Full text of "The Wilson bulletin"

See other formats

HARVARD UNIVERSITY

Library of the
Museum of
Comparative Zoology

IhcWilsonBulkUn

PUBLISHED BY THE WILSON ORNITHOLOGICAL SOCIETY

VOLUME 93

1981

QUARTERLY

EDITOR: JON C. BARLOW
REVIEW EDITOR: ROBERT RAIKOW
COLOR PLATE EDITOR: WILLIAM A. LUNK
ASSISTANT EDITOR: MARGARET L. MAY
SENIOR EDITORIAL ASSISTANTS: GARY BORTOLOTTI

NANCY FLOOD

EDITORIAL ASSISTANTS: C. DAVISON ANKNEY
KEITH L. BILDSTEIN
JAMES D. RISING
RICHARD R. SNELL

President — Abbot S. Gaunt, Department of Zoology, Ohio State University, Columbus,
Ohio 43210.

First Vice-President — Jerome A. Jackson, Department of Biological Sciences, P.O. Drawer
Z, Mississippi State University, Mississippi State, Mississippi 39762.

Second Vice-President — Clait E. Braun, Wildlife Research Center, 317 West Prospect St.,
Fort Collins, Colorado 80526.

Editor — Jon C. Barlow, Department of Ornithology, Royal Ontario Museum, 100 Queen’s
Park, Toronto, Ontario, Canada MSS 2C6.

Secretary — Curtis S. Adkisson, Department of Biology, Virginia Polytechnic Institute and
State University, Blacksburg, Virginia 24061.

Treasurer — Robert D. Burns, Department of Biology, Kenyon College, Gambler, Ohio 43022.

Elected Council Members — Richard C. Banks (term expires 1982); Mary H. Clench (term
expires 1983); Helen Lapham (term expires 1984).

DATES OF ISSUE OF VOLUME 93
OF THE WILSON BULLETIN

NO. 1—14 May 1981
NO. 2 — 20 August 1981
NO. 3 — 29 October 1981
NO. 4 — 22 February 1982

CONTENTS OF VOLUME 93

NUMBER 1

OBSERVATIONS OF SEABIRDS DURING A CRUISE FROM ROSS ISLAND TO ANVERS ISLAND, ANT-
ARCTICA Robert M. Zink 1

DISPLAY BEHAVIOR OF OVENBIRDS {Seiurus aurocapillus) II. SONG VARIATION AND SINGING

BEHAVIOR M. Ross Lein 21

THE MAYFIELD METHOD OF ESTIMATING NESTING SUCCESS: A MODEL, ESTIMATORS AND SIMU-
LATION RESULTS Gary L. Hensler and James D. Nichols 42

NARROWLY DISJUNCT ALLOPATRY BETWEEN BLACK-CAPPED AND CAROLINA CHICKADEES IN

NORTHERN INDIANA Peter G. Merritt 54

TOE FUSION IN OSCINES George A. Clark, Jr. 67

ENVIRONMENTAL EFFECTS ON ROOSTING BEHAVIOR OF CHIMNEY SWIFTS

Richard M. Zammuto and Edwin C. Franks 77

GENERAL NOTES

OBSERVATION OF A BROOD OF SHARP-SHINNED HAWKS IN ONTARIO, WITH COMMENTS
ON THE FUNCTIONS OF SEXUAL DIMORPHISM

Helmut C. Mueller, Nancy S. Mueller and Patricia G. Parker 85

FOOD DEPRIVATION AND TEMPERATURE REGULATION IN NESTLING FERRUGINOUS HAWKS

Diana F. Tomback and Joseph R. Murphy 92

AERIAL “play” OF BLACK VULTURES Walter A. Thurber 97

THE SHOULDER-SPOT DISPLAY IN RUFFED GROUSE Allan Garbutt 98

THE AGONISTIC REPERTOIRE OF SANDHILL CRANES

Stephen A. Nesbitt and George W. Archibald 99

NOTES ON THE SLENDER ANTBIRD Edwin O. Willis and Yoshika Oniki 103

NOTES ON THE UNIFORM CRAKE IN COSTA RICA F. G. Stiles 107

TRICHOMONIASIS IN BALD EAGLES Ward B. Stone and Peter E. Nye 109

Protocalliphora INFESTATION IN BROAD-WINGED HAWKS

Scott Crocoll and James W. Parker 110

HERRING GULL ATTACKS AND EATS ADULT MALE OLDSQUAW Richard R. Snell 110

RED-LEGGED KITTIWAKES FORAGE IN MIXED-SPECIES FLOCKS IN SOUTHEASTERN ALAS-
KA Douglas Siegel-Causey and Thomas E. Meehan 111

GROUND-FEEDING METHODS AND NICHE SEPARATION IN THRUSHES Alan Tye 112

AMERICAN COOT DISTRIBUTION AND MIGRATION IN COLORADO

Warner P. Gorenzel, Ronald A. Ryder and Clait E. Braun 115

REPRODUCTIVE RATE AND RENESTING OF RED-WINGED BLACKBIRDS IN MINNESOTA

Daniel W. Moulton 119

MIGRATION SPEEDS OF THREE WATERFOWL SPECIES

Frank C. Bellrose and Robert C. Crompton 121

ORNITHOLOGICAL LITERATURE 125

ORNITHOLOGICAL NEWS 135

NUMBER 2

THE RUFOUS-FACED CRAKE {LATERALLUS XENOPTERUS) AND ITS PARAGUAYAN CONGENERS

Robert W. Storer 137

RESOURCE USE STRATEGIES OF WADING BIRDS James A. Kushlan 145

AGE RATIOS AND THEIR POSSIBLE USE IN DETERMINING AUTUMN ROUTES OF PASSERINE

MIGRANTS C. John Ralph 164

WEATHER, MIGRATION AND AUTUMN BIRD KILLS AT A NORTH FLORIDA TV TOWER

Robert L. Crawford 189

CLIMATIC INFLUENCES ON PRODUCTIVITY IN THE HOUSE SPARROW W. Brace McGUUvray 196

CANADA GOOSE BROOD BEHAVIOR AND SURVIVAL ESTIMATES AT CREX MEADOWS, WISCONSIN

Michael C. Zicus 207

BREEDING BIRD POPULATIONS IN THE GREAT SMOKY MOUNTAINS, TENNESSEE AND NORTH

CAROLINA S. Charles Kendeigh and Ben J. Fawver 218

NEST-SITE SELECTION AMONG AD6LIE, CHINSTRAP AND GENTOO PENGUINS IN MIXED SPECIES

ROOKERIES Nicholas J. Volkman and Wayne Trivelpiece 243

COWBIRD PARASITISM AND EVOLUTION OF ANTI-PARASITE STRATEGIES IN THE YELLOW

WARBLER Karen L. Clark and Raleigh J. Robertson 249

INTERACTIVE BEHAVIOR AMONG BALD EAGLES WINTERING IN NORTH-CENTRAL MISSOURI

Curtice R. Griffin 259

president’s message 264

GENERAL NOTES

INTERSPECIFIC SONG MIMESIS BY A LINCOLN SPARROW

Luis F. Baptista, Martin L. Morton and Maria E. Pereyra 265

NOTES ON PURPLE GALLINULES IN COLOMBIAN RICEFIELDS Wallace D. McKay 267

AGONISTIC BEHAVIOR OF THE WHITE-BREASTED NUTHATCH Lawrence Kilham 271

EVASIVE BEHAVIOR OF AMERICAN COOTS TO KLEPTOPARASITISM BY WATERFOWL

Mark R. Ryan 274

ADDITIONAL EVIDENCE OF EGG-MOVING BEHAVIOR BY FEMALE GADWALLS

Robert J. Blohm 276

MALLARD USING MOVING VEHICLES FOR PREDATOR AVOIDANCE

Bruce C. Thompson and James E. Tabor 277

OCHRACEOUS WREN FAILS TO RESPOND TO MOBBING CALLS IN A HETEROSPECIFIC FLOCK

William H. Buskirk 278

FISH ATTACK ON BLACK GUILLEMOT AND COMMON EIDER IN MAINE Thomas W. French 279

CROWS STEAL GOLF BALLS IN BANGLADESH Richard M. Poche 280

NOTES ON THE STATUS OF THE COMMON AFRICAN WAXBILL IN AMAZONIA

David C. Oren and Nigel J. H. Smith 281

DISTRIBUTION AND REPRODUCTIVE SUCCESS OF ZONE-TAILED HAWKS IN WEST TEXAS

Sumner W. Matteson and John O. Riley 282

THREE CRESTED EAGLE RECORDS FOR GUATEMALA

David H. Ellis and Wayne H. Whaley 284

286

ORNITHOLOGICAL LITERATURE

NUMBER 3

SUBSPECIES VS FORGOTTEN SPECIES: THE CASE OF GRAYSON’S ROBIN {Turdus graysoni)

Allan R. Phillips 301

HYPERPHAGIA AND SOCIAL BEHAVIOR OF CANADA GEESE PRIOR TO SPRING MIGRATION

M. Robert McLandress and Dennis G. Raveling 310

A MULTIPLE SENSOR SYSTEM FOR MONITORING AVIAN NESTING BEHAVIOR

James A. Cooper and Alan D. Afton 325

FORAGING SPEEDS OF WARBLERS IN LARGE POPULATIONS AND IN ISOLATION

Douglass H. Morse 334

DIFFERENTIAL PASSERINE DENSITY AND DIVERSITY BETWEEN NEWFOUNDLAND AND OFF-
SHORE GULL ISLAND Monique /. Vassallo and Jake C. Rice 340

ENVIRONMENTAL INFLUENCE ON SOARING IN WINTERING RED-TAILED HAWKS

Charles R. Preston 350

BREEDING SUCCESS IN AN ISOLATED POPULATION OF ROCK DOVES

David E. Preble and Frank H. Heppner 357

RELATIVE ABUNDANCE OF GEORGIA CAPRIMULGIDS BASED ON CALL-COUNTS

Robert J. Cooper 363

NON-DRUMMING MALES IN A RUFFED GROUSE POPULATION Gordon W. GulUon 372

GENERAL NOTES

BEHAVIORAL IMPLICATIONS OF ABERRANT SONG OF A RED-EYED VIREO

Jake C. Rice 383

COURTSHIP FEEDING AND COPULATION OF ROYAL TERNS Lawrence Kilham 390

TWO CASES OF COMMENSAL FEEDING BETWEEN PASSERINES Mark B. Robbins 391

FOOD FINDING IN BLACK-CAPPED CHICKADEES: ALTRUISTIC COMMUNICATION?

Millicent S. Ficken 393

THE SENTINEL CROW AS AN EXTENSION OF PARENTAL CARE

Gloria M. D’Agostino, Lorraine E. Giovinazzo and Stephen W. Eaton 394

BEHAVIOR OF A MALE LEAST BITTERN INCUBATING AFTER LOSS OF MATE

B. T. Aniskowicz 395

NOTES ON BROWN PELICANS IN PUERTO RICO

Ralph W. Schreiber, David W. Belitsky and Bruce A. Sorrie 397

EGGS OF THE MARBLED MURRELET Lloyd F. Kiff 400

FIRST DOCUMENTED CINNAMON TEAL NESTING IN NORTH DAKOTA PRODUCED HYBRIDS

John T. Lokemoen and David E. Sharp 403

FIRST RECORD OF THE BLACK-CHINNED HUMMINGBIRD IN ALBERTA

Philip H. R. Stepney and Peter C. Boxall 405

ORNITHOLOGICAL LITERATURE 407

REPORT OF THE CONSERVATION COMMITTEE — 1980 438

NUMBER 4

ON AERIAL AND GROUND DISPLAYS OF THE WORLD’S SNIPES George Miksch Sutton

FORAGING OF FIVE BIRD SPECIES IN TWO FORESTS WITH DIFFERENT VEGETATION STRUCTURE

Brian A. Maurer and Robert C. Whitmore

AGE AND SEX DIFFERENCES IN WING LOADING AND OTHER AERODYNAMIC CHARACTERISTICS
OF SHARP-SHINNED HAWKS Helmut C. Mueller, Daniel D. Berger and George Allez

EVIDENCE FOR AERODYNAMIC ADVANTAGES OF TAIL KEELING IN THE COMMON CRACKLE

Scott Hickman

REPRODUCTIVE CORRELATES OF ENVIRONMENTAL VARIATION AND NICHE EXPANSION IN THE
CAVE SWALLOW IN TEXAS Robert F. Martin

HORNED LARK BREEDING BIOLOGY AT CAPE ST. MARY’S, NEWFOUNDLAND

Richard J. Cannings and William Threlfall

ASPECTS OF THE BREEDING BIOLOGY OF A SUBTROPICAL ORIOLE, Icterus gularis

Barbara Yohai Pleasants

GENERAL NOTES

AN EXAMPLE OF A HYBRID GREEN JAY X BLUE JAY

Warren M. Pulich and Rebecca M. Dellinger
DUSKY SEASIDE SPARROW FEEDS RED-WINGED BLACKBIRD FLEDGLINGS

James L. Rakestraw and James L. Baker
STATISTICAL SIGNIFICANCE AND DENSITY-DEPENDENT NEST PREDATION

Stephen D. Fretwell and Frank S. Shipley
A COMPARISON OF NEST-SITE AND PERCH-SITE VEGETATION STRUCTURE FOR SEVEN

SPECIES OF WARBLERS Scott L. ColUns

USE OF ARTIFICIAL PERCHES ON BURNED AND UNBURNED TALLGRASS PRAIRIE

Janet Jean Knodel-Montz

JUVENILE PEREGRINE FALCON SWOOPS ON ROSEATE SPOONBILLS E. Scott Clark

SYMBIOTIC INTERACTION BETWEEN STARLINGS AND DEER Robert K. Murphy

CATTLE EGRETS FEEDING IN ASSOCIATION WITH HUMAN WORKERS G. K. Menon

SCRUB JAY CAPTURES HERMIT THRUSH IN FLIGHT

M. Robert McLandress and Use McLandress
FOOD HABITS OF BLACK-BELLIED WHISTLING DUCKS OCCUPYING RICE CULTURE HABI-
TATS Godfrey R. Bourne

HOUSE SPARROWS FLUSHING PREY FROM TREES AND SHRUBS

Harland D. Guillory and Jack H. Deshotels
DIFFERENTIAL PREDATION BY TWO SPECIES OF PISCIVOROUS BIRDS

Fritz L. Knopf and Joseph L. Kennedy

RED PHALAROPE EATING CARRION Wade Wander

RE-MATING OF A LESSER SNOW GOOSE

Kenneth F. Abraham, Pierre Mineau and Fred Cooke

COMMON EIDER PLAYS “POSSUM” Douglas B. McNair

TERRITORIAL ATTACHMENT AND MATE FIDELITY BY HORNED GREBES

Robert S. Ferguson

EFFECTS OF REDHEAD NEST PARASITISM ON MALLARDS

Larry G. Talent, Gary L. Krapu and Robert L. Jarvis

SURVIVAL OF A DEMAXILLATE RED- WINGED BLACKBIRD Kent L. Fiala

MINIMIZING INVESTIGATOR DISTURBANCE IN OBSERVATIONAL STUDIES OF COLONIAL
BIRDS: ACCESS TO BLINDS THROUGH TUNNELS

Gary W. Shugart, Mary A. Fitch and Vern M. Shugart

ORNITHOLOGICAL LITERATURE

PROCEEDINGS OF THE SIXTY-SECOND ANNUAL MEETING
INDEX

457

478

491

500

506

519

531

538

540

541

542

547

548

549

550

551

554

557

559

560

562

563

565

570

575

589

w 1C
8V2,f

PUBLISHED BY THE WILSON ORNITHOLOGICAL SOCIETY

VOL. 93, NO. 1 MARCH 1981 PAGES 1-136

The Wilson Ornithological Society
Founded December 3, 1888

Named after ALEXANDER WILSON, the first American Ornithologist.

President — George A. Hall, Department of Chemistry, West Virginia University, Morgan-
town, West Virginia 26506.

First Vice-President — Abbot S. Gaunt, Department of Zoology, Ohio State University, Co-
lumbus, Ohio 43210.

Second Vice-President — Jerome A. Jackson, Department of Biological Sciences, P.O. Draw-
er Z, Mississippi State University, Mississippi State, Mississippi 39762.

Editor — Jon C. Barlow, Department of Ornithology, Royal Ontario Museum, 100 Queen’s
Park, Toronto, Ontario, Canada MSS 2C6.

Secretary — Curtis S. Adkisson, Department of Biology, Virginia Polytechnic Institute and
State University, Blacksburg, Virginia 24061.

Treasurer — Robert D. Burns, Department of Biology, Kenyon College, Gambler, Ohio 43022.

Elected Council Members — Sidney A. Gauthreaux, Jr. (term expires 1981); Richard C. Banks
(term expires 1982); Mary H. Clench (term expires 1983).

Membership dues per calendar year are: Active, $16.00; Student, $14.00; Sustaining, $25.00;
Life memberships $250 (payable in four installments).

The Wilson Bulletin is sent to aU members not in arrears for dues.

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. Now 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, thus
making available to all Society members the more important new books on ornithology and
related subjects. The fund will be administered by the Library Committee, which will be
happy to receive suggestions on the choice of new books to be added to the Library. William
A. Lunk, University Museums, University of Michigan, is Chairman of the Committee. The
Library currently receives 195 periodicals as gifts and in exchange for The Wilson Bulletin.
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: The Josselyn Van Tyne Memorial Library, University of Michigan
Museum of Zoology, Ann Arbor, Michigan 48109. Contributions to the New Book Fund
should be sent to the Treasurer (small sums in stamps are acceptable). A complete index of
the Library’s holdings was printed in the September 1952 issue of The Wilson Bulletin and
newly acquired books are listed periodically. A list of currently received periodicals was
published in the December 1978 issue.

The Wilson Bulletin
(ISSN 0043-5643)

The ofhcid organ of the Wilson Ornithological Society, published quarterly, in March, June, September, and December.
The subscription price, both in the United States and esewhere, is $20.00 per year. Single copies, $4.00. Subscriptions,
changes of address and claims for undelivered copies should be sent to the Treasurer. Most back issues of the Bulletin are
available and may be ordered from the Treasurer. Special prices wiU be quoted for quantity orders.

All articles and communications for publications, books and pubbcations for reviews should be addressed to the Editor.
Exchanges should be addressed to The Joselyn Van Tyne Memorial Library, Museum of Zoology, Ann Arbor, Michigan
48109. Known office of publication: Department of Zoology, Ohio State University, 1827 Neil Avenue, Columbus, Ohio
43210.

Second class postage paid at Columbus, Ohio and at additional mailing office.

Antarctic Petrel (Thalassoica antarctica). Amundsen Sea, Antarctica,
January 1980. Photograph by J.R. Jehl, Jr.

THE WILSON BULLETIN

A QUARTERLY MAGAZINE OF ORNITHOLOGY

Published by the Wilson Ornithological Society

VoL. 93, No. 1 March 1981 Pages 1-136

Wilson Bull., 93(1), 1981, pp. 1-20

OBSERVATIONS OF SEABIRDS DURING A CRUISE
FROM ROSS ISLAND TO ANVERS
ISLAND, ANTARCTICA

Robert M. Zink

The breeding avifauna of the Sub-Antarctic and Antarctic consists of
about 65 species, mostly seabirds. In the antarctic region, which includes
the Antarctic continent. Peninsula and adjacent islands, there are 18
breeding species; 8 are restricted as breeders to the Antarctic Peninsula
(Watson 1975). These 18 species depend largely on the sea for their ex-
istence and most are highly pelagic, only coming to land for breeding
during the short austral summer. Sub-adult and foraging adult seabirds
from subantarctic islands and the wintering Arctic Tern {Sterna paradi-
saea) are also found in antarctic waters during the austral summer. During
the austral spring, under the influence of warming temperatures, wind and
water currents, the continuous band of ice which surrounds the Antarctic
continent and much of the Peninsula in winter, breaks up into pack ice
and open water. Seabirds forage in continuous daylight of the austral sum-
mer and exploit plankton-abundant regions of the antarctic oceans.

Most information available on antarctic and subantarctic seabirds has
resulted from terrestrial studies of their breeding biology. Although little
is known about their pelagic distributions, behaviors and ecologies (Wat-
son 1975, Ainley et al. 1978), these are influenced by prolonged sexual
immaturity, nonbreeding periods when sexually mature, absence between
bouts of incubation and foraging for food for young. In addition, there is
little known about patterns in pelagic seabird community composition and
species interactions. Biotic and abiotic determinants of species occur-
rences are also poorly understood (Watson 1975) although data exist on
effects of sea water temperature (Szijj 1967), ice concentration (Cline et

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

al. 1969, Zink 1978, Ainley et al. 1978) and local plankton blooms (Cline
et al. 1969, Ainley et al. 1978) on bird distributions.

Early studies (Gain 1914; Siple and Lindsey 1937; Holgerson 1945, 1957;
Bierman and Voous 1950) established the broad patterns of seabird dis-
tributions and general pelagic ecology in Antarctica; recent summaries of
many aspects of antarctic avian biology can be found in Watson et al.
(1971) and Watson (1975). In addition, Watson (1975) provides a useful
discussion of the maritime and terrestrial physical environments.

This study contributes information on the pelagic distribution, abun-
dance and habits of 16 species of seabirds obtained during an austral
summer cruise aboard the USCGC Glacier, while it traveled through the
Ross, Amundsen and Bellingshausen seas, from McMurdo Station (Ross
Island) to Palmer Station (Anvers Island). Recent seabird observations
from the areas covered herein include those of Erickson et al. (1972) and
Darby (1970), and are compared with recent pelagic seabird surveys from
the Weddell Sea (Cline et al. 1969, Parmelee 1977, Zink 1978).

ITINERARY AND METHODS

The route and itinerary of the Glacier are shown in Fig. 1. This study was largely oppor-
tunistic as the purpose of this portion of the 1976 International Weddell Sea Oceanographic
Expedition (IWSOE) was merely to reach Palmer Station, consequently, the route
depended mostly on pack ice densities. Of the 136.1 h of observation, 99.1 h were
taken in the Ross Sea (170°E-135°W), 23.5 h in the Amundsen Sea (135°W-90°W) and 13.5
h were mostly in ice-free portions of the Bellingshausen Sea (90°W— 66°W).

Observations were made from the “flying bridge” of the Glacier, 16 m above the waterline,
during periods of various lengths (0.5— 4.5 h) throughout the day. A minute by minute account
(in GMT) was kept of numbers and behavior of birds and seals sighted within approximately
0.4 km of each side of the ship, providing a census strip width of 0.8 km. The data on seals
are discussed elsewhere (in prep.). To determine the transect boundaries a 12.7 x 38.1 cm
rectangular hoard with a line, describing an angle of 2.3°, drawn diagonally from a top corner
to a point on the opposite 12.7 cm side was used. While the top (i.e., 38.1 cm) edge of the
hoard was sighted to the horizon (and perpendicular to the ship’s course), a simultaneous
sighting was taken along the line, which intersected the water 0.4 km from the observer (for
details see Cline et al. [1969]). I tried to monitor specific birds following the ship to prevent
multiple entries of given individuals. When several individuals of the same species were
following the ship, all individuals seen at 5-10 min intervals were counted and the approx-
imate turn-over rate was estimated. Maximum number in sight during any count and turn-
over rate were used to estimate the number of a given species seen during the transect
period. Estimated numbers of birds in large flocks were probably conservative; direct counts
of flocks while the Glacier was stuck in pack ice showed that initial estimates were low by
as much as 30-40%.

Pack ice concentration during each observation period was recorded in oktas (0 indicating
open seas, 8 representing solid pack). Ice concentrations changed rapidly during many cen-
suses and prevented a precise correlation of each sighting with a specific okta value (see
discussion of Adelie Penguin [Pygoscelis adelioe] for an example). Also, the okta value
assigned to a flying bird or one foraging in a relatively narrow zone of several ice concentra-

Zink • SEABIRDS IN ANTARCTIC WATERS

Fig. 1. Approximate route of the Glacier from McMurdo Station to Palmer Station. Num-
bers refer to dates between 16 January and 7 February 1976. Crosses (-I-) indicate noon
(GMT) positions.

tions was somewhat arbitrary. Because of these factors, and insufficient data for each okta,
data were grouped into 1—4, 5-8 oktas and open seas. Large leads (channels of water through
ice floes) or polynyas (areas of open water in sea ice, distinguished from leads) were scored
as open seas even though they existed within the pack ice, accounting for some of the
apparent “open seas” occurrences of pack ice species, as it did in the study by Cline et al.

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

Table 1

Summary of Shipboard Observation Periods from McMurdo Station to Palmer
Station (16 Jan. -6 Feb., 1976)

Census h

Percent

Km^ of census

Percent

Open water

14.7

10.8

243.5

12.1

Pack ice (1-4 oktas)

100.2

73.6

1438.5

71.2

Pack ice (5-8 oktas)

21.2

15.6

337.2

16.7

Total

136.1

100.0

2019.2

100.0

(1969). The Glacier followed routes of easiest passage through pack ice and habitats sampled
were consequently biased towards open pack, leads and polynyas (Table 1).

Sightings of each species in the 3 habitats were totaled to ascertain general relationships
(Table 2). Information on pelagic zonal distribution of each species in antarctic (continental
and maritime) and suhantarctic (transitional, cold and temperate) waters is given (after Wat-
son 1975). General locations of known breeding sites along the census route are given in
Table 2. Density estimates (given as mean ± SD) were computed for the 3 most common
bird species by dividing the number of individuals seen during a census by the area censused
(Table 3); for some other species only maximum densities are given, and for the remaining
species densities were not computed because of low numbers or uncorrected biases in the
probability of detection. For instance, although albatrosses were identifiable at 0.4 km, smaller
birds, such as prions and storm-petrels, were not. The areas covered during the 69 censuses
ranged from 8.6-94.5 km^ (x = 29.3 ± 15.3 knP). Observation periods were not uniformly
spaced, either within or among days, hence there was insufficient basis for a daily comparison
of species or abundances. To show general patterns in species pelagic ranges, the days for
which each species was observed are given in Table 2. The Glacier often traveled 300+ km
per day, thus only approximate range estimates are possible from this presentation; specific
locations of important sightings are given in the text.

Data from Erickson et al. (1972) were used to calculate density estimates. They censused
birds from 23 January-15 February 1972 in areas of pack ice, from its eastern edge in the
Bellingshausen Sea at about 85°W to its western edge in the Amundsen Sea at 135°30'W.
'I'heir transects ranged from 68°S-72°S, which represented distances into the pack of from
39-330 km, respectively. Their 23 censuses sampled an area of 1255 km^ over 88 h; the
average census covered 54.6 ± 38.5 krn^ and lasted 3.8 ± 1.9 h. They did not partition their
observations according to ice concentrations, hence throughout the present paper, references
to their density and abundance figures are for pack ice in general (however, they stated that
ice concentrations governed their penetration into the pack, hence their censuses were
probably in pack ice of less than 609b concentration).

The densities given by Gline et al. (1969) in their study of birds of the summer pack ice
in the Weddell Sea were converted from birds/rnile^ to hirds/km‘^. They related the occur-
rence of birds to pack ice concentrations as follows: light (10-309b), medium (40—60%) and
heavy (70-100%). Parmelee (1977) provided observations of seabirds obtained during a study
in the Weddell Sea, primarily in open water, between 23 January and 26 February 1973;
there were no density estimates given or derivable from this study. Zink (1978) censused
birds in the northwestern Weddell Sea during the austral summer of 1976 and presented
data similar to those in the present report, however, pack ice densities represented averages

Table 2

Number, Dates Each Species was Sighted and Observed Habitat Preferences

Zink • SEABIRDS IN ANTARCTIC WATERS

c/2 -

O S -g

5 ?o > E j 2

h4 ^

^ . LO . . ^

^-Ph^^CL PhQhPh“ OhDh

<z<<< z<<zz<z

4^ <u a;

H H H

U CJ CJ CJ
U( j-T j-T s-T tT

Z Z H H H H H

S S S S S § §

4J OJ OJ

H H H

CJ CJ CJ CJ CJ

s-T ^ s-T s-T J J I

H Z H H H H I

H S

■B -S

S-H

'y c

I ^

£ CO
2 ^
o JS

O -jS
^ 3

rt o

£ 2

•S t =2

3 *- <
*=^ C _
£ .2 ^
^ o i

y 3
O s-i
y y
y _c

y y

H H

CJ CJ CJ

H H H

Cj"

cj"

CJ*"

cj'

cj''

S CJ

Cj"

(Jl

,-h'

,— H

r-C

■o’

r--

1 — 1

r-H

'Cf

1 1

r--

•^"

f-H

cm"

r-"

On"

r-H

lo"

r-"

On"

r— 1

■O'

,-H

-H

,-H

r--

^ 1

'Cf-

M ^3

a -a

^£oi3-..

CQ W CO CQ O CJ

CO 3

-3

"5 .„
P J

^ §

< -c

3 2 •£ .2

£ y y

J _Q "O ^

_3 ::3 3 3

bC < 3 g

^ <

2 i "

i ^ i -2

-3-03

“D 2 c
c -r o C

^ E as
E -2 ^ s

“O t o

iu
« cS O "

•S 5 t. CJ
o o
^ i! -S §

1) u _, -
i! S 5
-C ® S C

3 "O _c <

!■“*==
^ « -2 -2

^ : o|

C J= -o 3

N H ^ ^
■?

< 2

II S

^ i

2 2
^ *1
S I

"O 2;

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

Table 3

Density Estimates for Selected Species®

Density

Open water Pack ice (1^ oktas)

(N = 11) (N = 46)

Pack ice (5-8 oktas)
(N = 12)

Species

SD (range) x SD (range)

(range)

Adelie Penguin

—

_ _ 1.79b + 3 33 (0-18.92)

4.05

± 4.46

(0-11.65)

Antarctic Petrel

0.31

± 0.24 (0-0.74) 1.57 ± 4.99 (0-31.87)

0.23

± 0.53

(0-1.78)

Snow Petrel

0.10

± 0.25 (0-0.79) 0.48 ± 0.45 (0-2.55)

0.38

± 0.41

(0-1.35)

^ In birds/km^.

Based on N = 36; see range in Fig. 2.

over all oktas because of insufficient pack ice census data. Variance in densities, not given
in the original report, have been calculated from the original census data.

I'here was continuous daylight from 16 January-1 February, hut by 6 February there were
7 h of darkness per day. The daily mean sea water temperature from McMurdo to Palmer
varied from — 1.9-2. 6°C (3-12 readings per day, N = 193; x - -0.95 ± 0.88°C). Because of
the narrow range, bird distributions were not compared with sea water temperatures, as was
done by Szijj (1967) in Pacific suhantarctic waters. Data gathered during periods when vis-
ibility was less than 0.4 km were not included in the abundance or density estimates but
were used for distributional records and behavioral information.

RESULTS

A total of 7786 individuals representing at least 16 species were sighted
and their numbers and general habitat occurrence are given in Table 2.
The hypothesis that each species was randomly distributed with respect
to area censused, within each habitat category, was tested with Chi-square
and rejected {P < 0.005) (see Table 2) in all species except the Southern
Giant Fulmar {Macronectes g:ig(int€us). Thus, these arbitrarily chosen di-
visions do provide indications of general habitat preference.

The occurrence, absolute and relative abundance of each species are
functions of census effort per habitat and geographic region studied, as
well as of species spatio-temporal distributions and densities. Since most
of the census effort was in pack ice (1775.7 of 2019.2 km^), it is not sur-
prising that 4 species typical of pack ice (Antarctic Petrel [Thalassoica
antarctica]^ Adelie Penguin, Snow Petrel [Pagodroma nivea], Arctic Tern)
comprised 90.4% of the total birds observed; these species often occur in
high densities. Many of the infrequently sighted species are characteristic
of more open seas and/or lower latitudes. However, of all species observed,
only the Black-browed {Diomedea melanophris)^ Gray-headed {D. chry-
sostoma) and Wandering (D. exulans) albatrosses typically are not found
in continental antarctic waters (see zonal classification in Table 2). A con-

Zink • SEABIRDS IN ANTARCTIC WATERS

siderable range of latitudes (65°S-77°S) and longitudes (170°E-66°W) were
censused, in addition to different habitats, hence the absolute ranking of
species; perhaps the relative relationships reflect arbitrary distribution of
sampling effort.

Species typical of pack ice. — Antarctic Petrel. — This was the most fre-
quently sighted species during this cruise and occurred mostly in loose
pack ice (Table 2). The 73 individuals sighted in open water were in groups
of 4 or fewer. The mean density of 1.57 ± 4.99/km^ in 1—4 oktas was
biased by a concentration of 1750 birds sighted during a 3-h period on 27
January at 74°30'S, 179°0'W, in 2-4 oktas of pack ice (Table 3). Several
large flocks of 300-400 birds roosting on large tabular icebergs and nu-
merous other smaller flocks and individuals standing on pack ice were
observed, accounting for the maximum density of 31.9/km^. This concen-
tration was unexplained. Without these flocks the mean density was 0.90
± 2.04/km^ and was typical of most pack ice (1-4 oktas) areas censused.
The mean and maximum densities in 5-8 oktas, 0.23 ± 0.53/km^ and 1.78/
km^, respectively, were less than those recorded for 1^ oktas and were less
variable. However, while the preference for light pack ice was clear (Table
2), the densities in light and heavy pack ice were not statistically different
(^ - 1.121, df = 58, NS).

Erickson et al. (1972) sighted this species on all but 2 of their 23 tran-
sects and recorded a mean density of 4.31 ± 6.55/km^ and a maximum
density of 23.06/km^ in the pack ice of the Bellingshausen and Amundsen
seas; the 4894 individuals they counted constituted 47.7% of the total birds
they observed. Of the 23 density estimates calculated from their data, only
5 exceeded 4.0/km^ (5.76, 12.49, 15.48, 18.56 and 23.06).

Cline et al. (1969) observed no large concentrations of Antarctic Petrels
in the Weddell Sea and recorded a mean density of 0.35/km^ (no variance
given); they noted this species most often in light pack ice. In the pack
ice of the northwestern Weddell Sea, Zink (1978) observed this species
most frequently in 1-4 oktas and recorded a mean density of 3.0 ± 11.03/
km^ and a maximum density of 58/km^. Parmelee (1977) also found them
common in the Weddell Sea. Present data are insufficient to determine if
there are significant differences between densities of Antarctic Petrels in
the Weddell Sea and the Ross, Bellingshausen and Amundsen seas, al-
though there is a preference by this species for light pack ice concentra-
tions.

Adelie Penguin. — Occurrence of Adelie Penguins during this cruise is
shown in Fig. 2. Erickson et al. (1972) reported this species as rare in the
pack ice of the Bellingshausen and Amundsen seas; they observed 24
individuals in 1255 km^ of pack ice census. A possible reason for this is
the apparent lack of breeding sites along the coasts of the Bellingshausen

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

a region where Adelie densities were significantly higher than other areas in which they were
sighted.

and Amundsen seas (G. E. Watson, pers. comm.). The absence of Adelies
from these areas was substantiated during the present study, as most of
the Adelies (2746 of 2773) were observed in the Ross Sea. Twenty-seven
Adelies were sighted between 135°W and 127°W in 7.3 h (118.1 km^) and
none was seen east of 127°W during 18.3 h (294.2 km^) of pack ice census.
This species reportedly breeds on Peter I Island (Watson 1975), however,
none was observed from the Glacier (see Fig. 1) or during a helicopter
seal census covering 65 km^ at approximately 67°47'S, 90°47'W, which is
roughly 100 km north of Peter I. Because of the apparent absence of
Adelies from the Bellingshausen and Amundsen seas, censuses in these
areas were not used to calculate density estimates. In the Ross Sea, den-
sity estimates ranged from 0-18. 9/km^ and the mean densities were
1.79 ± 3.38/km^ in 1-4 oktas and 4.05 ± 4.46/km^ in 5-8 oktas of pack
ice (Table 3); the difference between these mean densities was insignifi-
cant (^ = 1.85, df 48, NS). On several occasions Adelies were observed
in 3-5 oktas during 1 census, and it was difficult to assign these obser-
vations to either (1^ or 5-8) pack ice category. Thus, while there appeared
to be more Adelies in heavy pack ice (Table 2), the insignificant relation-
ship may reflect consolidation of observations into 1-4 and 5-8 oktas,
whereas, possibly Adelies prefer concentrations of 3-5 oktas. Further-

Zink • SEABIRDS IN ANTARCTIC WATERS

more, the distribution of Adelies in the Ross Sea was not uniform. The
blackened portion of Fig. 2 represents an area where Adelies appeared to
be concentrated. Within these areas a density of 5.60 ± 5.13/km^ was
recorded (N = 15), whereas the density was 1.24 ± 1.78/km^ for the re-
mainder (N = 25) of the Ross Sea; using a t-test for unequal variances
and sample sizes (Snedecor and Cochran 1967:114-116) these differences
were statistically significant (^ = 3.179, df = 40, 0.005 < P < 0.01). This
concentration possibly consisted of nonbreeders attracted to a localized
food source.

Cline et al. (1969) found Adelies to be the most abundant bird in the
summer pack ice of the Weddell Sea and recorded a mean density of
10.96/km^ (no variance given) and a maximum of 90.81/km^; they found
them most frequently in heavy pack ice. They noted that the distribution
of Adelies was uneven and that the greatest numbers were seen in the
northern limits of the pack. Their maximum density estimate consisted of
1360 individuals observed on 15 March 1968, roughly 350 km from nearby
breeding sites on the Antarctic Peninsula. They noted that this concen-
tration could have comprised dispersing adults from peninsular breeding
colonies. Zink (1978) observed large numbers of Adelies in loose pack ice,
at the northern limits of the pack ice in the Weddell Sea, and estimated
mean and maximum densities of 7.0/km^ and 101.75/km^, respectively.
(The mean density was corrected downwards to reduce the effect of sev-
eral large concentrations; original figure was mean density of 28 ± 30.6/
km^ [Zink, unpubl.].) The apparent occurrence of higher densities in the
Weddell Sea, as compared to the Ross, Bellingshausen and Amundsen
seas, needs further documentation, as both Cline et al. (1969) and Zink
(1978) censused near breeding localities and high densities recorded in
these areas could have inflated mean densities.

Adelies usually were seen in groups of 6—20 birds, often on the leeward
side of ice hummocks and pressure ridges. Like the Antarctic Petrel,
Adelies seemed to prefer older pack ice, especially ice with uneven sur-
faces providing shelter. Few Adelies were observed in the water except
when forced to dive as the Glacier approached to within 100 m. Cline et
al. (1969) also observed this behavior in the Weddell Sea and noted that
Adelies during their post-breeding molt typically remain out of the water
and fast. I observed little direct evidence of molting. That few Adelies
were seen in the water could be a result of either restricted feeding periods
or the difficulty in seeing swimming penguins. On several occasions,
Adelies were seen to approach the ship when it was stuck in ice, hence,
curiosity and/or lack of fear may explain their reluctance to flee an ap-
proaching ship.

Leopard seals {Hydrurga leptonyx) prey on adult Adelies at sea, how-

THE WILSON BULLETIN • Vol. 93, No. I, March 1981

ever, the incidence of this, and adult mortality in general, are thought to
be low (Watson 1975). On 25 January, 2 crew members on the bridge of
the Glacier observed a killer whale {Orcinus orca) lunge out of the water
and apparently capture 2 Adelies standing at the edge of a floe. Further
observations are needed to document the validity of this claim and extent
of this predation.

Snow Petrel. — This species is a common antarctic bird generally re-
stricted to cold continental and maritime waters with pack ice and/or ice-
bergs present (Watson 1975). It was sighted on all but 2 days. Most Snow
Petrels (98.58%) were seen over pack ice and their distribution was usually
uniform. The densities in pack ice were 0.48 ± 0.45/km^ in 1-4 oktas and
0.38 ± 0.41/km^ in 5-8 oktas (Table 3); there was no significant difference
between these densities {t = 0.697, df = 56, NS). Of the 847 Snow Petrels
recorded, the largest flock had 40 individuals and only 5 other flocks had
10 birds or more (10, 12, 13, 19 and 25). In the pack ice of the Bellings-
hausen and Amundsen seas, Erickson et al. (1972) recorded 3516 Snow
Petrels, which represented 34.2% of the total number of birds they ob-
served. Their mean density was 2.83 ± 3.53/km^ and the maximum was
15.92/km^, suggesting that this species is more abundant in the Bellings-
hausen and Amundsen seas. The pelagic distribution map given in Watson
(1975) appears to exclude the southern Ross Sea. In the Weddell Sea,
Cline et al. (1969) recorded a mean density of 2.82/km^ and a maximum
density of 46.68/km^ and noted this species most frequently in light and
medium pack ice concentrations. The Snow Petrel was the most common
(24.5% of total) volant species sighted by Zink (1978); a mean density of
8.0 ± 5.50/krn^ and a maximum density of 98.69/km^ (not included in the
mean density estimate) were recorded in pack ice, primarily in ice con-
centrations of from 1—5 oktas. Few birds were sighted in open waters and
a maximum density of 5.93/km^ was recorded. The largest group of Snow
Petrels observed by Parmelee (1977) in the Weddell Sea was 47 and a total
of 1588 birds was sighted (sightings made on 32 of 35 days).

The specific habitat preference of Snow Petrels during this study was
not clear as the observed frequencies in pack ice (83.7:14.9) paralleled the
relative areas censused in each habitat (71.2:16.7). Although this differ-
ence was significant (x^ 100.23, df = 1, P < 0.005) the preferred hab-

itat may have been masked by consolidating the pack ice observations into
2 categories. As with the Adelie Penguin, perhaps 3-5 oktas is a better
approximation of their habitat (ice concentration) preference.

Snow Petrels use 2 types of flight. One was “very erratic, almost bat-
like” (Watson 1975); the birds flew low over the surface, often exhibiting
high maneuverability as they closely followed ice edges. This apparent
foraging behavior is consistent with the observation of Falla (1964) that

Zink • SEABIRDS IN ANTARCTIC WATERS

their primary food consists of dead or injured macroplankton. As Ainley
et al. (1978) noted, such plankton would be expected to accumulate at the
edges of cakes of ice. The other type of flight, steady and direct, was used
about 7-10 m above open water and occasionally over pack ice.

Snow Petrels frequently hunted along narrow strips of pack ice while
avoiding adjacent open water. Upon sighting a prey item a flying individual
quickly veered upwards and then fluttered down to the surface. Feeding
motions included pecking at the surface and submerging the head and
neck to catch subsurface prey, which, according to Watson (1975), consists
mainly of fish and some invertebrates. Birds that remained on the surface
for 15 sec or longer folded their wings and swam about after prey. Foraging
individuals that alighted on the surface for less than 15 sec usually kept
their wings unfolded and held at about 60° to the surface. This was also
noted by Ainley et al. (1978) and apparently facilitates more rapid or ef-
ficient take-offs.

Arctic Tern. — Adult Arctic Terns migrate annually from their arctic
breeding grounds to “winter” in the Antarctic during the austral summer.
The Arctic Tern was the fourth most abundant species observed and was
seen mostly in loose pack ice (Table 2) along the northern edge of the
pack; few were encountered south of the pack edge. The maximum density
recorded was 2.72/km^ on 3 February. Ainley et al. (1978) noted that there
are no records of Arctic Terns from the Ross Sea. The westernmost ob-
servations of terns were on 22-23 January, when 56 birds were sighted
between 72°S-72°50'S and 144°40'W-151°6' W. These observations per-
haps establish the limit of Arctic Tern distribution in the eastern Ross
Sea. Darby (1970) observed only 1 Arctic Tern in 4 north-south traverses
of the Ross Sea and that sighting was (actually in the southern Pacific) at
67°55'S, 174°41'E on 18 January 1968.

Arctic Terns occurred in flocks of 5-20 birds; 5 flocks of more than 20
were seen (60 + , 52, 41, 30 and 25); a total of 60 birds was observed in
groups of fewer than 5. The flock of 60+ terns was on an iceberg near
Peter I Island during a helicopter seal census and was not included in the
shipboard census data. Erickson et al. (1972) recorded a mean density of
1.27 ± 1.91/km^ and a maximum density of 7.95/km^.

Cline et al. (1969) recorded a mean density of 0.97/km^ and a maximum
density of 13.90/km^, mostly in light and medium pack ice concentrations.
Zink (1978) observed a mean density of 1.4 ± 1.14/km^ and a maximum
of 3.39/km^ in pack ice, as well as a mean density of 2.67 ± 5.36/km^ and
a maximum density of 27.54/km^ in open seas. These observations support
the conclusion that Arctic Terns are most abundant in the Weddell Sea
(Watson 1975, Parmelee 1977).

Arctic Terns often passed the Glacier at the outer limits of the transect

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

strip and it was not always possible to see plumage characteristics clearly.
Individuals observed closely were in non-breeding plumage and most ap-
peared to be in heavy wing and tail molt. The few feeding individuals
observed plunged headfirst into the water from a height of about 3.5 m.
Some birds worked low over brash ice and hovered above prey items
before picking them off the surface in flight.

