5^.
HARVARD UNIVERSITY
Library of the
Museum of
Comparative Zoology
H E
MC7
'"£5 2 J 1995
GREAT BASIN
NATURALIST
VOLUME 56 NO 1 — JANUARY 1996
BRIGHAM YOUNG UNIVERSITY
GREAT BASIN NATURALIST
Editor
KiciiAHi) \V. Baumann
29()MLBM
PO Box 2()2()()
Bri,i;;hani Yoiinjji; University
Provo, VT s'Ki()2-()200
801-378-5053
FAX 801-378-3733
Assistant Editor
Nathan M. Smiiii
190MLBM
PC) Box 26879
Brigham Young University
Provo, LIT 84602-6879
801-378-6688
E-mail: NMS(a)HBLLl. BYU.EDU
Associate Editor:
MiCllAKI. A. liOWKliS
HlaiuK l'A|HMiiiK'iilal Farm, Univcrsitx' of
Virginia, liox 175, Honxv, \'A 22620
J. H. ('Al.I.AllAN
Museum of SoutliwvsttMii Biology, University of
New Mexico, Alhuquerciue, NM
Mailing address: Box 3140, I lemet, CA 92546
JKI'IKKY J. JOIIANSI'.N
Department olliiologv jolm (Carroll llnixiMsity
University Heights, Oil'} 1 1 IS
Boius C KoxnuAriKi'i-
Department ol Isntomology, (Colorado State
Uiiiversit\, i'ort Collins, CO 80523
PaulC. Mahsii
Center for Kn\ ironmental Studies, Arizona
State Universitx, lempe, AZ 85287
SiANLKY D. Smith
Department ol Biology
University of Nevada-Las Vegas
Las Vegas, NV 89154-4004
Paui.T. TUEl.l.KK
Department olEnxironmental Hesonrce Sciences
University oINevada-lUMio, 1000 \alle\ Hoad
Reno, NV 89512
KoiU'Hi (;. WiirrMOHi'.
Division ol I'ori'stry, Box 6125, West Virginia
Uni\ersit\, Morgantown, \\'\" 26506-6125
Kdilorial Board, jerran T l'"liiiders, C^liairman, Botan\ and Range Science; Duke S. Rt)gers, Zoology;
VViUbrd M. Hess, Botany and Range Science; Richard R. Tolman, Zoology. All are at Brigham Young
Uni\ersit\. Kx Ollicio lulitorial Board members include Steven L. Taylor College of Biolo,g> and Agriculture;
11. Dnant- Smith, Director, Monte L. Bean Life Science Museum; Richard \V. Baumann, Editor Crcat Basin
Naturalisf.
The (Ural Hdsiii Ndtnnilist, founded in 1939, is puMished cinarterK !)>■ Brigham Young University.
I ni)ul)lished manuscripts that hutluM- our biological undcMstanding ol the Creat Basin and surrounding areas
in \\('st("rn Nortli America are accepted lor publie;ition.
Subscriptions. Annual snbscriiilions to the Clrcdt lidsiit Ndiiinilist for 1996 are $25 for indixidual sub-
scribers (.$.30 outside tlu- Uniti'd States) and $50 for institutions. The price of single issues is $12. .\11 back
issues are in i)rint anil ;i\ ailable for sale. .Ml matters pertaining to subseri|itions, b;ick issui>s, or other busi-
ness should be ilireeted to the P.ditor C.irat Basin Ndtwdlist, 290 MLBM, PO Box 20200, Brigham Young
UnivtMsilN, l'ro\(), UT 84602-0200.
Sclu)hirly Kxchanges. l,ibraries or other organizations interested in obtaining the Clrcdt Bdsin
NdtnralisI through a continuing exchange of seholarK publications should contact the E.xchange Librarian,
6385 IIBLL, I'O Box 26889, Brigham Young UniMisily Provo, UT 84602-6889.
Editorial Production Staff
joAunc Abel 'iechnii'al l^ilitor
jail Sptiieer Assist;iiit to the lulitor
(.iopNiiulit (i'^ 19% In Brigham Young I'niversity
OIReial piihlkation ilate; 31 January U)9(i
ISSN 0017-3614
1-96 750 16974
The Great Basin Naturalist
1^1 BIJSIIi:i) Al PlU)\(), U TAII, 15V
BUICMIAM Y()LIN(; UNIVKliSriY
ISSN ()()17-;]()14
VOIAMK 56
31 January 1996
No. 1
Groat Basin Naliiralist 56(1), © 1996, pp. 1-1 1
TEMPORAL AND SPATIAL DLSrUlBUTlON Ol ilK;ilWAY
MORTALITY OF MULE DEER ON NEWLY CONSTRUCTED ROADS
AT JORDAN ELLE RESERVOIR, UTAH
Laura A. Horniii' and jolin A. Bissoiictte^
AliSTKACT: — ill tliis pap rela-
tionships. There is limited information that
broadK characterizes mule deer use and kill
distributions on and near highway systems, or
that has investigated the influence of physical
1 an d scape featu res .
Study Area
The study area is located in the valley
between the Wasatch and Uintah mountain
ranges of northeastern Utah; the Provo River
originates in the Uintah mountains and bisects
the valley floor. Segments of three highways —
1995]
ITlCIIW AY MOKTALITV DISTRIBUTIONS OF DRER
US 40, state routes (SR) 32 and 248, totaliiit^
47.3 km on the eastern slope of the Wasateh
mountains in nortlieastern LI tali — were ehosen
for study. Construction of the roadways was
completed in 1989 and was necessitated by
inundation of existing roads following con-
struction of Jordanelle Reservoir. Filling com-
menced in spring 1993.
Dominant valley habitats consist of mesic
meadow, riparian areas, and pasture lands.
Surrounding drainage slopes are predomi-
nantly within a mountain brush and sage-
brush-grass zone (6000-7000 ft elevation), with
scattered pinyon pine and juniper. Limited
stands of aspen, cottonwood, and willow occur.
Mule deer utilize the area as year-long range
but usually are forced into the valley bottom
during winters with heavy snowfall.
Methods
Deer roadkill data were collected at least
once per week by research personnel from 15
October 1991 to 14 October 1993. UDOT and
UDWR personnel assisted with collection
efforts during their daily activities. Date, high-
way identification, and location of each kill
were recorded to the nearest 0.10 mile. Deer
initially were classified as adult or fawn;
incisors were removed from adult deer for age
determination by cementum annuli proce-
dures (Low and Cowan 1963).
Deer kill zones and nonkill zones were des-
ignated based on 1991-1993 deer-highway
mortality locations. A minimum of 5 kills per
mile had to have occurred for a segment of
roadway to be considered a kill zone. A kill
zone ended when a section of road did not
contain a kill for more than 0.10 mile. We ran-
domly selected 4 kill zone and 4 nonkill zone
paired locations of 0.10-mile length along each
highway and established transects to evaluate
respective road alignment and associated habi-
tat features.
We recorded the distribution of kills over
the entire study area, average traffic volume
and speed for each highway percent vegeta-
tive cover, and topography proximal to area
roads. Kills were recorded to the nearest 0.01
mile. Twice monthly spotlight counts were
conducted to document deer use and density
adjacent to study area roads. Counts were ini-
tiated at dark; each count averaged 3.2 h {s =
21 min). We began the spotlight run on a dif-
ferent route each night. We drove along both
sides of each road at a speed of 45-50 kph and
used a hantlheld 400, 000 candlepower spot-
light to locate deer. Deer were located to the
nearest 0.10 mile. We stopped when deer
were spotted to identify sex and age class, dis-
tinguishing fawns by size. The activity of deer
spotted in the right-of-way was classed as
feeding, bedding, walking, or standing. We used
statistical correlations to compare deer road-
kill locations between years and with locations
of live deer.
Rangefinder readings were recorded at each
0.1-mile interval to provide an estimate of ob-
servable area along each road (Fafarman and
DeYoung 1986). Mountainbrush habitats de-
creased deer visibility, and some areas along
roads were not visible from a vehicle due to
roadside rock cuts or steep declines bordered
by concrete barriers. From numerous spotlight
runs, we calculated the mean maximum visi-
ble distance to be 500 m.
Deer snow track counts were recorded
along the right-of-way once each during the
winters 1991-92 and 1992-93 to evaluate deer
approaches to the roads. We counted the num-
ber of trails within each 0.10-mile intei^val and
described them as either parallel or perpen-
dicular to the road. A parallel trail continued
its direction for at least 30 m.
Road alignment, right-of-way width and
slope, right-of-way vegetation, and vegetation
composition were characterized to a perpen-
dicular distance 100 m beyond the right-of-
way fence. Each highway was classified as
either 4-lane or 2-lane with passing lanes.
UDOT recorded traffic speed and volumes for
each road during 2 periods: 11 March to 15
March 1992 and 29 June to 5 July 1992. Road
alignments at each selected kill and nonkill
transect location were described as cun^e, hill,
or straight section. A cui-ve or hill was consid-
ered part of the road alignment if it was within
100 m of the transect. Deer further than 100 m
from the road are unlikely to be involved in an
immediate collision (Romin and Dalton 1992);
thus, beyond this distance, a hill or curve that
would have reduced driver visibility had less
significance.
We analyzed habitat features during Sep-
tember 1993. Stereoscopic aerial photography
(1:24,000) was used to describe habitat features.
We placed a transparent grid over photographs
to determine percent cover (mountain brush
Great Basin Naturalist
[Volume 55
XJS40
30
M "'
I 1^
^ 10
O
20
M 1^
10
3
.a.
(^
SR248
1^
1
-#
10 11 12 13
10 11 12 13
SR32
1
3S
30
2S
:s 15
eS 10
^^ s
o
1.
^
J
Fig. 1. Distribution of deer kill (%) by mile marker on 3 nevvlv built highways (US 40, SR 248, SR 32) at Jordanelle
Reservoir, Utah, 1991-1993.
and riparian areas) and topographical features
at deer-highway mortality locations beginning
at the road and extending 1.2 km distant. At
each paired kill and nonkill location we estab-
lished 3 habitat transect lines aligned perpen-
dicular to the road. The transects were spaced
100 m apart and extended through the right-
of-way zone 100 m past the right-of-way fence.
We measured the length of each habitat along
each transect line. Habitats included right-of-
way revegetation, mountain brush, sagebrush-
grass, grass-forb, aspen, cottonwood, willow,
agricultural pastureland, riparian, and river.
We calculated the proportion of each habitat
present along the combined transect lines for
each kill and nonkill location.
We identified roadkill and live deer loca-
tions, as well as descriptixe roadside features
to 0.1 mile, consistent with highway mile
marker delineation. We con\erted to metric
units for analysis where appropriate.
Results
Deer locations
We documented 397 deer roadkills during
the stud\ from 15 October 1991 to 14 October
1993; 278 (5.9 kills/km) kills occurred during
1995]
Highway Mortality Distributions of Deer
Kill Zone \
::$$>» Drainage Location
Heber
1 Mile
Fig. 2. Location of kill zones and associated drainages at Jordanelle Reservoir, Utah, 1991-1993.
the 1st year of study (15 October 1991 to 14
October 1992), and 119 (2.5 kills/km) were
documented during the 2nd year (15 October
1992 to 14 October 1993). Highway US 40
sustained the highest kill levels: 68% during
the 1st year and 55% during the 2nd yean
State routes 248 and 32 sustained similar kill
levels; during the 1st year we recorded 18%
and 14% of the total deer-highway mortality
on SR 248 and SR 32, respectively. During the
2nd year we recorded 25% of the total annual
kill on SR 248 and 19% on SR 32. Deer kills
averaged <20 before the roads were relocated.
Nineteen deer kill zones were identified
based on the spatial distribution of deer road-
kills during both years (Figs. 1, 2). The mean
length of kill zones was 1.0 km {s = 0.62).
Deer- vehicle collisions along US 40 occurred
most frequently between mile markers 6.0 and
9.0 during both the 1st (56%) and 2nd (48%)
years of the study. Twenty-eight percent of
deer roadkills along US 40 occurred from mile
marker 7.0 to 7.9 during the 1st year. Roadkill
locations were correlated between years along
US 40 at both the 1.0-mile (r = 0.69, P =
0.03) and 0.10-mile (r = 0.56, P < 0.001)
interval. Deer kill locations were not signifi-
cantly correlated between 1st and 2nd years
along SR 32 at either the 1.0-mile (r = -0.14,
P = 0.70) or 0.10-mile (r = 0.004, P = 0.968)
6
Great Basin Naturalist
[Volume 55
scale. Deer kill locations along SR 248 were
significantly correlated at the 1.0-niile intenal
(r = 0.72, P = 0.02) but not at the 0.10-mile
inten^al (r = 0.18, P = 0.07).
Deer spotlight counts were not signifi-
cantly correlated to kill locations at the 1.0-
mile interval for any road during either year:
SR 248 year 1 (r = 6.43, F = 0.19), year 2\r =
0.17, P = 0.61); SR 32 year 1 (r = 0.42, P =
0.23), year 2 (r = 0.12, P = 0.73); US 40 year 1
(r = 0.51, P = 0.14), year 2 (r = 0.15, P =
0.68). However, positive correlations were
stronger during the first year.
Fort>' percent of spotlighted deer were seen
on the right-of-way. We identified the behav-
ior of 968 (55%) of the deer along the right-of-
way. Thirty-three percent were standing when
first observed, 32% were feeding, 12% were
bedded, and 23% were walking along the right-
of-way or crossing the road.
Perpendicular snow tracks were not corre-
lated with deer-highway mortality locations (r
= 0.29 , P = 0.42). Parallel tracks constituted
48% and 32% of all deer trails counted during
the 1st and 2nd years, respectively.
Traffic Characteristics
Traffic characteristics contributed to deer-
highway mortality levels (Table 1). Highway
US 40 had the highest (3.7-9.9 times) mean
24-hr traffic totals of the 3 study area roads.
Mean traffic speed was highest along US 40
(69.3 mph) from 11 March to 15 March 1992;
however, over the 4 July weekend (29 June-5
July 1992), average speed along SR 248 (59.1
mph) was slightly higher than along US 40
(58.9 mph). Volume and speed were somewhat
higher along SR 248 than along SR 32 for both
test dates.
Highway US 40 is a 4-lant> road and SR 248
and SR 32 are 2-lane roads with occasional
passing zones. Road alignment (Table 2) was
similar for transect kill and nonldll zone loca-
tions (x^ = 1.2, df= 2, P = 0.70).
Habitat
From aerial photographs (1:24,000) we deter-
mined that percent cover was greater along
US 40 (63%) than along SR 248 (28%) or SR 32
(31%). Designated kill zones had higher mean
percent cover (40%) than nonkill zones (29%).
Highway deer kill along US 40 was highest in
an area (mile markers 6.0-9.0) of 88%) vegeta-
tive cover during both the 1st (56%) and 2nd
Table 1. Traffic speed and volume of new routes at Jor-
danelle Reser\'oir, Utiih, 1992.
Speed (mph)
Date
Location
Mean
Maximum
Vehicles/hr
11 Mareh-
US 40
69.3
76.0
172.2
15 March
SR248
56.9
72.0
37.9
SR.32
54.0
68.0
17.3
29 June-
US 40
58.9
68.0
264.6
5 July
SR248
.59.1
63.8
71.4
sr;32
.55.0
68.0
.37.8
(48%) years of study. Low mortality occurred
in predominantly sagebrush-grass/wet meadow
(mile markers 4.0-5.0) or agiicultm-al zones (mile
markers 12.3-12.9) with <20%) cover. Along
SR 248, agricultural zones sustained 1 deer
(1%) mortality during the 2-yr period. State
route 32 sustained 28% of its total deer road-
kill in agricultural areas. However, 50% of this
kill occurred at mile marker 9.0, located in a
riparian area at an agricultural pasture and
cliff interface. During spotlight censuses we
observed a larger proportion of deer along
right-of-ways associated with mountain brush
habitat than along agricultural areas (Table 3).
Paired / tests of microhabitat features showed
no significant difference in proportion of cover
100 m beyond the fence between kill and
nonkill locations {t = 0.13, df = 13, P = 0.90).
Proportion of cover on the right-of-way ne\'er
was higher than 29% for any transect.
We examined 19 kill zones and 19 nonkill
zones in the study area for associations with
drainages (Fig. 2). Since deer-vehicle collisions
occuned along nearly all of US 40, we evaluated
the 8 highest kill locations along this road.
Major drainages intersected the roads in 16
(79%) kill zones. Along US 40, large drainages
intersected the highway at 6 (75%) of the kill
locations. Two kill zone locations along US 40
weie at highway overpasses (mile markers 4.0
and 8.0); drainages were located within 0.2
miles. Two other kill zones extended past high-
way underpasses (mile markers 8.2 and 11.4)
Seven (37%) nonkill zones had drainages inter-
secting the roads. Howe\er, in 4 of the nonkill
zones, drainages were within 0.2 miles of a kill
zone.
Kill and nonkill locations did not differ in
right-of-way widtlis (^ = 1.1, df = 13, P = 0.30).
Deer kill per km was greatest when right-of-
way areas were inclined rather than declined
or level (Table 4).
1995]
HiciiwAY Mortality Distiubutions of Deer
Tablk 2. Road alitiniiu'iit at paiivtl (n — 42) kill and
nonkill locations along stiid\ areas routes at Jordanelle
Rcsenoir, Utali.
Cunt-
Straight
Hill
Kill
Nonkill
15
19
23
21
1.2, df'=2. f = 0.70.
T.\BLE 3. Deer observed (% of total deer) along right-of-
\\a>'S associated with agricultural or nioinitain bnish habitat
t>pes.
Habitat
US 40
SR 248
SR32
Agricultural
Mountain brush
49
19
40
23
44
juvenile
2.5 3.5
AGE CLASS
Fig. 3. Deer-highway mortality ages classes {n = 198),
Jordanelle Reservoir, Utah, 1991-1993.
Table 4. Deer kill per km relative to right-of-way slope
relief along both sides of study area roads at Jordanelle
Resen'oir Utah, 1991-1993.
Road
Right-of-way
US 40
SR 248
SR32
No incline
Incline 1 side
Incline 2 sides
6.7
22.3
17.1
0.9
6.8
9.3
2.6
7.1
10.6
Temporal Deer Roadkill
Distributions
During winter 1991-92, mean monthly
snowfall totaled 7.7 cm; mean monthly winter
snowfall for 1992-93 was 46.9 cm. Of 397 deer
mortalities documented during the study from
15 October 1991 to 14 October 1993, we clas-
sified 205 (51.6%) does, 75 (18.9%) bucks, 86
(21.7%) fawns, and 31 (7.8%) unknown. Sixty-
four fawns (16.1%) were female and 22 (5.5%)
were male (Fig. 3). There was a 57% decrease
from 278 (5.9 deer/km) deer roadkills during
the 1st year to 119 (2.5 deer/km) roadkills dur-
ing the 2nd year. We determined the age of
198 (70.7%) adult deer by cementum annuli
techniques. Sixty-seven percent (n = 133) adult
kills were < 2.5 yr old. The oldest recorded
deer roadkills (2.5%) were 6.5 yr old. The 1992
hunter buck hai-vest from the Kamas district,
east of the study area, also indicated a young
population (n = 85); 55% yearlings, 15% 2.5 yr
old, and 30% > 3.5 yr old (M. Welch, UDWR,
personal communication).
We located 4378 deer on 39 spotlight trips
driving a total of 1845 km. There was a 64.2%
decrease from an average 14.6 deer/km^ in the
1st year of the study to 5.23 deer/km^ during
the 2nd year UDWR estimated a similar 70%
reduction in the deer population on the Kamas
District, attributed to the harsh 1992-93 win-
ter (M. Welch, UDWR, personal communica-
tion). We identified sex and age of 1515 (34.6%)
spodighted deer: 987 (65.2%) does, 136 (8.9%)
bucks, and 392 (25.9%) fawns. We calculated
an obsei'vable area unobstructed by roadside
barriers or dense vegetation of 10.98 km^ for
the study area.
We identified monthly and seasonal peaks
in deer mortality (Table 5) by phenological
period: fall (September-November), winter
(December-Februaiy), spring (March-May),
and summer (June-August). The following anal-
yses treat the study period as year 1 (15 Octo-
ijcr 1991-30 August 1992) and year 2 (1 Sep-
tember 1992-14 October 1993), to allow inter-
pretation of seasonal deer distributions and
roadkill patterns. The highest roadkill peak
(25%) occurred during No\'ember 1991. Thirty
percent of the mortality in year 1 occurred
during the fall even though data collection did
not begin until 15 October 1991. Another peak
(33%) was evident during the summer of year
1; 15% of the mortality for the year occurred
in July. A similar fall peak (52%) occurred dur-
ing year 2; 20% of the mortality occurred in
October and 19% in November. A relatively
large peak (18%) occurred in April. Eleven
percent of the mortality occurred during the
summer. During year 1, 41.8% of the annual
Great Basin Naturalist
[Volume 55
Table 5. Seasonal roadldll distributions (%) for each deer
class at Jordanelle Resenoir, Utah, October 1991-August
1993.
YearI
Fall
\\'inter
Spring
Summer
Doe
30.0
16.0
10.2
44.0
Buck
14.5
27.3
16.4
41.8
? fawn
57.4
2S.6
11.4
2.9
6 fawn
47.0
40.0
13.0
0.0
Year 2
Doe
65.4
6.2
13.6
14.8
Buck
52.6
10.5
26.3
10.5
9 fawn
50.0
21.4
28.6
0.0
6 fawn
40.0
40.0
0.0
20.0
buck mortality and 44.8% of doe mortality
occurred during summer (Table 5). Fawn mor-
tality peaked for both males (47%) and females
(57.4%) in the foil. During year 2, the highest
mortality among all sex and age classes occuned
during fall.
Seasonal distributions of deer-highway mor-
tality' were compared to observed deer densi-
ties during the same periods. Seasonal deer
densities and highway mortalities were not
significantly correlated (r = 0.54, P = 0.14)
over the 2-year period (Fig. 4). For the period
of summer 1992 to summer 1993, deer-high-
way mortality and deer population density
were strongly correlated (r = 0.94, P < 0.01),
suggesting a density-dependent relationship. A
negative coirelation existed between deer den-
sities and kill/density (r = -0.68, F = 0.06).
During year 1, observed deer density was low
during fall (5.4 deer/km^) and winter (9.9 deer/
kni^) while highway mortality was high (fall =
71 deer, winter = 58 deer). Deer density (2.41
deer/ km^) and highway mortality (18 deer)
were low during the 2nd winter Following
winter 1992-93 deer density adjacent to study
area roads increased slightly during spring
(3.3 deer/km^) and summer (3.8 deer/km^).
Observed density never reached pre-winter
levels. Highway mortality levels of deer also
increased (n = 31) in spring 1993 but did not
return to pre-winter levels. Kill as a function
of density was lower than observed deer den-
sity from winter 1992 to winter 1993 but
exceeded density following the harsh winter of
1992-93 (Fig. 4)^
The roadkill buckxloe ratio during fall
(22.9:100) and early winter (78.9:100) of year 1
was greater than that observed in the living
population (fall = 6.7:100, winter = 4.4:100)
during the same periods (Table 6). Likewise,
30
25
20
^ .
•s^ 10
I
I
I
I
rl
□ Year 1
^ Year 2
I
!
I
Sep Oct Nov Dec
Jan
Feb Mar Apr May Jun Jul
MONTH
Fig. 4. Monthly deer4nghway ni()rtalit\ at Jordanelle Reserxoir Utah. 1991-1993.
Aug
1995]
lllCIIWAV MOKTALITV DiS THIIRITIONS OF DEER
Tabi.K (i. Seasonal hiickidoi' latios ol roadkill ami spolliglil cicci- at JorclaiR'lk' He,Sfi"v<)ir, Utah, Ottohcr 1991 -October
1993.
Se
asons'
•
Counts'
F91
W91
Sp92
Su92
F92
W92
Sp93
Su93
Kill
Spotlight
22.9
6.7
7S.9
4.4
75
2.9
44.2
31.3
18.9
5.6
40.0
0.0
45
13.3
16.7
12.5
■'Kill and spotliiiht count.s are recorded as l)ucks:100 does.
''Winter counts include only December and earK- Jannan'; sprini; counts include only April and May. Bucks arc prolialily underrcpresented.
the roadkill huck:cloe ratio during the fall of
year 2 (18.9:100) was larger than the ratio of
the living population (5.6:100). The summer
l:)uck:doe ratio was similar for roadkill and liv-
ing populations during both years. For the
months June-November 1992, the correlation
coefficient between number of fawns involved
in vehicular collisions and number observed
on spotlight runs was significant: r = 0.84 {P
= 0.04). For both summers the fawn:doe ratio
of road-killed animals was 8.3:100, higher than
the observed fawn: doe ratio (1.4:100) of the
living population.
Discussion
We distinguished aspects of deer mortality
based on traffic volume, habitat, topography,
and seasonal distribution. Traffic volume signif-
icantly influenced overall deer mortality levels.
Though total kill in the study area decreased
by 57% fi-om the 1st to the 2nd year, roadkills
remained higher along US 40 than either SR
248 or SR 32. The 4-lane alignment of US 40
contributed to higher deer kills. Traffic vol-
ume was higher and deer-vehicle collisions
occurred more frequently along SR 248 than
along SR 32 during both years.
Vegetative cover along the length of US 40
was greater than along state routes 248 or 32.
Likewise, percent cover was higher for desig-
nated kill zones compared to nonkill zones.
High percent cover appears to attract deer to
right-of-ways for foraging. Agricultural areas
provide abundant forage away from roadsides
and were associated with low deer-vehicle
collision levels. Deer usually approached roads
along drainages, and higher kill levels occurred
near large drainages.
The ability to predict kill locations requires
that kill locations remain similar over time. Kill
location correlations at the 0.10-mile interval
were low for SR 248 and SR 32 between the 2
yr The kill locations along US 40 were signifi-
cantly correlated; however, most of US 40 was
considered a continuous kill zone, which
would lead to a correlation simply by coinci-
dence.
Although drainages provide highway ap-
proaches, it is not possible to predict with exact-
ness where deer-car collisions will occur based
on habitat (% cover) and topography proximal
to the roads. Deer often move parallel along
the right-of-way after approaching a road. How-
ever, inclined right-of-ways flinneled deer along
the highway and were associated with higher
kills. Low correlations between spotlight and
kill locations further suggest that deer did not
immediately cross the roads where they entered
right-of-way areas. Snow trail counts also indi-
cated parallel movement of deer.
While seasonal deer-highway mortality dis-
tributions tracked large fluctuations in popula-
tion levels, behavior associated with life his-
toiy activities of deer, e.g. fawning, breeding,
and migration, also influenced year-round road-
kill levels and composition. During the 2-yr
study period, both roadkill and observed deer
density levels decreased. When harsh winter
conditions (1992-93) reduced population lev-
els, deer-highway mortality was proportionally
lower.
Variability in the association between live
deer density and roadkill numbers can be attrib-
uted in part to deer-use patterns. Between fall
and spring of year 1, highway mortality de-
creased and spotlight counts recorded increased
deer density. The mild winter that year allowed
deer access to large areas and they maintained
residence higher on drainage slopes. Weather
conditions did not force deer to remain near
area roads, although tliey fiequenfly approached
and crossed roads. We attributed the initial
increase in deer density during spring 1992 to
the approach and congregation of deer along
right-of-ways for foraging.
Fall peaks in deer-highway mortality ap-
peared related to activities associated with
10
Great Basin Naturalist
[Volume 55
25
DEEDING
SEVERE W
UNTING SEASON
100
<«'' ^'^'^
# # ^* #
SEASON
Fig. 5. Seasonal deer-highway mortahty (no.) and den-
sity (deer/km2) at Jordanelle Reservoir, Utah, 1991-1993.
hunting and breeding during this time (Fig. 5).
Deer were moving around the study area
more frequently than during other seasons.
Proportionally more bucks were involved in
vehicular collisions during the fall than were
obsei"ved in the population. The lireeding sea-
son of mule deer in Utah begins the last few
days of October, peaks between 20 November
and 2 December, and declines through Janu-
ary (Robinette and Gashwiler 1950). During
the study, Utah deer and elk hunting seasons
occurred from late August through October (T.
L. Parkin, UDWR, personal communication).
Fawns were involved in deer-vehicle colli-
sions most often during the fall and least often
during the summer of both years. The fawning
period for mule deer in Utah begins appro.xi-
mately 5 June, reaches and maintains a peak
11-20 June, and declines through 15 August
(Robinc>tte and Gashwiler 1950). Fawns are seen
inirec^uently during their first 6-8 wk because
their predator defense is based on a "hider"
strategy ((ieist 1981). Fawns were absent in
the observed population during the sunnner
but appeared during the fall.
Does were involved in collisions and ob-
served more frequently than males during
both years. Si.xty-eight percent of adult deer
roadkills were does, while 70% of fawns were
female during year 1. Similarly, 81% of adult
deer killed were does and 87.5% of fawais were
female during year 2. Does have heavy energy
demands associated with gestation, parturi-
tion, and lactation, which may explain their
association with high-(jualit>' roadside vegeta-
tion and subscciuent high mortalitv rates.
Management Recommendations
Certain topographic features and vegeta-
tion characteristics associated with roads, cou-
pled with deer movement dynamics, predis-
pose mule deer to highway mortality. Highway
alignment and right-of-way topography often
function to funnel deer to the right-of-way and
encourage movement of deer along the high-
way corridor, creating the potential for colli-
sions at numerous locations. Roads planned in
high deer-use areas that will sustain high traf-
fic volumes should be prioritized for mitigative
procedures during planning. Mitigative tech-
nologies, particularly fencing with crossing
stiiictures, should focus on the initial approach
of deer to the highway along large drainages
and take into account deer spatial dynamics
and population trends.
Continuing studies designed for species-
specific and habitat-specific conditions may
fiuther an understanding of why deer-vehicle
collisions occur on a spatial and temporal
basis, and promote development of appropri-
ate pre-construction designs and mitigation
strategies.
Acknowledgments
We thank die United States Bureau of Recla-
mation (BOR), Utah Department of Transpor-
tation (UDOT), Utah Division of Wildlife
Resources (UDWR), and the United States
Fish and Wildlife Senace (USFWS) for fund-
ing and support provided throughout this
study. We extend a special thanks to Lariy B.
Dalton (UDWR), whose efforts made imple-
mentation of this study possible. We sincerely
appreciate the efforts of personnel who
assisted with roadkill data collection: UDOT
(Kamas maintenance shed crew: Shane W.
13ushell, Doug C. Gines, Ken L. Moon, Tyler K.
Page, and Dave H. Sundquist) and Delmar C.
Waters, a private contractor. G. David Cook,
Justin L. Dalton, Larry B. Dalton, and Herb
C Freeman provided valuable assistance dur-
ing spotlight counts.
Literature Cited
Ai.LKN, R. E., AM) D. R. McCl;li,()Uc;ii. 1976. Deer-car
accidents in .southern Michigan. Journal of Wildlife
Management 40: 317-325.
Basiiore, T. L., W. M. Tzilkowski, and E. D. Bellis.
1985. AnaKsis ol deer-vehicle collision sites in
1995]
Highway Mortai,i iy Disti{ii5Uti()ns of Deer
11
PcnnsyKaiiia. Journal of Wildlife Manaut'iiK'iit 49;
769-774.
Bkii.is, E. D., and H. B. Gkavics. 1971. Deer mortality on
a Pennsylvania interstate highway. Jonrnal of
Wildlife Management 35: 232-237.
Carbaich, B., J. P Vaughan, E. D. Bellis, a.nd H. B.
Graxes. 1975. Distribution and activity of white-
tailed deer along an interstate highway. Journal oi
Wildlife M anagement 39: 570-58 1 .
DusEK, G. L., R. J. Mackie, J. D. Herrices, Jr., and B. B.
CoMPTON. 1989. Population ecology of white-tailed
deer along the lower Yellowstone River. Wildlife
Monographs 104: 1-68.
Fafarman, K. R., and C. A. DeYoung. 1986. Exaluation
of spotlight counts of deer in south Te.xas. Wildlife
Society Bulletin 14: 180-185.
Geist, V. 1981. Behavior: adaptive strategies in mule deer.
Pages 157-223 in O. C. Wallmo, editor, Mule and
black-tailed deer of North America. University of
Nebraska Press, Lincoln.
Goodwin, G. A., and A. L. Ward. 1976. Mule deer mor-
tality on Interstate 80 in Wyoming: causes, patterns,
and recommendations. USDA Forest Service
Research Note RM-332. Rocky Mountain Forest and
Range Experiment Station, Fort Collins, CO. Pages
1-4.
Jahn, L. R. 1959. Highway mortality as an index of deer
population change. Journal of Wildlife Management
2: 187-196.
Kasul, R. L. 1976. Habitat factors associated with mortal-
ity of southern Michigan wildlife on an interstate
highway. Unpublished master's thesis, Michigan
State University, East Lansing. 39 pp.
Kramer, A. 1971. Notes on the winter ecology of mule
and white-tailed deer in the Cypress Hills, Alberta,
Canada. Canadian Field-Naturalist 85; 141-145.
. 1973. Interspecific behavior and dispersion of two
sympatric deer species. Journal of Wildlife Manage-
ment 37; 288-300.
Kress, M. J. 1980. The effects of habitat on the distribu-
tion of white-tailed deer {Odocoileus virginianus)
along a Pennsylvania interstate highway. Unpublished
doctoral dissertation, Pennsylvania State University,
University Park. 56 pp.
Low, W. A., and I. M. Cowan. 1963. Age determination of
deer by annular stnacture of dental cementmn. Jour-
nal of Wildlife Management 27; 466—471.
Mansfield, T. M., and B. D. Miller. 1975. Highway
deer-kill district 02 regional study. Caltrans internal
report. Sacramento, CA. 49 pp.
Myers, G. T. 1969. Deer-auto accidents; serious business.
Colorado Outdoors 18; 38-40.
Pi'.EK, F W, and E. D. Bellis. 1969. Deer movements
and behavior along an interstate highway. Highway
Research News 36: 36-42.
Puc;lisi, M. J., J. S. Lindzey, and E. D. Bellis. 1974. Fac-
tors associated with highway mortality of white-
tailed deer. Journal of Wildlife .Management 38:
799-807.
Reed, D. F. 1993. Efficacy of methods advocated to reduce
cervid-vehicle accidents: research and rationale in
North America. Colorado Division Wildlife Resources,
Fort Collins, CO. 13 pp.
Reeve, A. F 1988. Vehicle-related mortality of mule deer
in Nugget Canyon, Wyoming. Wyoming Coopera-
tive Fisheries and Wildlife Research Unit, Laramie.
75 pp.
Reilly, R. E., and H. E. Green. 1974. Deer mortality on
a Michigan interstate highway. Journal of Wildlife
Management 38: 16-19.
ROBINETIE, W. L., AND J. S. Gashwiler. 1950. Breeding
season, productivity, and fawning period of the mule
deer in Utah. Journal of Wildlife Management 14:
457-469.
RoMiN, L. A., AND J. A. Bissonette. In press. Deer-vehicle
collisions: nationwide status of state monitoring
activities and mitigation efforts. Wildlife Society
Bulletin.
RoMiN, L. A., AND L. B. Dalton. 1992. Lack of response
by mule deer to wildlife warning whistles. Wildlife
Society Bulletin 20; 382-384.
SICURANZA, L. P 1979. An ecological study of motor vehi-
cle-deer accidents in southern Michigan. Unpub-
lished master's thesis, Michigan State University,
Lansing. 63 pp.
Utah Division of Wildlife Resources. 1992. Utah big
game annual report. Reported yearly summar>' of
deer-highway mortality 1982-1992. Utah Division
Wildlife Resources, Salt Lake City.
