bai'a

850

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4. December 2010

de Todos os Santos, northeastern Brazil, 788-
791

growth patterns of Dolichonyx oryzivorus, 432^38
growth rate of Zimmerius chrysops in Venezuela, 689-
698

nests, eggs, and young of Amazilia cyanocephala, 592-
597

Nighthawk, Antillean, see Chordeiles gundlachii
Common, see Chordeiles minor
Night-Heron, Black-crowned, see Nycticorax nycticorax
Yellow-crowned, see Nyctanassa violacea
Nightingale, Common, see Luscinia megarhynchos
Nightingale-Thrush, Orange-billed, see Catharus aurantiir-
ostris

Ruddy-capped, see Catharus frantzii
Slaty-backed, see Catharus fuscater
Nipponia nippon, 228-235

Nol, Erica, Simone Williams, and Brett K. Sandercock,
Natal philopatry and apparent survival of juvenile
Semipalmated Plovers, 23-28
Numenius phaeopus, 532-544
Nuthatch, Brown-headed, see Sitta pusilla
Pygmy, see Sitta pygmaea
Red-breasted, see Sitta canadensis
White-breasted, see Sitta carolinensis
Nyctanassa violacea, 1 14, 532-544
Nyctea scandiaca, 3, 10
Nycticorax nycticorax, 1 14, 232, 768, 770

O

O’Donnell, Sean, Anjali Kumar, and Corina Logan, Army
ant raid attendance and bivouac-checking behavior
by neotropical montane forest birds, 503-512
Odocoileus virginianus, 801

Oliveros, Carl, see Sdnchez-Gonzdlez, Luis A., ,

Nevong Puna, and Robert G. Moyle
Olson, Storrs L., Horace Loftin, and Steve Goodwin,
Biological, geographical, and cultural origins of the
Loon hunting tradition in Carteret County, North
Carolina, 716-724

Olson, Storrs L., review. The curse of the Labrador Duck.
My obsessive quest to the edge of extinction, by
Glen Chilton, 196-197
Ondatra zibethicus, 62, 613
Opisthocomus hoazin, 105, 109, 111, 115
Oporomis formosus, 503-512, 518-531
Philadelphia, 290, 725-737
tolmiei, 768, 771

opossum, Virginia, see Didelphis virginiana
white-eared, see Didelphis albiventer
Oreoscoptes montanus, 768, 769, 771
Oriole, Baltimore, see Icterus galbula
Bullock’s, see Icterus bullockii
Orchard, see Icterus spurius

Ornelas, Juan Francisco, Nests, eggs, and young of the
Azure-crowned Hummingbird (Amazilia cyanoce-
phala), 592-597

Ortiz, Javier Salgado, see Vargas-Soriano, Jesus, ,

and Griselda Escalona Segura
Ortiz-Maciel, Sonia Gabriela, Consuelo Hori-Ochoa, and
Ernesto Enkerlin-Hoeflich, Maroon-fronted Parrot

(Rhynchop sitta terrisi) breeding home range and
habitat selection in the northern Sierra Madre
Oriental, Mexico, 513-517
Osprey, see Pandion haliaetus
Ovenbird, see Seiurus aurocapilla
Owl, Bam, see Tyto alba
Barred, see Strix varia
Burrowing, see Athene cunicuaria
Great Homed, see Bubo virginianus
Long-eared, see Asio otus
Short-eared, see Asio flammeus
Snowy, see Bubo scandiacus [Nyctea scandiaca]

Tawny, see Strix aluco

P

Pachyptila belcheri, 762
Pachyramphus spp., 797
Pan spp., 183
Pandion haliaetus, 359
Panterpe insignis, 279
Panthera leo, 370

Pappas, Sara, Thomas J. Benson, and James C. Bednarz,
Effects of Brown-headed Cowbird parasitism on
provisioning rates of Swainson’s Warblers, 75-81
Parabuteo unicinctus, 364

Parakeet, White-eyed, see Aratinga leucophthalma
parasites

prevalence of Protocalliphora tundrae in nests of
Phylloscopus borealis in Alaska, 699-705

parasitism

Coccyzus americanus hatched and fed by an Agelaius
phoeniceus, 402^05

conspecific brood parasitism in Spiza americana, 186-
187

effects of Molothrus ater parasitism on provisioning
rates of Limnothlypis swainsonii, 75-81
Molothrus rufoaxillaris parasitism of nests of Cacicus
solitarius and Machetornis rixosa, 795-799
observations of a mixed brood of Parus major and
Poecile varius, 618-620

of an Anas discors nest by A. clypeata in South Dakota,
612-614

reproductive success and nestling growth of Agelaioides
badius parasitized by Molothrus rufoaxillaris
and M. bonariensis, 4 1 7^3 1
songbird nest survival in early-successional restoration
treatments in a large river floodplain, 307-317
Paroaria gularis, 1 15
Parrot, Ground, see Pezoporus wallicus
Maroon-fronted, see Rhynchopsitta terrisi
Thick-billed, see Rhynchopsitta pachyrhyncha
Parula americana, 302, 304, 518-544, 725-737
Pamla, Northern, see Parula americana
Parus major, 126, 326, 436, 604-608, 618-620, 755, 759
Passer domesticus, 62, 135-138, 207-216 (Frontispiece),
604-608

Passerculus sandwichensis, 144, 289, 290, 346-353, 436,
437, 455-464, 635-654, 704, 725-737, 743, 768,
771

Passerella iliaca, 768, 770
Passerina amoena, 768, 771

INDEX TO VOLUME 122

851

caerulea, 518-544
c/m, 532-544

cyanea, 291, 307-317, 518-531, 549, 550, 552, 655-
665, 738-743, 768, 771
Patagona gigas, 496
Patrobus foveocollis, 703, 704

Payne, Robert B., review. All about birds, a short illustrated
history of ornithology, by Val6rie Chansigaud,
814-816

Payne, Robert B., review. Birds of Europe, Second Edition,
by Lars Svensson, 631-632

Payne, Robert B., review. Handbook of the birds of the
world. Volume 14, Bush shrikes to old world
sparrows, edited by J. del Hoyo, A. Elliott, and D.
Christie, 627-628

Peer, Brian D., Conspecific brood parasitism in the
Dickcissel, 186-187

Peer, Brian D., see Atterberry-Jones, Megan R. and

Pelican, American White, see Pelecanus erythrorhynchos
Brown, see Pelecanus occidentalis
Pelecanus erythrorhynchos, 567
occidentalis, 212, 565

Penguin, Erect-crested, see Eudyptes sclateri
Macaroni, see Eudyptes chrysolophus
Magellanic, see Spheniscus magellanicus
Snares, see Eudyptes robustus
Western Rockhopper, see Eudyptes chrysocome

Pereyra, Maria E., see Morton, Martin L. and

Perisoreus canadensis, 773
Perognathus parvus, 62
Peromyscus maniculatus, 61, 62, 65
spp., 64, 456, 615-617, 800, 801

Perry, Margaret L., see Hagelin, Julie C., , Ben S.

Ewen-Campen, Derek S. Sikes, and Susan Shar-
baugh

Petrel, Great-winged, see Pterodroma macroptera
Southern Giant, see Macronectes giganteus
Petrick, John, see Klaus, Nathan A., Scott A. Rush, Tim S.

Keyes, , and Robert J. Cooper

Petrochelidon pyrrhonota, 135-138, 768, 770
Pezoporus wallicus, 516
Phacellodomus ruber, 797
sibilatrix, 797
spp., 421

Phaenicophaeus spp., 173
Phaetusa simplex, 108, 115

Phainoptila, Black-and-yellow, see Phainoptila melanox-
antha

Phainoptila melanoxantha, 279
Phalacrocorax africanus, 565
auritus, 116-125, 565
brasilianus, 105, 108, 114
spp., 380, 720

Phalarope, Red-necked, see Phalaropus lobatus
Wilson’s, see Phalaropus tricolor
Phalaropus lobatus, 768, 771
tricolor, 768, 771

Pham, Lisa H., see Butler, Christopher J., , Jill N.

Stinedurf, Christopher L. Roy, Eric L. Judd,
Nathanael J. Burgess, and Gloria M. Caddell
Pharomachrus mocinno, 608-6 1 1

Pheucticus ludovicianus, 291, 518-531, 725-737
melanocephalus, 625-626
philopatry

natal philopatry of juvenile Charadrius semipalmatus,
23-28

Philomis spp., 794
Phimosus infuscatus, 108, 109, 114
Phoebe, Eastern, see Sayornis phoebe
Phyllomyias virescens, 694
Phylloscopus borealis, 699-705
borealis kennicotti, 699-705
physiology

developmental changes in serum androgen levels of
Megascops asio, 755-761

stress responsiveness decreases with age in precocial
juvenile Alectoris chukar, 762-766
Pica hudsonia, 457, 768, 770

Pichorim, Mauro, Influence of age and season on
morphometric measurements of the Biscutate Swift
(Streptoprocne biscutata), 153-159
Picoides arcticus, 773

borealis, 46, 48, 128, 244-258, 518
pubescens, 128, 290, 518-544, 768, 770
villosus, 290, 302, 518-531

Pierce, Andrew J., see Pobprasert, Korakoch and

Pigeon, Rock, see Columba livia
Pilherodius pileatus, 1 14
Pintail, Northern, see Anas acuta
Pipilo aberti, 399-402, 806-809
albicollis, 399
chlorurus, 768-770
crissalis, 399

erythrophthalmus, 288, 290, 518-544
fuscus, 399—402

Pipit, American, see Anthus rubescens
Sprague’s, see Anthus spragueii
Piranga olivacea, 290, 518-531
rubra, 532-544
Pitangus sulphuratus, 421
Pituophis catenifer, 800, 801
melanoleucus, 457
Platalea [Ajaia] ajaja, 114
leucorodia, 233
Plectrophenax nivalis, 704
Plegadis falcinellus, 96, 108, 114
Plover, Collared, see Charadrius collaris
Grey, see Pluvialis squatarola
Pacific Golden, see Pluvialis fulva
Piping, see Charadrius melodus
Semipalmated, see Charadrius semipalmatus
Snowy/Kentish, see Charadrius alexandrinus alexandri-
nus