Southern Fulmar {Fulmarus glacial oides). — Most (287 of 289) of the
sightings of Southern Fulmars occurred between 3-6 February. The 2
birds seen west of 91°W were in loose pack ice on 19 January at 68°30'S,
161°30'W and 68°45'S, 159°0'W. Watson (1975) states that this species is
highly pelagic and avoids pack ice, but Johnstone and Kerry (1974) remark
that Southern Fulmars occur commonly in pack ice in the Australian sector
of the southern ocean. The majority of the fulmars sighted east of 91°W
(222 of 287), during the present study, were in loose pack ice (1-2 oktas),
and of these, 151 were in groups of fewer than 5 and the remainder were
in 9 flocks of fewer than 40. Most birds were sitting on the water or on
pack ice and few feeding birds were observed. Some fulmars followed the
ship briefly.

Erickson et al. (1972) observed this species on 9 of 23 transects, and of
the total of 53 individuals sighted, the maximum density was 0.42/km^ (25
birds sighted in 60 km^). The scarcity of sightings west of 91°W might be
explained by the relative lack of censuses in more northerly ice-free waters
(west of 91°W). In the Weddell Sea, Zink (1978) observed 3 of 709 birds
in pack ice and Cline et al. (1969) sighted none in pack ice.

South Polar Skua (Catharacta maccormicki). — Watson (1975) states that
during the austral summer breeding season, adult South Polar Skuas re-
main near breeding colonies, while sub-adults are found at sea. Of the 110
skuas observed, 102 were within 75 km of McMurdo Station, either in
dense pack ice or along the fast ice in McMurdo Sound. Skuas observed
near McMurdo Station were light phase birds. Of the remaining 8 skuas
sighted, 6 were seen within 115 km of McMurdo, 1 in the western Ross
Sea and 1 in the western Bellingshausen Sea. The skua in the western
Ross Sea (at 72°10'S, 157°20'W) was in loose pack ice (1 okta), and was
a dark phase bird that showed prominent golden hackles and appeared to
be in fresh plumage. Presumably an adult (based on plumage, see Watson
1975), it was some 560 km from the nearest known breeding site.

Skuas are more often found near land, thus, there are relatively few
pelagic sightings. Erickson et al. (1972) observed 8 individuals. In the
Weddell Sea, Cline et al. (1969) observed 4 and Zink (1978) sighted 10. In
spite of the paucity of pelagic sightings, individuals banded in Antarctica
have been recovered from temperate seas (Watson 1975) and Greenland
(Parmelee et al. 1977).

Zink • SEABIRDS IN ANTARCTIC WATERS

Emperor Penguin {Aptenodytes forsteri). — The largest number of Em-
peror Penguins seen at 1 time was 4. Most sightings were of groups of 2
or 3, widely dispersed in the pack ice of the Ross and Amundsen seas.
Most individuals occurred in light pack ice. Erickson et al. (1972) recorded
30 Emperor Penguins in the Bellingshausen and Amundsen seas. There
are no known breeding sites along the coasts of these seas and this species
apparently does not range far from its breedings sites, except for wander-
ing young birds. Larger concentrations were noted in the Weddell Sea by
Parmelee (1977, maximum number sighted at 1 time was 67, total sightings
were 363). Also in the Weddell Sea, Cline et al. (1969) found Emperor
Penguins most often in light and moderate pack ice concentrations and
recorded a mean density of 0.5/km^ and a maximum of 5.33/km^. Zink
(1978) observed a mean density of 0.14 ± 0.17/km^ and a maximum of
0.50/km^.

Southern Giant Fulmer. — Watson (1975) stated that this species is highly
pelagic throughout antarctic and subantarctic waters and that young in-
dividuals usually are distributed farther north than adults. Fourteen of 17
(82.3%) Giant Fulmars sighted in the Ross Sea were juveniles, based on
entirely (or nearly so) dark plumage (Watson 1975). There are both dark
and white phases of the Southern Giant Fulmar and the young of each
phase resemble the adult condition. The birds sighted during the present
study were all dark phase. Siple and Lindsey (1937) and Watson (1975)
suggested that the percentage of white phase birds increases to the south
or in pack ice; the present observations do not substantiate this. Four of
the other 8 Giant Fulmars observed in the Amundsen and Bellingshausen
seas were adults. Giant Fulmars appeared to be distributed randomly with-
in the 3 habitat types, although this is inconclusive because of the small
sample size (25).

Erickson et al. (1972) observed 65 individuals in 1255 km^ of pack ice
census in the Bellingshausen and Amundsen seas. In the summer pack
ice of the Weddell Sea, Cline et al. (1969) sighted 8 individuals. Zink (1978)
observed 106 Southern Giant Fulmars in the Weddell Sea and found them
randomly distributed between open seas and light and heavy pack ice.

Species typical of open seas. — Wilson’s Storm-Petrel {Oceanites ocean-
icus). — Wilson’s Storm-Petrels were observed in open water or loose pack
ice (Table 2); their occurrence closely followed the pelagic distribution
given by Watson (1975). Twenty-nine birds were sighted between 178°27'E
and 176°25'W (73°10'S-75°S) and 1 bird was seen between 176°25'W and
97°59'W (at 72°0'S, 145°0'W, on 22 January in 1 okta of pack ice) and the
remainder (86) were seen between 97°59'W and Palmer Station. The south-
ernmost sighting was at 76°23'S, 171°53'E, about 150 km from McMurdo
Station on 26 January. This is near the southwestern limit of the pelagic

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

range, and is approximately 450 km from a breeding site on the coast line
of Victoria Land, at 72°S, 170°E (Watson 1975). Ainley et al. (1978) noted
Wilson’s Storm-Petrels at Cape Crozier (Ross Island) and therefore the
limits of its pelagic range in the Ross Sea are equivocal.

Erickson et al. (1972) observed 7 individuals in the pack ice of the
Bellingshausen and Amundsen seas. Cline et al. (1969) observed few birds
in the pack ice of the Weddell Sea. Zink (1978) recorded a mean density
of 0.44 ± 1.15/km^ and a maximum density of 4.38/km^ in the pack ice of
the Weddell Sea in addition to a mean density of 1.44 ± 1.97/km^ and a
maximum density of 9.65/km^ in open seas. These studies indicate that
the occurrence of Wilson’s Storm-Petrels in pack ice is limited. Once,
(about 65°30'S, 67°20'W) approximately 50 birds were in view, mostly
astern and following the ship. Otherwise the maximum in view was about 4.

Blue Petrel {Halobaena caerulea). — All Blue Petrels seen in this study
and in the Weddell Sea by Zink (1978) were in open seas. A flock of 55
Blue Petrels was sighted at 70°40'S, 136°52'W on 22 January as the Glacier
passed through an extensive patch of open water in the eastern Ross Sea.
This sighting is approximately 300 km south of the previous southernmost
record (see Watson et al. 1971, Watson 1975). The other 22 individuals
were sighted on 4 February, in several small flocks, in the Bellingshausen
Sea.

Unidentified petrels. — Unidentified petrels were seen on 9 occasions in
the southern Ross Sea on 19, 20 and 26 January. Poor viewing conditions
precluded positive species determination but probably only the Mottled
Petrel {Pterodroma inexpectata) and/or Sooty Shearwater {Puffinus gris-
eus) were involved. All of these sightings are south of the known range of
the Sooty Shearwater by at least 400 km. Noteworthy is the sighting on 26
January at 76°10'S, 173°30'E, as this location is about 100 km south of the
pelagic range of the Mottled Petrel, but is 1500 km south of the known
pelagic range of the Sooty Shearwater. During 4 cruises from New Zealand
to McMurdo, between January and March 1968, Darby (1970) recorded her
southernmost Sooty Shearwater at 68°22'S, 170°18'E, however, she ob-
served no Mottled Petrels.

Antarctic Prion {Pachyptila desolata). — A single Antarctic Prion was
seen on 3 February at 68°50'S, 100°20'W over open water. Erickson et al.
(1972) did not see this species. In the large leads and polynyas of the Wed-
dell Sea pack ice, Cline et al. (1969) recorded 24 prions and a mean density of
0.04/km^. Zink (1978) recorded a mean density of 0.21 ± 0.40/km^ during
open water transects in the Weddell Sea and Parmelee (1977) observed
304 prions, which included a single observation of 100+ birds.

Cape Pigeon [Daption capense). — On 22 January, at 70°40'S, 136°52'W,
a Cape Pigeon was seen over open seas. This sighting represents a south-

Zink • SEABIRDS IN ANTARCTIC WATERS

ern extension of approximately 200 km based on the map in Watson (1975).
However, Ainley et al. (1978) noted records of Cape Pigeons in the Ross
Sea at 72°S and 76°54'S. Other records include sightings by Darby (1970)
at 73°46'S and by the first Byrd Antarctic Expedition {in Siple and Lindsey
1937) in Discovery Inlet, at 78°30'S. From these records it is possible to
conclude that only a few Cape Pigeons, perhaps sexually immature indi-
viduals, wander through the southern Ross Sea. As noted by Ainley et al.
(1978) the observation of Spellerberg (1971) that Cape Pigeons were abun-
dant off the northern tip of Ross Island during March 1964, is suspect. Of
the other 14 Cape Pigeons sighted during the present study, 5 were in
light pack ice and 9 were over open waters in the Bellingshausen Sea.
Erickson et al. (1972) sighted a Cape Pigeon in the pack ice of the Amund-
sen Sea at 67°47'S, 128°50'W on 11 February 1972. In the northwestern
Weddell Sea, Zink (1978) sighted 694 Cape Pigeons in open seas and
recorded a mean density of 1.0 ± 1.22/km^ and a maximum density of 4.2/
km^; few were observed in pack ice. Cline et al. (1969) observed a total
of 25 Cape Pigeons and a mean and maximum density of 0.04/km^ and
1.70/km^, respectively.

Albatrosses. — Four species of albatrosses (Black-browed, Gray-headed,
Wandering and Light-mantled Sooty [Phoebetria palpebrata]) were sighted
during open water censuses in the Bellingshausen Sea from 4-6 February.
These species are typically highly pelagic and avoid pack ice (Watson
1975). The largest number of adults of either the Black-browed Albatross
or Gray-headed Albatross in view at 1 time was 3. The actual number of
albatrosses observed is difficult to determine, because of their well known
ship-following habits (except for the gray-headed). Also, 2 or 3 Black-
browed or Gray-headed albatrosses were observed (discontinuously) within
a short time and there was suspicion as to whether there were 1, 2 or 3
individuals involved. Immature Black-browed and Gray-headed albatross-
es are difficult to distinguish at sea, and the records of “unidentified mol-
lymauk” in Table 2 refer to juveniles of these species.

DISCUSSION

Pelagic seabird observations obtained during the austral summer in Ant-
arctica are difficult to interpret because the relative abundances of im-
matures, unsuccessful breeders and foraging breeders are unclear. Darby
(1970) suggested that many birds seen far from land during the breeding
season were probably non-breeding or immature individuals. Delayed sex-
ual maturity is typical for most seabirds. Consequently, it is probable that
sexually immature procellariids and penguins constitute a high proportion
of birds seen at sea, since they evidently avoid regions near breeding sites

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

(Watson 1975). Because of the short austral breeding season, it is likely
that most unsuccessful breeders depart to sea rather than renest. Brook
and Beck (1972) discovered nesting Antarctic Petrels, Snow Petrels and
South Polar Skuas in the Theron Mountains, 250 km from the nearest
foraging grounds in the open waters of the Weddell Sea. Warham et al.
(1977) speculated that breeding Mottled Petrels, from colonies on islands
south of New Zealand, may range up to 2200 km into the southwestern
Ross Sea during off-duty periods. Beck (1969) noted that Cape Pigeons in
breeding condition have been collected 350 km from the nearest breeding
site. Breeding petrels range considerable distances from nest-sites and
undoubtedly contribute to the number of birds seen at sea; this contri-
bution may increase as observations are made nearer to breeding sites. It
is not generally possible to distinguish ages or sexes of most species on
the basis of sight observations. Therefore, the relative abundance of adults
and immatures is uncertain. Sightings of birds hundreds of km from the
nearest land (not necessarily breeding sites; see Table 2 for breeding sites
of species sighted during this study) provide data for species-level regional
distributions and information about species' pelagic ecology and behavior.
These data can also be used to ascertain regional patterns in the compo-
sition of pelagic seabird communities.

Density estimates are useful for providing a quantitative measure of i

avian occurrence. Densities (birds/area) are superior in most instances to '

measures of birds per unit of time or linear distance because of potentially I

serious biases in observer proficiency and ability, ship speed and uncor- |

rected biases in the nonuniform probability of sighting different species j

at a given distance. Densities can also be extrapolated to estimate the t

number of birds in a given region. Erickson et al. (1972) stated that the
area circumscribed by the outer limits of their sampling effort was about
250,000 krn‘^. If the mean density for the Antarctic Petrel (4.31/km^) is
applied to this area, the resulting number of birds is 1,077,500. However,
it should be recognized that there is considerable variance (SD = 6.55)
about this mean. The Antarctic Petrel often occurs in dense aggregates
and additional censuses are probably required. It is necessary to establish
the uniformity of the habitat within the bounds of the census effort and to
account for such clumped distributions of birds (or at least acknowledge
them). Also, because the age structure of this pelagic “population” is
unknown, such population estimates are probably area specific and not J
necessarily applicable to other geographic regions. I

In antarctic ecosystems, there are typically few species, but these are 1

often abundant (Watson 1975, Cline et al. 1969). In the present study, ;

most of the census effort (88% of the total census area) was in pack ice.

Ten species were observed in pack ice and of these, 4 (Antarctic Petrel,

Zink • SEABIRDS IN ANTARCTIC WATERS

Adelie Penguin, Snow Petrel and Arctic Tern) accounted for over 90% of
birds observed in pack ice. Erickson et al. (1972) found the same 10 species
in the pack ice of the Bellingshausen and Amundsen seas, although few
Adelie Penguins. In order of decreasing abundance, the Antarctic Petrel,
Snow Petrel and Arctic Tern accounted for 98.2% of the 10,270 total birds
they observed. In the pack ice of the Weddell Sea, the Adelie Penguin,
Snow Petrel and Antarctic Petrel accounted for 69.4% of the 14,376 total
birds (pack ice and open seas) observed by Zink (1978) even though only
21.6% of the 698 km^ of census area was in pack ice. Also in the summer
pack ice of the Weddell Sea, Cline et al. (1969) found that the Adelie
Penguin, Snow Petrel, Arctic Tern, Emperor Penguin and Antarctic Petrel
accounted for 98.8% of the 9451 total birds sighted. These studies sub-
stantiate the claim that the pack ice environment is dominated by a few,
rather abundant species. However, the same species are not equally dom-
inant between areas nor are the relative abundances equal.

Cline et al. (1969) concluded that pack ice concentrations influenced
bird distributions and densities and they found the highest densities of
most species in ice concentrations of 10-60%. The observations of Ainley
et al. (1978) further support this conclusion. During the present study, ice
concentrations of 1-5 oktas supported the greatest numbers of birds. How-
ever, the type of pack ice also was important. Pack ice varies considerably
in thickness, surface and subsurface structure; these vary with age of ice
and amount of compacted snow. Different kinds of ice probably offer dif-
ferent potentials to birds, in terms of resources or shelter. Older pack ice
(2 years and up) is thicker and generally has a rough, irregular surface
that provides wind shelters for resting or molting birds. In addition, its
subsurface has usually deteriorated into a matrix of small holes and chan-
nels which are often frequented by various invertebrates (Watson 1975)
such as krill (pers. obs.). There were qualitative indications that more
birds were in areas of older pack ice than in newer (thinner and flatter)
pack ice. Future studies should determine the type of ice as well as its
concentration for correlations with the occurrence of birds, seals and
whales.

Voous (1965) suggested that food availability and abundance are prob-
ably the primary factors influencing the distribution of birds in antarctic
waters. From the above discussion, it is apparent that the pelagic occur-
rence of seabirds results from an interaction of food availability and ice
concentration and structure.

There was insufficient census effort in open waters to determine, with
much certainty, the species characteristic of this environment. Species
typical of open, continental antarctic waters are the Southern Fulmar,
Wilson’s Storm-Petrel, Blue Petrel, Southern Giant Fulmar, Cape Pigeon,

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

Antarctic Prion and Light-mantled Sooty Albatross (Watson 1975). In gen-
eral, seabird diversity is greater in antarctic maritime and subantarctic
waters than in antarctic continental waters (Darby 1970, Szijj 1967, Watson
1975). Darby (1970) thought that the northern limit of the pack ice consti-
tuted a barrier to the southward expansion of many species and noted that
several species were consistently 80-100 km north of the pack, whatever
its northern limit. On 4 February, the Glacier passed out of what was
probably the northern extent of pack ice (68°S, 87°W). The species com-
position changed rapidly (within 2 km) from predominantly pack ice
species to speeies characteristic of open seas. The interface between pack
edge and open water potentially supports large numbers of birds because
of an accumulation of plankton (Routh 1949). However, no such concen-
trations of birds were noted.

The interactions and associations of various species at sea are poorly
understood. In the Bellingshausen and Weddell seas (Zink, unpubl.) Wil-
son’s Storm-Petrel appeared to be an “indicator” of localized patches of
food. Individuals seemed to search independently and were randomly in-
terspersed until a pateh of food was located. Others then flocked to the
site and fed as described by Alexander (1954) and Murphy (1936). They
headed into the wind, pattering their feet on the surface and feeding
through the patch of food and then returned downwind and worked through
this specific area again. Such a grouping of Wilson’s Storm-Petrels often
attracted other species such as the Cape Pigeon and Southern Giant Ful-
mar, and albatrosses in more northerly areas. Both the nature of specific
roles in such assemblages and the geographic trends in eomposition of
interspecific flocks are unknown.

SUMMARY

Observations of 16 species of seabirds in antarctic waters were obtained during the austral
summer of 1976 while the USCGC Glacier cruised from McMurdo Station (Ross Island) to
Palmer Station (Anvers Island). Information on distribution, abundance, habitat (ice con-
centration) preference and behavior of these species is given and comparisons are made with
recent seabird surveys in the Bellingshausen, Amundsen and Ross seas and the Weddell
Sea.

The Antarctic Petrel, Adelie Penguin and Snow Petrel were the most abundant species,
respectively, in pack ice, where most of the 136.1 h of observation occurred. The Adelie
Penguin, which was generally absent from the Amundsen and Bellingshausen seas, and the
Antarctic Petrel both occurred in large concentrations, while the Snow Petrel was more
uniformly spread throughout the pack ice encountered on this cruise. These species, the
Arctic Tern and the Emperor Penguin, probably are the primary species of the antarctic
pack ice ecosystem. Antarctic avifaunas are dominated by relatively few species. However,
the same species were not dominant throughout all regions examined and in many cases the
relative dominances were different. A part of these regional differences may be attributed
to the distribution of sampling effort between different regions and habitats. Pack ice con-

Zink • SEABIRDS IN ANTARCTIC WATERS

centrations of 1-5 oktas appeared to support the greatest numbers of birds. There were
indications that older, thicker ice is preferred by birds because its irregular surface provides
shelter and its subsurface structure is inhabited by various invertebrate prey items of sea-
birds.

Because there were few censuses in open seas, only general indications of characteristic
species were obtained. The Wilson’s Storm-Petrel, Blue Petrel, Black-browed Albatross,
Gray-headed Albatross, Cape Pigeon, Light-mantled Sooty Albatross, Wandering Albatross
and Antarctic Prion were sighted either entirely or mostly over open waters.

The relative abundance of adults and immatures in pelagic observations is unknown be-
cause they are not distinguishable at sea. Also, both sexually immature birds and foraging
breeders can be expected to occur long distances from breeding sites.

ACKNOWLEDGMENTS

This study was financed by NSF grant No. OPP74-21374, through the United States Ant-
arctic Research Program (USARP) to D. F. Parmelee. I thank Dr. Parmelee for his advice
and assistance throughout the project and Dr. G. Llano for his efforts in planning the field
season. The entire crew of the USCGC Glacier was very helpful, although I would like
especially to thank Captain C. P. Gillett, Cdr. J. Coste and Lt. B. Genez for their assistance.
S. M. Kozlechar assisted with observations on occasion and his help is greatly appreciated.
Drs. T. Foster and J. Middleton, and Lt. R. Buhl offered helpful advice throughout the
cruise. I thank D. G. Ainley, G. F. Barrowclough, L. M. Conroy, D. F. Parmelee, R. M.
Timm and H. B. Tordoff for comments on earlier drafts of the manuscript. I am especially
grateful to G. E. Watson for providing a detailed and critical review of this paper. Valuable
discussions were held with S. L. Frye, P. H. Jacobsma, H. Kermott, N. E. Lederer, R. J.
Oehlenschlager, D. B. Siniff, R. M. Timm, D. W. Warner and M. D. Witt regarding various
aspects of this study. K. A. Kohn prepared Fig. 1 and the map on which Fig. 2 is based. A.
Fosdick typed several drafts.

LITERATURE CITED

Ainley, D. G., R. C. Wood and W. J. L. Sladen. 1978. Bird life at Cape Crozier, Ross
Island. Wilson Bull. 90:492-510.

Alexander, W. B. 1954. Birds of the ocean. 2nd rev. ed. Putman, New York, New York.
Beck, J. R. 1969. Food, moult and age of first breeding in the Cape Pigeon, Daption
capensis Linnaeus. Br. Ant. Surv. Bull. 21:33-44.

Bierman, W. H. and K. H. Voous. 1950. Birds observed and collected during the whaling
expeditions of the ‘Willem Barendsz’ in the Antarctic, 1946-1947 and 1947-1948. Ardea
37:1-123.

Brook, D. and J. R. Beck. 1972. Antarctic Petrels, Snow Petrels, and South Polar Skuas
breeding in the Theron Mountains. Br. Ant. Surv. Bull. 27:131-137.

Cline, D. R., D. B. Siniff and A. W. Erickson. 1969. Summer birds of the pack ice in
the Weddell Sea, Antarctica. Auk 86:701-716.

Darby, M. M. 1970. Summer seabirds between New Zealand and McMurdo Sound. Notornis
17:28-55.

Erickson, A. W., J. R. Gilbert, G. A. Petrides, R. J. Oehlenschlager, A. S. Sinha
AND J. Otis. 1972. Populations of seals, whales and birds in the Bellingshausen and
Amundsen seas. Ant. J. of the U.S. 7:70-72.

Falla, R. A. 1964. Distribution patterns of birds in the antarctic and high latitude suhant-
arctic. Pp. 367-378 in Biologie Antarctique (R. Garrick et ah, eds.). Hermann, Paris,
France.

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

Gain, L. 1914. Oiseaux antarctiques. Deuxieme Exped. Ant. Fr. 1908-1910, 2:1-100.
Holgerson, H. 1945. Antarctic and sub-antarctic birds. Sci. Result. Norw. Ant. Exped.
1927-28, 2(23): 1-100.

. 1957. Ornithology of the ‘Brategg’ Expedition. Sci. Result. ‘Brategg' Exped. 1947-

48, 4:1-80.

Johnstone, G. W. and K. R. Kerry. 1974. Ornithological observations in the Australian
sector of the southern ocean. Proc. 16th Int. Ornithol. Congr., Canberra, Australia.
Murphy, R. C. 1936. Oceanic birds of South America. Vol. 1, Am. Mus. Nat. Hist., New
York, New York.

Parmelee, D. F. 1977. Adaptations of Arctic Terns and Antarctic Terns within Antarctic
ecosystems. Pp. 687-702 in Adaptations within Antarctic ecosystems. Gulf Publ. Co.,
Houston, Texas.

, W. R. Fraser and D. R. Neilson. 1977. Birds of the Palmer Station area. Ant.

J. of the U.S. 12:14-21.

Routh, M. 1949. Ornithological observations in the antarctic seas. Ibis 91:577-606.

SiPLE, P. A. AND A. Lindsey. 1937. Ornithology of the second Byrd Antarctic expedition.
Auk 54:147-159.

Snedecor, G. W. and W. G. Cochran. 1967. Statistical methods. The Iowa State Univ.
Press, Ames, Iowa.

Spellerberg, I. F. 1971. Arrival and departure of birds at McMurdo Sound, Antarctica.
Emu 71:161-171.

SziJJ, L. J. 1967. Notes on the winter distribution of birds in the western Antarctic and
adjacent Pacific waters. Auk 84:366-378.

VOOUS, K. H. 1965. Antarctic birds. Pp. 649-689 in Biogeography and ecology in Antarctica
(P. van Oye and J. van Mieghem, eds.). Dr. Junk, The Hague, Holland.

Warham, j., B. R. Keeley and G. J. Wilson. 1977. Breeding of the Mottled Petrel. Auk
94:1-17.

Watson, G. E. 1975. Birds of the Antarctic and Sub-antarctic. Am. Geophys. Union, Wash-
ington, D.C.

, J. P. Angle, P. C. Harper, M. A. Bridge, R. P. Schlatter, W. L. N. Tickell,

J. C. Boyd and \1. M. Boyd. 1971. Ant. Map Folio Series 14:1-18.

Zink, R. M. 1978. Birds of the Weddell Sea. Ant. J. of the U.S. 13:142-145.

J. F. BELL MUSEUM OF NATURAL HISTORY, UNIV. MINNESOTA, MINNEAPOLIS,
MINNESOTA 55455. (PRESENT ADDRESS: MUSEUM VERTEBRATE ZOOLOGY,
2593 LIFE SCIENCES BLDG., UNIV. CALIFORNIA, BERKELEY, CALIFORNIA
94720.) ACCEPTED 5 APR. 1979.

COLOR PLATE

The color plate ITontispiece of the Antarctic Petrel (Thalassoica antarctica) has been
made possible by an endowment established by George Miksch Sutton. Joseph R. Jehl, Jr.
provided the photograph.

Wilson Bull., 93(1), 1981, pp. 21-41

DISPLAY BEHAVIOR OF OVENBIRDS {SEIURUS
AUROCAPILLUS) II. SONG VARIATION
AND SINGING BEHAVIOR

M. Ross Lein

Song of the Ovenbird {Seiurus aurocapillus) is a characteristic sound of
late spring and early summer in woodlands over much of North America.
John Burroughs (1871) first used the onomatopoeic phrase teacher to de-
scribe its apparently double-syllabled phrases, characterizing it as a series
of repetitions of this phrase, beginning softly and building in a crescendo.
Although the Ovenbird was the subject of an intensive life-history study
(Hann 1937) and of experimental investigations of song recognition (Wee-
den and Falls 1959, Falls 1963), there has been no detailed examination
of song variation within local populations, or of the role of song in behav-
ioral interactions.

In addition, the Ovenbird has a second song, often referred to as the
“flight song,” heard far less frequently than the normal teacher song.
Many authors have commented on this display (Burroughs 1871, Gibbs
1885, Wright 1913, Allen 1919, Hann 1937, Kendeigh 1945), but it has not
been described carefully on the basis of tape recordings and its signifi-
cance has been largely a matter of speculation.

I previously have described the nature and use of non-song vocalizations
of breeding Ovenbirds (Lein 1980). This paper describes the song variation
of local populations, and documents singing behavior in detail.

METHODS

The study areas and methods are deseribed in a previous paper (Lein 1980) and are
summarized only briefly here. Fieldwork was conducted in 1970, 1971 and 1972 in Bedford,
Middlesex Co., Massachusetts and in South Lyndehoro, Hillsboro Co., New Hampshire. The
habitat consisted of second-growth mixed forest in both areas.

Detailed behavioral observations were made on 5-10 pairs of Ovenbirds each season,
including 9 individually color-banded males. More casual observations were made on nu-
merous other males. Individual males are identified by a 2-letter species designation (OB),
plus a suffix designating the individual (OB-A, OB-B, etc.). Behavior was recorded on a
cassette recorder and subsequently transcribed for analysis. Short-term singing rates (songs
per min) were calculated from the time required for a bird to sing 10 complete songs, and
were normally made only during regular singing not interrupted by other activity. Long-term
singing rates (songs per h) during different phases of the breeding cycle were calculated
from the number of songs males sang during entire observation periods. High-fidelity re-
cordings of songs for audiospectrographic analysis were made with a Nagra HIB tape recorder
and a Norelco D-119ES cardioid dynamic microphone and analysed with a Kay Elemetrics
6061B audiospectrograph. Because of background noise and echos in recordings, I use trac-

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

'1 M 'I ' 1 '1 '1

\ \ \ \ \ 1^

■ c

1.0 2.0
sec

Fig. 1. Complete “primary songs" of male Ovenbirds: (A) song of OB-A; (B) song of OB-
C; (C) song of an unhanded male. Note that the fifth and tenth phrases differ from the others;
only 5 songs hy this male, out of all songs recorded in this study, showed this phenomenon.

ings for illustration, rather than the actual sonagrams themselves. Terms describing the
structure of the songs are used in the manner proposed by Shiovitz (1975).

RESULTS
Song variation

Male Ovenbirds have 2 vocalizations referred to as song: the normal
“primary” or “territorial song” (Thorpe 1961), and the song referred to in
the literature as the “flight song.” This latter name is inappropriate be-
cause the vocalization is often given when the bird is not in flight. There-
fore, 1 refer to this song as “attenuated song” and restrict the term “flight
song” to those performed during a stereotyped aerial display (see below).

Primary Song

Description. — The “primary song” consists of a series of repetitions of
a single phrase (Fig. lA, B). Each phrase consists of 3-5 separate notes.

Lein • OVENBIRD SINGING BEHAVIOR

Most notes sweep rapidly downward in frequency with a concentration of
energy between 3.0 and 5.0 kHz. The highest-pitched notes start at about
9.0 kHz and some notes may fall as low as 2.5 kHz. The phrases range in
duration from 0.16-0.25 sec in different birds. The phrase is usually re-
peated from 8-13 times in complete songs. Songs longer than 14 phrases
were heard only rarely, although once a song of 26 phrases was recorded.
The phrases are separated by silent periods of about 0.05-0.1 sec. Com-
plete songs vary in length from about 2.5— 4.0 sec. Songs of an individual
can vary considerably in length, however, depending on how many repe-
titions of the phrase are involved.

Although the first phrase of a song may be separated from the second
by an interval longer than those occurring later in the song, the rate of
delivery is usually constant after the second phrase. However, the ampli-
tude increases for at least the first 5 or 6 phrases, producing the crescendo
effect. The song is harsh and certainly could not be called musical, but
it is delivered at such a loud volume that it carries long distances in the
woods.

Intra-individual variation. — Each male Ovenbird sings a single phrase
type. I recorded songs of 15 males repeatedly during this study (some over
several years), and approximately 30 other males were recorded less reg-
ularly. Only once did a male sing more than 1 phrase pattern. On 24 May
1971, an unbanded male at Bedford sang 5 songs which included 1 or 2
phrases of a type other than the predominant one (Fig. 1C). This occurred
about 5 min after a territorial encounter on the first day that the male was
on territory. The significance of these circumstances is uncertain. All other
songs of this male consisted of a single phrase type.

Complete songs of individuals showed minor variation in the number of
phrases. For example, during 4 bouts of singing on 28 May 1971, OB-I
sang 82 songs. These included songs with the following numbers of phras-
es: 7 phrases — 1 song; 8 phrases — 11 songs; 9 phrases — 56 songs; 10
phrases — 14 songs. Such variation was typical of all males, although the
modal song length differed between birds.

Incomplete songs (arbitrarily defined as songs of less than 8 phrases)
and muted songs were sung in a variety of situations described and ana-
lyzed in detail below. By definition, incomplete songs can vary from 1-7
phrases in length, but songs of 4 or 5 phrases predominated. Muted songs
were delivered at a much lower volume than normal. Most incomplete
songs are also muted since they are not long enough to develop the cres-
cendo.

Inter -individual variation. — There is much inter-individual variation in
the form of the phrases (Fig. 2). 1 did not attempt to sample all males over
wide areas, but the samples from both Bedford (Fig. 2A-L) and South

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

sec

Fig. 2. Individual phrases from the “primary songs” of male Ovenhirds: (A-L) phrases
from the songs of 12 males recorded at Bedford, Massachusetts; (M-X) phrases from the
songs of 12 males recorded at South Lyndeboro, New Hampshire.

Lyndeboro (Fig. 2M-X) probably indicate the extent of variation within
local populations.

In some cases, aural recognition of individuals was possible on the basis
of the rate of delivery of the phrases or the quality of the phrases them-
selves. For example, OB-G (Fig. 2J) sang a recognizably more rapid song

Lein • OVENBIRD SINGING BEHAVIOR

than did his neighbors. Similarly, OB-I’s song (Fig. 2L) was distinguishable
to the ear by the squeaky quality of the phrases. Such aural discrimination
could be made with certainty only for individuals with extremes of varia-
tion in these song characters.

The situation differs for audiospectrographic analysis. With a recording
of reasonable quality I could always identify the singer by comparison with
sonagrams of known individuals, even with males sharing a similar phrase
structure and whose songs could not be distinguished reliably by ear. For
example, OB-D (Fig. 2F) and OB-F (Fig. 2E) occupied neighboring terri-
tories. Their song phrases are very similar, but are characterized by a
number of minor differences. The first 2 notes of OB-F’s phrase terminate
at a lower frequency than those of OB-D. The down-up-down inflection of
the third note is more pronounced in OB-D’s phrase. Finally, the terminal
note of OB-F’s phrase begins at a lower frequency than that of OB-D.
Repeated recordings of these color-banded males established that these
minor differences were constant and could be used to identify them. Sim-
ilar consistent differences exist for other pairs of males with similar phras-
es (see OB-K and OB-J, Fig. 2Q, R).

Attenuated song

Description. — “Attenuated song” is a highly variable vocalization. Char-
acteristically it is introduced by a series of whink notes and a pie-bleep
vocalization (for a description of these calls see Lein 1980). Several normal
song phrases occur immediately or shortly after the ple-bleep. This initial
part of the song is followed by a rambling succession of other notes, many
possessing a large amount of gradual frequency modulation (see Figs. 3
and 4). This “rambling” section of the song includes 1 or more chip notes
and, in some examples, additional “primary song” phrases or ple-bleep
notes. “Attenuated songs” are variable in length, ranging from 4-7 sec.

“Attenuated song” is frequently given as part of an aerial display, var-
iously referred to as “love-song” (Burroughs 1871, Gross 1953), “passion
song” (Jones 1900) and most frequently “flight song” (Chapman 1907,
Hann 1937, Saunders 1951, Gross 1953, Gunn and Borror 1957). I use the
last term to avoid the subjective implication of motivation.

“Flight song” was heard frequently and was recorded on a number of
occasions. Because it was given irregularly, and most commonly at twi-
light, the performance was rarely seen. The male was initially perched in
a tree, usually at the height of the subcanopy. The male gave a series of
“soft sip” calls (Lein 1980) while perched. The rate of delivery of these
calls accelerated until the bird took flight and climbed to 3-15 m above
the treetops. He then flew in a hovering flight with spread wings and tail

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

while delivering the song. The flight appeared labored and the bird some-
times circled as it sang. Immediately upon completion of the song the bird
dropped back into the woods.

“Attenuated song” was also frequently given by perched males during
encounters with conspecifics. While the songs in such situations were
often incomplete, they were frequently as long as those of the “flight song”
display, and match the latter in form (Fig. 4B).

I ntra -individual variation. — The “attenuated songs” of individuals vary
within the constraints of the above description. In several cases, “flight
songs” of the same individual recorded on different dates can be matched
note by note throughout their entire length, indicating that there is some
stereotypy in the form of the song. However, other examples of “flight
songs” of the same male show differences in the rambling terminal portion,
including variation in the ordering of the elements, repetition of some
elements, or inclusion of a second series of phrases from “primary song”
(Fig. 3A, B).

Inter-individual variation. — “Attenuated song” always includes 2 in-
dividually-distinctive elements: the pie-bleep vocalization (Lein 1980) and
the “primary song” phrases. It is perhaps significant that both these ele-
ments occur together during the more stereotyped initial portion of the
display. The terminal part of the song is more variable both within and
between birds, but has the same quality and form in different males and
there is much overlap in elements between birds (Figs. 3, 4).

Singing behavior

Primary song

General pattern of singing. — Singing is almost the only feature of Oven-
bird biology that uses the vertical aspects of the territories. The Ovenbird
is a ground-nesting, ground-foraging species and seldom perches in trees,
except when singing or when alarmed. Prior to female arrival, undisturbed
males sing regularly while sitting still on perches near the bottom of the
canopy. Bouts of singing range from I or 2 to more than 20 min in length.
The mean height (estimated within I m) of 95 Ovenbird song perches was
8.8 m (SE = ±0.23 m, range = 1. 8-15. 2 m), while the mean height of the
trees in which they were perched was 15.5 m (SE = ±0.42 m, range =
3.0-24.5 m). Bouts of song are interspersed with bouts of feeding on the
ground, during which males are silent, or sing very sporadically. Males
feeding on the forest floor frequently would hop or flit up to a rock, stump
or twig before singing one of these sporadic “ground” songs; then they
would hop down and continue foraging. Several reasons may be suggested
for the obvious reluctance of males to sing from the ground. First, singing

Lein • OVENBIRD SINGING BEHAVIOR

sec

Fig. 3. “Attenuated songs” of male Ovenbirds, because of their length, have been broken
for the illustrations. The solid triangles on the time axis indicate the point of overlap of the
upper and lower portions of each song. (A-B) “Flight songs” of OB-C. Note the similar
introduction of the 2 examples, consisting in A, of 2 whink calls, 1 ple-bleep call and 3
“primary song” phrases. Also note the occurrence of the same elements in different se-
quences in the terminal parts of the 2 examples.

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

Fig. 4. “Attenuated Songs " of male Ovenbirds, because of their length, have been broken
for the illustrations. ▼ on the time axis indicate the point of overlap of the upper and lower
portions of each song. (A) “Flight song" of OB-A. Note the similarity of the elements to those
of the “flight songs" of OB-C, shown in Fig. 3. (B) “Attenuated song" of OB-I given from
the ground. Note the similarity in form to the “flight songs" illustrated in Fig. 3 and Fig. 4A.

on the ground may attract predators to an inconspicuously colored bird on
the forest floor. Second, songs delivered from the ground may not carry
well because of ground attenuation and interference by dense vegetation
with sound transmission (Wiley and Richards 1978). Song perches are at
a level of the vegetation profile which has an open structure, consisting

Lein • OVEINBIRD SINGING BEHAVIOR

Table 1

Song Rates of Male Ovenbirds during Steady Singing in Relation to Other
Features of the Situation"*

Situation

Mean song rate*"
(.songs/min

± SE)

Range

(songs/min)

No.

observa-

tions

Steady singing with no disturbance

3.20 ± 0.06

1.09-4.44

Steady singing with preening activity

2.80 ± 0.11

2.20-3.33

Counter-singing with another male

3.71 ± 0.12

2.98-4.41

Rapid singing in first morning bout‘s

4.24 ± 0.33

2.51-5.40

Rapid singing during territorial encounters*"

6.36 ± 0.93

4.05-10.34

® Short-term song rates were calculated from the times required for 10-song intervals.

” Song rates differ significantly between situations (Model 1 one-way ANOVA, P < 0.001); rate for steady singing with
no disturbance differs significantly from those for the other situations (Student-Newman-Keuls multiple range test, P <
0.05).

Both these situations involve many incomplete and/or muted songs.

only of trunks and dead branches, with very few leaves and little shrub-
bery.

During the bouts of steady singing, Ovenbirds sing at a rate of about 3
or 4 songs per min (Table 1). Several features of the external situation
influence the song rate. Bouts of intense preening activity result in a slight-
ly slower rate, while several other factors, such as counter-singing with
another male, increase the song rate (Table 1). In 2 situations in particular,
during bouts of song at dawn, and during mild territorial encounters (and
at the beginning and ending of more intense encounters), Ovenbirds may
sing very rapidly. The highest rate recorded, 10.34 songs per min, oc-
curred in such an encounter. However, in both of these situations, most
of the songs given were incomplete and many were muted.

Counter-singing (Armstrong 1963) was noted frequently. Two neighbor-
ing males would sing in phase, the songs of 1 male following immediately
after, or overlapping with, the songs of the leading male. Several facts
suggest that this is not merely the fortuitous result of 2 males singing at
approximately the same rate. First, the singing rate of such birds was
higher than during normal singing (Table 1). Second, the relationship of
“song a — song b — pause — song a — song b — pause ...” could continue for
30-40 songs. When the relationship changed it was not due to the 2 birds
drifting out of phase, as would be expected if it were fortuitous, but in
many instances by the “b” bird not “waiting” for “song a,” but rather
singing his own song toward the end of the pause. In such cases, the birds
would sometimes continue to counter-sing with the lead reversed. This
demonstrates that the males are “paying attention” to the songs of neigh-
boring males and responding to them, even at a considerable distance.

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

Table 2

Long-term Singing Rates of Male Ovenbirds in Relation to the Phase of the

Breeding Cycle

Phase in cycle

No.

observations'*

Singing rate**
(songs/h ± SE)

Observation time
(min)

Unmated

139.4 ± 9.7

633

Courtship period

65.9 ± 10.2

692

Incubation period

85.4 ± 12.6

456

Nestling period

87.9 ± 13.7

299

Fledgling period and later

26.9 ± 5.8

296

“ All observations during morning activity period.

Singing rates differ significantly between phases of the breeding cycle (Model 1 one-way ANOVA, P < 0.001); rates for
unmated and fledgling periods differ from those of other periods (Student-Newman-Keuls multiple range test, P < 0.05);
rates for courtship, incubation and nestling period do not differ significantly.

Daily pattern of singing. — The effects of daily activity rhythms on sing-
ing behavior have been well-described (Thorpe 1961, Armstrong 1963). As
in other birds, Ovenbirds show an activity peak in early morning and a
second, smaller peak in the evening. The morning peak is more pro-
nounced early in the breeding season, when the birds are very active until
noon. Later in the season strong singing becomes more restricted to early
hours. Extremely hot or cold weather inhibits singing, as does heavy rain.
Early in the season, however, birds sing strongly during light rain.

Many workers have shown that the beginning of song at dawn, and to
a lesser extent its cessation at dusk, is closely correlated with the light
intensity (Armstrong 1963). The times of first songs of Ovenbirds gradually
became earlier until the longest days of June, then became later as the
summer progressed. A comparable relationship in the time of the last song
of the evening was also noted.

Light intensity appears to influence singing in other ways as well. The
songs at the beginning of the initial bouts of the morning were character-
istically incomplete and/or muted. This was also true of late evening sing-
ing. Light level is also important in determining the occurrence of “flight
song” (see below).

Seasonal pattern of singing. — A seasonal decline in singing is known for
many species (Thorpe 1961, Armstrong 1963). Many studies have dem-
onstrated that the long-term singing rates of a male bird are related not
to the calendar date per se, but rather to the stage of the breeding cycle
(see summary in Armstrong 1963:152-156). Since the breeding cycles of
individual males are usually out of phase with one another, studies which
report only population averages in relation to calendar date may miss
important and abrupt changes.

Lein • OVENBIRD SINGING BEHAVIOR

The long-term singing rates of male Ovenbirds at various phases in the
nesting cycle are shown in Table 2. All figures are based on observations
made during the morning activity period (04:30-10:00 EDT). Unmated
males sing an average of 139.4 songs per h. If they were singing at the
mean rate calculated for undisturbed birds, 3.2 songs per min (Table 1),
this would represent steady singing for about 75% of the time.

Upon arrival of females, however, singing rate drops significantly to 65.9
songs per h, a decrease of over 50%. The decline is related to the male’s
close association with the female at this time. During such activity he
sings only sporadically and many of the songs are incomplete or muted,
a feature rarely observed prior to female arrival.

Once incubation begins, the singing rate may increase slightly, but it
never regains the level reached prior to mating. Song bouts are much more
irregular during incubation and feeding of young. By early July regular
singing is heard only for a short period at dawn.

Incomplete song. — The situations in which incomplete songs were used
range from the occurrence of 1 or 2 incomplete songs during bouts of
regular singing, to periods of up to 1 h when a male sang incomplete songs
repeatedly while moving with his mate (Table 3). Half the observations
involved association of paired males and females. Most of these were
during courtship; contact between mates during the incubation or nestling
periods is similar to that during courtship, although much less frequent or
prolonged. Males used incomplete songs when at intermediate distances
from their mates. When within 7 m or less of females, males rarely sang
at all (with the exception of “attenuated song,” see below). When more
than 25-30 m from their mates, males usually sang full songs. Males
changed from incomplete to full song (11 cases) as they moved away from
their mates. Switching from full to incomplete song (7 cases), and from
incomplete song to silence (9 cases), occurred as mates moved closer
together. Males gave incomplete songs on 4 occasions in association with
chases of their mates.