Vaughan, J. R 1970. Influence of environment on the
activity and behavior of white-tailed deer (Odocoileus
virginianus). Unpublished doctoral dissertation, Penn-
sylvania State University, University Park. 73 pp.
Received 9 May 1995
Accepted 9 October 1995
Great Basin Naturalist 56(1), © 1996, pp. 12-21
EXCEPTIONAL FISH YIELD IN A MID-ELEVATION UTAH TROUT
RESERVOIR: EFFECTS OF ANGLING REGULATIONS
Wayne A. Wurtsbaughl, David Barnard^ and Thomas Pettengill^
Abstract. — We used creel surveys to evaluate how a change from a 6-mon to a year-round fishing season affected
the sport fish harvest in East Canyon Reservoir (Utah), a 277-ha mesoeutrophic system. Under the year-round season,
fishing effort was 840 angler-h-ha-^-yr-l, and 360 trout ha~l were captured. Catch rates were proportional to estimated
trout densities in the resei"voir, ranging fi^om 1.06 during the winter ice fishery, to 0.18 fish angler~lh~l in July. Ninety-
nine percent offish hai-vested were rainbow trout {Oncorhijnchus mijkiss). Thirty-two percent of the 300,000 75-mm fin-
gerling trout stocked annually were captiu-ed by anglers within 2.5 yr, but return rates varied with the strain and/or size
of trout stocked. Annual fish yield was 102 kg/lia, among the highest yet reported for a temperate zone, lacustrine sys-
tem. Extending fishing from a 6-mon season to year-round increased the number of fish captured and provided almost
twice as many hours of recreational fishing in the reservoir The harvest period was changed from traditional
spring-simimer months to primarily a winter-spring fisher\' because relatively few trout sui-vived for more than 6 mon
after reaching hanestable size. Although salmonid production in East Canyon Resei^voir is veiy high, the fishery is in a
precarious state because high primaiy producti\'it\' dri\'en, in part, b\ cultural eutrophication, makes water quality sub-
optimal din-ing midsummer
Key words: reservoir, yield, trout, creel, harvest, strains, regulation, productivity, fish, management, growth,
Oncorhynchus mykiss.
Important goals of lake and reservoir man-
agement are to maximize both fish yield and
recreational use. Methods of increasing yield
include introducing different species or strains,
lake fertilization, and modifying fishing regu-
lations (Hall and Viin Den Avyle 1986, Stock-
ner 1992). Modification of littoral zone escape
habitat may also be important (Wurtsbaugh et
al. 1975, Trendall 1988, Tibor and Wurtsbaugh
1991). Changes in fishing regulations, how-
ever, offer a manager the most flexibility (Carl-
ton 1975), and these changes are less likely to
damage the ecosystem than are the other meth-
ods. In 1985 the State of Utah changed from a
6-mon open season for trout (late May-
November) with a daily limit of 8 fish, to a
year-round fisheiy with no seasonal closures
and a daily limit of 8 fish. To investigate how
this management change affected the fisher);
we conducted a 1-yr creel survey in 1986 to
determine timing and magnitude of harvest
from East Canyon Reservoir: we then com-
pared these results with harvest characteris-
tics measured in the reservoir in 1970 and
1972 under the 6-mon regulation. The 1986
creel survey also allowed us to measure the
high fish yield of the reservoir and to relate it
to various limnological parameters affecting
fish production (Carline 1986). We were also
able to investigate how different strains of trout
stocked in the resei^voir recruited to the fisheiy
(Brauhn and Kincaid 1982, Babey and Berry
1989). This work was part of a comprehensixe
study on the ecology and causes of mortality of
stocked rainbow trout in mid-elevation reser-
voirs in Utah.
Study Area
East Canyon Resenoir is located at an ele-
vation of 1734 m in northern Utah (Morgan
County; 4()°54'N, 110°35'W). East Canyon
Creek and other minor tributaries of the reser-
voir drain a 99,200-ha watershed in the cal-
careous Wasatch Moim tains. At full pool the
resei^voir is 5.6 km long, 60 m deep, and cov-
ers 277 ha (Table 1). The resei-voir is produc-
tive, with a mean summer (May-Oct) chloro-
ph\ll a concentration of 5.4 mg/ni'^ (1985-86
and 1989-90 mean), and a mean Secchi depth
of 4.6 m (W. Wurtsbaugh unpublished data).
Blooms of cyanobacteria occur frecjuentK' dur-
ing sunmier and lall. Annual total phosphorus
(TP) loading of 2.8 g m~-yr~^ is very high
'Dcparliiient of Fisheries and \Vildlife/EcoloR>' Center. Utuli Stale l.'ni\ersit\, Loyan. V'V 84322-5210.
^Utiili Division ol Wildlife Resources, l,59(i West \ortli Temple, Salt Lake C.'it). IT S4I16.
12
1996]
Trout Yield in Utah Reskhvoir
13
Table 1. Limnological characteristics of East Canyon
Reser\oir, Utah. Data sources: ''Utah Department of
Health (1982); ''Merritt et ah 1980. Other data are unpuh-
lislii'd data of W. Wurtshaiigh.
Elevation-' 1734 m
Area (hill pool)-' 277 ha
Volume (hdl pool)^' 63,200 nv^
Mean and m;L\inunn depths-'' 23 & 60 m
Shoreline length-' 16 km
Ch!oroph>ll a (May-Oct) 5.4 /i,g/L
Seechi depth (May-Oct) 4.6 m
AlkalinitN-' 3.4 mE(iui\'
Total hardness" 233 mg/L
Total dissolved solids" 328 mg/L
Annual phosphorus loading'' 2.8 g ni~^yr~^
Mean water colimin total phosphonis-'' 80 /U.g/L
tMit color of fliiorcsccMit pigment (Phinn(>y et
al. 1967, Vondracek et al. 1980).
Trout grow quickly in East Canyon Reser-
voir and enter the fishery within 5 nion. The
reservoir is intensively fished due to its prox-
imity to 2 major population centers, Salt Lake
City and Ogden. Creel sui'veys in the 1970s
indicated fishing effort at over 300 angler-
hha~^yr~^ Because anglers fish primarily
with bait, there is little catch-and-release fish-
ing. Most trout captured are less than 350
mm. Schrader (1988), Babey and Beny (1989),
and Tabor and Wurtsbaugh (1991) provide
additional information on the fish and fishery.
(Merritt et al. 1980), and mean water column
TP is 80 )ag/L (Utah Department of Health
1982). Algal growth in the reservoir, however,
is limited primarily by nitrogen (Wurtsbaugh
1988). The reservoir's water level fluctuates
widely because of water withdrawals for irri-
gation, and consequently there is little macro-
phyte development in the littoral zone. During
much of the summer, oxygen concentrations in
the hypolimnion drop below 1 mg/L. Epilim-
netic temperatures reach 22°C in July, and the
reservoir is typically ice covered from late
December through March. During much of
the year high densities (>10/L) oi Daphnia
pulex, D. galeota, and other crustacean zoo-
plankton are evident (Tabor and Wurtsbaugh
1991, W. Wurtsbaugh unpublished data).
Additional limnological information is given in
Table 1.
Dominant fishes in the lake, in approximate
order of biomass, are Utah suckers {Catosto-
mus ardens), redside shiners {Richardsonius
balteatus), and rainbow trout {Oncorhynchus
mykiss). Less-abundant species are cutthroat
trout (O. clarki), brown trout {Salmo trutta),
speckled dace {Rhinichthys osculiis), fathead
mii.nows {Pimephales promelas), and kokanee
(O. .lerka). Rainbow tiout are heavily parasitized
by anchor worms {Lernaea cyprinacea- Berry
et al. 1991).
In late May the Utah Division of Wildlife
Resources stocks 300,000 (1080/ha) rainbow
trout, approximately 75 mm in length, in East
Canyon Reservoir. Fish captured by anglers
during our 1986 creel survey were derived
from several strains of rainbow trout stocked
in 1984-1986 (Table 2). Each strain stocked in
the reservoir was spray marked with a differ-
Methods
Creel data were collected during 1970, 1972,
and 1986 by interviewing anglers and by count-
ing the total number of anglers on the reser-
voir. Sampling effort was stratified by weekday
and weekend, month, time of day (morning,
midday, and evening), and method of fishing
(ice, shore, and boat), with random samples
taken within each stratum (Malvestuto 1983).
The creel clerk determined the number of fish
released and the number, length, and weight
(1986 only) of each species or strain kept. In
1986, 25% of the weights were not measured.
These were subsequently estimated with an
empirically derived length-weight regression
for rainbow trout:
W = 1.619 10-5 • TL2 949 ; fi2 = 0.95,
where W = wet weight in grams and TL =
total length in mm. Sample estimates were
expanded to provide monthly and seasonal
totals for fish harvests and angler use.
Details of the methods varied somewhat be-
tween surveys in the 1970s and those in 1986.
In 1986 we sampled 5 weekdays and 4 week-
end/holidays each month of the year In 1970
and 1972 the sampling inteival lasted only fi-om
opening day (Memorial Day weekend) through
August. Creel surveys in 1970 and 1972 were
done on both days of the opening weekend:
during the remainder of the sampling period
the reservoir was randomly censused on 20
(1970) or 48 d (1972). Because catch infomiation
was unavailable for the September-November
periods in 1970 and 1972, we restricted com-
parison with the 1986 catch statistics to the
Januaiy-August intei^val. Nevertheless, in 1986,
14
Great Basin Naturalist
[Volume 56
Table 2. Sizes (± standard deviation) and percentages of rainbow trout strains planted in East Canyon Resei-voir from
1984 to 1986, and percent of those fish captured by anglers during 1986. Each year 300,000 fish were stocked in the
resei-voir Relative return of each strain was calculated: [100 (% returned / % stocked)] -100. A large (L) and small (S)
group of Ten Sleep strain were planted in 1986. Shepherd = Shepherd of the Hills strain.
Strain-Size
Mean stocked
weight (g) ±s
%
stocked
Number
captiued
%
captured
Relative
return
1984
Kamloop
Ten Sleep
McConaughy
4.7 ±1.4
5.0 ±1.6
5.8 ±3.1
32
36
32
2,300
2,400
4,500
1985
25
26
49
-22
-28
+53
Kamloop
Ten Sleep
Shepherd
7.5 ± 2.3
5.4 ±1.4
3.7 ±1.4
33
33
33
33,000
27,700
16,100
1986
43
36
21
+30
+9
-37
Ten Sleep-L
Ten Sleep-S
Shepherd
4.8 ±1.4
3.1 ±1.1
4.2 ±1.4
25
25
50
3.300
2,800
6,700
26
22
52
+4
-12
+5
85% of the effort and 81% of the annual rain-
bow trout harvest occurred by the end of
August (see below), indicating that earlier sur-
veys provided a reasonable assessment of the
fisher)'.
During 1986 we identified fluorescent-
marked rainbow strains using a portable, bat-
teiy-powered black light affixed within a light-
exclusion box. Fish captured during the year
they were planted were designated age 0, and
those captured during the 2nd and 3rd year
after planting as age 1 and age 2, respectively.
We analyzed creel data with the FORTRAN
program WCREEL, supplied by the Utah Divi-
sion of Wildlife Resources (B. Schmidt personal
communication).
Temporal changes in trout abundance in a
put-grow-and-take fisheiy such as that in East
Canyon Reservoir can be evaluated by the
number of fish removed from the system by
anglers because relativeK' little mortality occins
from other factors after trout attain a harvest-
able size. For example, estimated losses of all
sizes of trout to birds, which has been shown
to bc> important in some Utah resenoirs (Waso-
wicz 1991) and elsewhere (Matkowski 1989),
accounts for approximately 6% of planted trout
in East Canyon Reservoir (R. A. Tabor unpub-
lished data). Piscivorous fish eat over 25% of
stocked trout, but this loss is negligible once
prey reach 150 mm (Wiu-tsbaugh 1987 and
unpublished data). Furthermore, because the
reser\()ir has a deep release, located in the
hyi^olimnion, we believe that few fish emigrate,
although we lack quantitative data to support
this.
Had we used this approach to estimate abun-
dance of trout planted in 1985 that reached
harvestable size, we would have required
creel data from at least 3 consecutive years
(1985-1987), or until anglers had removed all
of the cohort. Because we measured harvest
only during 1986, and thus lacked a long-term
data set, we assumed that harvests of age 0
fish in 1985 and age 2 fish in 1987 were simi-
lar to the measured harvest of age 0 (stocked
1986) and age 2 (stocked 1984) fish during
1986. Because 80-90% of each strain was har-
\'ested as age-1 fish (see below), violations of
this assumption should not have seriously
affected our analysis.
To determine the effect of trout densit\' in
the reservoir on monthly success rates for
anglers, we graphed the estimated density of
fish remaining to be captured from the 1985
cohort against catch per hour for fish in that
cohort. At the beginning of the survey in Janu-
ar>' 1986, we estimated that 67,400 fish from
the 1985 cohort were available in the reser-
voir This density was based on total catch of
the cohort in 1986 plus an additional 9000 fish
estimated to have survived into 1987. Nine
thousand (3%) of the 1984 cohort sunived
over 1 yr and were captured by anglers in 1986.
Fish densities for subsequent months were
calculated b>' subtracting the previous month's
1996]
Trout Yield in Utah Reservoir
15
liarvest. The resulting regression from tliis
analysis may include some bias, since mea-
surements of fish densities each month were
not independent of each other For this reason
we did not calculate statistical significance
levels for the regression. Nevertheless, the
approach yields a useful estimate of the rela-
tionship between abundance and catch rates.
To estimate the mortality of trout that were
captured and then released by anglers, we
assumed a survival rate of 69% (Taylor and
White 1992). Because sizes and ages of these
released fish were unknown, we assigned pro-
portions to the different year classes. Our inter-
views with anglers indicated the main reason
fish were released was because of small size,
but a limited number were also returned be-
cause of the presence of ectoparasites {Lernaea)
or their scars. We therefore assumed that 90%
of returned fish were age 0 (i.e., returned be-
cause of small size), 10% were age 1 (returned
for cosmetic reasons), and no age 2 were
returned.
Results
The Fisheiy Under a Six-month Season
Under the 6-mon open season documented
in 1970 and 1972, fishing was concentrated
from the opening weekend in late May
through August. Fishing on the opening week-
end accounted for 16-21% of the estimated
total effort, and 28-38% of the rainljow trout
harvest (Table 3). Fishing pressure dropped
steadily through the summer, and catch rates
varied from 0.18 to 0.49 trout/li. Total fishing
effort was similar in 1970 and 1972, with the
lake providing over 350 angling-h/ha. Anglers
harvested an estimated 60,100 rainbow trout
during the survey period in 1970, but only
35,600 in 1972 (Table 4). The catch rate for
rainbow trout in July 1972 was much lower
than in other months. This was due, in part, to
anglers fishing for kokanee and a strain of
albino rainbow trout that made up 44% of the
July hai-vest. The total catch rate of 0.32 fish/h
was comparable to other months of the year
when kokanee and albino trout were har-
vested less (4% of the catch in June and 9% in
August).
The Year-round Fishery
Trout grew rapidly in East Canyon Reser-
voir, particularly during their 1st year (Fig. 1).
Fish were planted in May at a mean size of 75
mm and 3.8 g. When they first entered the
fisheiy in July, they were 178 mm and 77 g. By
July, the previous year's cohort of fish had
reached 305 mm and 420 g. By the end of the
3rd year, fish had reached 400 mm and 728 g.
In 1986 anglers spent over 230,000 h
(±9300, .sy) fishing in East Canyon Resei-voir,
or 840 angler-hha"lyr"l. Most of these hours
were by shore anglers (58%), followed by boat
Table 3. Pressure, harvest, and catch rates for rainbow trout for creel surveys conducted in 1970, 1972, and 1986 for
the January-August period. Eadier surveys lasted only fiom the opening weekend (Memorial Day — the last weekend in
May) through August. In 1986 the state changed to a year-round season, so there was no opening day. Only the Janu-
ary-August data of 1986 are shown here to facilitate comparisons between the 2 periods. Total catch for the year is
shown in Table 4.
Jan-May
Opening
weekend
June
Julv
August
Total
Jan-Aug
1970
Effort (h)
Harvest (nuinber)
Catch rate (fish/h)
—
19,100
18,300
0.96
58,600
21,900
0.37
1972
28,600
13,900
0.49
13,600
6,000
0.44
119,900
60,100
0.50
Effort (h)
Harvest (number)
Catch rate (fish/li)
—
22,100
13,400
0.61
40,700
12,600
0.31
1986
32,500
5,800
0.18
11,600
3,800
0.33
106,900
35,600
0.33
Effort (h)
Harvest (number)
Catch rate (fish/h)
118,000
66,800
0.57
—
42,400
8,300
0.20
22,700
3,200
0.14
15,900
2,500
0.16
199,000
80,800
0.41
16
Great Basin Naturalist
[Volume 56
Table 4. Total catch of salinonids from East CaiiNon
Reservoir in 1970, 1972, and 1986. In 1970 and 1972
yields were estimated from the start of the fishing season,
in June, through August. Data for 1986 show captures
during the entire year. Table 3 shows the comparable
catch in 1986 from the opening day through August.
450
I 350
1970
1972
1986
z
Ul
_l
_l
<
250
Tax.\
Rainbow trout
60,100
35,600
98,960
O
150
Albino rainbow trout^
—
1.200
—
1-
Brown trout
200
20
60
Cutthroat trout
0
500
700
50
Kokanee-'
—
3,900
100
Tot.^l
60,300
42,220
99,820
"First stocked in 1970
(24%) and ice anglers (18%). The relative dis-
tribution of angling type varied seasonally; In
January and February, nearly all fishing was
done through the ice, but subsequent fishing
pressure was dominated by boat and particu-
larly shore anglers (Fig. 2A). Total fishing pres-
sure reached a peak during May, the period of
the traditional opening day.
Monthly catch rates for rainbow trout var-
ied from a high of 1.06 fish/angler-h in Febru-
ary to 0.18 in July (Fig. 2B). Annual catch rates
were 0.92 for ice anglers, 0.34 for boat anglers,
and 0.30 fish/h for shore anglers. The average
for all types of fishing was 0.42 fish/h. Catch
rates for ice anglers in Januaiy and Februaiy
were the highest for any month or method for
the year (Fig. 2B).
There was a strong relationsliip between
the estimated density of trout from the 1985
cohort remaining to be captured and monthly
catch rates for those fish (Fig. 3). In Januaiy
and Februaiy when there were more than 200
fish/ha (0.2 fish/m^) in the reservoir, catch
rates were over 0.6 fish/angler-h. As densities
dropped, however, catch rates declined pro-
gressively, reaching a low of 0.1 fish/angler-h
in December.
We estimate that 99,300 ± 7500 (sy) game
fish were removed from East Canyon Reser-
voir by anglers in 1986. Of these, 99.1% were
rainbow trout, 0.7% were cutthroat trout, 0.1%
were kokanee salmon, and <0.1% were brown
trout. Sixty-eight percent of the annual har-
vest of rainbow trout occurred from Januar\'
through May, and 38%; of these were captured
in January and Februaiy during the ice-fishing
season (Fig. 2C). Rainbow trout planted the
previous year (1985) dominated the catch from
Januaiy to August of 1986 (Fig. 2C). Rainbow
TOTAL LENGTH
J 1984
y
' 1985
/
w^^
"
- 1
-
J^986
■
700
600
§ 500
£ 400
O
I 300
200
100
0
- WET WEIGHT
^'
-
1
/i984 :
-
r^J
-
^
^1985 COHORT
-
•V
-
-m*^
.
^^1986
, _
■
YEAR 1 YEAR 2 YEAR 3
Fig. 1. Changes in total lengths (above) and wet weights
(below) of the 1984, 1985, and 1986 cohorts of rainbow-
trout captured by anglers in East Canyon Resei^voir All
fish were captured during 1986 but are plotted over a 2.5-
\T period to show long-term growtli rates. Also plotted are
initial lengths and weights of the fish stocked in 1986.
Total lengths (TL) can be converted to standard lengths
(SL) by dividing by 1.15.
trout planted in May 1986 first entered the
fisheiy at a mean total length of only 178 mm
in Jul\', and by October this cohort dominated
the harvest. Although age 0 and age 2 fish
were important in the fishen' eaiK' and late in
the year, 78% of the total catch was of age 1
fish from the 1985 planting.
Anglers released 37,000 hooked fish during
1986, giving an estimated mortalitv' of 10,400
fish during the 1st >'ear the>' were in the reser-
voir and an additional 1100 in the 2nd xear.
Consequently, approximately 4% of stocked fish
are lost because of hooking mortalit). About
75% of this mortcilit)' occuned fiom JuK' througli
December when small trout first entered the
fisheiy.
Total fish yield in East Can\on Reservoir
during 1986 was 102 kg/ha. Most of the hanest
occuned before July (Figs. 2C, 2D). Fish planted
the previous year represented 82% of the bio-
mass of rainbow trout captured in 1986.
1996]
Trout Yield in Uiaii Hkskhvoik
17
50000
^■K
X
t-
40000
z
o
30000
^
«
20000
oc
3
O
10000
X
^•^
0
1.6
1.4
oc
1.2
3
O
1.0
X
0.8
T
W
0.6
u.
0.4
StMlO
0.2
0.0
^„^
30000
X
H-
Z
o
20000
^
cc
UJ
03
10000
s
3
Z
*-'
0
8000
^
X
6000
1-
z
o
4000
S
o>
)^
2000
: B.
CATCH RATES
-
jn ICE
: ^
\ ^ BOAT
-■••-.
■-♦-.
--■■'
^<.j:r-'
SHORE M B
^ ^ .'
.■«k ■
. D. YIELD
■
k
A. ^TOTAL
■
Biib^
^ Ota ^
■iffe*^
lilllllilH^' ^
>^
|i984lHHIIHHI
.: <---i:^iiSi
JAN FEB MAR APR MAY JUNE JULY AUG SEPT OCT NCV DEC
Fig. 2. Seasonal changes in fishing effort and rainbow trout captured during 1986 in East Canyon Reservoir A, Fre-
quency polygon of seasonal changes in effort expended in ice fishery (ice), boat angling, and shore angling (top line
shows total fishing effort); B, monthly changes in catch rates for the 3 fishing methods; C, numbers of rainbow trout cap-
tured each month during 1986 from the 1984, 1985, and 1986 cohorts of rainbow trout planted in the resenoir; D, total
and component yield of trout from each cohort captured during 1986.
Hai-vest of Different Strains
Four strains were in the reservoir during
1986 as a result of stocking in 1984, 1985,
and 1986 (Table 2). Relative proportions of
each strain harvested fluctuated seasonally.
McConaughy strain from the 1984 stock and
Kamloop trout from 1985 were captured more
than expected in the winter and spring catches
of 1986. In the summer, however, catch rates
of Kamloop and Ten Sleep from the 1985
stocking were similar for the rest of the year.
Shepherd of the Hills strain stocked in 1985
was han'ested less than the other two strains
planted that year. During 1986 there were sig-
nificant differences in harvest rates of differ-
ent strains planted in 1984 (X^ = 13.34, p <
0.05) and in 1985 (X^ = 7.76, P < 0.05), but
not in 1986.
A large percentage of each strain stocked
in the reservoir was eventually captured by
anglers. There were, however, considerable dif-
ferences in relative return of different strains.
We estimate that 40% of Kamloop, 32% of Ten
Sleep, and only 23% of Shepherd of the Hills
strain were captured during their first 2 1/2 yr
in the reservoir (Fig. 4). For all strains com-
bined, 32% of the fish stocked were eventually
captured by anglers.
18
Great Basin Naturalist
[Volume 56
^ 1.U
'
T ' 1
■
O
F
X
^ 0.8
.
_
z
y
^
;;: 0.2
-
"osXi
-
0
J/* J
9
lU
"
Y - 0.0053 ♦ 0.0031 X
r - 0.79
3
O. n n
^_^
.
1 > 1
■
100 200
TROUT DENSITY (No. / Ha)
300
Fig. 3. Relationship between monthly estimates of the
density of rainbow trout remaining in the 1985 cohort and
catch per unit effort (CPUE) for those fish in the reservoir.
The CPUE shown here is less than in Figure 2B because
it does not include fish from 1984 and 1986 cohorts that
were captured, nor the captiue of other species. Letters
on graph indicate months.
Discussion
The fishing regulation change in East Can-
yon Resei-voir resulted in an excellent winter
ice fishery but poorer summer angling than
when a 6-mon season was in effect. In 1970
and 1972 anglers harvested 30-37% of the
annual total during the intensive 3-d opening
(Table 3), but large numbers of fish still
remained in the lake to support a summer fish-
ery with catch rates of 0.3-0.5 fish/h. In 1986,
however, about 66% of the fish had been har-
vested in the winter and spring fisheiy by the
time of the traditional opening day. Monthly
estimates of pressure during the summer fish-
ery (June-August) for 1986 were similar to
those in the earlier studies (Table 3), but the
sununer harvest was only 33-64% of that in
previous years.
While failing to maintain the tiaditional catch
rate for summer months, the regulation change
may have provided a fishery that not only pro-
duced increased numbers offish over a longer
period of time, but also provided almost twice
as many hours of recreational fishing as under
the 6-mon open season (Table 3). If the popu-
larity of winter angling were to increase sub-
stantially, an even larger proportion of trout
would be captured then, leaving fewer for the
traditional spring and summer fisheries. To
spread the catch over a longer period, the State
of Utah reduced the winter bag limit to 4 fish
subsequent to our study. The differences noted
under the different angling regulations must
be treated cautiously, however, as only 1 yr of
data was available for the year-round season,
and substantial between-year differences were
noted for the 1970 and 1972 period. Factors
such as changing predation pressure from pisci-
vores and changes in nutrient loading to the
reservoir undoubtedly also contributed to
changes in the fishery.
Catch rates for the 1985 cohort of fish were
clearly related to monthly changes in the den-
sity of these fish (Fig. 3), but there may have
been additional factors influencing fishing
success. Catch rates in February were higher
than the prediction based on density'. The rea-
son for this is not clear, but it is possible that
catch rates were especially high during mid-
winter when available food was low. Catch
rates in June-August were somewhat below
the regression, perhaps because during warm
months of the year fish are concentrated in
deeper water near the thermocline where they
are more difficult for anglers to reach. Catch
rates increased, relative to the regression, in
the fall (September-November) when the reser-
voir began to cool. Despite relative minor sea-
sonal shifts, it appears that densities of rain-
bow trout available in the reservoir can explain
most of the variation in catch rates.
Significant differences in the relative har-
vest of different strains of rainbow trout were
not unexpected, as others have found that
strains stocked can have large effects on the
fisheiy (e.g., Brauhn and Kincaid 1982, Babe\'
and Berry 1989). The poor return for Shep-
herd of the Hills strain (Table 2) is consistent
with the poor return of this group in East
Canyon Reservoir reported by Babey and
Beny (1989). Nevertheless, 2 factors confound
the interpretation of these results. First,
despite efforts to control sizes offish planted,
there were sometimes substantial differences
in weights of different strains stocked. For each
annual cohort, the relative return of a strain
was correlated with its size at stocking (Table
2); groups stocked at a large size usualh sunived
better than smaller ones. Second, because our
creel survey lasted only 1 yr, we could not
determine if some strains entered the fisheiy
as (luickK as others. For example, the veiy high
relatixe return rate of the McConaughy strain
in 1986 may be a consequence of a very low
catch rate of these fish measured in 1984 and
1996]
Tkout Yield in Utah Rkseuvoiii
19
30-
^ 20 H
-I
3
10 -
El KAMLOOP
A— TEN SLEEP
---«-- SHEPHERD
T T ?
MJJASONDJFMAMJJASOND
YEAR 1 YEAR 2
J F M A M J J
YEAR 3
TIME IN RESERVOIR
Fig. 4. Cumulative monthly increase in the percent of fish captured from 3 strains of trout stocked in 1985. The creel
sun'ey was conducted for only 12 mon, but data were expanded to cover a longer period by using information on other
cohorts (see te.\t).
1985 (Schrader 1988). Consequently, in our
study and in many others (see Babey and Berry
1989) that have investigated the importance of
fish strains, results are confounded because
strain size and condition were not carefully
controlled, and because the harvest of fishes
was not measured over their entire life span.
The fish yield of 102 kg/ha in East Canyon
Reservoir is among the highest yet reported
for a temperate zone lake (Morgan et al. 1980,
Jones and Hoyer 1982, Schlesinger and Regier
1982) and is as high as yields in many tropical
systems (Morgan et al. 1980). It is also high in
relation to clilorophyll levels in the lake. Regres-
sions with summer chlorophyll levels would
predict yields ranging from 4 to 13 kg/ha, de-
pending on the model chosen (Ogelsby 1977,
Jones and Hoyer 1982; see Carline 1986). A
model based on total phosphoiiis would predict
salmonid production of only 22 kg/h (Plante
and Downing 1993), so the realized yield of
102 kg/ha is far above expectations (Downing
and Plante 1993). Even when the weight of
fish stocked (5 kg/ha) is subtracted from total
yield, hai-vest from this cold-water reservoir is
still remarkably high.
Several characteristics of the reservoir and
fishery may contribute to the high yield. First,
high nutrient loading (Merritt et al. 1980) pro-
duces high algal productivity that in turn sup-
ports a large zooplankton population domi-
nated by Daphnia (this, however, does not
explain why fish production is higher than that
predicted by chlorophyll or phosphorus levels).
Second, rainbow trout in East Canyon Reser-
voir are primarily first-order carnivores, feed-
ing throughout most of their lives on large
Daphnia spp. (Tabor et al. in press). They begin
feeding on other fish only when they exceed
about 370 mm total length (Wurtsbaugh 1987).
Third, the management agency takes full ad-
vantage of high productivity by stocking large
numbers of fish. Fourth, with intense fishing
pressure, most of the trout are hai-vested thor-
oughly and quickly while they are growing
rapidly (Fig. 1). The combined effects of high
reservoir productivity, high stocking density,
trout feeding close to the base of the food web,
and intensive fishing pressure contribute to
the very high fish yield.
Although East Canyon Reservoir has pro-
vided exceptional trout yields, there are indi-
cations that high nutrient loading from resi-
dential and recreational development in the
headwaters of the drainage may be pushing
the fisheiy toward collapse. Because the reser-
voir is already mesoeutrophic, increased pro-
ductivity resulting from development may fur-
ther deplete oxygen in the hypolimnion and
metalimnion. Oxygen and temperature pro-
files we took in July and August 1985 and
1986 demonstrated that water with O2 con-
centrations >5 /xg/L was found only at depths
above 10 m where temperatures were above
20
Great Basin Naturalist
[Volume 56
18° C. Summer metalimnetic and hypolimnetic
oxygen concentrations in 1985 and 1986 were
much lower than reported for the reservoir
during 1978-1980 (Merritt et al. 1980, Utah
Department of Health 1982). When oxygen is
lost from these layers, fish are forced into the
warm epilimnetic water. Because optimal tem-
peratures for rainbow trout are near 15-18° C
(Hokanson et al. 1977, Wurtsbaugh and Davis
1977), and because O2 concentrations for
salmonids should be at or above 5 /Ltg/L (Brett
1979, EPA 1986), the situation in East Canyon
Reservoir may become too stressful for rain-
bow trout, and they may be squeezed into a
narrow metalimnion where conditions are sub-
optimal. Indications that trout are stressed
include poor growth in midsummer (Fig. 1;
Babey and Beny 1989), increases in Lernaea
infestation from 20/fish in the 1970s to 40/fish
in the late 1980s (T. Pettengill unpublished
data), and complete failure of the 1989 and
1991 year-classes subsequent to our field study.
Loss of salmonid fisheries with increasing
eutrophication is common (Colby et al. 1972).
Consequently, urban planners and fisheiy man-
agers should limit reservoir nutrient loading to
maintain adequate summer oxygen levels and
thus ensure that the outstanding family fisheiy
for salmonids in the resei'voir is maintained.
Acknowledgments
We thank D. Neverman and K. Marine for
assistance in the field, and G. Blommer for
helping in the field and in revising the
WCREEL FORTRAN program. Danen Brandt
assisted in data analysis and preparation of fig-
ures. C. Beny encouraged the stud\' and pro-
vided \'aluable criti(|ues of the manuscript. D.
Hepworth, R. Tabor, D. Archer, R. Whaley,
and two anonymous reviewers provided valu-
able comments on drafts of the manuscript. D.
Pitman and D. Andriano carried out the origi-
nal creel studies in 1970 and 1972. The study
was supported by the Utah Division of Wildlife
Resources with Federal Sport Fish Restoration
funds (F47-R) and was administered bv the
USFWS C:()operative iMsh and Wildlife Re-
search Unit at Utah State University.
Literature Cited
B.\l!l';v, {;. J., AND C. R. Bkhuv. 19.S9. Posl-stocking pltIoi-
niance of three strains of rainbow trout in a reser-
voir. North American Journal of Fisheries Manage-
ment 9: 309-315.
Berry, C. R. Jr., G. J. Babey, and T. Schrader. 199L
Effect of Lernaea eijprinacea (Crustacea: Copepoda)
on stoclced rainbow trout (Oncorhijnchus intjkiss).
Journal of Wildlife Diseases 27: 206-213.
Br.\lh\, J. L., AND H. L. Kincaid. 1982. Sunival, growth,
and catchability of rainbow trout of four strains.
North American Journal of Fisheries Management 2:
1-10.
Brett, J. R. 1979. Environmental factors and growth.
Pages 599-675 in W. S. Hoar, D. J. Randall, and J. R.
Brett, editors. Fish physiology. Volume III, Bioener-
getics and growth. Academic Press, NY.
Carline, R. F 1986. Indices as predictors offish commu-
nity' traits. Pages 46-56 ('» G. E. Hall and M. J. Van
Den Avyle, editors, Resenoir fisheries management:
strategies for the 80s. American Fisheries Society;
Bethesda, MD.
Carlton, F E. 1975. Optimum sustainable yield as a man-
agement concept in recreational fisheries. American
Fisheries Society Special Publication 9: 45—49.
Colby, R J., G. R. Spangler, D. A. Hurley, and A. M.
McCoMBlE. 1972. Effects of eutrophication on
salmonid communities in oligotrophic lakes. Journal
of the Fisheries Research Board of Canada 29:
97.5-983.
Dow NiNG, J. A., and C. Pl.\nte. 1993. Production of fish
populations in lakes. Canadian Journal of Fisheries
and Aquatic Sciences 50: 110-120.
EPA (U.S. Environmental Protection Agency). 1986.
Quality' criteria for water. EPA report 440/5-86-001.
Washington, DC.
Hall, G. E., and M. J. Van Den A\tle. 1986. Resei-voir
fisheries management: strategies for the 80s. Ameri-
can Fisheries Society, Bethesda, MD. 327 pp.
Hok.\nson, K. E. E, C. E Kleiner, and T. W. Thors-
LU.ND. 1977. Effects of constant temperatures and
diel temperature fluctuations on specific growth and
mortality rates and yield of juvenile rainbow trout,
Salmo gairdneri. Journal of the Fisheries Research
Board of Canada 34: 639-648.
Jones, J. R., and M. V. Hoyer. 1982. Sportfish harvest
predicted by summer chlorophyll-« concentration in
midwestern lakes and resenoirs. Transactions of the
American Fisheries Society 111: 176-179.
Mal\'ESTUTO, S. P 1983. Sampling the recreational fishen.
Pages 397-419 in L. A. Nielsen and D. L. Johnson,
editors, Fisheiy techniques. American Fisheries
Society, Bethesda, MD.
M VI'KOWSKI, S. M. D. 1989. Differential susceptibilitx of
three species of stocked trout to bird predation.