Wilson’s, see Charadrius wilsonia
plumage

carotenoid-based male plumage predicts parental invest-
ment in Setophaga ruticilla, 318-325
description of Grallaria ridgelyi nestling, 392-395
feather structure of Cinclidae: water repellency and
resistance to water penetration, 563-568
flight feather molt of Cathartes aura, 354-360

852

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4. December 2010

variation in plumage coloration of Cardinalis cardiiuilb
in urbanizing landscapes. 326-333
Pliivicdis fidva. 27
squatarola. 27

Pobprasert. Korakoch and Andrew J. Pierce, Observations
and predation of a Coral-billed Ground Cuckoo
(Carpococcy.x reiunddi) nest in northeastern Thai-
land. 173-177
Podieep.s aurilus. 374-378
gri.'iegena. 374-378
lugricollis. 116-125. 217-227, 377
Poecile atricapilliis, 290, 302, 608. 662, 768. 770
curalinensis. 128. 518-544
gamhePi. 805
hudsonicit.s, 773
varius, 618-620

Poisbieau. Maud, see Demongin, Laurent, . Georgina

Strange, and Ian J. Strange
PoUoptilu caendea. 518-544, 742
Polyhici .scuteUaris, 421
Pomatomns scdtcitri.x, 720, 721

Pooecetes gramineiis. 144. 289. 290, 346-353, 458, 646,
652. 768. 771
population

seasonal Iluctuatiuon of Amazona aniazonica at a
roosting site in Amazonia, 88-94
status of Caprimulgits carolinensis in the Bahamas, 381-
384

Porphyria flavirostris. I 15
hochstetteri, 1 0
martinica. 108, I 15

Portier. Emmanuelle. see Kesler. Dylan C., Thomas

Ghestemme, , and Anne Gouni

Porzonci Carolina, 62, 65, 768, 77 1

Potvin. Dominique A. and Elizabeth A. MacDougall-
Shackleton. Paternal song complexity predicts
offspring sex ratios close to Hedging, but not
hatching, in Song Sparrows, 146-152

Powell, Emily M., see Stanback. Mark T. and

Powell. Melissa A., see Baldwin, Heather Q., Clinton W.

Jeske. . Paul C. Chadwick, and Wylie C.

Barrow Jr.

Powers. Donald R.. Jessamyn A. Van Hook. Elizabeth A.
Sandlin, and Todd J. McWhorter, Arthropod
foraging by a southeastern Arizona hummingbird
guild. 494-502

prairie dog, black-tailed, see Cynoinys htdovicianus
Pratt. Thane K., review. The ecology and conservation of
Asian Hornbills: farmers of the forest, by Margaret
E. Kinnaird and Timothy G. O'Brien. 412-414
predation

infanticide by Sayornis phoehe. 620-622
infanticide of chicks of Crotophaga ani. 369-374
Ixirns atlanticns depredation by sympatric Lariis dotnin-
icanns. 188-189

nest predators of ground-nesting birds in montane forest of
the Santa Catalina Mountains. Arizona, 614-617
predation of ;i Carpococcy.x renaiddi nest in northeastern
Thailand. 173-177

predator vocalizations affect foraging trade-offs of
Cardinalis cardinalis. 1 68- 1 73

snakes are important nest predators of Spiza americana
in an agricultural landscape, 799-803
Presidents, The Wilson Ornithological Society, 416
Price. Melissa R.. see Hayes, William K., Elwood D.

Bracey, , Valerie Robinette, Eric Gren, and

Caroline Stahala

Prion, Slender-billed, see Pachyplila helcheri
proceedings

ninety-first annual meeting, 820-832
Procyon lotor, 57, 404, 646, 799-803
Protocalliphora tiindrae. 699, 701, 703, 704
P rotonotaria citrea. 532-544
Psendoleistes virescens, 417, 418, 428, 795
Ptarmigan, Evermann's Rock, see Lagopns inuta evennanni
White-tailed, see Lagopns lenenra
Willow, see Lagopns lagopns
Plerodroma macroptera, 762

Puna, Nevong, .see Sanchez-Gonzalez, Luis A., Carl

Oliveros, , and Robert G. Moyle

Pyle. Peter, see Chandler, Robert M., , Maureen E.

Flannery, Douglas J. Long, and Steve N. G. Howell

Q

Quail. Common, see Cotnrni.x coinrni.x
Japanese, see Cotnrni.x Japan ica
Quail-Dove, Chirqui, see Geotrygon chiriqnensis
Quedins brnnnipennis, 703, 704
Quetzal, Resplendent, see Pharomachrns mocinno
Quinn. J. S., A. Samuelsen. M. Barclay, G, Schmaltz, and
H. Kahn. Circumstantial evidence for infanticide of
chicks of the communal Smooth-billed Ani (Cro-
tophaga ani). 369-374
Qniscains major. 532-544
Qni.scalns nie.xicanns, 1 87
qniscnia. 291, 768. 770

R

raccoon, common, see Procyon lotor
racer, eastern yellowbelly, see Coluber constrictor flavi-
ventris

Rail, Clapper, see Rallns longirostris

Grey-necked Wood, see Arainides cajanea
Virginia, see Rallns limicola
Yellow, see Cotnrnicops novehoracensis
Rallina enrizonoides, 786
Rallns limicola. 62, 768, 77 1
long i rostris. 532-544
rat, black [brown), see Rattns rattns
Ratcliffe, Laurene M., see Germain, Ryan R., Matthew W.

Reudink. Peter P. Marra, and

Ratnam. Rama, see Leonard. Wendy J., Jayne Neal, and

rattlesnake, Arizona black, see Crotalns cerherns
prairie, see Crotalns viridis
twin-spotted, see Crotalns price!

Rattns norvegiens, 62

rattns, 370, 560, 607, 679
spp.. 3

Raven, Common [Northern], see Corvns cora.x
Reboreda, Juan C., see De Marsico, Maria C., Bettina
Mahler, and

INDEX TO VOLUME 122

853

Reboreda. Juan C., see Di Giacomo, Alejandro G., Betlina

Mahler, and

Recurvirostm ainericana, 62
Redhead, see Aythya ainericana
Redstart, American, see Setophaga niticilla
Reetz, Matthew J., Jacob M. Musser, and Andrew W.
Kratter, Further evidence of breeding by Shiny
Cowbirds in North America, 365-369
Regnliis calendula, 302, 72,5-737, 741, 768, 771
satrapa, 290, 302, 725-737, 741

Reiley, Bryan M., see Anich, Nicholas M. and

Reithrodontomys niegalotis. 6 1 , 62
spp., 64, 457
reproduction

reproductive success of Agelaioides hadiiis parasitized
by Molothrus rufoa.xil laris and M. bonariensis,
417^31

reproductive success of Athene cimicuaria in urban and
grassland habitats in southern New Mexico, 51-
59

Retogral, Stanley, see Buden, Donald W. and

Reudink, Matthew W., see Germain, Ryan R., , Peter

P. Marra, and Laurene M. Ratcliffe
reviewers

for Volume 122, 833-834
Rhynchopsitta pachyrhyncha, 515
terrisi, 513-517

Rijke, Arie M. and William A. Jesser, The feather structure
of Dippers: water repellency and resistance to water
penetration, 563-568

Ritchison, Brandon T., see Ritchison, Gary and

Ritchison, Gary and Brandon T. Ritchison, Infanticide by
an Eastern Phoebe, 620-622

Ritchison, Gary, see Beckett, Matthew D. and

Robertson, Bob, see Hinnebusch, Daniel M., Jean-Fran9ois

Therrien, Marc-Andre Valiquette, , Sue

Robertson, and Keith L. Bildstein
Robertson, Bruce A., review, Farmland birds across the
world, edited by Wouter van der Weijden, Paul
Terwan, and Adriaan Guldemond, 629-631
Robertson, Sue, see Hinnebusch, Daniel M., Jean-Fran^ois
Therrien, Marc-Andre Valiquette, Bob Robertson,

, and Keith L. Bildstein

Robin, American, see Turdits niigratorius

Robinette, Valerie, see Hayes, William K., Elwood D.

Bracey, Melissa R. Price, , Eric Gren, and

Caroline Stahala

Robinson, Gabrielle L., Courtney J. Conway, Chris
Kirkpatrick, and Dominic D. LaRoche, Response
to nestling throat ligatures by three songbirds, 806-
809

Robsoniiis rabori, 587-591
sorsogonensis, 589

Rodewald, Amanda D., see Graves, Bret M., , and

Scott D. Hull

Rodewald, Amanda D., see Jones, Todd M., , and

Daniel P. Shustack

Rodrigues, Marcos, see Gomes, Henrique Belfort and

Romero, Michael, see Dickens, Molly J. and —
Ronconi. Robert A., see Timoney, Kevin P. and

Rong, Dong, see Xiao-Ping, Yu, Xi Yong-Mei, Lu Bao-
Zhong, Li Xia, Gong Ming-Hao, Shi Liang, and

Root, Brian G., see Burhans, Dirk E., , Terry L.

Shaffer, and Daniel C. Dey

Roy, Christopher L., see Butler, Christopher J„ Lisa H.

Pham, Jill N. Stinedurf, , Eric L. Judd.

Nathanael J. Burgess, and Gloria M. Caddell
Ruggera, Roman and Thomas E. Martin, Breeding biology
and natural history of the Slate-throated Whitestart
in Venezuela, 447^54

Rush. Scott A., see Klaus, Nathan A., , Tim S. Keyes,

John Petrick, and Robert J. Cooper

Ruthrauff, Daniel R., see Johnson, Matthew, , Brian

J. McCaffery, Susan M. Haig, and Jeffrey R.
Walters

Rynchops niger, 1 09, 1 1 5

S

Sage-Grouse, Greater, see Centrocercus urophasianus
Salpinctes obsoletus, 67 1

Samuelsen, A., see Quinn, J. S., , M. Barclay, G.