Incomplete songs were a regular feature of territorial encounters be-
tween males. Males gave incomplete songs at either a normal or a reduced
and irregular rate during male-female interactions; during male-male en-
counters there frequently was an elevated rate of singing (Table 1). In
some encounters, males sang incomplete songs almost continuously, with
very short pauses between them. Incomplete songs were not used during
intense interactions involving vigorous or prolonged chasing; these were
characterized by silence or by the use of non-song vocalizations such as
chep, whink, ple-bleep and pink (Lein 1980). Incomplete song was used
during mild encounters, or at the beginning or end of encounters, or during
pauses between bouts of active chasing.

THE WILSON BULLETIN • Vol. 93, No. I, March 1981

Table 3

Summary of Situations in Which Male Ovenbirds Used Incomplete Songs

Situation

No. cases

Major category

Subcategory®

Male-female interactions

During association

During male-female chasing

Male-male interactions

During territorial encounter

During male-male chasing

Response to approach of male

At territorial boundary

Non-encounter situations

During dawn or dusk singing

While carrying food to young

During vigorous preening

At end of singing hout

Total

® Total for subcategories may be greater than number of cases in major category since 1 observation may fall into several
subcategories.

Birds also switched from full to incomplete songs in other circum-
stances. Several times a male began to sing incomplete songs when a
neighbor, who had been silent or singing at a distance, suddenly ap-
proached the mutual territory boundary. Encounters did not ensue in these
instances and the males resumed singing full songs when they separated.
Once, a male switched to incomplete song when he moved near a known
territorial boundary. The neighbor was not singing at the time although I
had previously observed several territorial encounters in the same area.
The bird resumed full song again when he moved away from the boundary.

Incomplete song was noted on 34 occasions not involving interactions.
Twenty-seven cases occurred during the first bouts of song in the morning
or during late evening singing. In view of the influence of light level on
the onset or cessation of daily song, it is not surprising that the birds
should show a transition from silence to incomplete song to full song during
the morning twilight, or vice versa in the evening. An important difference
was that the morning performance usually involved incomplete songs given
very rapidly and regularly (Table 1), whereas the singing in the evening
was typically slow and sporadic.

Other non-encounter circumstances in which incomplete song was used
suggest that they were correlated with a conflict between a tendency to

Lein • OVENBIRD SINGING BEHAVIOR

Table 4

Summary of Situations in Which Male Ovenbirds Used Muted Songs

Situation

No. cases

Major category

Subcategory^

Male-female interactions

During association

During male-female chasing

Male-male interactions

During territorial encounter

At territorial boundary

Non-encounter situations

During dawn singing

While carrying food to young

Response to playback of song

Total

® Total for subcategories may be greater than number of cases in major category since 1 observation may fall into several
subcategories.

sing and a tendency to carry out some other activity. In 3 cases, males
sang incomplete songs during vigorous preening while on a song perch;
when the preening ceased, full song was resumed. Twice incomplete songs
came at the end of a bout of full songs, just before the male ceased singing
and began to forage. In 2 other instances, a male sang incomplete songs
while foraging and carrying food to its young. Males were typically silent
during this activity.

Muted songs. — Incomplete songs frequently sound muted, but since nor-
mal songs increase in volume for the first 4 or 5 phrases it is impossible
]; to determine this with certainty. Therefore, only songs of 5 or more phras-
I es are considered here. The situations in which muted songs were record-
ed (Table 4) are similar to those in which incomplete songs were employed
, (Table 3). Therefore, I will not consider the uses of muted song in detail.

I believe that muted song is intermediate in motivation between incom-
plete song and full song, or that it represents a “less inhibited” variation
[ of full song than does incomplete song. On several occasions, the first
^ morning bout of singing of a male began with a series of incomplete (and

*1 certainly muted) songs. These gradually lengthened until they were rec-

* ognizable as complete, but still muted, songs. Then there was a gradual
^ increase in volume until the male was singing normal volume, full songs.

I believe that this represents the usual pattern of onset of song in the
I? morning, but I was rarely close enough to a singing male to ascertain that

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

Table 5

Distribution of Ovenbird Flight Songs by Time of Day

Beginning h

^ No. displays'*

Beginning h

No. displays

01:00

13:00

02:00

14:00

03:00

15:00

04:00

16:00

05:00

17:00

06:00

18:00

07:00

19:00

08:00

20:00

102

09:00

21:00

10:00

22:00

11:00

23:00

12:00

24:00

® All times are Eastern Daylight Time.
Total of 225 displays recorded.

the intermediate songs were really muted. Similar transitions from full to
muted to incomplete songs were occasionally noted at the beginning of
male-male or male-female interactions; switching in these situations was
often abrupt and frequently omitted 1 or more of the intermediate steps
from full song to silence.

Attenuated song

Flight song. — 1 observed or heard 225 performances of the Ovenbird
“flight song” display. There is a major concentration of displays at dusk
(Table 5), with 155 (68.9%) occurring between 19:00 and 21:00. Most
“flight songs” heard in an evening occurred in a 15-20 min interval when
it was quite dark in the woods, although the sky was fairly bright and
visibility was good in open areas. A second, smaller peak occurs at sunrise.
Similar daily patterns of distribution of Ovenbird “flight songs” have been
reported by Hann (1937) and Eaton (1957).

Because “flight songs” were noted in the course of other activities and
there were no controlled observation periods, it is difficult to assess the
remainder of the distribution pattern. The twilight peaks are so pro-
nounced as to leave no question as to their reality. The rarity of “flight
songs” during the morning is also real, since the majority of my fieldwork
was between 05:00 and 11:00. “Flight songs” may occur at night (21:00—
03:00) more frequently than recorded since almost no observations were
made during this period.

Lein • OVENBIRD SINGING BEHAVIOR

Table 6

Summary of Situations in Which Male Ovenbirds Used
(Excluding Flight Song)

“Attenuated Song”

Situation

No. cases

Major category Subcategory“

Male-female interactions

During association

During male-female chasing

During attack by male

During flight toward female

During copulation attempts

Male-male interactions

During territorial encounter

During male-male chasing

After chasing

At end of encounter

Non-encounter situations

At dawn or dusk

Late in breeding season

Total

® Total for subcategories may be greater than number of cases in major category since 1 observation may fall into several
subcategories.

The predisposing influence of low light intensity on the performance of
“flight song” is evident from the twilight peaks. This is also suggested by
the occurrence of “flight songs” during the middle of drizzly and heavily-
overcast days. At least 9 (20%) of the 45 “flight songs” recorded between
06:00 and 19:00 occurred during such weather. On a dark afternoon, dur-
ing light rain, I sometimes heard several “flight songs” in an hour. This
never occurred during more clement weather.

Attenuated song during encounters. — Incomplete “attenuated songs”
are often given, but all cases considered here included some of the “ram-
bling” terminal portion of the song and may thus be considered to be
complete. Of the 33 instances when “attenuated song” was recorded (ex-
cluding “flight song” performances), 30 involved interactions between con-
specifics (Table 6). Its use in these situations is similar to those of incom-
plete and muted song, but with one difference. Twelve of 17 records during
male-female interactions involved chases, aggressive (?) attacks by the
male, or copulation attempts. This is a much higher association with very
intense interaction than was recorded for incomplete or muted song (see
Tables 3 and 4).

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

This association of “attenuated song” with high activity was also noted
during male-male encounters. Almost half (6 of 13) of the records in ter-
ritorial encounters were during aerial chases. This is in contrast to incom-
plete or muted songs which were rarely used during chases.

Attenuated song during non-encounter situations. — On 3 occasions,
male Ovenbirds sang 1 or more “attenuated songs” from a perch, or from
the ground, in the apparent absence of conspecifics (Table 6). Twice it
occurred during the twilight of dusk or dawn when one might expect to
hear “flight songs.” The third record was somewhat different and warrants
an extensive description.

On 7 July 1971, 1 was observing OB-1, who had been singing irregularly
for at least 20 min. At 07:05 he flew down to the ground and in the next
4 min sang at least 16 “attenuated songs.” All were muted in volume, but
were full-length and had the form of other “attenuated songs” (Fig. 4B).
Several times 2 or more of the songs were run together, so that the bird
was singing the soft, rambling song for periods of 10-30 sec. This perfor-
mance of “attenuated song” resembles what has been called “sub-song”
(Thorpe 1961:64-70, Thorpe and Pilcher 1958, Armstrong 1963:58-69).
Sub-song in other species is a low-volume, rambling vocalization, usually
hearing little resemblance to the typical song, although it frequently con-
tains call notes and isolated elements of “primary song.” It may be pro-
duced almost continuously for extended periods of time. The similarity to
the performance of OB-I is striking and suggests that a relationship may
be involved.

DISCUSSION

Song variation. — The Ovenbird’s pattern of song variation is a relatively
simple one which is exhibited by many passerines (Borror 1961, Thielcke
1969). Each male possesses 1 “primary song” pattern, and the song pat-
terns of different males in a local population show considerable variation.
This is in contrast to the pattern shown by the Chestnut-sided Warbler
{Dendroica pensylvanica) (Lein 1978) and some other members of that
genus (Lein 1972; Morse 1966, 1967, 1970), in which each male possesses
2 or more song patterns shared by all males in a local population.

This type of individual song variation may facilitate individual recogni-
tion of neighbors’ songs by male Ovenbirds. The experiments of Weeden
and Falls (1959) clearly demonstrated that males discriminate between the
songs of familiar and strange birds. The aggressive responses of males
were more prompt and pronounced toward the songs of strangers. This
implies that territorial establishment in this species results in a relatively
stable relationship between neighbors. The outcome is that males are more
tolerant toward neighbors, who presumably pose less of a threat to their

Lein • OVENBIRD SINGING BEHAVIOR

territorial security, than they are to strangers, who are likely to he newly-
arrived birds searching for a territory. Subsequent work by Falls (1963)
demonstrated that the pitch, form and arrangement of the component
sounds of the song, and the length of the sounds and the intervals between
them, are important in eliciting normal responses from male Ovenbirds
during playback experiments. These are the features that vary between
the songs of different individuals and they probably form the basis of
individual discrimination.

The wide range of variation in phrase structure within a local population
(Fig. 2), and the similarity of the phrases of some individuals in different
populations (compare Fig. 2H and Q, and Fig. 2L and V), suggest that a
system of “dialects” (Marler and Tamura 1962, Baptista 1975) is unlikely.

Little can be said regarding the pattern of variation in “attenuated song”
because of the small sample of recordings. However, since it contains
individually-distinctive “primary song” phrases and pie-bleep notes, it
probably shows patterns of variation comparable to those of “primary
song.”

Singing behavior. — Several features of the singing behavior of Oven-
birds deserve comment, especially in relation to the differences found in
a parallel study of the Chestnut-sided Warbler (Lein 1978). First, Oven-
birds rarely sing while foraging on the ground; Chestnut-sided Warblers
commonly sing while foraging in foliage. Whatever the reasons for the
Ovenbird’s reluctance to sing from the ground, simultaneous singing and
foraging are largely incompatible in this species. In contrast, the Chestnut-
sided Warbler combines singing and foraging with only a slight decline in
song rate (Lein 1978) and with little apparent interference with foraging.
This difference has interesting consequences for time and energy budgets.
If we assume, for the sake of argument, that males of both species require
the same time and expend the same energy in acquiring the food necessary
for subsistence, then it follows that Ovenbirds will have less time available
for territorial proclamation than will Chestnut-sided Warblers. Although
it is improbable that this assumption is strictly true, it suggests a manner
in which the features of the foraging niche of the species could influence
territorial advertisement. Species that can combine singing and foraging
may be better suited to situations of intense territorial competition, or to
conditions which require extended and continuous advertisement for other
reasons.

The second difference is in the influence of pair formation on singing
behavior. The large decline in singing of Ovenbirds at pairing is due to
the almost continuous association of mates at this time, and its inhibitory
effect on song. In contrast, pairing seems to have little influence on the
singing of the Chestnut-sided Warbler, and perhaps on the singing of

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

sexually dimorphic warblers in general (Lein 1978, Morse 1966, Ficken
and Ficken 1969). Chestnut-sided Warbler pairs spend relatively little time
in association in the period between female arrival and the start of incu-
bation (16.4% of 373 min of observations on birds whose exact breeding
status was known) in contrast to the Ovenbird (83.4% of 452 min).

This sharp decline suggests that song functions in attracting females to
unmated males. This is supported by the observation that males who lose
their mates resume singing in a manner similar to that of newly-arrived
males. However, because new males may be arriving and establishing
territories after the first males are already mated, song probably continues
to function in territorial proclamation and defense after pairing. The
amount of time that a male Ovenbird devotes to pair-related activities
during the courtship period possibly decreases his efficiency in territorial
defense. Other males frequently intrude on a territory during the courtship
period, but this may be due to the attractiveness of the highly vocal fe-
males to neighboring males at this time (pers. obs.) rather than to a decline
in the efficiency of defense.

Communicative function of song. — The broad range of circumstances in
which “primary song” is used indicates that, as with other Ovenbird vocal-
izations (Lein 1980), it encodes rather general behavioral messages (Smith
1969, 1977). It is rich in identifying messages, indicating that the singer
is a male Ovenbird, in breeding condition and on his territory. In addition,
a recipient with previous experience may be able to identify the singer as
a specific individual. By monitoring singing rate over a period of time, the
listener may also be able to determine the mating status of the singing
male.

It is difficult to suggest which behavioral selection messages (Smith
1977) the signal encodes. Probably the message is one relating to inter-
actional behavior, indicating that the singer is prepared to interact in any
of the ways typical of territorial male Ovenbirds. This is perhaps the only
behavioral selection message encoded. It could still have a variety of
meanings to different recipients. To a male who was a potential intruder
it could signify a threat; to an unmated female it could mean the presence
of a potential mate.

Additional messages for muted or incomplete songs are equally difficult
to determine. Although both variants are used in clearly agonistic situa-
tions, they also occur in circumstances not involving attack or escape, and
hence these agonistic messages probably are not encoded. The only gen-
eral feature of their use is that the bird is experiencing some type of
inhibition of, or conflict with, his tendency to sing. Hence they may encode
a message of indecisive behavior. This may provide information regarding

Lein • OVENBIRD SINGING BEHAVIOR

the communicator’s probable future behavior, although the meanings of
the signals are certainly dependent upon the class of recipient and the
context. The frequent association of these variants with interactions sug-
gests that they may encode a supplemental message indicating a greater
probability of interaction than that of full “primary song.”

“Attenuated song” used during interactions probably encodes infor-
mation similar to that borne by incomplete or muted “primary song,” since
all these vocalizations are used in similar circumstances. In encounter
situations “attenuated song” frequently accompanies chasing and attack.
However, its use in other situations, such as copulation attempts and
“flight song,” argues against the encoding of a specific attack message.
An attack meaning may be available to the recipient from the context of
the signal. The main difference between the messages of “attenuated
song” and muted or incomplete songs probably relates to differences in
the supplemental messages of probability and intensity of the behavioral
selection. “Attenuated song” may indicate a higher probability of a more
intense interaction, either aggressive or sexual.

It seems impossible to assess the communicatory function of the “flight
song,” regardless of how dramatic it is. Its rare occurrence and peculiar
situation of use, plus the obvious influence of low light level in its elici-
tation, make it difficult to suggest its function.

SUMMARY

Song variation and singing behavior of Ovenbirds were studied at 2 locations in New
England. Each male possesses a single, distinctive “primary song,” which may also be given
in muted or incomplete forms. Males also possess a second type of song, “attenuated song,”
often used as part of an aerial display (“flight song”).

Prior to pair formation, male Ovenbirds sing strongly from perches near the bottom of the
forest canopy. Few songs are given from the ground, where the majority of foraging occurs.
There is a sharp decline in singing at pairing, the result of the inhibitory effect of the female’s
presence on her mate’s singing. Muted and incomplete “primary songs“ are associated with
both male-male and male-female interactions, but are used in other situations as well. “Flight
song” is given primarily at dusk and dawn but may occur at other times, particularly during
heavy overcast. Other performances of “attenuated song” are associated almost entirely with
very intense interactions.

The relatively simple pattern of song variation described may facilitate individual recog-
nition of neighbors’ songs by males. The incompatibility of singing and foraging in this
species suggests that features of the foraging niche may influence the ability of birds to
advertise territories. Similarly, the reduction in singing produced by the extensive courtship
interactions may affect the efficiency of territorial defense at this time.

The songs of Ovenbirds appear to be rich in the identifying messages they encode, but
their message regarding future behavioral selections is probably very general, indicating only
that the singer is prepared to interact in any manner typical of territorial male Ovenbirds.
Muted and incomplete songs, and “attenuated songs,” probably encode different supple-
mental messages about the probability and intensity of interaction.

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

ACKNOWLEDGMENTS

Professor Ernst Mayr supervised this study, and encouraged me in all its phases. He and
Mrs. Mayr graciously allowed me to use their summer retreat in New Hampshire as a study
area. W. John Smith provided valuable assistance with techniques and theoretical ap-
proaches to animal communication. M. S. Ficken and R. Zach improved the manuscript with
their comments. Facilities and equipment were made available by the Museum of Compar-
ative Zoology, Harvard University. Financial support was provided by Harvard University,
the Carl A. and Katharine F. Richmond Fund, and National Science Foundation Grants
GB7346 and GB19922 to Harvard University (R. C. Rollins, Principal Investigator).

LITERATURE CITED

Allen, F. H. 1919. The evolution of bird song. Auk 36:528-536.

Armstrong, E. A. 1963. A study of bird song. Oxford Univ. Press, London, England.
Baptista, L. F. 1975. Song dialects and demes in sedentary populations of the White-
crowned Sparrow (Zonotrichia leucophrys nuttalli). Univ. Calif. Publ. Zool. 105:1-52.
Borror, D. J. 1961. Intraspecific variation in passerine bird songs. Wilson BuU. 73:57-78.
Burroughs, J. 1871. Wake-robin. Hurd and Houghton, New York, New York.

Chapman, F. M. 1907. The warblers of North America. D. Appleton and Co., New
York, New York.

Eaton, S. W. 1957. A life history study of Seiurus noveboracensis. St. Bonaventure Univ.,
Sci. Stud. 19:7-36.

Falls, J. B. 1963. Properties of bird song eliciting responses from territorial males. Proc.
13th Int. Ornithol. Congr. 259-271.

Ficken, M. S. and R. W. Ficken. 1969. Responses of Blue-winged Warblers and Golden-
winged Warblers to their own and the other species’ song. Wilson Bull. 81:69-74.
Gibbs, M. 1885. Song of the Golden-crowned Thrush. (Siurus auricapillus). Ornithol. and
Ool. 10:191-192.

Gross, A. O. 1953. Eastern Ovenhird. Pp. 457-476 in Life histories of North American
wood warblers (A. C. Bent, ed.). U.S. Natl. Mus. Bull. 203.

Gunn, W. W. H. and D. j. Borror. 1957. Interpretations of some warbler songs. Pp. 26-
36 in I'he warblers of America. (L. Griscom and A. Sprunt, Jr., eds.). Devin-Adair Co.,
New York, New York.

Hann, H. W. 1937. Life history of the Oven-bird in southern Michigan. Wilson Bull. 49:145-
237.

Jones, L. 1900. Warbler songs (Mniotiltidae). Wilson Bull. 12:1-57.

Kendeigh, S. C. 1945. Nesting behavior of wood warblers. Wilson Bull. 57:145-164.

Lein, M. R. 1972. Territorial and courtship songs of birds. Nature 237:48-49.

. 1978. Song variation in a population of Chestnut-sided Warblers (Dendroica pen-

sylvanica): its nature and suggested significance. Can. J. Zool. 56:1266-1283.

. 1980. Display behavior of Ovenhirds {Seiurus aurocapillus). I. Non-song vocaliza-
tions. Wilson Bull. 92:312-329.

Marler, P. and M. Tamura. 1962. Song “dialects” in three populations of White-crowned
Sparrows. Condor 64:368-377.

Morse, D. H. 1966. The context of songs in the Yellow Warbler. WJlson Bull. 78:444-455.

. 1967. The contexts of song in Black-throated Green and Blackburnian warblers.

Wilson Bull. 79:64-74.

. 1970. Territorial and courtship songs of birds. Nature 226:659-661.

Saunders, A. A. 1951. A guide to bird songs. Rev. Ed. Douhleday and Co., Inc. Garden
City, New York.

Lein • OVENBIRD SINGING BEHAVIOR

Shiovitz, K. a. 1975. The process of species-specific song recognition by the Indigo Bunt-
ing, Passerina cyanea, and its relationship to the organization of avian acoustical be-
haviour. Behaviour 55:128-179.

Smith, W. J. 1969. Messages of vertebrate communication. Science 165:145-150.

. 1977. The behavior of communicating. Harvard Univ. Press, Cambridge, Massa-
chusetts.

Thielcke, G. 1969. Geographic variation in bird vocalizations. Pp. 311-339 in Bird vocal-
izations (R. A. Hinde, ed.). Cambridge Univ. Press, Cambridge, England.

Thorpe, W. H. 1961. Bird-song. Cambridge Univ. Press, Cambridge, England.

AND P. M. Pilcher. 1958. The nature and characteristics of sub-song. Br. Birds

51:509-514.

Weeden, J. S. and J. B. Falls. 1959. Differential responses of male Ovenbirds to recorded
songs of neighboring and more distant individuals. Auk 76:343-351.

Wiley, R. H. and D. G. Richards. 1978. Physical constraints on acoustic communication
in the atmosphere: implications for the evolution of animal vocalizations. Behav. Ecol.
Sociobiol. 3:69-94.

Wright, H. W. 1913. Morning awakening and even-song. Second paper. Auk 30:512-537.

DEPT. BIOLOGY, UNIV. CALGARY, CALGARY, ALBERTA T2N 1n4 CANADA. AC-
CEPTED 2 AUG. 1979.

MEANDARRA ORNITHOLOGICAL FIELD STUDY UNIT

In December 1980 the Meandarra Ornithological Field Study Unit (MOFSU) was formally
established as a research group of the University of Queensland, Australia. Active in field
research mainly near Meandarra, 300 km west of Brisbane, MOFSU’s major interest is in
aspects of the behavior and ecology of communally breeding species. Visitors and volunteer
field assistants are welcomed, and although no funding is available, students are accepted
to work on field projects towards an M.Sc. or Ph.D. For further information write to: Dr.
Douglas Dow, Director, MOFSU, Dept. Zoology, University of Queensland, Brisbane, Aus-
tralia, 4067.

INTERNATIONAL COMMISSION ON
ZOOLOGICAL NOMENCLATURE

The following opinion has been published by the ICZN in the Bulletin of Zoological
Nomenclature, Vol. 38, Pt. 1, 26 Feb. 1981: Opinion No. 1168 (p. 69) ''Cacatua ducorpsii
Pucheran, 1853 (Aves): conserved.” The Commission cannot supply separates of Opinions.

Wilson Bull., 93(1), 1981, pp. 42-53

THE MAYFIELD METHOD OF ESTIMATING
NESTING SUCCESS: A MODEL, ESTIMATORS
AND SIMULATION RESULTS

Gary L. Hensler and James D. Nichols

Mayfield (1960, 1961, 1975) proposed a method of estimating nesting
success which removes potential sources of bias often associated with
other estimates of this parameter. Despite the intuitive appeal of May-
field’s method and the general recognition that it is appropriate (e.g.. Mil-
ler and Johnson 1978, Custer and Pitelka 1977, Johnson 1979), it is
still not widely used. In this paper we present a probabilistic model for
the experimental situation considered by Mayfield (1960, 1961, 1975). We
then obtain maximum likelihood estimators based on this model and pre-
sent results of Monte Carlo simulations designed to evaluate the esti-
mators. Sample size considerations are also discussed.

Mayfield’s method is based on the concept of “nest days.” The model
he employs assumes the following: (a) the complete period to success,
which we will call the nesting period, (for example, the period of incubation
of eggs) is the same number of days, say J days, for all nests; (b) there is
a constant unknown probability, p (0 < P < 1), over this period that a nest
observed on day j will survive to day j + 1; the probability of a nest suc-
ceeding from day 1 to full term is then p-^; (c) there is a fixed unknown
probability, that an observed nest will have been first found on day j
of the nesting period of J days (for j = 1, 2, . . . , J).

Assume that we observe K nests under the above model. For each of
these nests we observe a random vector = (Y^, TJ, k = 1, 2, . . . , K,
where (i) is a random variable taking the value 1 if the k‘*^ nest is
successful (i.e., survives the complete nesting period) and the value 0 if
the nest fails at any time, and (ii) is a random variable denoting the
number of days the k**^ nest is observed until it either succeeds or fails.
For example, an observation of (0, 10) would mean a nest was seen on 10
days but on the IF^ day visit it had failed, while a value of (1, 10) would
mean a nest was seen on 10 days and on the IF^ day was observed to have
succeeded (for example, hatchlings were present on day 11). Given the
random vectors X,, . . . , we wish to estimate p, the daily probability
of survival.

To do this we consider the joint distribution of (Y^, TJ:

(1) f(y, t|p) = [0j-,+,p‘P'[p'“‘(l - p) 2

L j=i J

Table 1

Results of Monte Carlo Simulations Designed to Evaluate the Maximum Likelihood Estimators for Daily Survival

Probability and Its Variance^

Hensler and Nichols • A TEST OF THE MAYFIELD METHOD 43

(On

(ON

(On

,-H

r-H

£ 2

-o fc

•rf<

c O
o w

<On

(ON

Cfl

l-S

<ON

(On

<0N

(ON

<o.

rj*

<o-

(On

(ON

Tf<

(On

(ON

(On

(ON

l-H

1— 1

l-H

1 — 1

(ON

alues

Tf-

(ON

^ 1

■Ct

(—1

(On

■rf<

(On

(ON

"(t

(ON

(— (

(ON

Actual confidence

Mean estimated values interval coverage

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

d d r-H

CO 00 fO
O CvJ Tf-
0\

o r- 00

CO vO LO
0^ On 0^

CO 00
lO lO lO
lC lO lO

00 nO lO
lO lO LO

o o o
o o o

NO NO fO
NO 00 fO

^ o o
o o o

fO CO fO

d d d

00 ON o

Tt UO

On On On

On On §

d d d

^ lO
CO lC NO
O' O' On

d d d

lO lO NO
O' 00
O' O^ On

fO Cn| O
fO CO CO

o o o
o o o

lO Tt LO
CSJ o CM

^ o o
o o o

CO ^

CO Tf- Tt
LO LO LO

o o o

O O' O'

d d d

(O'

r’

CTn

'—1

(O'

■rf<

'-H O O
ON o o
On On O'

CO O O
ON o o
OC CO o

CO o o
Tt -rj- lO
O' O' O'

00 o o

NO NO
On O O'

d d d

Actual confidence

Mean estimated values interval coverage

Hensler and Nichols • A TEST OF THE MAYFIELD METHOD

(On

(ON

•rj'

Tj<

r—

(On

8 8

-N

<>>

<>-

(On

Tf-

<Q.

r--

r-;

1— 1

t B ~

" " =
£ "c 'I

.Cl. ^ ~
-o a £

-o -c

E I

C 0..

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

for y = 0, 1 and t = 1, 2, . . . , J.

The log likelihood function (see Cramer 1946:498-506) for our random
sample is:

(2)

logfl

k=l

U-T.+ l

+ 2 (T. - D(1 - YK)log p + (k - I; YK)log(l - p)

k=i \ k=i

+ logf]

k=l

J-Tk+l

■- j=l

1-Yk

Differentiating (2) with respect to p, setting the derivative equal to zero
and solving for p, yields the maximum likelihood estimate (m.l.e.) of p,
say p (see Cramer 1946:498-506). Here we have

(3)

i T, + 2 Y. - K

k=l k=l

I T,

k = 1

Mayfield (1960, 1961, 1975) proposes the following estimator for p:
Count the total number of nest days observed (i.e., 2] count

the total number of failures (i.e., K — ^ Ykj, and estimate p by

\ U— 1 /

1 - K - S Yk

j-dU which is in fact p, the m.l.e.

ST,

The theory of maximum likelihood yields that the asymptotic distribu-
tion of (P - P) is Normal with mean zero and variance 1/I(p) where
l(p) is the Fisher information and

l(p) = -E

d^log f(Y, T|p)

dp2

(see Cramer 1946:498-506). As usual, E denotes expected value. Thus,
the asymptotic variance of p is

(4) 1 ^ J_ P‘^(l - P)^

K l(p) K ET(1 - pf + (EY - 1)(1 - 2p)

which we can estimate as

Hensler and Nichols • A TEST OF THE MAYFIELD METHOD

(K)

P^d - P)^

T(1 - p)2 + (Y - 1)(1 - 2p)

if Y#1

K T

if Y = 1.

Here T and Y denote the sample means of T and Y, respectively. Ap-
proximate \ — a confidence intervals for p are then given by

(P - Za/2V, P + Z«/2V)

where Za/2 is the upper a/2 value for the standard normal distribution,

i.e.,

^0/2 1

P=exp(-zV2) dz = 1 - a/2.

-OC ^2t7

Similarly, approximate level a tests for the equality of p values from 2
populations of nests are given by the following: reject Hq, the null
hypothesis that pi = p2, in favor of the alternative hypothesis that
Pi 7^ P2 if and only if

(5)

I Pi - fe I

Za/2 •

The behavior of these confidence intervals and tests depends on the
efficacy of T, Y, and p as estimators of ET, EY and p respectively. To
investigate this behavior we performed Monte Carlo simulations of a nest-
ing experiment which met the assumptions of our model. We chose
several values of J (nesting period), K (the number of observed nests)
and p (the daily survival probability). The 6] probabilities were chosen to
be in proportion to the available number of nests from the day of
nesting given that the same number of new nests are started each day
and only p of them survive to the second day, p^ to the third, etc.; i.e..

we set 6j =

P^-^1 - P)

1 - p^ ’

j = 1, 2,

We randomly divided the K nests into J groups using the distribution
given by the 0^ values. Each nest from the group was then followed
until it survived for J — j -I- 1 days or until it failed. The probability of
daily survival was p, and the probability of full term success for a nest for
J days was p*^. The appropriate (Y,T) vector was recorded for each of the
K nests, and p, v, and estimated confidence intervals (90%, 95%, 99%)
were calculated. It was then determined whether or not the computed

Table 2

Results of Simulations Investigating the Power of the Suggested Test Statistic, |p, - pa]/

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

CSl

Tf-

'tf

i-H

Tj-

>—1

-rf

Tj>

CSl

CNI

— 1

Table 2
Continued

Hensler and Mchols • A TEST OF THE MAYFIELD METHOD

S3 §

odd

lO Tj- CSJ

d d d

r- r— I Os

fO (M o
d d d

2^8
d d d

^ CM

9 R
o o o

8 g 8 8 8 8

os lO
sO

d d

2
d d

2 8
d d

S 8

d d

§83

d d d

o q 8

d d d

o d o
d d d

8 8 8

d d d

8 8
d d d

s s B

d d d

3 8 2

d d o

8 3 8

o
d d

8 3

d d

8 ^
d o

lO fO

d d

oc o
d d

8 8

d d

5RR I I

o o o ^ ^

R R R

8 8 8

8 8 8 3

d d d d

3 3 8^
d d d d

R R R R
d d d d

8 8

o o o o

3 8 8

o o o o

I I

§ §

d d

^ 3
d d

§ 8

o d

-H O

d d

o o o o o o

I I I I
I I I I
I I I I
I I I I

8 § § ^
d d d d

g ::: g §

d d o d

3 8 8 8

d d d d

8 2 8 3

d d d d

8 8 8 3

d d d d

0 o o o

2; g g g

d d d d

1 888

8 8 8 8

d r-l ^ -H

8 3

d d

sO Tf;

d d

3 8

d d

I I I
I I I
I I I
I I I

8 8 3

^ -H d

CM ^ O

d d d

2 3 §
d d d

Os » sO

d d d

8 883 888

d d d d ^ — I ^

—I cc o O O

CM Os Os CC O O O

d d d d — :

§ 8 8 § 8 8 8

O r~: — I d ^ r-l ^

8 888 888

■■c E

2 c ^ ^

■ .£ll‘

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

confidence intervals contained the true value of p. The entire above pro-
cedure was repeated 100 times for each combination of J, p and K values.
We computed the proportion of the 100 times in which the estimated
confidence interval in fact covered the true value. We also determined the
mean, minimum and maximum values of the 100 estimates of p, the mean
value of the 100 estimates v and the mean of the 100 estimates Kv^ (which
should estimate l/l(p)). These values are compared with the theoretical
values, p, l/l(p), and the exact confidence proportions, in Table 1. The
means of the estimated values p and Kv^ appeared to be good estimators
of p and l/l(p) in virtually all simulated cases. In addition, the actual
confidence interval coverages were close to the theoretical values, espe-
cially when it is remembered that proportions represent results of only
100 simulations.

We also calculated p-^ as an estimate of p^ p*^, the probability of full
term success. An alternative estimate of p^ which is commonly used in

nesting studies, is Y = ^ Y^/K, the ratio of the number of observed

k-l

successful nests to total observed nests. The comparisons of p*^ and Y
as estimates of p^ show the superiority of p*’ in cases where the model
assumptions are met (Table 1).

All of the results presented in Table 1 were obtained assuming

= £

i-‘(i - p)

1 - p'’

j = 1, 2,

In order to assess the robustness of the above procedures to changes in
this assumption we set all 0^ equal (i.e., dj = 1/J, j = 1, 2, . . . , J) and
conducted additional simulations for many of the situations examined in
Table 1. Results were virtually identical to those presented in Table 1,
indicating that the estimators are quite robust with respect to reasonable
changes in the 6j values.

In addition, we ran Monte Carlo simulations of tests of equality of two
p values using the test statistic in (5). We assessed both type I and type
11 error probabilities under several experimental situations. These results
are presented in Table 2 and can be used as empirical approximations to
the power of these hypothesis tests under various conditions. It should be
noted that the power curve is not symmetric. Thus, for a specified value
of A a test of the null hypothesis that pi = p2 given that P2 = Pi + A (for
p, > 0.5) is more powerful for A > 0 than for A < 0.

We note that the estimator v^ can be useful in planning an experiment
in which the daily survival probability is to be estimated. If we express
the desired precision of p in terms of a specific coefficient of variation, cv
(where cv = v/p), then we can substitute estimates or guesses for p, ET, EY

Hensler and Nichols • A TEST OF THE MAYFIELD METHOD

Table 3

Sample Sizes (Number of Nests) Needed to Estimate Daily Survival Probability
(p) With Specified Levels of Precision‘s

Nesting
period (J)

Daily

survival

probability

(P)

Desired coefficient of variation (v/p)

0.050

0.040

0.030

0.020

O.OIO

0.005

0.75

Sample

109

size .

245

980

3918

0.85

—

105

419

1675

0.90

—

239

957

0.95

—

104

415

0.85

—

301

1205

0.90

—

154

615

0.95

—

239

0.97

—

130

0.99

—

0.90

—

128

513

0.95

—

178

0.97

—

0.99

—

0.95

—

149

0.97

—

0.99

—

® Sample sizes were computed from (6). Values less than 20 were not presented, because we do not believe it is appropriate
to recommend such small sample sizes. Reasons for this belief are: (I) we question the applicability of results relying on
asymptotic theory to such small sample sizes; and (2) since (6) involves guesses of p, ET and EY, we feel the resulting
uncertainty would never warrant our recommending a sample size of less than 20 nests.

(denote these guesses by p*, T*, Y*) into (4) and obtain the approximate
number of nests, K*, we need to observe:

(6)

(1 - P*f

T*(l - p)^ + (Y* - 1)(1 - 2p*)(cv)2

In the absence of other estimates of ET or EY we may wish to specify
and p to compute EY and ET in the standard manner using (1). As
an example of sample sizes needed to estimate p with various levels of
precision, we have computed values of K* using several reasonable com-
binations of J and p (Table 3). AU values in Table 3 were computed
using ET and EY under the assumption that

^ - p)

j = 1, 2, .

. . , J.

Einally, we wish to indicate some uses of the tables for the field biologist.
Table 1 shows that when the model assumptions are met and the field
biologist uses the approximate confidence interval estimates herein sug-

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

gested, the actual frequency of coverage is quite close to the theoretical
in a wide range of situations. In any one case, however, the estimate of
p (and hence also of p*^) can be considerably different from the true value
especially when the number of nests in the sample is small (see Max p
and Min p in Table 1). Comparisons of p*^ and Y show that p-^ is always a
better estimate of nesting success when the assumptions of the model
obtain, and that the difference in these two estimates is greater in cases
of lower overall nesting success.

The precision of the Mayfield estimator in a field situation is of course
dependent on how closely the assumptions of the model are met by the
population in question. It is doubtful whether this model (or any other
probability model of a biological phenomenon for that matter) will reflect
exactly the reality of nature. However, the traditional estimator Y is almost
sure to overestimate nesting success in all situations where nests are found
on other than the first day of the nesting period (see for example, Mayfield
I960, 1961, 1975; Custer and Pitelka 1977). If the assumptions of this
model approximate the reality of a population, then we suggest its use to
correct for the obvious, known bias associated with Y. In cases where this
model seems totally inappropriate we know of no way to accurately esti-
mate nesting success if nests other than first day nests are to be used.

Table 2 gives empirical estimates of the power in testing the null hy-
pothesis that Pi = p2 against the alternatives that P2 = Pi + A. This table
gives, for selected values of A, the probability of rejecting the null hy-
pothesis given that in fact P2 = Pi + A. Note that for A 7^ 0 rejecting the
null hypothesis is the correct decision, and hence we would hope the
probability of rejecting would be large. For A = 0 this probability should
be the level of significance a for the test. Table 2 shows how the power
function varies with changes in A, Pi , J, and K. A more detailed discussion
of the concept of the power of a statistical test can be found in Cohen
(1977).

Table 3 is a guide for the field biologist to determine the number of
nests needed in his or her sample in order to achieve a given precision in
the estimator. We feel that a sample size of at least 20 nests is needed in
all cases (our reasons are given in Table 3) so only calculated sample sizes
greater than 20 are presented. As mentioned, the biologist must first make
guesses of p, EY and ET or of p and dj, j = 1, . . . , J, in order to calculate
the sample size required for a specified coefficient of variation using equa-
tion (6). Table 3 covers several cases, but direct calculation using (6) is a
simple matter in cases not covered in the table.

SUMMARY

Using a nesting model proposed by Mayfield (1960, 1961, 1975) we show that the estimator
he proposes is a maximum likelihood estimator (m.l.e.). M.l.e. theory allows us to calculate

Hensler and Nichols • A TEST OF THE MAYFIELD METHOD

the asymptotic distribution of this estimator, and we propose an estimator of the asymptotic
variance. Using these estimators we give approximate confidence intervals and tests of sig-
nificance for daily survival. Monte Carlo simulation results show the performance of our
estimators and tests under many sets of conditions. A traditional estimator of nesting success
is shown to be quite inferior to the Mayfield estimator. We give sample sizes required for a
given accuracy under several sets of conditions.

ACKNOWLEDGMENTS

Jim Hines handled most of the programming associated with the Monte Carlo simulations.
Madaline Schumacher and Peggy Bowley typed the manuscript. We wish to thank L. Stokes,
F. Percival, P. Geissler, H. Mayfield and A. Dunham for reading the manuscript and pro-
viding their comments.

LITERATURE CITED

Cramer, H. 1946. Mathematical methods of statistics. Princeton Univ. Press, Princeton,
New Jersey.

Cohen, J. 1977. Statistical power analysis for the behavioral sciences. Academic Press,
New York, New York.

Custer, T. W. and F. A. Pitelka. 1977. Demographic features of a Lapland Longspur
population near Barrow, Alaska. Auk 94:505-525.

Johnson, D. H. 1979. Estimating nest success: the Mayfield method and an alternative.
Auk 96:651-661.

Mayfield, H. 1960. The Kirtland’s Warbler. Cranbrook Inst. Sci., Bloomfield Hills, Mich-
igan.

. 1961. Nesting success calculated from exposure. Wilson Bull. 73:255-261.

. 1975. Suggestions for calculating nest success. Wilson Bull. 87:456-466.

Miller, H. W. and D. H. Johnson. 1978. Interpreting the results of nesting studies. J.
Wildl. Manage. 42:471-476.

PATUXENT WILDLIFE RESEARCH CENTER, LAUREL, MARYLAND 20811 AND
MIGRATORY BIRD AND HABITAT RESEARCH LABORATORY, LAUREL,
MARYLAND 20811. ACCEPTED 15 DEC. 1979.

Wilson Bull., 93(1), 1981, pp. 54-66

NARROWLY DISJUNCT ALLOPATRY BETWEEN
BLACK-CAPPED AND CAROLINA CHICKADEES
IN NORTHERN INDIANA

Peter G. Merritt

The Black-capped Chickadee {Parus atricapillus) inhabits wooded
areas across the northern United States, Canada and high elevations in
the southern Appalachian Mountains. This species is replaced geograph-
ically by the closely related Carolina Chickadee (P. carolinensis) in the
mideastern and southern United States (A.O.U. Check-list 1957). Where
the ranges adjoin, a variety of situations prevail. In Kansas, the western-
most state where the ranges meet, the southern range of P. atricapillus
is contiguous with the northern range of P. carolinensis. Rising (1968)
provided evidence based on multivariate analyses of morphological char-
acters that some interbreeding resulting in hybrid birds may take place
there. Following the boundary eastward, the ranges remain contiguous
through Missouri and into southern Illinois. There, Brewer (1961, 1963)
described a zone of interbreeding where many birds appeared to be hy-
brids. Eastward from Illinois through Indiana and Ohio, Brewer (1963)
reported a narrow gap (about 24 km wide in eastern Illinois) between the
breeding ranges. Ward and Ward (1974) reported contiguous ranges in
southeastern Pennsylvania and they provided evidence based on song that
hybridization may occur there. Johnston (1971) described a hybrid popu-
lation at the contact zone in the mountains of southwestern Virginia, where
the chickadee ranges are separated by elevation. In contrast. Tanner
(1952) found an elevational gap of about 180 m (855—1035) between the
nesting ranges in the Great Smoky Mountains.

The existence of a narrow gap between the chickadee ranges is peculiar
because apparently suitable nesting habitat occurs within the gap regions
(Tanner 1952, Brewer 1963). Furthermore, both species appear to be able
to exist under essentially the same climatic conditions. In the Great Smoky
Mountains where P. atricapillus occurs only on some peaks. Tanner (1952)
found that P. carolinensis nested at higher elevations on the peaks where
P. atricapillus did not occur. In Illinois, Brewer (1963) found that P.
atricapillus and P. carolinensis occurred along parallel river systems at
the same latitude only 32 km apart.

On Mt. LeConte (Great Smoky Mountains), Tanner (1952) observed that
P. atricapillus dispersed to lower elevations in the winter, invading the
range of P. carolinensis. As the nesting season approached most P. atri-
capillus withdrew up the slopes resulting in the formation of the gap.

\lerritt • ALLOPATRY BETWEEN TWO CHICKADEES

Similar observations are lacking for the latitudinal gap, but the hiatus
seems to be most prominent during the breeding season (Brewer 1963).
Brewer (1963) suggested that the gap evolved as a reproductive isolating
mechanism. Since the hybrid population in Illinois had a relatively low
rate of reproductive success (Brewer 1961), selective pressures may have
favored an annual movement of chickadees away from the range interface
in certain regions; this would act to increase the fitness of these birds.
This hypothesis can be falsified by demonstrating that the gap does not
form prior to reproduction.

This paper describes a field study examining the range relationship of
P. atricapillus and P. carolinensis in northern Indiana. The study was
designed to test the prediction that after a dispersal of either species of
chickadee towards the range interface during the winter, these individuals
withdraw from that region, forming a gap between the ranges before re-
production takes place.

STUDY AREA AND METHODS

Fieldwork was conducted in Kosciusko, Wabash and adjacent counties in Indiana. Drain-
age for most of the study area is by way of the Wabash River system which empties south
into the Ohio River. The northern part of the study area consists of gently rolling hills
characteristic of glacial landforms; the southern part is a flat till plain. Around 1820, beech-
maple {Fagus-Acer) and oak-hickory {Quercus-Carya) forests were the dominant vegetation
types in the study area (Lindsey et al. 1965). Today most of the land is farmed, the main
crops being corn, wheat and soybeans. Wooded areas occur along river systems or as isolated
woodlots. Lindsey (1966) gives additional information on climate and other aspects of natural
history of the area.

Preliminary fieldwork indicated that P. atricapillus occurred near the city of Warsaw,
Kosciusko Co., and P. carolinensis occurred along the Wabash River in Wabash County.
Six relatively undisturbed stands of mesic forest were chosen as trapping stations along a 72
km north-south line transecting this region (Merritt 1977). Feeders erected at each trapping
station were filled with sunflower seeds when originally set up, and a constant supply of seed
was maintained until May 1976. Trapping was conducted with 6 McCamey chickadee traps
(McCamey 1961) on at least 4 separate days totaling about 25 h at each station, from 24
December 1975-4 April 1976. Captured chickadees were held briefly for banding, measure-
ment and observation of plumage coloration. The birds were marked with U.S. Fish and
Wildlife Service aluminum bands and from 1-3 plastic color leg bands for individual rec-
ognition.