North American Journal of Fisheries Management 9:
184-187.
MiTuuiT, L. B., A. W Miller, R. N. Winget, S. R. Rush-
forth, AND W. H. Brimhall. 1980. East Canyon
Resen'oir water quality assessment. Mountainland
Association of Governments, Provo, UT. 193 pp.
Morgan, N. C, et al. 1980. Secondary production. Pages
247-340 in E. D. LeCren and R. H. Lowe-McConnell,
editors. The functioning of freshwater ecosystems.
Cambridge University Pres.s, London.
Ogelsry, R. T. 1977. Relationships of fish \ield to lake
pin toplankton standing crop, production and moipho-
edaphic factors. Journal of the Fisheries Research
Board of Canada .34: 2271-2279.
1996]
Trout Yield in Utah Rkservoir
21
I'liiNNEY, E. E., D. M. MiiJj'.K, AM) M. L. Daiilberg.
1967. Mass-marking young salnionids with fluores-
cent pigment. Transactions oi tlie Ameriian ImsIi-
eries Society 96: 157-162.
Plante, C, and J. A. Downing. 1993. Relationship of
salmonine production to lake trophic status and tem-
perature. Canadian Journal of Fisheries and Aquatic
Science 50: 1324-1328.
SCHLESINGER, D. A., AND H. A. Regier. 1982. climatic
and morphoedaphic indices of fish yields from nat-
ural lakes. Transactions ol tlic American Fisheries
Society 111: 141-150.
Schrader, T. M. 1988. Performance of three strains of
rainbow trout in East Canyon Reservoir. Unpub-
lished master s thesis, Utah State University, Logan.
Stockner, J. G. 1992. Lake fertihzation: the enrichment
cycle and lake sockeye salmon {Oncorhijnchus iwrka)
production. Pages 199-214 in H. D. Smith, L. Mar-
golis, and C. C. Wood, editors, Sockeye salmon [Oiico-
rJiynchus nerha) population biology and fiiture man-
agement. Canadian Special Pubhcation in Fisheries
and Aquatic Science 96: 198-215.
Tabor, R. A., and W. A. Wurtsbaugh. 1991. Predation risk
and the importance of cover for juvenile rainbow
trout in lentic systems. Transactions of the American
Fisheries Society 120: 728-738.
Tabor, R. A., C. Luecke, and W. A. Wurtsb.wgh. In
press. Effects of Daphnia availability on the growth
and food consumption rates of rainbow trout in two
Utah Resei^voirs. Transactions of the American Fish-
eries Society.
T.WLOR, M. J., and K. R. White. 1992. A meta-anahsis of
hooking mortality of nonanadromous trout. North
American Journal of Fisheries Management 12:
760-767.
Trend.all, J. 1988. Recruitment of juvenile mbuna (Pisces:
Cichlidae) to experimental rock shelters in Lake
Malawi, Africa. Em ironmental Biology of iMshes 22:
117-132.
Utah Department of Health. 1982. State of Utah clean
lakes inventory and classification, volume 1. Depart-
ment of Health, Salt Lake City, UT. 519 pp.
Vondracek, B., W. Wurisbaugh, and J. Cecu. 1980.
Mass marking of Gamlmsia. California Mostjuito and
Vector Control Association 48: 42—44.
Wasowicz, a. F 1991. Influence offish and avian preda-
tors upon the trout population of Minersville Reser-
voir. Unpublished masters thesis, Utah State Univer-
sity, Logan.
Wurtsbaugh, W. A. 1987. Importance of predation by
adult trout on mortality rates of fingerling rainbow
trout stocked in East Canyon Resei^voir, Utah. Pages
14-17 in Proceedings of the Bonneville Chapter of
the American Fisheries Society, Salt Lake City, UT,
1987.
. 1988. Iron, molybdenum and phosphonis limita-
tion of N, fi.xation maintains nitrogen deficiency of
plankton in the Great Salt Lake drainage (Utah,
USA). Verhandlungen Internationale Vereinigung
fur Theoretische unci Angewandte Limnologie 23:
121-130.
Wurtsbaugh, W A., and C. E. Davis. 1977. Effects of
temperature and ration level on the growth and food
con\'ersion efficiency of Sahno gairdneri, Richard-
son. Journal of Fish Biologv' 11: 87-98.
Wurtsbaugh, W. A., R. W Brocksen, and C. R. Gold-
man. 1975. Food and distribution of underyearling
brook and rainbow trout in Castle Lake, California.
Transactions of the American Fisheries Society 104:
88-95.
Received 26 May 1995
Accepted 18 September 1995
Great Basin Naturalist 56(1), © 1996, pp. 22-27
CONSUMPTION OF DIFFUSE KNAPWEED BY TWO SPECIES
OF POLYPHAGOUS CRASSHOPPERS (ORTHOPTERA: ACRIDIDAE]
IN SOUTHERN IDAHO
Dennis J. Fielding^-, M. A. Brusvenl, and L. P Kish^
Abstract. — Consumption of diffuse knapweed {Centaurea diffusa Lam.) by 2 polyphagous grasshopper species,
Melanoplus sanguinipes (E) and Oedaleonotus enigma (Scudder), was studied using microhistological analysis of
grasshopper crop contents. Grasshoppers were confined to cages containing C. diffusa and Sisymbrium altissimum L., a
member of the mustard family known to be readily eaten by these 2 grasshopper species. Preference indices for knap-
weed were lower than for S. altissimum in 4 of 5 trials. An uncaged population of A/, sanguinipes on a knapweed-infested
site consumed only small amounts of knapweed until late summer when most other plants were senescent. Results sug-
gest that diffuse knapweed's low palatability to generalist herbivores may confer to it a competitive advantage over other
rangeland plants.
Key words: Centaurea diffusa Lam., diffuse knapweed, herbivory, insects, competition.
Diffuse and spotted knapweed, Centaurea
diffusa Lam. and C. maculosa Lam., respec-
tively, were introduced to the Pacific North-
west around 1900 (Watson and Renney 1974).
Since then they have rapidly spread through-
out the area (Fig. 1; Forcella and Hai-vey 1981).
Heavy infestations of knapweed reduce produc-
tion of more desirable species of forage plants,
thus reducing the value of rangeland for graz-
ing and wildlife habitat. Several specialist
insect herbivores have been introduced in
attempts to control knapweed (Story and Ander-
son 1978, Maddox 1979). To date, no studies
have reported on the consumption of knapweed
by polyphagous insect herbivores.
Cnicin, a sesquiteipene lactone, is produced
by spotted and diffuse knapweed (Drodz 1966,
Locken and Kelsey 1987). Pieman (1986) sug-
gested that sesquiterpene lactones have toxic
effects on many herbivores and may function
as deterrents to herbivoiy Locken and Kelsey
(1987) suggested that nonpalatability of knap-
weeds may afford them a competitive advantage
over many other plant species by protecting
them from herbivoiy. (grasshoppers (Orthoptera:
Acrididae) are a conspicuous and important
class of herbivores on rangeland in the \\'est-
em U.S.
Rangeland grasshopper populations in south-
ern Idaho occasionally reach outbreak propor-
tions. Two species in particular, Melanoplus
sanguinipes (F). and Oedaleonotus enigma
(Scudder), are capable of attaining very high
densities (>30/m-). Both species feed upon a
broad range of forbs (Brusven and Lamley
1971, Banfill and Brusven 1973, Sheldon and
Rogers 1978). Pfadt (1992) suggested that an
increase in introduced weeds is a factor lead-
ing to outbreaks of O. enigma. Fielding and
Brusven (1993) found that both species prefer
disturbed rangeland habitats dominated by
e.xotic annual plants. This study assessed the
utilization of diffuse knapweed as food by these
2 grasshopper species to determine if knap-
weed represents a significant and expanding
resource for grasshoppers and if grasshopper
herbivor)' may be a constraint to knapweed
populations.
Previous studies (Brusven and Lamle\ 1971)
have shown Sisymbrium altissimmn L., an intro-
duced annual forb, to be preferred by many
forb-feeding grasshoppers. Both species of
weeds initiate growth as a basal rosette of
leaves and later develop erect, sparsely leaved
stems that bear flowers. Because C. diffusa is
usually a biennial, it does not develop beyond
the basal rosette until the 2nd year Sisymbrium
alfissi}num constituted a large proportion of
the forbs present in this stud)'; therefore utili-
zation of C. diffusa and S. altissinuim was
compared.
' l)c|>,irliMciit of Plant, Sciil, and luiloniolDjiual Scicntos, Univcrsit) oC Ulalm, Moscow. W 83844-2339.
-Present addtcss; PO Box 75010. Uni\eisit\<)l Alaska, Fairbanks, AK 9MT7.">0102.
22
1996]
Knapweed Consumption by Crassiioim'krs
23
Fig. 1. Idaho counties reporting infestations of dirfuse knapweed, Centuurea diffusa.
Materials and Methods
The study site is about 3 km south of Sho-
shone, Idaho (Lincoln County), in a knapweed-
infested area that had been seeded with crested
wheatgrass {Agropyron ciistatum [L.] Gaertn.)
in 1975. Grasshopper food preferences were
identified by microhistological analysis of grass-
hopper crop contents (Brusven and Mulkern
1960, Sparks and Malechek 1968, Fielding and
Bnisven 1992). Grasshoppers were confined to
cages so that relative amounts of different
plant species could be precisely determined.
Five trials were conducted during the summer
of 1989: O. enigma 4th- and 5th-instar nymphs
in early June; O. enigma adults and M. san-
guinipes 4th and 5th instars in late June; M.
sanguinipes adults in July and again in August.
For each trial, 4 wire-mesh (5-mm pore size),
conical cages covering 0.5 m^ each were placed
in the field such that at least 1 plant each of C.
dijfiisa and Sisymbrium altissimiim L., along
with assorted common grasses, occurred within
each cage. Twelve to 15 grasshoppers of a sin-
gle species were placed in each cage. Grass-
hoppers used in the tests were collected from
rangeland and placed in the cages within 20 h
of collection. A 4-d interval was estimated to
be sufficient to completely void previous meals
and to accurately assess preferences in choice
tests. After 4 d, 10 grasshoppers were removed
from each cage and immediately preserved in
95% ethanol for crop analysis.
Species composition of plants in each cage
was determined on an air-dry basis by clipping
and sorting by species aboveground portions
of plants in each cage after each trial. Clipped
plants were stored in air-tight plastic bags, and
fresh weight was obtained within 4 h of clip-
ping. Clipped plants were then allowed to air-
dry until they quit losing weight (10-15 d),
after which dry weights were obtained (to the
nearest 0.1 g). Percent moisture of above-
ground portions of each plant species was then
determined.
Plants were rated after each trial according
to phenology as follows: 1, vegetative growth
only; 2, flowering; 3, seed set; 4, seed maturity;
5, senescent or dormant (USDA-Soil Conser-
vation Service 1976). Grass species present
included Poa sandbergii Vasey, Bromus tecto-
riim L., and Agropyron cristatum. Centaurea
dijfusa and S. altissimum composed about 97%
of aboveground biomass of forbs. Both 1st-
and 2nd-year C. diffusa were present in each
of the cages. Other forbs present were Helian-
thus annuus L., Lactuca serriola L., and Epilo-
bium L. sp.
Grasshopper crops were removed and the
contents mounted on glass slides in glycerin
and safranin stain. Plant fragments in the crops
were identified by comparing them with refer-
ence slides made from fragments of known
plants collected at the study site, similar to the
methods described by Fielding and Brusven
(1992). Frequency counts were made for each
plant species by determining their presence or
absence in 20 microscope viewing fields per
grasshopper crop. Trichomes, hairs, and pollen
were not counted. Frequencies from the 10
grasshoppers per cage were summed. Relative
frequency was calculated by dividing the fre-
quency of a plant species by the total fre-
quency of all plant species (Sparks and
Malechek 1968, Pfadt and Lavigne 1982).
Holecheck and Gross (1982) demonstrated the
24
Great Basin Naturalist
[Volume 56
near equivalence of relative frequency to actual
diy weight percentage of plants consumed.
Relative availability of different plant species
within an area has been shown to influence
diet composition in many grasshopper species
(Ueckert et al. 1972, Mitchell 1975). To account
for the effect of availability on consumption,
preference values for plant species constitut-
ing more than 10% of either cage or crop con-
tents were calculated by dividing relative fre-
quency of a plant species in the crops by that
species' percentage of the dry-weight of all
plants within the cages (Ueckert and Hansen
1971). A preference value >1 indicates feed-
ing in greater proportion to the plant's avail-
ability, whereas a preference value <1 indi-
cates low preference in relation to a plant's
availability.
Possibly, total diy weight of a plant may not
accurately portray the amount of plant mater-
ial available to grasshoppers, thus introducing
bias into the preference values. In this study
our obsei"vations indicated that both species of
weeds had similar ratios of leaves to stems.
Also, we have obsei^ved grasshoppers feeding
on stems of both weed species. Because we
had no way to determine more precisely
exactly what proportion of the plant was avail-
able as food to the grasshoppers, we used total
aboveground biomass as a reasonably objec-
tive measure of availability. The presence of
Ist-year rosettes of C dijfusa in the cages
ensured that each replication included a rep-
resentative choice of plant material.
Differences bet\\'een plant species in rela-
tive frequency and preference values were
tested using the Wilcoxon 2-sample test (PROC
NPARIWAY, SAS 1985), with each cage rep-
resenting 1 replication. Comparisons between
plant species were made for each trial of a sin-
gle grasshopper species and with data from
different trials pooled by grasshopper species.
The same statistical methods were also used to
test for differences in relative frequency and
preference values between grasshopper
species for C. diffusa and S. (dtissimurn.
Food selection was monitored in an uncaged
population of M. sanguiwipes near the cage
study. Thirty to 50 individuals were collected
on each of 5 dates from June through October
from an area of ca 1 ha infested with knap-
weed. Food preference in this population was
determined by microhistological methods de-
scribed above.
Plant species composition at the site was
determined by visual estimates, in 5% incre-
ments, of the ground cover of each plant
species in forty 0.1-m- quadrats, arranged in 4
transects of 10 quadrats each. Ground cover
estimates were made in July and again in
October after precipitation caused abundant
germination of cheatgrass. Because accurate
estimates of food axailability (biomass) in the
field were not available, preference values were
not calculated and the results are presented
for comparative puiposes only.
Results
Cages were placed such that C. diffusa was
equally as abundant as or more abundant than
S. altissimum in each trial (Table 1). Percentage
moisture of both species of weeds declined
throughout the season (Table 1). Sisijmhnum
altissimum tended to be slightly more advanced
phenologically than C. diffusa throughout the
season, partly due to the presence of Ist-year
rosettes of C. diffusa in the cages, but also
because of earlier flowering bv S. altissimum
(Tlible 1).
Although C. diffusa constituted a substan-
tial percentage (10-46%) of the caged grass-
hoppers diet, preference values for C. diffusa
were <1 in eveiy trial, indicating that it was
not consumed in proportion to its diy weight
composition within the cages (Table 1). Prefer-
ences values for S. altissimum were > 1 in each
trial, indicating that it was consumed in pro-
portions greater than its relative availability.
After flowering in Jul\', a large portion of
the C. diffusa plant material in the crops of i\/.
sanguinipes consisted of floral parts (44% and
30% of the C diffusa material consumed, in
the July and August trials, respectively). Other
forbs represented in in situ caged trials were
not present in sufficient quantity to ade-
(|uately assess their preference values.
More S. altissimum than C. diffusa was con-
sumed b>' grasshoppers in 3 of the 5 trials
(Table 1). Preference values for S. altissimum
were greater than those for C. diffusa in 4 of
the trials (Table 1). Combining data from the 3
trials with M. sanguinipes, crop contents and
preference values for S. altissimum, 42% and
2.0, respectively, were greater than for C. dif-
fusa, 16% and 0.5, respecti\'el\' (Wilcoxon test,
P < 0.01 for both tests). For O. enigma, the
overall preference value for S. altissimum, 3.5,
1996]
Knapweed Consumption by Grasshoppers
25
Table I. Relatixe availaliilit\' and consumption h\' grasshoppers of plant species.
Plant
PerccTit
Mean
Relative
Relative
Mean
pheuologic;
il moisture
clr\ weigh
t ;
ixailability
frequency
preference
Plant species
stage'
ol plants
in cages
in cages^
in crops
index
4th- and 5th-instar Oedaleonotus enigma nymphs on 6 June
1989
Ccntaurca dijfusu
1
77
13.4
23
lOa'5
0.38a
Sisyinhriuin altissiiiiuiii
1
81
8.8
15
48b
5.06b
Other ibrbs
1
85
0.6
1
<1
—
A^roptjroit cristatiiin
1-2
57
18.1
31
<1
<0.05
Poo sandbi'i-fiii
4
24
4.7
8
2
—
Broiuus tcctoriiin
4
21
12.8
22
32
1.42
Detritus
6
adult Oedaleonotufi
1 enigma on 26 June
1989
Ccntaurca diffusa
1
64
59.0
61
46a
0.76a
Sisymbrium altissimum
1-2
67
27.1
28
48a
1.9.3a
Other forbs
1-2
79
1.0
1
1
—
Agropyron cristatum
3-4
45
6.8
7
0
—
Poa sandbergii
5
15
0
0
0
—
Bromus tcctoruin
5
12
3.6
3
1
—
Detritus
4
—
4th- and 5th
-instar Melanoplus
sanguinipes
nymph
s on 26 June 1989
Ccntaurca diffusa
1
64
60.8
59
16a
0.25a
Sisymbrium altissimum
1-2
67
25.8
25
74b
3.00b
Other forbs
1-2
79
2.1
2
0
—
Agropyron cristatum
3-4
45
6.2
6
0
—
Poa sandbergii
5
15
4.1
4
5
—
Bromus tectorum
5
12
3.1
3
4
—
Detritus
2
—
adult Melanoplus sanguinipes on
21 July 1989
Ccntaurca diffusa
1-2
63
25.4
29
16a
0.56a
Sisymbrium altissimum
2-3
55
24.5
28
44b
l.,55b
Other forbs
1-2
75
8.8
1
3
—
Agropyron cristatum
4
45
7.0
8
4
—
Poa sandbergii
5
9
17.5
20
3
0.17
Bromus tectorum
5
14
11.4
13
27
2.52
Detritus
3
—
ad
ult Melanoplus sanguinipes on 25 August 1989
Centaurea diffusa
1,3-4
22
38.8
38
23a
0.70a
Sisymbrium altissimum
4-5
11
18.4
18
24a
1.48b
Other forbs
2-3
65
8.2
8
5
—
Agropyron cristatum
4
18
15.3
15
3
0.37
Poa sandbergii
5
8
4.1
4
5
—
Bromus tectorum
5
7
13.3
13
25
2.83
Detritus
16
—
'l. vegetative giowtli uiiK; 2, fluwfriiig, .3, seed set; 4, seed ni;iturit\'; 5, seneseent ur donnant
-Mean (JV = 4) percentage ot aboveground plant biomass (air-dn' basis) within cages
•'Means for C. diffusa and S. altissimum within columns of each trial followed by different letters are significantly different, P < 0.0.5, Wilco.xon 2-saniple test.
was greater than for C. diffusa, 0.6 (Wilcoxon
test, P < 0.05). There was no difference in con-
sumption by O. enigma between S. altissimum
and C. diffusa, 48% and 27%, respectively
(Wilcoxon test, P > 0.05). There were no dif-
ferences between the 2 species of grasshop-
pers in relative frequency or preference values
for either S. altissimum or C. diffusa (Wilcoxon
test, F > 0.10 for both comparisons).
Of the grass species, only Bromus tectorum
was eaten in greater proportion than its per-
centage of air-dry biomass. Even though O.
enigina is generally considered to be a forb-
feeder (Sheldon and Rogers 1978, Pfadt 1992),
B. tectorum constituted 32% of the diet of O.
e7iigma in early June (Table 1). Adult O. enigma
in late June ate very little B. tectorum. Melanop-
lus sanguinipes consumed B. tectorum through-
out the summer, with 4-27% of its diet com-
posed of B. tectorum, even though the grass
was completely senescent by 26 June (Table 1).
26
Great Basin Naturalist
[Volume 56
Table 2. Relative frequency of food items in crops of Af. sanguinipes on 5 dates and percentage ground cover in JuK
and October 1989.
Sisyinhriuin (iltissiinit)n
Ccntaiirca diffusa
Other forbs'*
Bromiis tectorutn
Other grasses
Litter, detritus
Relati\e fre(juenc\ of crop components
30
20
14
6
13
June
Jul\
Aug
Sep
Oct
46
23
22
/
6
18
30
32
55
1
19
25
7
24
6
7
9
12
6
76
1
4
15
1
10
9
9
13
7
1
Percentage
ground cover
July
2
6
<1
October
1
4
<1
16
6
••Iiicludes rahl)itlirusli iChnisotluimiis rwti'irosiis [Rill] Biitt.l, lupine iLiipiiais L spl, am! suiifloufr Hh'lianthus (iiuuiiis L.l.
Knapweed was the most common forb grow-
ing on the site where the uncaged population
of M. sanguinipes was studied (Table 2). In June,
S. oltissimiDn was the largest single food item,
but consumption declined as the season pro-
gressed. Knapweed was a substantial food item,
especially in August and September when it
remained succulent after other forbs had dried.
After rainfall stimulated germination of B. tec-
toriim in late September and October (Table
2), it became the primaiy food item for M. san-
guinipes, and forbs constituted onb' a minor
portion of the diet.
Discussion
The evolutionar)' histoiy of an herbivorous
species, by shaping its food habits and other
life history traits, determines its present rela-
tionships with exotic plant species. The 2 grass-
hopper species in this study consume a wide
variety of plants, especialK' forbs (Banfill and
Brusven 1973, Sheldon and Rogers 1978, Pfadt
1992), and will readily accept exotic plant
species. Melunophis sangninipcs is a veiy oppor-
tiniistic feeder Egg hatch in this species is often
spread out over a long period, resulting in a
large proportion of a population maturing dur-
ing the diy periods typical of late summer in
the intermoimtain region. At such times man\
late-maturing plants that still retain some suc-
culence, such as rabbitbrush, sagebrush, and
some lupine species, are primary food items
for M. sanguinipes. The results of this stud>
indicate that this was the case with C. diffusa:
even though it was not highly preferred 1)\ M.
sanguinipes, it was a major food item in late
summer when most other plants were dry.
SisyniJjriuni alfissiinuin tended to become sene-
scent earlier than C. dijfusa, which would re-
duce the qualit)' of S. altissimum relative to C.
dijfusa, especially when Ist-year rosettes, con-
sisting mostly of leaves, are considered.
Locken and Kelsey (1987) reported that
cnicin concentrations in C. maculosa xaiy con-
siderably within and among indi\ idual knap-
weed plants. Cnicin is stored within glandular
trichomes on the surface of knapweed tissues.
Highest concentrations of cnicin were found
in leaves surrounding the inflorescence. Only
trace quantities were found by Locken and
Kelsc)' (1987) in the stem epidermis and flow-
ers. Leaf concentrations were lowest in spring
and increased with flo\\ ering. We assume that
cnicin concentrations in C. dijfusa follow much
the same pattern. Variability in cnicin concen-
tration may result in selective consumption by
grasshoppers of knajDweed tissues w ith low
cnicin concentrations. Our results suggest that
this is the case: In late-summer trials much of
the knapweed tissue consimied by grasshoppers
consisted of flo\\'ers. This implies that during
years of high grasshopper densities, feeding
by grasshoppers, especialK' on the flowers,
could result in a modest reduction in seed
production in this plant.
Residts of this stud)' pro\'ide support for the
h> pothesis that knapweed is protected from
herbivoiy by its chemical constituents (Pieman
1986, Locken and Kelsey 1987). When com-
pared to S'. altissimum, diffuse knapweed was a
2nd-choice food item for these generalist
grasshopper species. Its low palatability may
confer a competitive acKantage to knapweed
when herl)i\on' is a strong selection factor.
.\lthough it is conceivable fliat at high densities
grasshoppers may consume significant amounts
of knapweed and reduce seed production,
man> other plants would be affected to a greater
degree, thus reducing competition to knapweed.
1996]
Knapweed Consumption by Giusshoppers
27
Grasshopper species used in this trial are
the dominant species contributing to outbreaks
in soutliern Idaho. It appears that increasing
knapweed infestations do not represent a sig-
nificant increase in food resources for these
grasshoppers. However, because knapweed
stays green longer during the summer than
many other rangeknid phmts, it may provide
sustenance for polyphagous grasshoppers dur-
ing kvte-summer droughts in southern Idaho.
Acknowledgments
The authors thank the staff of the Bureau of
Land Management s Shoshone District office
for technical and logistic support. Russell
Biggam assisted with field studies. Bahman
Shafii advised on statistical matters. Robert H.
Callihan, Dave Koehler, and Don Hostetter
provided helpful comments on earlier versions
of the manuscript. This study was supported
in part by the Bureau of Land Management as
Cooperative Agreement ID 910-CA7-05. It is
published with the approval of the director of
the Idaho Agricultural Experiment Station as
paper 94724.
Literature Cited
Banfill, J. C, AND M. A. Brusven. 1973. Food habits and
ecology of grasshoppers in the Seven Devils Moun-
tains and Sahnon River breaks of Idaho. Melanderia
12; 1-21,
Brusven, M. A., and J. D. Kwiley. 1971. The food habits
and ecology of grasshoppers from southern Idaho
rangeland. University of Idaho in cooperation with
USDA, Project Completion Report 1967-1971.
Brusven, M. A., and G. B. Mulkern. 1960. The use of
epidermal characteristics for the identification of
plans recovered in fragmentary condition from the
crops of grasshoppers. North Dakota Agricultural
Experiment Station Research Report .3. 11 pp.
Drodz, B. 1966. Isolation of cnicin from the herbs of Ccn-
faiirea dijfusa Lam. Dissertationes Pharmaceuticae
et Pharmacologicae 18: 281-283.
Fielding, D. J., and M. A. Brusven. 1992. Food and habi-
tat preferences of Melanophi^ sanguinipes and Aiilo-
cara elliotti (Orthoptera: Acrididae) on disturbed
rangeland in southern Idaho. Journal of Economic
Entomologx' 85: 78.3-788.
. 1993. Grasshopper (Orthoptera: Acrididae) com-
mimity composition and ecological disturbance on
southern Idaho rangeland. Environmental Entomol-
ogy 22: 71-81.
FoRCELLA, E, and S. J. Hahvey. 1981. New and exotic
weeds of Montana. Volume II: Migration and distri-
bution ol 100 alien weeds in northwestern U.S.A.,
1880-1980. Montana Department of Agriculture,
Helena.
Holecheck, J. L., and B. D. Gross. 1982. Evaluation of
diet calculation procedures for microhistological
analysis. Joimial of Range Management 3.5: 721-723.
LoCKEN, L. J., and R. G. Kelsey. 1987. Cnicin concentra-
tions in Centaurea maculosa, spotted knapweed.
Biochemical Systematics and Ecology 15: 313-.320.
Maddo.x, D. M. 1979. The knapweeds: their economics
and biological control in the western states, U.S.A.
Rangelands 1: 139-141.
Mitchell, J. E. 1975. Variation in food preferences of
three grasshopper species (Acrididae; Orthoptera) as
a function of food availability. American Midland
Naturalist 94; 267-283.
Pfadt, R. E. 1992. Valley grasshopper Oedaleonotm enigma
species fact sheet. USDA-Animal and Plant Health
Inspection Service, Wyoming Agricultural Experi-
ment Station Bulletin 912.
Pfadt, R. E., and R. J. Lavigne. 1982. Food habits of
grasshoppers inhabiting the Pawnee site. University
of Wyoming Agricultural Experiment Station Science
Monograph 42. 72 pp.
PiCMAN, A. K. 1986. Biological activities of sesquiterpene
lactones. Biochemical Systematics and Ecology 14:
2.55-281.
SAS Institute. 1985. SAS/STAT user's guide. SAS Insti-
tute, Carey, NC.
Sheldon, J. K., and L. E. Rogers. 1978. Grasshopper food
habits within a shrub-steppe community. Oecologia
32: 8.5-92.
Sparks, D. R., and J. C. Malechek. 1968. Estimating per-
centage dry weights in diets using a microscopic tech-
nique. Journal of Range Management 21: 264—265.
Story, J. M., and N. L. Anderson. 1978. Release and
establishment of Urophora ajfinis (Diptera; Trypeti-
dae) on spotted knapweed in western Montana. Envi-
ronmental Entomology 7: 44.5-448.
Ueckert, D. N., and R. M. Hansen. 1971. Dietary over-
lap of grasshoppers on sandhill rangeland in north-
easteiTi Colorado. Oecologia 8; 276-295.
Ueckert, D. N., R. M. Hansen, and C. Terwilliger, Jr.
1972. Influence of plant fi-equency and certain mor-
phological variations on diets of rangeland grasshop-
pers. Journal of Range Management 25; 61-6.5.
USDA-SoiL CONXERSATION SERVICE. 1976. National range
handbook. USDA-SCS, Washington, DC.
Watson, A. K., and J. Renney 1974. The biolog>' of Cana-
dian weeds. 6. Centaurea dijfusa and C. maculosa.
Canadian Journal of Plant Science .54: 687-701.
Received 11 July 1994
Accepted 2 February 1995
Great Basin Naturalist 56(1), © 1996, pp. 28-37
FIRE FREQUENCY AND THE VEGETATIVE MOSAIC OF
A SPRUCE-FIR FOREST IN NORTHERN UTAH
Linda Wadleighl and Michael J. Jenkins^
Abstract. — Fire scar and vegetative analysis were used to constnict a fire histoiy for the Engelmann spruce/sub-
alpine fir {Picea engelmannii/Abies lasiocarpa) vegetation type of the Utah State University (USU) T. W. Daniel E.xperi-
mental Forest. Three distinct periods of fire frequency were established — presettlenient (1700-1855), settlement
(1856-1909), and suppression (1910-1990). Mean fire intei-val (MFI) decreased during the setdement period and greatly
increased during the suppression era. The difference was attributed to the influ.x of ignition sources during the settle-
ment of nearby Cache Valley, located 40 km to the west. Logging and livestock grazing appear to have led to the
reduced MFI, which in turn worked as a factor to create the vegetative mosaic now obsei"ved on the study area. The
increase in MFI during the suppression era permitted the advancement of shade-tolerant species in the understoiy of
the shade-intolerant lodgepole pine (Pimis contoiia \ar latifoUa) and quaking aspen {Popiihis treiiuilokles). Continued
suppression of disturbance fi-om wildfire will allow the lodgepole pine cover type, which experienced the lowest MFI
during the settlement period, to be further invaded by shade-tolerant species, decreasing spatial stand diversity and
increasing the risk of more intense fires.
Key words: fire jrequeney. subalpine spruce-fir joresl. fire sear.
Absence of natural fire in wilcUand ecosys-
tems, due to removal of fine fuels by livestock,
reduction in Native American ignitions, and a
suppression policy instituted in the early 1900s
has led to extensive alterations in natural vege-
tative succession patterns. Human disruption
of natural fire regimes in fire-dependent com-
munities limited natural diversity and altered
the long-term stability of fire-adapted plant
species (Heinselman 1973, Gruell 1986, Agee
1993). Previously, natural ecosystems had
evolved under episodic fires (Parsons 1981,
Gruell 1983). Gruell's (1983) interpretation of
paired photos from the Northern Rockies
showed early stages of forest succession were
more common from 1870 to 1940 than they
are today; however, Gruell (1983) also found
the absence of fire has contributed to a marked
alteration of natural vegetation mosaics by
favoring woody species such as shrubs and
trees over grasses.
Lightning-ignited fires in Engelmann spiiice/
subalpine fir {Picca cn'^chntnuui/ Abies lasio-
carpa) forests are less frequent than fires in
drier vegetation types. Arno (1980) estimated
a fire return interval of 50 to 130 yr for spruce/
fir habitat types. Veblen et al. (1994) found a
mean fire-rctnrn intenal of ca 200 \'r in a Kock>
Mountain subalpine forest in northwestein
Colorado. In these subalpine fir forests, historic
fire allowed the dominance of serai species
and created a mosaic of species and age com-
positions. Where serai species such as lodge-
pole pine {Piniis contorta) or aspen {Populus
tremidoides) occurred, a higher fire frequency
favored their dominance (Bradley et al. 1992).
In the lodgepole pine-dominated communi-
ties that occur in the lower portion of the sub-
alpine fir forest, fire was more frequent with
intensit)' depending on amount of precipitation
received in the summer months. Abundant evi-
dence was found in the lodgepole pine forests
of northern Utah of nondestmctive groimd fires,
more intense "thinning fires, "stand-replacing
fires, and severe double bums" (Arno 1980).
Fire histoiy studies provide land managers
with estimates of past fire fi-equencies, mean
fire-return intenals, and effects of natural fire
on stand composition and structure (Arno and
Sneck 1977). Such studies help to determine
the return inten al of fires on a site, intensit)'
and size of fire, effects of past fire on stand
dviiamics, and effects of an era of modem sup-
pression. Managers may also use the natund fire
cycle or regime of an area to determine if the
present disturbance regime is within the histor-
ical range ol \ariatiou. A variet\ ol techni({ues
are used to exaluate fire historx; including
'USDA Fmcsl Service, Onden, IJT
^DepartTiieiit of l-oresl Resotirees, Utah State Univcisit\', Ix)Kaii, UT 84322-5215. Addiess correspoiicleiiee to this autlior.
28
1996]
Spkuce-Fih Fire Frequency
29
mapping stand types, correlating tire dates
from fire-scarred trees to establisli a fire
chronolog)', and determining age-class distri-
butions, using increment cores to establish the
extent of fires (Arno 1980, T^mde 1979).
The objective of this study was to deter-
mine il the existing vegetative mosaic of the
T. W. Daniel Experimental Forest is correlated
with the fire histon' of the study area, primarily,
whether fire frequency has changed between
3 distinct periods; presettlement, settlement,
and suppression. Additionally, if fire frequency
has changed, is that change reflected in the
vegetation structure visible today.
Study Area
The USU T W. Daniel Experimental Forest,
located about 40 km east of Logan, Utah, is
1036 ha in area and ranges in elevation from
2377 m to 2651 m (Fig. 1). Topography ranges
from higher plateaus dissected by deep
drainages to gentle slopes and small meadows.
No permanent lakes or streams are wdthin the
study area (Schimpf et al. 1980); however,
intermittent streams do cany runoff from the
site. Winters are cold and wet, and summers
are warm and dry. Mean annual precipitation
is 104 cm per yr, mostly falling as snow (Hart
and Lomas 1979).
The major vegetation component is the
Engelmann spruce/subalpine fir type in late
successional stages, with serai lodgepole pine
{Piniis contorta van latifolia) and ciuaking aspen
stands, and small meadows distributed througli-
out. A young conifer understory consisting pri-
marih' of subalpine fir is often present in the
aspen stands (Schimpf et al. 1980).
Methods
Fourteen sampling transects were estab-
lished along contours spaced 61 m apart based
on slope distance. A continuous log of forest
cover type, the predominant vegetative type,
was kept along each transect to create a stand
map. As the contour intervals were traversed,
trees with fire scars were identified and re-
corded. The number of fire scars was recorded
for each "catface" — an open scar resulting from
lire damage. Fire scars are fonned when flames
near the trunk raise the temperature of the
cambium to a lethal level, or actually consume
bark, phloem, and xylem (McBride 1983). Trees
with the largest number of sound scars were
marked for further studv.
Great Salt Lake
Fig. 1. Map showing approximate location of the T. W. Daniel Experimental Forest between Cache and Rich counties
northeast of Logan, Utah.