Schmaltz, and H. Kahn

Sanchez-Gonzalez, Luis A., Carl Oliveros, Nevong Puna,
and Robert G. Moyle, Nests, nest placement, and
eggs of three Philippine endemic birds, 587-591
Sandercock, Brett K., see Kaler, Robb S. A., Steve E.

Ebbert, Clait E. Braun, and

Sandercock, Brett K., see Nol, Erica, Simone Williams, and

Sandlin, Elizabeth A., see Powers, Donald R., Jessamyn A.

Van Hook, , and Todd J. McWhorter

Sandoval, Luis and Ignacio Escalante. Nest description of
the Garden Emerald (Chlorostilbon assimilis) from
Costa Rica. 597-599

Sandpiper, Pectoral, see Calidris melanotos
Semipalmated, see Calidris pusilla
Solitary, see Tringa solilaria
Spotted, see Actitis macularius [macularia]

Stilt, see Calidris himantopus
Upland, see Bartranua longicauda
Western, see Calidris mauri

Sankaran, Ravi, see Manchi. Shirish S. and

Sapsucker, Yellow-bellied, see Sphyrapiciis varius
Sayornis phoebe, 518-531, 618-620, 695. 725-737
Scaup, Greater, see Aythya niarila
Lesser, see Aythya affinis

Schlesinger, Matthew D., see Jablonski. Kevin E.. Stacy A.
McNulty, and

Schmaltz, G., see Quinn. J. S.. A. Samuelsen. M. Barclay,
, and H. Kahn

Schmutz, Joel A., see Lewis, Tyler L., Paul L. Flint. ,

and Dirk V. Derksen

.Schreffler, Lisa, John K. Leiser, and Terry L. Master. Costs
and benefits of foraging alone or in mixed-species
aggregations for Forster's Terns. 95-101

Schwagmeyer, P. L., see Mock, Douglas W. and

Sciurus niger. 801

Scomberomorus niacidatus. 720, 721
Scoter, Black, see Melanitta nigra
Surf, see Melanitta per.spicillata

854

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4, December 2010

White-winged, see Melaniua fusca
Screamer, Homed, see Anhima cornuta
Screech-Owl, Eastern, see Megascops asio
Scrub-Jay, Florida, see Aphelocoma coerulescens

Sealy, Spencer G., see Klippenstine, Dwight R. and

Seewagen, Chad L. and Christopher G. Guglielmo, Effects
of fat and lean body mass on migratory landbird
stopover duration, 82-87

Segura, Griselda Escalona, see Vargas-Soriano, Jesus,

Javier Salgado Ortiz, and

Seiurus aurocapilla, 82-87, 290, 302, 507, 518-531, 545,
552, 655, 725-737
noveboracensis, 725-737
Selasphorus flammula, 279
platycercus, 496, 768, 770
rufus, 496

Sephanoides sephanoides, 496
Serinus canaria, 655, 755, 756, 762
Setophaga rulicilla, 301, 302, 318-325, 518-531, 549-551,
725-737

Seymour, Adrian S., Graham Hatherley, Francisco Javier
Contreras, James Aldred, and Fergus Beeley,
Hatching synchrony, green branch collecting, and
prey use by nesting Harpy Eagles {Harpia harpyja),
792-795

Shaffer, Terry L., see Burhans, Dirk E., Brian G. Root,

, and Daniel C. Dey

Shag, Imperial, see Leucocarbo atriceps

Shan, Fu-Hsien, see Lin, Shang-Yao, Fang-Cian Lu, ,

Shu-Ping Liao, Jing-Ling Weng, Wei-Jen Cheng,
and Chao-Nien Koh

Sharbaugh, Susan, see Hagelin, Julie C., Margaret L. Perry,

Ben S. Ewen-Campen, Derek S. Sikes, and

shark, tiger, see Galeocerdo cuvier
Sheldon, Frederick H., review. Birds of Borneo, by Susan
Myers, 410

Sherry, Thomas W., review, A photographic guide to the
birds of Jamaica, by Ann Haynes-Sutton, Audrey
Downer, Robert Sutton, and Yves-Jacques Rey-
Millet, 817-818

Shoveler, Northern, see Anas clypeata
shrew, northern tree, see Tupaia belangeri
Shrike, Bull-headed, see Lanius bucephalus
Great Grey, see Lanius excubitor
Grey-backed, see Lanius tephronotus
Loggerhead, see Lanius ludovicianus
Long-tailed, see Lanius schach
Red-backed, see Lanius collurio
Southern Grey, see Lanius meridionalis
Woodchat, see Lanius senator
Shriver, W. Gregory, Thomas P. Hodgman, James P. Gibbs,
and Peter D. Vickery, Home range sizes and habitat
use of Nelson’s and Saltmarsh sparrows, 340-345
Shustack, Daniel P., see Jones, Todd M., Amanda D.

Rodewald, and

Sialia mexicana, 132

sialis, 126-134, 137, 289, 532-544
Siegfried, Dennis G., Daniel S. Linville, and David Hille,
Analysis of nest sites of the Resplendent Quetzal
iPharomachrus mocinno): relationship between
nest and snag heights, 608-61 1

Sikes, Derek S., see Hagelin, Julie C., Margaret L. Perry,

Ben S. Ewen-Campen, , and Susan Sharbaugh

Siskin, Pine, see Spinus pinus

Sitta canadensis, 290, 296, 301, 302, 304, 725-737, 805
carolinensis, 290, 302, 518-531
pusilla, 128, 532-544
pygmaea, 132

Skimmer, Black, see Rynchops niger
Skua, Great, see Siercorarius skua

Brown [Subantarctic], see Siercorarius antarcticus
skunk, striped, see Mephitis mephitis
sloth, three-toed, see Bradypus tridactylus
Smallwood, John A., review. The second atlas of breeding
birds in New York State, edited by Kevin J.
McGowan and Kimberley Corwin, 414-415
snake, black rat, see Elaphe obsoleta
fox, see Elaphe vulpina
garter, see Thamnophis spp.
plains ganer, see Thamnophis radix
Texas rat, see Elaphe obsoleta
Snipe, South American, see Gallinago paraguaiae
Wilson’s, see Gallinago delicata
Solenopsis invicta, 370, 470
Sora, see Porzana Carolina
Sorex spp., 64
vagrans, 62

Sparrow, Bachman’s, see Aimophila aestivalis
Baird’s, see Ammodramus bairdii
Brewer’s, see Spizella breweri
Chipping, see Spizella passerina
Clay-colored, see Spizella pallida
Field, see Spizella pusilla
Fox, see Passerella iliaca
Grasshopper, see Ammodramus savannarum
Henslow’s, see Ammodramus henslowii
House, see Passer domesticus
Le Conte’s, see Ammodramus lecontii
Lincoln’s, see Melospiza lincolnii
Mountain White-crowned, see Zonotrichia leucophrys
oriantha

Nelson’s, see Ammodramus nelsoni
Rufous-collared, see Zonotrichia capensis
Saltmarsh, see Ammodramus caudacutus
Savannah, see Passerculus sandwichensis
Seaside, see Ammodramus maritimus
Song, see Melospiza melodia
Swamp, see Melospiza georgiana
Vesper, see Pooecetes gramineus
White-crowned, see Zonotrichia leucophrys
White-eared Ground, see Melozone leucotis
White-throated, see Zonotrichia albicollis ■
Spermophilus franklinii, 800-802
richardsonii, 457
tridecemlineatus, 800, 801
variegatus, 51-59, 616
Spheniscus magellanicus, 39, 762
Sphyrapicus varius, 290
Spinus pinus, 768, 77 1

tristis, 291, 518-531, 768, 770
Spiza americana, 186-187, 307-317, 646-654, 799-803
Spizella breweri, 768, 770

INDEX TO VOLUME 122

855

pallida, 289, 646, 652
passerina, 290, 302, 518-531, 768, 770
pusilla, 290, 307-317, 518-531, 782
Spoonbill, Eurasian, see Platalea leucorodia
Roseate, see Platalea [Ajaia] ajaja
squirrel, Aben’s, see Sciurus abeni

Franklin’s ground, see Spermophilus franklinii
Richardson’s, see Spermophilus richardsonii
rock, see Spermophilus variegatus
southern flying, see Glaucomys volans
thirteen-lined ground, see Spermophilus tridecemlineatus
variable, see Callosciurus finlaysonii
Stahala, Caroline, see Hayes, William K., Elwood D.
Bracey, Melissa R. Price, Valerie Robinette, Eric
Gren, and

Stanback, Mark T. and Emily M. Powell, Predator
vocalizations affect foraging trade-offs of Northern
Cardinals, 168-173

Starling, European, see Stumus vulgaris

Steadman, David W., review. Birds and bats of Palau, by H.