Captured chickadees were identified on the basis of tail-to-wing ratio and feather color-
ation. Wing chord and tail measurements were taken as suggested by Baldwin et al. (1931,
see Merritt 1978). Inspection of plumage color was limited to the outer edge of the lateral
tail feathers and the lateral edges of the secondary wing feathers. These areas are charac-
teristically distinct and white in P. atricapillus, but are gray and less distinct in P. caroli-
nensis. Notes were taken on the coloration of these feathers as compared with a color chart
consisting of 5 gradations ranging from white to gray. The chart was constructed by reference
to feathers of specimens of P. atricapillus, P. carolinensis and suspected hybrids from
Illinois.

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

TRAPPING STATION

Fig. 1. The number of chickadees banded at each trapping station; P. atricapillus is
shown in solid bars and P. carolinensis in open bars. The relative north-south distance
between stations is represented on the abscissa.

Other distributional information was gained by soliciting vocal responses from chickadees
in non-trapping areas in the winter (27 January-26 February 1976) and spring (29 April-31
May 1976). P. atricapillus typically sings a 2-noted whistle song, fee-bee, which is generally
distinct from the 4-noted song, fee-bee-febay, of P. carolinensis. Less distinct, yet diagnostic
to each species (especially when an on-the-spot comparison with a pre-recorded tape can be
made), is the general call note. P. atricapillus gives this as a relatively slow chickadee-dee-
dee, whereas P. carolinensis gives the same call hut faster.

A portable cassette tape recorder was used to broadcast pre-recorded chickadee songs.
One min of song and call notes for each species was recorded on separate cassettes. Songs
of both species and call notes for P. atricapillus were recorded from Kellogg and Allen
(1971) and call notes for P. carolinensis were recorded from Borror (1970). I located wooded
areas containing seemingly favorable habitat for chickadees. Up to 30 min were spent walking
through each area trying to elicit song responses from chickadees. This was accomplished
by continuously broadcasting the pre-recorded tapes (once a minute I would pause to rewind
the tape and listen for chickadees). During the winter survey I broadcast the atricapillus
tape at areas north of the Eel River, the carolinensis tape at areas south of the Wabash
River and both tapes, interchanging them at 5-min intervals, between the Eel and Wabash
rivers. Because of changes in the chickadee populations found at the trapping stations this
procedure was modified for the spring survey. The atricapillus tape was broadcast at areas
north of the Eel River, the carolinensis tape at areas south of the Eel River and both tapes
at areas along the Eel River. The trapping stations were also included in the spring survey.

Approximately 10 h of observation on 4 separate visits were spent at each of the 6 trapping
stations from 6 April— 22 May 1976. An additional 3 h of observation were spent at trapping

Merritt • ALLOPATHY BETWEEN TWO CHICKADEES

stations 5 and 6 on 25 May. The observation time was spent searching for and following
chickadees.

TRAPPING RESULTS

The winter ranges of the 2 species were found to overlap by at least 20
km (Fig. 1). Thirty individuals of P. atricapillus were banded at trapping
stations 1-5 and 23 individuals of P. carolinensis were banded at stations
3-6. Characteristics of the captured chickadees are described and com-
pared in a separate paper (Merritt 1978); only features necessary for
species identification are included in the following discussion.

The tail-to-wing ratio ranged from 0.846-0.938 (.r = 0.890, SD =
±0.023) in P. carolinensis and 0.919-0.992 (x = 0.955, SD = ±0.016) in
P. atricapillus. The overlap in tail-to-wing ratio (9.4% of all chickadees
captured) is not uncommon; Lunk (1952), Tanner (1952), Simon (1959),
Brewer (1963) and Johnston (1971) have reported slight overlap in the tail-
to-wing ratio of these species.

Accurate classification of plumage color was difficult in the field because
lighting conditions varied, but some differences in coloration were evident.
Four P. atricapillus (13.3%) and 4 P. carolinensis (17.4%) showed plum-
age coloration tending towards intermediacy. These percentages of indi-
viduals deviating from the characteristic forms are within the range of
variation observed by Brewer (1963:16-17) in Illinois.

One P. atricapillus banded at station 2 and 1 P. carolinensis banded
at station 5 were intermediate in tail-to-wing ratio and feather coloration.
Their intermediate morphology may have resulted from interbreeding;
however, it is not likely that hybridization is widespread within the study
area since these individuals represent only 3.8% of all chickadees cap-
tured.

WINTER DISTRIBUTION

The winter survey of vocal responses indicated that the ranges of P.
atricapillus and P. carolinensis overlapped by about 25 km (Fig. 2). Dif-
ferences in time and day of sampling did not seem to affect this survey
significantly; chickadees responded with call notes or song at all 58 areas
sampled. The distributional map in Fig. 2 may be biased to some extent
by the choice of songs broadcast at a particular area. However, this bias
is probably minimal since on several occasions an individual of 1 species
responded to the taped broadcast of the heterospecific song (Merritt 1978)
and the amount of range overlap detected (Fig. 2) corresponds with that
found by the trapping (Fig. 1).

Typical vocal responses were elicited at most areas permitting species
identification, but at 1 location an individual chickadee gave the song of

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

Fig. 2. Winter distribution of chickadees as determined by vocal responses. The trapping
station symbols are solid if P. atricapillus was present, open if P. carolinensis was present
and half-filled-in if both species were present.

Merritt • ALLOPATRY BETWEEN TWO CHICKADEES

both species and at 3 other locations abnormal songs were elicited (Merritt
1978). Tanner (1952), Johnston (1971) and Ward and Ward (1974) also
observed individual chickadees singing the songs of both species. Similar
cases of closely related species singing mixed songs have been described
in buntings {Passerina cyanea and P. amoena)^ meadowlarks {Sturnella
neglecta and S. magna), towhees {Pipilo fuscus and P. albicoLlis), tree-
creepers {Certhia familiaris and C. brachydactyla). Old World warblers
{Acrocephalus scirpaceus and A. palustris, Phylloscopus trochilus and P.
collybita, Sylvia communis and S. atricapilla) and wrens {Troglodytes
aedon and Thryomanes bewickii) (Lemaire 1977:227-230). As suggested
by Emlen et al. (1975) for the case of the buntings {Passerina cyanea and
P. amoena), it is likely that this behavior promotes interspecific recogni-
tion and facilitates heterospecific spacing.

The atypical songs given by 2 chickadees consisted of 4 high-pitched
whistle notes of equal tone. The other atypical song consisted of a trill
followed by 3 high-pitched whistle notes. In each case, these songs seemed
to be given with unusually high intensity. Vocal anomalies apparently sim-
ilar to these were also reported in Illinois (Brewer 1961, 1963), Kansas
(Rising 1968), southwestern Virginia (Johnston 1971) and southeastern
Pennsylvania (Ward and Ward 1974). Brewer (1963) and Johnston (1971)

! noted an unusually high percentage of atypical vocalizations near the hy-
brid zones in Illinois and Virginia, respectively. It seems probable that the
vocal anomalies are the result of interbreeding. In this study only 3 indi-
viduals in the 58 areas surveyed responded with abnormal songs. This
suggests, as does the trapping data, that isolated cases of interbreeding
may occur in northern Indiana but a significant zone of hybridization is
not present.

SPRING DISTRIBUTION

The spring survey of vocal responses revealed that the ranges of P.
atricapillus and P. carolinensis were separated by a gap of about 30 km
(Fig. 3). In general, chickadees of both species were sparse throughout
the region surveyed; rarely did more than 1 chickadee in any particular
1 area respond to the taped broadcast. Most chickadees occurred in river
bottom forests or lowland woods along streams. This was especially true
for P. atricapillus; only twice were individuals located in isolated wood-

I lots.

Only 1 P. atricapillus was found south of the Tippecanoe River and
^ only 1 P. carolinensis was found north of the Eel River. Perhaps the most
? striking feature of the gap was the absence of chickadees along the Eel
\ River northeast of trapping station 3. This area offers habitat similar to
J that existing along the Elkhart, Tippecanoe and Wabash rivers, yet in

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

)

spite of repeated tries at various times on different days no chickadees
could be found here.

TRAPPING STATION OBSERVATIONS

Of the 30 banded individuals of P. atricapillus, only 9 were seen past
the middle of March and only 2 were observed after 1 April (Fig. 4). One
of these (at station 2) was seen on both 8 and 15 April accompanied by an
unbanded individual. Of the 23 banded individuals of P. carolinensis, 16
were observed in April and 4 individuals (at stations 4, 5 and 6) were seen
near the end of May. During the observations at stations 5 and 6, typical
carolinensis songs were commonly heard and a few instances of what
appeared to be territorial combat were observed.

The only other sightings of chickadees at any of the trapping stations
during April and May were as follows: On 3 April, 1 followed 3 unbanded
chickadees from 16:00-16:10 as they foraged from the center of station 2
to the north edge of the woods. At least one of the birds gave the 2-noted
whistle song characteristic of P. atricapillus several times, but the song
was given softly. When they reached the edge of the woods they hesitated
for a minute, then flew north over a corn field out of sight. On 13 April
and 16 May, I observed an unbanded chickadee foraging in woods along
the Eel River about 0.4 km north of station 3; no vocalizations were heard.
On 25 May, I saw a group of 4 chickadees as they foraged slowly through
some dense vegetation at station 6. One bird wore a band and appeared
to be a parent leading young.

DISCUSSION

Comparison of the spring distribution (Fig. 3) with the winter distribution
(Fig. 2) makes it evident that most P. atricapillus withdrew from the range
interface. Observations at the trapping stations indicate that this took
place primarily during the last 2 weeks of March (Fig. 4). The withdrawal
included P. atricapillus occupying woodlots in Elkhart County as far as
80 km north of the range interface. Interestingly, most of the chickadees
found in this region were in riparian habitat. In Kalamazoo County, Mich-
igan, 175 km north of the range interface, P. atricapillus is common in
woodlots as well as riparian habitat at all times of the year (pers. obs.). It
seems likely that in northern Indiana chickadees were found mainly in
riparian habitat because the only continuous stretches of woodland habitat
occur along the river systems and these areas act as avenues of dispersion
for the southernmost breeding P. atricapillus.

Seasonal migration in P. atricapillus has been discussed by Butts
(1931), Wallace (1941), Bent (1946) and Lawrence (1958). Examination of
unpublished banding records obtained in December 1978 from the U.S.

Merritt • ALLOPATHY BETWEEN TWO CHICKADEES

Fig. 3. Spring distribution of P. atricapillus and P. carolinensis. (Refer to legend in Fig. 2.)

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

TRAPPING STATION

■

■ "
■

■ ■

■

□

■

r-i

□

■

r-i r-i

□

□—

— O
Q— O

JAN

FEB

MAR

APR

MAY

MONTH

Fig. 4. Summary of captures and observations at the trapping stations. Squares indicate
the bird was captured and circles indicate it was observed. P. atricapillus is represented by
solid symbols and F. carolinensis by open symbols.

Fish and ildlife Service indicates that long distance movements of P.
atricapillus are most common in a northeast-southwest direction through-
out the eastern and midwestern U.S. and southeastern Canada. Since
numerous handing studies show that many individuals of P. atricapillus
are permanent residents in this region (e.g., Odum 1942, Glase 1973,

Merritt • ALLOPATHY BETWEEN TWO CHICKADEES

Weise and Meyer 1979), it appears that during the winter the gap between
the ranges may become occupied by individuals of P. atricapillus from
the northeast. The gap becomes apparent only after the withdrawal of P.
atricapillus in the spring. Observations in Ohio by Thomas (1958) also
agree with this conclusion.

The disjunct nature of the breeding ranges of P. atricapillus and P,
carolinensis appears to be relatively stable; the gap has remained at ap-
proximately the same latitude in northern Indiana for at least 40 years.
From 1935-1939 Ernest M. Shull frequently went on early morning walks
from Manchester College along the Eel River, upstream about 3.2 km to
the town of Liberty Mills and then back. This area lies only a few kilo-
meters northeast of trapping station 3. ShuU (unpubl.) recorded sightings
of all birds along this route. Downy Woodpeckers {Picoides pubescens),
Tufted Titmice (P. bicolor) and White-breasted Nuthatches {Sitta caro-
linensis) were common during every month of the year. Chickadees (most
reported to be P. atricapillus)^ however, were found in January through
April and September through December, but none in May or July and only
1 in each of June and August during these years.

It is still not clear what factors cause the gap to occur. The fact that
the elevational gap found by Tanner (1952) in the Great Smoky Mountains
and the latitudinal gap found in northern Indiana both occurred only during
the breeding season and became obvious after a withdrawal of P. atri-
capillus suggests that similar factors maintain the gap at both locales.
Considering ultimate factors. Brewer (1963) suggested that the gap evolved
as a reproductive isolating mechanism functioning to reduce unsuccessful
interbreeding. No active chickadee nests were found in this study so to
test Brewer’s reproductive isolation hypothesis it is necessary to estimate
the time of reproduction. The expected date of first egg-laying by chick-
adees at the range interface in northern Indiana can be approximated by
adjusting data for chickadees nesting in Illinois. Both species tend to begin
laying about 3.5— 4.5 days later for each degree of latitude northward
(Brewer 1961), so the first egg-laying at 41°N latitude should occur from
20 April-26 May for P. atricapillus and 30 April-21 May for P. caroli-
nensis. Nest-building, excavation and the establishment of territorial
boundaries precede the first egg-laying by about 20 days. Therefore, chick-
adees remaining in the study area might engage in reproductive activities
as early as 1 April for P. atricapillus and 10 April for P. carolinensis. The
main withdrawal of chickadees away from the range interface took place
prior to these dates so the prediction that the gap forms before nesting
activities begin is supported. This, plus the fact that only a few possible
hybrid individuals were found in this study, indicate that the gap functions
as a reproductive isolating mechanism, but it is not clear that the gap

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

evolved for this purpose. If an individual chickadee’s fitness can be in-
creased by migrating to another area to breed, then natural selection may
have favored the spring withdrawal of P. atricapillus from the range in-
terface; however, this hypothesis remains speculative.

Two other hypotheses attempting to explain narrowly disjunct allopatry
in general, and thought possibly to apply to the case of the chickadees,
rely chiefly on proximate factors. Cornell (1974) suggested that parasites
might be transmitted between chickadee species in the overlap zone. If
these had an adverse effect on the reproductive success of chickadees
breeding at the range interface a gap could develop. Expanding on
MacArthur’s (1972) model of exploitative competition, Slade and Robert-
son (1977) suggested that a change in resource availability as a result of
the cost of interspecific competition in the overlap zone may have caused
a competitively-induced gap. These hypotheses do not assume that the
withdrawal of P. atricapillus is related to the maintenance of the gap.
They both require that interspecific contact occurs between chickadees
that are permanent residents at the range interface. If the interactions
severely reduce the fitness of chickadees attempting to breed in this region
a gap might develop. For an hypothesis of this type 2 predictions can be
made: (1) the width of the gap should be dependent upon the degree of
overlap exhibited by sedentary individuals of both species; and (2) the
width of the gap should fluctuate in time according to the rate of recolo-
nization. A gap developing under these circumstances might persist for a
relatively long period of time if the rate of recolonization is low. Therefore,
these hypotheses might only be testable by a long term study monitoring
the distribution and abundance of permanent residents at the range inter-
face.

It is also possible that the gap is caused and maintained by the winter
influx of P. atricapillus. These individuals might provide additional com-
petition for limiting food resources, which may result in decreasing the
winter survival rate of resident chickadees of both species at the range
interface. This might be especially important in forming the gap if hybrid
chickadees are inferior competitors. This hypothesis could be examined
by testing the prediction that year-to-year fluctuations in sedentary chick-
adee populations near the range interface are directly related to the in-
tensity of the winter influx of P. atricapillus.

SUMMARY

During the winter and early spring of 1975-1976 the ranges of Parus atricapillus and P.
carolinensis overlapped hy about 25 km in northern Indiana. Evidence based on morphology
and song suggest that isolated cases of interbreeding may occur. During the last 2 weeks of
March most P. atricapillus withdrew from the range interface. A survey of vocal responses

Merritt • ALLOPATHY BETWEEN TWO CHICKADEES

conducted from 29 April-31 May revealed that the breeding ranges were separated by a gap
of about 30 km. The withdrawal of P. atricapillus included individuals occupying woodlots
as far as 80 km north of the range interface.

It is still not clear what factors cause the gap to occur. The resemblance of the elevational
gap found by Tanner (1952) in the Great Smoky Mountains and the latitudinal gap found in
northern Indiana suggests that the gap is maintained by similar factors at both locations.
Since the gap becomes obvious with the withdrawal of P. atricapillus just prior to the
breeding season, it is possible that these movements were selected for as a reproductive
isolating mechanism.

Other hypotheses attempting to explain the presence of the gap rely chiefly on proximate
factors. The most likely of these hypotheses are: (1) the gap forms as a direct result of
interspecific interactions between sedentary chickadees at the range interface; and (2) the
gap is caused and maintained by the winter movement of P. atricapillus to the range inter-
face. Competitive interactions reducing the fitness of sedentary chickadees in and near the
gap may be important for either hypothesis.

ACKNOWLEDGMENTS

I am grateful to Richard Brewer for providing guidance throughout all phases of this
project. Richard F. Johnston, Julian C. Lee, Oscar T. Owre, James D. Rising and Edwin O.
Willis read drafts of this manuscript and provided useful suggestions. James Bostwick, Neil
Case, James B. Cope, David Filer, Russell E. Mumford, Ernest M. Shull, Paul Steffen and
Rex Watters contributed useful information. Doug and Christine Campbell, John Olson and
John Stiner helped with the fieldwork. This project was completed in partial fulfillment of
the Master of Arts Degree at Western Michigan University, Kalamazoo, Michigan. Funding
was provided by grants from the Frank M. Chapman Memorial Fund of the American Mu-
seum of Natural History and the Western Michigan University Research Fund.

LITERATURE CITED

American Ornithologists’ Union Check-list Committee. 1957. Check-list of North
American birds, 5th ed. Lord Baltimore Press, Baltimore, Maryland.

Baldwin, S. P., H. C. Oberholser and L. G. Worley. 1931. Measurements of birds.
Sci. Publ. Cleveland Mus. Nat. Hist. 2.

Bent, A. C. 1946. Life histories of North American jays, crows, and titmice. U.S. Natl.
Mus. Bull. 191.

Borror, D. J. 1970. Common bird songs (sound recording). Dover Publ., Inc., New York,
New York.

Brewer, R. 1961. Comparative notes on the life history of the Carolina Chickadee. Wilson
BuU. 73:348-373.

. 1963. Ecological and reproductive relationships of Black-capped and Carolina

chickadees. Auk 80:9-47.

Butts, W. K. 1931. A study of the chickadee and White-breasted Nuthatch by means of
marked individuals. Pt. II: the chickadee (Penthestes atricapillus atricapillus). Bird-
Banding 2: 1-26.

Cornell, H. 1974. Parasitism and distributional gaps between allopatric species. Am. Nat.
108:880-883.

Emlen, S. T., J. D. Rising and W. L. Thompson. 1975. A behavioral and morphological
study of sympatry in the Indigo and Lazuli buntings of the Great Plains. Wilson BuU.
87:145-179.

THE WILSON BULLETIN • Vol. 93, \o. 1, March 1981

Gl\SE. J. C. 1973. EcoIog> and social organization in the Black-capped Chickadee. Living
Bird 12:235-267.

Johnston. D. W. 1971. Ecological aspects of hybridizing chickadees {Parus) in Virginia.
Am. Midi. Nat. 85:124-134.

Kellogg. P. P. and A. A. Allen. 1971. A field guide to bird songs of eastern and central
North America (sound recording). Houghton Mifflin Co.. Boston. Massachusetts.

Lawrence, L. DE K. 1958. On regional movements and body weight of Black-capped
Chickadees in winter. Auk 75:415-443.

Lemaire, F. 1977. Mixed song, interspecific competition and hybridization in the Reed and
Marsh warblers {Acrocephalus scirpaceus and palustris). Behaviour 63:215-240.

Lindsey. A. A. (ed.) 1966. Natural features of Indiana. Indiana Acad. Sci., Indianapolis,
Indiana.

, . B. Crankshaw and S. A. Qadir. 1965. Soil relations and distribution map of

the vegetation of presettlement Indiana. Botan. Gaz. 126:155-163.

Ll nk. . A. 1952. Notes on variation in the Carolina Chickadee. ilson BuU. 64:7-21.

MacArthlr, R. H. 1972. Geographical ecology : patterns in the distribution of species.
Harper and Row. Publishers. Inc.. New \ork, New' ^ork.

McCamey, F. 1961. The chickadee trap. Bird-Banding 32:51-55.

Merritt, P. G. 1977. Gap formation: a reproductive isolating mechanism for Parus atri-
capillus and P. carolinensis in northern Indiana. M. A. thesis. estern Michigan Univ.,
Kalamazoo, Michigan.

. 1978. Characteristics of Black-capped and Carolina chickadees at the range inter-
face in northern Indiana. Jack-Pine arbler 56:171-179.

Odum, E. P. 1942. Annual cycle of the Black-capped Chickadee — 3. Auk 59:499-531.

Rising, J. D. 1968. A multivariate assessment of interbreeding between the chickadees,
Parus atricapillus and P. carolinensis. Syst. Zool. 17:160-169.

Simon. S. W . 1959. Occurrence and measurements of Black-capped Chickadees at Monk-
ton. Md. Maryland Birdlife 15:3^.

Slade, N. A. and P. B. Robertson. 1977. Comments on competitively-induced disjunct
allopatry. Occ. Pap. Mus. Nat. Hist., No. 65, Univ. Kansas, Lawrence, Kansas.

Tanner. J. T. 1952. Black-capped and Carolina chickadees in the southern Appalachian
Mountains. Auk 69:407-424.

Thomas. E. S. 1958. The Black-capped Chickadee in central Ohio. Vi heaton Club Bull. 3
(new ser.):8-ll.

W ALL ACE, G. J. 1941. \\ inter studies of color-banded chickadees. Bird-Banding 12:49-67.

ARD, R. AND D. A. \\ ARD. 1974. Songs in contiguous populations of Black-capped and
Carolina chickadees in Pennsylvania. ilson Bull. 86:344-356.

W ELSE, C. M. AND J. R. Meyer. 1979. Juvenile dispersal and development of site-fidelity
in the Black-capped Chickadee. Auk 96:40-55.

DEPT. BIOLOGY. UMV. MIAMI, CORAL GABLES, ELORIDA 33124. ACCEPTED
1 DEC. 1979.

Wilson Bull., 93(1), 1981, pp. 67-76

TOE FUSION IN OSCINES

George A. Clark, Jr.

Although striking differences in extent of webbing between the toes of
nonpasserines are well-known and often used to illustrate adaptation in
birds, the degree of connection between the toes in oscines has received
much less attention. Ridgway (1901-07) in scattered keys and descriptions
commented on integumental fusion of the toes of numerous New World
oscine taxa, but provided little interpretation on possible significance of
the taxonomic variations, apart from their use in distinguishing taxa. Al-
though others (e.g., Rand and Traylor 1953) have occasionally commented
on fusion in oscines, a comprehensive survey of the families is lacking. 1
attempt here to interpret major taxonomic differences in fusion in relation
to systematics and behavioral differences and to indicate problems for
future study.

MATERIALS AND METHODS

I examined study skins of 1941 species of oscines in the collections of the National Museum
of Natural History (Washington, D.C.), American Museum of Natural History (New York),
British Museum (Nat. Hist., Tring) and the University of Connecticut. In addition, I made
more than 175 observations of toe positions in perching or standing for wild or captive birds
representing 30 species. The taxonomic sequence follows Morony, Bock and Farrand (1975).

To examine fusion among the 3 forward toes I used a hand lens or binocular dissecting
microscope, except for species of large size. Degree of fusion of the middle (HI) and outer
(IV) toes is primarily emphasized, but extent of fusion of the inner (II) and middle toes was
noted for 349 species, as discussed below. I selected as a major landmark for comparison
the articulation between the first and second phalanges of the middle toe, located by bends
in that toe, by the plantar flexion creases and, in many cases, by an overlying scute termed
the proximal cap (Clark 1977). Fusion of toes HI and IV is rated low (L) if not reaching
distally to the region of articulation between tbe first and second phalanges of HI, moderate
(M) if reaching that region, and high (H) if extending further distally (Fig. 1). These ratings
are arbitrary divisions of a continuum, but tbe extremes of low vs high fusion are markedly
different. This summary of fusion in 3 categories provides less detail than sometimes given
by Ridgway (1901-07), but is advantageous in facilitating comparisons across a wide range
of taxa.

RESULTS AND INTERPRETATION

In tabulating data on fusion of toes III and IV (Table I), I emphasize
particularly the distribution of extreme differences (low vs high) and the
condition in many genera considered atypical in their assigned families.
Such a summarizing list (Table I) necessarily obscures many finer taxo-
nomic differences. For example, nearly all genera of the Mimidae have

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

Fig. 1. Examples of toe fusion, from left to right: low. House Sparrow {Passer domesti-
cus); moderate. White-breasted Nuthatch {Sitta carolinensis); and high. Red-eyed Vireo
iVireo olivaceus). The arrows mark the approximate level of the joint between phalanges 1
and 2 of the middle toe.

low fusion, but the inclusion of Donacobius extends the mimid range to
the moderate level.

Partial associations exist between degree of fusion and behavior. The
terrestrial larks (Alaudidae) and pipits (Motacillidae) have low to moderate
fusion, whereas many predominantly arboreal Old World families have
moderate to high fusion, e.g., Campephagidae, Irenidae, Dicaeidae, Nec-
tariniidae, Zosteropidae, Meliphagidae, Oriolidae, Dicruridae. Families
that climb on tree trunks or other vertical surfaces often have moderate
to high fusion, e.g., Sittidae, Certhiidae, Climacteridae. Still other Old
World families or subfamilies range from low to high fusion, e.g., Laniidae,
Timaliinae, Sylviinae, Malurinae. Finches, sparrows and buntings
throughout the world (Emherizinae, Cardinalinae, Carduelinae, Estrildi-
dae, Ploceidae) have low to moderate fusion, a level widespread in the
New World 9-primaried assemblage, including Parulidae, Drepanididae
and Icteridae. However, vireos (Vireonidae), including peppershrikes (Cy-
clarhis) and shrike-vireos {Vireolanius), have greater fusion.

Species within a genus are usually similar in the broad categories of toe
fusion used here (Table 1), hut a few Old World genera, the warblers
Cettia and Bradypterus and the hush-shrikes Telophorus, exhibit excep-
tional interspecific differences. Within Cettia, for example, the low fusion
of C. squarneiceps, C. major and C. brunnifrons contrasts with high fusion
of C. fortipes; other species are intermediate. Among the 10 examined
species of Bradypterus, only B. seebohmi has high fusion, the other 9
being low. Telophorus bocagei, T. sulfureopectus, T. olivaceus, T. nigri-
frons and T. multicolor have high fusion in contrast to low to moderate in
T. zeylonus, T. viridis, T. quadricolor and T. dohertyi. In Telophorus,
separation of groups of species by degree of fusion matches taxonomic
units recognized on other characters (Hall and Moreau 1970), but fusion
differences within Cettia and Bradypterus do not parallel taxonomic

Clark, Jr. • OSCINE TOE FUSION

Table 1

Fusion of Toes III and IV in Oscines

Taxa

No. species examined

Fusion*

Alaudidae

L-M

Hirundinidae

L-M

plus

Atticora

Neochelidon

Motacillidae

Campephagidae

M-H

plus

Chlamydochaera

Pycnonotidae

M-H

plus

Spizixos

Hypsipetes

Irenidae

M-H

Laniidae

Prionopinae

M-H

Malaconotinae

M-H

plus

Telophorus

L-H

Laniinae

L-M

Pityriasinae

Vangidae

plus

Hypositta

Bombycillidae

plus

Phainoptila

Dulidae

Cinclidae

Troglodytidae

L-M

Mimidae

L-M

Prunellidae

Muscicapidae

Turdinae

261

L-M

Orthonychinae

Orth onyx

Androphobus

Psophodes

Sphenostoma

Cinclosoma

Eupetes

Melampitta

I frit a

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

Table 1

Continued

Taxa

No. species examined

Fusion'

Timaliinae

L-M

plus

Garritornis

Stachyris

M-H

Rhopocichla

Macronus

M-H

Micromacronus

Timalia

Pteruthius

M-H

Alcippe

M-H

Yuhina

M-H

Panurinae

L-M

Picathartinae

Polioptilinae

Microhates

Ramphocaenus

Polioptila

Sylviinae

128

L-M

plus

Psamathia

Cettia

L-H

Hradypterus

L-H

Acrocephalus

M-H

Hippo! ais

M-H

Hathmocerciis

Macrospheruis

Malurinae

Malurini

M-H

plus

A mytornis

Stipiturus

Acauthizini

L-M

plus

Gerygone

M-H

Moliouini

M-H

Eptliianurini

L-M

Genus incertae sedis

Larnprolia

Muscicapinae

Hradornis

L-M

Melaenornis

L-M

Fraseria

Rhinomyias

L-M

Ficedii la

L-M

Niltava

L-M

Clark, Jr. • OSCINE TOE FUSION

Table 1
Continued

Taxa

No. species examined

Fusion'*

Muscicapa

L-M

Myioparus

L-M

Humblotia

Newtonia

M-H

Microeca

M-H

Peltops

M-H

Petroica

L-M

Tregellasia

Eopsaltria

Philentoma

Poecilodryas

M-H

Peneothello

Pachycephal apsis

Platysteirinae

M-H

Monarchinae

M-H

Rhipidurinae

M-H

Pachycephalinae

M-H

plus

Hylocitrea

Genus incertae sedis

Turnagra

Aegithalidae

M-H

Remizidae

M-H

Paridae

Sittidae

M-H

Certhiidae

M-H

Rhabdornithidae

Climacteridae

Dicaeidae

M-H

Nectariniidae

M-H

Zosteropidae

M-H

Meliphagidae

110

M-H

Emberizidae

Emberizinae

L-M

Catamblyrbyncbinae

Cardinalinae

Tbraupinae

L-M

Tersininae

Parulidae

plus

Zeledonia

Drepanididae

L-M

Vireonidae

M-H

Icteridae

L-M

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

Table 1
Continued

Taxa

No. species examined

Fusion®

Fringillidae

Fringillinae

Carduelinae

L-M

Estrildidae

L-M

Genus incertae sedis

Pholidornis

Ploceidae

L-M

Sturnidae

plus

Buphagus

Oriolidae

M-H

Dicruridae

M-H

Callaeidae

(irallinidae

Artamidae

M-H

Cracticidae

M-H

Ptilonorhynchidae

L-H

Paradisaeidae

M-H

Corvidae

L-M

plus

Platylophus

Crypsirina

M-H

Tern minis

“ Symbols: I. = low, M = moderate, H = hifih.

groups of species. Unfortunately, little has been reported about the use of
the feet in Cettia, Bradypterus and Telophorus; species of the first 2 are
widely noted as difficult to observe as they skulk in brush.

Fusion of toes II and III. — Fusion of toes II and III extends less far
distally than that between III and IV. The relatively few taxa with fusion
of II and III reaching distally to the vicinity of the articulation of phalanges
I and 2 of toe III also have high fusion of toes III and IV, e.g., Vangidae,
Orthonyx, Microbates, Ramphocaenus, certain muscicapid flycatchers,
Hypositta, Certhia, Climacteris, Vireonidae.

Evolution and systematics. — The taxonomic distribution of different de-
grees of fusion including variation within genera, subfamilies and families
shows that evolutionary convergence has been frequent. Among birds as
a whole, and among oscines, high fusion between toes is probably usually
a derived, rather than primitive, condition. However, reduction of high

Clark, Jr. • OSCINE TOE EUSION

fusion remains a hypothetical possibility and might have occurred occa-
sionally. Raikow (1978) suggested that in situations where evolutionarily
primitive and derived conditions are indeterminable, systematists should
use characters phenetically while recognizing that resulting hypotheses on
relationships will be relatively weak. Any systematic suggestions for os-
cines based heavily on similarity of toe fusion would be at best tentative,
but where fusion agrees with other characters in differing markedly be-
tween genera traditionally hypothesized to be closely related, reconsider-
ation of affinities seems warranted, as in certain of the following examples.

The monotypic Chlamydochaera from Borneo is the sole genus of the
cuckoo-shrikes (Campephagidae) with low fusion. Ames (1975) found that
Chlamydochaera was unique among examined campephagids in having a
thrush-like syrinx and concluded that the genus belongs in the thrushes
(Turdidae). The low fusion of the toes is also like that of thrushes. In
addition, Ames found thrush-like syringes in the muscicapine genera Bra-
dornis, Melaenornis, Rhinomyias, Ficedula, Niltava and Muscicapa,
which, unlike most other Old World flycatchers, also have relatively low
toe fusion like that of thrushes.

Harrison (1967) suggested that the babbler Ifrita from New Guinea is
closely related to the blue wren group of Clytomyias, Chenorhamphus ,
Todopsis and Malurus of Australia and New Guinea. I find that all these
genera share a high fusion of toes 111 and IV. In addition, Ifrita has a
ridged culmen like that of Clytomyias and shares the unusual feature of
blue feathering on the head with Todopsis, Malurus and male Chenorham-
phus. Thus, several characters link Ifrita with the malurids, indicating a
possible relationship not reflected in traditional classification. Clytomyias,
Chenorhamphus, Todopsis and Malurus have an unusual gap in the inter-
scapular zone of the spinal feather tract, present also in the Australian
grass-wrens {Amytornis) and emu-wrens {Stipiturus", Harrison 1969); the
condition of Ifrita in this regard has not been reported. As the relatively
terrestrial Amytornis and Stipiturus have low fusion, caution is necessary
in using fusion as a taxonomic character in this group.

Dorst (1960) proposed that the monotypic genera Tylas and Hypositta
from Madagascar belong in the family Vangidae, which is endemic to that
island. The high fusion in all these birds is compatible with his suggestion,
although high fusion also occurs in other families with which Tylas has
often been placed, including bulbuls (Pycnonotidae). In Hypositta, toe
fusion is greater than in typical nuthatches (Ridgway 1904:439; this study),
and Hypositta is thus more like the vangids in this respect. Also from
Madagascar is Newtonia, the sole genus of the Muscicapinae outside the
Australian region having moderate to high fusion. Although superficial
appearances of study skins can be highly misleading concerning evolu-

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

tionary affinities, Newtonia brunneicauda and females of the vangid Cal-
icalicus madagascariensis have some resemblance. Further consideration
of the affinities of Newtonia would be desirable.

The high fusion of Cyclarhis, Vireolanius, Vireo and Hylophilus sup-
ports the idea that these genera constitute a monophyletic group (Barlow
and James 1975, Raikow 1978). Such high fusion is unusual among New
World oscines, being known otherwise from Certhia and 2 genera of swal-
lows, Microbates, Ramphocaenus. Vireos thus differ markedly from most
New World 9-primaried oscines, including warblers, tanagers, blackbirds
and buntings.

Use of the feet. — As relationships between the degree of fusion and use
of the feet are not well understood, I have emphasized here the most
conspicuous taxonomic differences, for associations between structure
and behavior might be most prominent in such cases. My findings support
Riiggeberg’s (1960) conclusions, based on a much smaller sample of
species, that high fusion often occurs in arboreal species and that low
fusion is typical for terrestrial species. Bock and Miller (1959) indicated
that the high fusion of syndactyly in nonpasserines was advantageous in
arboreal perching because the parallel position of the toes applies all the
force of flexion directly against a branch; separated toes would presumably
he mechanically less efficient. In those climbing oscines with syndactyly,
the forward toes are restrained in a roughly parallel orientation that pos-
sibly helps to ensure a secure grasp on vertical surfaces.

My observations of live oscines and of published photographs indicate
that birds with low fusion vary the spread of the forward toes considerably
according to the kind of perch. On the ground or other flat surfaces, these
toes are widely separated, presumably providing a stable base for standing
or moving. However, on horizontal perches of a small diameter relative to
foot size the forward toes are held close together, a position equivalent to
that of syndactyly, with presumably similar advantages. On sharply in-
clined perches of small diameter, toe II of the lower foot is often abducted
from III and IV, which are held close together (Leisler 1972; see also
Willis 1969, 1972); the application of forces in 2 directions against the
perch by toe II, as opposed to III and IV, presumably helps to prevent
the foot from slipping down the perch. Leisler (1972) has provided further
details on the relationships between toe position and body orientation of
small oscines perched on vertical stems.

A lack of absolute associations between degree of syndactyly and use
of the feet prevents the use of structure to predict habits of species not
studied alive. For example, the climbing Black-and-white Warbler {Mni-
otilta varia) has low fusion, like that of allied nonclimbing parulids (Parkes
1978; this study), in contrast to the moderate to high fusion of many trunk-

Clark, Jr. • OSCINE TOE FUSION

climbers in other families. The tree creepers (Certhia), which have highly
fused forward toes and stiffened tail feathers, are climbers. Because Or-
thonyx from Australia and New Guinea have the same structural charac-
teristics they might be expected also to be climbers; however, Orthonyx
actually forage terrestrially, propping themselves on the ground with 1 leg
and stiffened tail while scratching in the litter with the other foot (Zusi
1978). As another example of absence of absolute associations between
toe fusion and degree of arboreality, many arboreal species have only low
to moderate fusion, e.g., kinglets (Regulus) and New World orioles (Ic-
terus). Possibly such arboreal birds with low fusion differ in perching hab-
its from those with high fusion, but evidence is lacking.

Although asynchronous terrestrial gaits (walking and running) are char-
acteristic for terrestrial oscines, and synchronous (hopping) for arboreal
species, there are many exceptions (Clark 1975), and degree of toe fusion
is not absolutely associated with gait, except that walking is apparently
exceptional in oscines with high fusion. Furthermore, no direct association
exists between degree of fusion and the ability to hold food with the feet
(Clark 1973). Relationships between fusion and uses of the feet may not
be apparent in many cases without simultaneously considering many other
aspects of structure and behavior.

SUMMARY

The degree of toe fusion, here summarized for the oscine families, probably has only
limited taxonomic usefulness, but may serve along with other characters to detect genera
possibly needing further systematic study, e.g., Chlamydochaera, Ifrita, .\eivtonia. Terres-
trial species often have lower fusion than do arboreal or climbing species, but there are
important exceptions. Low fusion apparently aids balance on flat substrates, whereas high
fusion facilitates perching or climbing. The large number of arboreal species with low fusion
has not yet been satisfactorily explained.

ACKNOWLEDGMENTS

I thank G. E. Watson, J. Farrand, Jr. and I. C. J. Galbraith for providing access to
specimens in their care. R. L. Zusi gave helpful advice and J. A. Slater made useful sug-
gestions on an earlier version of the manuscript. Miss Mary Hubbard prepared the illustra-
tion.

LITERATURE CITED

Ames, P. L. 1975. The application of syringeal morphology to the classification of the Old
World insect eaters (Muscicapidae). Bonn. Zool. Beitr. 26:107-134.

Barlow, J. C. and R. D. James. 1975. Aspects of the biolog>- of the Chestnut-sided Shrike-
vireo. Wilson Bull. 87:320-334.

Bock, W'. J. and W. Dew. Miller. 1959. The scansorial foot of the woodpeckers, with
comments on the evolution of perching and climbing feet in birds. Am. Mus. Novit.
1931.

THE WILSON BULLETIN • VoL 93, \o. 1, March 1981

Clark, G. A., Jr. 1973. Holding food with the feet in passerines. Bird-Banding 44:91-99.

. 1975. Additional records of passerine terrestrial gaits. Wilson BuU. 87:384-389.

. 1977. Foot-scutes in North American oscines. Bird-Banding 48:301-308.

Dorst, J. 1960. A propos des affinities systematiques de deux oiseaux malagaches: Tylas
eduardi et Hypositta corallirostris. L, Oiseau 30:259-269.

Hall, B. P. and R. E. Moreau. 1970. An atlas of speciation in African passerine birds.
Br. Mus. (Nat. Hist.), London, England.

Harrison, C. J. O. 1967. The apparent affinities of Ifrita. Bull. Br. Orn. Club 87:97-

100.

. 1969. The affinities of the blue wren genus Malurus and related genera: with special

reference to the grass-wren genus Amytornis. Emu 69:1-8.

Leisler, B. 1972. Artmerkmale am Fuss adulter Teich- und Sumpfrohrsanger (Acroceph-
alus scirpaceus, A. palustris) und ihre Funktion. J. F. Orn. 113:366-373.

Moron Y, J. J., Jr., W. J. Bock and J. Farrand, Jr. 1975. Reference list of the birds of
the world. Dept. Ornithology, Am. Mus. Nat. Hist., New York, New York.

Parkes, K. C. 1978. Still another parulid intergeneric hybrid (Mniotilta X Dendroica) and
its taxonomic and evolutionar>- implications. Auk 95:682-690.

Raikow, R. j. 1978. Appendicular myolog> and relationships of the New orld nine-pri-
maried oscines (Aves: Passeriformes). BuU. Carnegie Mus. Nat. Hist. 7:1-43.

R-AND, a. L. and M. a. Traylor, Jr. 1953. The systematic position of the genera Ram-
phocaenus and Microbates. Auk 70:334-337.

Ridgway, R. 1901-07. The birds of North and Middle America. U.S. Natl. Mus. BuU. 50,
Pts. I-IV.

Ruggeberg, T. 1960. Zur funktionellen Anatomic der hinteren Extremitat einiger mitteleu-
rop’aischer Singvogelarten. Z. ^ iss. Zool. 164:1-106.

ILLIS, E. O. 1969. On the behavior of five species of Rhegmatorhina, ant-followdng ant-
hirds of the Amazon Basin. \\ ilson BuU. 81:363-395.

. 1972. The behavior of Spotted Antbirds. Ornithol. Monogr. No. 10.

Zusi, R. L. 1978. Notes on song and feeding behaviour of Ort/jony.v Emu 78:156-

157.

BIOLOGICAL SCIENCES GROUP, UMV. CONNECTICUT, STORRS, CONNECTI-
CUT 06268. ACCEPTED 25 APR. 1980.

SEVENTH INTERNATIONAL CONEERENCE ON BIRD
CENSUS ORK AND EIETH MEETING OE THE
EUROPEAN ATLAS COMMITTEE

An international conference for those involved or interested in bird census and/or atlas
work will convene 8-12 September 1981 at the Universidad de Leon, Spain. For further
information write: Prof. Francisco Purroy, Departamento de Zoologia, Facultad de Biologia,

Leon, Spain.

Wilson Bull., 93(1), 1981, pp. 77-84

ENVIRONMENTAL EFFECTS ON ROOSTING
BEHAVIOR OF CHIMNEY SWIFTS

Richard M. Zammuto and Edwin C. Franks

The Chimney Swift (Chaetura pelagica) is widespread and abundant in
eastern North America. The lengthy migration and long-unknown winter-
ing grounds of this species early attracted attention, but detailed ecological
research on this species began only in the middle of this century (Fischer
1958). Others have focused on responses by swifts to environmental con-
ditions. Ramsey (1970), for example, studied the effect of changing am-
bient temperatures upon internal body temperatures, and Michael and
Chao (1973) showed associations between roosting behavior, time of sunset
and light intensity. Here we relate roosting and other behaviors to several
environmental variables.

STUDY AREA AND METHODS

Our studies were conducted in Macomb, McDonough Co., Illinois. Midsummer daily tem-
peratures range from 18-36°C. The center of the city has several blocks of contiguous 3- to
5-story buildings; the central area is surrounded on each side by about 10 blocks of 1- to 3-
story homes averaging 10 m apart. Because Macomb is surrounded by cultivated farmland
and is the largest city (population 23,000) in a 60 km radius, it holds the major portion of the
Chimney Swift population in the area.

The first Chimney Swifts usually arrive in mid-April. Numbers are low until late April and
early May, when the first large flocks are seen. The city contains a large population of swifts
(1000-3000) until nest-building begins in late May, when numbers decline to fewer than 1000.
During June and July, most chimneys with swifts contain 1 breeding pair with occasional
visitors or nest helpers (see Dexter 1952, 1974). A few chimneys contain flocks consisting of
non-breeding swifts (up to 300) or both a nesting pair and a flock (Zammuto and Franks
1978). Nesting occurs throughout the city from early June to August. Population numbers
peak during September and slowly decrease until mid-October when all the swifts are gone
for the winter.

Roosts of 6 or more individuals were analyzed. We located most roost-sites from a car at
dusk by watching for circling flocks of swifts. Individual roost-sites were studied for several
mornings and/or evenings. Those sites used most by swifts were observed most often.

A photometer placed on top of a car roof facing the open sky was used to measure light
intensity. For each foot-candle (fc) change in light intensity during exit or entrance by swifts,
the time and number of birds (tallied on hand counter) that left or entered a chimney were
recorded. These data were tested using exponential curvilinear correlation and regression
analyses.