30
Great Basin Naturalist
[Volume 56
Sixty-two trees with the greatest number of
visible, individual fire scars were sampled by
taking a partial cross section from the pith to 1
side of the catface (Arno and Sneck 1977). The
wedges were sanded and annual growth rings
counted, recording the number of years back
to each fire and the number between fires.
Trees may be scan'ed in a number of ways in-
cluding mechanical damage by nearby falling
trees, root rot infection, lightning, or strip
attacks by mountain pine beetle {Dendroctoniis
ponderosae Hopkins, Coleoptera: Scolytidae)
(Johnson and Gutsell 1994); however, there
were no blue stains, lai^val galleries, or beetle
emergence holes in the scars sampled (Stuart
et al. 1983), which would suggest they had
resulted from causes other than fire. Because
pockets of obscured rings or rot may also cause
inaccurate counts, tree records were combined
into a master fire chronology (Arno and Sneck
1977).
Individual tree ring counts were arranged
horizontally on paper, geographically ordered
so that neighboring trees were adjacent. Ten-
year increments were placed on the left verti-
cal axis, beginning with the sample year at the
top, and the oldest ring year recorded at the
bottom. The number of trees scarred in a year
was compared to the number of trees suscep-
tible to scarring. If a tree was consistently out
of order, a number of years was added or sub-
tracted to bring it into alignment (Arno and
Sneck 1977). The maximum number of years
added or subtracted equaled 3; and 16 trees
were adjusted.
Variable-radius plots were laid out along the
sampling transects at a spacing of 200 m. Tree
species present were recorded to determine
cover type, and a site tree — a dominant or
codominant tree on the plot — ^was aged for each
species. Increment cores were taken at breast
height for each site tree and were adjusted for
total age for each species. A 74()th-ha regener-
ation plot was recorded, tallying seedlings and
saplings by species and diameter, at the center
of the variable plot to aid in detennining suc-
cessional patterns.
Cover type, dated scars, and stand age data
collected from these plots were incorporated
into a stand map to show the extent of stands
that might have resulted from a fire distiu-
bance (McBride 1983). The stand map was
supplemented by remotely sensed satellite
imagery obtained in 1986.
Fire frequency, "the number of fires per
unit of time ' (Romme 1980), on an area was
calculated for 3 fire frequency periods to por-
tray the effects of settlement, logging, grazing,
and modem fire suppression on the fire regime.
Mean fire intei^vals, "an arithmetic average of
all fire intervals determined in a designated
area' (Romme 1980), were calculated for each
period. Determining mean fire intei^vals for
distinct land-use periods is useful in under-
standing human impact on forest ecology and
fire histoiy (McBride 1983). The periods were
"suppression" (1910-1990), when U.S. Forest
Sei'vice fire suppression was initiated, "settle-
ment era" (1856-1910), and "presettlement"
(prior to 1856). Mormon pioneers established
the first settlement by Europeans in the Cache
Valley in 1856 (Bird 1964). The presettlement
period began the year just prior to the age of
the oldest tree sampled — 1700 (Romme and
Despain 1989). A fire history is limited by
longevity of trees on the site and durabilit\' of
wood exposed when scarred (Heinselmann
1973).
Total number of years in each period was
then divided by the number of fires in that
period to obtain mean fire interval. Docu-
mented evidence of historical fires was used to
verify dates in the settlement-era and fire-sup-
pression periods (Bird 1964).
A master fire chronolog)' was developed for
each stand experiencing fire in the study area
as indicated by scars and the presence of
even-aged stands of lodgepole pine (Romme
and Despain 1989) or aspen stands (Brown
and Simmerman 1986, Debyle et al. 1987).
Stands were considered even-aged if deviation
in the increment core age of site trees was
< 20% (Daniel et al. 1979).
Results
Three forest cover types consisting of 15
stand types were identified. Species repre-
sented in pure stands were lodgepole pine,
Engelmann spruce, subalpine fir, and quaking
aspen, but the area in pure stands was rela-
tively small compared to that of mixed stands:
280 ha in pure stands \ersus 580 ha in mixed
stands out of a total 1036 ha.
Ol the 15 delineated stand t>'pes, subalpine
fir, the climax species in the habitat type pre-
sent (Schimpf et al 1980), was a major sec-
ondary stand component in 9 types and the
1996]
Spkuce-Fik Fikk Fiu:yuENCY
31
Tahle 1. Percent of regeneration by species witliin stand type. Suhalpine fii- is flie primary comi^onent in regenera-
tion in all stands except aspen.
Stand
type
Snbalpine
fir
Engelniann
sprnce
Aspen
Lodgepol
pine
Donglas-
fir
DF/PF^'
DF/ES/AF
LP
LP/AF/AS
DF/AF
DF/ES
LP/AF
LP/AF/ES
ES
ES/AF
AF
AF/AS
AS/ES/AF
AS
100
0
0
0
0
100
0
0
0
0
75
19
6
0
0
67
11
22
0
0
67
33
0
0
0
67
33
0
0
0
65
35
0
0
0
61
31
0
8
0
60
25
15
0
0
57
32
11
0
0
56
33
11
0
0
52
0
48
0
0
46
27
27
0
0
41
3
53
0
3
"Stand hpe abl)reviatit>ns: AF = siihalpinc lir. AS = aspen, ES = Engelmanii spnice, LP = lodgepole pine, DF = Douglas-lir, PF = linilier pine.
principal component in 2. Regeneration sur-
veys conducted at each plot showed subalpine
fir to be the primaiy regeneration component
in 13 of the 15 types (Table 1). Aspen regener-
ation was the primary component in the aspen
stand t\'pe. Overstoiy ages ranged fi"om 63 to
284 yr in lodgepole pine, 106 yr in aspen, 188
yr in subalpine fir, and 193 yr in Engelmann
spruce.
Sixty-two fire-scar wedges were collected
fiom fire-scarred trees, 22 fiom Engelmann
spiiice, 1 fi-om subalpine fir, and 39 fi^om lodge-
pole pines. All scar and pith dates were used
in the master fire chronology, but only 6 of the
spruce scars were used to indicate fire years,
while 37 lodgepole pine scars were utilized.
The remaining scars were not used due to rings
obscured by decay.
Sixteen fire years were represented in scar
and/or regeneration data. Where scars were
not present, but vegetation was even-aged,
e.g., stands L20, F24, L18, and L17 (Tables 2,
3), a fire year was determined from the age of
dominant lodgepole pine or aspen trees pre-
sent. Two of the 16 fire yeai's, 1700 and 1860,
were represented solely by age-classes on the
site. Two fire years during the settlement per-
iod, 1890 and 1895, were documented by Bird
(1964). Bird's account stated that numerous
small fires were reported in Logan Canyon in
1890, while the 1895 fire year was substanti-
ated by a large fire reported in Stump Hollow
in Logan Canyon, an area north of the study
area (Bird 1964).
Those stands where the major component
was lodgepole pine exhibited 13 fire years, 4
in the presettlement fire period from 1700 to
1855, 9 in the settlement period from 1856 to
1909, and no fires in the suppression period
from 1910 to the present. Ten of the 13 fires
were represented by fire scars in the present
stands (Table 2).
There were 7 fire years in stands in the
spruce/fir cover type, which predominantly
comprised spruce/fir and secondary compo-
nents of lodgepole pine, aspen, Douglas-fir, and
limber pine. There were no fires in the pre-
settlement period, 6 in the settlement period,
and 1 in the suppression period. Five of the 7
fires were recorded by scars and validated by
age of the present stand (Table 3). There were
4 fire years in the aspen cover type. Three of
those fires were validated by both fire scars
and age-class analysis. One fire occurred in
the presettlement period and 4 in the settle-
ment period (Table 3). Only 1 fire year, 1903,
was common to all 3 forest types. Four fire
years (1860, 1890, 1902, and 1903) were
shared between the spruce/fir and lodgepole
pine cover types (Figs. 2, 3).
Mean fire inter\'als estimated for the entire
study area, for each cover type, and for each
fire frequency period are shown in Table 4.
Mean fire interval for the entire study area
was 18 yr, i.e., a fire occurred about every 18 yr
somewhere within the study area. Mean fire
interval was shortest in lodgepole pine and
longest in aspen. During the presettlement
32 Great Basin Naturalist [Volume 56
Table 2. Fire frequency in the lodgepole cover type by stand and fire year Stands consist of a predominant lodgepole
component or mixed species with the priman,' overstoiy component of lodgepole pine. (Adapted fiom Arno and Sneck
1977.)
Stands
year
L2^'
L3
L4
L.5
L6
L7
L22
LI I LIO
L12
LL5
L20
LLS
L17
L9
L13
Suppression period
1942 — —
Settlement period
1909 ______ — — _ — — — — — — —
1903 3rl^ 2r Ir 4r 2 1 1 _,■_ — — — — — — —
1902 _______ Ir 1 Ir 1 Ir 1 5r — —
1899 __________ _ _____
1895 -r____l ___
1890 ________1
1887 ___-,_____
1883 _r _ _ Ir — — — — —
1877 -r _ _ -r — 1 _ _ _
I860 _________
1858 ________2
Presettlement period
1847 -r _r _ 2r — — — — —
1834 ________ ir
1822 ________ Ir
1700 ___,■ _____
aStand description: Lodgepole = L3, L5, L6, L22, L12, L17, L9, L1.3; LP/AF/AS = L2, L7, L1.5; LP/AF = L4. Lll, L20; LP/AF/ES = LIO. LLS, LP = iudgt-pole
pine, AF = subalpine fir, AS = aspen, ES = Engelmann spruce.
"Digit (L2, etc ...) = number of trees in stand with fire-scar date; r = regeneration in stand, determined from increment cores.
Table 3. Rre freciiiencies in the Engelmann spriice/subalpine fir and aspen cover t\'pes. (Adapted from Arno and
Sneck 1977.)
Aspen
Engelmann spnice/suhtilpine fir stands stands
Fire
year E2'' E4 E5 F21 F24 F7 F23 A3 A8
Supression period
1942 ll' ________
Settlement period
1909 ______1 __
1903 Ir Ir 2r — — — — 1 Ir
1902 1 ___rl ____
1899 1 ________
1890 ______r ___
1883 _______1 _
1877 _______! _
1860 _____r ____
1858 _________
Presettlement period
1834 _________
1822 _________
1700 _________
"Stand description: ES/AF = E2. E.5; AF/AS = F,3, F23; DF/ES = D2; AF/LP = V5. F7; ES/AF/AS = E4, F24: AF/ES = n). Alv'LP/ES = F2L AS = A3. .\S.
AS = aspen, AF = subalpine fire, DF = Douglas-fir, ES = Engelmann spruce, and LP = lodgepole pine.
•^Digit (1, 2) = number of trees in stand with fire-scar date: r = regeneration in sl:md determined from increment cores that correspond to fire data.
1996]
Spruce-Fih Fire pREyuENCY
33
^
1 : 2 4.000
0.5 1 1.5
KI L OMET E RS
Legend
1847 \ZA 1860 ^ 1877
Regeneration occurring after 1847
♦ I Regeneration occurring after 1860
\W]\ Regeneration occurring after 1 877
\M.\ Fire scars that recorded the particular fire date
Fig. 2. GlS-pi-oduced diagram of fires in the study area from 1700 through 1877 based on stand mapping, regenera-
tion, and fire-scar data.
34
Great Basin Naturalist
[Volume 56
1 : 24.000
KILOMETERS
Legend
E^ 1883 ZZ 1890 E] 1902
V77A 1887 en] 1895 E] 1903
3E Regeneration that occurred after the fire date
r^\ I Fire scars that recorded the particular fire date
Fig. 3. CIS-produced diauiaiii of fiic-s in the stucK area iioiii 1(SS3 through 1942 l)ased on .stand mapping, regenera-
tion, and fire-sear data.
1996]
Spruce-Fir FikI'; Frequency
35
Table 4. Mean fire intenal In conlt t\pe and firi' irciinenty [leiiod. Mean lire interval i.s an aritlnnelic average in
years of tlie nnniher of years in a period dixided by the nnmher ol iires oecurring in that ])eriod. A d()nl)le hyphen
denotes that no exidenee of fire occurring in tliat period was found. Ranges of intervals are in parentheses.
Presettlement
Settlement
Suppression
Total
(1700-1855)
(1856-1909)
(1910-1988)
(289 years)
Stud\ area
39
(1-122)
4.9
(1-30)
79
18.1
Cover types
ES/AF
—
9
(1-30)
79
41.3
LP
39
(12-122)
6
(1-17)
—
22.2
AS
156
13.5
(4-16)
~
57.8
and settlement periods, mean fire intei"val was
shortest in lodgepole pine. Mean fire internals
were longest in the suppression period (e.g.,
spruce/fir) or no fires occurred (e.g., lodgepole
pine and aspen; Table 4).
Discussion
Stand Age and Regeneration
The widespread occurrence of subalpine fir
in the cover types, both in the overstoiy com-
ponent and in the regenerating understory, is
associated with later stages of succession
(Schimpf et al. 1980). Stands sustaining the
most recent extensive fires, 1902 and 1903,
have less of a subalpine fir component than
those not withstanding recent fires (Figs. 2, 3).
However, subalpine fir is apparent as a com-
ponent of regeneration following these fires
and now as a tolerant understoiy.
Fire frequencies declined during the last
century, a trend that would favor the establish-
ment of stands of Engelmann spruce and sub-
alpine fir that are less resistant to fire. When a
subalpine fir climax is reached, overtopping
intolerant serai species, it is not easily replaced
due to its tolerant reproduction, unless a dis-
turbance interferes, such as fire, insects, dis-
ease, or logging (Eyre 1980). Aspen stands also
have a component of subalpine fir present and
will require a disturbance if they are not to be
replaced by the tolerant subalpine fir climax
(Mauk and Henderson 1984).
Fire Frequencies
Compared to the mean fire interval in the
presettlement period, there was a large increase
in fire frequency in the settlement period in
all 3 cover types (Tables 2, 3). Both Bird (1964)
and Roberts (1968) stated that ignition sources
increased while settlement was occurring in
Cache Valley
Size and number of fires in the mountains
surrounding Cache Valley coincided with the
heaviest use period (Bird 1964). The 1880 cen-
sus stated 1%-10% of the timbered area of
Cache County buiTied, or 5000 to 50,000 acres.
Heavy grazing of the period undoubtedly
reduced fine fuel loads, but use by loggers and
sheepherders increased ignition hazards.
Fires were largely untended until 1906, when
the U.S. Forest Service arrived. An employee
of the U.S. Forest Sei^vice in 1906 stated that
3/4 of the Bear River Forest Reserve (later to
become part of the Wasatch-Cache National
Forest) had been burned over in tlie last 20 yr,
probably due to careless sheepherders (Bird
1964). Fires were recorded in Blacksmith Fork
Canyon in 1878, as well as a "large fire in
Stump Hollow in Logan Canyon in 1881 (Bird
1964).
Compared to the settlement period, fire fi"e-
quency decreased during the suppression per-
iod and there was no evidence of fire in the
lodgepole pine and aspen types. Forest Ser-
vice suppression techniques decreased the size
and occurrence of fires, which also coincided
with a large reduction in allowable grazing,
lessening an ignition hazard (Bird 1964).
The lack of evidence of fire since 1910 can-
not be attributed to deterioration of fire-scar
evidence. A fire severe enough to scar stand-
ing trees should be recorded in the present
stands. The actual fire frequency may be
higher than recorded; fires may not have been
severe enough to scar trees (Lorimer 1984) or
were suppressed before they became extensive.
36
Great Basin Naturalist
[Volume 56
Mean fire intervals in all cover types de-
creased in the settlement period and increased
or there were no fires during the suppression
era (Table 4).
There were few if any fires found in this
study in the presettlement period. The fire scars
in aspen may have been lost to natural mortal-
ity and decay, and fires may not have been
severe enough to produce fire-scarred trees.
Evidence of additional fires in the lodgepole
pine and spruce/fir cover types may also have
been destroyed, and actual mean fire intei"vals
for this period may be substantially shorter.
Fire hazard in a lodgepole pine stand is
highest shortly following a fire due to standing
snags and remaining ground fuels from the
previous fire, and later when crowns of the
tolerant underston' reach into crowns of mature
lodgepole pine creating ladder fuels (Brown
1975, Romme 1982). In tlie study area, less fire-
resistant Engelmann spruce and subalpine fir
have begun to reach into the crowns of the
lodgepole pine and aspen stands, increasing
fire hazard. Both spruce and fir are highly sus-
ceptible to fire, due to their low-branching
habits and thin bark (Schimpf et al. 1980). Evi-
dently, fuel was also available to allow several
nonlethal fires to bum in lodgepole pine stands,
as occurred between 1877 and 1903 in the
study area. One stand apparently burned 4
times during this 26-yr period, and several
areas burned more than once (Table 2).
Conclusions
The lack of disturbance by fire on the USU
T. W. Daniel Experimental Forest in the last 80
yr has allowed succession to proceed towards
a climax of subalpine fir. The increase in fire
frequency following settlement was probably
due to efforts to exploit natvnal resources and
the concomitant increase in ignition sources.
Freciuent disturbance by fires dining the
settlement period resulted in die present mature
vegetative mosaic. These earlier frequent fires
favored lodgepole pine, and the less-fre(iuent
fires of the suppression period favored more
tolerant species, as demonstrated 1)\ the abun-
dance of subalpine fir regeneration in all coxcr
types. The continued lack of disturbance will
allow the more tolerant species of subalpine fir
and Engelmann spruce to overtop the intoler-
ant lodgepole pine and aspen. Eventually the
area will lose its diverse appc>arance and will
be similar to that in the areas where fire dis-
turbance is less frequent.
Acknowledgment
This research was supported by the Utah
Agricultural Experiment Station, Utah State
University, Logan, Utah, as Journal Paper 4689.
Literature Cited
At;EE, J. K. 1993. Fire ecology of Pacific Northwest
forests. Island Press, \V;ishington, DC. 493 pp.
Arno, S. E 1980. Forest fire histoiy in the Northern Rock-
ies. Journal of Forestry 78(8): 460-465.
Arno. S. F, and T. D. Petersen. 1983. Variation in esti-
mates of fire intei^vals; a closer look at fire histoiy on
tlie Bitterroot National Forest. USDA Forest Service
Research Paper INT-301.
Arno, S. F, and K. M. Sneck. 1977. A method for deter-
mining fire history in coniferous forests of the
Mountain West. USDA Forest Service, General
Technical Report INT-42.
Barreit, J. W. 1980. Regional sil\ iculture of the United
States. 2nd edition. John Wile\ ic Sons, Inc., New
York.
Bird, D. W. 1964. A histoiy of timber resource use in the
development of Cache Valley, LI tali. Unpublished mas-
ter's thesis, Utah State University, Logan.
BiuDLEY, A. F, N. V NosTE, AND W. C. FisCHER. 1992. Fire
ecology of forests and woodlands in Utah. USDA
Forest Sei^vice, General Technical Report INT-287.
Brown, J. K. 1974. Handbook for inventorying downed
woody material. USDA Forest Senice, General Tech-
nicalReport INT-16.
Brown, J. K., and D. G. Simmerman. 1986. Appraising fuels
and flammabilit)' in western aspen: a prescribed fire
guide. LISDA Forest Sendee, General Technic;il Report
INT-20.5.
Daniel, T W, J. A. Hel.ms, and F S. Baker. 1979. Princi-
ples of silviculture. McGraw-Hill, Inc., New York.
Debvle, N. V, C. D. Bevins, and W. C. Fischer. 1987.
Wildfire occurrence in aspen in the interior western
United States. Western Journal of Applied Forestn
2(3): 73-76.
Eyre, F H. 1980. Forest cover types of the United States
and Canada. Societ\' of American Foresters, Bethesda,
MD.
Gruell, G. E. 1983. Fire and vegetative trends in the
Northern Rockies: interpretations from 1871-1982
photographs. USDA Forest Senice, General Techni-
cal Report INT-158.
. 1986. The importance of lire in the Greater Yellow-
stone Ecosvstem. Western Wildlands, I'all 1986;
14-18.
Mart, G. E., and D. A. Lomas. 1979. Effects of clearcut-
ting on soil water depletion in an Engelmann spruce
stand. Water Resources Research 1.5: 1598-1602.
llllNSELMAN, M. L. 1973. Fire in the virgin forests of the
Boundan Waters Canoe Area, Minnesota. Quateman
Research ,3: 329-382.
Johnson, E. A, and Gutsell, S. L. 1994. Fire frequent)
models, methods and interpretations. Advances in
I-Aologieal Research 25: 2.39-287.
1996]
Sphuce-Fih Fire Frequency
37
LoiUMER, C. G. 1984. Mrthodolo^ical considerations in
the analysis of forest clisturl)anee history. Canadian
Journal of Forestiy Research 15: 200-213.
Mauk, R. L., .\nd J. A. Henderson. 1984. Coniferous for-
est habitat types of northern Utah. USl^A Forest
Sei^vice, General Technical Report INT-170.
McBride, J. R. 1983. Analysis of tree rings and fire scars
to establish fire history. Tree-Ring Bulletin 43:
.51-67.
Pakson.s, D. J. 1981. Role of fire management in maintain-
ing natural ecosystems. Pages 469-488 /';; Proceed-
ings of the conference — fire regimes and ecosystem
properties. US DA Forest Senice, General Technical
Report \VO-26.
Roberts, R. B. 1968. Histon' of the Cache National For-
est. 3 volumes. USDA Forest Sei-vice, Washington,
DC.
ROMME, W. H. 1980. Committee chairman of fire histoiy
terminolog\ : report of the ad hoc committee. Pages
1135-1137 in Proceedings of fire histoiy workshop,
20-24 October 1980, Tucson, AZ. USDA Forest Ser-
vice, General Technical Report RM-81.
. 1982. Fire and landscape diversity in subalpine
forests of Yellowstone National Park. Ecological
Monographs 52: 199-221.
RoMMK, W. II., and D. G. Despain. 1989. Historical per-
spective on the Yellowstone fires of 1988. BioScience
39: 695-699.
SciiiMPK, D. J., J. A. Henderson, and J. A. MacMahon.
1980. Some aspects of succession in the spruce-fir
forest zone of northern Utah. Great Basin Naturalist
40: 1-26.
Stuart, f. D., D. R. Geiszler, R. I. Gara, and J. K. Acee.
1983. Mountain pine beetle scarring of lodgepole
pine in south central Oregon. Forest Ecology Man-
agement 5: 207-214.
Tande, C. F 1979. F"ire histoiy and vegetative patterns of
coniferous forests in Jasper National Park, Alberta.
Canadian Journal of Botany 57: 1912-1931.
Veblen, T. T, K. S. Hadley, E. M. Nel, T. Kitzberger,
M. Reid, and R. VlLl^LBA. 1994. Disturbance regime
and disturbance interactions in a Rocky Mountain
subalpine forest. Journal of Ecolog)' 82: 125-135.
Received 18 November 1994
Accepted 2 October 1995
Great Basin Naturalist 56(1), © 1996, pp. 38-47
ARIZONA DISTRIBUTION OF THREE SONORAN DESERT
ANURANS: BUFO RETIFORMIS, GASTROPHRYNE OLIVACEA,
AND PTERNOHYLA FODIENS
Brian K. Sullivan 1, Robert W. Bowker^, Keith B. Malmos^, and Erik W. A. Gergus'^
Abstract. — We surveyed historic collecting localities in south central Arizona dining JuK, August, and September
1993-94 to determine the presence of 3 little-known Sonoran Desert anurans, Biifo retifoniiis. Gastrophnjne olivacea,
and Pternoliyla fodiois. All 3 species were present at most historic localities visited under appropriate conditions (fol-
lowing rainfall in JuK' and August). Pternohijla fodiens was restricted to San Simon Wash and associated tributaries in
south central Pima County. Gastrophnjne olivacea ranged from Vekol Valley in extreme southern Maricopa County
south to the Mexican border, and southeast near Tucson and Nogales in Pima and Santa Cruz counties. Bitfo retifonnis
occuiTcd over the widest area, from southern Rainbow Valley in Maricopa Coimty southwest to the vicinit>' of Organ
Pipe Cactus National Monument, and southeast to the vicinity of Tucson and Sasabe in Pima County.
Key tcords: Bufo retifonnis, Gastrophiyne olivacea, Pternohyla fodiens, historic distribution, present distribution,
ainpliibian decline, Arizona, Sonoran Desert.
Three relatively little-known anurans, Bufo
retiforDiis, Gastrophnjne olivacea, and Pterno-
hyla fodiens, occur in the Sonoran Desert in
south central Arizona. Although placed in sep-
arate families (Bufonidae, Microhylidae, and
Hylidae, respectively), they are superficially
similar in behavioral ecology. Each is inactive
for more than 10 mon each year, emerging
only to reproduce and forage following intense
rainfall during the summer "monsoon' season.
All exhibit "explosive" breeding behavior (Wells
1977) in which males form high-density aggre-
gations for a few nights (sometimes only one)
following a major rainstorm and call to attract
females. Within Arizona all 3 species are largely
restricted to a small portion of the Sonoran
Desert in the extreme south central part of the
state, so it is perhaps not suiprising that they
are relatively unknowii. Indeed, Bufo retifonnis
was described in 1951 based on specimens
collected southeast of Ajo in 1948 (Sanders
and Smith 1951), and Pternohyla fodiens was
first documented in Aiizona in 1957 (Chrapliw>'
and Williams 1957, Williams and ChraplivvT
1958).
Given limited information on these Arizona
aniuans, this investigation was undertaken in
1993 and 1994 to ascertain th(>ir present dis-
tribution in Maricopa, Pima, Pinal, and Santa
Cruz counties, Arizona. First, we describe
methods used in conducting the suney. Then,
for each target species sun^eyed, we describe
distinguishing acoustic characteristics and out-
line historic and present distributions. Last,
we present observations on breeding behavior.
Materials and Methods
Suney Methods
All surveys were conducted along paved
roads throughout the known ranges of the 3
target species following rainstorms during
July, August, and September 1993-94. Given
the highly unpredictable and variable nature
of summer rainfall and the need for monitor-
ing the entire south central portion of Arizona,
we could only crudely estimate (e.g., weather
reports) the appropriateness of field condi-
tions (i.e., le\el of rainfall) for anuran activit\
prior to each field excursion. Whenever suffi-
cient rainfall appeared to have fallen in the
study area, we traveled to that particular area
on the night of the rainfall exent, or the fol-
lowing night, to surve\' for amphibians along
roadways. Frequently, 2-3 nights of surveying
occurred for each rainfall exent. Occasionally,
siu\'e\- plans were adjusted to take advantage
of local conditions (e.g., localized flooding).
I|)i-pai(iiit'iil of Life Sciences, Arizona State Universit\- West, PC) Box 37101). Phoeni.v, .\/. 85069.
-Department of Biolosy, Clendale Comnuniit>' College, Glendale. AZ 85302.
^Department of Zoology; Arizona State University', Tenipe, AZ 85287.
38
1996]
SONOIUN Dksert Anuiuns
39
To conduct surveys we drove slowly (40-65
kniph) along paved roadways scanning for
anurans on the road surface and listening lor
chorus acti\'it\' adjacent to the roadway. Most
roads in the study area are located in valley
floodplains crossed by numerous washes so that
collection of large lain pools immediately ad-
jacent to roadways occurs commonly. If insuf-
ficient rainfall had occurred so that anuran
surface activity was initiated but no chorusing
activit\' was apparent (i.e., no calling or breed-
ing), we continued driving, scanning for and
recording all anurans foimd on the road. When
activity was relatively high (e.g., >20 anurans/
km) and/or associated with an area of interest
(e.g., historic or suspected locality for one of
the target species), we recorded eveiy individ-
ual anuran seen on the roadway (for a minimum
of 1 km) until lack of moisture resulted in
reduced anuran activity (e.g., <5 anurans/km).
Whenever we detected choiaising activity or
pools of water along the roadway, we stopped
and scanned the area adjacent to the roadway.
If none of the target species were detected
either visually or acoustically, we resumed the
road survey. If target species were present, we
attempted to record a series of voucher calls
(see below) and collect a small series of voucher
specimens {N < 10). Unfortunately, summer
rainfall in south central Arizona was below
average during the sui-vey period, resulting in
few actual breeding aggregations. All speci-
mens are deposited in the ASU Vertebrate
Collection.
Field Observations
Each target species possesses distinctive
vocalizations. Advertisement calls were recorded
in the field with a Marantz PMD 430 stereo
recorder and Sennheiser ME 80 microphone
with K3-U power module, or a Sony WM-D6C
cassette recorder and Sony ECM-909 stereo
microphone. Males generally ceased calling
when they were approached {Gastrophryne
and Pternohyla were easily disturbed); only if
the observer remained relatively motionless
would apparently normal calling behavior be
resumed. Release calls were recorded either
in the field or in the laboratory by gently com-
pressing the sides of a male held between
thumb and forefinger directly above a micro-
phone (following Sullivan 1992). Only slight
pressure was necessary to elicit a series of re-
lease calls. Cloacal temperatures were measured
with a Weber quick-recording thermometer
within 5 sec of recording the final advertise-
ment call or release call. Water and air tem-
peratures were generally within 3°C] of cloacal
temperatures during field recordings.
Acoustic Analysis
Advertisement calls were digitized with a
DATA Precision model 610 plug-in digitizer at
a sampling rate of 10 kHz (Nyquist frequency
= 5 kHz) and analyzed with a DATA Precision
6000 waveform analyzer. Release calls were
digitized at a capture rate of 22 kHz on a Macin-
tosh LC computer using a Farallon Corpora-
tion MacRecorder and analyzed with Sound-
Edit software (version 2.03). Call durations
were measured to the nearest 0.01 sec with
the Waveform analyzer (<2 sec) or with a stop-
watch. Pulse rates of advertisement calls were
measured over a 0.5-sec interval spanning the
call midpoint; all pulses were counted to deter-
mine the pulse rate of release calls using the
oscilloscope mode of SoundEdit. Dominant fre-
quencies were estimated to the nearest 10 Hz
over a 0.25-sec intei^val spanning call mid-
points using the waveform analyzer. Neither
advertisement nor release calls are frequency
modulated to any large extent in any of the 3
anurans under study. For each male used in
analysis of advertisement and release calls,
mean values were generated for each of the 3
call variables from 3 or more calls.
Historic Distributions
We obtained specimen listings from the fol-
lowing institutions: American Museum of Nat-
ural Histoiy (AMNH), Arizona State Univer-
sity (ASU), Brigham Young University (BYU),
California Academy of Sciences (CAS), Carne-
gie Museum of Natural Histoiy (CMNH), Los
Angeles County Museum (LACM), Museum
of Vertebrate Zoology (MVZ), University of
Arizona (UA), University of Michigan Museum
of Zoology (UMMZ), University of New Mex-
ico (UNM), and United States National Muse-
um (USNM). It is important to note that we
examined only specimens deposited in the ASU
collection and a portion of those housed at the
USNM. We assume that anurans listed by the
other institutions are conectly identified. Given
that these 3 anurans are quite distinct from
other Sonoran Desert forms and therefore un-
likely to be misidentified, it seems reasonable
to accept these listings in lieu of a physical
40
Great Basin Naturalist
[Volume 56
examination of all specimens. We did, however,
obtain detailed information from collectors for
any specimen collected outside or on die periph-
ery of the range (e.g., San Xavier region).
Results and Discussion
Bufo retifonnis
Relative to other toads (genus Bufo) found
in south central Arizona, B. retifonnis pos-
sesses an unusually high-pitched, short-dura-
tion advertisement call, often described as an
"insect-like buzz" (see Stebbins 1985, Hulse
1978). However, given similarities in adver-
tisement calls of B. retiformis and G. olivacea,
identification based on calls can only be confi-
dently determined with analysis of signals in
the laboratory (Sullivan unpublished data). On
average, B. retiformis calls are longer (/x = 3.0
sec, range = 2.0—4.3 sec at approximately 26°
C body temperature) and lower in frequency
ilJL = 3112 Hz) than calls of Gastrophryne
(typically 1-2 sec duration at =4000 Hz).
Historic distribution. — Bifo retiformis is
known from west central Sonora and south
central Arizona (Hulse 1978; Fig. 1). Since it
was described in 1951, this anuran has been
obsei-ved in Arizona at sites ranging from near
San Cristobal Wash, just west of Organ Pipe
Cactus National Monument, north to tribu-
taries of Waterman Wash near Mobile, south-
east to the vicinity of Tucson (San Xavier Mis-
sion), and southwest to the international bor-
der near Sasabe. Across this region it occurs in
creosote flats, upland saguaro-palo verde asso-
ciations, and relatively high-elevation (>900
m) desert grassland.
One historic locality deserves special dis-
cussion: southern Vekol Valley, Pinal Count)'.
At this site Jones et al. (1983) reported bodi B.
retiformis and B. debilis. We have examined
the single voucher specimens for B. retiformis
(USNM 252797) and B. debilis (USNM 252776;
SVL = 43 mm, reproductive female) and deter-
mined by comparison with juvt-niles in the
ASU collection (ASU 23099-23102) that the
putative B. debilis is not simply a juvenile B.
retijormis. Using the morphometric methods
proposed 1)\ Ferguson and Lowe (1969), we
scored diis indixidual close to B. debilis in all
respects; hence, the B. debilis individual can-
not be disnussed as a simiije nnsidentification
or hybrid. The presence of /i debilis well with-
in the range of R retiformis is especially prob-
lematic. No B. debilis have been recorded from
appropriate habitat spanning the 240 km be-
tween Vekol Valley and the otherwise western-
most previous locality for this eastern relative
of B. retiformis (near Benson, Arizona). Unfor-
tunately, we were unable to sui'vey Vekol Val-
ley when conditions were suitable for anuran
activity.
Present distribution. — In 1993-94, we
obsei^ved B. retiformis at or near most historic
localities, except San Xavier and Vekol Valley,
and at additional sites (Fig. 1). They were
especially abundant along Indian Route (IR)
15, 0-40 km north of Quijotoa, associated with
the Santa Rosa Wash floodplain. Surveys in
which every anuran was identified along a
roadway segment (1-65 km) revealed that B.
retiformis constituted up to 63% of all anurans
sighted on this route (Table 1), whereas they
were absent or composed a small proportion
(<1%) of total anurans sighted on roadways on
the peripheiy of their distribution near Mobile
and Sasabe (Table 1). Similarly, this toad was
not abundant along State Route (SR) 85 near
Organ Pipe Cactus National Monument. Din-
ing 1993 and 1994 we never obserxed this
species on SR 85 or SR 86 in this westernmost
portion of the range. Philip Rosen (personal
communication) has observed only a few B.
retifonnis near the international border, and a
number of individuals near Why, Arizona, dur-
ing the course of extensive fieldwork near
Organ Pipe Cactus National Monument over
the past 6 yr
Contraiy to the suggestion ol Hulse (1978;
see also Nickerson and Mays 1968), Bufo reti-
formis does not appear to be expanding its
range northward into areas of agricultural activ-
ity (e.g., soutiiern Pinal County). We conducted
many sin"\ eys in southern Pinal County: south
of Stantield and south of Arizona City, 2 areas
directly north of known localities for B. reti-
formis (Fig 1). We also extensively surveyed
the Avra Valley region, Pima County, immedi-
ately west of Tucson, and the \icinity of Mobile,
Maricopa County. These habitats are similar to
areas inhabited by B. retiformis directb' to the
south or west, except that agricultural activity
is relati\t'K higher in these areas. It appears
that B. retiformis is less conmion on die periph-
eiy of its range: near Organ Pipe Cactus Nation-
al Monument in the west, near Mobile in the
noitli, and in Altar Vallev in the east.
1996J SoNoiUN Desert Anuiuns
a) Historic collecting localities for Bufo retiformis in south central Arizona.