Douglas Pratt and Mandy T. Etpison, 411-412
Stelgidopteryx serripennis, 596
Stercorarius antarcticus, 44, 361
skua, 189
Sterna caspia, 232
eurygnatha, 39

forsteri, 95-101, 232, 768, 770
hirundo, 361
maxima, 39
sandvicensis, 116-125

Stemula antillarum [Sterna antillarium], 96, 116-125
superciliaris, 109, 115

Stilt, Black-necked, see Himantopus mexicanus
Stinedurf, Jill N., see Butler, Christopher J., Lisa H. Pham,

, Christopher L. Roy, Eric L. Judd, Nathanael

J. Burgess, and Gloria M. Caddell
Stork, Maguari, see Euxenura maguari
Marabou, see Leptoptilos crumeniferus
White, see Ciconia ciconia
Wood, see Mycteria americana
Strange, Georgina, see Demongin, Laurent, Maud Pois-

bleau, , and Ian J. Strange

Strange, Ian J., see Demongin, Laurent, Maud Poisbleau,

Georgina Strange, and

Streptoprocne biscutata, 153-159
biscutata seridoensis, 153
zonaris, 158
Strix aluco, 755, 756
varia, 128, 290, 518-544

Strobel, Brad N. and Clint W. Boal, Regional variation in
diets of breeding Red-shouldered Hawks, 68-74
Stumella magna, 436, 437, 646-654

neglecta, 62, 346-353, 455-464, 768, 771
Stumus vulgaris, 62, 71, 136, 187, 242, 608, 672, 759, 768,
770

Suiriri affinis, 695
islerorum, 695
Sula grand, 756
sula, 361-365

Sun, Yue-Hua, see Wang, Jie, Chen-Xi Jia, Song-Hua Tang,
Yun Fang, and

Sunbittern, see Eurypyga helias
Sungrebe, see Heliornis fulica

Suomala, Rebecca W., Sara R. Morris, Kimberly J. Babbitt,
and Thomas D. Lee, Migrant songbird species
distribution and habitat use during stopover on two
islands in the Gulf of Maine, 725-737

survey

avian communities of the Altamaha River estuary in
Georgia, USA, 532-544

survival

hatching and fledging success of Larus dominicanus on
the Brazilian coast, 39-45

influence of woody vegetation on grassland birds within
reclaimed surface mines, 646-654
mountain biking trail use affects reproductive success of
Dendroica chrysoparia, 465-474
nest survival of Myioborus miniatus in Monteverde,
Costa Rica, 29-38

nesting success of Vermivora chrysoptera and V. pinus in
two habitats, 273-278

nest-site choice and nesting success of Turdus rufiventris
in a montane forest in northwestern Argentina,
674-680

of breeding Calidris mauri on the Yukon-Kuskokwim
River Delta, Alaska, 15-22
of juvenile Charadrius semipalmatus, 23-28
of Lagopus muta evermanni in the Aleutian Islands, 1-14
Passer domesticus associated with reduced Petrocheli-
don pyrrhonota nesting success, 135-138
songbird nest survival is invariant to early-successional
restoration treatments in a large river floodplain,
307-317

survival, site fidelity, and population trends of Falco
sparverius wintering in southwestern Florida,
475-483

Sus scrofa, 556, 559, 561

Suzuki, Toshitaka N. and Yuko Tsuchiya, Feeding a foreign
chick: a case of a mixed brood of two tit species,
618-620

Swallow, Barn, see Hirundo rustica
Cliff, see Petrochelidon pyrrhonota
Northern Rough-winged, see Stelgidopteryx serripennis
Tree, see Tachycineta bicolor
Violet-green, see Tachycineta thalassina
Swift, Alpine, see Tachymarptis melba
Biscutate, see Streptoprocne biscutata
Chapman’s, see Chaetura chapmani
Chimney, see Chaetura pelagica
Common, see Apus apus
White-collared, see Streptoprocne zonaris
Swiftlet, Edible-nest, see Aerodramus fuciphagus inexpec-
tatus

Glossy, see Collocalia esculenta affinis
Himalayan, see Aerodramus brevirostris
Indian, see Aerodramus unicolor
Mountain, see Aerodramus hirundinaceus
Pygmy, see Collocalia troglodytes
Uniform, see Aerodramus vanikorensis
White-rumped, see Aerodramus spodiopydius assimilis
Sylvia atricapilla, 236
Sylvilagus nuttalii, 62

856

THE WILSON JOURNAL OL ORNITHOLOGY • Vol. 122, No. 4. December 2010

Synallaxis spp., 42 1
Syrigma sibilatrLx. 108, 114

T

Tcichybaptiis dominicus, 105, 108, 109, 114
Tachycineta bicolor, 289, 574, 581, 622, 741, 768, 771
thalassina, 132
Tuchymarptis nielba, 157
Taeniopygia guttata, 755, 759
Takahe, see Porphyria hochstetteri
Tamias dorsalis, 614—617
striatus, 171, 275

Tanager, Scarlet, see Piranga olivacea

Sooty-capped Bush, see Chlorospingus pileatus
Summer, see Piranga rubra

Tang, Shan, see Cao, Lei, Guixia Zhao, , and

Hanzhao Guo

Tang, Song-Hua, see Wang, Jie, Chen-Xi Jia, , Yun

Fang, and Yue-Hua Sun
Taxidea taxus, 58, 457, 801
tayra, .see Eira barbara
Teal, Blue-winged, see Anas discors
Brazilian, .see Aniazonetta brasiliensis
Cinnamon, see Anas cyanoptera
Green-winged, see Anas crecca

Telles, Marina, see Kanegae, Mieko F., , Severino A.

Lucena, and Jose Carlos Motta-Junior
Tern, Caspian, see Sterna caspia
Cayenne, see Sterna eurygnatha
Common, see Sterna hirundo
Forster’s, see Sterna forsteri
Gull-billed, see Gelochelidon [Sterna] nilotica
Large-billed, see Phaetusa simplex
Least, see Sternula antillarum [Sterna antillariuin]
Royal, see Sterna maxima
Sandwich, see Sterna sandvicensis
Yellow-billed, see Sternula superciliaris
Tetrastes sewerzowi, 177-1 80
Thamnophis radix, 800
spp., 456, 801
Theristicus caudatus, 1 14

Therrien, Jean-Franijois, see Hinnebusch, Daniel M., ,

Marc-Andre Valiquette, Bob Robertson, Sue Ro-
bertson. and Keith L. Bildstein
Thick-knee, Double-striped, see Burhinus bistriatus
Thomomys talpoides, 62, 65
Thornbird. Greater, see Phacellodomus ruber
Little, .see Phacellodomus sihilatrix
Thrasher, Brown, see Toxostoma rufum
Curve-billed, see Toxostoma curvirostre
Sage, see Oreo.scoptes montanus

Western Curve-billed, see Toxostoma cundrostre pai-
nt eh

Thrush, Hermit, .see Catharus guttatus
Rufous-bellied, see Turdtis rufiventris
Swainson's, .see Catharus ustulatus
White-throated, see Turdus assimilis
Wood, see Hylocichla mustelina
Thryothorus ludovicianus, 126-134, 518-544
rufalhus, 503-5 1 2
Tihicen spp., 404

Tiger-heron, Rufescent, see Tigrisoma lineatum
Tigrisoma lineatum. 1 14

Timoney, Kevin P. and Robert A. Ronconi, Annual bird
mortality in the bitumen tailings ponds in north-
eastern Alberta, Canada, 569-576
Tit, Great, see Pams major
Varied, see Poecile varius

Titman, Rodger D., see Frei. Barbara, David M. Bird, and

Titmouse, Tufted, see Baeolophus bicolor
Todiramphus chloris, 48
cinnamominus, 48
gambieri. 48
veneratus. 46-50

Todirostrum chiysocrotaphum. 694
Toucanet, Blue-throated, see Aulacorhynchus prasinus
Towhee, Abert’s, see Pipilo aberti
California, see Pipilo cris.salis
Canyon, .see Pipilo fuscus
Eastern, see Pipilo eiythrophthalmus
Green-tailed, see Pipilo chlorurus
White-throated, see Pipilo albicollis
To.xostoma curvirostre. 625-626
curvirostre palmeri, 625-626
rufum, 290, 5 1 8-544, 782
Trichastoma spp., 589
Tringa flavipes, 115. 569

melanoleuca. I 15, 569, 768, 770
semipalmata, 532-544
solitaria. 1 15
spp., 109, 115

Troglodytes aedon. 35, 290. 419, 420, 425, 608, 622, 679,
768. 770

troglodytes. 290, 302. 304

Tsuchiya, Yuko. see Suzuki, Toshitaka N. and

Tupaia belangerl, 175, 176
Turdinus macrodactyl us. 590
spp., 589, 590
Turdus assimilis. 507
merula, 436

migratorius. 290. 296, 301, 302, 351, 666-673, 738-743,
768, 770

rufiventris. 674-680

Turkey, Domestic, see Meleagris gallopavo
Wild, see Meleagris gallopavo
Turnstone, Ruddy, see Arenaria interpres
Tursiops truncatus, 1 1 7
Tympanuchus spp., 2

Tyrannulet, Golden-faced, see Zimmerius chrysops
Tyrannus forficalus. 695
melancholicus, 236-243
melancholicus despotes, 236, 242
melancholicus melancholicus, 236, 237, 242
melancholicus .xatrapa, 236, 242
tyrannu.s, 290. 532-544. 695, 738-743, 768, 770
verticalis. 695, 768, 771
Tyrant, Cattle, see Machetornis ri.xosa
Pied Water, see Fluvicola pica
White-headed Marsh, see Arundinicola leucocephala
Tyto alba. 19. 60-67. 801

INDEX TO VOLUME 122

857

LI

Urocyon cinereocirgenteits, 614-617

V

Valiquette, Marc-Andre, see Hinnebiisch, Daniel M., Jean-

Fran^'ois Theirien, , Bob Robertson, Sue

Robertson, and Keith L. Bildstein

Van Hook, Jessamyn A., see Powers, Donald R., ,

Elizabeth A. Sandlin, and Todd J. McWhorter
Vargas-Soriano, Jesus, Javier Salgado Ortiz, and Griselda
Escalona Segura, Breeding phenology and nesting
success of the Yucatan Wren in the Yucatan
Peninsula, Mexico, 439^46
Veery, see Catharus fitscescens
Vennivora celata, 768, 771

chrysoptera, 273-278, ,307, 518-,33l
piniis, 273-278, 290
ruftcapilla, 290, 302, 304, 725-737
spp., 451

Vickery, Peter D., see Shriver, W. Gregory, Thomas P.