The times when the majority of swifts left or entered a roost-site were determined with a
stopwatch and an average time, designated as the time of peak exit or entrance, was computed.
Temperatures and wind velocities were measured 1.5 m above ground near each roost-site.
Cloud cover was estimated to the nearest 5%. These measurements and the times of sunrise
and sunset were examined in relation to the time of peak exit and entrance at the roost-site
using multiple regression and correlation analyses. The mean difference between time of

THE WILSON BULLETIN • Vol. 93, \o. 1, March 1981

peak exit and time of sunrise and between time of peak entrance and time of sunset were
analysed with regard to sky haziness and precipitation using t-tests. Sky haziness indicated
that a fair sky was somewhat hazy, not clear blue. Precipitation was recorded as present or
absent. Throughout this report, means appear with ±1 SD unless otherwise indicated.

RESULTS AND DISCUSSION

The mean number of Chimney Swifts recorded leaving a roost-site (N =

32) was 70 on 203 mornings, and the mean number entering was 83 on 166
evenings. A total of 224 flock departures or entries was observed at the
32 different chimneys. Of the 224, 65% of the departures or entries were
of less than 100 individuals (Fig. 1).

Behavior at departure. — The swifts could usually be heard calling inside
the chimney when we arrived at the roost 0.5 h before daylight. The calling
became louder as daylight approached (about 0.2 fc), and continued during |
exit by swifts. |

On many mornings, 1-10 swifts arrived at a roost-site about 20 min j

before any swTfts left. Sometimes they circled the chimney and then flew^ |

away but more often they entered the chimney. The roosting birds called I

more loudly if any of these swifts entered the chimney, but the calling j

subsided after about 15 sec.

Swifts departed roost-sites singly or in small groups, almost always in
a steady flow at rates of 4-150 per min. Sometimes a 30-min interval
occurred between departure of 2 substantial portions of a flock. '

Morning returns to roost-site. — After swifts left a chimney, some birds i
reentered on about 50% of the mornings. About 18% of 14,144 departing |
birds reentered wTthin 30 min after first departure, and 3% reentered j

between 30 and 60 min. On about 3% of the mornings, more swifts entered !

a chimney than had left it only minutes before.

Swifts often circled the chimney without reentering for many minutes.
When 1 circling swift entered, many other circling birds immediately fol-
lowed in quick succession; the rest circled without reentering for several
more minutes until another bird entered, followed by another portion of
the group.

Weather conditions seemed to affect reentry into the roost. On cold or
rainy mornings, over 90% of the swifts that left a chimney reentered it
within 30 min. On these mornings, some were reentering while others '
were leaving. A shortage of insects in the air may also cause reentry. The
number of insects in flight was likely very high at daybreak but probably ,
declined sharply after sunrise (McClure 1938; Click 1939, 1957). Reduced |
aerial prey may have caused the swifts to reenter the roost-site at sunrise j

(on the average of 11 min after departure) where they remained until later i

in the morning. |

Evening observations of behavior. — Entry into the roost in the evening

Zammuto and Franks • CHIMNEY SWIFT ROOSTING BEHAVIOR

NUMBER OF SWIFTS PER FLOCK

Fig. 1. Size of 224 flocks.

is described in detail by various authors (James 1950, Zammuto 1978).
Often more than 50 swifts per min entered the chimney if light levels
dropped to 0.5 fc with most of the flock still outside. Reentry rate in the
morning was usually lower, so we believe approaching darkness is a major
I stimulus for entering the roost at dusk. However, as reported by Calhoun
(1938), there were many evenings when only a few swifts were seen at any
one time near a roost-site. As soon as they entered, additional small groups
moved into the area and entered the chimney until the whole flock was
: inside.

i Alterations in roosting behavior. — Various human-related activity dis-
rupted swift roosting patterns. Fumes from furnaces, noises from loud

I vehicles, slamming doors and voices sometimes caused the swifts to leave
ji the roost early, or caused circling swifts to disperse from the roost-site.
4 On 60% of the summer evenings loudly calling swifts flying near roost-

ii sites seemingly caused all the swifts in the area to disperse, although most
I usually returned within 3 min.

I On nearly 50% of the evenings, late-arriving individual swifts circled
I the roost a few times but then flew off (Coffey 1936). Koskimies (1950) felt
I; that such behavior in the Common Swift {Apus apus) indicated it was too
•I dark for the swifts to see well enough to enter the roost, and thus swifts
'I overtaken by nightfall spent the night on the wing.

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

Table 1

Monthly Mean Light Intensity (Foot-Candles) When the Swifts Entered the

Roost

Date

No. swifts

No. evenings
observed

Mean ± SE

Light

intensity mode

Sept. 1976

207

0.6 ± 0.35

0.5

Oct. 1976

623

0.7 ± 0.17

0.5

April 1977

401

0.5 ± 0.28

0.5

May 1977

1937

1.8 ± 0.30

0.5

June 1977

2853

2.7 ± 0.23

0.5

July 1977

2474

3.3 ± 0.33

0.5

-Aug. 1977

886

3.5 ± 0.44

1.5

Sept. 1977

1934

1.0 ± 0.14

0.5

Oct. 1977

183

1.1 ± 0.37

0.5

Total

11,498

103 overall x

= 1.7 ± 1.2«

0.6

® Standard deviation of the 9 monthly means shown.

Light intensity. — Morning departure of the swifts was more widely dis-
tributed at very low light intensities than was evening descent. Based on
12,430 birds, 70% left the chimney during light of 0 and 7 fc, while a
similar percentage of 11,498 l)irds entered between a more restrictive 0
and 2 fc.

Light intensity was negatively correlated with the number of swifts leav-
ing the roost each morning [r = —0.87, P < 0.001) and entering the roost
each evening (r = —0.63, P < 0.001). The regression formula for the

Table 2

Monthly Mean Light Intensity (Foot-Candles) V\ hen the Swifts Left the Roost

Date

No. swifts

No. mornings
observed

Mean ± SE

Light

intensity mode

Oct. 1976

634

6.7 ± 0.62

0.5

April 1977

391

8.0 ± 1.56

2.5

May 1977

1668

2.3 ± 0.21

1.5

June 1977

2104

3.7 ± 0.24

2.5

July 1977

2982

7.0 ± 0.46

2.5

Aug. 1977

1900

4.1 ± 0.49

3.5

Sept. 1977

2185

8.1 ± 0.44

7.5

Oct. 1977

394

4.4 ± 0.57

2.5

Total

12,430

105 overall .v

= 5.5 ± 2.2^

2.9

Standard deviation of the 8 monthly means shown.

Zammuto and Franks • CHIMNEY SWIFT ROOSTING BEHAVIOR

Table 3

Equations and Correlations Showing the Effect of Environmental Variables on
THE Mean Time of Peak Exit and Entrance at the Roost-site

Exit Entrance

Independent

variable

Equation®

Coeffi-
cient of
determi-
nation**

No.

morn-

ings

Equation®

Coeffi-
cient of
determi-
nation**

No.

even-

ings

Time of
sunrise
Time of
sunset

Temperature

= -0.37 + l.OX

0.96***

114

Y = 0.40 + l.OX

0.98***

114

(°C)

Wind speed

= 7.1 - 0.072X

0.43***

114

Y = 18.2 + 0.092X

0.32***

114

(km/h)

Percent cloud

= 6.1 - 0.044X

0.05*

113

Y = 20.1 + 0.06X

0.05**

114

cover

= 6.0 - 0.002X

0.02

109

Y = 20.2 + 0.002X

0.01

112

® Where Y is the mean time (24-h clock, CDT) of peak exit from or entrance into the roost-site and where X is the
independent variable listed.

Levels of significance as determined by F tests: *P < 0.05, **P < 0.01, ***P < 0.001.

number of swifts (Y) leaving and entering the roost with respect to each
level of light intensity (fc) was Y = 17. at dawn and Y = 13.
in the evening.

The swifts entered the roost at significantly higher light intensities in
the warmer months (May through August) than in the colder months (April,
September, October) {t = 5.68, df = 7, P < 0.001) (Table 1). In a Texas
study by Michael and Chao (1973), May through August was also when
swifts entered the roost at higher mean light intensities, although in their
study, light intensities at entrance were much higher (ranging between 2.0
and 14.2 fc, x = 5.9 ± 3.7) than the ones we observed (0.5-3. 5 fc, x =
2.0 ± 1.1; for 1977, Table 1). Mean monthly light intensities at which
swifts left roost-sites are shown in Table 2.

Sunrise and sunset. — The times of sunrise and sunset were more closely
associated with the mean time of peak exit or entrance at the roost than
temperature, wind speed, or cloudiness (Table 3).

Swifts left the roost about 11 min before sunrise and entered it about
21 min after sunset (Table 4). There was less than a 10 min difference
among monthly means. In spring and summer in Texas, swifts entered the
roost much sooner after sunset (Jc == 14 ± 2 min, range 12-16 min) (Mi-
chael and Chao 1973) than did swifts in our study for all months.

Temperatures. — Swifts left the roost significantly later and entered it

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

Table 4

Times Before Sunrise and After Sunset When Swifts Left and Entered the
Roost-site at the Fastest Rate

Date

No.

mornings

Monthly mean i
before sunrise

minutes

± se

No.

evenings

Monthly mean minutes
after sunset ± SE

Sept. 1976

9.8

1.8

22.4

± 1.3

Oct. 1976

11.7

1.7

22.6

± 1.7

April 1977

7.7

2.9

21.8

± 0.8

May 1977

17.3

0.9

23.1

± 1.3

June 1977

12.2

1.7

21.9

± 1.2

July 1977

10.7

0.9

18.0

± 2.1

Aug. 1977

12.3

2.7

17.1

± 0.8

Sept. 1977

9.0

-h

1.2

18.6

± 0.7

Oct. 1977

8.9

-h

1.9

Total

114

overall x — 11.1

2.8^

total 113

overall x — 20.7

± 2.4b

“ Standard deviation of the 9 monthly means shown.
.Standard deviation of the 8 monthly means shown.

significantly earlier on colder days (Table 3). Koskimies (1950) found that
A. apus did not leave the roost-site until the air temperature was high
enough for normal numbers of flying insects to be available. Chimney
Swifts may behave similarly. In southern Texas, Click (1939, 1957) found
that the greatest numbers of insects were in the air at 25°C, and that
numbers of insects decreased considerably below 18°C; very few were
flying at temperatures below 15.6°C.

Wind speed, cloud cover and haziness. — With higher wind speeds, the
swifts leave the roost earlier in the morning and stay out later in the
evening (Table 3). Numbers of flying insects may be an important factor;
Click (1939, 1957) found that the abundance of insects in flight decreased
as winds dropped below 8 km/h. Therefore, fewer insects were flying and
presumably less food was available for swifts on calmer days. It may be
more advantageous to be roosting when the food supply reaches some
lower threshold. Neither cloud cover (Table 3) nor sky haziness had sig-
nificant {P > 0.05) effects on the time of peak exit or entrance.

Precipitation. — Rainfall signihcantly delayed the time of peak exit from
the roost (mean exit time on 10 rainy mornings = 0.3 min before sunrise,
mean on 103 fair mornings = 12.5 min before sunrise, P < 0.001). On rainy
mornings, the swifts usually stayed in the chimney until the rain stopped,
but if they did depart in the rain, most quickly reentered. In sporadic rain,
the reentry corresponded to the periods of rain. Rainy weather also delayed
the onset of daily activity in A. apus (Koskimies 1950), probably due to a re-
duction of tbe food supply. If the rain continued for several consecutive

Zammuto and Franks • CHIMNEY SWIFT ROOSTING BEHAVIOR

days, the Chimney Swifts did not return to the roost in the morning after the
first 2 days. They were probably forced to hunt in the poor weather to keep
from starving, or moved elsewhere.

On 5 days. Chimney Swifts circled roost-sites when summer afternoon
thunderstorms threatened. During most storm threats. Chimney Swifts
flew low, resuming their normal clear day flying height when the threat
passed as observed for Black Swifts {Cypseloides niger) (Rathbun 1925).
Between intermittent rain showers, the swifts we were studying usually
flew in small flocks, flying lower than usual. They may have been following
their food supply; Koskimies (1950) found that many insects were washed
down to the ground by rain. Some swifts in our study were repeatedly
observed flying close to lawns when it rained (Zammuto and Franks 1979).
A. apus reduced foraging activity in rain (Koskimies 1950, Lack and Lack
1951).

When it rained. Chimney Swifts entered the roost-site earlier (mean of
8 rainy evenings = 16.3 min after sunset, mean of 104 fair evenings =
20.9 min after sunset, P < 0.05). On most rainy evenings about 10% of
any flock entered the roost 30 min before the rest of the birds. If the rain
stopped after such early entrances some birds left the roost.

James (1950), Bowman (1952) and Fischer (1958) noticed increases, but
Dexter (1966, 1968) noticed decreases in roosting flock sizes during pro-
longed, cool, rainy periods. In our study, swift numbers increased at 10
' roost-sites during 10 such periods. Some flock sizes doubled, reaching 200
' or more birds. Since in the summer months these added birds in our study
were probably not migrants (Zammuto 1978), we postulate that they moved
from smaller roosts. Presumably, loss of body heat would be minimized
j when many swifts formed tight roosting clusters than when only a few

I swifts were present. This would allow the swifts to survive longer without

I food since they stayed in the roost-site during cool, rainy weather. Dexter

(1966) noted that flocks he observed decreased in numbers in inclement
^ weather and he hypothesized that absent birds were roosting in warmer

. chimneys elsewhere. Our experience suggests that they were in a larger

' roost, but we never studied small roosts during rainy periods to see if swift

f numbers decreased.

SUMMARY

Roosting of Chimney Swifts was studied in relation to several environmental variables in
September and October 1976 and April through October 1977. An average of 70 swifts were
recorded leaving and 83 entering 32 different roost-sites on 203 mornings and 166 evenings.
Of 224 flocks, 65% numbered below 100 individuals. About 18% of the swifts that left the
roost-site at dawn returned to it to reenter around sunrise. In the morning, 70% of the swifts
left the chimney during light intensities of 0-7 fc, but in the evenings 70% entered between
a more restrictive 0 and 2 fc. Swifts left the roost-site at 5.5 ± 2.2 fc, 11.1 ± 2.8 min before
sunrise and entered at 1.7 ± 1.2 fc, 20.7 ± 2.4 min after sunset. Light intensity, time of

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

sunrise and sunset, temperature, wind speed and precipitation were all associated with the
time of departure from and entrance into the roost-site. Swifts left the roost-site later and
entered it earlier on colder days, on days with precipitation, and on calmer days. The effect
of sky haziness and cloud cover upon time of peak exit and entrance was not significant. All
the environmental variables reported to affect flying insect abundance were similarly asso-
ciated with swift activity patterns.

ACKNOWLEDGMENTS

Donald Campbell and Robert Reason assisted with the statistical analyses, and Tate Lin-
dahl helped with the computer programming. Julie Zammuto helped in the field, and she
and Judy Zielinski provided valuable secretarial assistance, Charles Collins, Ralph Dexter,
Gary Schnell and Jerrold Zar suggested improvements for earlier drafts.

LITERATURE CITED

Bowman, R. I. 1952. Chimney Swift banding at Kingston, Ontario from 1928 to 1947. Can.
Field-Nat. 66:151-164.

Calhoun, J. B. 1938. 1938 Swift banding at Nashville and Clarksville. Migrant 9:77-81.
Coffey, B. B. 1936. Chimney Swift migration at Memphis. Migrant 7:77-82, 96-98.
Dexter, R. W. 1952. Extra-parental cooperation in the nesting of Chimney Swifts. Wilson
Bull. 64:133-139.

. 1966. Analysis of Chimney Swift returns at Kent, Ohio, in 1964 and 1965, with

notes on a declining nesting population. Bird-Banding 37:120-121.

. 1968. Analysis of the 1966 and 1967 returns of Chimney Swifts at Kent, Ohio. Bird-

Banding 39:56-57.

. 1974. Unusually large numbers of Chimney Swifts at a nest. Bird-Banding 45:365.

Fischer, R. B. 1958. The breeding biology of the Chimney Swift, Chaetura pelagica (Lin-
naeus). N.Y. State Mus. Bull. 368.

Click, P. A. 1939. The distribution of insects, spiders and mites in the air. U.S. Dept.
Agric. Tech. Bull. 673.

. 1957. Collecting insects by airplane in southern Texas. U.S. Dept. Agric. Tech.

Bull. 1158.

James, P. 1950. Spring flocking of Chimney Swifts (Chaetura pelagica Linnaeus) at Cornell
University. Bird-Banding 21:9-11.

Koskimies, J. 1950. The life of the Swift, Micropus apus (L.) in relation to the weather.
Ann. Acad. Sci. Fennica 15:1-151.

Lack, D. and E. Lack. 1951. The breeding biology of the Swift, Apus apus. Ibis 93:501-
548.

McClure, H. E. 1938. Insect aerial populations. Ann. Ent. Soc. Am. 31:504-513.
Michael, E. D. and W. H. Chao. 1973. Migration and roosting of Chimney Swifts in east
Texas. Auk 90:100-105.

Ramsey, J. J. 1970. Temperature changes in Chimney Swifts (Chaetura pelagica) at lowered
environmental temperatures. Condor 72:264-265.

Rathbun, S. F. 1925. The Black Swift and its habits. Auk 42:497-516.

Zammuto, R. M. 1978. Seasonal activity of the Chimney Swift (Chaetura pelagica) popu-
lation in Macomb, Illinois. M. Sc. thesis. Western Illinois Univ., Macomb, Illinois.

AND . 1979. Chimney Swifts apparently feeding near surfaces of lawns and

from trees. Inland Bird Banding 51:72-73.

AND E. C. Franks. 1978. Forty adult Chimney Swifts at an active nest. Bird-Banding

49:278-279.

DEPT. BIOLOGICAL SCIENCES, WESTERN ILLINOIS UNIV., MACOMB, ILLINOIS
61455. (PRESENT ADDRESS RMZ: DEPT. ZOOLOGY, UNIV. WESTERN ON-
TARIO, LONDON, ONTARIO N6A 5B7 CANADA.) ACCEPTED 20 DEC. 1979.

Wilson Bull., 93(1), 1981, pp. 85-92

GENERAL NOTES

Observation of a brood of Sharp-shinned Hawks in Ontario, with comments on
the functions of sexual dimorphism. — The Sharp-shinned Hawk {Accipiter striatus)
exhibits the greatest “reversed” sexual dimorphism in size of any North American bird, with
males averaging 100 g and females 170 g (Mueller and Berger, Auk 87:452-457, 1970). Of
the several, controversial hypotheses to explain the extent of sexual dimorphism in raptors,
that of Snyder and Wiley (Ornithol. Monogr. 20, 1976) is perhaps most amenable to verifi-
cation. Snyder and Wiley have suggested that food stress late in the period of dependency
of the young has acted to select for adults of greatly different body sizes, and thus different
feeding niches. This increases the range of size of prey available to maintain the adults and
their rapidly growing young. Casual observations of breeding Sharp-shinned Hawks and
small bird populations in Ontario led us to question this hypothesis and served as the in-
spiration for this study.

The senior author is preparing an extensive discussion of the hypothesis of Snyder and
Wiley. The purpose of this paper is simply (1) to present our intensive observations of 1
brood of Sharp-shinned Hawks, (2) to compare our results with those of 3 nests studied by
Snyder and Wiley, and (3) to offer alternative explanations for some of the data. Our obser-
vations are unique in the amount of time and effort expended in observations during the
post-fledgling period and in that ours is the first detailed study from eastern Canada, which
appears to be the center of abundance of the species (Bent, U.S. Natl. Mus. Bull. 167, 1937).

We found a brood of 4 Sharp-shinned Hawks (3 females and 1 male) near their nest in
Burpee Township, Manitoulin Island, Ontario, on 22 July 1977. We began observations on
the brood that afternoon and continued daily until noon on 8 August when we had to depart.
Two to 5 observers (usually 3) watched for 6.7 ± 2.67 (SD) h per day with the least observation
(1.95 h) on 2 August, when it rained most of the day. The nest was about 11 m up in a white
spruce (Picea glauca) in a grove dominated by both this species and northern white cedar
{Thuja occidentalis), with scattered small groups of quaking aspen (Populus tremul aides)
and an occasional balsam fir [Abies balsamea), white birch (Betula papyrifera) and white
pine [Pinus strobus). The nest was in the southwest corner of the grove about 35 m from the
western edge and 30 m from the beach of Lake Huron. The grove extended about 130 m
north and 200-350 m east of the nest, the edge was irregular, and a narrow strip of trees
connected the grove to 2 larger forests to the east and north. The area surrounding the grove
is locally called a “prairie,” an open parkland covered with lichens, sparse grasses and some
bare dolomite rock. Individual and small groups of trees, scattered through the prairie,
formed 5-20% of the vegetative cover.

At least 1 observer remained near the young; the other observers were stationed at points
which offered the best view of approaching adults. The young habituated to our presence
rapidly and would tolerate approach to within less than 5 m after the second day of obser-
vations. The adult female also habituated rapidly. She perched quietly near us on a number
of occasions. The adult male flew within 10 m of us several times, but was not observed
perching near us. The young usually occupied rather exposed perches, either in aspens or
dead trees of various species, although well-fed young were occasionally found on secluded
perches. The center of activity of the young, as they grew older, moved northward from the
nest and we moved our observation posts accordingly.

On 28 July, the young appeared to have achieved full feather growth and flight skills
comparable to those of the adults; identification of age was possible only if we saw the
plumage or heard begging calls. Camp (in Platt, unpubl. M.S. thesis, Brigham Young Univ.,
Provo, Utah, 1973) indicates that full feather growth is attained at an age of 38-40 days. We

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

age of young

I^IG. 1. Mean prey deliveries per h by the adults and mean “predatory episodes” per h
involving young. FDF: food deliveries by the female. FDM: food deliveries by the male. FDT:
sum of food deliveries for both sexes. PEY: predatory episodes involving young (see text).

thus estimated the age of our birds as 33 days w^hen we discovered them on 22 July and that
the females had left the nest about 6 days and the male about 9 days before we began
observations.

Identification of prey items was rarely possible, particularly since most prey items were

GENERAL NOTES

Fig. 2. Mean estimated g of food delivered per h by the adults; F: delivered by female, M:
delivered by male, T; sum of the 2.

delivered plucked and remains were rarely found. If we obtained a sufficiently good look at
the prey as it was being carried by the adult, we estimated its size and weight.

We did not attempt to census prey-bird populations in the range of the hawks, but a
subjective judgement is that the populations were about the same as on 2 previous visits to
Manitoulin at the same time of year.

Food deliveries and behavior of the adults. — In our 120.5 h of observation, there were 73
deliveries of food by the adults, 40 by the male, 23 by the female and in 10 cases we could
not determine the sex of the adult. In 34 of the 73 deliveries (male 22, female 10, unknown
sex 2), the prey item was seen sufficiently well, or sufficient remains were found to permit
estimation of size. The 22 such prey items delivered by the male had a mean weight of 26 ±
14.8 g (range 10-80 g). The 10 items delivered by the female had a mean weight of 35 ± 21.5
g (range 10-90 g). The difference in weight of prey delivered by the sexes was not statistically
significant (t-test, P > 0.26, 1-tailed).

To examine changes in the feeding rates by the adults, we divided our 18 days of obser-
vations into 6 three-day intervals, thus insuring a reasonable amount of observation time per
interval (mean 19.9 h, range 14.0-25.9 h). Within a 3-day interval, food deliveries by adults
of unknown sex were assigned to the sexes in proportion to the deliveries by adults of known
sex. Weights of prey items not seen were assigned to the mean weight of prey delivered by
that sex. The male made 0.53/h food deliveries during the first 12 days of study and only
0.18/h during the last 6 days (Fig. 1). The female showed less temporal change in food

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

deliveries: 0.24/h for the first 12 days and 0.16/h for the last 6 days. The estimated weight
of food delivered per h shows similar trends: 13.6 g/h by the male and 7.7 g/h by the female
during the first 12 days and 4.2 g/h by the male and 5.0 g/h by the female during the last 6
days (Fig. 2). There was a precipitous decline in total food deliveries between the end of the
sixth week of age of the young and the beginning of the seventh week: 65% fewer deliveries
per h and 63% fewer g/h. Daily mean deliveries/h for this 6-day period, when the young were
42 through 47 days old were: 1.05, 0.63, 1.54, 0.42, 0.25, 0.32. The dramatic drop from 1.54
deliveries per h to 0.42/h is further intersting because the latter day (3 August) marked the
first easily noticeable influx of migrant passerines (mostly warblers) into the area.

The adult female was observed to loiter in the nest grove on 13 occasions for a mean of
0.83 ± 0.67 h. These 13 occasions of loitering occurred throughout our period of observation
and as late as 7 August. The female may have spent even more time in the nest grove. She
usually flew below canopy for some distance before relinquishing food to the young and then
exited the grove at low altitude, sometimes unobserved. The male either dropped prey to
the young above canopy, or dipped briefly below the treetops. The male was known to loiter
in the grove on only 3 occasions for a mean duration of 0.15 ± 0.08 h. Neither adult showed
any behavior which appeared defensive of the young in our presence.

We observed the female apparently hunting on 2 occasions, about 0.7 and 1.3 km from
the nest. She was once observed carrying food 1.2 km from the nest. The male was never
observed hunting. On the basis of our observations, we suspect that most of the hunting was
done more than 1.5 km from the nest, although prey appeared to be abundant in the im-
mediate vicinity. Chipping Sparrows (Spizella passerina), Yellow-rumped Warblers (Den-
droica coronata) and American Redstarts (Setophaga ruticilla) were observed with fledged
young within 20 m of the nest.

Behavior of the young. — The first predatory efforts by the young were observed on 29 July,
when they were 40 days old and food deliveries by the adults were quite frequent (Fig. 1).
Of the 80 “predatory episodes by young” depicted in Fig. 1, 52 were stoops or obvious
pursuit flights at prey hidden from our view by vegetation. Of these 52, 23 were definitely
unsuccessful, and for the other 29 cases we could not determine the outcome. During the
last few days of our observations, young would often disappear for several hours, apparently
hunting. Our efforts to follow and find such young were rarely successful, probably because
of the extremely cryptic behavior of the birds. Distress calls of potential prey birds account
for 16 episodes. In 2 episodes, young birds were seen apparently eating a tiny item (grass-
hoppers were extremely abundant in the prairie surrounding the nest grove). On 1 other
occasion, 3 young were flushed from the ground by one of us twice in 3 minutes. The young
remained together and their behavior strongly suggested that one had food hut the dense
vegetation prevented us from confirming this. No food delivery by an adult had occurred
during the previous 3.13 h. On 9 occasions, we saw the bird at which the young hawk
stooped. These included 4 pursuits of Gray Jays {Perisoreus canadensis), 2 stoops at small
passerines, 1 at an unidentified warbler, 1 at a \ ellow-rumped Warbler and 1 at a Common
Flicker (Colaptes auratus). The most spectacular pursuit of prey occurred on 7 August, when
we saw the young male suddenly leave a perch about 15 m up in a dead white pine. Accel-
erating rapidly with flicking wing-beats, he intercepted a flying warbler about 80 m away,
missing by less than 30 cm as the warbler dived. The hawk wheeled and plunged after the
warbler which disappeared into a spruce. We did not observe any successful capture of prey
by the young, hut the hawks had ample opportunities when they were not under surveillance
by us. During the last few days of our observations, known positions of individual young
were sometimes more than 1.5 km apart and on our last day 1 young disappeared 2 km to
the north, flying high, and did not return during our last 40 min of observation. It is possible
that we witnessed the departure of 1 young.

GENERAL NOTES

On 8 August, our last day of observations, all 4 young were under surveillance for only
0.7% of the 5.05 h of observation. Three young were under surveillance 29.7% of the time,
2 young 40.1%, 1 young 37.0% and 12.3% of the time no young were seen.

Discussion. — The 65% decline in food deliveries by the adults that occurred between 42-
44 and 45^7 days of age of the young is far too sudden to be explained by a decline in prey
availability or vulnerability. Furthermore, we noted a marked influx of migrant passerines
into the area on day 44, just when the decrease in prey deliveries began. The decrease can
be explained by a reduction in hunting by the adults. Sharp-shinned Hawks become inde-
pendent of the parents at an age of about 7 weeks (Platt 1973, Snyder and Wiley 1976). A
reduction in food deliveries by the adults 6 days before the young become independent and
the continued begging of the young is consonant with the parent-offspring conflict hypothesis
of Trivers (Am. Zool. 14:249-264, 1974). During the “weaning” period the attempts of the
offspring to maximize their inclusive fitness come into conflict with that of the parents. The
young selfishly demand more care than the parents, selfishly, should provide. Davies (Ibis
120:509-514, 1978) has suggested that if parents time the decline in feeding rates in response
to cues from the young, they are open to being “cheated” in the conflict of maximizing
fitness. If, on the other hand, the adults feed for a fixed period irrespective of the performance
of the young in feeding themselves, young would be lost in times of food scarcity. The optimal
strategy would involve flexibility in the behavior of the adults with parental care being pro-
longed when food is scarce. Davies presents the results of several studies which indicate
that the length of the parental care period in birds is affected by the availability of food. We
believe that this is the key to the understanding of the differences in feeding behavior of
adults observed at various nests.

Newton (J. Zool. London 184:465-487, 1978), in summarizing his observations of the Spar-
rowhawk {A. nisus), suggested that the male usually brings as much food to the young as it
can and that the female is induced to hunt chiefly when the efforts of the male do not meet
the needs of the young. Newton noted that female A. nisus often loiter in the vicinity of the
nest, a phenomenon observed by us, and by Platt (1973) in A. striatus. The available data
on the Sharp-shinned Hawks are in complete agreement with Newton’s hypothesis. The male
Puerto Rican Sharp-shinned Hawk of Snyder and Wiley (1976) was able to provide adequate
food for his family, and the female did no hunting (at least not to provide food for the young).
For the spruce-fir brood of Snyder and Wiley and our Manitoulin Island brood, the male
could not provide all of the needs of the young, and the female did some hunting. In the
case of the riparian canyon brood of Snyder and Wiley, which we readily agree was under
conditions of food stress, the female had to do considerable hunting to help feed the young.
The same was possibly true of the oak-juniper brood of Snyder and Wiley, although the data
are limited.

The peak in food deliveries to the Puerto Rican brood of Snyder and Wiley occurred during
week 5 (the week after fledging) with a sharp drop in week 6 and little further decline in
week 7. The spruce-fir brood of Snyder and Wiley showed a peak in week 4, with a consid-
erable and steady decline through week 5 and weeks 6 and 7 (the 2 weeks were lumped
because of limited observations). However, if we look at the amount of food delivered per
young, the spruce-fir brood also shows a peak during week 5 (1 young disappeared at the
end of the fourth week, cause unknown). A recalculation of the data shown in Fig. 2 into
weekly intervals shows that our Manitoulin Island brood received the same amount of food
during the last 3 days of week 5 (21.2 g/h) as they did in week 6 (21.3 g/h) with a sharp drop
to 11.5 g/h in week 7. The lack of a peak in week 5 for our brood might possibly be due to
the fact that we did not begin observations until the last 3 days of that week. Another possible
partial explanation for the differences observed between nests in the temporal pattern of
food deliveries is in the sex ratios of the brood. Newton (1978) found no difference in the

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

Table 1

Per Cent Distribution of Size Classes of Prey Taken and Prey Available

Size class®

Prey taken

124

11.3

33.1

40.3

11.3

0.8

3.2

(Storer 1966, Table 5)

125

1.6

21.6

37.6

23.2

1.6

8.0

Spruce-fir nest
(Snyder and Wiley 1976)
Prey taken

48.5

47.1

4.4

(Table 5)

50.0

Prey available

(Table 17)

14.3

61.0

12.3

5.6

2.9

3.8

Riparian canyon nest
(Snyder and Wiley 1976)
Prey taken

22.2

58.3

16.7

2.7

(Table 5)

20.0

6.7

26.7

20.0

Prey available

(Table 17)

22.0

44.9

7.1

9.5

7.3

9.3

Oak-juniper nest
(Snyder and Wiley 1976)
Prey taken

36.8

63.2

(Table 5)

25.0

Prey available

(Table 17)

14.3

40.7

20.5

12.1

5.1

7.2

Manitouiin Island

18.2

50.0

22.7

4.5

Prey taken

10.0

40.0

10.0

30.0

10.0

® The size classes are those of Storer (1966):

1 = 3.4-8

g. 2 = 8-

-15.6 g, 3

= 15.6-27 g.

4 = 27-42.9 g, 5 =

42.9-64 g.

6 = 64-91. 1 n.

food consumption of nestling male and female Sparrowhawks and attributed the high food
consumption of the smaller males to more rapid development than the larger females. It is
unlikely that ecpial food consumption hy the sexes persists for very long after fledging; we
suspect that males soon consume less than females. Our Manitouiin Island brood consisted
of 3 females and 1 male, the spruce-fir brood of Snyder and Wiley consisted of 2 males and
1 female (after the loss of 1 male), and the Puerto Rican brood consisted of 2 males. Assuming
that the adults were responding to the needs of the young, the sex ratio of the broods thus
relates rather well with changes in food deliveries after the fifth week. If we take the above
factors into consideration, the temporal patterns of food deliveries to the 3 broods discussed
above are basically similar.

In contrast, the riparian canyon brood of Snyder and Wiley showed a slight increase in
food deliveries between week 5 and weeks 6 and 7 (the last 2 weeks were lumped because
of limited observations). This high rate of food delivery at the end of fledgling dependency,
and the considerable participation of the female in providing food, suggest that this brood

GENERAL NOTES

was suffering from food stress and that the parents were responding in accord with the
hypotheses of Davies (1978) and Newton (1978). The riparian canyon nest was studied in
1971 during the worst drought in 50 years, and bird populations were judged to be low
(Snyder and Wiley 1976). Our results show a precipitious drop in food deliveries early in the
seventh week. It is possible that a similar, sudden decline occurred in most of the broods
studied by Snyder and Wiley (1976) hut was not recognized because of the very limited
observations conducted during week 7.

Newton’s (1978) hypothesis suggests that the relatively meager participation of the female
in delivering food to the spruce-fir brood of Snyder and Wdley (1976) was due to adequate
provisioning by the male. Snyder and Wiley suggest “. . . both sexes of adult accipiters hunt
pretty much full time toward the end of the breeding cycle . . .” and suggest that the
relatively low contributions of the female to the provisioning of the spruce-fir brood was due
to her bringing relatively small prey. She also brought fewer prey items (11% of the total)
than did the riparian canyon female (29% of the total). Snyder and Wiley attribute the
difference in the feeding behaviors of the females at the 2 nests to a greater abundance of
prey in the size range preferred by females in the riparian canyon habitat than in the spruce-
fir habitat. We interpret the data in another way. A shortage in prey of the size classes
preferred by the male in the riparian canyon habitat resulted in inadequate food deliveries
by the male and increased participation by the female, while in the spruce-fir habitat, a
reasonable abundance of prey in the size classes preferred by the male resulted in reasonable
food deliveries by the male and the female was not induced to hunt regularly.

We feel that the data of Storer (Auk 83:423-436, 1966) offer the best estimate of prey-size
preferences in sharp-shins. Storer examined the gut contents of 223 sharp-shins of which
82% were taken during migration, a period when one might reasonably expect random avail-
ability of the various size classes. The analysis of 249 prey items reveals that males prefer
size classes 2 and 3 (73% of the total) and females prefer size classes 3, 4 and 2 (82% of the
total). The preference of females for size class 4 is only very slightly greater than for size
class 2 (Table 1).

An examination of Table 1 suggests that the male at the spruce-fir nest of Snyder and
Wiley (1976) was taking larger prey than expected either in comparison to Storer's (1966)
data or the prey available. The female took few prey, and it is not surprising that none were
from the presumably few individuals available in larger size classes. The riparian canyon
male took considerably smaller prey than the males of Storer (1966) or those of any other
brood studied. We believe this is strong evidence of food stress. It appears that the male
was doing the best he could during this “worst drought in 50 years,*' low bird populations
and probably limited breeding by prey-birds. The riparian canyon female brought a larger
proportion of the prey during the fledgling period than at any other nest: 53% (Manitoulin
37%, spruce-fir 22%, Puerto Rico 0%). In view of the high participation in feeding by the
riparian canyon female, and the apparent scarcity of prey, the high proportion of large-sized
prey taken is not surprising. It is possible that the male considerably depressed populations
of smaller birds early during this unusually bad year. The limited data from the oak-juniper
nest suggest that the male was preying on relatively large prey. The 4 prey items taken by
the female are insufficient for comment. The Manitoulin Island male took larger prey than
predicted on the basis of Storer’s data and the same is true, to a lesser extent, for the female.

We believe that our intensive observations on Manitoulin Island, and our interpretation of
the data of Snyder and Wiley, suggest that food stress late in the period of dependency of
the young in Sharp-shinned Hawks is an exceptional phenomenon. We believe it occurred
only in the riparian canyon brood of Snyder and Wiley (1976) during an unusually unfavorable
year. We therefore conclude that food stress during breeding, ameliorated by separate feed-
ing niches of the adults, is an unlikely explanation for the remarkable reversed dimorphism

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

in size exhibited by Sharp-shinned Hawks. It appears that, under most conditions, the female
could contribute whatever prey is needed by the brood without having to be 1.7 times as
large as the male and capturing slightly larger prey.

Acknowledgments. — We thank Mary Tasker for her exceptional hospitality to often unso-
ciable field biologists and James Tasker for aid in the observations. J. A. Feduccia, K. D.
Meyer and R. H. Wiley offered helpful comments on the manuscript. We thank T. J. Cade,
R. T. Reynolds and N. F. R. Snyder for constructive criticisms of an earlier version of the
manuscript; this does not mean that we have followed all of their suggestions or that they
endorse our interpretations. — Helmut C. Mueller, Dept. Zoology and Curriculum in Ecol-
ogy, Univ. North Carolina, Chapel Hill, North Carolina 27514, Nancy S. Mueller, Dept.
Biology, North Carolina Central Univ., Durham, North Carolina 27707, AND Patricia G.
Parker, Dept. Zoology, Univ. North Carolina, Chapel Hill, North Carolina 27514. Accepted
10 Nov. 1979.

Wilson Bull., 93(1), 1981, pp. 92-97

Food deprivation and temperature regulation in nestling Ferruginous Hawks. —

Nestling deaths from heat prostration may occur regularly in some falconiform species (e.g.,
Fitch et al.. Condor 48:207-237, 1946; Nelson, pp. 64-72 in Peregrine Falcon Populations,
Their Biology and Decline, Hickey ed., Univ. Wisconsin Press, Madison, Wisconsin, 1969;
Olendorff, U.S.I.B.P. Rep. No. 211, 1973; Beecham and Kochert, Wilson Bull. 87:506-513,
1975). Beecham and Kochert (1975) concluded that 41% of Golden Eagle (Aquila chrysaetos)
nestling mortality in their Snake River, Idaho, study area was caused by overheating and
observed that young birds in nests with a southern or western exposure are most vulnerable
to heat stress.

The Ferruginous Hawk (Buteo regalis) nests in the semi-arid regions of southern Canada
and the western L^nited States (Olendorff 1973; Tomhack and Murphy, unpubl.). Throughout
the breeding range Ferruginous Hawks nest primarily on rocky outcrops, tops of trees (es-
pecially junipers lJuniperus spp.] and cottonwoods [Populus spp.R and occasionally on sage-
brush {Artemisia spp.) or the ground (Smith and Murphy, Brigham Young Univ. Sci. BuU.
18:1-76, 1973; Oldendorff 1973; W offinden, Ph.D. diss., Brigham Young Univ., Provo, Utah,
1975; Howard and W olfe, J. Range Manage. 29:33-37, 1976; Lokemoen and Duebbert, Con-
dor 78:464-470, 1976; Fitzner et al.. Condor 79:245-249, 1977). Despite the high tempera-
tures encountered in the latter stages of the nesting cycle (e.g.. Smith and Murphy 1973;
Fitzner et al., 1977), apparently shade availability is not a nest-site requirement for the
species. In the Great Basin west of Utah Lake, W offinden (1975) examined 56 nests of
Ferruginous Hawks on rocky outcrops, on the ground, and in trees. Almost half of these
nests were unshaded throughout the day. The nests on steep slopes received some shade
only in morning or afternoon, depending on slope aspect.

Here, we present field data suggesting that underfed Ferruginous Hawk nestlings are
especially vulnerable to heat stress. W e predict that combined effects of inadequate food
provisions and high temperatures may cause much nestling mortality in years when prey
f)opulations are low.

On 15 June 1977. we surveyed Cedar and Rush valleys, Utah Co. and Tooele Co. (elev.
1760-1895 m; 40°00'N, between 111°55'W and 112°35'W), for Ferruginous Hawk nests (for
detailed description of study area see Smith and Murphy 1973). Thirteen light phase nestlings
between ca 5 and 7 weeks old remained in 5 nests, including 2 tree nests in Utah juniper
(Juniperus osteosperma), 2 nests on rocky outcrops on steep (ca 40°), west-facing slopes, and

GENERAL NOTES

1 ground nest on a southeast-facing slope (ca 30°). Using a YSl Tele- thermometer (Model
43E), a YSI black bulb probe and 2 YSI plastic-tipped probes, we measured ambient and
cloacal temperatures of 2 nestlings 5.5-6 weeks old at a rocky outcrop nest from 14:00-17:00
on 24 June. A plastic-tipped probe was inserted 3-4 cm into the cloaca of each nestling,
taped in place and checked each time the nestlings defecated. We erected a low cardboard
barrier, non-reflective and brown in color, to prevent the young from leaving the nest. The
nestlings showed no signs of agitation whenever we left our tent blind and approached the
nest to take temperature readings.

On 23 June 1979, in the Raft River Valley, Box Elder Co., Utah, and Cassia Co., Idaho
(elev. 1400-1700 m; 42°00'N, 113°30'W), we found 3 Ferruginous Hawk nestlings in 2 nests
in Utah juniper (for description of study area see Howard and Wolfe 1976). One nest con-
tained a melanistic bird, dubbed “Othello,” ca 5 weeks old, and the second nest contained

2 light phase nestlings, “lago” and “Desdemona,” ca 5.5 and 6 weeks old, respectively.

Prior to the trip, we calibrated 3 miniaturized temperature-sensitive radio transmitters

(Mini-Mitter Co., Inc., Model T). On 23 June, a transmitter was wrapped in Black-tailed Jack-
rabbit (Lepus californicus) meat and fed to each Ferruginous Hawk nestling, and the 3 birds
were placed on a recently active Ferruginous Hawk ground nest which was unshaded through-
out the day. We hobbled the young so they could not leave the nest to find shelter. Within 30 min
the nestlings no longer struggled against the restraint and settled into position. We monitored
transmitter signals from the nestlings 15 m from the nest by means of a Lafayette 3-Channel
1.5 Watt Receiver (Mini-Mitter Co., Inc., Model Ch receiver) and recorded ambient tem-
peratures with a YSI black bulb probe and YSI Tele-Thermometer (Model 43TD). As indi-
cated by signal strength, the transmitters lodged in the crops of the nestlings until they were
regurgitated in pellets ca 24 h later. Transmitter signals were converted into body temper-
atures as follows: the receiver was switched to 1 of 3 channels and the time required for 100
clicks was recorded on a stopwatch. This time interval was compared to the appropriate
calibration curve. Every 15 min from 13:30-17:00, we took body temperature readings and
paired them with a black bulb ambient temperature reading.

The nestlings were transported in and maintained for 2 nights in individual grass-lined
cardboard cartons. We fed them several times a day with fresh jackrabbit meat,
hamburger and water. Cast transmitters were recalibrated and again fed to the birds. On 24
and 25 June 1979, we placed the nestlings on a recently active Ferruginous Hawk rocky
outcrop nest built on a steep (ca 40°), west-facing slope in Cedar Valley, Utah Co., Utah.
The nest was not shaded in the afternoon and the nestlings were again hobbled. On 24 June,
we measured ambient and crop temperatures from 14:50-18:00. We did not feed the young
on 25 June until 15:00 in the hopes of duplicating the effects of low food availability; black
bulb ambient and crop temperatures were taken only from 12:00-14:00 to avoid the extreme
late afternoon temperatures.

The behavior of the 5-7-week-old nestlings encountered in Cedar and Rush valleys in June
of 1977 strongly suggested that ambient temperatures were uncomfortable, if not stressful.
All of the nests were unshaded for half the day or longer. Three nestlings had wandered
several meters from their ground nest and crouched under a small sagebrush {Artemisia
tridentata). At 1 rock outcrop nest, the 2 nestlings had moved to the shade of a nearby
man-made, small rock shelter. Two young hawks had moved 3-5 m from the other outcrop
nest to lay their heads in the shade of a large rock. The 3 nestlings in each of the juniper
nests were apparently unable to leave the trees. However, they all tended to pant rapidly
and salivate. Angell (Living Bird 8:225-241, 1969) also reported “shade-seeking” behavior
in nestling Ferruginous Hawks 4-4.5 weeks old. This thermoregulatory behavior is obviously
an important means by which nestlings avoid overheating in the weeks prior to fledging.