41
b) Recent collecting localities for Bufo retiformis in south central Arizona.
Fig. 1. Map of a) historic distribution (•) and b) present distribution (•) of Bufo retifonim in south central Arizona.
Breeding ACTiviri'. — Like many explosive
breeding desert anurans, B. retiformis will take
advantage of a variety of water sources for repro-
duction. We observed chorusing activity in
cattle tanks and roadside pools associated with
washes. We obsei-ved B. retiformis breeding in
the same pool with all other explosive breed-
ing anurans that occur in south central Arizona:
B. ulvarius, B. cogmitus, B. punctatus, Gastro-
phnjne olivacea, Pternohylafodiens, Scaphiopus
concha, and Spea multipUcata. We never ob-
served B. retiformis breeding in the absence of
other anurans — minimally, B. cognatus and S.
concha bred sympatrically with B. retiformis.
Male B. retiformis typically call positioned
beneath vegetation (e.g., small shrubs or grass),
42
Great Basin NATUii\LiST
[Volume 56
Table 1. Numbers of anurans individualK' identified on road surface over a specified distance. Bal = B. alvarius, Bco
= B. cognatus, Bpu = B. punctatiis. Ere = B. rctifonnis, Sco = Scaphiupus cotichii, IR = Indian Route, SR = State
Route. MM = mile marker
Location
(appro.ximate)
Siuvey
distance (km)
Species
Date
Bal m
Bco {9c)
Bpu m
Bre {%)
Sco
m
Total
7/18/94
SR 286
40
31 (33)
13 (14)
—
—
49
(53)
93
7/28/94
.\rizona Cit>'
24
4 (40)
2 (20)
—
—
4
(40)
10
7/29/94
SR 286
72
13 (18)
13 (18)
5 (7)
—
39
(56)
70
8/7/94
Mobile
25
5 (18)
—
3 (11)
1 (3)
19
(68)
28
8/8/94
SR 286
24
3 (23)
4 (31)
—
—
6
(46)
13
8/8/94
Mobile
30
10 (14)
4 (6)
5 (7)
—
51
(73)
70
8/13/94
IR 15. MM 11
4.8
1 (3)
1 (3)
1 (3)
9 (28)
20
(63)
32
8/15/94
IR 15, M.\l 11
3.4
2 (25)
—
—
5 (63)
1
(13)
8
9/10/94
Stanfield
5.3
64 (75)
9 (10)
1 (1)
—
11
(13)
85
1-5 111 from the water's edge. Amplexus is ini-
tiated on land with the t>'picall\' larger female
earning the male to water for o\iposition. In
high-density aggregations, satellite males can
be common — we saw as man\' as 3 non-calling
males near 1 calling male.
Chorusing males and ample.xing pairs were
obsen^ed on onK' 4 occasions. Three breeding
aggregations along IR 15 were relatively large
and located at sites used regularly in the past
(e.g., 1984, 1986, 1988; Sullivan and Bowker
unpublished). At mile marker (MM) 18.7 on
IR 15 north of Quijotoa, a large aggregation
formed in a shallow roadside pool (8/9/93).
Unfortunately, direct counts of all indi\iduals
present were not possible due to restricted
property access, but complete counts of all
males and females along an open section of
the pool shoreline (23 calling and satellite
males, 5 females in 75 m) allow a rough mini-
mum estimate of >2()0 males and females for
the entire pool (=600 m circumference).
Obsenations at a 2nd site that same night, a
cattle tank (=25 X 50 m) near MM 8.5, nortli
of Quijotoa, indicate a thriving population in
spite of hybridization with B. piinctatiis (see
below). On the 1st night (8/9/93) following
hea\->- rainfall in this area, we counted 20 male
B. retifonnis at 0300, calling with numerous B.
alvariiis, B. cognatus, and B. punctatus. On the
following night (8/10/93), appro.\imat(>l\- 40
male B. retifonnis were obsened, in additit)ii
to a niiiiimimi of 5 pairs in ample.xus. A 3rd
l^reeding aggregation (8/25/94) at a roadside
pool (=50 X 25 m) at MM 11 on IR 15 west of
Santa Rosa comprised 19 calling males and 5
amplexing pairs (direct count of all individu-
als). In contrast to these relativeb' vigorous
aggregations, onK' 6 males and a single female
were obsened at a "first-night" choiois (8/20/93)
in a large cattle tank (=25 X 75 m) near Gun-
sight Wash along SR 85.
Hybridization with Bvfo puxctatvs. —
Bowker and Sullivan (1991) documented a
naturally occurring hybrid between B. reti-
fonnis and B. punctatus. and we obsened 3
additional h\ brids during oiu" in\estigation (all
in August 1993). These Inbrids were obsened
along IR 15, 10-20 km north of Quijotoa.
Hybrids are intermediate to the 2 parental
forms and unlikeK to be confused with an>
other anurans in the \ icinit). Gi\ en the appar-
ent rareness of hybrids, it is unlikeK that they
present a significant concern for the popula-
tion status of either parental form.
Hxbridization between B. punctatus and B.
retifor))iis is somewhat surprising gi\en dra-
matic differences in their advertisement calls
and habitat preferences (Ferguson and Lowe
1969). Three factors may facilitate li\bridiza-
tion between B. punctatus and B. retifonnis
along IR 15 north of Quijotoa. First, along IR
15 we observed relatively high numbers of B.
retifonnis compared to B. punctatus. and we
also noted satellite males near calling males in
these aggregations. Male mating tactics such
1996]
SoNORAN Desert Anurans
43
as active searching and satellite behavior can in-
crease the probability of heterospecific crosses
since these tactics subvert active choice by
females. Second, although B. retifonnis is typi-
cally found in desert flats and B. punctatus
generally occurs in rockier, upland regions,
the "hybrid zone" along IR 15 (MM 6-12) rep-
resents a transition between lowland (Lower
Colorado River Subdivision) and upland (Ari-
zona Upland Subdivision) desert habitats that
would allow coexistence of both species. Third,
habitat modification at the site, namely, road
construction and development of cattle tanks,
may overcome ecological separation between
the species and provide opportunities for
hybridization.
Gastrophryne olivacea
As noted above, the advertisement call of
G. olivacea can be confused with B. retifonnis.
In the hand, this small, narrow-mouthed toad
cannot be confused with any other species
found in Arizona (Nelson 1972a, 1972b, 1973,
Stebbins 1985). Identification based on calls
(insect-like buzz) alone must be corroborated
by laboratoiy acoustic analysis.
Although Lowe (1964) listed G. carolinensis
from the mountains near Nogales, Arizona,
Nelson (1972a, 1972b) showed that these indi-
viduals do not differ significantly from nearby
populations of G. olivacea from lower-eleva-
tion sites. Having examined specimens from
throughout the range in Arizona, we concur
with Nelson that only a single taxon occurs
north of the international boundary.
Historic distribution. — The range of G.
olivacea largely overlaps that of B. retifonnis
(Fig. 2), except in Santa Cruz County (e.g.,
near Pena Blanca) where Gastrophryne occurs
farther east. Of the 3 anurans surveyed, this
species occurs in the widest variety of habitats
in Arizona, ranging from low-elevation cre-
osote flats through grasslands to oak-woodland
communities near Ruby, Arizona (>1200 m).
Wake (1961) reported calling G. olivacea
4.8 km southeast of Ajo. Because no individu-
als were visually confirmed and because of the
difficulty of identifying this species by call,
we are inclined to discount the record.
Present distribution. — In 1993-94 we
obsei-ved G. olivacea at most historic localities
except those on the eastern margin of the
study area (San Xavier and vicinity of Pena
Blanca), and at some new sites (Fig. 2). We
observed a small chorus near Lukeville, just
north of the international boundaiy, a site that
extends the range of Gastrophryne approxi-
mately 58 km southwest of the previous west-
ernmost locality (San Simon Wash, SR 86) in
the United States. Philip Rosen (personal com-
munication) suggests that Gastr()})hryne is more
abundant in Mexico to the south and southeast
of Lukeville. The absence of previous distribu-
tional records from Organ Pipe Cactus National
Monument substantiates the notion that G.
olivacea reaches its northwestern range limit
in this area.
We were unable to document G. olivacea
anywhere along SR 286 (Altar Valley, Buenos
Aires Refuge) in spite of apparently adequate
habitat and the presence of G. olivacea to the
east. Philip Rosen (1994 personal communica-
tion) obsei-ved a number of G. olivacea breed-
ing choruses in southwestern Santa Cruz
County, just east of the Buenos Aires Refuge
boundary, during summer 1994. Hence, this
species likely occurs in the area but, like B.
retifonnis, may be less abundant along SR
286. We did not find G. olivacea in the vicinity
of San Xavier Mission or along SR 289,
although we visited these sites after rainfall on
several occasions. Our failure to document
Gastrophryne in areas with appropriate habi-
tat may be an artifact of its secretive habits
(i.e., individuals may not come on road sur-
faces) and small size (i.e., they are difficult to
detect when on a road).
Breeding activit\'. — Gastrophnjne olivacea
aie usually well concealed in vegetation when
calling and possess a call that is extremely dif-
ficult to localize. They call next to water sources
or from floating vegetation. Male satellite
activity was not observed. Although G. olivacea
has been observed in choruses with all other
sympatrically breeding anurans (see above list-
ing under B. retifonnis), on many occasions we
observed it in large, relatively monotypic
aggregations (e.g., MM 26.7 and 35, IR 15). In
these areas Gastrophryne often breeds in dense
stands of mesquite shrubs growing in the flood-
plain of Santa Rosa Wash.
Choruses of Gastrophryne are easily de-
tected, and we were led to a number of new
Gastrophryne localities by their distinctive
vocalizations. Because of their secretive nature,
we never observed pairs in amplexus, and thus
no definitive estimates of population size were
obtained for breeding choruses of G. olivacea.
44
Great Basin Naturalist [Volume 56
a) Historic collecting localities for Gastrophryne olivacea in south central Arizona.
b) Recent collecting localities for Gastrophryne olivacea in south central Arizona.
Fig. 2. Map of a) historic' distrihutioii (•) aiul h) present distriliiition (•) of Gastrophryne olivacea in sonth central Ari-
zona.
By walking the perimeter oi rain-formed i)o()ls,
we obtained rough estimates of >2()() ealling
males at 2 sites along IR 15, 43 and 56 km
north of Quijotoa, respectively, on recent
(8/9/93) and previous sinveys (19(S4: Sulh\an
and Bowker impublished). UnfortunateK, since
these pools contained considerable vegetation
(mesquite shrubs, grass), chorus sizes can onl>
be considered approximate (individual toads
were not visually verified). B\ contrast, at Luke-
\ ille (8/9/94) ouK- 5 calling males were present
in a small pool (5 X 10 X 0.25 m). Rain had
fallen the previous 2 nights (8/7-8/8), and sev-
eral small egg masses were obseiA'ed.
Ptcniohyla fodicns
Hie advertisement call oi Ptcrnohijlafodiens
is a distinctixe "wonk" repeated at a relatively
high rate (2/sec: "wonk- wonk- wonk . . . ," etc;
see Trueb 1969). Males also produce a call,
19961
SoNOiuN Desert Anukans
45
a) Historic collecting localities for Pternohyla fodiens in south central Arizona.
b) Recent collecting localities for Pternohyla fodiens in south central Arizona.
1 '
-MARICOPA COUNTY
-,^4^'
^ J
r—(¥>.
4
\ ...
\^ '
Hickiwan
■- ^ '^*
\
t^yj
' \
Organ
Pipe /
m
PINAL COUNTY
Queens Well
^Quijotoa
Sells
PIMA COUNTY
I SANTA CRUZ
COUNTY
.Nogales
Fig. 3. Map of a) historic distribution (. ) aiid h) present distribution (•) oi Pternohyla fodiens in south central Arizona.
which, based on simihirities with other hyhds,
can be tentatively classified as a territorial call.
This putative territorial call sounds much like
the advertisement call of Pseiidacris triseriata
or the sound of a finger sliding across a comb.
Historic distribution. — This anuran has
been obsei-ved at a few sites (Fig. 3). All locali-
ties but Santa Rosa Wash are associated with
washes that flow south toward Mexico: San
Simon Wash, and its 2 largest tributaries, Hicld-
wan and Sells washes. Randy Babb (personal
communication) has heard the distinctive
vocalization of Pternohyla fodiens many times
and visually identified at least 1 individual
approximately 16 km north of Quijotoa, west
of IR 15, in the floodplain of Santa Rosa Wash.
Present distribution. — In 1993-94 we
obsei-ved P. fodiens at most historic localities
except Santa Rosa Wash and the vicinity of
Sells, and at some additional sites (Fig. 3).
46
Great Basin Naturalist
[Volume 56
More than the other target species, P. fodiens
is found in association with washes. The 2 new
locahties we documented are both associated
with small tributaries of Sells Wash, a tribu-
taiy of San Simon Wash.
During the preparation of this report,
Thomas R. Jones and Ross J. Timmons (per-
sonal communication) found a single male P.
fodiens near Santa Rosa Wash, 1 km north of
the Pinal County line and west of IR 15 (12
July 1995). This record confirms the presence
of P. fodiens in Santa Rosa Wash, well north of
the San Simon Wash system.
Pteniohyla fodiens is only rarely found on
road surfaces, although specimens can be taken
near washes when roads are wet (e.g., SR 86 at
San Simon Wash). Similar to Gastrophrync,
Pternohyla can be easily missed unless chorus
activity is underway when a survey is con-
ducted. Because of their extremely explosive
breeding habits and the lack of sufficient rain-
fall near Sells during the survey period, it is
not surprising that we obsei-ved no Pternohyla
at the historic localities along Sells Wash near
SR 86.
Breeding activit\\ — We observed breed-
ing aggregations of Pternohyla fodiens only in
rain-formed pools associated with washes.
Calling males are always in or near water, and
of the 3 survey anurans Pternohyla seems
more dependent on heavy rainfall to initiate
breeding activity. This species appears to
exhibit the most explosive mating system of
the 3 species. We never obsei"ved Pternohyla
chorusing more than 36 h after rainfall; by
contrast, both Gastrophryne and Bufo were
observed in chorus activity 1-4 nights follow-
ing rainfall.
The only significant Pternohyla chorusing
that we observed occiured near Hickiwan
(7/13/93) and San Simon Wash (7/13/93).
Although direct coimts were not possible, esti-
mates from chorusing intensities suggest that
dozens, if not hundreds, of calling males may
have been present at San Simon Wash along
SR 86; however, only a single pair in amplexus
was obser\'ed. Large aggregations of Pterno-
hyla have been observed at these sites regu-
larly over the past 30 yr (Sullivan and Bowker
unpublished).
Summary
Oui- siu-\ey indicates that all 3 target species
are present at most historic localities in south
central Arizona. We documented range exten-
sions to the northwest and southeast for B.
retiformis (Mobile/SR 286) and to the south-
west for Gastrophryne olivacea (Lukeville).
These forms probably occur at all historic
localities, since our inability to verify their
presence at some sites undoubtedly resulted
from the absence of sufficient rainfall. It is
critical to note that our survey methods,
although allowing rapid coverage of a rela-
tively large area, were limited by unpredict-
able rainfall and the secretive nature of the
target species (especially Pternohyla and Gas-
trophryne). Unless chorusing activity was
undei^way when we visited an area, the pres-
ence of any of the 3 forms may have been
overlooked. In the absence of chorusing activ-
ity, Bufo retiformis was the only target species
regularly found on road surfaces.
Minimally, the presence of these anurans at
most historic localities suggests no widespread
decline as experienced by other anuran amphib-
ians in the United States (e.g., ranid frogs of
the Southwest; Michael Sredl personal com-
munication). Future work should address esti-
mation of population levels through mark-
recapture methods in conjunction with inten-
sive monitoring of single sites throughout as
many consecutive activity' periods (June— Sep-
tember) as possible. An understanding of fac-
tors contributing to variations in species abun-
dance will require long-term study.
Acknowledgments
This research was supported by an IIPAM
award (192004) from the Arizona Game and
Fish Department Heritage Fund. We grate-
fulK' acknowledge the assistance of the Tohono
O Odham Nation, especially the Department
of Public Safety personnel. In addition, Henn'
Ramon of the Hickiwan District, Norbert
Manuel of the Sells District, and Madeline
Sakiestewa and Jefford Francisco of the Babo-
quivari District were especially helpful in co-
ordinating activities. Mike Demlong, Robert
Dudley, Matthew Goode, Matthew Flowers,
and Michael Sredl pro\'ided assistance with
field observations. Randy Babb, Darrel Frost,
Jeff Howland, K. Bnice Jones, Thomas R. Jones,
Cla\t()n Ma\, Phil Rosen, Cecil Schwiillie, Nonn
Scott, and Michael Sredl graciousK' shared
their field records and experiences.
1996]
SoNORAN Desert Anurans
47
LiTER.\TURE Cited
Appendix 1
BOWKER, R. W., AND B. K. SULLlVAN. 1991. Anura: Bitfo
punctatus X B. retiformis natural liybridization. Her-
petological Review 22: 54.
CnRAPLiwT, R S., AND K. L. Williams. 1957. A species of
frog new to the fauna of the United States: Ptemo-
hyla fodiens Boulenger. Chicago Academy of Sci-
ence, Natural Histoiy Miscellaneous Publication
160: 1-2.
Ferguson, J. H., and C. H. Lowe. 1969. Evolutionaiy
relationships of the Bufo punctatus group. American
Midland Naturalist 81; 435-446.
Hl'LSE, A. C. 1978. Bufo rctifonnis: Sonoran green toad.
Catalogue of American Amphibians and Reptiles
207: 1-2.
Jones, K. B., L. Porzer Kepner, and W. G. Kepner.
1983. Anurans of Vekol Valley, central Arizona.
Southwestern Naturalist 28: 469—170.
Lowe, C. H. 1964. The vertebrates of Arizona. University
of Arizona Press, Tucson.
Nelson, C. E. 1972a. Gastrophnjne olivacea: western
narrow-mouthed toad. Catalogue of American
Amphibians and Reptiles 122: 1-4.
. 1972b. Systematic studies of the North American
microhylid genus Gastrophnjne. Journal of Herpe-
tology6(2): 111-137.
. 1973. Gastrophnjne: narrow-mouthed toads. Cat-
alogue of American Amphibians and Reptiles 134:
1-2.
NiCKERSON, M. A., AND C. E. Mays. 1968. Bufo retiformis
Sanders and Smith from the Santa Rosa Valley, Pima
County, Aiizona. Journal of Heipetology 1(1-4): 103.
Sanders, O., and H. M. Smith. 1951. Geographic varia-
tion in toads of the dehilis group of Bufo. Field and
Laboratory 19(4): 141-160.
Stebbins, R. C. 1985. A fieldguide to western reptiles and
amphibians. Houghton Mifflin Press. 589 pp.
Sullivan, B. K. 1992. Calling behavior of the southwest-
ern toad [Bufo inicroscaphus). Herpetologica 48:
383-389.
Trueb, L. 1969. Pternohijla fodiens: bunowing treefi^ogs.
Catalogue of American Amphibians and Reptiles 77:
1-4.
Wake, D. B. 1961. The distribution of the Sinaloa nanow-
mouthed toad Gastrophnjne mazatlanensis (Taylor).
Southern California Academy of Science Bulletin
60(2): 88-92.
Wells, K. D. 1977. The social behaviour of anuran
amphibians. Animal Behaviour 25: 666-693.
Williams, K. L., and E S. Chrapli\w. 1958. Selected
records of amphibians and reptiles from Arizona.
Transactions of the Kansas Academy of Science 61:
299-301.
Specimen numbers for historic collecting localities for
Bufo retiformis, GastropJmjne olivacea, and Pternohijla
fodiens. Institutional abbreviations: AM Nil = American
Museum of Natural Histoiy, ASU = Arizona State Uni-
versity vertebrate collection, BYU = Brigham Young Uni-
versity collection, CAS = California Academy of Sciences,
CMNH = Carnegie Museum of Natural History, LACM
= Los Angeles County Museum, MVZ = Museum of Ver-
tebrate Zoology, UAZ = University of Arizona, UMMZ =
University of Michigan Museum of Zoology, UNM =
University of New Mexico, USNM = United States
National Museum.
Bufo retiformis: AMNH 59189, 60671, 85357-65, 91953-
54, 1022.34-36; ASU 3298-3300, 3894-3902, 3942-48,
8002, 8004, 8005, 22775-76, 23099-102, 23252, 24038-39,
24273-74, 25552-53; BYU 42119; CAS 91.501-04, 94390-
95, 98055-56, 188354-55; CMNH 51562, .53841-42, 538.55,
63,520, 89782-95; LACM 26086-88, 64180-84, 88380-400,
91833, 105719, 11.5266-314, 12.3234-41, 137788-89; MVZ
71906-07, 73751-52, 74206-32, 76620-28, 81269, 139130,
180219-22, 180358-59; UAZ 12369-75, 14848-49, 25847-
48, 31381, 4,3011; UMMZ 133460, 1,36,395, 134077; UNM
30993-995, 31268, 40207, 41686-87; USNM 226443-45,
24,5988, 252797, 322966.
Gastrophnjne olivacea: AMNH 88986, 91971-80, 119746;
ASU 14014, 22059-60, 22224-25, 22969-70, 22771-74,
2,3095, 23411, 24259-60, 25664-66; CMNH 63138-,39;
LACM 26576-81, 91896, 115511, 112480, 12,3293; MVZ
49479-,504, 58922, 72304-05; UAZ 26993-96, 29101-04,
29107, 42187-91, 38181, 35163-64, 38179, 38200-01,
38180, 38197-99, 29027; USNM 252817; UMMZ 136400,
75737-38, 757,53, 92300.
Pternohyla fodiens: AMNH 91964-70, 95147; ASU 3301,
1,39,52-68, 22777-80, 24276, 25,556-61; CAS 91505; CMNH
63188-89; LACM 90170-82, 11,5447-75; MVZ 71905,
73747-48, 80104-21, 81271, 178447, 76629-,33; UNM
40201, 40204.
Received 4 May 1995
Accepted 1 September 1995
Great Basin Naturalist 56(1), © 1996, pp. 48-53
HABITAT AFFINITIES OF BATS FROM NORTHEASTERN NEVADA
Mark A. Ports l and Peter V Bradley^
Abstract. — Bat surveys were completed in 6 habitat types in eastern Nevada between 1980 and 1994. Twelve
species of bats and 578 individuals were identified fioni 33 trap localities in 144 trap nights. There were weak correlations
between bat species richness and Januan maximum temperatures (0.728, P < 0.05) and mean annual days widi 0° C or
lower (-0.704, P < 0.05). Bat species richness exhibited no correlation with annual normal precipitation, Januaiy mini-
mum temperatiu-es, July minimum temperatures, and July maximum temperatures. It appears that bat species richness
is highest in portions of northeastern Nevada typified by sedimentary' deposits (limestone, dolomite). Igneous mountain
ranges (basalt, volcanic ash) generally had moderate bat species richness, and metamoiphic mountain ranges (quartzite)
t^'pically had low bat species richness. Notable range extensions include Antrozoiis paUidus (from central Nye Countv'
north to the Nevada-Idalio border, approximately 450 km), Tadarida brasiliensis (approximately 350 km north), and Pip-
istrellus hespenis (approximately 350 km north). Also, the presence of Lasiomjcteris noctivagans. Ldaiunis cinereus, and
Corijiiorluiuis fownsendii was confirmed.
Key uords: bats, Chiropfera. Nevada, habitat.
Although the distribution of mammals of the
Great Basin has been studied in some detail
(Hall 1946, Dun-ant 1952, Brouai 1971, Thomp-
son and Mead 1982, Wells 1983, Grayson 1987),
bats remain poorly known. There are verv" few
recent records of bats from the northern Great
Basin of Oregon, Idalio, and Nevada (Hall 1946,
Durrant 1952, Larrison and Johnson 1981).
Here we present new information on habitat
affinities and distribution of 12 species of bats
from eastern and northeastern Nevada. Such
information may prove valuable to land man-
agers and wildlife biologists who make deci-
sions on how to deal with the impact of human
activities on bats.
Methods
Study Area
Northeastern Nevada is part of the Great
Basin Division of the Intermountain Floristic
Region (Holmgren 1972), an area of continen-
tal climate with fairly hot summers and cold,
snowy winters. Some 30 north/south-trending
fault-block mountain ranges (3000— 1000 m) are
separated by high-ele\'ation (15()()-2()0() m)
xeric basins.
Mountain ranges in northern Elko, Eureka,
Humboldt, and Lander counties are mosth'
igneous and metamorphic fault blocks, coNcred
with \ arious mountain brush communities and
fragmented coniferous and deciduous forests.
Perennial streams produce riparian habitats in
most canyons. Vertical cliffs and stands of de-
ciduous and coniferous trees provide sites for
da\' roosting and shelter for maternity acti\i-
ties. Valle\' floors are mosth' xeric, co\ered
with salt-tolerant shrubs {Atriplex spp., Sarco-
batus spp.) and sagebrush {Artemisia spp.).
Occasional perennial streams extend onto val-
le\' floors and are lined with narrow coiridors
of deciduous woodlands and mesic shrubs.
Mountain ranges in eastern Nevada (White
Pine and southern Eiueka and Lander coun-
ties) are predominantK' limestone and dolomite
fault blocks and tend to have more xeric plant
communities. A large number of natural caves
and vertical cliff sites provide excellent habi-
tats for bat maternit) and hibernation roosts.
Natural perennial springs found near the val-
ley/mountain fault lines often provide the only
dependable water for miles around. Contigu-
ous coniferous forests on some of the higher
mountain slopes provide suitable tree roosts.
Abandoned mine shafts and adits are abundant
in northeastern Nevada and are criticalK' im-
portant to some bat species, botli siunmer and
winter.
Sur\ e\' Methods
Sun eys began in the smnmer of 1980 and
extended through the fall of 1994. Capture
'Biolo)^- Departim-nt, Great Basin Colliue, 1.500 College Parkway, Elko. N\' 89801.
^Nevada Division orWikllilc, 137.5 Mountain Citv Hut.. Elko, NV 89801.
48
1996]
Bats from Northeastern Nevada
49
methods included mist nets, hand capture, and
harp trap (Kunz and Kurta 1990). Mist nets
and the harp trap were used over perennial
streams, small springs, beaver ponds, livestock
tanks, in forest canopies, and adjacent to mine
shafts, adits, and natural caves. Captured bats
were identified, sexed, reproductive status
recorded, aged, weighed, and then released.
Some indi\'idiials were taken as voucher speci-
mens and are temporarily held in tlie vertebrate
collection of Great Basin College. S. Altenbach
(personal commimication) and M. OFairell (per-
sonal communication) assisted in identifica-
tions. Localities were identified on 1:100,000
scale metric topographic maps.
To describe habitat affinities, we delineated
6 general habitat types for the region: C-river
canyons in igneous or metamoiphic rock, above
low-gradient, perennial streams lined with
Cottonwood {Popiilus spp.), willow {Salix spp.),
and mesic shrubs {Roso spp. and Ribes spp.),
elevation approximately 2200 m; S-foothill
and valley springs, with or without deciduous
trees and a surrounding area of salt-tolerant
shrubs {Athplex spp., Sarcobatiis spp.) or
mountain brush {Artemisia spp., Amelanchier
spp., Sambiicus spp., Syrnphoricarpos occiden-
talis, Purshia tridentata) communities, eleva-
tion approximately 2000 m; F-mid- to high-
elevation coniferous forests of juniper {Jiini-
penis osteosperma), fir {Abies concolor and A.
lasiocarpa), spruce {Picea engehnannii), and
pine {Pinits monophylla, P. flexilis, and P. lon-
gaeva) often with cliff sites and natural caves
in the proximity, elevation approximately
230O-.3000 m; D-mid- to high-elevation decidu-
ous forests of aspen {Populus tremuloides), Cot-
tonwood {Populus spp.), and mesic shrubs
{Amelanchier spp., Prunus spp., Betula occi-
dentalis, Ahius tentdfolia) often along high-
gradient, perennial streams, elevation approxi-
mately 2300-2800 m; U-natural caves and
underground mine shafts/adits with surround-
ing plant communities described in habitats
C, F, S, and D; and B-buildings in towns and
on ranches. There may also be additional
important bat habitats not yet identified in this
region.
Results and Discussion
A total of 578 individuals of 12 species of
bats were identified from 33 trap localities in
144 trap nights from eastern and northeastern
Nevada (Tables 1, 2 and Appendix 1). Three
species of Myotis, (M. evotis, M. volans, and
M. ciliolahrum) were the most widespread
(Appendix 1) and had the highest occurrence
(Tables 1, 2) of bats from eastern Nevada. M.
evotis was one of the most abundant species of
Myotis in eastern Nevada and occurred in all
habitats except towns and around buildings.
This species is most often associated with mid-
elevation pinyon pine and Utah juniper wood-
lands (Manning and Jones 1989). We, too,
found this species to be most abundant in this
habitat type (localities 8, 9, and 18, TdhXe 1).
M. evotis depended heavily on the presence of
natural springs within these woodlands as their
sole source of water. M. volans was also found
to utilize a variety of habitats in eastern Nevada,
including pinyon-juniper woodlands such as
those found near Old Man's Cave. Eight lac-
tating females were examined at this site, sug-
gesting a nearby nurseiy colony. Upon release,
4 individuals flew into the cave while the oth-
ers flew to nearby rock outcrops. The litera-
ture suggests that this species uses cracks in
cliff sites and areas beneath bark as roost sites
and caves only as hibernacula (Warner and
Czaplewski 1984). It is possible that M. volans
is using caves in easteiTi Nevada as maternity
roosts, although more data are needed to con-
firm this. M. ciliolabrum also occurred in a
variety of habitats in eastern Nevada (Table 1),
including river canyons with sunounding sage-
brush deserts (locality 14, Appendix 1). Lamson
and Johnson (1981) found this species in simi-
lar canyon and desert habitat in central Idaho.
Only 6 individuals of M. htcifiigus were
caught. This species was uncommon and more
restricted in its habitat affinities. Unidentified
specimens of Myotis were sent to Dr. Scott
Altenbach and Dr. Mike O'Farrell to deter-
mine whether or not M. californiciis is present
in this region (Table 2, Myotis spp.). Tentative
identifications suggest that M. californicus
may be found in southern White Pine County,
while M. ciliolabrum is more common in the
remainder of the region.
The 3 high-elevation, tree-roosting species
(L. noctivagans, E. fusciis, and L. cinereus) were
found in order of decreasing occuirence (Table
1). These species were found repeatedly in
several mountain ranges of eastern Nevada
that have a combination of coniferous and/or
deciduous trees (aspen, cottonwood, white fir,
subalpine fir, and Engelmann spruce) for
50
Great Basin Naturalist
[Volume 56
Table 1. Occurrence of bat species by locality (see Appendix 1). Habitat affinities (C-river canyons, S-springs,
F-high-elevation coniferous forests. D-mid-elevation deciduous forests, U-underground caves and mines, B-buildings)
for each species and relative frequencies for each species examined.
Bat species
Localities
(Appendix 1)
Habitat
affinities
Mtjotis ciliolabrinn
Mijotis evotis
Mijotis hicifugiis
Mijotis volans
Lasiiinis cinereus
Lasionycteris nocfivagans
Eptesicus fnsciis
Pipisf reikis hespcnis
Conjnorhiniis townsendii
Antrozous palUdus
Tadarida hrasiliensis
2,6,8,9, 10, 11, 12,14,
17, 20, 25, 26, 29, 32, 33
1,3,4,6.8,9,11, 12,
15-19, 21, 22, 25, 32, 33
5, 12, 15-17
1, 2, 6, 7, 9-12, 15, 17-19,
24, 25, 27, 32
10, 17, 20
10-12, 17, 23, 28, 29, 32
10, 12, 17, 23, 26, 29, 32
10,29
5, 9, 10, 13-15.
24-27. 30. 32
10, 14, 15, 25
10, 29. 31. 32
C, S, E D, U, B
C, S, E D, U
C, E D, U
C, S, E D, U
S.ED
C, S, E D, B
C, S, E D, U, B
S, B
C, S, U
C, S, U
S. U. B
roosting and open water in the form of beaver
ponds, stock tanks, and perennial streams for
foraging and drinking sites. In the mountains
of the West, these 3 species are known to com-
monly forage together in similar habitats along
with 2-4 species of Mijotis (Kunz 1982). In
eastern Nevada high-elevation deciduous and
coniferous forests are limited to watered
drainages and north-facing slopes in the larger
mountain ranges. This suggests that these
species are uncommon when compared to
populations in the northern Rocky Mountains
and may be negatively impacted by deteriora-
tion, fiagmentation, and/or total removal of for-
est habitats by hard-rock mining, livestock graz-
ing, and logging.
Foothills covered with pinyon pine and Utah
juniper, caves, and river canyons with high
cliffs provided habitats for 2 lower-elevation
breeding species, Corynorhinus townsendii and
A. pallidus. C. townsendii had 4 times the fre-
quency of occurrence as A. pallidus and
appeared to be more evenly distributed across
the region (Table I). C. townsendii and A. pal-
lidus depend heavily on cliff sites, natural caves,
and mine shafts/adits for maternity, hiberna-
tion, and day roosts in eastern Nevada. They
are found to utilize similar situations in other
arid regions of the West, such as California,
Montana, Washington, and Utah (Kunz and
Martin 1982). Hermanson and O'Shea (1983)
rarely found A. pallidus using caves, but rather
found them depending heavily on crevices
and cliff sites for maternity roosts, day roosts,
and hibernacula. We found this species using
caves (localities 15, 25), cliff sites (14), and val-
ley springs (10) in eastern Nevada.
A large, historic colony of T. hrasiliensis
was found occupied in July 1994. Vandalism
may have caused this population to roost else-
where in 1992 and 1993. Outside of Las Vegas
and Reno, this colony is the largest known
concentration of mammals in Nevada. Based
on visual techniques suggested by Kunz and
Kurta (1990), we estimate the population at
between 54,000 and 82,000 animals.
P. hesperus was found in low numbers in
this region. Two individuals were caught 320
km apart, and no meaningful habitat patterns
were identified for this species.
Species found in and around abandoned
mine shafts and adits included C. townsendii,
M. ciliolahnnn, and M. volans. C. townsendii
was found using mines dining both winter and
summer Mijotis species were found only in
summer Pat Brown (personal communication)
recently docimiented a maternity colony of
Antrozous pallidus in an abandoned mine shaft
in northern Lander Count)' as well.
Climatological data from Elko in the north-
eastern part of the state, Ely in the east central,
and Las Vegas in the south were compared to
1996]
Bats from Northeastern Nevada
51
Table 2. Number of bats examined, percent freciuency by species, and nnmber of specimens collected and preserved
from eastern Nevada (1980-1994).
Bat species
Number of bats
% frequency
Specimens
collected
Mijotis ciliolabnnn
73
Mijotis evotis
112
Mijotis Iticifiif^u,s
6
Mijotis volans
186
Mijotis spp.
16
Lasiiiriis cinereus
3
Lasioni/cteris noctivagans
39
Eptesicus fiiscits
52
Pipistrellus Hesperus
2
Corynorhinm townsendii
69
Antrozoiis paUidiis
15
Tudarida brasiliensis'^
5
TO'IAL
578
13.0
19.0
0.4
32.0
3.0
0.1
7.0
10.0
0.1
12.0
3.0
0.4
100.0
2
3
1
3
2
1
4
2
0
1
1
2
22
''Roost cavern not included in calculations.
bat species richness from each of these
regions (Hall 1946, Durrant 1952). Pearson's
3i and Spearman's Rho tests were used to test
for correlations. Bat species richness exhibited
no correlation with the following climatologi-
cal data: annual normal precipitation, January
minimum temperatures, July minimum tem-
peratures, and July maximum temperatures.