Hodgman, James P. Gibbs, and

Vidoz, Julian Quillen, Jahn, Alex E., Douglas J. Levey,

Izeni Pires Farias, Ana Maria Mamani, , and

Ben Freeman
Vidua chalybeata, 796

Vielliard, Jacques M. E., see de Moura, Leiliany Negrao,

, and Maria Luisa da Silva

Vilella, Francisco, J., Mark S. Gregory, and Guy A.
Baldassarre, Erratum: 577, Abundance and distri-
bution of waterbirds in the Llanos of Venezuela,
102-115

Vireo, Blue-headed, see Vireo solitarius
Philadelphia, see Vireo philadelphicus
Red-eyed, see Vireo olivaceus
Warbling, see Vireo gilvus
White-eyed, see Vireo griseus
Yellow-throated, see Vireo flavifrons
Vireo flavifrons, 5 1 8-544
gilvus, 768, 771
griseus, 532-544

olivaceus, 290, 296, 301, 302, 518-544, 549-551, 725-
737, 742

philadelphicus,, 507, 725-737
solitarius, 290, 5 1 8-53 1 , 725-737
vocalization

effects of breeding stage and behavioral context on singing
behavior of male Passerina cyanea, 655-665
interspecific song imitation by Dendroica cerulea. 583-
587

nocturnal social cues attract migrating Icleria virens,
780-783

occasional mimicry and night-time singing by To.xo-
stoma cun’irostre pahneri, 625-626
paternal song complexity predicts offspring sex ratios
close to fledging, but not hatching, in Melospiza
melodia, 146-152

variation of Type B song in the endangered Dendroica
chrysoparia, 777-780
Vulpes vulpes, 457, 801
Vulture, Black, see Coragyps atratus
Turkey, see Cathartes aura

W

Walter, W. David, see Davis, Craig A., David M. Leslie Jr.,
, and Allen E. Graber

Walters, Jeffrey R., .see Johnson. Matthew, Daniel R.
Ruthrauff, Brian J, McCaffery, Susan M. Haig, and

Wang, Chen, see Lu, Xin, , and Tonglei Yu

Wang, Jie, Chen-Xi Jia, Song-Hua Tang, Yun Eang, and
Yue-Hua Sun, Nests and breeding of the Giant
Laughingthrush (Garndax maximus) at Lianhua-
shan, southern Gansu, China, 388-391
Warbler, Adelaide’s, see Dendroica adelaidae
Arctic, see Phylloscopus borealis
Bay-breasted, see Dendroica castanea
Black-and-white, see Mniotilta varia
Blackburnian, see Dendroica fusca
Blackpoll, see Dendroica striata
Black-throated Blue, see Dendroica caerulescens
Black-throated Green, see Dendroica virens
Blue-winged, see Vennivora pimis
Canada, see Wilsonia canadensis
Cape May, see Dendroica tigrina
Cerulean, see Dendroica cerulea
Chestnut-sided, see Dendroica pensylvanica
Golden-cheeked, see Dendroica chrysoparia
Golden-crowned, see Basileuterus culicivorus
Golden-winged, see Vennivora chrysoptera
Hooded, see Wilsonia citrina
Kentucky, see Oporornis formosus
MacGillivray’s, see Oporornis tolniiei
Magnolia, see Dendroica magnolia
Mangrove, see Dendroica petechia bryanti
Mourning, see Oporornis Philadelphia
Nashville, see Vennivora ruficapilla
Orange-crowned, see Vennivora celata
Palm, see Dendroica palmarum
Pine, see Dendroica pinus
Prairie, see Dendroica discolor
Prothonotary, see Protonotaria citrea
Red-faced, see Cardellina rubrifrons
Rufous-capped, see Basileuterus rufifrons
Swainson's, see Limnothlypis swainsonii
Three-striped, see Basileuterus tristriatus
Wilson’s, see Wilsonia pusilla
Worm-eating, see Hehnitheros vennivorum
Yellow, see Dendroica petechia
Yellow-rumped, see Dendroica coronata
Yellow-throated, see Dendroica dominica
Ward, Michael P., see Alessi, Mark G., Thomas J. Ben.son.
and

wasp, paper, see Polybia scutellaris
Waterhen, White-breasted, see Amaurornis phoenicurus
Waterthrush, Northern, see Seiurus novehoracensis
Waxwing, Cedar, see Bombycilla cedrorum
weasel, long-tailed, see Mustela frenata
mountain, see Mustela altaica
Weidensaul, Scott, review. Invisible connections: why
migrating shorebirds need the Yellow Sea, by Peter
Battley, Brian McCaffery, Danny Rogers. Jae-Sang
Hong, Nial Moores, Ju Yung-Ki, Jan Lewis, and
Theunis Pier.sma, 816-817

858

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 122, No. 4, December 2010

Weng, Jing-Ling, see Lin, Shang-Yao, Fajig-Cian Lu, Fu-

Hsien Shan, Shu-Ping Liao, , Wei-Jen Cheng,

and Chao-Nien Koh
whale, right, see Eubalaena australis
Whimbrel, see Numenius phaeopus
Whitestart, Brown-capped, see Myioborus brunniceps
Collared, see Myioborus torquatus
Painted, see Myioborus pictus
Slate-throated, see Myioborus miniatus
Spectacled, see Myioborus melanocephalus
Whitney, Bret, review. Field guide to the songbirds of
South America; the passerines, by Robert S.
Ridgely and Guy Tudor, 406-408
Wigeon, American, see Anas americana
Willet, see Tringa semipalmata

Williams, Simone, see Nol, Erica, , and Brett K.

Sandercock

Wilsonia canadensis, 35, 36, 290, 507, 518-531, 549-552,
725-737

citrina, 36, 518-544, 549-552, 583-587
pusilla, 35, 36, 507, 509, 725-737, 768, 771
spp., 29, 36

wolf, gray, see Canis lupus

Wolfenbarger, L. LaReesa, see Klug, Page E., , and

John P. McCarty

Wood, Petra Bohall, see McDermott, Molly E. and

Woodcreeper, Ruddy, see Dendrocincla homochroa
Spot-crowned, see Lepidocolaptes ajfmis
Streak-headed, see Lepidocolaptes souleyetii
Tawny- winged, see Dendrocincla anabatina
Woodpecker, Acorn, see Melanerpes formicivorus
Black-backed, see Picoides arcticus
Downy, see Picoides pubescens
Hairy, see Picoides villosus
Helmeted, see Dryocopus galeatus
Hispaniolan, see Melanerpes striatus
Lineated, see Dryocopus lineatus
Pileated, see Dryocopus pileatus
Red-bellied, see Melanerpes carolinus
Red-cockaded, see Picoides borealis
Red-headed, see Melanerpes erythrocephalus
Syrian, see Dendrocopos syriacus
Wood-Pewee, Eastern, see Contopus virens
woodrat, see Neotoma mexicana
Wren, Cactus, see Campylorhynchus brunneicapillus
Carolina, see Thryothorus ludovicianus
House, see Troglodytes aedon
Marsh, see Cistothorus palu.stris
Rock, see Salpinctes obsoletus
Rufous-and-white, see Thryothorus rufalbus
Rufous-naped, see Campylorhynchus rufmucha

Song, see Cyphorhinus phaeocephalus
Stripe-backed, see Campylorhynchus nuchalis
Winter, see Troglodytes troglodytes
Yucatan, see Campylorhynchus yucatanicus

X

Xanocephalus xanocephalus, 768, 771

Xia, Li, see Xiao-Ping, Yu, Xi Yong-Mei, Lu Bao-Zhong,

, Gong Ming-Hao, Shi Liang, and Dong Rong

Xiao-Ping, Yu, Xi Yong-Mei, Lu Bao-Zhong, Li Xia, Gong
Ming-Hao, Shi Liang, and Dong Rong, Postfled-
ging and natal dispersal of Crested Ibis in the
Qinling Mountains, China, 228-235

Y

Yasukawa, Ken, Yellow-billed Cuckoo hatched and fed by
a Red-winged Blackbird, 402-405
Yaukey, Peter H., Concentrated migratory morning flight at
Lake Pontchartrain, Louisiana, USA, 738-743
Yellowhammer, see Emberiza citronella
Yellowlegs, Greater, see Tringa melanoleuca
Lesser, see Tringa flavipes
Yellowthroat, Common, see Geothlypis trichas

Yong-Mei, Xi, see Xiao-Ping, Yu, , Lu Bao-Zhong,

Li Xia, Gong Ming-Hao, Shi Liang, and Dong
Rong

Yosef, Reuven, see Keynan, Oded and

Yosef, Reuven, Thermoregulatory behavior in migratory
European Bee-eaters {Merops apiaster), 378-380

Youngman, Joseph A., see Binford, Laurence C. and

Yu, Tonglei, see Lu, Xin, Chen Wang, and

Yue-Hua, Sun, see Jie, Wang, Yang Chen, Lu Nan, Fang
Yun, and

Yun, Fang, see Jie, Wang, Yang Chen, Lu Nan, , and

Sun Yue-Hua

Z

Zapus spp., 457, 801

Zenaida macroura, 290, 518-544, 768, 771

Zhao, Guixia, see Cao, Lei, , Shan Tang, and

Hanzhao Guo
Zimmerius acer, 693, 694
chrysops, 689-698
improbus, 693
vilissimus, 693-695

Zonotrichia albicollis, 291, 302, 663, 725-737
capensis, 602
leucophrys, 768, 771
leucophrys nuttalli, 762
leucophrys oriantha, 671

Wilson Journal

of Ornithology

Published by the Wilson Ornithological Society

Volume 122

2010 Quarterly

EDITOR:
EDITORIAL BOARD:

REVIEW EDITOR
INDEX EDITOR
EDITORIAL ASSISTANT

CLAIT E. BRAUN
RICHARD C. BANKS
JACK CLINTON EITNIEAR
SARA J. OYLER-McCANCE
LESLIE A. ROBB
ROBERT B. PAYNE
LESLIE A. ROBB
NANCY J. K, BRAUN

THE WILSON ORNITHOLOGICAL SOCIETY
FOUNDED 3 DECEMBER 1888

Named after ALEXANDER WILSON, the first American ornithologist.