In 1977, there were few active Ferruginous Hawk nests in the Cedar and Rush valley study

THE WILSON BULLETIN • Vol. 93, \o. 1, March 1981

Fig. 1. Ambient temperatures and cloacal temperatures of 2 Ferruginous Hawk nestlings
vs time of day at Cedar Valley nest, 24 June 1977.

areas (Murphy et al., unpubl.). The primary reason for the decline appeared to be a low
population of Black-tailed Jackrabbits (Murphy et al., unpubl.). This jackrabbit is the
principal food of the Ferruginous Hawk in the Great Basin region (Smith and Murphy 1973;
Smith and Murphy, Raptor Research 13:1-14, 1979), Woffinden and Murphy (1977) showed
a relationship between the annual numbers of Ferruginous Hawk nesting pairs and jackrabbit
abundance. They also noted a high nestling mortality (an average of 49.59?^) in 2 years of low'
jackrabbit density in 1973 and 1974.

In June 1977, we found no fresh jackrabbit remains at any of the nests we examined,
although remnants of small prey items such as kangaroo rats {Dipodomys sp.), lizards (Cne-
midophorus sp., Phrynosoma sp.) and birds (Icteridae) occurred at 2 nests. One or both

GENERAL NOTES

Ambient Temperature °C

Fig. 2. Crop temperatures monitored by telemetry vs ambient temperatures for Ferru-
ginous Hawk nestlings lago (1), Desdemona (2) and Othello (3), June 1979.

parent birds circled overhead at all the nests we visited, indicating that the young were still
attended.

At the rock outcrop nest where we took temperature readings, no fresh food was present
during the 3 days of our visits. At all nests it appeared that young were inadequately pro-
visioned during a time when food consumption rates should be maximal (Olendorff, Condor
76:766-768, 1974). A graph of black bulb ambient vs cloacal temperatures for the afternoon
of 24 June (Fig. 1) suggests the nestlings had difficulty thermoregulating. The sky was partly
overcast that day, and when the sun was exposed, black bulb temperatures at the nest
increased rapidly from 30°-36.5°C. Nestling cloacal temperatures (range 33-37°C) increased
and decreased rapidly with ambient temperature. These body temperatures are lower than
those daytime temperatures reported for other falconiforms (e.g., McNab, Condor 68:47-55,
1966; Fitch, Condor 76:331-333, 1974; Rudeen and Powers, Condor 80:447-449, 1978), pos-
sibly a consequence of both the cloacal temperature (rather than deep core temperature
measurement) and an impaired thermoregulatory capacity. Both nestlings began panting

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

continuously each time their temperatures reached 35°C. Fluctuations in body temperature
of nestling #2 were more extreme than those of #I (F-test NS); nestling #2 was not found
during our last visit on 25 June.

The 3 nestlings from Raft River Valley in 1979 were seemingly well-fed prior to our ex-
periments, judging by the fresh jackrabbit remains. Body temperatures of the 3 nestlings
monitored by telemetry ranged from 39.9-43.6°C (Fig. 2), and black bulb ambient temper-
atures ranged from 29.5-43°C. These body temperatures are higher than those of the 1977
Cedar Valley nestlings and indicate either that there is a difference between temperatures
measured in cloaca and crop, that the 1977 nestlings were suffering from starvation and
could not thermoregulate well, and/or that the 1979 Raft River Valley nestlings were stressed
by heat and restricted movement on the nest. This latter possibility is unlikely, since we
obtained similar crop temperatures for the Raft River Valley nestlings in a resting state in
the laboratory (Tomback and Murphy, unpubl.). However, the 3 initial 23 June 1979 readings
of 42.7°C for lago and 1 initial reading of 43.2°C for Desdemona at the Raft River Valley
nest (Fig. 2) may have been elevated by our handling.

Means and standard deviations (°C) for ambient and nestling body (crop) temperatures for
the 3, 1979 experimental periods are as follows: Raft River Valley, 23 June — T^ 31.4 ± 1.06,
lago 42.5 ± 0.21, Desdemona 42.8 ± 0.25, Othello 42.4 ± 0.24; Cedar Valley, 24 June — T^
39.4 ± 2.74, lago 41.2 ± 0. 11, Desdemona 42.3 ± 0.27, Othello 42.5 ± 0.20; and Cedar Val-
ley, 25 June— T^ 35.1 ± 1.06, lago 40.7 ± 0.72, Desdemona 43.1 ± 0.74, Othello 43.0 ± 0.29.
F-tests indicated that ambient temperature varied more on 24 June than on either 23 June
{F ^ 0.01) or 25 June {F ^ 0.01). Yet, the crop temperatures of both lago and Desdemona
fluctuated most extremely on 25 June, when the nestlings were deprived of food, than on
either 23 June {F = 0.01, both nestlings) or 24 June (F = 0.01, F = 0.05, respectively). The
crop temperatures of lago varied more on 23 June than 24 June (F = 0.05), whereas those of
Desdemona varied much the same. For Othello crop temperature varied similarly on all 3
days. However, the crop temperatures measured for both Othello and Desdemona under
conditions of food deprivation were significantly higher than those temperatures monitored
on either 23 or 24 June ( Mann-Whitney f/-test, F ^ 0.001), even though the ambient tem-
peratures of 25 June were lower than those of 24 June. Body temperatures for lago were
lower on 25 June than on previous days (Mann-Whitney f/-test, F < 0.001), even though
they fluctuated more, lago rarely panted on 24 June, while the other nestlings panted most
of the time and occasionally held their wings out to the sides to facilitate cooling. The high
temperatures on 24 June affected Othello most severely, as the panting and posturing of this
nestling were pronounced and almost continuous. It is possible that Othello's dark plumage
increased his heat burden (Hamilton and Hepner, Science 155:196-197, 1967; Lustick, Sci-
ence 163:387-390, 1969; hut see U ahlsherg et al., J. Comp. Physiol. 126:211-222, 1978).

In conclusion, the temperature fluctuations of the underfed nestlings at the Cedar Valley
nest in 1977 and the significantly higher and more varied body temperatures of the underfed
nestlings in 1979 suggest that food deprived nestling Ferruginous Hawks cannot cope with
heat stress. Reduction in prey probability are known to result in high nestling mortality in
the Ferruginous Hawk (WOffinden and Murphy 1977). Heat prostration may he a major cause
of nestling deaths under such conditions, especially since Ferruginous Hawk breeding
grounds are characterized by high summer temperatures and nests are unshaded for all or
part of the day.

Acknowledgments. — W e thank Dan Johnson, Kent Keller and Philip Murphy for their in-
valuable help in the field and Clayton W hite for Raft River Valley nest information. Fieldwork
in 1977 was funded in part by a grant from the Chapman Memorial Fund of the American
Museum. At that time D. F. Tomback held a postdoctoral fellowship from Brigham Young
I'niversity. Data and manuscript preparation by D. F. Tomback was supported by the De-

GENERAL NOTES

partment of Biology, University of California at Riverside. Keith L. Bildstein, a reviewer,
provided helpful suggestions concerning the manuscript. — DiANA F. ToMBACK, Dept. Zo-
ology and Entomology, Colorado State Univ., Fort Collins, Colorado 80523, AND JOSEPH R.
Murphy, Dept. Zoology, Brigham Young Univ., Provo, Utah 84602. Accepted 10 Jan. 1980.

Wilson Bull., 93(1), 1981, p. 97

Aerial “play” of Black Vultures. — Occasionally I have seen Black Vultures (Coragyps
atratus) engage in playlike, aerial acrobatics at Cerro Verde (13°50'N, 89°38'W; 2000 m
elev.), El Salvador. On 15 November 1971, I was present during a violent “Norte,” a pro-
longed windstorm which may attain gale velocity on mountain tops. I noticed, without heeding
at first, sounds I attributed to a child tooting a musical toy. Then a trio of Black Vultures
shrilled past my head, producing a sound like that of an aeolian harp, caused undoubtedly
by wind passing through the feathers of wings and/or tails.

A dozen or more Black Vultures were soaring in the strong upslope winds on the north
side of Cerro Verde, being carried upwards as much as 5(X) m above the summit. From time
to time one or more birds “peeled off’ to dive precipitiously towards the southeast, the wind
behind them. Some leveled off where I could see them; others continued out of sight, de-
scending more than 700 m. Recordings of their sounds, made at the time, suggest diving
wire-strutted biplanes of the First World War.

One trio was especially notable, diving again and again as a team. I watched them make
some 25 dives, 3 of which ended near me. The birds dove sometimes in V-formation, some-
times in line and attained impressive speeds. They ended the dives in 2 steps; (1) a slight
increase in angle of attack which checked their speed slightly and flattened the dive; and
(2) a sharp increase in attack angle plus spreading and lowering the tail which forced them
into a shallow climb. They then used their forward momentum to circle along the lee side
of the mountain into the upward current again. The leader of a dive also led in the following
ascent but I could not ascertain whether it retained that position in subsequent dives.

I have heard since, under less favorable conditions, the shrill of diving Black Vultures
during strong Nortes. I add only that twice I saw single vultures diving as described above.
I noted that a single bird produces several tones, suggesting that several feathers are in-
volved.

Bent (U.S. Natl. Mus. Bull. 167:29, 1938) described similar diving sounds produced by
courting Black Vultures. Brown and Amadon (Eagles, Hawks and Falcons of the World,
McGraw-Hill, New York, New York, 1968:181) reported sounds like ripping heavy paper as
Black Vultures dive for food. The birds at Cerro Verde were not diving for food and did not
seem to be courting. Brown and Amadon (1968:101) refer to certain otherwise unclassified
aerial maneuvers of falconiformes as communal displays. However, a display implies com-
munication between a sender and an intended observer; this does not apply to lone birds.
I have no notion of the incentive which governed the vultures I watched, but to me the
exuberant quality of their behavior, so unlike our usual impression of the species, had “all
the appearances of play and seemed to serve no other function than the release of pent-up
energy” (Pettingill, Ornithology in Laboratory and Field, Burgess Publ. Co., Minneapolis,
Minnesota, 1970:254). — Walter A. Thurber, Cornell Univ. Laboratory of Ornithology, 159
Sapsucker Woods Road, Ithaca, New York 14853. Accepted 30 Nov. 1979.

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

Wilson Bull., 93(1), 1981, pp. 98-99

The shoulder-spot display in Ruffed Grouse. — Although the “shoulder-spot” display
is widespread in Tetraonidae, Lumsden (Living Bird 9:65-74, 1970) and Hjorth (Viltrevy
7:184-596, 1970) were both unable to find evidence of this display in Ruffed Grouse {Bonasa
umbellus). The shoulder-spot display is most frequently seen in ambivalent (also termed
conflict by Lumsden 1970) situations containing strong elements of fear. The behavior pat-
terns associated with fear are difficult to observe in wild Ruffed Grouse because of the
general wariness of the species and the restricted visibility characteristic of Ruffed Grouse
habitat. These observational difficulties may have resulted in this behavior being overlooked
when Ruffed Grouse were observed in a natural setting.

A colony of captive Ruffed Grouse at the University of Guelph, Guelph, Ontario, provided

Fig. 1. A male Ruffed Grouse displaying a shoulder spot.

GENERAL NOTES

the opportunity for extended observations of grouse behavior. The shoulder-spot display was
observed in many males on numerous occasions. By contrast, no females were observed
using this display.

Copulation is the activity during which the shoulder-spot display has been most frequently
seen in females of other grouse species (Lumsden 1970). However, our captive females rarely
permit normal copulation, and the observation of copulation is very rare in captivity. There-
fore, failure to observe the shoulder-spot display in female Ruffed Grouse may be attributable
to the rarity of copulation by captive hens rather than to the absence of this display in female
Ruffed Grouse.

The shoulder-spot display was most frequently observed in captive males performing what
has been termed the “intimidation” display (Aubin, M.Sc. thesis, Univ. Alberta,
Edmonton, Alberta, 1970) or “upright-cum-ruff’ (Hjorth 1970) display. When a male per-
forming the intimidation display is approached by an observer, the male usually attacks or
retreats within a short time. However, some males are reluctant to do either, leading to an
ambivalent situation. If the ambivalence is sufficiently intense, the male assumes a semi-
upright posture, with all feathers except the crest sleeked. In this posture, he alternately
approaches and retreats from an intruder, with his body held at a slight angle to the intruder.
He may threaten to strike with his bill. Before this strike intention movement is made, the
wings are withdrawn from beneath the contour feathers, and are slightly extended. The
shoulder-spot display is performed just before the wings are extended (Fig. 1).

Close examination of birds performing this display revealed that the shoulder-spot is
formed by exposure of the underwing coverts on the upper surface of the wing, as Lumsden
(1970) thought. From examination and manipulation of the wings of live birds, it does not
seem possible that a simple re-alignment of the underwing coverts could produce a shoulder-
spot of the dimensions seen on many males. The exposure seems to be effected by movement
of the patagial skin, accompanied by repositioning of the feathers. Apparently, the skin is
drawn over the leading edge of the wing, and onto the upper surface. The coverts are then
exposed to form the shoulder-spot. By varying the degree of skin movement and feather
rearrangement, it would be possible for the bird to alter the dimensions of the shoulder-spot
displayed.

Similarities between Ruffed Grouse and other grouse species in the method of effecting
this display, and the context within which it is performed indicate the origin of the shoulder-
spot display is similar in all grouse. These observations support the suggested evolutionary
development, whereby the display is derived from what was originally a flight intention
movement (Hjorth 1970, Lumsden 1970).

Financial support for the maintenance of the grouse colony was provided by the Ontario
Ministry of Natural Resources and the University of Guelph. — Allan Garbutt, Dept.
Zoology, Univ. Guelph, Guelph, Ontario NIG 2W1 Canada. (Present address: 1424 Carlyle
Rd., Calgary, Alberta T2V 2Vl Canada.) Accepted 3 Feb. 1980.

Wilson Bull., 93(1), 1981, pp. 99-103

The agonistic repertoire of Sandhill Cranes. — Detailed descriptions of agonistic dis-
plays are lacking for wild Sandhill Cranes (Crus canadensis). Walkinshaw reports that all
cranes have some of the same aggressive displays (Walkinshaw, Cranes of the World, Win-
chester Press, New York, New York, 1973). Masatomi and Kitagawa (J. Fac. Sci., Hokkaido
Univ., Ser. IV, Zool. 19:834-878, 1975) give a thorough description of agonistic behavior in
the Japanese Cranes {G. japonensis) that facilitates description of such behavior in Sandhill
Cranes. Voss (pp. 63-85 in Eastern Greater Sandhill Crane Symposium, R. D. Feldt, com-

100

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

piler, Michigan City, Indiana, 1977) describes agonistic behavior in captive and wild Greater
Sandhill Cranes (G. c. tabida) on the summer range in Wisconsin. Here we present obser-
vations of agonistic behavior of migratory Greater Sandhill Cranes on their winter range in
Florida and summer range in Wisconsin and year-round observations of the resident Florida
Sandhill Crane {G. c. pratensis) based on approximately 700 h of fieldwork at Paynes Prairie,
Alachua Co., Florida and at the International Crane Foundation, Baraboo, Wisconsin. We
also include the contexts under which these agonistic behaviors were given. Approximately
80% of the birds observed in Florida had been captured (Nesbitt, pp. 299-303 in Proceeding
of the International Crane Workshop, J. C. Lewis, ed., Oklahoma State Univ., Stillwater,
Oklahoma, 1975), individually color marked and their sex and approximate age deter-
mined. Sex was determined by laporatomy or from postures assumed during Unison Calling,
a sexually distinct display. Age was determined from plumage characteristics.

Highly dominant displays are given by individuals that have little or no fear of other cranes.
If 2 highly aggressive cranes confront each other Bill Sparring usually results. This display
begins when the 2 birds approach each other, some preliminary bill jabbing ensues, then
with wings extended, neck feathers erect and bill tips close together (Fig. LA), the birds
vocalize and vault into the air throwing feet and wings forward (Fig. IB). Substantial contact
does not usually occur, but there is a risk of injury from feet or bill. In many instances, this
display is very brief, lasting only 2 or 3 sec and does not progress beyond the initial bill
jabbing. The victor stands his ground while the loser retreats giving any of the 6 escape
postures described by Masatomi and Kitagawa (1975). Archibald reported a similar display
in the Hooded {Grus monacha) (Animal Kingdom 77:19-24, 1974) and White-naped (Grus
vipio) cranes (Animal Kingdom 76:17-21, 1973) as does Walkinshaw (1973) for Sandhills. Bill
Sparring probably is important to the establishment of a dominance hierarchy. Kepler (pp.
177-196 in Proceeding of the International Crane Workshop, J. C. Lewis, ed., 1975) dis-
cusses the occurrence of a linear dominance hierarchy in a flock of 9 captive-reared Whoop-
ing Cranes {G. americana). A similar hierarchical system was noticed in the wild among
adult male Sandhill Cranes and may exist through other social groups of sandhills as well.
During another high intensity agonistic display, the Head Lowered Charge (Fig. ID), the
aggressor rushed quickly at another bird, neck extended, head, neck and body held hori-
zontal, wings usually tight to the body. Sometimes flapping begins as the other bird is
approached, the bill is open, and the aggressor often grabs the other bird by the wing or
scapular feathers. A charge can occur during feeding where the aggressor, apparently feeding
normally, moves closer to the offending individual. Then from the feeding position a charge
erupts catching the other crane by surprise. Walkinshaw (1973) has generally described this
behavior as occurring with all cranes. The Head Lowered Charge often leads to Aerial
Pursuit, especially during the period when the birds are defending nesting territories. During
Aerial Pursuits the aggressor may attempt to strike the fleeing bird with its feet. A dominant
crane will displace a subordinate individual from a feeding or drinking site with a Bill Stab
(Fig. IE) directed at the back or back of the neck with bill either open or closed. The attacked
bird simply moves a few steps, and the dominant bird assumes the feeding or drinking spot.
Charges or stabs from the dominant individual often follow a full Bill Sparring episode.
Following all high intensity agonistic displays, the dominating individual usually gives a Low
Bow display (Masatomi and Kitagawa 1975, Voss 1977) during which the neck is arched and
the head slowly lowered toward the ground displaying an expanded bright red comb.
The display terminates with the head between the bird’s legs and bill held vertically (Fig.
IF). As the head is lowered the bird emits a low growl. A Low Bow is often given by territorial
birds upon landing near an intruder before any other signs of aggression are shown.

During agonistic episodes where the motivation levels are lower, any one of a series of
generalized ambivalent displays may be seen. To drive other cranes from a defended territory

GENERAL NOTES

101

Fig. 1. Common agonistic displays of Sandhill Cranes: (A, B) Bill Sparring; (C) Directed
Walk Threat; (D) Head Lowered Charge; (E) Bill Stab; (F) Low Bow; (G) Generalized Body-
Wing Shaking; (H) Neck Retracted Submissive Posture.

or a feeding area, the dominant individual or pair (and their chick, if present) give a Directed
Walk Threat or Adornment Walking (Masatomi and Kitagawa 1975, Voss 1977) often with
a characteristic vocalization (Slow-rattle Family Call; Nesbitt and Bradley, in press) given
in unison, by all defending cranes. This display is an exaggeration of the normal upright
walking movement directed at a particular individual. The displaying bird circles the op-
ponent with a stiff gait characterized by animated head pumping and tertials slightly raised.
With each step the neck is extended and the bill pointed at the other crane. Periodically,
the bill is angled downward displaying an expanded, bright red comb (Fig. 1C). The Directed
Walk Threat is apparently equivalent to a display described by Archibald (1974) for the
Hooded Crane. The individual toward which this behavior has been addressed usually moves
away quickly with head lowered and body held horizontally. Other adornment displays de-
scribed by Masatomi and Kitagawa (1975) for G. japonensis seem to represent variations in

102

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

the intensity of adornment walking in sandhills. If the offending individual does not retreat
as a result of the Directed Walk Threat a charge or stab may occur. If the aggression level
is not high, one of several ritualized general threat displays may ensue. The Generalized
Body-Wing Shaking (Voss 1977) (Fig. IG), similar to the Low Bow, has been described for
several species of cranes (Archibald 1974; Walkinshaw 1973; Paulsen, Dansk Ornithol. For-
enings Tidsskr. 69:119-122, 1975). But it is less intense and general rather than directed.
Body-Wing Shaking begins as a stylized feather maintenance movement, contains a less
dramatic bow and terminates with displacement preening of the legs or belly. Again a low
growl is given as the head is lowered. In Sandhill Cranes other forms of bowing described
by Masatomi and Kitagawa (1975) appear to be less intense versions of the Low Bow. The
Unison Call may function as a generalized aggressive display (Archibald, Ph.D. diss., Cornell
Univ., Ithaca, New York, 1976). A pair will often Unison Call before aggressive episodes or
after successfully driving intruders from a defended territory. The Crouch Display observed
by Voss (1977) principally in young cranes was noticed among the Florida birds only once,
given by a young crane. In the wild it is perhaps a seldom given display that in adult birds
transmits little or no agonistic information. When 2 cranes with lower aggressive motivation
are close together, they often engage in displacement foraging or preening. These are tran-
sitional behaviors leading to normal feeding or preening and away from aggressive encoun-
ters.

In fearful situations, birds of the year and other subordinant individuals wishing to avoid
attack, assume a Neck-retracted Submissive Posture (Voss 1977) (Fig. IH), during which the
comb is constricted and dull in color. A crane, fearful of attack from another crane, may
feign the precopulatory display with wings spread and beak held above horizontal at a 45°
angle, thus changing the motivation of the attacker. When approached by a mammalian
predator. Sandhill Cranes give a predator threatening Spread-Wing Display (Voss 1977),
consisting of an upright posture with wings held high and half-open. The head is held high
with the beak pointed directly at the predator. If the predator does not retreat, then the bird
approaches, thrusts the bill forward and hisses. A crane was observed successfully fending
off the attack of a juvenile Bald Eagle (Haliaeetus leucocephalus) using bill jabs with vaults
and forward thrusting of the feet and wings. Such movements, similar to Chasing and Kicking
(Voss 1977), are generally equivalent to the movements that have been stylized into Bill
Sparring, but without the characteristic vocalizations. A similar attack behavior and hissing
has been observed during trapping operations when oral tranquilizers were used. Attacks
were directed at tranquilized cranes that did not react normally to unaffected birds. Pre-
sumably this is the type of attack that has resulted in severe injuries to cranes (Walkinshaw,
Michigan Acad. Sci., Arts, Letters 40:75-88, 1965). The attack method described by Altmann
(J. Mammal. 41:525, 1960) employed by a pair of adult cranes with a chick to drive off a
moose (Alces alces) incorporated these same movements.

Aggression was often observed during and after unpaired dancing bouts between non-
breeding cranes. Aggressive attacks were also launched at individuals apparently preoccu-
pied in another behavior such as Bill-Raising (Masatomi and Kitagawa, 1975:Fig. 119), uni-
lateral stretching or sitting.

All of these displays have been observed in both sexes, though Bill Sparring appears more
pronounced in males. Sub-adult cranes exhibited the same aggressive displays, but the
frequency and intensity of the displays were much reduced when compared with paired
males. Among 4 of 5 distinctly marked breeding pairs tbe male consistently took the leading
role in territorial defense. W ithin the fifth pair, the male and female were equally aggressive,
both initiating an equal number of encounters. Females and young of the year commonly
participate in Directed Walk Threats, and occasionally Charges. Other displays appear much
less frequently. The hierarchical position of the pair or family seems to depend on the

GENERAL NOTES

103

position of the male. The aggressiveness and therefore the hierarchical position of a male
relates to the presence of a chick. Pairs without chicks tend to be lower in the order. For
example, the dominant pair of the 5 marked pairs had 1 chick in 1977. In 1978, they were
chickless and were dominated by 2 previously subordinate pairs, both with chicks.

The frequency of agonistic encounters involving adult Florida cranes increased during the
period when young of the previous year were separated from the family group (February and
March). The frequency remained high until just before the eggs hatched, then the level
declined and the number of encounters remained low until several weeks after hatching.
This lowered aggression period corresponds with the period in which Bennett (Auk 95:411^13,
1978) noticed little response from territorial cranes to the play back of tape recorded calls.

We wish to thank K. S. Voss and C. B. Brownsmith for their comments on early drafts of
this manuscript. This study was in part a contribution of the Federal Aid to the Wildlife
Restoration Program, Florida Pittman-Robertson Project W-41. — Stephen A. Nesbitt,
Florida Game and Fresh Water Fish Commission, Wildlife Research Laboratory, 4005 S
Main St., Gainesville, Florida 32601 AND George W. Archibald, International Crane
Foundation, Baraboo, Wisconsin 53913. Accepted 1 Nov. 1979.

Wilson Ball., 93(1), 1981, pp. 103-107

Notes on the Slender Antbird. — The Slender Antbird (Rhopornis ardesiaca), first col-
lected somewhere in eastern Brazil by Prince Maximilian von Wied (Beitrage zur Naturges-
chichte von Brasilien, Vol. 3, 1831), was until recently known from 3 specimens: the male
type, another male from Ituagu, in south-central Bahia, and a female from the town of Boa
Nova just down the Rio de Contas (Naumberg, Bull. Am. Mus. Nat. Hist. 76:231-276, 1939).
Emil Kaempfer collected the last 2 specimens in 1928. Naumberg suggested that Kaempfer’s
“Boa Nova” was another town with the same name, northwest across the Rio Sao Francisco;
but Kaempfer was at the second Boa Nova in 1927, not 1928. Moreover, Wied is likely to
have collected the type near the first Boa Nova, which he passed en route from Vitoria da
Conquista to Salvador.

From 3-9 December 1974, we studied Slender Antbirds in patches of dry forest on Fazenda
Alvorada, just north of the first Boa Nova (14°20'S, 40°H'W). A good, if scattered, population
exists in these patches, which are gradually being cleared for cattle pastures. In 1977, H.
Sick collected a male at Boa Nova after we mentioned our observations to him.

Habitat and foraging. — Boa Nova lies at 700 m elev., below 800-1000 m ridges of the
northern end of a broad plateau that stretches southwest past Vitoria da Conquista nearly
to the valley of the Rio Pardo in the state of Minas Gerais (Fig. 1). This plateau is the main
ridge of southeastern Bahia, forming a border between wet coastal forests (which include
some patches of dry forest in the upper basins of small rivers) and the desert scrub or
“caatinga” of the interior.

The natural vegetation of this rolling plateau varied within short distances from wet cloud
forests (15(K)-2000 mm annual rainfall) on the eastern escarpments to caatinga in such rain-
shadow areas as the lee slopes around Boa Nova, but the summit was mainly a dry forest
(8(X)-1000 mm rainfall) with many “cipos” or lianas — a “mata de cip6.” The scattered re-
maining patches of dry forest have a strange appearance, with scattered white trunks of
small trees above a dense layer of midstory trees and vines. The understory is fairly open,
but blocked here and there by lianas and by patches of huge terrestrial bromeliads [Aechmea
sp.). In the forest, bromeliads tend to sit high on tree trunks; but at the borders between dry

104

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

Eig. 1. Coastal Bahia, Brazil, showing the region inhabited by Slender Antbirds. The
only roads shown are ones traveled by the authors. '

forest and caatinga, or between dry forest and pastures, there is enough light for the hro-
tneliads to descend to the ground. These hromeliads, and surrounding undergrowth near the
forest edge, are the habitat of Slender Antbirds. j

Just inside the forest borders — occasionally in tall scrub nearby — the antbirds hop on the !

ground, low vines and on the tops of hromeliad leaves, pausing now and then to swipe dead j

leaves from atop the hromeliads or from spots on the ground. They wander silently much of j

the time, peering up and down and capturing small grasshoppers or other insects by short j

flights or by hopping down to peck. We found them surprisingly like Gray Catbirds {Du- \
metella carolinensis) of North America in their appearance and actions, although they stay j
on the ground much more. j

Members of a pair wander separately or together, often disappearing among the dense
hromeliads for 15 min or more, only to reappear less than 5 m from where they entered.

Each pair we watched seemed to have a very limited home range, barely 50 m across; but ■

home ranges of pairs were usually separated by 100 m or more because patches of hromeliads i

GENERAL NOTES

105

Fig. 2. Audiospectrographs of Slender Antbird vocalizations: (A) song of male; (B) alarm
chip (logarithmic scale); (C) song of female; (D) faint chirps and grunts of a pair; (E) rattle,
starting with a chip; (F) rattle of another bird.

were seldom close together. Occasionally we found birds hopping on the ground through
fairly open undergrowth between patches of bromeliads, and once 2 such pairs were singing
as they hopped along a presumed territorial boundary. In the center of the main forest tract
of Fazenda Alvorada, we encountered only 1 bird near a bromeliad zone in a treefall clearing.
Maximum densities were in a rather scrubby second-growth woodlot, where pairs were 100-
200 m apart.

We found no Rhopornis following army ants within the main woodlot; there the related
White-winged Fire-Eyes (Pyriglena leucoptera) and White-bibbed Antbirds (Myrmeciza lor-
icata) split the niche of the Slender Antbird. Its niche is essentially that of another slender
and long-legged forest-edge antbird, the White-bellied Antbird (M. longipes), a species that
follows army ants occasionally in Panama, Trinidad and other areas.

Song and vocal behavior. — The simple and loud peer peer peer peer peer peer song of
Slender Antbirds (Fig. 2A, C) is audible up to 500 m from the forest edge, and seems well
adapted for birds that live widely scattered along forest edges in isolated patches of dense
cover. Songs began 15—60 min after the first light and after the first songs of other diurnal
birds, hence between 05:20 and 06:00. Sessions of song were irregular, mostly during the
morning; few birds sang at any time. Songs were rare after 13:00 or 14:00. Often the male
started singing when the female disappeared, she would finally answer him a few times, and
the two would move together again. One male sang off-and-on for nearly 2 h (05:40-07:30)

106

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

until the female appeared. At times, 2 males or 2 pairs sang back and forth for short periods,
mostly in boundary disputes.

When the pair met and foraged close together, they exchanged faint sif or prit and other
notes (Fig. 2D), some rather like the “bubbling” call of White-backed Fire-Eyes. The male
occasionally started a “serpentine-song” of the type common in antbirds: an alternating
series of slow and fast chirps, eu, eu, u-u-u-u-u, eu, eu, eu, eu . . . and so on. One male fed
his mate a small grasshopper after serpentine-singing. Another pair carried small insects to
a patch of bromeliads and rattled repeatedly at the observer; but on other days the female
rattled only near another patch of bromeliads. Otherwise there was no evidence of breeding
activity.

Individual birds varied widely in their reactions to us. Some foraged quietly and tamely,
others quickly fled to dense vegetation and others scolded vigorously. None was as difficult
to observe as are antbirds of the rain forest, perhaps because Slender Antbirds live next to
good cover rather than in open undergrowth. The usual alarm calls are a sharp chipping and
a loud rattling. Chipping is a loud sibilant tsiekl or psiefl (Fig. 2B) as the bird pounds the tail
downward before or after fleeing to cover. At times the call becomes a loud feeyoul or phewl,
perhaps a different note. An approaching or mobbing bird opens the beak widely and rattles
loudly, \ (Fig. 2E, F). The song and all these alarm calls are like calls of Black-

headed Antbirds (Percnostola rufifrons), but also resemble calls of antbirds of the genera
Myrmeciza and Pyriglena. The song and rattle are especially like those of all the species of
Pyriglena, and can be confused with notes of Pyriglena leucoptera at Boa Nova. The songs
of Narrow-billed Antbirds (Formicivora iheringi), common in lower midlevels of dry forest
at Boa Nova, are like slow and faint versions of the songs of Slender Antbirds. Competitive
mimicry (Cody, Ann. Rev. Ecol. Syst. 4:189-211, 1973) in voice seems possible, although
the 3 species do not overlap much in foraging.

Conservation. — Dry forest in central southern Bahia is rapidly being cleared for cattle,
according to CEPLAC, the Cocoa Institute at Itabuna and Renato Aragao, former director
of the Institute Brasileiro de Desenvolvimento Florestal at Salvador. The initial stages of
clearing for cattle, creating many zones of forest edge, probably benefit Slender Antbirds.
However, patches of forest are decreasing in size and length of edge. At Fazenda Alvorada,
cutting a corner of the woodlot near the ranch houses to plant corn and beans had trapped
1 pair of Slender Antbirds in a patch 50 x 100 m. Early each morning they hopped out
through newly felled trees, searched a bit and gave alarm notes, and soon returned to
bromeliads in the decreasing bit of forest. Another edge of the main woods was recently
cleared to plant introduced grass, even though wide areas of pasture nearby were reverting
to scrub. On the far side of the main woodlot, a neighbor was also clearing new areas. One
can see cleared slopes all around Boa Nova, so the fate of Slender Antbirds is certainly in
doubt. The pattern in similar areas in Sao Paulo state has been to clear all forests, then plant
eucalyptus or pine for crops when hills start to erode and pastures are no longer worth
weeding.

Slender Antbirds should be kept on the Endangered or Vulnerable lists of the International
Council for Bird Preservation. Another species restricted to mata-de-cip6, Formicivora iher-
ingi, should probably he added to the list. A forest reserve of mata-de-cip6 is certainly
desirable, just as it would be good to have reserves in all distinctive types of vegetation, or
even better to preserve a natural framework of vegetation in all regions. One idea for Boa
Nova would be a reserve in conjunction with an agricultural experiment station, which is
needed on the plateau because of its distinctive climate, soils, etc. The problem remains
that species often disappear from reserves (Willis, Ecol. Monogr. 44:153-169, 1974), espe-
cially small ones. W Idle the Slender Antbird has dense populations and may resist extinction,
it could be lost.

GENERAL NOTES

107

Acknowledgments. — We appreciate a grant from the Committee for Research and Explo-
ration of the National Geographic Society, the courtesy of ranchers Joel Almeida Sampaio
and Humberto Gomes Sampaio of Fazenda Alvorada and of other people of Boa Nova and
information furnished by the agronomists of CEPLAC at Itabuna. Eugene Eisenmann and
John Farrand checked the records of Rhopornis and suggested the correct Boa Nova for us.
Edwin O. Willis and Yoshik.4 Oniki, Dept. Biology, Univ. Miami, Coral Gables, Florida
33124. Accepted 15 Jan. 1980.

Wilson Bull., 93(1), 1981, pp. 107-108

Notes on the Uniform Crake in Costa Rica. — Although the Uniform Crake (Amau-
rolimnas concolor) is found from Mexico to Bolivia and Brazil, and formerly occurred on
Jamaica, virtually nothing is known of its habits. During fieldwork in 1971-1973 at Finca La
Selva, in the wet lowlands of NE Costa Rica, I obtained information on the behavior and
vocalization of this elusive bird, as well as the first unequivocal data on nesting of the species.

Although not reported for La Selva by Slud (Bull. Am. Mus. Nat. Hist. 121:49, 1960), the
Uniform Crake is fairly common in forested swamps, heavy vine-tangled thickets along
forested streams and in dense second growth adjoining forest, especially favoring the maze
of hanging dead and decaying leaves in Heliconia thickets (cf. also Orians and Paulson,
Condor 71:426, 1969; Kiff, Condor 77:101, 1975). In these dense, tangled habitats, the birds
are not particularly shy and may approach a motionless observer closely, but seldom leave
the densest available cover.

In life, A. concolor resembles a diminutive wood-rail (Aramides sp.) in build, posture and
soft-part colors: red iris, greenish-yellow bill and orange to reddish legs. The bird usually
has an erect stance and walks with head high and tail cocked, except while foraging. The
tail may be pumped in agitation and is carried low as the bird scurries, mouselike, across
an opening. The birds forage deliberately, walking slowly and pecking into leaf litter, hanging
dead leaves and detritus. Apparently they also dig in soft mud, as I have observed birds with
obviously muddy beaks on several occasions. I have seen Uniform Crakes seize, beat and
swallow spiders (Lycosidae), a very small frog (Eleutherodactylus sp.) and a small lizard
(Anolis sp.), which was killed with a few swift pecks, then swallowed headfirst.

Like Aramides wood-rails, the Uniform Crake possesses loud, arresting whistled calls that
often provide the only clue to its presence. To date 1 have noted the following vocalizations:

(a) A series of 6—9 clear, upslurred whistles, in which successive notes first become louder
and higher pitched, then accelerate, drop in pitch and fade away — tooeee, Tooeee, Toooeee,
TOOOEEE, Tooee, tooee-tuee-tui. A bird giving this call was often answered by another some
distance away. 1 could often decoy single birds to within 1—2 m by imitating it; this call is
probably a territorial advertisement or “song.” At very high intensities (as when answering
my imitation at very close range), the loudest notes of the call often had a flutelike break in
the middle: toourleee, etc.

(b) At close range in the last-mentioned situations a soft, low-pitched, pigeon-like cu-
uuuhuuuu is audible, possibly an aggressive note.

(c) Two birds, perhaps a mated pair, may call back and forth with one or several clear,
not very loud, whistled toooo notes, either level in pitch or slightly downslurred.

(d) A sharp, nasal kek is given by a startled bird.

The loud “song” of A. concolor was heard at La Selva chiefly from late August to Decem-
ber, which is probably the breeding season. On 14 November 1973, I found a nest with eggs
in an area where I had seen and heard much A. concolor activity in the preceding 2 months.

108

THE WILSON BULLETIN • VoL 93, \o. 1, March 1981

The nest consisted of a loose cup of leaves filling a hollow in the top of a vine-covered stump
beside a seldom used trail in a small treefall clearing in swamp forest. The stump was about
5 m from a stream, and 1 m from the nearset dense thicket, into which the adult bird flushed
from the nest promptly disappeared, allowing a brief glimpse of reddish-brown plumage and
orange legs. The nest contained 4 slightly incubated, sub-elliptical, slightly glossy eggs (set
no. 78177 of the estern Eoundation of Vertebrate Zoolog> ). The ground color of the eggs
is pale huffy, marked mostly near the large end with bold superficial reddish-brown splotches
and subsurface spots and blotches of grey and purplish-brown. Measurements (length and
largest diameter) and dry shell weights of each egg are: 33.40 x 26.11 mm, 0.774 g; 44.41 x
25.90 mm, 0.740 g: 33.28 x 26.18 mm, 0.789 g: and 33.60 x 25.74 mm, 0.675 g.

The first description of a putative .4. concolor egg (from Brazil) was by Nehrkorn (Katalog
der Eiersammlung nebst Beschreibungen der Aussereuropaischen Eier von Adolf Nehrkorn,
11 Auflage, R. Friedlander und Sohn. Berlin, 1910): “reddish-grey with very sparse violet
and rust-brown flecks, 33 x 26.5 mm" (translation mine). The measurements, but not the
colors, fit the set described here. Schonwetter (Handbuch der Oologie, Lieferung 5, Aka-
demie-verlag, Berlin, 1961) repeated Nehrkorn's description but could not locate the egg in
question, which is probably lost. Bond (Birds of the est Indies, averly Press, Baltimore,
Maryland, 1936) wrote of .4. concolor that “the egg of the continental form is said to be ashy-
grey with a reddish tinge, speckled dusky (34 x 31.5 mm)." This description was not re-
peated in later editions of Bond’s book, and its original source is unknown: since neither
colors nor measurements agree with the present set, the identification was probably erro-
neous. Finally, Wetmore (Birds of the Republic of Panama, Pt. 1, Smithson. Misc. CoU.
\ ol. 150, 1965) described eggs he believed to pertain to .4. concolor collected on Isla San
Jose, in the Las Perlas Archipelago off the Pacific coast of Panama on 1 September 1944.
The description, measurements and date agree well with the present set. but as the eggs
had been picked up by a worker who stated only that they “had been found in a low nest,"
the identification clearly required corroboration.

The eggs of .4. concolor are quite similar to those of Rallus spp. and Aramides spp., but
differ markedly in these respects from eggs of Loterallus spp. or Porzana spp. Behavior and
vocalizations of Amaurolimnas also indicate close affinities with Aramides, as suggested by
Olson (\^ ilson Bull. 85:381. 1973). Peters (Birds of the orld. Vol. 2. Harvard Univ. Press,
Cambridge. Massachusetts, 1934) had placed Amaurolimnas among a group of probably
unrelated Old \\ orld genera. Ripley (Rails of the orld. David Godine, Boston. Massachu-
setts. 1977) recognized an Amaurolimnas-Aramides relationship by lumping both into the
Old \\ orld Eulabeornis. Relationships to Old W orld rails are beyond the scope of the present
paper, but my observations definitely support the conclusion that Aramides is the closest
relative of Amaurolimnas.

1 thank L. F. Kiff for supplying weights and measurements of the eggs of A. concolor and
for bibliographic assistance and critical commentary, k. C. Parkes and S. Olson suggested
several improvements in the manuscript and S. M. Smith provided help in the field. ork
at La Selva was supported financially by a Chapman-Naumberg Postdoctoral Fellowship
from the American Museum of Natural History, and logistically by the Organization for
Tropical Studies. — F. G. Stiles, Escuela de Biologia, L'niversidad de Costa Rica, Cuidad
i niiersitaria. Costa Rica. Central America. Accepted 14 Jan. 198U.

GENERAL NOTES

109

Wilson Bull., 93(1), 1981, p. 109

Trichomoniasis in Bald Eapjles. — The protozoan parasite Trichomonas gallinae has
been reported causing disease in the upper digestive tract of a variety of birds including the
Golden Eagle (Aquila chrysaetos); the Rock Dove (Columba livia) is its primary host (Levine,
Protozoan Parasites, Burgess Publishing Co., Minneapolis, Minnesota, 1973). This flagellate,
in birds of prey, causes a disease called “frounce,” characterized by yellow caseous nodules
in the upper alimentary canal and emaciation (Stabler, Exper. Parasitol. 3:368-402, 1954).
This note reports 2 cases of trichomoniasis found in Bald Eagles (Haliaeetus leucocephalus)
that were subsequently treated with Emtryl (l,2-dimethyl-5-nitroimidazole. Dr. Salshury’s
Laboratory, Charles City, Iowa). Stabler and Kitzmiller (N. Am. Falcon. J. 7:47-48, 1967)
recommend this drug as a safe and effective treatment for trichomoniasis in hawks.

On 25 February 1977, a debilitated 6.3 kg adult Bald Eagle was found in Sullivan County,
New York. The bird was treated for shock and given an antibiotic by a veterinarian. When
the senior author examined the bird on 27 February 1977, yellow-brown caseous lesions were
present over most of the surface of the hard palate, and the saliva was blood-tinged. Nu-
merous live trichomonads were demonstrated microscopically from the lesions and saliva,
and lesion smears stained with Giemsa stain showed organisms fitting the description of T.
gallinae (Levine 1973). The bird was given orally three 125 mg tablets of Emtryl. On 1 March
1977, the mouth lesions were nearly gone. However, rare trichomonads were found micro-
scopically from material taken from the esophagus on 4 March and 3 more 125 mg tablets
were given. This was repeated on 5 March. No ill effects were noted from the medication.
The eagle rapidly gained strength and showed greatly increased aggression toward its care-
takers. Further examination for trichomonads were negative, and the bird was released to
the wild on 11 March.

The second Bald Eagle was an immature, captured in a weak, emaciated condition in a
farm field near Cutchogue, Suffolk Co., Long Island on 26 July 1978, after it flew weakly
against a slow moving pick-up truck. This bird was treated for shock and x-rayed at the
North Fork Animal Hospital, Southhold, Long Island. No broken bones, or reason for the
sickness were found. On 28 July 1978, the bird was transferred to the Delmar Wildlife
Resources Center at Delmar, New York. The eagle was thin (3.6 kg), weak and had numerous
yellow-brown circumscribed lesions on the hard palate, pharynx, tongue and anterior esoph-
agus. Microscopic examination of oral scrapings revealed numerous organisms typical of T.
gallinae. Five hundred mg of Emtryl were administered to the bird on 28 July by mouth.
The dose was repeated 24 h later as the lesions were disappearing. The eagle showed no
I untoward effects from an approximate dose of 278 mg of Emtryl per kg of body weight given

i! in a 24 h period. The bird fed voraciously on venison and fish, and appeared stronger. On
! 31 July the lesions had almost entirely healed and the bird was given another 250 mg of
Emtryl. No trichomonads could be found in the mouth or esophagus. The bird gained weight
i and vigor, remained negative for trichomonads and was released on 10 August at Shelter
I Island, New York.

I The eagles may have contracted the trichomoniasis from eating Rock Doves and/or Mourn-
l) ing Doves {Zenaida macroura). Tangredi (N.Y. Fish and Game J. 25:89-90, 1978) reported
I trichomoniasis in Mourning Doves from Long Island, and many Rock Doves were seen near
( the capture site of the eagle from Sullivan County.

These seem to be the first cases of trichomoniasis reported in the Bald Eagle. Emtryl
q appears to be a very quick and effective way to eliminate trichomoniasis in this species. —
I Ward B. Stone and Peter E. Nye, N.Y.S. Dept. Environmental Conservation, Delmar
i* Wildlife Resources Center, Delmar, New York 12054. Accepted 27 Nov. 1979.