There were weak correlations between bat
species richness and January maximum tem-
peratures (Pearson's % 0.728, P < 0.05) and
mean annual days with 0° C or lower (Pear-
son's % -0.704, P < 0.05).
Bat records were pooled by mountain ranges
with similar rock types — sedimentaiy, igneous,
or metamoiphic. Bat species richness was high-
est in portions of northeastern Nevada typified
by sedimentary rock (limestone, dolomite).
Igneous mountain ranges (basalt, volcanic ash)
generally had moderate bat species richness,
and metamorphic mountain ranges (quartzite)
typically had low bat species richness.
Several bat localities from eastern Nevada
represent notable range extensions. Four locali-
ties (10, 14, 15 and 25, Appendix 1) for A. pal-
lidus extend its range from central Nye County
(Hall 1946) north to the Nevada and Idaho
border, approximately 450 km. Two specimens
of T. hrasiliensis at Swallow Canyon (locality
10, Appendix 1), the recent confirmation of a
large roost colony, and the two specimens from
Elko (locality 29, Appendix 1) represent the
first records of this species for Elko and White
Pine counties (Hall 1946) and extend its range
approximately 350 km north. The capture of
single specimens of P. hesperus at Swallow
Canyon (locality 10, Appendix 1) and in Elko
(locality 29, Appendix 1) also suggest a north-
ern range extension and, based on spring and
late-summer capture dates, may represent mi-
grating individuals.
Although certain bat species have long been
suspected of occuning in this region (Hall 1946,
Durrant 1952, Kunz 1982, Kunz and Martin
1982), the localities listed in Appendix 1 rep-
resent the first range confirmations for L. noc-
tivagans, L. cinereus, and C. townsendii in east-
em and northeastern Nevada.
On examination of contributing abiotic fac-
tors such as geological features, precipitation,
and average temperatm^es, one can see patterns
in eastern Nevada's bat fauna beginning to
emerge. The greatest diversity of bat species
from eastern Nevada was recorded in east
central Nevada. The lower maximum January
temperatures and more annual days below 0°
C in east central Nevada contradicted the cor-
relations in our data and suggested that factors
other than climate were contributing to zoo-
geographical patterns. East central Nevada's
mountain ranges are primarily sedimentary in
nature and provide abundant caves, cliff sites,
and high-elevation forests for roosting and
hibernation. In northeastern Nevada most of
the mountain ranges are igneous or metamor-
phic in structure, thus reducing the number of
potential roost sites for bats. Climatic factors
undoubtedly play a large role in defining bat
52
Great Basin Naturalist
[Volume 56
distribution. However, the density of suitable
roost sites may prove to be an even greater
influence on bat distribution where roost site
availability becomes a limiting factor Inasmuch
as most bat species probably do not migrate
more than 1500 km from maternity roosts to
hibernacula (Hill and Smith 1992), an abun-
dance of suitable hibernation roosts would
probably provide any given bat fauna the best
chance of survival in an area where severe
winters are commonplace.
Manning, R. W., and J. K. Jones, Jr. 1989. Myotis evotis.
Mammalian Species 329; 1-5.
Thompson, R. S., and J. I. Mead. 1982. Late Quaternary
environments and biogeography in the Great Basin.
Quaternary Research 17: 39-55.
Warner, R. M., and N. J. Czaplewski. 1984. Myotis
volans. Mammahan Species 224: 1^.
Wells, R V. 1983. Paleobiogeography of montane islands
in the Great Basin since the last glaciopluvial. Eco-
logical Monographs 53: 341-382.
Received 24 March 1995
Accepted 15 August 1995
Acknowledgments
We wish to thank the numerous people who
accompanied us in the field, especially our
families, Lois and Susan and the boys, Roger,
Mark, Boden, and Jedediah. Thanks also to
Dr Scott Altenbach (University of New Mex-
ico); Dr. Mike O'Fanell (O'Farrell Wildlife
Consulting); Lariy Hyslop and Len Seymour
(Great Basin College); Linda White-Trifaro
and Mitchell White (USFS); Cristi Baldino
and Vidal Davila (Great Basin National Park);
Curt Baughman, Lany Gilbertson, Sara Gran-
tham, Gary Herron, Rory Lamp, and Tyler
Turnipseed (Nevada Division of Wildlife); and
the Northeastern Nevada Naturalists.
Literature Cited
Brown, J. H. 1971. Mammals on mountaintops: nonequi-
lihrium insular biogeography. American Naturalist
105: 467-478.
DURRANT, S. 1952. Mammals of Utah: taxonomy and dis-
tribution. University of Kansas, Museum of Natural
Histoiy Publication 6: 1-159.
Grayson, D. K. 1987. The biogeographic histoiy of small
mammals in the Great Basin: obsen'ations on the last
20,000 years. Journal of Manmialogy 68: 359-375.
Hall, E. R. 1946. The mammals of Nevada. University of
Galifornia Press, Berkeley. 710 pp.
Hermanson, J. W, and T. J. O'Shea. 1983. Antrozoiis pal-
lidus. Mammalian Species 213: 1-8.
Hill, J. E., and J. D. Smith. 1992. Bats: a natural histoiy
University of Te.xas Press, Austin. 243 pp.
HoLMCREN, N. H. 1972. Plant geography of the Inter-
mountain Region. Pages 77-161 in A. Cronquist, N. H.
Holmgren, and J. L. Reveal editors, Intermountain
flora. Volume 1. Hafner Publishing Go., New York.
KuNZ, T. H. 1982. Lcmonycteris noctiixigens. Mannnalian
Species 172: 1-50.
Kunz, T. H., and Allen Kurta. 1990. Capture methods
and holding devices. Pages 1-29 in T. H. Kunz, edi-
tor. Ecological and behavioral methods for the stud\'
of bats. Smithsonian institution Press, Washington,
DG. 533 pp.
Kunz, T. H., and R. A. Martin. 1982. Plecotus tnivnsendii.
Mammalian Species 175: 1-6.
Larrlson, E. J., and D. R. Johnson. 1981. Manmuils of
Idaho. University of Idaho Press, Moscow. 166 pp.
Appendix 1
Bat Survey Localities and
Animals E.xamined
1. Stump Greek, 8.2 mi S and 7.6 mi W of Northfork,
Independence Mountains, Elko Go., Nevada. T40N,
R53E, SWl/4 sec 12. 2325 m. 17 Julv 1980, Mijotis evotis
(1), M. volans (2).
2. Sheep Greek, 8.5 mi S and 7.8 mi W of Northfork,
Independence Mountains, Elko Go., Nevada. T40N,
R53E, NWl/4 sec 13. 2320 m. 6-7 August 1980, Myotis
volans (1 lactating female), M. ciliolabnnn (1 male).
3. Jim Greek, 10.4 mi S and 7.2 mi W of Northfork,
Independence Mountains, Elko Co., Nevada. T40N,
R53E, NEl/4 sec 25. 2155 m. 15 July 1981, Myotis evotis
(2 nonscrotal males).
4. Jarbidge River, 5.5 mi S and 1.2 mi E of Jarbidge,
Jarbidge Mountains, Elko Co., Nevada. T45N, R58E,
SEl/4 sec 10. 2460 m. 26 July 1981, Myotis evotis (1).
5. Northfork of tlie Humboldt River, 12.4 mi S and 2.5
mi E of Northfork, Elko Co., Nevada. T.39N, R55E, center
sec 3. 1850 m. 7 Sept. 1981, Corynorhinus townsendii (1);
30 August 1989, Myotis hicifugus (1).
6. Mouth of Cave Creek, Ruby Lake National Wildlife
Refuge, east slope of the Ruby Mountains, Elko Co.,
Nevada. T27N, R57E, SWl/4 sec 24. 1850 m. 25 July
1986, Myotis volans (2), M. evotis (1): 15 June 1987,
Myotis evotis (1), M. ciliolabnim (1).
7. Ferguson Springs, 1/4 mi W of Ferguson Station on
St. Hwy 93, Elko Co^, Nevada. T30N, R69E, NEl/4 sec
33. 187.5 111. 17 Sept. 1989, Myotis volans (1).
8. Arizona Springs, southeast end of the East Hum-
boldt Range, Elko Co., Nevada. T33N, R61E, SWl/4 sec
20. 2050 m. 21 June 1991, Myotis evotis (9 males, 18 lac-
tating females), M. ciliolahrwn (3 males).
9. Sidehill Spring, 6.4 mi S and 11.8 mi W of Wend-
over, Goshute Mountains, Elko Co., Nevada. T32N, R68E,
SWl/4 sec 14. 22.55 m. 7 June 1991, Myotis evotis (6 males,
2 lactating females), M. volans (4 males, 2 lactating females),
.\/. ciliolahruni (1), Corynorhinus townsendii, 1 male.
10. Swallow Canyon, spring site at the mouth of the
canyon. Snake Range, White Pine Co., Nevada. TllN,
i^68E, sec 5. 2100 iii. 21 August 1991, A/(/()^/,s' ciliolahrwn
(1), Lasionycteris noctivagans (1 male, 2 females), Lasiurus
cinerius (\ male), Tadarida hrasiliensis (2 males), Antro-
zous pallidas (1 lactating female); .30 August 1991, Myotis
volans (1 male), Lasionycteris noctivagans (19 males), Pip-
istrelhis hesperiis (1 male), Eptesiciis fiisciis (2 males), Cory-
norhinus tounscn(hi (1 male); 22 August 1994, Myotis
1996]
Bats from Northeastern Nevada
53
Lolaus (8), M. evotis (1), M. ciliolahrwu (11), Coryiwrliiiiiis
townscndii (1), Lasiomictens noctiv(i;i:,(ut.s (2), Ei)tc'sicits
fiiscus (1).
11. Headwaters of" McCall Cheek, Bull Run Moun-
tains, Elko Co., Nevada. T45N, R52E, middle sec 23.
2420 m. 6 July 1991, Mijotis volans (2), M. evotis (1), M.
ciliolahrum (1), Lasiomjcteris noctivagans (2).
12. Man's River, 6.5 mi S and 2 mi W of Maiy's River
Peak, Jarbidge Mountains, Elko Co., Nevada. T44N,
R58E, SWl/4 sec 35. 2220 m. 30 July 1990, Mtjotis evotis
(2 males), M. ciliolahnnn (1 lactating female), Eptesicus
fiiscus (1 lactating female); 31 July 1990, Mijotis lucifugus
(2 males), M. evotis (1 lactating female), M. volans (1 male,
2 lactating females, 5 nonlactating females), M. ciliolahrum
(2 females), Eptesicus fuscus (1 male, 1 female), Lasiomjc-
teris noctivagans (2 males); 1 August 1990, Mijotis volans
(2 males, 3 lactating females), M. evotis (1 lactating female),
Eptesicus fuscus (2 males), Lasiomjcteris noctivagans (2
males).
13. Complex of mine shafts in Snowstorm Mountains,
canyon 1.5 mi N of Midas, Elko Co., Nevada. T39N, R46E,
NVVl/4 sec 16. 1950 m. 31 May 1992, Conjnorhinus town-
scndii (3).
14. Salmon Falls Creek, 1.6 mi W of Jackpot, Elko Co.,
Nevada. T47N, R64E, center sec 10. 1500 m. 23 May 1992,
Mtjotis ciliolahrum (1); 24 June 1992, Mijotis ciliolahrum
(1), Antrozous pallidus (3 males, 1 lactating female), Coryno-
rhimts townscndii (1 lactating female).
15. Goshute Cave^ Cheny Creek Range, White Pine
Co., Nevada. T25N, R63E. 20 June 1992, Mijotis evotis
(2), M. lucifugits (1), Conjnorhinus townscndii (3), Antro-
zous pallidus (3); 16 August 1992, Mijotis evotis (2 males, 4
lactating females), M. volans (2 scrotal males), Mijotis spp.
(either ciliolahrum or californiciis) (1 scrotal male), Antro-
zous pallidus (3 scrotal males, 1 nonscrotal male), Conjno-
rhinus townsendii (5 scrotal males).
16. Bruneau River, junction of Cottonwood Creek and
the Bruneau, Elko Co., Nevada. 1725 m. T45N, R57E,
NWl/4 sec 20. 7 June 1992, Mijotis evotis (1), M. lucifugus
(1 pregnant female); 22 July 1992, Mijotis evotis (1).
17. Mill Creek, 1.6 miN and 2.4 mi W of Jack Creek
Campground, Independence Range, Elko Co., Nevada.
T42N, R53E, SWl/4 sec 16. 2620 m. 15 July 1992, Mijotis
evotis (1), M. ciliolahrum (5 males, 6 lactating females), M.
volans (1), M. lucifugus (1), Eptesicus fuscus (3 males, 4
lactating females), Lasiomjcteris noctivagans (3), Lasiiirus
cinereus (1).
18. Water Canyon and Buck Springs, southwest slope
of the Ruby Mountains, White Pine Co., Nevada. T25N,
R56E, NW'l/4 sec 1 and T26N, R56E, center of sec 35,
respectively. 2300 m. 6 July 1992, Mijotis evotis (3 scrotal
males, 2 females), M. volans (2 scrotal males, 3 females),
M. californiciis (1 scrotal male); 22 July 1993, Mijotis evo-
tis (17), M. volans (2), M. californicus (2).
19. Middlefork of Doby George Creek, 1.2 mi S of
Maggie Creek Summit, Bull Run Mountains, Elko Co.,
Nevada. 2050 m. 27 July 1992, Mijotis evotis (4), M. volans
(1).
•^Because of the sensitivity of natural caves, location descriptions are li;
ited to township and range information.
20. Horse Creek, 5.2 mi W and 0.4 mi N of Secret
Pass, East Humboldt Range, Elko Co., Nevada. T34N,
R61E, NEl/4 sec 16. 2520 m. 4 August 1993, Myotis cilio-
lahrum (7), Lasiurus cinereus (1).
21. USES campground on Northfork of Berry Creek,
Schell Creek Range, White Pine Co., Nevada. T17N,
R65E, SEl/4 sec 10. 2550 m. 9 July 1993, Myotis evotis (3
lactating females).
22. Worthington Canyon, Schell Creek Range, White
Pine Co., Nevada. T17N,'R65E, center sec 16. 2550 m. 10
July 1993, Myotis evotis (3).
23. Currant Creek, USES campgrounds, 1.8 mi E and
0.8 mi S of Currant Mountain, White Pine Co., Nevada.
2650 m. 11 July 1993, Eptesicus fuscus (1), Lasiomjcterus
noctivagans (1).
24. Old Man's Cave, North Snake Range, White Pine
Co., Nevada. T15N, R70E. 16 August 1993, Corijnorhinus
townscndii (4 scrotal males, 1 nonscrotal male, 4 lactating
females, 4 nonlactating females), Myotis volans (1 scrotal
male, 5 nonscrotal males, 9 females), Myotis spp. (2 males,
1 lactating female, 1 nonlactating female); 7 September
1994, Corijnorhinus townsendii (7 males, 17 femtiles), Myotis
volans (2 females).
25. Snake Creek Cave, Snake Creek, South Snake
Range, White Pine Co., Nevada. T12N, R70E. 17 August
1993, Myotis ciliolahrum (1 scrotal male, 3 females), M.
californiciis (1 lactating female), M. evotis (1 male, 1 female),
M. volans (1 female), Corijnorhinus townsendii (1 scrotal
male), Antrozous pallidus (3 scrotal males).
26. Pescio Cave, Schell Creek Range, White Pine Co.,
Nevada. T19N, R64E. 18 August 1993, Myotis ciliolahrum
(2 scrotal males), M. californicus (1 scrotal male, 2
females), Eptesicus fuscus (1 scrotal male), Conjnorhinus
townsendii (1 scrotal male, 1 lactating female).
27. Mine shafts near Emigrant Canyon, Edna Moun-
tain, Humboldt Co., Nevada. T36N, R40E, sec 36. 1400
m. 28 Sept. 1993, Myotis volans (1), M. ciliolahrum (2),
Corijnorhinus townsendii (3).
28. North Fork Little Humboldt River, 3.5 mi S and 9
mi E of Table Mountain, Santa Rosa Range, Humboldt
Co., Nevada. T44N, R41E, sec 1. 2270 m. 10 August 1991,
Lasiomjcterus noctivagans (1).
29. Elko, town center, Elko Co., Nevada. T34N, R55E,
center sec 15. 22 Sept. 1992 and 23 Sept. 1991, Lasiomjc-
terus noctivagans (1); 15 May 1992, Pipistrellus hesperus
(1); 19 Aug. 1991, Myotis ciliolahrum (1); 15 July 1992,
Eptesicus fuscus maternity roost; 15 Nov. 1994 and 6 Jan.
1995, Tadarida hrasiliensis (2).
30. Mine shaft near Contact, Elko Co., Nevada. T45N,
R64E, sec 19. 1800 m. 21 Dec. 1993, Conjnorhinus town-
sendii (3 hibernating).
31. Cave in Spring Valley, White Pine Co., Nevada.
T15N, R68E. 2300 m. 27 July 1994, Tadarida hrasiliensis
roost (54,000-82,000).
32. Muiphy Wash, South Snake Range, White Pine Co.,
Nevada. TION, R68E, sec 2. 2250 m. 29 July 1994, Cory-
norhinus townsendii (3), Lasiomjcterus noctivagans (1),
Myotis evotis (11), M. volans (42), Myotis spp. (2), Eptesi-
cus fuscus (1), Tadarida hrasiliensis (1); 21 Sept. 1994,
Myotis volans (35), M. evotis (4), M. ciliolahrum (1).
33. Rock Creek, Sheep Creek Range, Eureka Co.,
Nevada. T34N, R48E, sec 8. 1450 m. 21 May 1994, Myotis
ciliolahrum (10), M. evotis (1).
Great Basin Naturalist 56(1), © 1996, pp. 54-58
NUPTIAL, PRE-, AND POSTNUPTIAL ACTIVITY OF THE THATCHING ANT
FORMICA OBSCURIPES FOREL, IN COLORADO
John R. Conway^
Abstract. — Obsei-vations and excavations of thatching ant nests from 1990 to 1994 at 2560 m in Colorado provided
infomiation on the numbers and behavior of males and winged and wingless queens. Nuptial activit)' was compared to
that reported by other investigators at lower altitudes. Reprodiictives were obsewed from 24 June to 15 August. Activity
was greatest in 1993 when reprodiictives were on 10 of 98 mounds in the area. Mating and swarming occuned on rab-
bitbmsh 4 m from 1 nest 2-6 July. The number of wingless queens in 4 excavated nests varied fioni 0 to 198.
Key words: nuptial flight, Formica obscinipes, Colorado, thatching ant.
Information on the reproductive activity of
the thatching ant, Formica ohscuripes Forel, in
Colorado is sparse (Gregg 1963). The puipose
of this study is to help remedy the deficiency
and to compare nuptial and pre- and postnup-
tial activity of the thatching ant at high alti-
tude in Colorado with similar studies on this
species at lower elevations in North Dakota
(McCook 1884, Weber 1935, Kannowski 1963,
Wheeler and Wheeler 1963), Michigan (Talbot
1959, 1972), Illinois (Herbers 1978, 1979),
Idaho (Cole 1932), and Nevada (Clark and
Comanor 1972). The Nevada site north of Reno
at 1550 m most closely approximates the Colo-
rado study area in elevation and vegetation.
Mating flight plays a major role in the
reproduction and dispersal of most social in-
sects (Holldobler and Wilson 1990). Males and
queens of F. ohscuripes fly to "swarming
grounds" as reported by Talbot (1972). There
males lly back and forth in search of queens,
which alight on low vegetation and release
pheromones to atti'act males (Cheilx et al. 1993).
Materials and Methods
The main Colorado study area (64.6 X 114
m) has 85 mounds and is dominated by big
sagebrush {Artemisia tridentata Nuttall). It is
adjacent to a quaking aspen grove {Popuhis
trermdoides Michau.x) at an elevation of about
2560 m. The site is located in Gunnison C^ount)'
north of Blue Mesa Resenoir and west of Soap
Creek road. Other plants in the study area are
Chrysothamnus nauseosus (Pallas) Britton (rub-
ber rabbitbrush), Purshia tridentata (Pursh) de
Candolle (antelope bitterbiTish), Lupimis argen-
teus Pursh (silvery lupine), SympJioricarpos
rotiimlifolius A. Gray (mountain snowberry),
Rosa woodsii Lindley (Woods rose), Urtica
gracilis Alton (stinging nettle), Penstemon
strictus Bentham (Mancos penstemon), Ipo-
mopsis aggregata (Pursh) Grant ssp. aggregata
(trumpet gilia), 1 Saskatoon serviceberry tree
{Amelanchier alnifolia var pumila), and 1 Doug-
las-fir {Pseiidotsiiga sp.). Observations in this
area took place on 5-6 August 1990; 20-28
June, 22-27 July 13-15 August, 12-13 Sep-
tember, and 11 October 1992; 28 June-16
August 1993; and 29 June-31 July and 14-16
August 1994. Observations before 20 June
were not possible due to academic commit-
ments. A nest was excavated on each of the
following dates: 6 August 1990, 27-28 June
1992, 12-14 July 1993, and 11-25 July 1994.
The 1993 mound was poisoned with 1 1/2
cups Hi-Yield ant killer granules (Diazinon)
wetted down with about 2 gal of water prior to
excavation.
Results and Discussion
Reproductives
Reproductives (males, winged and wingless
queens) were observed in Colorado from 24
June to 15 August over 3 summers. Activity
was greatest in 1993 when reproductives were
found on 10 mounds scattered among 98 nests
in the area: males, winged queens, and wing-
less queens on 5 mounds; males and winged
'Department of Biology, Univcrsit\ of Scranton, Scranton, P.V 18.510.
54
1996]
Thatchinc Ant in Colorado
55
queens on 3 mounds; a winged queen on 1
mound; and a wingless queen on 1 mound.
Observations of both male and female alates
on Colorado mounds support Herbers s (1978)
observations that some nests produce a mix-
ture of sexes. We were unable to confirm
reports that some nests produce all males or
all females (Kannowski 1963, Herbers 1978),
or that a changeover from early all-male flights
to later all-female ones occurs (Talbot 1959,
1972, Clark and Comanor 1972).
Males. — Males were observed on 8
mounds from 28 June to 13 July 1993 and at 1
mound on 5-6 July 1994. Males seemed to
prefer the shady side of 1 mound built around
a fencepost. Workers sometimes chased males
and once one carried a male on a mound. Oth-
ers have reported males earlier in the year.
Talbot (1959, 1972) saw males flying 16-24
June, and Clark and Comanor (1972) saw
males from 15 April to 4 May.
Although males were observed from 0740
to 1635 hours in Colorado, they were most
numerous and flew from 0938 to 1101. Talbot
(1959) saw them fly even earlier, between
0608 and 1000. Clark and Comanor (1972) also
saw morning flights, but noted males through-
out the day (0840 to 1445).
The largest number of males on 1 Colorado
mound was 10 on 3 July 1993, about the same
maximum per mound (12) reported by Clark
and Comanor (1972). Herbers (1979) noted up
to 1264 males. Talbot (1959, 1972) reported
even more males (up to 4500) but noted that
the ratio of males to females varies from
colony to colony and from flight to flight.
One male was found in a Colorado nest exca-
vated in July 1993; none were in 3 other exca-
vated nests. Wheeler and Wheeler (1963) re-
ported males in nests from 23 May to 12 July.
Winged queens. — Winged queens were
observed on 9 Colorado mounds from 28 June
to 16 July 1993, and one was on a mound on 5
July and 10 July 1994. Workers pulled queens
by their wings and antennae on mounds and
were in turn sometimes dragged by queens.
Queens were noted with tattered, spread, and
partial wings from 30 June to 6 July. Others
reported winged queens at nests earlier and
later in the season than in Colorado. Clark and
Comanor (1972) saw them as early as 1 May,
and Wheeler and Wheeler (1963) reported
winged females in nests as late as 8 August.
Winged queens were observed from 0654
to 1640 hours in Colorado, but most often in
the morning. Clark and Comanor (1972) also
saw them throughout the day, from 0830 to
1720. Those found later in the day were pre-
sumably remnants of the morning activity.
The maximum number of winged queens
on 1 Colorado mound was about 50 on 3 July
1993. Odiers reported greater numbers per nest:
78 (Clark and Comanor 1972) and 230 (Talbot
1959). Winged queens were more abundant
than males on Colorado mounds as reported
by Clark and Comanor (1972), except on 1
occasion when males were more numerous.
No winged queens were found in 4 excavated
Colorado nests.
Wingless queens. — Dealation was not ob-
sei'ved in Colorado, but wingless queens were
seen on 6 mounds and on trails from 24 June
to 15 August between 0757 and 1742 hours.
The greatest number on 1 mound was 7. Wing-
less queens were usually sunounded by a group
of workers on the mounds who often pulled
them by their antennae and legs and some-
times lunged at queens as if attacking them.
Some were carried on the trails by workers.
Dead wingless queens were observed being
carried on a mound and a nearby dirt road.
The number of wingless queens in 4 nests
excavated in Colorado varied greatly: 0, 1, 32,
and 198. Five of the 198 queens from 1 nest
were found with numerous workers amid a
clump of rabbitbrush roots 1.5 m away from
the excavated moimd. Workers probably moved
the queens along a trail from the main nest to
a secondaiy nest at the rabbitbrush for safety
during the prolonged excavation.
Kannowski (1963) stated that many species
of Formica have more than 1 dealate queen
per colony, and Cole (1932) reported 2 or more
per F. obsciiripes nest. The significance of the
highly varible number of dealated queens per
Colorado nest is unclear, and more excavations
are necessary to determine the normal state of
affairs. Observations of wingless queens on
trails suggest that they may be transferred be-
tween mounds or adopted by existing colonies
after the nuptial flight (Weber 1935).
Flight Season and Period
The time of year during which alates of a
species in a given area fly is termed the flight
season. Kannowski (1963) noted that species
such as F. obscuripes, with a large geographical
56
Great Basin Naturalist
[Volume 56
distribution, may have a very long flight sea-
son over their range. In Colorado, queens flew
1-8 July and males 1-9 July. Although others
noted flights as early as 1 May (Clark and
Comanor 1972) and as late as September
(McCook 1884), flights were more common in
June and July (Cole 1932, Weber 1935, Talbot
1972). Talbot (1972) noted that the flight sea-
son varies greatly fi-om colony to colony in any
year and that colonies may have 5-16 flights.
Interestingly, she found that colonies in shel-
tered nests or those on west-facing slopes flew
later than those on open east slopes.
Each ant species has a flight period — the
time of day that flights take place. Kannowski
(1959) reported that most species of Formica
have early morning flights. Queens flew be-
tween 0950 and 1141, and males between
0938 and 1101 in Colorado. Colorado flights
did not begin as early (0500) or end as early
(0750) as some reported by Talbot (1959) in
Michigan, perhaps due to colder temperatures
at high altitude in the morning. Reproductive
activity subsided at Colorado nests between
1040 and 1107, or approximately at the same
times (1030-1145) reported by T^ilbot (1972).
Emergence and Positioning
Reproductive emergence and positioning be-
havior in Colorado is similar to that reported
by Kannowski (1963) and Weber (1935). Mates
emerged, walked around, and went back into
the entrances before leaving the mound and
climbing nearby structures. Workers some-
times chased emerging alates or held onto their
wings; at other times they seemed to ignore
the sexuals. Males ignore winged queens at
this time. Winged queens left Colorado mounds
1-8 July 1993 between 0818 and 1145 hours.
Winged queens and males were found on the
ground as far away as 7.85 m and 5.28 m from
the mounds, respectively.
Reproductives often climb prior to flight.
In Colorado they climbed nearby sagebrush,
rabbitbrush, lupine, and grass, as well as dead
sagebrush and a fencepost protruding from
mounds. At the most active mound thev climbed
3 sagebrushes, 0.48-0.89 m high, and 0.91-2.57
m away. Others have reported alates on nearby
sagebmsh and rabbitbmsh (Clark and Comanoi-
1972), grass and herbs (Weber 1935), and tim-
othy and bluegrass (TlUbot 1959).
Although a number of Colorado reproduc-
tives flew from their perches, many did not.
Some queens descended 1-6 min after arrival,
and one was pulled down by workers. Kan-
nowski (1963) saw some alates wait longer
(10-30 min) before flying from their perches.
Tapping and blowing on perched queens did
not induce them to fly.
A correlation between temperature and
emergence and positioning was noted by Tal-
bot (1972). She reported that alates began
leaving mounds when the air temperature
reached 17.2° C and began climbing plants at
temperatures above 18.3° C.
Flights
In Colorado alates flew from grass, sage-
brush, rabbitbrush, and lupine; a few took off
from the ground. Prior to flying, some queens
released their front legs and fanned their
wings, as reported by Kannowski (1963). On
the other hand, Talbot (1959) reported that
queens flew quickly with little preliminary
wing fluttering.
One Colorado queen flew east at least 13.1
m at an estimated altitude of 4 m. Another
flight lasted about 20 sec at an estimated alti-
tude of 9 m. Other winged queens moved away
from mounds by alternately walking on the
ground and making short, low flights between
plants. One queen using this method moved
7.85 m away from a mound over a period of 37
min. Most queen flights were low and down-
hill to the east. Males generally had short (2.5
cm-1.5 m), flitting or hovering flights about
a meter above the ground, sometimes reland-
ing on the same vegetation from which they
departed.
Reproductive activity was greatest in Colo-
rado on clear, warm, windless days. All investi-
gators agree that these are the most favorable
conditions for flight. Wind supressed repro-
ductive activity at 0918 hours on 3 Jul\- 1993.
Weber (1935) noted alates leaving the nest
when the air temperature was above 15.5° C,
humidity exceeded 50%, and the sk-)- was clear.
Others reported first flights at an air tempera-
ture at least 5 ° C higher. A Colorado male flew
at 22.7° C. Talbot (1972) reported that alates
flew at temperatures between 20.5° C and
27.2° C, and Clark and Comanor (1972) saw
flights between 20.5° C and 26.5° C, but at a
relative humidity of only about 18%. Talbot
(1959, 1972) noted that wind gusts, rain, low
temperatures, and dark skies stopped flights,
and wet grass and gray skies delayed flying.
1996]
Thatching Ant in Coloiuuo
57
Colorado flights involved relatively few re-
prodiictives, but reports in the literature vary
considerably. Weber (1935) believed there is
no marriage flight because only 1 sexual or a
few sexuals fly at a time. Kannowski (1963)
saw 1 mass flight, but noted most flights were
sparse or moderate. Talbot (1959), on the other
hand, reported that 695 females and an esti-
mated 4500 males flew over time. Rates of fly-
ing of 4-14 queens/min and 1-10 males/min
have been reported (Talbot 1959, Clark and
Comanor 1972).
There appears to be no agreement on the
flight pattern. Talbot (1959) noted that most
queens flew downhill and westward, but some
had short, sporadic flights from plant to plant
or to the ground as sometimes observed in
Colorado. Colorado flights were generally at
low altitude (estimate 4-9 m), downhill, and
eastward toward the sun. Kannowski (1963)
also noted that alates fly in the general direc-
tion of greatest light intensity. Others report
that flights are often upward and out of view
(12 m or more; Weber 1935, Kannowski 1963,
Clark and Comanor 1972).
Swarming and Mating
Swarming is the process whereby alates
aggregate to mate in the air or on the ground
and vegetation (Kannowski 1963). Most swann-
ing and mating in Colorado occurred 2-6 July
1993 bet^veen 1008 and 1125 hours on rabbit-
brush 4.01 m from 1 mound. Mating was also
obsei^ved on rabbitbiaish beside another mound
on 2 July and 6 July 1993. Talbot (1972) noted
swarming earlier in the year and over a longer
time period, namely, 4-17 June between 0700
and 1200.
Swarming in Colorado was similar to that
described by Kannowski and Johnson (1969)
and Talbot (1972). Queens anived first on rab-
bitbrush, followed by males. Queens perched
on the upper parts of plants often with their
heads down and their abdomens pointing
upward or toward the nest. Presumably they
emit a pheromone to attract males (Kannowski
and Johnson 1969, Walter et al. 1993). Once
the female's pheromone is detected, males fly
upwind to the general location of the female,
fly quickly from stem to stem until they find
her, alight, and then attempt to mate (Kan-
nowski 1963). After mating, males usually fly
off while the queen remains and sometimes
inspects her abdomen.
Up to 7 in copulo alates were noted at 1
time at the Colorado swarming site 4.01 m
away, 6 pairs on rabbitbrush and 1 pair on an
adjacent lupine. Some pairs fell off the plants.
One queen appeared to mate 2 or 3 times.
Kannowski (1963) reported a queen mating 4
times. Two Colorado males tried to simultane-
ously mate with a queen for 1 min 40 sec and
remained attached to each other for 20 sec
after the queen left. Talbot (1972) noted 3 or 4
males tiying to mate a queen, and Kannowski
(1963) reported a single male may mate sev-
eral times before flying away.
The durations of 6 Colorado matings ranged
from 1 min 40 sec to 3 min 40 sec (mean = 2
min 43 sec), or within the 1- to 5-min dura-
tions reported by Talbot (1972).
Talbot (1959, 1972) noted larger, more
diverse, and more heavily populated swarming
areas than the small rabbitbrush area in Colo-
rado. Some of her swarming areas were over
short grass; others were on shrubs. One swami-
ing area involved thousands of males hovering
over hundreds of females from 3 colonies and
covered an oval-shaped area 27.5 X 11 m.
Males usually flew near grass level, but some-
times as high as 1.2-1.5 m. Another swarming
area shifted somewhat from day to day and
increased to approximately 41.3 X 32.1 m.
She found that these areas were maintained
throughout the flying season, and some were
used year after year.
Conclusions
Preliminary studies of the reproductive
behavior of the thatching ant, F. obscuripes, in
Colorado are in general agreement with the
literature. Time constraints on our seasonal
obsei"vations probably explain why we did not
observe reproductive behavior as early in the
year as that reported in the literature. The
most notable finding was the paucity of repro-
ductive activity: swarming and mating were
obsei-ved only 2-6 July 1993; 9 of 98 mounds
(9%) in the area had winged reproductives;
mating occurred near 2 mounds (2%); and a
swaniiing area was found 4.01 m from 1 mound
(1%). The numbers of males and winged queens
were relatively low and the swarming area was
small. Other notable findings were the highly
variable number (0-198) of dealated queens
per nest and the almost complete absence of
winged alates in excavated nests.
58
Great Basin Naturalist
[Volume 56
Further studies are needed to determine
whether our findings are anomahes or whether
they represent the normal state of affairs for
this species at high altitude.
Acknowledgments
I thank 4 University of Scranton students,
John Bridge, Tom Sabalaske, Antliony Musingo,
and Jeanne Rohan, who conducted fieldwork
in Colorado in 1993-94. Support for this re-
search was provided by a grant from the
Howard Hughes Medical Institute through
the Undergraduate Biological Sciences Edu-
cation Program. Barry C. Johnston, ecologist
at the U.S. Forest Service in Gunnison, Colo-
rado, identified plant specimens.
Literature Cited
Cherix, D., et at. 1993. Attraction of the sexes in Formica
htgiibris Zett. Insectes Sociaux 40: 319-324.
Clark, W. H., and R L. Comanor. 1972. Flights of the
western thatching ant, Formica obsciiripes Forel, in
Nevada. Great Basin Naturalist .32: 202-207.
Cole, A. C, Jr. 1932. The thatching ant, Formica
obscuripes Forel. Psyche 39: 30-33.
Gregg, R. E. 1963. The ants of Colorado. University' of
Colorado Press, Boulder 792 pp.