President — E. Dale Kennedy, Biology Department, Albion College, Albion, Ml 49224,
USA; e-mail: dkennedy@albion.edu

First Vice-President— Robert C. Beason, P. O. Box 737, Sandusky, OH 44871, USA;
e-mail: Robert.C.Beason@gmail.com

Second Vice-President — Robert L. Curry, Department of Biology, Villanova University,
800 Lancaster Avenue, Villanova, PA 19085, USA; e-mail: robert.curry@viIlanova.
edu

Editor — Clait E. Braun, 5572 North Ventana Vista Road, Tucson, AZ 85750, USA;
e-mail: TWILSONJO@comcast.net

Secretary — John A. Smallwood, Department of Biology and Molecular Biology,
Montclair State University, Montclair, NJ 07043, USA; e-mail: smallwoodj@
montclair.edu

Treasurer — Melinda M. Clark, 52684 Highland Drive, South Bend IN 46635, USA;
e-mail: MClark@tcservices.biz

Elected Council Members — Jameson F. Chace, Sara R. Morris, and Margaret A. Voss
(terms expire 2011); Mary Bomberger Brown, Mary Garvin, and Mark S. Woodrey
(terms expire 2012); Mark Deutschlander, Paul G. Rodewald and Rebecca J. Safran
( terms expire 2013).

Living Past-Presidents — Pershing B. Hofslund Douglas A. James, Jerome A. Jackson,
Clait E. Braun, Mary H. Clench, Richard C. Banks, Richard N. Conner, Keith L.
Bildstein, Edward H. Burtt Jr., John C. Kricher, William E. Davis Jr., Charles R.
Blem, Doris J. Watt, and James D. Rising.

DATES OF ISSUE OF VOLUME 122 OF

THE WILSON JOURNAL OF ORNITHOLOGY

NO. 1 — 1 March 2010
NO. 2 — 18 May 2010
NO. 3 — 1 September 2010
NO. 4 — 1 December 2010

1

15

23

29

39

46

51

60

68

75

82

88

95

102

116

126

135

139

146

153

160

162

CONTENTS OF VOLUME 122

NUMUKR 1

Major Articles

Demography ol a reintroduced population ol Evermann’s Rock Ptarmigan in the Aleutian Islands
Robb S. A. Kaler, Steve E. Ebbert, Clait E. Braun, and Brett K. Sandercock

Apparent survival ol breeding Western Sandpipers on the Yukon-Kuskokwim River Delta, Alaska
Matthew Johnson, Daniel R. Ruthrauff, Brian J. McCaffery, Susan M. Haig, and JeJJrey R. Walters

Natal philopatry and apparent survival of juvenile Semipalmated Plovers
Erica Nol, Simone Williams, and Brett K. Sandercock

Breeding biology and nesting success of the Slate-throated Whitestart {Myioborus miniatus) in Monteverde,
Costa Rica
Ronald L. Mumme

Breeding biology ol Kelp Gulls on the Brazilian coast
Gisele Fires de Mendonga Dantas and Jodo Stenghel Morgante

Cooperative breeding of the Society Kingfisher {Todiramphus veneratus)

Dylan C. Kesler, Thomas Ghestemme, Emmanuelle Fortier, and Anne Gouni

Reproductive success of Burrowing Owls in urban and grassland habitats in southern New Mexico
Daniele Berardelli, Martha J. Desmond, and Leigh Murray

Dietary trends ol Barn Owls in an agricultural ecosystem in northern Utah
Carl D. Marti

Regional variation in diets of breeding Red-shouldered Hawks
Brad N. Strobel and Clint W. Boal

Effects of Brown-headed Cowbird parasitism on provisioning rates of Swainson’s Warblers
Sara Fappas, Thomas J. Benson, and James C. Bednarz

Effects of fat and lean body mass on migratory landbird stopover duration
Chad L. Seewagen and Christopher G. Guglielmo

Seasonal fluctuation of the Orange-winged Amazon at a roosting site in Amazonia
Leiliany Negrdo de Moura, Jacques M. E. Vielliard, and Maria Luisa da Silva

Costs and benefits of foraging alone or in mixed-species aggregations for Forster’s Terns
Lisa SchreJJler, John K. Leiser, and Terry L. Master

Abundance and distribution ofwaterbirds in the Llanos of Venezuela
Erancisco J. Vilella and Guy A. Baldassarre

Phenology of six migratory coastal birds in relation to climate change
Charles R. Eoster, Anthony E Amos, and Lee A. Euiman

Nest-site limitation of secondary cavity-nesting birds in even-age southern pine forests
Karl E. Miller

House Sparrows associated with reduced Cliff Swallow nesting success
Douglas R. Leasure, Ragupathy Kannan, and Douglas A. James

Home-range size and site tenacity of overwintering Le Conte’s Sparrows in a fire managed prairie
Heather Q. Baldwin, Clinton W. Jeske, Melissa A. Fowell, Faul C. Chadwick, atid Wylie C. Barrow Jr.

Paternal song complexity predicts oflspring sex ratios close to fledging, but not hatching, in Song Sparrows
Dominique A. Fotvin and Elizabeth A. MacDougall-Shackleton

Influence ol age and season on morphometric measurements of the Biscutate Swift {Streptoprocne biscutata)
Mauro Fichorim

Short Communications

Cooperative breeding by Red-headed Woodpeckers
Megan R. Atterberry-Jones and Brian D. Feer

Observations on the breeding biology of the Collared Crescentchest {Melanopareia torquata)

Mieko E Kanegae, Marina Tettes, Severino A. Lucena, and Jose Carlos Motta-Junior

165

168

173

177

181

183

186

188

190

194

196

207

217

228

236

244

259

273

279

288

296

307

Effects of a flood on foraging ecology and population dynamics of Swainson’s Warblers
Nicholas M. Anich and Bryan M. Reiley

Predator vocalizations affect foraging trade-offs of Northern Cardinals
Mark T Stanback and Emily M. Powell

Observations and predation of a Coral-billed Ground Cuckoo {Carpococcyx renauldi) nest in northeastern
Thailand

Korakoch Pobprasert and Andrew J. Pierce

Diet of Chinese Grouse {Tetrastes sewerzowi) during preincubation
Wangjie, Yang Chen, Lu Nan, Fang Yun, and Sun Yue-Hua

Observations of a possible foraging tool used by Common Ravens
Paul F Frame

Hanging behavior of the Hooded Crow [Corvus cornix)

Mario Melletti and Marzia Mirabile

Conspecific brood parasitism by the Dickcissel
Brian D. Peer

First report of Olrog’s Gull depredation by sympatric Kelp Gulls
Luciano La Sala and Sergio Martorelli

Second and third records of Snares Penguins {Eudyptes robustus) in the Falkland Islands
Laurent Demongin, Maud Poisbleau, Georgina Strange, and Ian J. Strange

A new bird species for Costa Rica; Sapphire-throated Hummingbird {Lepidopyga coeruleogularis)

Esteban Biamonte

Ornithological Literature
Robert B. Payne, Review Editor

NUMBER 2

Major Articles

Not the Nice sparrow (The 2007 Margaret Morse Nice Lecture)

Douglas W. Mock and P L. Schwagmeyer

The postbreeding migration of Eared Grebes
Joseph R. Jehl Jr. and Annette E. Henry

Postfledging and natal dispersal of Crested Ibis in the Qinling Mountains, China
Yu Xiao-Ping, Xi Yong-Mei, Lu Bao-Zhong, Li Xia, Gong Ming-Hao, Shi Liang, and Dong Rong

Morphological and genetic variation between migratory and non-migratory Tropical Kingbirds during
spring migration in central South America

Alex E. Jahn, Douglas J. Levey, Izeni Pires Farias, Ana Maria Mamani, Julian QuilUn Vidoz, and
Ben Freeman

Red-cockaded Woodpecker male/female foraging differences in young forest stands
Kathleen E. Franzreb

Foraging habits and habitat use by Edible-nest and Glossy swiftlets in the Andaman Islands, India
Shirish S. Manchi and Ravi Sankaran

Golden- and Blue-winged warblers: distribution, nesting success, and genetic differences in two habitats
John L. Confer, Kevin W. Barnes, and Erin C. Alvey

Genetic and morphological variation of the Sooty-capped Bush Tanager (Chlorospingus pileatus), a highland
endemic species from Costa Rica and western Panama

Tania Chavarria-Pizarro, Gustavo Gutiirrez-Espeleta, Eric J. Fuchs, and Gilbert Barrantes

Long-term changes in avian community structure in a successional, forested, and managed plot in a

reforesting landscape

Elizabeth W. Brooks and David N. Bonter

Scale-dependent response by breeding songbirds to residential development along Lake Superior
Michelle T. Ford ana David J. Flaspohler

Songbird nest survival is invariant to early-successional restoration treatments in a large river floodplain
Dirk E. Bur bans, Brian G. Root, Terry L. Shaffer, and Daniel C. Dey

318 Carorcnoid-bascd male plumage predicts parental investment in the American Redstart
Ryan R. Germain, Matthew W. Reudink, Peter P. Marra, and Laurene M. Ratcliffe

326 Variation in plumage coloration ol Northern Cardinals in urbanizing landscapes
Todd M. Jones, Amanda D. Rodewald, and Daniel P Shustack

334 Annual precipitation affects reproduction of the Southern Grey Shrike {Lanius meridionalis)

Oded Keynan and Reuven Yosef

340 Home range sizes and habitat use of Nelson’s and Saltmarsh sparrows

W Gregory Shriver, Thomas P. Hodgman, James P. Gibbs, and Peter D. Vickery

346 Assessing generalized egg mimicry: a quantitative comparison of eggs of Brown-headed Cowbirds and
grassland passerines

Dwight R. Klippenstine and Spencer G. Sealy
354 Flight feather molt of Turkey Vultures

Robert M. Chandler, Peter Pyle, Maureen E. Flannery, Douglas J. Long, and Steve N. G. Howell
Short Communications

361 The first reported case of cooperative polyandry in the Red-footed Booby: trio relationships and benefits
Lei Gao, Guixia Zhao, Shan Tang, and Hanzhao Guo

365 Further evidence of breeding by Shiny Cowbirds in North America
Matthew J. Reetz, Jacob M. Musser, and Andrew W. Kratter

369 Circumstantial evidence for infanticide of chicks of the communal Smooth-billed Ani {Crotophaga ani)

J. S. Quinn, A. Samuelsen, M. Barclay, G. Schmaltz, and H. Kahn

374 Flight speeds of migrating Red-necked and Horned grebes
Laurence C. Binforaand Joseph A. Youngman

378 Thermoregulatory behavior in migratory European Bee-eaters {Merops apiaster)

Reuven Yosef

381 Population status of Chuck-will’s-widow {Caprimulgus carolinensis) in the Bahamas

V/illiam K Hayes, Elwood D. Bracey, Melissa R. IVice, Valerie Robinette, Eric Gren, and Caroline Stahala