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

Wilson Bull., 93(1), 1981, p. 110

Protocalliphora infestation in Broad-winged Hawks. — In spring and summer 1978,
during study of the productivity of the Broad-winged Hawk {Buteo platypterus) in Chautauqua
County, New York, we discovered infestation of nestlings by dipteran larvae later identified
as Protocalliphora avium Shannon and Dobroscky (Calliphoridae). This is the first reported
occurrence of infestation of the Broad-winged Hawk by this ectoparasite. Bohm (Wilson BuU.
90:297, 1978) listed Protocalliphora spp. in Great Horned Owls (Bubo virginianus). Long-
eared Owls (Asio otus). Red-tailed Hawks (Buteo jamaicensus). Red-shouldered Hawks (Bu-
teo lineatus) and Cooper’s Hawks (Accipter cooperii).

Larvae were first noticed and collected from the ear cavities of nestling broad-wings from
18-26 days old. Each nestling in 3 broods of 2 averaged 9 larvae (range 2-15) per pair of ear
cavities. Three nestlings in an additional brood, killed by a predator at 6-9 days of age, were
not infestpd. The infested nests were widely separated, and the nest-sites were variable in
characteristics.

The nestlings appeared to suffer no major deleterious effects from infestation by the blood-
sucking larvae. However, bleeding was observed in the ear cavities, the skin around the ear
openings was swollen and scabs sometimes covered the ear cavities. No behavioral abnor-
malities were noticed. Bohm (1978) stated that infestation by these dipterans caused no
serious harm to large species, but did cause some mortality in small passerines. Because
mortality is known in small birds, and because at least 1 other ectoparasite (Mexican chicken
hug [Haematosiphon indorus]) is known to cause nestling mortality in raptors (Platt, Wilson
Bull. 87:557, 1975), Protocalliphora might be expected to cause or contribute to mortality
of small or undernourished nestling Broad-winged Hawks, and potentially other raptor nest-
lings. The effects of ectoparasites should be looked for in bird species showing brood re-
duction strategies for growth and reproduction (see O’Conner, Living Bird 16:209-239, 1977)
in which the young are often undernourished and weak, and this includes raptors.

We wish to thank Allen Benton, Robert Bohm and especially C. W. Sabrosky for their aid
in identifying the larvae. — ScoTT Crocoll and James W. Parker, Dept. Biology and
Environmental Resources Center, State Univ. Coll., Fredonia, New York 14063. Accepted 30
Oct. 1979.

Wilson Bull., 93(1), 1981, pp. 110-111

Herring (iiill attacks an«l eats adult male Oldsquaw. — Herring Gulls (Larus argen-
tatus) have been observed preying on a wide variety of small adult birds (Witherby, Jourdain,
Ticehurst and Tucker, The Handbook of British Birds, Vol. 5, H. E. and G. Witherby Ltd.,
London, England, 1952: Harris, Ibis 107:43-53, 1965). Such prey items are almost always
small passerines or shorebirds, and as such are much smaller than the guUs themselves.
Harris (1965) mentions Herring Gull predation on Manx Shearwaters (Puffinus puffinus).
Razorbills (Alca torda) and Common Puffins (Fratercula arctica), and Peter Fetterolf (pers.
comm.) has observed Herring Gulls preying on juvenile Ring-billed Gulls (L. delawarensis).
Few, if any, instances of Herring Gulls preying on birds larger than these have been reported.
This note reports an instance of predation on an adult male Oldsquaw (Clangula hyemalis)
by an adult Herring Gull. The average weight of an adult male Oldsquaw in December is
about 580 g (Peterson and Ellarson, W ilson Bull. 91:288-300, 1979). The average weight of
an adult Herring Gull is 1098 ± 151 g, based on a sample of 15 male and 11 female specimens
at the Royal Ontario Museum, Toronto, Canada.

GENERAL NOTES

111

Daily mid-December populations of Oldsquaws in the inner harbor at Toronto, Ontario,
average about 1500 individuals (Alison, M.Sc. thesis, Univ. Toronto, Toronto, Ontario, 1970).
On 28 December 1978, at 10:23, while watching Oldsquaw behavior at about 1000 m from
the 30 m Toronto Harbor Police observation tower with a 20x spotting scope, I observed an
adult Herring Gull attack an adult male Oldsquaw. The gull dived at and struck the duck
from a height of 20-30 m. Similar attacks were observed at 11:00, 11:19 and 11:55 and others
may have occurred. The Oldsquaw was never observed diving or flying and swam only when
attacked. At 12:06, aU the Oldsquaws (>100 individuals) within 1500 m flocked and flew in
a tight 50-100 m diameter circle above the stricken duck. The Herring Gull, having appar-
ently killed the duck, was seen using its bill to bite and pick at the carcass, apparently
feeding; after about 3 min the flock of circling Oldsquaws began to disperse. The gull alter-
nately swam within 3 m of the Oldsquaw carcass or fed on it until 12:47, at which time the
gull departed. At 13:01, either it or another Herring Gull landed beside the carcass and
remained within 3 m of it occasionally feeding until 14:17, when the gull departed. No
subsequent visits to the carcass were made by Herring Gulls or any other birds, and by 14:40
the duck was no longer visible. Presumably it sank; Schorger (Wilson Bull. 59:151-159, 1947)
reported that Oldsquaws with completely water saturated plumage have negative buoyancy.

I thank the Toronto Harbor Police for permitting access to their tower. Comments and
criticism by P. M. Fetterolf, J. D. Rising, N. J. Flood, G. R. Bortolotti and J. C. Barlow are
appreciated. — Richard R. Snell, Dept. Zoology, Univ. Toronto, Toronto, Ontario MSS lAl
Canada. Accepted 30 Jan. 1980.

Wilson Bull., 93(1), 1981, pp. 111-112

Red-legged Kittiwak.es forage in mixed-species flocks in southeastern Alaska. —

The foraging behavior of Red-legged Kittiwakes {Rissa brevirostris) away from their breeding
sites is virtually unknown. Between 1-7 September 1978, we observed adult and juvenile
Red-legged Kittiwakes foraging in a mixed-species flock of adult Mew Gulls [Lams canus),
juvenile Bonaparte’s Gulls {L. Philadelphia) and juvenile Glaucous-winged GuUs {L. glau-
cescens) in the lower Green’s Creek drainage on Admiralty Island in southeastern Alaska.

During ebb tide the exposed delta mud flats at Green’s Creek are used by thousands of
guUs and shorebirds, particularly during spring and fall migration. The lower portion of
Green’s Creek is used by spawning salmon during late July through September: hundreds
of humpbacked salmon (Oncorhynchus gorbusche) were present during our observations.

We observed 5 mixed-species flocks of about 50 individuals each at the open meadow
bordering Green’s Creek, always during ebb tide; never when the stream bank and adjacent
meadow were flooded. The flock composition was nearly constant: Glaucous-winged Gulls,
10%; Mew Gulls, 25%; Bonaparte’s Gulls, 35%; and Red-legged Kittiwakes, 30% (adults,
10%; juveniles, 20%). (The identification of the kittiwakes in the flocks was difficult at first;
however, the juveniles were discriminated from juvenile Bonaparte’s Gulls by the kittiwakes’
well-marked, dark cervical collar, their dusky eyes, and unbarred tail. The adult kittiwakes
differ from adult Mew Gulls by their solid black wing tips and red legs.) Glaucous-winged
Gulls arrived first and remained near the deep, still water or the gravel shore, coming closest
to the forest edge. Mew Gulls arrived next, followed soon by Bonaparte’s Gulls. Red-legged
Kittiwakes were the last to arrive, and they frequented the faster stream riffles and stayed
farthest from the forest edge. Mew and Bonaparte’s gulls always stayed between the Glau-
cous-winged Gulls and Red-legged Kittiwakes. None of the gulls or kittiwakes left the stream
banks, nor flew into the forest. We observed no interactions between species.

112

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

Glaucous-winged and Mew gulls spent most of their time sitting on the water or the rocky
shore. They pecked the exposed dorsal surface of the spawning salmon, but we could not
confirm feeding. Bonaparte’s Gulls and Red-legged Kittiwakes flew almost continuously and
dived frequently. Bonaparte’s Gulls appeared to be “pursuit diving,” and the kittiwakes,
“dipping” (terms from Ashmole, pp. 223-286 in Avian Biology, D. A. Earner and J. R.
King, eds.. Academic Press, New York, New York, 1971). Close observation showed that
kittiwakes generally dived between salmon and ate something from the stream bottom, pre-
sumably salmon eggs. On at least 2 occasions, kittiwakes pecked at the exposed dorsal
surface of salmon spawning in the stream. Although dead salmon were abundant on sand
banks, we never saw gulls or kittiwakes eat any of them.

Adult and juvenile kittiwakes foraged similarly. No pattern of dominance, aggression, or
indication of feeding hierarchy was detected. Juveniles foraged next to, and independent of,
adults.

Red-legged Kittiwakes are commonly found near the Pribilof Islands during the breeding
season; they are rare, post-breeding visitors to the northeastern Bering Sea and Aleutian
Islands. There are a few accidental records in southeastern Alaska and the Yukon River
(Kessel and Gibson, Stud. Av. Biol. 1:48-49, 1978), but they are usually seen at sea, if at all.
They have only been reported to feed on small fish and cephalopoda in the waters south of
the Pribilofs (Hunt, pp. 196-382 in Environmental assessment of the Alaskan continental
shelf, Vol. 2, Natl. Ocean. Atmos. Admin. Environ. Res. Lab., Boulder, Colorado, 1977).

Bonaparte’s and Mew gulls are common visitors to southeastern Alaska in fall, and the
Glaucous-winged Gull breeds there — it probably is not unusual to find these 3 gulls foraging
together (Bent, U.S. Natl. Mus. Bull. 113:65-73, 1921). Our observations are significant in
that range extensions in both location and time are established for the Red-legged Kittiwake,
and the post-breeding feeding habits in southeastern Alaska are described for the first time.

I'his study was partially supported by VTN, Inc. We thank V. Byrd, G. Hunt, K. Vermeer
and the Editor for helpful comments. — DOUGLA.S Siegel-CauSEY, Dept. Ecology and Evo-
lutionary Biology, Univ. Arizona, Tucson, Arizona 85721 AND Thomas E. Meehan, VTN
Consolidated, Inc., 2301 Campus Dr., Irvine, California 92713. Accepted 10 Eeb. 1980.

Wilson Bull., 93(1), 1981, pp. 112-114

(iroiincI-f‘e«Mliiig: methods arul niche separation in thrushes. — Recent papers by
Clark (Wilson Bull. 83:66-73, 1971) and Henty (Wilson Bull. 88:497-499, 1976) have de-
scribed a foraging method used by a number of species of birds, consisting of lateral sweeps
of the bill to move aside loose material. They termed it “bill-sweeping” and reported its
occurrence in several thrushes (Turdus), namely the White-necked Thrush (Turdus albicol-
lis), European Blackbird (T. merula), American Robin (7\ migratorius) and Songthrush (T.
philornelos). 1 have recently observed such behavior, together with other foraging methods,
in 2 additional species, the Fieldfare (7’. pilaris) and European Redwing {T. iliacus) and
incidentally in the European Blackbird. The intention of this note is to relate bill-sweeping
to other principal foraging techniques used by thrushes when feeding on the ground, and to
comment on niche separation in the genus.

Observations were made between October 1975 and February 1976 on wild birds in Cam-
bridgeshire, England, using a hide or vehicle stationed at the edge of a field. Birds regularly
approached to within 20 m, and sometimes to within 10 m of a concealed observer, providing
detailed views of their searching and handling techniques.

GENERAL NOTES

113

Table 1

Relative Use of Food Categories by Fieldfares and Redwings

Fieldfare

Redwing

N (%)

Surface items (all <10 mm long)

266 (43.7)

176 (66.9)

Soil items

342 (56.3)

87 (33.1)

Soil items >10 mm long

154 (45.0)

19 (21.8)

The 2 main species were wintering together on areas of cattle-grazed permanent pasture.
For most of the winter this was their preferred habitat, in which they collected invertebrate
food. Foods taken comprised 2 major categories: (1) “soil items,” including earthworms,
centipedes, slugs and larvae of beetles and craneflies (Tipulidae), which required extraction
from soil or grass tufts; and (2) “surface items,” exposed on the soil or vegetation, and
comprising mainly flies, beetles and spiders, for which handling was minimal. This division
was based on the difference in technique required for the capture of prey items in each
category.

Similar methods were used by both Fieldfares and Redwings when taking prey from a
given category. In both species searching for both categories consisted of running along the
ground in short bursts, usually of 1-5 paces or hops, halting after each run, and apparently
scanning the ground in the immediate vicinity. If a potential food item was spotted, a few
paces might be taken towards it. In the case of a surface item the bird usually pecked
immediately at it (although there might be a brief hesitation) during which the bird might
cock its head to one side before the peck. Occasionally, more than 1 peck was made. With
a soil item, the behavior was similar to the bill-sweeping described by Clark (1971), although
he found bill-sweeping used “to move aside twigs, leaves, dry soil, or snow” and did not
mention use on relatively hard-packed substrates such as the damp soil with dense roots
found on the present study area. On hard substrates the behavior is more appropriately
termed digging, as described by Heppner (Condor 67:247-256, 1967) for American Robins
hunting earthworms. As in robins, a Redwing or Fieldfare would stop near potential prey,
hesitate, often cocking the head to one side, sometimes take a short step backwards or to
one side, then stab downwards. Often several pecks were made, when the first few did not
usually secure the food. Instead the downward stab was followed by a head flick, often
causing soil to be thrown to one side. The flick might be more or less pronounced, and
occasionally was not lateral.

The major difference between the bill-sweeping described by Clark (1971) and Henty (1976)
and digging as described by Heppner (1967) and the present paper is that the former is a
search technique while the latter is a “pursuit” technique used after a potential prey has
been spotted. However, the two are seemingly related.

Bill-sweeping was also used by Fieldfares and Redwings to move aside loose material.
Redwings occasionally fed in dead leaves in the manner described by Henty (1976), walking
through the litter with almost continuous, rapid, lateral head movements. Redwings used a
similar method when searching cattle dung for dipteran larvae, except the bird stood still by
a pat while searching. When Redwings foraged in cowpats in this way, the technique was
intermediate in action between bill-sweeping and digging, and was used for both search and
“pursuit.” Fieldfares searched litter much less frequently than Redwings, and were never
observed feeding at cowpats.

114

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

The 2 species differ in size — the Fieldfare weighing 120-140 g and the Redwing 60-80 g.
These differences were reflected in differences in diet. I recorded the numbers of surface
and soil items taken by thrushes in many large flocks over the whole winter (Table 1), In
order to avoid bias due to observations containing different numbers of items captured I
calculated the proportion of surface items taken during each feeding record. A Mann-Whit-
ney U-test on the difference of this proportion between the species was highly significant
(P < 0.001). The Redwings took more surface items than soil items and the Fieldfares took
more soil items than surface.

Digging required more time and energy than picking items from the grass. However, soil
items were generally larger than surface items (Table 1); also Fieldfares took larger soil
items than did Redwings (Table 1, Mann-Whitney U-test, P < 0.01).

Thus, the 2 species subdivided the habitat primarily on spatial and behavioral differences;
differences in prey size were to some extent consequential upon these, supporting Hespen-
heide (pp. 158-180 in Ecology and Evolution of Communities, M. L. Cody and J. M. Dia-
mond, eds., Belknap Press, Harvard, Cambridge, Massachusetts, 1975). However, prey size
differences were also evident within a foraging zone, indicating either that each species had
prey-size preferences or that some undetected difference in foraging technique was involved
(such as depth of digging).

In late February and March, when the availability of larger insects on the surface and
vegetation increased (Tye, unpubl.), the Fieldfare took proportionately more surface items
(75.4%) than in winter (43.7%, October to mid-February) and more closely resembled the
Redwing in feeding techniques, suggesting that prey-size preferences may have been im-
portant in determining the Fieldfare’s feeding behavior.

Litter-feeding was most common when the ground was frozen or snow-covered. At such
times most Fieldfares left the study area completely, and the remaining few fed on small
clear patches of pasture or garden lawns. In contrast. Redwings moved into hedge-bottoms
and gardens and switched to litter-feeding. Prey items found in litter resembled surface
items of the open pasture, mostly small arthropods and slugs. Large items were uncommon
in the litter which may explain why Fieldfares did not often feed there.

The European Blackbird apparently used the same techniques as the Fieldfare and Red-
wing, if not in the same proportions. The Blackbird tended to bill-sweep in a strict sense
more than the Fieldfare, especially in litter, and to dig more than the Redwing, and was
probably intermediate in its use of soil and surface items on pastures. It is also intermediate
in size (ca 90-120 g).

Therefore, when feeding on the ground, the Tardus species studied so far all seem to use
the same range of related feeding techniques. Within a habitat they achieve niche separation
by the differential use of these techniques, and therefore of the foraging zones for which
each technique is appropriate. Further niche separation may also be achieved by prey-size
preferences within a foraging zone. The behavioral, spatial and prey size differences may be
related to body size.

These observations were made while working for a Ph.D., supervised by the late Prof. R.
K. Murton and Dr. P. P. G. Bateson, at Monks Wood Experimental Station, and financed
by a Natural Environment Research Council Studentship. P. P. G. Bateson, J. C. Barlow,
J. P. Dempster, I. Newton and N. J. Westwood commented on the manuscript. N. J. West-
wood provided some bird weights. — Alan Tye, Monks Wood Experimental Station, Abbots
Ripton, Huntingdon PE 17 2LS, Cambridgeshire, England. (Present address: Dept. Zoology,
Fourah Bay Coll., Univ. Sierra Leone, Freetorun, Sierra Leone.) Accepted 20 Feb. 1980.

GENERAL NOTES

115

Wilson Bull., 93(1), 1981, pp. 115-118

American Coot distribution and migration in Colorado. — American Coots (Fulica
americana americana) are widespread in North America, breeding primarily on fresh water
wetlands and wintering on both brackish and fresh water habitats (Fredrickson, in Manage-
ment of Migratory Shore and Upland Game Birds in North America, Sanderson, ed., Int.
Assoc. Fish and Wildl. Agencies, Washington, D.C., 1977).

Despite being one of the more common marsh birds and a game species, the coot is often
overlooked or ignored during waterfowl inventories or marsh evaluations. Current data con-
cerning the coot in Colorado are limited. Cooke (The Birds of Colorado, Colo. Agric. Exper.
Stat. Bull. 37, 1897), Keyser (Birds of the Rockies, McClory & Co., Chicago, Illinois, 1902),
Sclater (A History of the Birds of Colorado, Witherby and Co., London, England, 1912),
Niedrach and Rockwell (The Birds of Denver and Mountain Parks, Colo. Mus. Nat. Hist.
Pop. Ser. No. 5, 1939) and Bailey and Niedrach (Birds of Colorado, Denver Mus. Nat. Hist.
Vol. I, 1965) summarized general information on nest observations, early and late occur-
rences and distribution. More recently Kingery and Graul (Colorado Bird Distribution Lati-
long Study, Colo. Field Ornithol. and Colo. Div. Wildl., Denver, Colorado, 1978) updated
distribution data and Lane and Holt (A Birder’s Guide to Eastern Colorado, L & P Press,
Denver, Colorado, 1979) gave average arrival and departure dates for eastern Colorado. We
initiated surveys of wetlands in August 1976 to further delimit distribution, seasonal abun-
dance and aspects of coot biology in Colorado.

Study areas and methods. — Counts of coots were conducted on 4 study areas to document
spring and fall migration. Study areas were selected to represent the major topographic areas
of Colorado and were located at Beebe Draw near LaSalle, Weld Co., on the eastern plains;
Lake John, near Walden, Jackson Co., in North Park; Ice Pond, near Buena Vista, Chaffee
Co., in the central mountains; and Hog Lake, part of Brown’s Park National Wildlife Refuge
(NWR), Moffat Co., in extreme northwestern Colorado. Each study area included an emer-
gent marsh dominated by cattail {Typha spp.) and/or bulrush {Scirpus acutus) and areas of
open water.

Weekly counts were initiated in 1977 to document spring migration at Beebe Draw and
Lake John. Ice Pond and Hog Lake were not visited until after the peak of migration.
However, counts by Brown’s Park National Wildlife Refuge personnel from 14 March to 22
May 1977 were used to indicate coot migration at Hog Lake. In 1978, regular visits to all
areas began by mid-March.

Distribution and status were determined from 108 questionnaires returned by field per-
sonnel of the Colorado Division of Wildlife and amateur ornithologists and by observations
of 230 wetlands encountered while traveling between study areas throughout the state from
August 1976 through November 1978. Most regions of the state were visited, with the ex-
ception of the far eastern plains, especially along the South Platte and Arkansas rivers.

Date or period of observation, dominant emergent vegetation, geographic location, coot
numbers and status (breeding, resident, migrant only, etc.) were recorded for each wetland
visited. Vegetation was classified as either cattail and/or bulrush, or other. Omitted were
roadside ditches and wetlands completely filled with cattail and/or bulrush with no open
water. “Other” included areas with sedges {Carex spp.), willows (Salix spp.), or grasses
(Graminae) as the dominant emergents, or areas without any emergents, such as many lakes,
reservoirs, and stockponds. Geographic location classifications were eastern plains, high
mountain valleys and west of the Continental Divide. High mountain valleys refer specifically
to North, Middle and South parks (large inter-mountain depressions devoid of extensive
woodlands) and the San Luis Valley. Coots were considered to be breeding if territorial

116

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

Table 1

Relationship of Coot Breeding Distribution and Vegetation Type in Colorado

Number of areas surveyed

Dominated by Dominated by

Typha or Scirpus other vegetation

Area

With

breed-

ing

coots

None
or non-
breed-
ers

Total

With

breed-

ing

coots

None
or non-
breed-
ers

Total

(df = 1)

Prob-

ability

Statewide

136

151

63.13

<0.001

Eastern plains

42.02

<0.001

High mountain valleys
West of Continental

12.44

<0.001

Divide

25.45

<0.001

behavior (as defined by GuIIion, Wilson Bull. 64:83-97, 1952) and nests and/or young coots
were observed during 15 April through 15 August, and nonbreeding if gregarious and non-
territorial. Coots observed from 16 August through 14 April were classed as migrating or
wintering. Distribution and status classifications were assigned according to the latilong
system (Kingery and Graul 1978), with the state divided into 27 blocks based on lines of
latitude and longitude.

Distribution and status. — Coots bred throughout most of Colorado. Breeding was not doc-
umented in 2 latilong blocks, one located in the southcentral mountain region and the other
in the southeastern plains. Major breeding areas were Brown’s Park on the western slope,
and North Park and the San Luis Valley in the high mountain valleys. Breeding coots were
not observed in Middle or South parks, although previously reported (Cooke 1897, Lane and
Holt 1979) and to 3045 m elev. at Kenosha Pass (Bailey and Niedrach 1965). On the plains,
coots were locally abundant where suitable habitat occurred.

Nesting coots were primarily associated with cattail and bulrush marshes (Table 1). A Chi-
square test of independence between the presence or absence of breeding coots and the
occurrence of cattail and/or bulrush dominated marshes indicated a strong relationship {P <
0.001) in each region and statewide. A similar analysis for the migration-wintering period
also indicated a strong relationship (P < 0.001) statewide. Not all cattail or bulrush marshes
were used by breeding, migrating or wintering coots, but they were used in strong preference
to other vegetation types. Other vegetation types used for nesting included tamarix (Tamarix
gallica), spikerush {Eleocharis macrostachya), willows and sedges. Sedges growing as emer-
gents were characteristic of many high elevation wetlands used by coots, especially wetlands
in North Park.

Coots were resident in low numbers (< 1000), mainly along the western boundary of the
plains from near Port Collins south to Pueblo. The presence of wintering coots was dependent
on mild winter weather. Coots were present near Port Collins during the winter of 1976-77,
hut absent during the 1977-78 winter when water areas froze. In the Port Collins area, coots
associated with wintering waterfowl and fed on feces and waterfowl carcasses. Christmas
bird counts sponsored by The National Audubon Society (1960-1977) and listed in Au-
dubon Field Notes and American Birds have consistently noted coots along the Front Range
from Fort Collins to Pueblo:

Spring migration. — At Beebe Draw and Hog Lake coots arrived in late February or early

Table 2

Total Number of Coots Observed During Spring and Fall Migrations 1977 and 1978, 4 Study Areas, Colorado

GENERAL NOTES

117

(ON

rf*

i— '

1— 1

rf-

f-H

•-H

§§

i;) '

-o

-a

-Q

C/)

-C

r-H

r— 1

<On

Pi!

-J

I—!

■o

-Q

Pi!

-J

o ON o 'O
^ r— NO ON o
cs ^

ON 00 lO
r-l O ON
NO On

o o lo csi

CM On UO On O

LO CM CM ^ CM

CM CM

I I

CO NO CM
Tj- C^ O CM
NO NO 00 CM
CM CM

CM ^ NO lO i-H

OO nO oo On lO

lO O CO OO CM

^ CM

OO lO O NO

CM O uO Tf O

^ OO CM 00 Tf

OO CM

I-H r- O CM 00 CM

NO 00 LO Tf CM

I CM NO CM Tt CM

CM ^

LO NO 00
r- NO ^ ^
Tf' CO CO CM
CM

CM On On LO NO
ON CM
^ CM NO oo

00 O O lO ON o
NOooONCMNor-
^ ON LO CM

CO NO
ON oo

I g I ^ I

ON ON Li ^

Xh r— I

On On
^ ^ ^ On On

''3 ^ C C

Q Q -i ^ 1 ^

_Q _Q OJ <D Ph

O O P»S

o t> o ca

m cQ j j

OJ O

Hog Lake 1977 — — 575 273 518 367 402 435 336 301 — 215

Hog Lake 1978 — 345 307 368 577 499 652 620 678 549 422 349

118

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

March. Peak numbers were recorded in 1977 and 1978 during the second week of April
(Table 2). At Lake John coots arrived in 1977 and 1978 during the second week of April,
although coots were present elsewhere in North Park the previous week. Arrival date was
dependent on ice thaw. In both years numbers increased rapidly for 2 weeks, then decreased
to resident levels within 3-4 weeks. Although coots arrived at Ice Pond earlier than Lake
John, peak numbers were also recorded the third week of April. The decrease of coot
numbers from peak to resident levels in 1977 and 1978 was consistent with other areas.

Timing of migration and relative numbers observed on each area were consistent between
years. The general pattern statewide was arrival in late February or early March (first week
of April at Lake John), increasing to peak numbers during the second to third week of April,
then decreasing to resident levels by the second to fourth week of May. Timing of peak
migration was similar for the eastern plains and west of the Continental Divide. The peak
occurred 1 week later in the high mountain valleys. The literature suggests a similar pattern
of coot migration in Colorado.

Summer movements. — Numbers of adult coots increased gradually on the study areas and/
or peripheral marshes starting in mid- to late July. Increases in numbers of immature coots
30 days of age or older occurred in mid- to late August. An influx of adult and immature
coots on Hog Lake in 1977 began about 2 weeks earlier than the general pattern found on
other areas.

The late summer buildup was influenced by local conditions. Adult and immature coots
will leave a marsh as water levels recede and it dries (Ryder, Ph.D. thesis, Utah State Univ.,
Logan, Utah, 1958). This situation was particularly applicable to Beebe Draw and Hog Lake,
where breeding coots occurred on nearby marshes. Movements onto Hog Lake were in part
related to pumping schedules and water levels on nearby marshes. Water levels were allowed
to recede on some marshes in Brown’s Park NWR as the summer progressed. Immature
coots crossed the Green River in the direction of Hog Lake on at least 1 occasion.

Fall migration. — Peak numbers were recorded in 1977 on all areas during late August to
mid-September, before the waterfowl hunting season of 1-14 October (Table 2). In 1978,
peak numbers were observed 5 weeks later at Beebe Draw, 2 weeks later at Lake John, 1
week later at Ice Pond and 6 weeks later at Hog Lake. Peak numbers occurred before the
30 September-13 October 1978 waterfowl season at Ice Pond and during or after the
hunting season for the other areas. In November 1977 and 1978, coot numbers decreased to
minimal levels. No coots were known to winter on the study areas.

Compared to spring, fall migration occurred over a more widespread period, with less
dramatic changes in numbers and with less consistency in timing. Successive migratory
waves may explain the fluctuating numbers observed on some Colorado areas.

Acknowledgments. — Financial assistance was provided by the Accelerated Research Pro-
gram for Migratory Shore and Upland Game Birds and the Colorado Division of Wildlife,
Federal Aid to Wildlife Restoration Project W-88-R. We are grateful to J. Creasey, refuge
manager. Brown's Park National W ildlife Refuge, and his staff, G. Deutscher and J. Sellers,
for their assistance throughout the field investigations. We thank the Beebe Draw Gun Club
for access to its property and the Mt. Princeton Fishing Club, especially R. M. Stabler for
access to Ice Pond, (/ratitude is expressed to all who returned questionnaires concerning
coot observations in Colorado. — Warnek P. Goreinzel, Ronaed A. Ryder, Dept. Fishery
and Wildlife Biology, Colorado State Univ., Fort Collins, Colorado 80523 AND Clait E.
Braun, Colorado Div. Wildlife, Wildlife Research Center, 317 West Prospect, Fort Collins,
Colorado 80526. (Present address WPG: Extension Wildlife and Sea Crant, 554 Hutchinson
Hall, Univ. California, Davis, California 95616.) Accepted 26 Jan. 1980.

GENERAL NOTES

119

Wilson Bull., 93(1), 1981, pp. 119-121

Reproductive rate and renesting of Red-winged Blackbirds in Minnesota. — Red-
winged Blackbirds {Agelaias phoeniceus) have an economic impact upon the industry of wild
rice {Zizania aquatica) cultivation in northern Minnesota (Moulton, J. Wildl. Manage.
43:747-751, 1979). Red-wings nest in the marsh-like, emergent vegetation that borders the
peripheral drainage ditches of the wild rice paddies. Very few Red-winged Blackbird studies
have provided data on the number of young fledged per female, per territorial male, or per
unit area; or on the extent and nature of renesting and movements of females during the
nesting season (Dolbeer, Auk 93:343, 1976). The objectives of this study were: (1) to estimate
the size and reproductive success of a population of Red-winged Blackbirds on a typical
group of wild rice paddies; (2) to estimate fledging rates per male territory and per nesting
female; and (3) to estimate the extent of renesting and movement by marked females during
the nesting season in this habitat.

The study was conducted from late April to early August 1977 on commercial wild rice
paddies (total area 53.2 ha) located 185 km north of Minneapolis, in Aitkin, Minnesota. Nest
searching began in early May and continued through July. Active nests (1 or more eggs or
nestlings) were checked daily. Prior to nesting, some birds were captured in large mist nets
and in a large, walk-in decoy trap baited with oats and live blackbirds. Territorial males
were captured in wire traps that used a live, adult male as a decoy (Bray et al.. West. Bird
Bander 50:4-7, 1975). Nesting females (with nestlings) were captured by placing small pieces
of mist net around their nests. Each bird was marked with a USFWS band on 1 leg and a
numbered, plasticized-nylon streamer, secured around the tarsus with an aluminum grommet
(Arnold and Coon, Bird-Banding 42:49-50, 1971; DeHaven, West. Bird Bander 50:48-50,
1975), on the other leg. Birds were classed as either second-year (SY) or after-second-year
(ASY) on the basis of plumage. Females were aged by color of marginal wing coverts (Payne,
Univ. California Publ. Zool. 90:57, 1969). An error rate of from 18 (Dolbeer 1976) to 20%
(Payne 1969) must be expected when using this technique.

Nesting chronology and success. — A total of 182 nests (154 active) was located and marked
along the 9566 m of paddy ditches. Dominant plant species bordering the drainage ditches
were broad-leaf cattail {Typha latifolia) and narrow-leaf cattail {T. angustifolia), sedges
{Carex spp.), bulrushes {Scirpus spp.), various grasses (Gramineae) and some water plantain
{Alisma spp.) and arrowhead {Sagittaria spp.). The first eggs were laid on 17 May and the

Table 1

Nesting and Fledging Rates for 45 Defined Male Territories and for the Entire

Study Area

Male

territories

No. of
active nests

Min.

nesting

females

Max.

nesting

females

Entire

study

area®

120

115

53.2^=

X nests or nesting females

—

2.67*’

2.16*’

2.56*^

2.48“

X young fledged

2.55

0.95

1.2

1.0

2.6

® Undetermined number of territories.
** Per male territory.

' Ha.

Nests per ha.

120

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

Table 2

Nesting Parameters for 36 Individually Marked Female Red-winged Blackbirds

Nesting

females

Young

fledged/

female

Females

renesting

X days
from end
of first
nest to
start of
next nest

Females
attempting
second
brood
on area

Renesting

females

that

switched

male

territories

distance

(m)

between

successive

nest-sites'

SY females

8 (22)^

1.5

2 (33)

—

ASY females

28 (78)

2.04

4(17)

1 (6.6)

3(75)

169

Total

1.93

6 (20)

9.6

1 (5.6)

3 (50)

169

“ Parenthetical values are percent of total within category.
” For the 3 females that switched male territories.

last young fledged on 16 July — a nesting season of 61 days. The median dates for nest starts
and fledging were 24 May and 18 June, respectively. The average clutch-size was 3.7 eggs
(mode 4). A total of 137 young was fledged from 48 of 154 (31.2%) active nests. Predators,
primarily raccoons (Procyon lotor), destroyed 99 nests (54.4%). About 80% of all egg mortality
and over 50% of all nestling mortality was due to predation.

Territoriality of marked males. — Most males began territorial defense in late April. Of 5
ASY males captured in the walk-in decoy trap and tagged in early May, 2 later established
territories on the area. Between 12 and 26 May, 30 ASY males were captured on-territory,
in small decoy traps, and tagged. Seven marked males abandoned their territories and were
not seen again. Table 1 gives nesting and fledging rates observed on 45 male territories (25
marked and 20 unmarked males). About 2.6 young were fledged per male territory, a low
reproductive rate compared to most other studies (Dolbeer 1976). One ASY male abandoned
its initial territory and established a second territory 1.4 km away. The first territory con-
tained 1 nest which was depredated with 3 eggs in it on the night of 23-24 May. The second
territory contained 1 nest in which the first egg was laid on 6 June and which fledged 3 young
on 30 June. The nesting female on the first territory was not marked so it is not known
whether or not it moved with the male. Two males in S\ plumage successfully defended
territories that attracted nesting females. One of the S\-male territories fledged 3 young.

Nesting and renesting hy marked females. — Prior to the start of nesting in mid-May, 11
females (5 AS\ and 6S\ ) were captured in large mist nets and marked. Of those, 4 remained
on the study area and 3 (2 ASY and 1 SY) nested, hut no nest was located for the other SY
female. Between 2 and 21 June, 33 nesting females (26 .\SY and 7 SY), with nestlings, were
captured on their nests and marked. Of the 36 marked females that nested, 8 (22%) were
classed as SY birds. The 36 marked females accounted for 42 nests (41 active) on the study
area. One SY female built a second nest after the loss of its first nest, but did not lay a
second clutch. Table 2 gives values for young fledged per marked female. Of 30 marked
females that could have been observed renesting, 6 (20%) renested on the study area. Eighteen
marked females successfully fledged young from nests started on or before 3 June. Only 1
marked female successfully produced a second brood on the study area. Three of 6 females
that renested switched male territories, moving considerable distances in the process. The
phenomenon of territory switching by individually marked, renesting, red-wing females was
also observed by Dolbeer (1976) and Fankhauser (Bird-Banding 35:120, 1964). This study and
that of Dolbeer (1976) suggest that this kind of movement may be common. Renesting and
second-nest values (Table 2) represent minimum estimates since females that left the study

GENERAL NOTES

121

area may have renested elsewhere. Eour cases of renesting where clutch-sizes were known
for both initial and subsequent nests were observed. In all cases, the females involved were
ASY and laid initial clutches of 4 eggs. Only 1 female laid 4 eggs in its second clutch; the
other 3 females each laid only 3 eggs in their second clutches.

Acknowledgments. — I thank the C. K. Blandin Foundation and Wild Rice Growers’ As-
sociation, Inc. for financial support; K. C. Carr and R. S. Wetzel (USFWS) for field support;
H. Jacobson for permission to study the rice paddies; and M. W. Weller (University of
Minnesota) for reviewing the manuscript. This paper is No. 10401, Scientific Journal Series,
University of Minnesota Agricultural Experiment Station, St. Paul, Project 17-92. — Daniel
W. Moulton, Dept. Entomology, Fisheries and Wildlife, Univ. of Minnesota, St. Paul,
Minnesota 55108. (Present address: 70 Edgewood Dr., Central Valley, New York 10917.)
Accepted 1 Nov. 1979.

Wilson Bull., 93(1), 1981, pp. 121-124

Migration speeds of three waterfowl species. — To better understand the physical
nature of bird migration and its energy requirements, it is important to evaluate the speed
at which various species fly, and how speed may be affected by environmental factors. This
paper reports on ground and air speed of migrating flocks of Canada Geese (Branta cana-
densis interior). Lesser Snow Geese {Anser c. caerulescens) and Mallards (Anas platyrhyn-
chos).

In 1966, we began recording ground speeds of migrating waterfowl in central Illinois. By
late 1978, we had obtained 160 records, all but 3 from a car driven for 1.5-24 km parallel
to the birds’ flight. The 3 additional records were obtained from mapping the course and
time interval of migrating birds observed from a light aircraft. Migrating flocks of Canada
Geese composed 79% of the records. Lesser Snow Geese 16% and Mallards 5% (Table 1).

At the time the waterfowl were observed, the wind direction and velocity were estimated.
Direction data were more reliable than velocity estimates, which were based on radio reports
and local clues (flag, foliage, smoke and the like). Most flocks were between 100 and 365 m
above the ground. At those altitudes, wind direction was approximately the same as at ground
level, but average wind velocity was probably higher. The wind force striking the migrating
flocks was vectored on the basis of cosine of the angle of wind to migration track X wind
velocity.

Table 1 shows the ground speed of the migrants, the vectored air velocity assisting or
impeding their passage, the calculated air speed that resulted from the deletion of the wind
force and the statistical significance of the results. Data for the Canada Goose were separated
into fall and spring periods to determine whether the stronger winds in the spring or the
proximity of the wintering grounds were factors affecting the air speed of these geese.

A comparison of ground speed to the vectored wind speed shows that migrating Canada
Geese adjusted their flight speed within certain constraints to compensate for wind velocity.
Although the ground speeds of Canada and Snow geese flying into the wind were reduced,
their effort (as measured by air speed) averaged 13.1 km/h; more when they flew against the
wind than when they flew with it. The F value derived from an analysis of variance dem-
onstrated a highly significant relationship between wind speed and the air speed of Canada
Geese (F = 18.5, P < 0.01 for fall and 20.7, P < 0.01 for spring), but no statistically signif-
icant difference in the Snow Goose (F = 3.6, NS). (The small sample measured in the op-
posed-wind category appears responsible.)

122

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

Table 1

The Effect of Vectored Wind Velocity on the Ground Speed of Migr.\ting
Flocks of Waterfowl in Central Illinois, 1966-1978

No.

flocks

Ground
speed (km/h)

Wind

speed (km/h)

Corr.

coeff. P

Calcu-
lated air
speed
( km/h)

Species

Season

X ± SD

X ±

Canada Goose

fall

72.7 ± 6.4

+9.3 ±

7.6

0.45 < 0.01

63.4

fall

64.7 ± 9.2

-12.2 ±

9.8

-0.69 < 0.01

76.9

spring

70.3 ± 10.0

+ 13.0 ±

10.5

0.22 < 0.10

57.3

spring

60.7 ± 6.0

-12.2 ±

7.6

-0.39 < 0.10

72.9

Lesser Snow Goose

fall

83.3 ± 5.1

+ 12.7 ±

6.6

0.17 < 0.10

70.6

fall

67.3 ± 24.0

-13.4 ±

8.8

-0.77 < 0.10

80.7

Mallard

fall

71.8 ± 11.1

-18.8 ±

12.2

-0.78 < 0.01

90.6

Within each wind force category (Table 1), linear regression demonstrated change in
ground speed with wind speed. As the tail-wind force increased, migrants failed to corre-
spondingly increase their ground speed. The correlation coefficient showed the best linear
fit for the Canada Goose fall data. Spring wind data were more variable than faU data,
perhaps as a result of a greater frequency of strong gusts at that season. Fig. 1 shows the
effect of vectored wind velocity on the ground speed of migrating Canada Geese during the
fall and the ground speed that might be expected without adjustment by the geese for wind
force. The difference between the actual and projected ground speeds suggests that as
favorable winds increase, the birds decrease their air speed and that as head winds increase,
they fly faster. Several different plot tests showed that the relationship was linear rather
than curvilinear.

The air speed of individual flocks of Canada Geese varied between 25 and 88 km/h. The
instance of the highest speed recorded is especially interesting. At 07:15, 10 March 1977,
Glen Sanderson and Bellrose were driving north on 1-57 at Dix, Jefferson Co., Illinois, when
they noticed 500 Canada Geese in 2 flocks migrating due north at an estimated altitude of
300 m. The geese had a ground speed of 96 km/h which they steadily maintained for the
next 24 km. Local weather on the car radio indicated a south wind at 8 km/h. These geese
had apparently just departed nearby Rend Lake, Jefferson Co., on a flight of about 185 km
to a traditional spring concentration point on the llliniois River near Bureau, Bureau Co.

Because of this high air speed at the apparent onset of a migratory flight, we thought that
spring air speeds might be higher than those in the fall because of energy expended on the
longer fall flights. Central Illinois is closer to the winter grounds in southern Illinois than to
James Bay or even Horicon Marsh, Wisconsin, points from which fall flights through Illinois
emanate. However, air speeds of migrants were 4.2 km/h lower in the spring than in the fall
(F = 4.4, P < 0.05). Apparently distance of flight does not appear to be a factor affecting
flight effort.

With a following wind in the fall, the air speed of Snow Geese was 7.2 km/h greater (P <
0.01) than that of Canada Geese. The difference was significant (F = 12.0, P < 0.01). Flying
against the wind in the fall. Mallard air speed also averaged 13.7 km/h faster than Canada
Geese (F = 21.4, P < 0.01).

There is conflicting evidence regarding the effect of wind speed on the air speed of mi-
grating birds. Blokpoel (Can. \^ ildl. Serv. Rept. Ser. 28:1-30, 1974) compared a small num-

GENERAL NOTES

123

Fig. 1. Linear regression of the response of Canada Geese in ground speed to favorable
wind speed and adverse wind speed. Dashed line indicates projected ground speed with no
change in air speed.

her of air speeds with wind speed records for migrating Lesser Snow Geese and concluded
that wind speed resulted in little, if any, change in the air speed. Although Tucker and
Schmidt-Koenig (Auk 88:97-107, 1971) found that air speeds of local birds varied with the
component of the wind (head, tail or cross), they minimized the importance of energy con-
servation resulting from a bird altering its air speed, stating: “The air speeds we measured
are too variable to support the hypothesis that birds fly at closely regulated air speeds to
conserve energy.”

On the other hand, Schnell (Living Bird 4:79-87, 1965) measured the local flight speeds
of birds near the ground by Doppler radar and found that wind velocity affected the air speed
of most birds. Bellrose (pp. 281-309 in Proc. XIV Int. Ornithol. Congr., 1967) noted that the
speed of bird migrants on radar was not proportional to increases in wind speed, and sug-
gested that migrants adjust their energy output in relation to the degree of wind assistance
or resistance. Bruderer and Steidinger (pp. 223-258 in Animal Orientation and Navigation,
S. R. Galler et ah, eds., SP-262, U.S. Gov't. Print Off., Washington, D.C., 1972) used radar
to ascertain that ground speeds of Chaffinches {Fringilla coelebs) in migration did not parallel
increases in wind velocity; their decrease in air speed equalled about one-third the increase
in wind force. In a further study of the flight speeds of 12 species of gulls (Lams sp.), terns
(Sterna sp.) and skimmers (Rynchops sp.) near their nesting colonies, Schnell and Hellack
(Am. Nat. 113:53-66, 1979) obtained additional evidence that bird air speeds usually varied
inversely to wind speeds. Additional work by Tucker (pp. 298-333 in Avian Energetics, R.
A. Paynter, Jr., ed., Publ. Nutt. Ornithol. Club No. 15, 1974) apparently led him to modify
his views on the energy expended in flight: “At both sea level and altitudes of 6000 m, it is
beneficial from an energetic point of view to fly faster into a head wind, and slower with a
tail wind than is the case in still air.”

124

THE ILSON BULLETIN • VoL 93, \o. 1, March 1981

Varying results obtained when comparing air speed of birds with wind speed may be a
function of the wide range of air speeds that birds can use without the undue loss of energy.
Schnell and Hellack (1979) concluded that . . air speeds can be increased or decreased
considerably with only a relatively small increase in metabolic rate or cost of transport." As
shown by Greenewalt (Trans. Am. Philos. Soc. 65:1-67, 1975) cost of transport curves are
relatively flat near their minima. — Frank C. Bellrose and Robert C. Crompton, Illinois
.\atural History Survey, Havana, Illinois 62644. Accepted 19 Dec. 1979.