Herbers, J. M. 1978. Trends in sex ratios of the reproduc-
tive broods of Formica obscuripes. Annals of the
Entomological Society of America 71: 791-793.
. 1979. The evolution of sex-ratio strategies in
Hymenopteran societies. American Naturalist 114;
818-8.34.
HOLLDOBLER, B., AND E. O. WlLSON. 1990. The ants. The
Belknap Press of Harxard University Press, Cam-
bridge, MA. 732 pp.
Kannowski, R B. 1963. The flight activities of formicine
ants. Symposia Genetica et Biologica Italica 12:
74-102.
Kannovvskl R B., and R. L. Johnson. 1969. Male patrol-
ling behaviour and sex attraction in ants of the genus
Fonnica. Animal Behaviour 17: 42.5—129.
McCooK, H. C. 1884. The nifous or thatching ant of Dakota
and Colorado. Proceedings of the Academy of Nat-
ural Sciences, Philadelphia, part 1: .57-6.5.
Talbot, M. 1959. Flight activities of two species of ants of
the genus Formica. American Midland Naturalist 61:
124-132.
. 1972. Flights and swarms of the ant Formica
obscuripes Forel. Journal of the Kansas Entomologi-
cal Society 45: 254-258.
Walter, F, et al. 1993. Identification of the sex phero-
mone of an ant, Formica lugubiis. Naturwissenschaften
80: 30-34.
Weber, N. A. 1935. The biology of the thatching ant
Formica obscuripes Forel in North Dakota. Ecological
Monographs 5: 16.5-206.
Wheeler, G. C., and J. Wheeler. 1963. The ants of Nortli
Dakota. University of North Dakota Press, Grand
Forks. 326 pp.
Received 17 Jaituanj 1995
Accepted 21 June 1995
Great Basin Naturalist 56(1), © 1996, pp. 59-72
TRACHYTES KALISZEWSKU, N. SE (ACARI: UROPODINA), FROM THE
GREAT BASIN (UTAH, USA), WITH REMARKS ON THE HABITATS AND
DISTRIBUTION OF THE MEMBERS OF THE GENUS TRACHYTES
Jerzy Bloszyk^ and Pawe4 Szymkowiak^
Abstract. — Trachytes kaliszewshii, n. sp., is described fiom the Great Basin, Utah, USA. SEM photography illustrates
moiphological detail. An annotated list is included of cuirently recognized species of the genus Trachytes, with comments
on their distribution and habitat characteristics.
Key words: mites. Trachytes kaliszewskii, Uropodina, Great Basin, Utah.
Mites of the genus Trachytes Michael, 1894,
are a morphologically distinct entity of the
Uropodina. The genus consists of 31 species
known mainly from the Palearctic region of
Europe and Japan. Wisniewsld and Hirschmann
(1993) mention two species from the USA: T.
aegrota (C. L. Koch, 1841) and T. traegardhi
(Hirschmann and Zirngiebl-Nicol, 1969). Tra-
chytes traegardhi is regarded as nominum
nudum. The USA listing for T. aegrota is con-
sidered either a mistake in determination or
an accidental introduction.
Taxonomic studies on mites of the genus
Trachytes are found in Hirshmann and Zirn-
giebl-Nicol (1969), Hutu (1983), and Pecina
(1970). Information on their biology, ecology,
and zoogeography is found in Athias-Binche
(1978, 1979, 1980, 1981, 1985), Pecina (1980),
Bloszyk (1980, 1982, 1984, 1985, 1990, 1991,
1992, 1993), Bloszyk and Athias-Binche (1985),
Bloszyk and Miko (1990), Bloszyk and Ols-
zanowski (1985a, 1985b, 1985c, 1986), and
Bloszyk et al. (1984).
We found a new species of the genus Tra-
chytes in soil collected from Rock Canyon near
Provo, Utah, USA. It is most similar to those
described by Hiramatsu (1979, 1980) from
Japan: T. aoki and T. onishii. Moiphological dif-
ferences between our species, those mentioned
from Japan, and Trachytes aegrota are shown
in Table 1. Our new species is dedicated to the
Polish acarologist. Dr. Marek Kaliszewski, who
was a faculty member at Brigham Young Uni-
versity/, Provo, Utah, USA, until 1993, when he
died tragically in an automobile accident.
Systematic Status of the Genus
Trachytes Michael
SUPERFAMILY. — Polyaspidoidea sensu Athias-
Binche & Evans, 1981
Family. — Trachytidae Tragardh, 1938
Genus. — Trachytes Michael, 1894
Type species. — Celano aegrota C. L. Koch,
1841 { = Trachynotus pyrifonnis Kramer, 1876)
Mites of middle size, strongly sclerotized,
dorsoventrally flattened. Idiosoma triangular,
"vertex" distinct with smoodi or slightly seiTated
edges. Corniculus simple, laciniae longer than
corniculi. Hypostomatic setae: hi very long,
simple; h2 shorter than hi, simple; h3 very
long, massive; h4 very short, serrated. Fixed
digit of the chelicera longer than moveable
digit, shaiply pointed distally. Base of tritoster-
num wide, not covered by coxae I.
Trachytes kaliszewskii, n. sp.
Diagnosis. — The fonn of the body is typical
for the genus Trachytes Michael. Vertex with
lamella. Dorsal shield with polygonal patteiTi
and irregular cavities in central part (similar to
T aegrota). Marginal shield is not divided as in
European species, without polygonal pattern.
Dorsal setae long and massive. Small pygidial
shield present in female. Epigynial shield tiape-
zoidal with net pattern, front margin slightly
convex and produced laterally into little corns.
Sternal setae short. Operculum of male rounded,
with a pair of long genital setae. Ventroanal
shield separated from sternal and metapodal
shields by a wide zone of interscutal membrane.
'Department of Animal Taxonomy and Ecolog\', Adam Mickiewicz University, Szamarzewskiego 91A, 60-569 Poznari, Poland.
59
60
Great Basin Naturalist
[Volume 55
Ventral setae long. One pair of paranal setae.
Postanal seta present.
Adult female. — Length of idiosoma 900-
907 ^tni, width 535-574 /xni.
Dorsmn: Lamellae with characteristic pat-
tern. Marginal shield not divided posteriorK;
\\ith irregular caxities in posterior part. Dorsal
shield with poKgonal pattern lateralh' and
irregular cavities in central and posterior parts
(Figs. 1, 10, 11). Dorsal setae long and mas-
sive. Two pairs of setae on vertex; no unpaired
medial dorsal setae. Marginal setae on small
scutellae; 4 pairs of setae situated medialK on
marginal shields. Pygidial shield \\ith pattern
as on marginal shield.
Veiitruin: Sternal shield (Fig. 2) fused to
parapodals. Ventroanal shield separated from
steiTial and metapodal shields by a zone of in-
terscutal membrane bearing 4 pairs of platelets
(Fig. 13).
Sternal shield smooth, bearing 5 pairs oi
short stenial setae. Setae: stl situated between
coxae II at the \eye\ of their front margins; st2
and st3 placed abo\ e anterior edge of epig> -
nium; st4 and st5 situated laterally of epig)-
nium. Opisthogastric setae generally long,
simple or delicateh' serrated, most anterior
pair short, similar to sternal setae. First pair of
opisthogastric setae situated below posterior
margin of epig\iiium, 2nd pair on metapodal
shields, with 4 pairs on interscutal membrane
and 2 pairs on ventroanal shield. One pair of
adanal setae; short and serrated. Postanal seta
long. E.xopodal and metapodal shields with
o\'al or irregular cavities. Ventroanal shield
smooth anteriorly, with polygonal patterns in
the posterior regions.
Epigynial shield trapezoidal, with front
margin slightly convex and produced laterally
into little corns; measurements: 175-199 /xm
length and 137-156 fim width (N = 3). Sur-
face of epig>aiium with delicate polygonal net
in anterior and central areas.
Peritrema simple, without poststigmatic
section, extending from the level of the poste-
rior border of the foramen pedale III (with
stigma) to beyond coxae II.
GnatJwsoma: Laciniae (internal mala) longer
than corniculi, serrated. Hypostomatic setae
(Fig. 4) smooth except for setae /i4 which are
delicately serrated; hi very long, /j2 shorter
than hi, /i3 long as hi but more massi\'e, h4
shorter than h2. Three transversal rows of
h\pognathal denticles between setae ^3 and
/i4.
Appendages: Shape of chelicerae typical for
Trachytes; fixed digit of the chelicera longer
than moxeable digit, shaped distalK. Pedipalp
xentral, setae of trochanter (vl, v2) massive
and serrated (Fig. 5).
Shape of legs tvpical for family. Tarsi of legs
II-IV \\'ith 4 long setae (3 times longer than
Table 1. SiininiaiA of major differences between closely related Trachytes species.
Character
T. acp'ota
T. (loki
T. onishii
T. kaliszewskii
Sex
parthenogenic
bisexual
■?
bise.xual
Female
Lamella
transverse
trans\erse
trans\'erse
oblong
Setae on interscutal nienibr
me
absent
present
absent
present
Unpaired mediodorsal seta
present
absent
absent
absent
Bod\- incasmements (in fim
)
600 X 68.5
400 X 450
400 X 600
535-574 X 900-907
H\postomal setae h3
simple
massive
massive
massive
Setae on ventroanal shield
different
ecjual
equal
equal
Epig\-ninm
smootii
smooth
with poKgon
al net
V'ential seta on metapodal s
lields
long
short
short
long
Seta Pa
short
short
short
long
1995]
Tlh\CHYTES KALISZEWSKII, N. SH, I-^HOM UTAH
61
Fig, 1. Trachytes kaliszeicskii. n. sp., dorsal view of female idiosoma.
62
Great Basin Naturalist
[Volume 55
Fig. 2. Trachytes kaliszcwskii, n. sp., \ential view oi icinalf idiosonia.
1995]
TrACHYTI'S hiMJSZEWSKII, N. SH, FKOM UlAll
63
Fig. 3. Trachytes kaliszewskii, n. sp., ventral view of male idiosoma.
64
Great Basin Naturalist
[Volume 55
Figs. 4—5. Trachytes kaliszew.skiL n. sp., female: 4, gnatliosoma, ventral view; 5, ventral setae of palpal trochanter.
others), small claws, and a veiy long distal seta.
Shape of dorsal setae on tarsus, tibia, genu,
and femur of legs I as in the genera Polijaspis
and Polyaspimis. Chaetotaxy of legs I and IV is
shown in detail in Figures 6 and 7.
Sexual dimorphism observed on femora II
(Figs. 8, 9).
Adult male. — Bodv measurements 830-
862 Aim X 538-540 )Ltm.'
Dorsum: Male dorsum slightly changed in
posterior part; pygidial shield absent (Fig. 12).
Sculpture and dorsal chaetotaxy as in the
female.
Ventnim: SteiTial shield with numerous oval
cavities and bearing 5 pairs of short sternal
setae (Fig. 3). Genital operculum rounded
(74-79 X 72 fxm), located a little below coxae
IV, with 1 pair of long genital setae. Opistho-
soma separated b)' transverse suture with in-
terscutal membrane. Seven pairs of long ven-
tral setae on rounded platelets; 1st pair short,
located below operculum. With 1 pair of deli-
cately serrated adanal setae and long unpaired
postanal seta (Pa). Opisthosoma with poly-
gonal sculpture on metapodal and anal shields
and small oval cavities on central portion.
Deutonymph. — Body measurements 624
X 396 lam.
Dorsum: Dorsum with polygonal pattern
(Fig. 15). Podonotal shield trapezoidal, fused
with lamellae. Mesonotal shields large, trian-
gular, with 4 setae. Pygidial shield arched,
with 2 pairs of setae. Dorsal setae strong, mas-
sive. Setae on interscutal membrane and mar-
ginal setae inserted on small platelets.
Ventrum: Ventrum with polygonal pattern
(Fig. 16). Sternal shield elongated, with 5 pairs
of short sternal setae; most posterior pair deli-
cately senated. Opisthogastric setae situated on
interscutal membrane, delicately serrated, sit-
ting on small platelets. Large ventroanal
shield with 2 pairs of short adanal setae (Ad),
postanal seta (Pa) longer than Ad; both setae
serrated.
Protonymph. — Bodv measurement 528 X
295 Aim.
Dorsum: Dorsum with poKgonal pattern
(Fig. 17). Podonotal shield trapezoidal. Meso-
notal shields large, oval-triangular, without
setae. P\gidial shield arched, with 2 strong,
massive setae. Dorsal setae strong, massive.
No setae on intersutal membrane. Marginal
setae numerous, inserted on small platelets.
Ventrum: Sternal shield smooth, elongate,
with 4 pairs of simple sternal setae (Fig. 18).
Four massive, serrated opisthogastric setae
1995]
Trachytes kaliszewsku, n. sf., from Utah
65
Figs. 6-9. Trachytes kaliszewskii, n. sp., legs chaetotaxy: 6, leg I of female; 7, leg IV of female; 8, chaetotaxy of male
femora II; 9, chaetotaxy of female femora II.
situated on intersutal membrane. Large ven-
troanal shield with 1 pair simple adanal setae
and a long postanal seta.
Material examined. — All specimens were
collected from soil under a maple tree in Rock
Canyon near Provo, Utah, 10 September 1992;
leg. J. Bloszyk (holotype and 5 paratype females,
7 paratype males, 7 deutonymphs, 5 proto-
nymphs).
The holotype is deposited in the Canadian
National Collection, Biosystematics Research
Cenbe, Ottawa, Canada. Paratypes are deposited
66
Great Basin Natur.'VLIst
[Volume 55
Figs. 10-14. Trachytes kaliszewskii, n. sp.; 10, dorsal polygonal pattern of feniiile (550X); 11, female, general dorsal
view (llOX); 12, posterior part of male idiosoma (220X); 13, opisthosoma of female (200X); 14, marginal setae of female
(750X).
in the Monte L. Bean Life Science Museum,
Brigham Young Univer.sity, Provo, Utah, USA;
in CSIRO, Canberra, Austraha; and in J.
Bloszyk's collection (Acarological Association,
ul. Lisowsldego, 16/1, 61-606 Poznari, Poland).
List of the Trachytes Species with
Remarks on Distribution and
Habit.at Preferences
Hirshmann (1993) listed 31 species refer-
able to the genus Trachytes. In view of the
1995]
Trachytes kaliszewsku, n. sp., from Utah
67
Fig. 15. Trachytes kaliszewskiu n. sp., dorsal view of deutonymph idiosonia.
above, we recognize 31 species in the genus
Trachytes as follows^:
Trachytes aegrota (C. L. Koch, 1841) is one
of the most numerous Uropodine species in
-Some data from Poland originate from an unpublished investigation
carried out by J. Bloszyk in the thematic program Bank of Invertebrate Fauna;
data on the distribution may be found in Hirschmann (1979, 1993). Hufu
(1973, 1983), Hiramatsu (1979, 1980), and Athias-Binche (1981).
central Europe. This species is parthenogenetic
and nonphoretic; males are rarely found (sex
ratio is 1:10,000). This eurytopic species lives
in all kinds of biotypes, but it prefers forest lit-
ter. It most often occurs below 500 m elevation
but is considered a tychoalpine species (i.e.,
lives in the mountains as well as the lowlands).
In Poland the spring-summer season is the
best time to observe the larva.
68
Great Basin Natuiulist
[Volume 55
Fiu;. 16. Trachiilcs kalisznvskiL n. sp., vcMitral view ol dcutonymiili idiosoina.
Trachytes aoki I liramatsu, 1979. Japan. In
litter.
Trachytes arcuatus Hirschmann and Zirn-
gicbl-Nicol, 1969. Austria, Koniania, H unwary.
Habitat unknown.
Trachytes hah)'clioalpine species. The best time to observe
the lai'va is during the spring-summer season.
Trachytes pecinaia Iluju, 1983. Romania.
In htter.
Trachytes pi Berlese, 1910. West and Cen-
tral Europe. In htter.
Trachytes romanica Huju, 1983. Romania.
In litter.
Trachytes splendkla Huju, 1983. East Car-
pathian species — Romania, Poland, Slovakia.
In litter and moss.
Trachytes stammeri Hirschmann and Zirn-
giebl-Nicol, 1969. Locality and biotype un-
known.
Trachytes tesquorwn Pecina, 1980. Czech
Republic. In grass.
Trachytes traeghardi Hirschmann and Zirn-
giebl-Nicol, 1969. Locality and biotype un-
known.
Trachytes tubifer Berlese, 1914. Italy, Austria.
In litter.
Trachytes welhournia Moraza, 1989. Spain.
In litter.
Trachytes wisniewski Huju, 1983. Romania.
In litter.
Acknowledgments
Dr. J. Bloszyk wishes to thank the adminis-
trators and workers of the Department of
Zoology and Monte L. Bean Life Science
Museum at Brigham Young University (BYU),
Provo, Utah, USA, for providing facilities and
an atmosphere that encouraged scholarship.
The authors are greatly indebted to Dr.
Richard Baumann, Department of Zoology,
BYU, for his kind help in reviewing the manu-
script and for his judicious remarks and advice;
and to Dr. John S. Gardner, electron micros-
copist from BYU, for his valuable scanning
photography. This study was completed with
financial assistance from the Department of
Zoology, Brigham Young University, and
Acarological Association (Poznaii, Poland).
Literature Cited
Athias-Binche, E 1978. Etude quantitative des Uropodes
edaphiques de la hetraie de la Tillaie en foret de
Fontainebleau (Acariens, Anactinotriches). Revue
d'Ecologie et de Biologie du Sol 1.5: 67-88.
. 1979. Effects of some soil features on a uropodide
mite community in the Massane forest (Pyrenees-
Orientales, France). Pages 567-.573 in Recent advances
in acarology — proceedings of the 5th International
Congress on Acarology.
. 1980. Contribution a la connaissance des Uro-
podidcs librcs (Arachnides: Anactinotriches) de
quelques ecosysternes forestiers Europeens. These
d'Etat, Universite de Paris VI, Paris.
1981. Differents types de structures des peuple-
ments d'Uropodides cdaphicjues de trois ecosysternes
forestiers (Arachnides: Anactinotriches). Acta Oeco-
logica-Oecologia Ceneralis 2: 153-169.
BiDSZYK, J. 1980. Mites of the genus Trachytes Michael,
1894 (Acari: Mesostigmata) in Poland. Prace Komisji
Biologicznej, PTPN 54: 5-52.
. 1982. Uropodina Polski (Acari, Mesostigmata).
Thesis, Biblioteka Glowna UAM, Poznan. 543 pp.
. 1984. Altitudinal distribution of the Uropodina
fauna (Acari) in Poland. Przeglad Zoologiczny 28:
69-71.
. 1985. Contribution to knowledge of the mites in
the mole nests {Talpa eiiropea L.). I. Uropodina (Acari,
Mesostigmata). Przeglad Zoologiczny 29: 175-181.
. 1990. Fauna of Uropodina mites (Acari: Meso-
stigmata) of decayed tree stumps and hollows in
Poland. Zeszyty Problemowe Post^pow Nauk Rol-
niczych 373: 217-2.35.
. 1991. State of investigation of Uropodina (Acari:
Anctinotrichida) in Polish National Parks. Parki Nar-
odowe i Rezei-waty Przyrody 10 (1,2): 115-122.
. 1992. Materials to the knowledge of the acaro-
fauna of Roztocze Upland. III. Uropodina (Acari:
Mesostigmata). Fragmenta Faunistica 35 (11):
323-344.
. 1993. Uropodina (Acari: Mesostigmata) of pine
forests in Poland. Fragmenta Flumistica 36 (11):
175-183.
BiDSZYK, J., AND E Athias-Binche. 1985. Urban ecosys-
tems and ecological studies: example of soil uropo-
did community in Poznan Park. Pages 278-282 in
Soil fauna and soil fertility — proceedings of the 9th
International Colloquium on Soil Zoology.
BU5SZYK, J., AND L. MiKO. 1990. Podna fauna Pienin. I.
Uropodina (Acarina: Anactinotiichida). Entomologicke
Problemy 20: 21-47.
BtDSZYK, J., AND Z. Olszanowski. 1985a. Contribution to
the knowledge of mites of birds nests. I. Uropodina
and Nothroidea (Acari: Mesostigmata et Oribatida).
Przeglad Zoologiczny 29: 69-74.
. 1985b. Mites of the genus Trachytes Michael,
1894 (Acari: Mesostigmata) in Poland. III. Sporadic
appearance of males in some populations of partheno-
genetic species. Przeglad Zoologiczny 29: 313-316.
. 1985c. Contribution to the knowledge of biology
of some Uropodina (Acari: Anactinotrichida) juvenile
stages. Przeglad Zoologiczny 29: 487-490.
_. 1986. Contribution to the knowledge of mites of
ant hills in Poland (Acari: Uropodina). Przeglad Zoo-
logiczny .30: 191-196.
BiDSZYK, J., I. ChOJNACKI, AND M. Kaliszewski. 1984.
Study on the mites of the genus Trachytes Michael,
1894. I. Seasonal population changes of Trachytes
aegrota (Koch, 1841) in deciduous resei^ves "Jakubowa"
and "Las Gr^dowy" near Pniewy, Poland. Pages 893-
900 in D. A. Griffiths and C. E. Bowman, editors,
Acarology VI. Volume II. Ellis Hai"wood, Chichester
Hirschmann, W. 1979. Bestinmibare Uropodiden-Arten
der Erde (ca. 1200 Arten), geordnet nach dem Gang-
system Hirschmann, 1979 und nach Adulten Gnippen
(Stadien, Heimatliinder, Synonym, Literatur). Acarolo-
gie (Niiemberg) 26: 15-57.
72
Great Basin Naturalist
[Volume 55
HmscHMANN, W, AND I. ZiRNGlEBL-NicOL. 1969. Gangsys-
tematik der Parasitifonnes Teil 57. Typus der Gattung
Trachytes Michael, 1894. Acarologie (Niirnberg) 12;
76-81.
HiRAMATSU, N. 1979. Gangsystematik der Parasitiformes
Teil 3322. Stadien einer neuen Traclujtes-Art aus
Japan (Uropodini, Uropodinae). Acarologie (Niirn-
berg) 25: 76-77.
. 1980. Gangsystematik der Parasitiformes Teil 360.
Teilgang und Stadien von 2 neuen Trachytes -Arten
aus Japan (Uropodini, Uropodinae). Acarologie (Niirn-
berg) 27; 26-27.
HUXU, M. 1973. Gangsystematik der Parasitifonnes Teil 145.
Zur Kenntnis der Uropodiden-Fauna Rumaniens
Neue Uropodiden-Arten der Gattungen Trachytes
Michael, 1894, Dinychus (Kramer, 1886) und Tra-
chyuropoda (Berlese, 1888). Hirschmann u. Zinigiebl-
Nicol 1961 nov. comb. Acarologie (Niirnberg) 19;
45-51.
. 1983. Gangsystematik der Parasitiformes Teil 428.
Teilgange, Stadien von 6 neuen Trachytes- hxien aus
Rumanien und Schweden (Uropodini, Uropodinae).
Acarologie (Niiniberg) 30; 51-66.
PeCi.na, P 1970. Czechoslovak uropodid mites of the genus
Trachytes Michael, 1894 (Acari, Mesostigmata). Acta
Universitatis Garolinae, Biologica 1969; 39-59.
. 1980. Additional knowledge of members of the
genus Trachytes Michael, 1894 (Acari, Mesostig-
mata) from Czechoslovakia. Acta Universitatis Gar-
olinae, Biologica 1978; 389-407.
WiSmewski, J., a.nd W. Hirschmann. 1993. Gangsystem-
atik der Parasitiformes Teil 548. Katalog der Gang-
gattengen, Untergattungen, Giiippen und Alien der
Uropodiden der Erde. Acarologie (Niirnberg) 40;
1-220.
Received 22 September 1994
Accepted 25 September 1 995
Great Basin Naturalist 56(1), © 1996, pp. 73-84
PRODUCTIVITY, FOOD HABITS, AND ASSOCIATED VARIABLES OF
BARN OWLS UTILIZING NEST BOXES IN NORTH CENTRAL UTAH
Sandra J. Loonian^ Dennis L. Shirley^, and Clayton M. White^
Abstract. — Productivit)' and food habits of the Bam Owl {Tyto alha) utilizing nest bo.xes in Juab, Utah, and Salt Lake
counties, Ut;ili, during 1979-1984 were examined. Average clutch size was 5.8 eggs for the 6-yr period; mean number
fledged was 3.9 yoimg per successfiil nest. While severe weather during the 1981-82 winter did not result in a significant
decrease in productivit)' during the 1982 breeding season, it may have resulted in a significant oveiproduction of female
\oung. BaiTi Owls in north central Utah fed almost exclusively on mammalian species, particularly Microtus spp. Differ-
ences in clutch size between areas and years may be a response to availability as well as abundance of prey.
Key words: Barn Owl food. Barn Owl reproduction, nest boxes, Utah, Tyto alba.
The Barn Owl {Tyto alba) is a nearly cosmo-
politan species that uses diverse nest sites, in-
cluding man-made ones (Voous 1988). Although
Barn Owls were reported in Utah as early as
1899 (Smith and Marti 1976), they were con-
sidered uncommon and rare breeders prior to
1976 (Smith and Marti 1976). The first Barn
Owl nesting record was reported by Behle
(1941) near Kanab in Kane County. Woodbury
et al. (1949) proposed that Barn Owls were
probably residents and widely distributed in
valleys and lower elevations throughout the
state. Smith et al. (1972, 1974) and Smith and
Marti (1976) presented information on Barn
Owl food habits, nesting ecology, and distribu-
tion throughout the state. While these studies
indicated prey was abundant in irrigated agri-
cultural areas, nesting sites were not adequate
in those areas to allow growth of the popula-
tion (Marti et al. 1979).
Marti et al. (1979) installed 8 nest boxes in
abandoned concrete silos in north central Utah
during 1977 and an additional 22 in 1978 in an
effort to increase numbers of nesting Bam Owls.
Of those boxes, 50% were used by breeding
owls in 1977 and 80% in 1978. A total of 154
young fledged fiom nest boxes during the 2 yr
In 1979 a similar program of installing nest
boxes in silos was adopted in central Utah by
the Utah Division of Wildlife Besources
(UDWR). Between 1979 and 1984, 41 nest
boxes were installed in Juab, Utah, and Salt
Lake counties. An ongoing investigation of
Barn Owl population and feeding habits was
undertaken in 1979. Herein we document
reproductive activities, dispersal, sui^vival, and
food habits of Barn Owls utilizing these nest
boxes from 1979 to 1984.
Study Area
This study was conducted on the 15- to 25-
km-wide strip of farmland and suburban area
between the Wasatch Mountains on the east
and Utah Lake on the west. The climate is arid,
characterized by hot, diy summers, cold win-
ters, and cool, wet springs. Precipitation aver-
ages 40 cm annually, falling mainly as winter
snow. Extensive agricultural irrigation and the
presence of a large freshwater lake have cre-
ated broad areas of habitat, especially for voles
{Microtus spp.), a major Barn Owl prey. Trees
occur sporadically along rivers and irrigation
canals and on farmsteads.
Preliminary surveys by UDWR in 1979
revealed that 50 silos were used for roosts by
Barn Owls, as indicated by presence of regur-
gitated pellets, fecal stain, and/or presence of
owls. Silos were in rural or semirural areas
throughout the counties and generally close to
corn or alfalfa fields; a few were located in
suburban areas within 2 km of an agricultural
area (dairy or cattle ranch). Silos not used by
fanners provided roosting owls protection from
Department of Zoolog)-, Brigham Young University, Prove, UT 84602. Present address: Department of Biological Sciences, and Institute of Arctic Biology,
Universit\' of Alaska, Fairbanks, AK 99775.
^Utah Division of Wildlife Resources, Regional Office, Springville, UT S4663.
Department of Zoology, Brigham Young University, Provo, UT 84602.
73
74
Great Basin Naturalist
[Volume 56
predation and disturbance; however, none pro-
vided adequate nest sites. Most bams and other
structures in the area also lacked adequate
nesting sites.
Forty-one wooden nest boxes were built,
after Marti et al. (1979), and installed between
1979 and 1984 (18 installed in 1979, 6 in 1980,
5 in 1981, 9 in 1982, and 1 each in 1983 and
1984). Three nest cavities (2 in silos and 1 in a
school building) were discovered and moni-
tored during these years; data from these sites
are included herein.
Methods
All nest boxes were examined at least once
monthly throughout the year to determine
presence of adult owls or fresh regurgitated
pellets. Behavior of adults was recorded on all
visits, and adults were caught and banded if
possible. Pellets were collected during each
visit. Presence of cached food and prey remains
inside boxes and on silo floors was noted.
Sites where nesting occurred were visited
appro.ximately eveiy 2 wk throughout the breed-
ing season, Januaiy-August, in 1979-1981 and
1984. During 1982 and 1983, a study to develop
a sexing technique (Looman 1985) was started,
and therefore we increased our efforts and vis-
ited active nest boxes more frequently (usually
once a week) throughout most of the breeding
season (May-August) during these years. Nests
were considered active if an adult owl was
obseived in the nestbox or signs of recent occu-
pation were evident (i.e., eggs, eggshells, fresh
pellets in nestbox, nestlings). Onset of egg lay-
ing was determined by direct observation or
by backdating from known-age nestlings or
date of fledging. For backdating, we used 30 d
as an incubation period (Smith et al. 1974,
Marti 1992), with 2 d between individual eggs
(Bunn et al. 1982). Clutch size and productivity'
(fledgling number) data were determined by
direct obsei^vation.
Behavior of adults and nesdings was recorded
at each visit. All young were banded when
approximately 5-6 wk old, and during 1982
and 1983 each young was weighed at fledging
(approximately 8 wk) and sexed according to
the sexing method described by Looman
(1985). While pellets collected during a 5-yr
period (1979-1983) were available for food
habit assessment, only pellets collected in 1982
and 1983 were separated into 4 time group-
ings, each representing a seasonal period of
Barn Owl activity and roughly corresponding
with 1 of the 4 seasons. The spring period
(March-May) corresponded with early repro-
ductive activities, summer (June-August) with
adult attentiveness to fledgling but still depen-
dent young. The autumn period (September-
November) included abandonment and subse-
quent dispersal of most young, and winter
(December-Februaiy) corresponded with the
period that remaining owls moved into well-
protected residential structures.
Pellet analysis followed Marti (1974). Verte-
brate prey remains were identified by compar-
ison with mammal (see Durrant 1952) and bird
specimens at M. L. Bean Museum, Brigham
Young University. Prey weights for estimation
of biomass were means obtained from these
specimens and from reported weight esti-
mates (Marti 1974, Steenhof 1983). Estimated
age of prey for use in biomass calculations was
based on cranial features (ossification of sutures
and auditory bullae and tooth eruption and
wear).
Diversity of Barn Owl diet was determined
using the multivariate statistical package
MVSP (Kovach 1987). To allow comparisons
with other published diversity indices of Bam
Owl diet, diversity indices were calculated
using the modified Shannon-Weiner diversity^
index formula
n = -i(pi){\ogpi\
i=l
where s is the number of species and p,- is the
proportion of the number of indixiduals in the
ith species. Species evenness (E = H/log2;
Magurran 1988) was also calculated.
Results
Breeding Chronoloy
Dates of onset of egg laying range from earh-
Januaiy (date obtained by backdating) through
early August, with 36% commencing egg lay-
ing during the first half of March and 25%
beginning in late FebiTiaiy (Fig. 1). The earliest
date on which eggs were obsened in a box was
February 12, the latest September 14 (eggs
and nestlings observed).
Length of the nesting season for this popu-
lation, defined as the period from deposition
of first egg to fledging of last young, averaged
6.6 mon for the 5->'r period (range 4.0 mon in
1996]
Barn Owls in North Central Utah
75
15 -T
10 -
5 -
ONSET OF EGGLAYING
1979-1983
23 123
9023
9023
1
1
1
1
12
012
012
012
012
012
123 90123
123 90123
3
3
3
1 3
1 3
123
123
9
9012
0
0
0
903
23
1-15
16-31
Jan
1-14 16-28
Feb
1-15 16-31
Mar
"1 1
1-15 16-30
Apr
NESTS
9=1979 n= 7
0=1980 n=16
1=1981 n=20
2=1982 n=19
3=1983 n=23
1213'
1-15 16-31
May
— I
1-15
Jun
— T"
1-15
July
1-15
Aug
'Onset of second clutch
Fig. L Dates of first egg laying by Bam Owls in north central Utah, 1979-1983.
1979 to 9.8 mon in 1983). This is long com-
pared to 5.3 mon in south Texas (Otteni et al.
1972) and in Utah (Smith and Marti 1976) dur-
ing 1974 and 1975; no late autumn nests were
found, however, as have been previously found
in Utah (Smith et al. 1970). Individual nesting
cycles, from deposition of first egg to fledging
of last young in the nest, were approximately
3.3 (3.25 ± 0.2, n = 10) mon in length.
Where egg deposition intervals were known,
the intei-val was 2 d between eggs (2.1 ± 3, n
= 10); this is similar to deposition data (2.3 d)
found for Barn Owls in Springville during
1973 (Smith et al. 1974). Known incubation
times averaged 32.3 d (±3 d, n = 10). Fledging
occurred at 62 d (±4 d), and young remained
in the area until approximately 13 wk of age.
Similar incubation and fledging times are
reported for Barn Owls elsewhere (Pickwell
1948, Reese 1972, Smith et al. 1974).
Nests
Owls made no attempt at nest construction.
However, prenesting behavior of adults, in
which they spent a great deal of time at the
nest site, resulted in a layer of broken down
pellets, incidental feathers, and fecal material
which produced a soft bed for eggs. Eggs
were laid in a shallow area in the middle.
Productivity
Four hundred twenty-eight young were
fledged from 104 (106 including 2nd broods)
nest boxes over a 6-yr period (Table 1), averaging
3.9 young/box with a nest failure rate of 16.6%.
Productivity ranged from 0.8 young fledged/
box (2.0 young/active box) and a failure rate of
25% in 1979, to 4.37 young fledged/box (5.4
young/active box) and a failure rate of 9.1% in
1981.
Mean clutch size for the 5-yr period was
5.8 eggs/clutch (±1.72) and ranged from 5.3
(1979, 1983) to 6.5 (1981) (Table 2). Modal
clutch size was 7 (22%); modal brood size was
7 (21%) (Table 3). Clutch size in 19 nests in
1982 ranged from 2 to 10 eggs and averaged
5.8 (±2.0); broods in these nests ranged from 2
to 7 and averaged 4.0 (±1.9) young hatched/
nest. Thirty-one percent of eggs failed to hatch,
and nestling mortality was approximately 8%.
Productivity in 16 nests where young success-
fully fledged averaged 4.4 (±1.4); however,
productivity fell to 3.7 (±2.1) young fledged/
total nesting attempt.
Clutch size in 23 nests in 1983 ranged from
3 to 9 and averaged 5.3 (±1.8) (Table 2). Brood
number ranged fi-om 2 to 8 and averaged 3.95
(±2.1) young hatched/nest. Twenty-five per-
cent of the eggs failed to hatch, and nestling
76
Great Basin Naturalist
[Volume 56
Table 1. Productivity of Bam Owls using artificial nest boxes in Juab, Utah, and Salt Lake counties, Utah, 1979-1984.
1979
1980
1981
1982
1983
1984
Total
.V
•s
# nest boxes suneved
20
25
27
29
28
29
158
26.3
3.44
# boxes used as nests
8
16
22
19
23^
17
106
17.5
5.39
# fledged
16
63
118
71
80
80
428
71.3
33.04
# Hedged/box (.? )
0.8
2.5
4.4
2.3
2.9
2.8
—
2.6
1.15
# fledged/used box (x )
2.0
3.9
5.4
3.7
3.5
4.7
—
4.0
1.16
# unsuccessful boxes
2
4
2
2
3
2
15
2.5
.83
% unsuccessful boxes
25
25
9.1
15.8
13.0
11.8
—
15.8
7.25
^Single nests at which 2iid liroods occurred are counted twice.
mortality was 12.5%. Nests that successfully
fledged young averaged 4.0 (±1.8) fledglings,
but net productivity for total attempt was 3.5
(±2.2).