385 Yellow Rails wintering in Oklahoma

Christopher J. Butler, Lisa H. Pham, Jill N Stinedurf, Christopher L. Roy, Eric L. Judd, Nathanael J. Burgess, and
Gloria M. Caddell

388 Breeding of the Giant Laughingthrush {Garrulax maximus) at Lianhuashan, southern Gansu, China
Jie Wang, Chen-Xi Jia, Song-Hua Tang, Yun Fang, and Yue-Hua Sun

392 First desermtion of the nest of Jocotoco Antpitta {Grallaria ridgelyi)

Harold E Greeney and Mery E. Juiha J

395 Nesting ecology of the Grey-backed Shrike [Lanius tephronotus) in South Tibet
Xin Lu, Chen Wang, and Tonglei Yu

399 The first reported hybridization of Abert’s and Ganyon towhees [Pipilo spp.)

R. Roy Johnson and Steven L. Hopp

402 Yellow-billed Cuckoo hatched and fed by a Red-winged Blackbird
Ken Yasukawa

406 Ornithological Literature

Robert B. Payne, Book Review Editor

NUMBER 3

Major Articles

417 Reproductive success and nestling growth of the Baywing parasitized by Screaming and Shiny cowbirds
Marla C. De Mdrsico, Bettina Mahler, and Juan C. Reboreda

432 Bobolink egg mass variability and nestling growth patterns
Barbara Frei, David M. Bird, and Rodger D. Titman

439 Breeding phenology and nesting success of the Yucatan Wren in the Yucatan Peninsula, Mexico
Jesus Vargas-Soriano, Javier Salgado Ortiz, and Griselda Escalona Segura

447

455

465

475

484

494

503

513

518

532

545

556

563

569

577

578

583

587

592

597

600

604

608

Breeding biology and natural history of the Slate-throated Whitestart in Venezuela
Roman A. Ruggera and Thomas E. Martin

Reproductive biology of a grassland songbird community in northcentral Montana
Stephanie L. Jones, ]. Scott Dieni, and Paula J. Gouse

Mountain biking trail use affects reproductive success of nesting Golden-cheeked Warblers
Craig A. Davis, David M. Leslie Jr., W. David Walter, and Allen E. Graber

Survival, site fidelity, and population trends of American Kestrels wintering in southwestern Florida
Daniel M. Hinnebusch, Jean-Erangois Therrien, Marc-Andre Valiquette, Bob Robertson, Sue Robertson, and
Keith L. Bildstein

Temporal and spatial shifts in habitat use by Black Brant immediately following flightless molt
Tyler L. Lewis, Paul L. Elint, Joel A. Schmutz, and Dirk V Derksen

Arthropod foraging by a southeastern Arizona hummingbird guild

Donald R. Powers, Jessamyn A. Van Hook, Elizabeth A. Sandlin, and Todd J. McWhorter

Army ant raid attendance and bivouac-checking behavior by neotropical montane forest birds
Sean ODonnell, Anjali Kumar, and Gorina Logan

Maroon-fronted Parrot {Rhynchopsitta terrisi) breeding home range and habitat selection in the northern
Sierra Madre Oriental, Mexico

Sonia Gabriela Ortiz-Maciel, Consuelo Hori-Ochoa, and Ernesto Enkerlin-Hoejlich

Short-term effects of fire on breeding birds in southern Appalachian upland forests
Nathan A. Klaus, Scott A. Rush, Tim S. Keyes, John Petrick, and Robert J. Cooper

Avian communities of the Altamaha River estuary in Georgia, USA
Ross A. Brittain, Vicky J. Meretsky, and Chris B. Craft

Influence of cover and food resource variation on post-breeding bird use of timber harvests with residual
canopy trees

Molly E. McDermott and Petra Bo hall Wood

Distribution, abundance, and status of Guban Sandhill Granes [Crus canadensis nesiotes)

Xiomara Galvez Aguilera and Eelipe Chavez-Ramirez

The feather structure of Dippers: water repellency and resistance to water penetration
Arie M. Rijke and William A. Jesser

Annual bird mortality in the bitumen tailings ponds in northeastern Alberta, Ganada
Kevin P Timoney and Robert A. Ronconi

Abundance and distribution of waterbirds in the Llanos ot Venezuela
Erancisco J. Vilella, Mark S. Gregory, and Guy A. Baldassarre

Short Communications

Piping Plover foraging distribution and prey abundance in the pre-laying period
Jonathan B. Cohen and James D. Eraser

Interspecific song imitation by a Cerulean Warbler
Than J. Boves, David A. Buehler, and Phillip C. Massey

Nests, nest placement, and eggs of three Philippine endemic birds

Luis A. Sdnchez-Gonzdlez, Carl Oliveros, Nevong Puna, and Robert G. Moyle

Nests, eggs, and young of the Azure-crowned Hummingbird {Amazilia cyanocephala)

Juan Erancisco Ornelas

Nest description of the Garden Emerald [Chlorostilbon assimilis) from Costa Rica
Luis Sandoval and Ignacio Escalante

The nest of the Cipo Canastero {Asthenes luizae), an endemic hirnariid Irom the Espinha90 Range,
southeastern Brazil

Henrique Belfort Gomes and Marcos Rodrigues

Nest box use by Great Tits in semi-arid rural residential gardens
Motti Charter, Yossi Leshem, Shay Halevi, and Ido Izhaki

Analysis of nest sites of the Resplendent Quetzal {Pharomachrus rnocinno): relationship between nest and snag
heights

Dennis G. Siegfried, Daniel S. Linville, and David Hille

612

614

618

620

623

625

627

635

646

655

666

674

681

689

699

706

716

725

738

744

755

762

Parasitism ol a Blue-winged leal nest by a Northern Sboveler in South Dakota.

Thomas E. Lewis and Pamela R. Garrettson

Nest predators ol ground-nesting birds in montane forest ol the Santa Catalina Mountains, Arizona
C.lms Kirkpatrick a>id Courtney J. Conway

Feeding a foreign chiek: a case ol a mixed brood of two Tit species
Toshitaka N. Suzuki and Yuko Tsuchiya

Infanticide by an Eastern Phoebe
Gary Ritchison and Brandon T. Ritchison

Marsh Wren eats small fish

Andrea J. Ayers and James W. Arrnacost Jr.

Occasional mimicry and night-time singing by the western Curve-billed Thrasher {Toxostoma curvirostre
palmeri)

R. Roy Johnson and Lois T. Haight

Ornithological Literature
Robert B. Payne, Book Review Editor

Major Articles

NUMBER 4

Breeding patterns of Henslow’s Sparrow and sympatric grassland sparrow species
L. Lynnette Dornak

Influence of woody vegetation on grassland birds within reclaimed surface mines
Bret M. Graves, Amanda D. Rodewald, and Scott D. Hull

Effects of breeding stage and behavioral context on singing behavior of male Indigo Buntings
Matthew D. Beckett and Gary Ritchison

Development of incubation temperature and behavior in thrushes nesting at high altitude
Martin L. Morton and Maria E. Pereyra

Flexibility in nest-site choice and nesting success of Turdus rujiventris (Turdidae) in a montane forest in
northwestern Argentina

Silvia B. Lomdscolo, A. Carolina Monmany Agustina Malizia, and Thomas E. Martin

Breeding biology of the Taiwan Barbet {Megalaima nuchalis) in Taipei Botanical Garden

Shang-Yao Lin, Fang-Cian Lu, Fu-Hsien Shan, Shu-Ping Liao, Jing-LingWeng, Wei-Jen Cheng, and Chao-Nien

Koh

Breeding biology of the Golden-faced Tyrannulet [Zimmerius chrysops) in Venezuela
William Goulding and Thomas E. Martin

Habitat type is related to nest mass and fledging success of Arctic Warblers

Julie C. Hagelin, Margaret L. Perry, Ben S. Ewen-Campen, Derek S. Sikes, and Susan Sharbaugh

Home range, habitat use, and nest site characteristics of Mississippi Kites in the White River National

Wildlife Refuge, Arkansas

Troy J. Bader and James C. Bednarz

Biological, geographical, and cultural origins ol the loon hunting tradition in Carteret County, North
Carolina

Storrs L. Olson, Horace Loftin, and Steve Goodwin

Migrant songbird species distribution and habitat use during stopover on two islands in the Gulf of Maine
Rebecca W. Suomala, Sara R. Morris, Kimberly J. Babbitt, and Thomas D. Lee

Concentrated migratory morning Hight at Lake Pontchartrain, Louisiana, USA
Peter H. Yaukey

Night migrant fatalities and obstruction lighting at wind turbines in North America
Paul Kerlinger, Joelle L. Gehring, Wallace P. Erickson, Richard Curry, Aaftab Jain, and John Guarnaccia

Developmental changes in serum androgen levels of Eastern Screech-Owls {Megascops asio)

Corinne P. Kozlowski and D. Caldwell Hahn

Stress responsiveness decreases with age in precocial, juvenile Chukar
Molly J. IDickens and L. Michael Romero

767

111

111

780

784

788

792

795

799

803

806

809

810

812

820

833

834

835

Short Communications

Composition dynamics of an avian assemblage
Fritz L. Knopf

A digital spot-mapping method for avian field studies

Kevin E. JabLonski, Stacy A. McNulty, and Matthew D. Schlesinger

Variation of Type B song in the endangered Golden-cheeked Warbler {Dendroica chrysoparia)

Wendy J. Leonard, Jayne Neal, and Rama Ratnam

Nocturnal social cues attract migrating Yellow-breasted Chats
Mark G. Alessi, Thomas J. Benson, and Michael P. Ward

Range expansion of the White-breasted Waterhen {Amaurornis phoenicurus) into Micronesia
Donald W Buden and Stanley Retogral