Wilson Bull., 93(1), 1981, pp. 125-134

ORNITHOLOGICAL LITERATURE

Two diverftent reviews of the followingj title, Nomina Anatomica Avium: an An-
notated Anatomical Dictionary of Birds, have been received by the Editor. Both
are published for the benefit of the readers.

Nomina An.atomica Avium: an Annotated Anatomical Dictionary of Birds. By J.
J. Baumel, A. S. King, A. M. Lucas, J. E. Breazile and H. E. Evans, (eds.), with R. L. Zusi
(consultant for taxonomy) and L. Malinovsky (consultant for classical languages). Academic
Press, London and New York, 1979;xxv + 637 pp. $64.50. — The study of avian anatomy is
pursued by workers in a variety of fields including systematics, neurobiology, veterinary
medicine, physiology, poultry science and others. Communication between workers in dif-
ferent areas, and even in the same area, has long been hampered by the lack of a universal
system of names for the parts of the avian body. This problem has now been remedied by
the publication of the Nomina Anatomica Avium (NAA) after a decade of work by the ap-
proximately 80 scientists constituting the International Committee on Avian Anatomical No-
menclature. The purpose of NAA is to provide a list of terms for the parts of the avian body,
and thereby to advance the anatomical study of birds. Subcommittees dealing with the
different body systems attempted to minimize changes in well-established terms while pro-
viding for each structure a single term that is short, easy to remember and informative.
Topographically related structures are given similar names, and eponyms are avoided. The
nomenclature is in Latin for the sake of international communication.

The book opens with a brief history of the project and information on the use of the NAA,
followed by a series of chapters dealing with the individual organ systems. Few readers will
make use of the whole book, but workers who study avian anatomy for any reason should
find one or more chapters dealing with organ systems relevant to their investigations, and
which will provide them with a standardized terminology for use in publication and other
technical communications. The book is much more than just a collection of terms, however.
It is extensively illustrated with labeled drawings of various anatomical structures, some of
them taken from published research reports, but many of them newly drawn for this volume.
The lists of terms are heavily annotated with explanations of the reasons for the choice of
terms, homologies, synonyms and variations in different groups of birds. There is an exten-
sive list of references establishing the authority for the decisions made in the choice of terms,
and a lengthy index to the terms themselves and their major synonyms. The content of the
book is accurately described by its subtitle.

The ultimate success of this venture will depend upon the extent to which the NAA
nomenclature is adopted by researchers and writers. If the ideal of a unified nomenclature
is to be achieved, it will be necessary for virtually all workers to use the system in their
publications, even when it differs significantly from that to which they are accustomed. This
may pose some temporary inconvenience but should be beneficial in the long run. Most
investigators will probably find that the new nomenclature is not radically different from
older ones because changes were not made for the sake of novelty but only to introduce
clarity and eliminate confusion. Nevertheless, there will be some workers who will find
themselves unwilling to adopt the NAA nomenclature. In such cases it would be useful to
provide a table correlating the terms used with their NAA counterparts. If the reason for
reluctance to adopt NAA terms is substantive then I recommend strongly that the individual
communicate with the appropriate subcommittee chairman (listed in the book) and explain
the objections to the NAA terms. If the reasons for rejection are sufficiently compelling, it

125

126

THE WILSON BULLETIN • VoL 93, No. 1, March 1981

may be possible to change them in a future edition of the NAA. Workers whose investigations
reveal new variations in anatomical structure, or who discover that the current terminology
is based on incorrect or inadequate information should also communicate their discoveries
by sending reprints or comments to the appropriate subcommittee chairman. The first edition
of NAA is not intended to be the final word on avian anatomical nomenclature. It is intended
that revised editions will be prepared in future years so that the work will increase in
effectiveness as a basis for communication among avian anatomists. — Robert J. Raikow.

Nomina Anatomica Avium: an Annotated Anatomical Dictionary of Birds. A sec-
ond review. — The Nomina Anatomica Avium (NAA) is an ambitious work, necessitated by
the nomenclaturally confusing mosaic of over- and under-represented areas confronting avian
morphologists. Eighty contributors labored for more than a decade to produce the NAA,
taking as their objectives the “promotion of international communication by establishing an
agreed list of terms in a universally acceptable language’" and “the advancement of anatom-
ical knowledge of birds.” If these intentions had been met the NAA would indeed be a pearl
beyond price and justify the claim on the flyleaf that “it is an essential work of reference for
all avian scientists and every zoological library.'" Unfortunately, neither objective has been
completely fulfilled and while many sections are excellent, the resultant hybrid may be
problematical enough to dissuade ornithologists from perusing subsequent editions or follow-
ing current recommendations.

There is no question that communication is enhanced by nomenclatural stability, so any
work, such as the NAA, that attempts to provide the framework for standardization is to be
commended. However, if stability is to be maintained in derivative studies, it is imperative
that the syntax of the terminology be intelligible, and this is a facet of standardization for
which little provision has been made in the NAA. In common with other anatomical reference
works (e.g., Nomina Anatomica Veterinaria, Jena Nomina Anatomica, etc.), Latin was the
nomenclatural source chosen by the International Committee on Avian Anatomical Nomen-
clature (ICAAN) for the NAA.

The fact that most ornithologists (including anatomists, who tend to work in their own
vernacular) are unfamiliar with the language was recognized in the introduction to the NAA,
where it was suggested nevertheless that “the Nomina should not be neglected simply be-
cause of the unfamiliarity of Latin. It bears repeating that the official Latin terms should be
used in scientific articles and books in order to enhance international scientific communi-
cation." It may also reasonably be argued that many researchers are already familiar with
the vernacularized Latin of textbooks and that attempts have been made in the NAA to
provide some Latin equivalents at least suggestive of the terms already used in the Romance
languages and English. However, textbook Latin is hardly adequate preparation for the
nomenclature of the N AA, where nouns and associated adjectives mostly exhibit the number,
gender and case of the unadulterated forms. To ornithologists with little or no formal knowl-
edge of the language, the rules which distinguish Os palatinum (palatine bone). Processus
palatinus (palatine process) and Facies articularis palatina (surface which articulates with
the palatine) are not clear. In addition, those unfamiliar with the manner in which word
stems affect the declension of a noun would he hard pressed to recreate the procedure which
derives a nominative plural of tractus from Tractus but musculi from Musculus. As the list
of terms given in the NAA is deliberately not exhaustive, the potential for a new era of
nomenclatural confusion in subsequent studies is obvious. The remedy for this difficulty is
simple enough, and it is hard to rationalize the absence (in a work already comprising 637
pages) of a short appendix which comprehensively lists plural and adjectival forms and which
explains the syntax of the present terminology clearly enough to serve as a guide for future

ORNITHOLOGICAL LITERATURE

127

use. Criticism of the NAA for failure to provide a Latin grammar, however brief, should not
be dismissed as yet another example of pedantic nit-picking. Haphazard or arbitrary ‘Latin-
ization’ of terms by ornithologists anxious to comply with NAA principles, but unable to
locate appropriate sources or to deduce suitable terms from questionable analogues will
promote neither communication nor standardization.

The second intent, that of advancing anatomical knowledge of birds, suffers from the same
theme of critical omission as the first objective. The NAA is subtitled ‘an annotated anatom-
ical dictionary of birds’ giving the misleading impression that the annotations are supple-
mental to the regular dictionary format of comprehensive definition. Although there is some
variation (for example, the chapter on the respiratory system is comprehensively annotated
and illustrated), in general the list of terms which prefaces each chapter is incompletely
annotated, and even fewer structures are figured. As a result, many features are both un-
described and unillustrated. The practice of differential annotation is easily justified when
a system of nomenclature has stabilized sufficiently that some terms require no further
definition. It is an inappropriate principle for any study that attempts to be the definitive
work within a discipline, especially if the raison d’etre for that study has been long-standing
nomenclatural confusion. Non-morphologically oriented ornithologists seeking, for example,
to extend their knowledge of the oral cavity will find the listing of terms such as Radix linguae
or Fenulum lingualis less than informative in the absence of further description. The dic-
tionaries (Donath and Crawford 1969, Kenneth 1966) cited in the introduction as useful in
the development (in the sense of etymology, not syntax) of Latin terms, are limited in the
extent to which they may be used to supplement the NAA. Many terms (e.g., Ovogonium,
Polocytus secundarius) undescribed and unfigured in the NAA are similarly absent in the
references mentioned. In addition, the dictionary by Donath and Crawford is pertinent to
human, not avian, anatomy and there are often radical differences in the meaning of similar
terms. For example, the principal aortic vessel in birds is derived from the right member of
the embryonic fourth pair of aortic arches while the principal vessel in mammals is the left
member of the fourth pair. Extreme caution is therefore required in extrapolating explana-
tions from other dictionaries to undefined terms of the NAA.

Differential annotation is equally problematical for avian anatomists familiar only with the
vernacular terms of their specialties. As the index is exclusively in Latin and as vernacular
synonyms are generally (though again, variably) lacking in the text, a dubious combination
of translation and elimination is required to match undescribed Latin terms with their ver-
nacular equivalents. Anatomical knowledge of birds seems hardly likely to be advanced while
ornithologists are unable to readily locate or identify features of interest. Exhaustive anno-
tation and a vernacular index would do much to eliminate these problems.

Quite apart from the difficulties described above, the NAA contains a number of incon-
i sistencies, errors and ambiguities, illustrated by the following examples. In the chapter on
I muscles, the term ‘Pars’ (to denote a distinct and consistent subdivision of a muscle) is
^ capitalized throughout the list of terms, whereas the lower case is used throughout the

I annotations; in the list of terms the muscles of the jaw are referred to Figs. 2 and 3, which

A in fact illustrate the muscles of the hyoid apparatus and tongue, and Musculus pectoralis is

^ used both to indicate an entire muscle (said to comprise 3 parts: pars subcutanea thoracica,
I pars subcutanea abdominalis and pars propatagialis) and, in Figs. 5 and 6, a prominent,

I previously unidentified, subdivision (? M. pectoralis pars thoracicus, sensu George and Ber-

I ger 1966) ventrocaudal to the M. pectoralis pars propatagialis. Other examples abound: the
k Lamina parabasisphenoidalis (p. 90, annotation 98) was previously referred to as the Lamina
i basiparasphenoidalis (p. 89, annotation 96) while the Tuba pharyngotympanica communis
Ij (p. 89, annotation 94) is shortened to Tuba pharyngotympanica on p. 90 (annotation 98); the
ik Canalis olfactorius and Palatum are not on the pages (108 and 282, respectively) cited in the

128

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

index; the Rostrum sphenoidale is missing from Fig. I of the osteology section, in contrast
to the reference given in the list of terms, and although Musculus subscapularis is described,
no corresponding description is given of Musculus subcoracoideus. The list could be ex-
tended, but the examples given are sufficient to demonstrate that the NAA could have
benefited from more careful proof-reading.

Despite these difficulties, there is valuable information to be gleaned from the NAA by
those ornithologists willing to brave the problems imposed by the format. The notes appended
to the chapter on the respiratory system, for example, are models of informed comment
unimpeded by jargon, while critical analysis of the literature in the light of new (often pre-
viously unpublished) information elevates many other annotations from the realm of simple
description. Many of the illustrations throughout the book are superb, although it must be
admitted that others (e.g.. Fig. 1) seem more likely to amuse than enlighten. Finally, more
than 900 citations in the bibliography ensure adequate extension of the anatomical foundation
laid by the text, regardless of the area of specialty. Given that the problems previously
described can be resolved, the excellence of much of the text and many of the figures
indicates the potential of subsequent editions of the NAA for meeting the objectives of the
current contributors. For the present, let the buyer beware. — R. JOHNSON.

Donath, T. and G. N. C. Crawford. 1969. Anatomical dictionary with nomenclatures
and explanatory notes. 1st English Edition. Pergamon Press, London, England.
George, J. C. and A. J. Berger. 1966. Avian Myology. Academic Press, New York, New
York.

Kenneth, J. H. 1966. Henderson’s dictionary of biological terms. 8th Edition. Van Nos-
trand, New York, New York.

British Birds — A Field Guide. By Alan J. Richards. David & Charles Ltd., North
Pomfret, Vermont, 1979:192pp., 186 color photographs, 186 line drawings. $14.00. — This
is an attractive hook in many ways. Well produced, with good illustrations, clear type on
good paper, and a sturdy binding, it is a pleasure to thumb through. It contains species
accounts of 186 British birds, one to a page, each with a color photograph and a line drawing.
I'he photographs are of excellent quality and accurately reproduced; the adequate, if not
inspiring drawings by Rob Hume supplement the photographs by showing the birds in flight
or in different plumages or poses. The species accounts are divided into five sections: char-
acteristics, voice, habitat, nest and status. Under “characteristics" are included a descrip-
tion, often rather abbreviated, and notes on locomotion, behavior, food, roosting and other
points of general interest. Under “status " are included population figures for each species,
taken from Sharrock's Atlas of Breeding Birds in Britain and Ireland (1976). This is a useful
feature for the layman who may not have access to the Atlas. Some species accounts have
a sixth section, “similar or allied species," where some of the more common British birds
not among the favored 186 are briefly described.

As far as it goes, this is a worthwhile book. But how far does it go? For an ornithologist
or a keen birder, not nearly far enough. The British list contains ca. 475 species, of which
some 270-290 occur every year. A great many have been left out of this hook, thus rendering
it useless to the person who wants to identify every bird he sees. According to the very
skimpy introduction, the hook is aimed at “those thousands of people interested in the birds
in their gardens and local parks, in the countryside or on the moors ..." etc.; in other
words, the armchair or junior-intermediate birder who has perhaps graduated from the bunny
slopes hut is not yet ready for the advanced runs. The 186 species treated are described as

ORNITHOLOGICAL LITERATURE

129

“those which might be seen in Britain without too much effort over the course of a year,”
although the choice sometimes favors distinctive or photogenic species over those less well
endowed. The drab Rock Pipit (50,000 breeding pairs) only appears under “similar or allied
species,” whereas the Dotterel (100 pairs). Red Kite (30 pairs) and Osprey (several pairs) are
accorded full treatment.

In spite of its title, this is only a partial field guide to British birds, and it will certainly not
replace any of the existing works. It is a book for the educated layman. My advice to readers
of The Wilson Bulletin is: if someone gives it to you for Christmas, accept and enjoy; hut
don’t spend hard-earned money adding it to your library. — Stuart Keith.

Hawks and Owls of North America. By Donald S. Heintzelman. Universe Books, New
York, New York, 1979:197 pp., 68 photographs (8 color), 4 figs., 1 table. $18.50. — While the
title of this book recalls the excellent but out-of-print classic Hawks of North America by
Dr. John B. May, and its successor North American Birds of Prey by Alexander Sprunt, Jr.,
the present book is less a systematic treatment than a more popular book addressed, in the
author’s words, “to raptor enthusiasts at less than the professional level.” Its chief feature
is its display of photographs, some in color, of most of our raptors. The informal text of
discursive, rather than ordered systematic accounts, is a vehicle to show more of the rapidly-
growing files of hawk photographs. Many of these are excellent (but that of a disheveled
captive White Gyrfalcon is not). Some half of these photographs are by the author, but the
rest are by many different photographers. The author acknowledges his debt to many
sources, but for the most part there is little to indicate sources of information. (Who supplied
the information about the subspecies of Black Hawk which lives “in southern Florida”?)

There are separate chapters on vultures, kites, accipiters, soaring hawks, etc., and the
individual species are afforded a variable selection of information and comment, but I find
little here to commend it to other than the general reader. The species accounts are followed
by chapters on matters of current interest, such as endangered species, habitat loss, ecology
(an over-worked term in this text), migration, conservation and chemical pollution.

At recent meetings of Raptor Research, the number of attending raptor enthusiasts doubled
that of A.O.U. and Wilson Society meetings. Many of these have seemingly little other
contact with wildlife biology, but my acquaintance with them leads me to believe they are
more sophisticated than the audience to which this book is addressed. It seems unfortunate
that we in North America do not have a volume in any way comparable to Leslie Brown’s
superb British Birds of Prey. — Walter R. Spofford.

The Peregrine Falcon in Greenland: Observing an Endangered Species. By James
T. Harris. University of Missouri Press, Columbia and London, 1979:255 pp., 39 plates, 1
map. $15.95. — This is neither a scientific treatise nor the definitive monograph on Green-
land’s Peregrine Falcons {Falco peregrinus). A “narrative of a summer’s research in Green-
land,” it is much more than an ordinary diary. Harris has written an unusual and fascinating
book that should be a must for all collectors of good bird books.

The title and subtitle should be reversed, as the book deals more with the adventure of
observing falcon cliffs, their birds and their weather than with the Peregrine Falcon in
Greenland. Thus, at first glance, I was disappointed: I expected graphs, tables and statistics.
Many of these data are included, woven skillfully into the personal narrative of the Greenland

130

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

experience of 1972. It is difficult, however, to find them quickly: the index is short and
emphasizes the work of other authors rather than Harris’ own observations. Apart from
these slight disappointments for the raptor expert there is much to admire in this handsome
book, and we must congratulate the publisher for an excellent and esthetic production. The
cover design (by Jerry Dadds), the selection of black-and-white photographs and even the
italic print of the whole book enhance the author’s objectives and style. Harris attempts to
blend three different stories together: the search for falcon eyries (partly a travelogue and
diary), the observations of falcon activities and behaviors at one eyrie (the proper topic of
the book) and “the development of my feelings toward the species and the individual falcon”
(the most difficult part to write).

The book serves as an excellent introduction to the tundra ecosystem and to the care and
preparation that must go into such a wilderness study. Thus, I recommend it highly as
background reading for anyone bound for the Arctic. Songbirds, mammals and other raptors
are not only described but also followed through the seasonal cycle as long as they occur
within the “falconscape.”

The peregrine story is two-fold. The author makes an admirable attempt to provide ex-
tensive background notes on the species. He begins by recounting the general population
decline and its inter-relationship with the use of pesticides, then introduces the species and
subspecies concepts, and talks about distribution, population surveys and other topics. Fi-
nally, we learn about captive breeding at Cornell up to 1977, and the latest breeding data for
Greenland. These sections come somewhat abruptly and distract from the ongoing narrative.
Yet, for most readers, Harris has written an excellent summary of the complex peregrine
problem.

J'he account of breeding activities is rather general and contains only the occasional “bon-
bon” for the specialist, e.g., the observation of “hop hunting” Lapland Longspurs (Calcarius
lapponicus) on the ground. This is not mentioned in the index. The interested reader must go to
various journals where scientific papers by some of Harris’ companions have been published
in the meantime. Thus, the jacket's promise, “Harris' reporting is a model of scientific
accuracy,” is certainly exaggerated. Today, the term scientific accuracy should mean more
than occasional behavioral anecdotes.

Finally, the most difficult part of this review, an evaluation of Harris’ feelings and his
writing about them. Everyone who has watched birds in the wilderness will share some of
the conscious and unconscious thoughts and moods that he felt sitting day after day amidst
flowers and mosquitos. Occasionally, however, he slips into metaphors and statements too
grandiose for this context. Examples are: “A sense of kinship had crept upon me, a feeling
that we shared the difficulty of living, and their presence, their successes, even the humming
of moscpiitoes, encouraged me.” (p. 63); “The mammals and I seemed to share an immediate
recognition. An illusion took hold of me, that we had a mutual understanding of and interest
in our parallel lives, because we all had fur or four-chambered hearts or some bond.” (p.
64); “Most oologists alive today are aging retired men with rich memories.” (p. 85); “Almost
always humans are individuals. Animals are not.” (p. 193).

There are many dialogues in this book that do not sound at all like field ornithologists
talking in the field. They are too clean, too literate compared to my experience of many
years. J'his introduces some element of artificiality, and quite a bit of distance from such
writers as Matthiessen and .\bbey, to whom Harris is compared on the book jacket. Still, we
learn quite a bit about the author. He immerses himself in the tundra wilderness, becomes
extremely protective of the falcons, and finds himself ill at ease with the sudden appearance
of human visitors. Harris — in one of the best dialogue areas of the book — is caught breath-
lessly almost stunned by the sudden confrontation with another world, and escapes back into
the tundra night.

ORNITHOLOGICAL LITERATURE

131

This beautifully produced book is a successful and rare (albeit not perfect) blend of or-
nithological adventure, scientific information and personal touch. It is w^ell worth its price. —
Hartmut Walter.

Shorebirds in Marine Environments. By Frank A. Pitelka (ed.). Studies in Avian
Biology, No. 2, Cooper Ornithological Society, 1979:261 pp., 70 figs., 52 tables. $8.00. — This
volume is a collection of papers from a symposium on shorebirds held at the 1977 meeting
of the Pacific Seabird Group. The symposium, organized by F. Pitelka, was composed of
two sessions: one on distribution, migration and conservation (15 papers); and the other on
ecology of shorebirds (5 papers, 4 abstracts). The objectives of the symposium were to look
at distribution, migration and ecology of shorebirds in terms of the basic information currently
available, and to examine shorebird biology in terms of the conservation and management
of coastal wetlands. Pitelka sets the stage for the discussion by examining shorebird distri-
bution on the Pacific Coast. Such a larger view of an entire coast is essential since shorebirds
are long-distance migrants. Pitelka lists the new world shorebirds, and comments on tbeir
status on the Pacific Coast. He then analyzes the world shorebird fauna (fig. 1). One partic-
ularly helpful figure shows the occurrence by five degree latitude intervals of shorebird
species in North and South America. He briefly outlines tbe current problems in shorebird
distribution: the importance and occurrence of staging areas, migration behavior (group
dynamics), age and sex differences in patterns, and winter site tenacity. His introduction is
an excellent starting point for the collection of papers, and for graduate students interested
in shorebirds.

The papers on distribution, migration and conservation cover a wide range of topics in-
cluding censuses of restricted areas and wide-ranging areas, habitat use, timing of migration,
migration patterns of particular species and tbe evaluation and conservation of coastal wet-
lands. Some papers merely report on census techniques; illustrating the kinds of information
available from large-scale censuses (see Prater), or in-depth censuses of small bays (see
Gerstenberg). Other papers illustrate the importance of particular bays for migrating shore-
birds. The Copper River Delta is a critical habitat for migrating shorebirds as some 20 million

I shorebirds pass through this area each spring (see Isleib). The delta is particularly important
to species such as Western Sandpiper (Colidris mauri) and Dunlin (C. alpina) that forage
j there and lay down fat reserves for continued northward migration (Senner). Senner found
I that Dunlin migrate collectively, shift from one place to another as flocks, and show greater

I weight gain than do Western Sandpipers that migrate independently of each other in less

) organized flocks while in the Copper River Delta. The information provided by Senner is

5! critical to protecting these areas against human activities (such as oil spills). Similarly, Gill
t and Jorgensen present quantitative data on shorebirds' use of another Alaskan bay.

The paper by Page et al. is an excellent example of tbe information that can be gained by
censuses for many years. The mass of data fall into species patterns, and provide insights
■I into habitat requirements and site tenacity. Many species returned to the same foraging
1 areas year after year. Winter feeding site fidelity was also found by Smith and Stiles, who
banded shorebirds in 2.5 years at a mudflat in Costa Rica.

Two excellent papers on migration patterns deserve special mention: Jehl's on the autum-
' nal migration of Baird’s Sandpiper (C. bairdii), and Harrington and Morrison's on migration

' of Semipalmated Sandpiper (C. pusilla). Jehl's is an innovative examination of the migration

pattern of Baird’s Sandpiper using specimens from 35 museum and university collections
(if I count the acknowledgments correctly!). The project involved a good deal of work and
imagination, and the end results of the country-wide survey indicate sexual and age differ-
ences in the migration pattern of this species. Adults may migrate some 9000 miles in 5

132

THE WILSON BULLETIN • Vol. 93, Vo. I, March 1981

weeks, whereas juveniles migrate more leisurely over a broad front (Jehl). Although I would
have liked to know exactly how many specimens were examined, the paper illustrates how-
museum specimens can be used to answer ecological questions.

Harrington and Morrison's paper on the Semipalmated Sandpiper uses a variety of tech-
niques (banding studies at James Bay and Massachusetts, and examinations of museum
specimens) to show that different populations have different morphological characteristics
that allow for identification of the source of migrants. The paper clearly indicates that a
variety of strategies are employed by one species while migrating. Their migration pattern
is a very complex system in which breeders from different geographical areas use different
routes and their patterns vary seasonally. Such large-scale, cooperative studies can provide
information necessary for an overview of migration and habitat use that is simply not possible
by individual investigators working by themselves in one bay.

Two other papers in this section (Speth, Goss-Custard) comment on the management of
wetlands, and the effects of habitat loss on overwintering shorebirds. The paper by Goss-
Custard is excellent in that it makes predictions about the food base of shorebirds, and
provides data relevant to them. JehlN concluding remarks are concise and point out topics
requiring extensive work.

The second part of the book deals with the ecology of shorebirds and includes papers on
flocking behavior, winter and summer ecology, feeding ecology-, energetics, and territoriality
of wintering shorebirds. This section is mainly papers on particular research topics with one
or a few species. As such, the papers are excellent, and provide good reviews of topics,
quantitative data, and fruitful discussions. The inclusion of the abstracts was frustrating, as
I wished to see the entire papers.

Strauch and Abele's paper on feeding ecology of plovers wintering in Panama was partic-
ularly interesting in illustrating that some species feed only at low tide whereas others are
tidally independent. The study by Myers, Connors and Pitelka on territoriality in non-breed-
ing shorebirds is a landmark study showing that a wide variety of species will defend terri-
tories during the winter, although its expression differs among species, individuals and hab-
itats. These data suggest that extensive cost-benefit analysis of foraging may lead to more
(juantitative predictions of the point at which particular species will stop defending territo-
ries. (b)ss-Custard's analysis of the energetics of foraging Redshanks {Tringa totanus) is an
excellent example of an examination of these cost-benefit considerations. J. Wiens’ con-
cluding remarks point out some of the areas that need further research, making a plea for
long-term studies.

Altogether this volume is essential for all ornithologists, particularly those working with
shorel)irds, migration patterns, foraging behavior, territoriality, habitat selection and con-
servation. Although papers vary from brief essays without quantitative data, to hypothesis-
testing and data-based papers, the overall quality is excellent. The papers provide docu-
mentation on several important aspects of shorebird biology (i.e., age and sex differences in
migration patterns, site tenacity during the winter, winter territoriality). In the very least
they provide quantitative data on migration patterns essential to understanding the impor-
tance of coastal wetlands. I was disappointed in the lack of any papers on non-Pacific coast
areas (except for Harrington and Morrison). Although it would have taken a lot of time, an
index would have made this volume more useful for beginning graduate students or others
interested in tracing particular ideas (for example, feeding site tenacity over the winter).
.'Similarly, the volume was a long time in coming out, which resulted in some key papers not
being cited. All in all, I cannot recommend this volume more highly. And for $8.00, it's
clearly the bargain of the year. — Joanna Burger.

ORNITHOLOGICAL LITERATURE

133

A Field Guide to Western Birds* Nests. By Hal H. Harrison, illus. by the author
unless otherwise credited; map, endpapers and logo by Mada Harrison. Houghton Mifflin
Company, Boston, 1979:xxx + 279 pp., 32 color plates with captions, 153 black-and-white
photographs, 1 map, glossary. $11.95 (hard cover). — Here is another attractive addition to
The Peterson Field Guide Series — the eagerly awaited companion volume to Hal Harrison's
I 1975 A Field Guide to Birds’ Nests of 285 species found breeding in the United States east
of the Mississippi River. The latest work represents a more exhaustive review of the literature
than the earlier one; included are all species, even the casual or rare ones, known to breed
in the contiguous United States west of the Mississippi River. For nearly each of the 520
species there is a succinct description of its breeding range, habitat, nest, eggs and other per-
tinent notes. This handy pocket-sized book is packed full of practical information that I am al-
ready putting to good use.

The color illustrations are excellent despite their tiny size (only 40 x 53 mm). The nests
centered in the 256 color photographs are so graphically portrayed that one can identify
many of them without using the captions; the same can be said of the eggs whose markings
and colors for the most part are clearly defined and lifelike. The eggs especially are so well
illustrated and described that one wonders why the author failed to include them in the
book’s title. However well done, I find it somewhat disappointing to look at photographs of
' nests situated in artificial nest boxes, but so many birds these days raise broods in human-
contrived sites that the author logically illustrated a number of them. If that unavoidable
little problem over natural vs artificial sites disturbed me, I find it hard to fault the numerous
black-and-white photos — all of very high quality and nicely placed throughout the text.

Harrison’s use of the Mississippi as a boundary between east and west — a method devised
earlier by Olin Sewall Pettingill, Jr. — seems to work well. Inasmuch as I spend a fair share
of my time on the upper Mississippi, I can vouch for Harrison’s accuracy for this region. I
also note that he left few stones unturned in other areas of the country familiar to me, e.g.,
Kansas with its many unusual records, including that remarkable Harris’ Hawk {Parabutea
unicinctus) nesting far beyond the species’ usual breeding range. The few places where I find
omissions in the text are readily referrable to unpublished material — hardly the fault of the
author! By and large, Harrison’s careful coverage of the huge block of 22 states appears to be
nothing less than outstanding.

One is impressed by the fact that nearly all of the photographs were taken by the author.
Only those who know how difficult it is to find the nests of a good many of our birds will
appreciate Harrison’s special talents, perseverance and endurance. I personally have seen
i so few nesting birds west of the Rocky Mountains that I hope to spend my retirement days
I chasing them down in the far west. One reliable reference that will accompany me will be
li this western guide. — David F. Parmelee.

i North American Ducks, Geese and Swans. By Donald S. Heintzelman. Winchester
Press, New York, New York, 1978:xiv -I- 236 pp., color and black-and-white photographs,
t wildlife refuge maps. $15.00. — A Guide to North American Waterfowl. By Paul A.
Johnsgard. Indiana Univ. Press, Bloomington and London, 1979:274 pp., color plates, black-
and-white drawings, range maps. $15.95.

In these volumes we have a mixed bag of information concerning the waterfowl of North
America. One offers an invariably sparse treatment of field recognition, size, flight habits
, and ranges in North America, and the second a relatively complete coverage of each species

134

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

(food habits, range maps, behavior, etc.)- Both books contain numerous photographs of
varying quality.

Heintzelman states his purpose as helping hunters and birders to identify waterfowl and
to highlight selected national and state wildlife refuges where waterfowl are prominent. The
species accounts are brief and general with most of the limited text devoted to field recog-
nition. Habitat for the Gadwall {Anas strepera) is “ponds, lakes, rivers, freshwater marshes”
and its North American range described as “southern Canada and the United States.” I
doubt that the comments under the rubric “flight style” (e.g. , “direct and rapid,” “graceful
and rapid,” etc.) will really be of much aid to even novice observers. The photographs
accompanying the species accounts are black-and-white, and average in quality, but a se-
lection of better quality color photographs graces the book’s centerfold. The author employs
a taxonomic scheme of subfamilies that is generally outdated in usage (e.g., Aythyinae) but
otherwise has presented current nomenclature for the species.

A naive chapter devoted to “Techniques of Waterfowl Study” tells us that “a pair of
binoculars is very helpful to people looking at waterfowl,” mentions that telescopes are also
useful (“a minimum magnification of 20x is necessary”), notes that blinds are also helpful,
since hunters have long used them, and sketches the vicissitudes of common and scientific
names and “the species problem.” Subsequent chapters briefly — some are scarcely more
than two pages in length — describe eclipse plumages and hybrids, migrations (with flyway
maps), prairie pothole breeding grounds and local viewing areas. The final chapter is more
extensive, and presents maps and succinct descriptions of waterfowl refuges in the U.S.
Eour appendices treat accidental sightings, Canadian wildlife areas, conservation organiza-
tions and homes for Wood Ducks (Aix sponsa).

The book contains some errors (e.g., Hockbaum for Hochbaum in the suggested reading
section) and poor layout in some places (e.g., the figure on p. 226 describing construction of
predator guards for Wood Duck nest boxes has no caption or cross-reference with the as-
sociated text on p. 220). On balance, this is not a book for ornithologists and, considering
the competition available, it may be of marginal value (at $15.00) to anyone.

Johnsgard's prolific pen adds yet another waterfowl book to his long list of credits. To be
sure, this is a shortened version of his treastise Waterfowl of North America (1978), as the
author freely admits. The range maps and “in hand,” “in field” identification sections are
unchanged for the most part, whereas a “natural history” section condenses the more de-
tailed material appearing in the earlier volume. The condensed text remains a useful refer-
ence to habitat, foods, behavior, breeding and conservation for each species, but without
extensive literature citations. Most species accounts are accompanied, however, by at least
one suggested reading and, while some of these may be overly general or indeed quite
narrow, it is not easy to assign a single reference or two with equal coverage for the likes of
a Mallard (Anas platyrhynchos) or, conversely, a Steller's Eider {Polysticta stelleri).

The color plates are largely the same as appeared before but some new drawings (the
author's) are included. Sketches of head profiles (where plate numbers are missing but the
sketches are cross-referenced with a numbered species list) are new and supplement the
dichotomous key repeated from the 1978 volume.

Johnsgard intended this volume to reach the “middle ground” of his reading audience,
those whose needs lie between heavier works and a simple field guide. Perhaps one might
have wished for a slightly smaller-sized book, with soft cover, to make it more field-worthy,
but I believe Paul Johnsgard adequately has met his goal, and 1 can recommend the book,
as is, for its stated purpose. — Eric C. Bolen.

Wilson Bull., 93(1), 1981, pp. 135-136

ORNITHOLOGICAL NEWS

ANNUAL MEETING— THE WILSON ORNITHOLOGICAL
SOCIETY, 1981

The 62nd annual meeting of The Wilson Ornithological Society will be held on the campus
of Mount Allison University, Sackville, New Brunswick, Canada, 4-7 June 1981. This will be
only the second Wilson Ornithological Society meeting to be held in Canada, and the only
one to feature excursions in 3 provinces. The Maritimes feature small towns and largely
undeveloped surroundings, with a leisurely pace of life that should provide a relaxing change.

The meeting will be hosted by the Canadian Wildlife Service — Atlantic Region, Mount
Allison University, and the Chignecto Naturalists Club. Information concerning accommo-
dations, transportation, excursions and related matters, and a call for papers will be mailed
to the society membership in early March 1981. The deadline for submission of abstracts
is 15 April 1981.

Chairman of the local arrangements committee is Anthony J. Erskine, Canadian Wildlife
Service, P.O. Box 1590, Sackville, New Brunswick, Canada EOA 3C0. Chairman of the
program committee is Jerome A. Jackson, Department of Biological Sciences, P.O. Drawer
GY, Mississippi State University, Mississippi State, Mississippi 39762 U.S.A.

EDITOR’S ACKNOWLEDGMENTS

The processing of manuscripts received by The Wilson Bulletin during 1980 was greatly
facilitated by careful evaluations by more than 75 persons who served as referees. I and
i the rest of the editorial staff are most grateful to them for their constructive criticism on
behalf of our contributing authors. I also thank all authors for their courteous cooperation
with our staff as we work together to produce the best papers, both with respect to scientific
I content and quality writing. The regular editorial staff and others who provided assistance in
• 1980 were: Dave Ankney, Melinda Barlow, Tim Barlow, Keith Bildstein, Gary Bortolotti,

David Brooks, Ann Crabtree, Brete Griffin, Rosemary Johnson, Dot Richardson, Jim Rising,
i| Richard Snell, Pat Urquahart and especially Margaret May and Nancy Flood. We also
appreciate the fine efforts in our behalf from the staff at Allen Press, who direct their
>1 best energy to our journal. Thanks also for support from Jerry Jackson, Ken Parkes, Bob
i; Storer and George Miksch Sutton. — ^Jon C. Barlow, Editor

MEMORIAL NOTE

» On 21 December 1980, Helen Van Tyne, widow of Josselyn Van Tyne, died in Carmel,
1 California. During her life, most of which was spent in Ann Arbor, Michigan, she was very
1 active in juvenile court matters, Mrs. Van Tyne donated her late husband’s extensive library
( ‘ to The Wilson Ornithological Society, where it is now part of the library which bears his name.

135

136

THE WILSON BULLETIN • Vol. 93, No. 1, March 1981

AARON M. BAGG STUDENT MEMBERSHIP AWARDS

Student Membership Awards in The Wilson Ornithological Society have been made avail-
able through funds generously donated in memory of Aaron M. Bagg, a former president
of the society. The Student Membership Committee has designated the award recipients
for 1981 as follows: Douglas A. Bell, Westfalische Wilhelms-University, Munster; Thomas
Bicak, University of Montana; Scott P. CarroU, University of Oklahoma; Dominick A.
Della Sala, Wayne State University; Sylvia L. Halkin, University of Wisconsin; Elizabeth G.
Henderson, University of South Carolina; William J, Hilton, Jr., University of Minnesota;
Mark L. Hoffman, University of Florida; Charles T. LaRue, Northern Arizona University;
David M. Lyons, Frostburg State College, Maryland; Kelvin G. Murray, University of Florida;
James S. Quinn, University of Oklahoma; Roland L. Redmond, University of Montana; David
E. Seibel, University of Kansas; David A. Spector, Stockton State College, New Jersey; and
Kathy A. Winnett-Murray, University of Florida.

FIFTH ANNUAL MEETING OF THE COLONIAL WATERBIRD GROUP

The Colonial Waterbird Group’s fifth annual meeting will be held 22-25 October 1981. A
symposium on the factors affecting reproductive success in colonial birds is also planned;
papers may be submitted to Colonial Waterbirds, the Group’s new journal. Symposium
abstracts should be sent to J. Burger, Dept. Biology, Livingston College, Rutgers University,
New Brunswick, New Jersey 08903 by 1 August 1981; regular session paper abstracts
should be received by J. Burger by 1 September 1981. For information on registration,
contact Brian Chapman, Dept. Biology, Corpus Christi State University, Corpus Christi,
Texas 78412.

SOUTHEASTERN COASTAL AND ESTUARINE BIRDS
CONFERENCE-WORKSHOP

This conference-workshop will be held 11-13 September 1981 at the field laboratory of
the Belle W. Baruch Institute for Marine Biology and Coastal Research, University of
South Carolina, near Georgetown. This meeting will focus on a total ecosystem approach
and will assemble past and present federal, state and academic researchers from Delaware,
Maryland, Virginia, North and South Carolina, Georgia and Florida in an attempt to deter-
mine levels and directions of current research. Results of research on coastal and/or
estuarine species, avian prey bases, avian competitors, etc. can be presented in conven-
tional or poster format. Graduate students are especially welcome. A limited number of
graduate student awards (room and registration fee waivers) are available. For information
on presentations and student awards write Keith Bildstein, Dept. Biology, Winthrop Col-
lege, Rock Hill, South Carolina 29733 or phone (803) 323—2111. Information on registra-
tion and housing will be mailed to A.O.U., W.O.S. and C.O.S. members in the southeast.
Others wishing to be placed on the mailing list should write Bobbie Christy, Baruch Field
Lab., P.O. Box 1630, Georgetown, South Carolina 29440 or phone (803) 546-3623.

This issue of The VI i I son Bulletin was published on 14 May 1981.

The Wilson Bulletin

Editor Jon C. Barlow

Department of Ornithology
Royal Ontario Museum
100 Queen’s Park

Toronto, Ontario, Canada MSS 2C6
Assistant Editor MARGARET L. May
Senior Editorial Assistants Gary Bortolotti
Nancy Flood

Editorial Assistants Keith L. Bildstein Richard R. Snell

C. Davison Ankney James D. Rising

Review Editor Robert Raikow Color Plate Editor William A. Lunk

Department of Biological Sci- 865 North Wagner Road

ences Ann Arbor, MI 48103

University of Pittsburgh
Pittsburgh, PA 15260

Suggestions to Authors

See Wilson Bulletin, 87:144, 1975 for more detailed “Suggestions to Authors.”
Manuscripts intended for publication in The Wilson Bulletin should be submitted in tripli-
cate, neatly typewritten, double-spaced, with at least 3 cm margins, and on one side only of
good quality white paper. Do not submit xerographic copies that are made on slick, heavy
paper. Tables should be typed on separate sheets, and should be narrow and deep rather
than wide and shallow. Follow the AOU Check-list (Fifth Edition, 1957) and the 32nd Sup-
plement (Auk, 90:411^19, 1973), insofar as scientific names of U.S. and Canadian birds are
concerned. Summaries of major papers should be brief but quotable. Where fewer than 5
papers are cited, the citations may be included in the text. All citations in “General Notes”
should be included in the text. Follow carefully the style used in this issue in listing the
literature cited; otherwise, follow the “CBE Style Manual” (1972, AIBS). Photographs for
illustrations should have good contrast and be on gloss paper. Submit prints unmounted and
attach to each a brief but adequate legend. Do not write heavily on the backs of photographs.
Diagrams and line drawings should be in black ink and their lettering large enough to permit
reduction. Original figures or photographs submitted must be smaller than 22 x 28 cm.
Alterations in copy after the type has been set must be charged to the author.

Notice of Change of Address

If your address changes, notify the Society immediately. Send your complete new address
to Ornithological Societies of North America, % Sandra L. L. Gaunt, Box 21160, Columbus,
Ohio 43221.

The permanent mailing address of the Wilson Ornithological Society is: do The Museum
of Zoology, The University of Michigan, Ann Arbor, Michigan 48109. Persons having business
with any of the officers may address them at their various addresses given on the back of
the front cover, and all matters pertaining to the Bulletin should be sent directly to the
Editor.

Membership Inquiries

Membership inquiries should be sent to Dr. Robert C. Whitmore, Division of Forestry,
West Virginia University, Morgantown, West Virginia 26506.

CONTENTS

OBSERVATIONS OF SEABIRDS DURING A CRUISE FROM ROSS ISLAND TO ANVERS ISLAND, ANT-
ARCTICA Robert M. Zink \

DISPLAY BEHAVIOR OF OVENBIRDS {Seiurus aurocapillus) II. SONG VARIATION AND SINGING

BEHAVIOR M. Ross Lein 21

THE MAYFIELD METHOD OF ESTIMATIIS[G NESTING SUCCESS: A MODEL, ESTIMATORS AND SIMU-
LATION RESULTS - Gary L. Hensler and James D. Nichols 42

NARROWLY DISJUNCT ALLOPATRY BETWEEN BLACK-CAPPED AND CAROLINA CHICKADEES IN

NORTHERN INDIANA Peter G. Merritt 54

TOE FUSION IN OSCINES George A. Clark, Jr. 67

ENVIRONMENTAL EFFECTS ON ROOSTING BEHAVIOR OF CHIMNEY SWIFTS

Richard M. Zammuto and Edwin C. Franks 77

GENERAL NOTES

OBSERVATION OF A BROOD OF SHARP-SHINNED HAWKS IN ONTARIO, WITH COMMENTS
ON THE FUNCTIONS OF SEXUAL DIMORPHISM

Helmut C. Mueller, Nancy S. Mueller and Patricia G. Parker 85

FOOD DEPRIVATION AND TEMPERATURE REGULATION IN NESTLING FERRUGINOUS HAWKS

Diana F. Tomback and Joseph R. Murphy 92

AERIAL “play” OF BLACK VULTURES Walter A. Thurber 97

THE SHOULDER-SPOT DISPLAY IN RUFFED GROUSE Allan Garbutt 98

THE AGONISTIC REPERTOIRE OF SANDHILL CRANES

Stephen A. Nesbitt and George W. Archibald 99

NOTES ON THE SLENDER ANTBIRD Edwin O. Willis and Yoshika Oniki 103

NOTES ON THE UNIFORM CRAKE IN COSTA RICA F. G. Stiles 107

TRICHOMONIASIS IN BALD EAGLES Ward B. Stone and Peter E. Nye 109

Protocalliphora INFESTATION IN BROAD-WINGED HAWKS

Scott Crocoll and James W. Parker 110

HERRING GULL ATTACKS AND EATS ADULT MALE OLDSQUAW Richard R. Snell 110

RED-LEGGED KITTIWAKES FORAGE IN MIXED-SPECIES FLOCKS IN SOUTHEASTERN ALAS-
KA Douglas Siegel-Causey and Thomas E. Meehan 111

GROUND-FEEDING METHODS AND NICHE SEPARATION IN THRUSHES Alan Tye 112

AMERICAN COOT DISTRIBUTION AND MIGRATION IN COLORADO

Warner P. Gorenzel, Ronald A. Ryder and Clait E. Braun 115

REPRODUCTIVE RATE AND RENESTING OF RED-WINGED BLACKBIRDS IN MINNESOTA

Daniel W. Moulton 119

MIGRATION SPEEDS OF THREE WATERFOWL SPECIES

Frank C. Bellrose and Robert C. Crompton 121

ORNITHOLOGICAL LITERATURE 125

ORNITHOLOGICAL NEWS 135

COMP. ZOOL
ARY

VOL. 93, NO. 2 JUNE 1981 PAGES 137-300