First reproductive record of Wilson’s Plover in Baia de Todos os Santos, northeastern Brazil
Vitor O. Lunardi and Regina FI. Macedo

Hatching synchrony, green branch collecting, and prey use by nesting Harpy Eagles {Harpia harpyjd)
Adrian S. Seymour, Graham Hatherley, Francisco Javier Contreras, James Aldred,
and Fergus Beeley

Screaming Cowbird parasitism of nests of Solitary Caciques and Cattle Tyrants
Alejandro G. Di Giacomo, Bettina Mahler, and Juan C. Reboreda

Snakes are important nest predators of Dickcissels in an agricultural landscape
Page E. Klug, L. LaReesa Wolfenbarger, and John P. McCarty

Interspecific cavity-sharing between a Helmeted Woodpecker {Dryocopus galeatus) and two White-eyed
Parakeets {Aratinga leucophthalma)

Kristina L. Cockle

Response to nesding throat ligatures by three songbirds

Gabrielle L. Robinson, Courtney J. Conway, Chris Kirkpatrick, and Dominic D. LaRoche

Editors of The Wilson Bulletin {iSS8-xoo^) and The Wilson Journal of Ornithology
(2006-Z010)

William and Nancy Klamm Service Award for zoio

Ornithological Literature
Robert B. Payne, Book Review Editor

Proceedings of the Ninety-First Annual Meeting

Reviewers for Volume izz

Editor’s Comments

Index to Volume izz

Contents of Volume izz

-A

* \

. 1

fci<g:3P>'V» *flf

*■— - Iff— >'<f lHfj

'W*

. ^ ^ ♦■■<5|[:>

£:

9

THE WILSON JOURNAL OL ORNITHOLOGY

Editor CLAIT E. BRAUN

5572 North Ventana Vista Road
Tucson, AZ 85750-7204
E-mail: TWILSONJO(S)comcast.net

Editorial NANCY J. K. BRAUN
Assistant

Editorial Board RICHARD C. BANKS

JACK CLINTON EITNIEAR
SARA J. OYLER-McCANCE
LESLIE A. ROBB

Review Editor ROBERT B. PAYNE

1306 Granger Avenue
Ann Arbor. MI 48104, USA
E-mail: rbpayne(S)umich.edu

GUIDELINES LOR AUTHORS

Please consult the detailed “Guidelines for Authors” found on the Wilson Ornithological Society web site
(http://www.wilsonsociety.org). All manuscript submissions and revisions should be sent to Clait E. Braun,
Editor, The Wilson Journal of Ornithology, 5572 North Ventana Vista Road, Tucson, AZ 85750-7204, USA.
The Wilson Journal of Ornithology office and fax telephone number is (520) 529-0365. The e-mail address is
TWilsonJO@comcast.net

NOTICE OF CHANGE OF ADDRESS

Notify the Society immediately if your address changes. Send your complete new address to Ornithological
Societies of North America, 5400 Bosque Boulevard, Suite 680, Waco, TX 76710.

The permanent mailing address of the Wilson Ornithological Society is: %The Museum of Zoology, The
University of Michigan. Ann Arbor, MI 48109, USA. Persons having business with any of the officers may
address them at their various addres.ses given on the inside of the front cover, and all matters pertaining to the journal
should be sent directly to the Editor.

MEMBERSHIP INQUIRIES

Membership inquiries should be sent to Timothy J. O'Connell. Department of Natural Resources Ecology and
Management, Oklahoma State University, 240 AG Hall, Stillwater, OK 74078; e-mail: oconnet@ okstate.edu

THE JOSSELYN VAN TYNE MEMORIAL LIBRARY

The Josselyn Van Tyne Memorial Library of the Wilson Ornithological Society, housed in the University of
Michigan Mu.seum of Zoology, was established in concurrence with the University of Michigan in 1930. Until
1947 the Library was maintained entirely by gifts and bequests of books, reprints, and ornithological magazines
from members and friends of the Society. Two members have generously established a fund for the purchase
of new books; members and friends are invited to maintain the fund by regular contribution. The fund is administered
by the Library Committee. Jerome A. Jackson, Florida Gulf Coast Univeristy, is Chairman of the Committee. The
Library currently receives over 200 periodicals as gifts and in exchange for The Wilson Journal of Ornithology. For
infonnation on the Library and our holdings, see the Society’s weh page at http://www.wilsonsociety.org. With the
usual exception of rare books, any item in the Library may be boiTowed by members of the Society and will be .sent
prepaid (by the University of Michigan) to any address in the United States, its possessions, or Canada. Return
postage is paid by the borrower. Inquiries and requests by borrowers, as well as gifts of books, pamphlets, reprints,
and magazines, should be addressed to: Josselyn Van Tyne Memorial Library. Museum of Zoology, The University
of Michigan. I 109 Geddes Avenue, Ann Arhor, MI 48109-1079. USA. Contributions to the New Book Fund should
be sent to the Treasurer.

This issue of The Wilson Journal of Ornithology was published on I December 2010.

767

111

111

780

784

788

792

795

799

803

806

809

810

812

820

833

834

835

Continued from outside back cover

Short Communications

Composition dynamics of an avian assemblage
Fritz L. Knopf

A digital spot-mapping method for avian field studies

Kevin E. Jablonski, Stacy A. McNulty, and Matthew D. Schlesinger

Variation of Type B song in the endangered Golden-cheeked Warbler {Dendroica chrysoparia)

Wendy J. Leonard, Jayne Neal, and Rama Ratnam

Nocturnal social cues attract migrating Yellow-breasted Chats
Mark G. Alessi, Thomas J. Benson, and Michael P. Ward

Range expansion of the White-breasted Waterhen [Amaurornis phoenicurus) into Micronesia
Donald W Buden and Stanley Retogral

First reproductive record of Wilson’s Plover in Bala deTodos os Santos, northeastern Brazil
Vitor O. Lunardi and Regina FI. Macedo

Hatching synchrony, green branch collecting, and prey use by nesting Harpy Eagles {Harpia harpyja)
Adrian S. Seymour, Graham Hatherley, Francisco Javier Contreras, James Aldred,
and Fergus Beeley

Screaming Cowbird parasitism of nests of Solitary Caciques and Cattle Tyrants
Alejandro G. Di Giacomo, Bettina Mahler, and Juan C. Reboreda

Snakes are important nest predators of Dickcissels in an agricultural landscape
Page E. Klug, L. LaReesa Wolfenbarger, and John P. McCarty

Interspecific cavity-sharing between a Helmeted Woodpecker {Dryocopus galeatus) and two White-
eyed Parakeets {Aratinga leucophthalma)

Kristina L. Cockle

Response to nestling throat ligatures by three songbirds

Gabrielle L. Robinson, Courtney J. Conway, Chris Kirkpatrick, and Dominic D. LaRoche

Editors of The Wilson Bulletin (iSSS-zoo^) and The Wilson Journal of Ornithology
(2006-2010)

William and Nancy Klamm Service Award for 2010

Ornithological Literature
Robert B. Payne, Book Review Editor

Proceedings of the Ninety-First Annual Meeting

Reviewers for Volume 122

Editor’s Comments

Index to Volume 122

Contents of Volume 122

The Wilson Journal of Ornithology

(formerly The Wilson Bulletin)

Volume 122, Number 4 CONTENTS December 20 1C

Major Articles

635 Breeding patterns ol Henslows Sparrow and sympatric grassland sparrow species
L. Lynnette 'Dornak

646 Influence of woody vegetation on grassland birds within reclaimed surface mines
Bret M. Graves, Amanda D. Rodewald, and Scott D. Bull

655 Effects of breeding stage and behavioral context on singing behavior of male Indigo Buntings
Matthew D. Beckett and Gary Ritchison

666 Development of incubation temperature and behavior in thrushes nesting at high altitude
Martin L. Morton and Maria E. Pereyra

674 Flexibility in nest-site choice and nesting success of Turdus rufiventris (Turdidae) in a montane forest in
northwestern Argentina

Silvia B. Lomdscolo, A. Garolina Monmany, Agustina Malizia, and Thomas E. Martin

68 1 Breeding biology of the Taiwan Barbet {Megalaima nuchalis) in Taipei Botanical Garden

Shang-Yao Lin, Eang-Cian Lu, Eu-Hsien Shan, Shu-Ping Liao, Jing-Ling Weng, Wei-Jen Cheng, and Chao-
Nien Koh

689 Breeding biology of the Golden-faced Tyrannulet {Zimmerius chrysops) in Venezuela
William Goulding and Thomas E. Martin

699 Habitat type is related to nest mass and fledging success of Arctic Warblers

Julie C. Hagelin, Margaret L. Perry, Ben S. Ewen-Campen, Derek S. Sikes, and Susan Sharbaugh

706 Home range, habitat use, and nest site characteristics of Mississippi Kites in the White River National
Wildlife Refuge, Arkansas
Troy J. Bader and James C. Bednarz

716 Biological, geographical, and cultural origins of the loon hunting tradition in Garteret County, North
Carolina

Storrs L. Olson, Horace Lojiin, and Steve Goodwin

725 Migrant songbird species distribution and habitat use during stopover on two islands in the Gulf of
Maine

Rebecca W. Suomala, Sara R. Morris, Kimberly J. Babbitt, and Thomas D. Lee

738 Concentrated migratory morning flight at Lake Pontchartrain, Louisiana, USA
Peter H. Yaukey

744 Night migrant fatalities and obstruction lighting at wind turbines in North America

Paul Kerlinger, Joelle L. Gehring, Wallace P Erickson, Richard Curry, Aajiab Jain, and John Guarnaccia

755 Developmental changes in serum androgen levels of Eastern Screech-Owls {Megascops asio)

Gorinne P. Kozlowski and D. Caldwell Hahn

762 Stress responsiveness decreases with age in precocial, juvenile Chtikar
MollJ J. Dickens and L. Michael Romero

Continued on inside hack cover

•

[.

I * f ■■

t

* .. «

*

" I

%

I

^-4*-

4

P

■

-

I

MCZ ERNST MAYR LIBRARY

lllllll

III

lllllll

III

III

3 2:

)44 118 689

4

05