Lower clutch sizes (2, 3, 4 eggs/clutch) had
a relatively higher percent success than larger
clutches (>4 eggs/clutch); however, clutch
sizes of 8 produced the highest number of
fledglings (x = 5.3 ± 3.8, n = 3). Clutches
with 5 (n = 8) and 10 (n = 1) eggs were least
productive, with approximately 50% hatching
and fledging success. Seven-egg clutches were
among the more productive clutch sizes, fledg-
ing an average of 5 young (±2.3), with 82%
hatching success and 71% fledging success.
Three instances of 2nd broods occurred
(Table 1). One female (1982) produced 7 fledg-
lings from 1 silo and then from another silo
located approximately 200 m away produced 4
fledglings from a 2nd clutch. The "alternate"
nest site was consistently used for roosting
throughout the previous winter and spring by
a male and during the latter part of the first
nesting period by the nesting pair Since only
the female of the nesting pair was banded, it is
not known whether the male using the "alter-
nate site" during winter and spring was a mem-
ber of the nesting pair, or whether the same
male fathered both clutches. The 2nd and 3rd
instances of 2nd brood occurred in 1983. Each
female produced both clutches in the same box.
Of 19 Barn Owl nesting attempts in 1982
with known outcome, 3 failed to fledge young
(15.8% failure); in 1983, 3 of 23 nests failed to
fledge young (13.0% failure). Nest failures were
believed to have occurred during incubation
or shortly after eggs hatched, judging from the
lack of accumulation of fecal matter and fresh
pellets. Reasons for most nest failures are un-
known, but 1 case of failure was due to human
disturl)ance (use of silo for silage storage).
Other probable causes were loss of 1 or more
parents or desertion, particularly in 1983, when
clutches were abandoned after a long, cool,
wet period followdng egg laying.
Although reasons for all brood reductions
are unknown, some ma>' be attributable to
human disturbance, particularly where there
was evidence of human activit\' at silos. Fratri-
cide may have accounted for at least 2 brood
reductions, where remains of young were in
the nestbox or in pellets. Two reductions were
investigator related and occurred when nest-
lings fell fiom the nestbox after the adult female
flushed.
Sex Ratios
Of 65 fledglings sexed in 1982, 26 were
males and 39 females; this is a significant
overproduction of females (x^= 2.6, 0.5 < P
< 0.10; df = 1). However, the number of
males and females produced during 1983 (of
49 fledglings sexed: 26 females, 23 males) was
not significantly different from the expected
1:1 ratio.
Dispersal
Thirty-five juveniles banded in the study
area between 1979 and 1983 were recovered.
Of these, 61% were within 25 km of their natal
site, 12% within 50 km, and the remainder
within 350 km. Most recovered juxeniles (54%)
dispersing more than 25 km tended to fly
northwest, with most live returns found occu-
pying nestboxes in northern Utah. Twenty-
three percent dispersed to the southwest.
Eleven (31%) recovered owls were less than
6 mon old; these were mostly within 1 km of
the natal site and probably died while dispers-
ing. Nineteen (54%) were approximately 1 yr
old when recovered, 3 (9%) were recovered
approximately 2 yr after banding, and 2 birds
were 3 xr old when recovered alive. One was
captured as a breeding bird at her natal site 3
vr in a row.
1996]
Barn Ow ls in Nohiii Ckntkal Utah
77
Tabi.K 2. Clutch sizes (% of yearly total) of Barn Owls in Jnah, LItah, and Salt Lake counties, Utah, 1979-1983.
1979
1980
1981
1982
1983
'ihtal
# nests
(7)
(16)
(20)
(19)
(23)
(85)
# eggs
1
0
0
0
0
0
0
2
0
0
1(5)
1(5)
0
2(2)
3
0
1(6)
0
1(5)
5 (22)
7(8)
4
2 (28.6)
1(6)
2(10)
3(16)
4(17)
12 (14)
5
2 (28.6)
3(19)
1(5)
4(21)
4(17)
14 (16)
6
2 (28.6)
5(31)
4(20)
2(10)
4(17)
17 (20)
7
1(14)
5(31)
5 (25)
5 (26)
3 (13)
19 (22)
8
0
1(6)
7(35)
1(5)
2(9)
11(13)
9
0
0
0
1(5)
1(4)
2(2)
10
0
0
0
1(5)
0
1(1)
Total eggs
37
95
130
111
121
494
Mean {s)
5.3(1.13)
5.9(1.29)
6.5(1.67)
5.8 (2.03)
5.3(1.81)
5.8 (1.72)
Mortality
Collision with automobiles, shooting, acci-
dents, and severe winter weather coupled with
food shortage have been cited as causes of
mortality of adult Bani Owls (Henny 1969, Fleay
1972, Smith and Marti 1976). At least 12 road-
kills were seen during summer and autumn
1982 in the study area, and accidental deaths
occur frequently, particularly with dispersing
juveniles (Smith and Marti 1976). Of 9 known
accidental deaths of fledglings in 1982 and 7 in
1983, most were due to collisions with cars.
During the winter of 1981-82, at least 55
dead Barn Owls were found in north central
Utah. During this same period, Marti and
Wagner (1985) reported 77 dead Barn Owls in
northern Utah. These birds were emaciated
and death was attributed to starvation result-
ing fi'om cold weather and deep snow. During
the period most deaths occurred, mean tem-
peratures were -9.7°C, 2.4° below normal.
Snow cover was estimated at 20-25 cm, and
this likely interfered with capture of Microtus
spp., the Barn Owl's main prey.
Additional Observations
Adults and fledglings were not color marked;
however, on 1 occasion, a banded fledgling
from 1 silo was found among a same-age brood
in a nearby (ca 0.75 km) silo. The fledgling was
9 wk old and was present at the nearby silo on
2 different occasions. Activity at the silo was
monitored the night of the discoveiy, and the
"foster" fledgling was observed accepting food
brought by the adults. No territorial behavior
was noted by adults or fledglings on this occa-
sion. The only occurrence of territorial behav-
ior noted during the 1982-83 period was
aggressive behavior by a female Bam Owl nest-
ing in a silo in Lehi toward an American Kestrel
{Falco sparverius) nesting in a nearby building.
Pellet and Prey Analysis
A total of 2179 individual prey items were
identified from 888 pellets and pellet frag-
ments gathered from silo floors. An additional
44 prey items were identified from remains on
silo floors (Table 4). At least 16 mammal species
(94% of total prey), 11 bird species (4.8%), and
4 insect groups (0.5%) were identified. By
individuals, Microtus spp. (ca 77%) and Per-
oiuyscus spp. (ca 7%) accounted for over 84%
of total prey. Other important mammalian
species included the western harvest mouse
{Reithrodontomys megalotis), house mouse {Miis
mmculiis), and pocket gopher {Thomomys spp.),
although none constituted over 3% on an
annual basis. The European Starling {Sturnus
vulgaris) and Yellow-headed Blackbird {Xan-
thocephalus xanthocephalus) were the most
frequently taken birds, each comprising 1% of
the total prey.
Percent frequency of each class of food
identified was strongly correlated with per-
centage biomass of the same class of food.
Mammals (over 94% by number) made over
92% by biomass, while birds (over 4% by num-
ber) made over 7% by biomass. Microtus spp.
made up a large proportion (73%) of the bio-
mass, with M. montanus alone accounting for
38% of the biomass consumed (Table 5).
Seasonal comparisons of prey (Appendix 1)
indicate that changes in relative abundance of
prey items occurred during the study. Some
78
Great Basin Naturalist
[Volume 56
Table 3. Number of nestlings (% yearh' total) fledged from artificial nest boxes in Juab, Utah, and Salt Lake counties,
1979-1984.
1979
1980
1981
1982
1983
1984
Total
# nests
(6)
(12)
(20)
(16)
{■20f
(15)
(89)
# fledged (%)
1
2(33)
0
0
0
1(5)
1(7)
4(4)
2
0
2(17)
1(5)
1 (12.6)
2(10)
0
6(8)
3
2 (33)
0
0
3 (19)
7(35)
3 (20)
15 (17)
4
2 (33)
1(8)
4(20)
5 (31)
4(20)
1(7)
17 (19)
5
0
3 (25)
0
4(25)
1(5)
0
8(9)
6
0
2(17)
6 (30)
1(6)
3 (15)
5(33)
17 (17)
7
0
4(33)
8(40)
2 (13)
1(5)
4(27)
19 (21)
8
0
0
1 (5)
0
1(5)
1(7)
3 (3)
Mean (s)
2.7(1.4)
5.3 (1.8)
5.9(1.6)
4.4(1.5)
4.0 (1.8)
5.3 (2.0)
4.8 (1.9)
"Counts 2nd clutches in single nests twice
changes appeared to be seasonal, while others
may be of a long-term nature. While Microtiis
was the most heavily used group throughout
the collecting period, it was used much more
frequently during winter and spring. Peroimjs-
ciis spp. and Thomomys spp. were more fre-
quent in pellets collected during summer and
autumn months. Sorex spp. were present in
pellets during autumn, winter, and spring but
not summer.
Birds were used throughout the year but
were least represented during summer. No sin-
gle bird species was represented in pellets
fi"om all 4 seasons; however, the European Star-
ling, House Sparrow {Passer domesticus), and
Red-winged Blackbird {Agelaiiis phoeniceiis)
were represented in 3 seasons.
Analysis of prey diversity (Table 5) gives
further characterization of the Barn Owl prey
base. Prey species diversity of Barn Owls in
north central Utah was 2.96; ma.\imum diver-
sity possible was 3.434. While this shows some
variation and an ability to take locally abun-
dant prey species, it indicates a degree of sin-
gular specialization on Microtiis spp. Diversity
of north central Utah Barn Owl's food habits is
roughly similar to recorded values observed in
other areas in North America and Europe
(Selleck and Glading 1943, Hawbecker 1945,
Evans and Emlen 1947, Uttendorfer 1952,
Glue 1974, Marti 1974), but it is higher than
values reported from the same area in 1976
(Smith and Marti 1976; T^ible 5). Evenness, the
actual diversity of prey base as a percentage of
maximum diversity possible, was 59%; this indi-
cates Barn Owls were not sampling possible
prey evenly, but rather were taking a higher
percentage of more common species.
Food Brought to Nest
Food stockpiles were found at most nests
during the incubation period. Stockpiling began
slightly before deposition of the first egg and
continued throughout the hatching period.
Initial stockpiles were small, 2-5 prey items,
but stockpile sizes increased as the season pro-
gressed. The largest stockpile consisted of 23
microtines, 3 starlings, and 16 Yellow-headed
Blackbirds. Wallace (1948) reported a stock-
pile of 190 mammals, primarily rodents.
At least 9 prey species were recorded: 53%
microtines, 28% Yellow-headed Blackbirds,
6% starlings, and 3% each of Red-winged
Blackbirds and deer mice. Other species were
the Brown-headed Cowbird {Molothrus ater,
1.5%), Black-billed Magpie {Pico pica, 1.5%),
vagrant shrew {Sorex vagrans, 1%), and Noi-way
rat {Rattns norvegiciis, 1%).
Discussion
Breeding and Productivity
It appears that variability of clutch size in
Barn Owls is more closely related to factors
other than latitude. The 5-yr mean clutch size
(5.8 eggs/clutch) for north central Utah (Lat.
39°— 40°N) reported herein was much higher
than average clutch size of 4.2 eggs reported
for areas of higher latitude, as well as for a
breeding colony studied in the same area in
1973 (Smith et al. 1974); however, this was
much lower than the 4->'r mean clutch size of
7.0 eggs reported by Marti and Wagner (1985)
for northern Utah Barn Owls (Lat. 41 °N).
Additionally, there was a wide discrepancy
between the niodal clutch and brood sizes
1996]
Barn Owls in North Central Utah
79
Table 4. Total prey identifietl for Barn Owls utilizing artificial nest boxes in Juab, Utah, and Salt Lake counties Utah
1982-83,
Number
Percent
Total
Percent total
i^re\ species
frequency
biomass
biomass
Mammals
Microtus pcunsijlvanicus
2L5
9.8
8600.0
9.5
Microtiis immtanus
887
40.4
35480.0
38.7
Microtus longicaudus
377
15.4
15080.0
14.7
Microtus spp.
239
10.9
9560.0
10.4
Mus musculus
51
2.3
969.0
1.1
Neotonm cinerea
9
0.4
2493.0
2.7
Pcromijscus maniculatus
102
4.7
2142.0
2.3
Pcromyscus tniei
2
0.1
42.0
<.l
Pcromijscus spp.
63
2.9
1323.0
1.4
Sorex cinereus
6
0.3
30.0
<.l
Sorex obscurus
14
0.6
84.0
.1
Sorex vagrans
18
0.8
108.0
.1
Sorex spp.
28
1.3
154.0
.2
Spennophihts variegatus
1
tr.
177.0
.2
Rattus norvegicus
5
0.2
1100.0
1.2
Reithrodontoimjs megalotis
32
1.5
350.0
.4
Thomomijs bottae
50
2.3
4250.0
4.6
Thomomys talpoides
1
tr
85.0
.1
Mephitis mephitis
5
0.2
4110.0
4.5
Total mammalian individuals
2105
94.6
86236.0
92.4
Birds
Agelaius phoeniceus
9
0.4
432.0
.5
Cohimba livia
1
tr.
.332.0
.4
Icterus galbula
7
0.3
231.0
.2
Molothrus ater
1
tr
41.0
<.l
Passer domesticus
8
0.4
216.0
.2
Passerculus sandwichensis
2
0.1
42.0
<.l
Pica pica
2
0.1
360.0
.4
Sturnus vulgaris
23
1.0
1817.0
2.0
Turdus migratorius
1
tr.
79.0
<.l
Tyto alba
1
tr.
525.0
.6
Xanthocephalus xanthocephalus
23
1.0
1702.0
1.8
Unidentified birds
30
1.4
1260.0
1.4
Total avian individuals
108
5.9
7037.0
7.5
Total vertebrate individuals
2213
99.5
93273.0
99.9
Invertebr.'\tes
Carabidae
2
0.1
.4
<.l
Tenebrionidae
3
0.1
1.8
<.l
Orthoptera
2
0.1
1.2
<.l
Unidentified Coleopterans
3
0.1
.6
<.l
Total invertebrate individuals
10
0.5
4.0
<.l
Total prey individuals
2223
100.0
93281.0
100.0
''tr = trace
reported herein (7, clutch; 7, brood) and those
reported elsewhere (Bunn et al. 1982 [5, 2],
Ottenietal. 1972 [5, 3]).
Lack (1949) found mean clutch size of owls
to increase with latitude and abundance of
rodents. Otteni et al. (1972) found that clutch
size for 112 clutches in southern Texas (Lat.
28° N) averaged 4.9 and was identical to aver-
age clutch size for 68 Maryland clutches (Lat.
38°- 43 °N; Henny 1969). A mean clutch size
of 5.3 eggs for Barn Owls nesting in Switzer-
land (Lat. 46°-47°N) was also reported by
Henny (1969); Glue (1974) reported an aver-
age clutch size of 4.7 in Great Britain (Lat.
50°-55°N).
Lack (1954) suggested the number of eggs
laid by each species has been established to
correspond with the number of young that can
80
Great Basin Naturalist
[Volume 56
Table 5. DiversiW indices oi Barn Owl piedation for Utah and odier areas.
# prey items
# prey species
Diversity-'
Location
Mammals
Birds
Source
Utah
North central
2173
16
11
2.96
this stud>^
Box Elder Co.
178
8
1
2.31
Smith and Marti 1976
Utah Co.
3004
12
12
1.45
Smith and Marti 1976
California
Southern
933
10
13
2.19
Selleck and Glading 1943
Central
948
20
11
3.10
Hawbecker 1945
Sierras
513
8
0
1.95
Fitch 1947
Northern
739
8
6 +
2.41
Evans and Emlen 1947
Colorado
4366
6
16
2.76
Marti 1974
Idaho
202
9+
1 +
1.79
Roth and Powers 1979
Michigan
6815
5
13
0.98
Wallace 1948
Ohio
1060
9
5
0.98
Phillips 1951
Pennsylvania
6165
7
17
1.46
Latham 1950
Texas
2056
6+
10
3.35
Otteni et al. 1972
Chile
3417
13+
0
2.82
Herrera and Jaksic 1980
England
47865
8+
17
2.29
Glue 1974
3546
8
0
1.60
Webster 1973
Germany
76664
51
32
2.69
Uttendorfer 1952
Spain
12351
11 +
0
2.11
Herrera and Jaksic 1980
^Diversih. calculated usiny .Sliannon-Weiner's diversih index (H ):
-'^(POilogPi)
be successfully raised, and successful rearing
is based on the amount of food available and
provided to young by adults. Otteni et al. (1972)
found that southern Texas Barn Owls seemed
to adjust reproductive efforts to rodent popula-
tion fluctuations. They produced slightly low-
ered mean clutch size and number of complete
clutches during periods of lower rodent prey
population sizes and increased the number of
young raised/pair during periods of abundant
rodent prey populations. Similar findings were
reported in Europe by Glue (data from Bunn
et al. 1982) and Baudvin (1975), whose studies
indicated that variations in fledging success
were entirely linked to vole numbers. Marti
and Wagner (1985) reported that a winter die-
off of northern Utah Barn Owls in 1981-82
resulted in a later egg-laying season, a 40%
decline in breeding attempts, and a decline in
average clutch size from 7.0 to 5.8 eggs; how-
ever, decline in productivity was not paral-
leled in our study area during this period.
These findings indicate that Barn Owl produc-
tivity may be closely tied to availability of
prey, and that differences between clutch and
brood sizes reported herein, and those reported
in the same and in different areas of the Barn
Owl range are likely correlated with fluctua-
tions in prey populations and weather as they
affect prey availability.
Production of 2nd broods by Barn Owls is
thought to be triggered by an abundance of
prey (Honer 1963). All 3 pairs producing 2nd
clutches during this study, 1 in 1982 and 2 in
1983, successfully fledged young from their
1st brood. In these cases, deposition of the 1st
egg of the 2nd brood occurred several weeks
after the last young of the 1st brood fledged.
Second broods are often less successful than
1st broods, since pre>' numbers decline later
in the season when hatchlings still require
feeding (Bunn et al. 1982). This was not the
case with our obsei'vations. All 3 second nests
were successful, with 2 nests 100% successful
in hatching and fledging, and 1 sustaining 60%
mortality of eggs but 100% success in fledging
young. Furthermore, the 3 pairs successfully
fledged 27 young for the breeding season, an
average of 9 young per pair.
Henny (1969) suggested that in northern
environments high biotic potential of Barn
Owls may serve as a "built-in compensating fac-
tor" that affords protection against low years
in rodent cycles and allows rapid restoration
of Barn Owl populations to previous "good
rodent year" size. Second-clutching during
1996]
Bahn Owls in North Central Utah
81
1982 and 1983 nia\' be 1 response to lowered
population numbers resulting from the winter
die-off of 1982 and abandoned clutches result-
ing from cool, wet weather following egg
deposition in 1983.
Food Habits
Barn Owls in Juab, Utah, and Salt Lake
counties sustained themselves almost exclu-
sively by consuming mammals and birds, de-
spite seasonal abundance of large invertebrates,
reptiles, and amphibians. Year-round presence
of microtine species in the diet is in agree-
ment with other data throughout the Barn
Owl's range (Wallace 1948, Phillips 1951,
Otteni et al. 1972, Smith et al. 1972, Webster
1973, Lovari 1974, Lovari et al. 1976, Smith
and Marti 1976, Roth and Powers 1979, Her-
rera and Jaksic 1980).
Webster (1973) and Wallace (1948) noted that
numbers of secondary prey species captured
by Barn Owls are inversely proportional to
numbers of microtines captured, particularly
when Sorcidae spp. form the main alternative
to Microtinae. Although Sorex spp. were uti-
lized frequently by Barn Owls in north central
Utah, no inverse relationship could be seen
between proportions of Sorex spp. and Micro-
tiis spp. An inverse relationship was noted for
proportions of Microtus spp. and Feromyscus
spp. Peromyscus spp. were clearly the main
alternative to Microtus spp. In studies where
numbers of secondary prey species are inversely
proportional to numbers of microtines, the
correlation has been linked with relative pro-
portions of woodland and open areas in the
owls' territories (Bunn et al. 1982). Woodlands
exist in isolated areas throughout the study
area, adjacent to lakes, streams, and foothills,
but open field areas are more common. Thus,
during summer and autumn, one or both adults
may have been foraging more frequently in
woodland areas (represented by Peromyscus
spp.) than in open field areas (represented by
Microtus spp.). During winter and spring, for-
aging may have shifted more to open field
habitats. Alternatively, increased occurrence
or availability of Peromyscus spp. resulting
from increased reproductive activity during
summer and autumn months may account for
the shift in diet.
Only a few unusual prey items are notewor-
thy: predation on a group of striped skunks
{Mephitis mephitis: 2 adults, 3 juveniles) at a
silo in Nephi [C. Marti (personal communica-
tion) doubts that the owl would have killed so
large an animal, but the evidence foimd clearly
indicated that owls nonetheless fed on skimks];
presence of a stockpiled rock squirrel [Sper-
mophilous variagatus) and a sora {Porzana car-
oHmi); cannibalism indicated by presence of a
juvenile Barn Owl skull among loose pellets
collected in autumn, as well as the discovery
of what looked like a partly consumed juvenile
Barn Owl in another nestbox.
Cannibalism has been reported in Califor-
nia (Henny 1969) during years when food sup-
plies were low, and Baudvin (1975) reported
cannibalism as the major source of Barn Owl
nestling mortality in France. Often during this
study, owlets (as well as eggs) seem to have
"disappeared " without a trace. These may have
been cannibalized, they may have died and been
moved to another site, they may have been eaten
by an adult or a sibling, or they may have been
predated by another species. While asynchro-
nous hatching characteristic of Barn Owls is
thought to facilitate cannibalism (O Connor
1978), care should be taken in ascribing Barn
Owl remains in pellets to cannibalism.
Sex Ratios
Mendenhall (1983) reported an equal pro-
duction of sexes in captive Barn Owls at
Patuxent Wildlife Research Center, Maiyland,
but data from the wild are few. The higher
proportion of female fledglings observed in
north central Utah during 1982 was significant
(X-= 2.6, P < 0.10; df = 1), particularly in
view of the high adult winter-kill observed
during the severe winter of 1981-82, and the
hypothesis of sex-biased brood reduction
favoring female offspring during periods of food
(or other environmental) stress (Howe 1977,
Newton 1979, Bildstein 1981) is supported.
While a single season's deviation fiom expected
unity could well be stochastic, differential pro-
duction of sexes during environmentally stress-
ful periods has been obsei-ved in a number of
vertebrate groups (Howe 1977, Bull 1980,
Charnov 1982).
Polygynous behavior by Bani Owls (Baudvin
1975, Bunn et al. 1982, Marti 1990) should be
considered when addressing the differential
sex ratio. Differential sex ratios among polygy-
nous birds are fairly well established (Newton
82
Great Basin Naturalist
[Volume 56
1979, Fiala 1981, Charnov 1982). Polygynous
species tend to show differential production of
sexes more fiequendy dian monogamous species
(Lack 1954, Verner 1964, Zimmerman 1966),
although hypotheses regarding proximate and
ultimate causes vaiy. Olsen and Cockbuni (1991)
have shown that raptors frequently have a nat-
urally biased sex allocation toward females.
The reasons for such an allocation were not
clear although their data did not implicate
polygyny. No verified polygynous behavior was
noted during this study; however, the close
association between the "foster" fledgling and
parents of a separate brood reported herein
indicates a possibility of shared parentage,
particularly since the foster fledgling's natal
site was so close. Unfortunately, adult males
from either silo were never captured for band-
ing, so pairing was unknown. An alternative
explanation of the "foster" fledgling behavior
is that the dispersing fledgling obsei-ved adults
leaving and entering the adjacent silo, and in
stereotypic behavioral fashion it followed the
adults. Once near the nest, normal brood beg-
ging would have elicited feeding response
from the adults.
Further information on Barn Owl mating
behavior and dispersal is needed to elucidate
the differential production of females observed
during this study. More importantly, documen-
tation of sex ratios, both at birth and fledging,
over many years is required to place the ob-
sei-ved skewed sex ratio into perspective.
Addendum: Since the final editing of this
paper a major review of Barn Owls by Marti
(1992) appeared. One should consult that paper
for recent details relevent to our findings.
Acknowledgments
Financial support was provided by the
Associated Students Research Council and
Zoology Department at Biigham Young Univer-
sity. Logistical support was provided by the
Utah Division of Wildlife Resources. Field
assistance was provided by D. Boyce, K. Fris-
tensky, J. Hebdon, K. Keller, R. Meese, K.
Rauhaufer, K. Rhodes, and S. Stewart. Norma
Konrad assisted in identifying and aging mam-
malian prey remains. We thank B. Sample, J.
Flinders, C. Marti, H. D. Smith, R. C. Whit-
more, and an anonymous reviewer for com-
ments on the manuscript.
Literature Cited
Baudvin, H. 1975. Biologie de reproduction de la chou-
ette effraine {Ttjto alba) en Cote d'Ore; premiers
resultates. La Jean le Blanc 14; 1-15.
Behle, W. H. 1941. Barn Owls nesting at Kanab, Utah.
Condor 43:160.
BiLDSTEiN, K. L. 1981. Reversed sexual size dimorphism
in raptors; selective forces acting on non-breeding
birds. Abstract, Proceedings of the 1981 annual meet-
ing of the Raptor Research Foundation, October
1981, Montreal, Quebec.
Bull, J. 1980. Sex determination in reptiles. Quarterly
Review of Biology 55: 3-21.
Bunn, D. S., a. B. Warburton, and R. D. S. Wilson.
1982. The Barn Owl. Buteo Books, Vermillion, SD.
264 pp.
Charnov, E. 1982. The theoiy of sex allocation. Princeton
University Press, Princeton, NJ. 355 pp.
DURIUNT, S. D. 1952. Mammals of Utah. Museum of Nat-
ural History, University of Kansas Publications 6:
1-549.
Evans, E C, and J. T. Emlen, Jr. 1947. Ecological notes
on the prey selected by a Barn Owl. Condor 49; .3-9.
Fiala, K. L. 1981. Sex ratio constancy in the Red-winged
Blackbird. Evolution 35: 898-910.
Fitch, H. 1947. Predation by owls in the Sierran foothills
of California. Condor 49:137-154.
Fleay, D. 1972. Nightwatchmen of bush and plain. Tap-
linger Publishing, Co., NY.
Glue, D. E. 1974. Food of the Bam Owl in Britain and
Ireland. Bird Study 21: 200-210.
Hawbecker, a. C. 1945. Food habits of the Barn Owl.
Condor 47: 161-166.
Henny, C. J. 1969. Geographical variation in mortality
rates and production requirements of the Bam Owl
Tyto alba. Bird-banding 40: 277-290.
Herrera, C, and F M. Jaksic. 1980. Feeding ecology of
the Barn Owl in central Chile and southern Spain: a
comparative study. Auk 97: 760-767.
HONER, M. R. 1963. Obsei-vations on the Barn Owl {Tijfo
alba guttata) in die Netlierlands in relation to its ecol-
ogy' and population fluctuations. Ardea 51; 158-195.
Howe, H. F 1977. Se.x-ratio adjustment in the Common
Crackle. Science 198; 744-745.
Kovach, W. L. 1987. MVSP— Multivariate Statistical Pack-
age, version 1.3, user's manual.
Lack, D. 1949. The significance of clutch size. Part 1: Intra-
specific variations. Ibis 89: 302-352.
. 1954. The natural regulation of animal numbers.
Claredon Press, Oxford.
Lathanl R. M. 1950. The food of predaceous animals in
northeastern United States. Final report, Pittman-
Robertson Project 36-R, PennsyKania Game Com-
mission.
Logman, S. J. 1985. Pr()clucti\ it\, food habits and sexing of
Barn Owls in Utah. Unpublished masters thesis,
Brigham Young University, Provo, UT.
Lo\'ARl, S. 1974. The feeding habits of four raptors in cen-
tral Italy Raptor Research 8; 45-57.
LovARi, S., A. J. Renzoni, and R. Fondi. 1976. The
predator}' liabits of the Barn Owl {Tyto alba) in rela-
tion to the vegetation cover. Bulletin of Zoolog\' 43:
173-191.
1996]
Barn Owls in North Central Utah
83
MAGURa\N, A. E. 1988. Ecological diversity and its mea-
surement. Princeton University Press, Princeton, NJ.
179 pp.
Marti, C. D. 1974. Feeding ecology of four sympatric
owls. Condor 76: 45-61.
. 1990. Same-nest polygxny in the Bam Owl. Condor
92:261-263.
. 1992. Barn Owl. //*; A. Poole, R Stettenheim, and
E Gill, editors. The birds of North America, No. 1.
Academy of Natural Sciences, Philadelphia, and
American Ornithologists" Union, Washington, DC.
Marti, C. D., and P W. Wagner. 1985. Winter mortality
in common Barn Owls and its effect on population
density and reproduction. Condor 87: 111-115.
Marti, C. D., R W Wagner, .\nd K. W. Denne. 1979. Nest
boxes for the management of Barn Owls. Wildlife
Society Bulletin 7: 145-148.
Mendenhall, v. 1983. Captive Barn Owl breeding at
Patu.xent Wildlife Research Center, Laurel, MD.
Unpublished manuscript.
Newton, I. 1979. Population ecology of raptors. Buteo
Books, Vermillion, SD. 399 pp.
O'Connor, R. J. 1978. Brood reduction in birds: selection
for fratricide, infanticide and suicide? Animal Be-
haviour 26: 79-96.
Olsen, R D., and a. Cockburn. 1991. Female-biased sex
allocation in Peregrine Falcons and other raptors.
Behavioral Ecology and Sociobiology 28: 417—423.
Otteni, L. C, E. G. Bolen, and C. Cottam. 1972. Preda-
tor-prey relationships and reproduction of the Barn
Owl in southern Texas. Wilson Bulletin 84; 434-448.
Phillips, R. S. 1951. Food of the Barn Owl, Tyto alba
pratincohi, in Hancock County, Ohio. Auk 68;
239-241.
Picknvell, G. 1948. Barn Owl growth and behaviourisms.
Auk 65: 339-373.
Reese, J. G. 1972. A Chesapeake Barn Owl population.
Auk 89; 106-114.
Roth, D., and L. R. Powers. 1979. Comparative feeding
and roosting habits of three sympatric owls in south-
western Idaho. Murrelett 60; 12-15.
Selleck, D. M., and B. Gladinc;. 1943. Food habits of
nestling Bam Owls and Maish Hawks at l^une Lakes,
California, as determined by the cage nest method.
California Fish and Game 29: 122-131.
Smith, D. G., C. R. Wilson, and H. H. Frost 1970. Fall
nesting of Barn Owls in Utah. Condor 72: 492.
. 1972. Seasonal food habits of Barn Owls in Utah.
Great Basin Naturalist 32: 229-234.
. 1974. Histoiy and ecology of a colony of Barn Owls
in Utah. Condor 76: 131-136.
Smith, D. G., and C. D. Marti. 1976. Distributional sta-
tus and ecology of Barn Owls in Utah. Raptor
Research 10: 33-44.
Steenhof, K. 1983. Prey weights for calculating percent
biomass in raptor diets. Raptor Research 17: 15-27.
Uttendorfer, O. 1952. Neue Ergebnisse uber die Erna-
ghrung der Greifvogel und Eulen. Eugen Ulmer,
Stuttgart.
Verner, J. 1964. Evolution of polygamy in the Long-
billed Marsh Wren. Evolution 18; 252-261.
Voous, K. H. 1988. Owls of the northern hemisphere.
MIT Press, Cambridge, MA.
WalL/\ce, G. J. 1948. The Barn Owl in Michigan; its dis-
tribution, natural histoiy and food habits. Technical
Bulletin of the Agricultural Experiment Station,
Michigan 208: 1-61.
Webster, J. A. 1973. Seasonal variation in mammal con-
tents of Barn Owl castings. Bird Study 20: 185-196.
Woodbury, A. M., C. Cottam, and J. W Sugden. 1949.
Annotated checklist of birds in Utah. University of
Utah Biological Series 39: 1-40.
Zimmerman, J. L. 1966. Polygyny in the Dickcissel. Auk
83; 534-546.
Received 9 August 1995
Accepted 14 September 1995
(Appendix 1 begins on the following page.
84
Great Basin Naturalist
[Volume 56
Appendix 1. Vertebrate food items taken seasonally bv Barn Owls in Juab, Utah, and Salt Lake eounties, LI tab,
1982-83.
Season
Spring
'
Summe
.1,
Autumn
^■
Winter
1
Totab'
%f
Total
%
Total
%
Total
%
Mammals
Mephitis mephitis
0
0
0
0
5
0
0
Mierotus longieaiidus
191
87
54
50
108
73
24
100
Microtus pennsijhaniciis
138
75
7
33
41
64
29
67
Microtus montamts
489
100
67
100
280
100
51
100
Microtus spp.
109
—
19
—
78
—
33
—
Miis iniiscuhis
25
56
3
33
21
64
2
67
Neotoma cinerea
1
6
0
0
8
46
0
0
Peromijseus maniculatus
27
44
23
67
49
91
3
67
Peromyscus truei
0
0
0
0
2
18
0
0
Peromijseus spp.
16
—
11
—
36
—
0
—
Rattus norvegictis
4
6
0
0
1
9
0
0
Reithrodontomijs megalotis
20
69
1
17
8
46
3
67
Sorex vagrans
12
44
0
0
3
18
3
67
Sorex cinereus
5
19
0
0
0
0
1
33
Sorex obscurus
14
25
0
0
0
0
0
0
Sorex spp.
22
—
0
—
3
—
3
—
Spermophilus variegatus
0
0
1
17
0
0
0
0
Thomomijs hottae
6
31
6
50
38
46
0
0
Thomomijs talpoidcs
1
6
0
0
0
0
0
0
Total mammalian individuals
1080
—
192
—
681
—
152
—
Birds
Agelaius phoenieeus
5
25
1
17
3
18
0
0
Cohimba livia
1
6
0
0
0
0
0
0
leterus galhuhi
3
19
0
0
4
27
0
0
Molothrus (iter
1
6
0
0
0
0
0
0
Passer domesticiis
4
6
0
0
2
18
2
67
Passercuhis sandwichensis
0
0
2
17
0
0
0
0
Pica pica
2
12
0
0
0
0
0
0
Stiirnus vulgaris
14
38
0
0
7
55
2
67
Turdiis migraforiiis
0
0
0
0
1
9
0
0
Tijto alba
0
0
0
0
1
0
0
0
Xanthoeephahis xanthocepluil
US 19
31
0
0
4
18
0
0
Unidentified birds
11
2
13
4
Total avian individuals
60
—
5
—
35
—
8
—
Total vertebrate individuals
1140
197
716
160
^total pellets collected; 467; total nests surveyed: 15
"total pellets collected: 61; total nests surveyed: 6
'^total pellets collected: 287; total nests surveyed: 11
"total pellets collected: 76; total nests surveyed: 3
•"total individuals identified
'fre(|iiency of occurrence in nests sui'veyed
Great Basin Naturalist 56(1), © 1996, pp. 85-86
ASTRAGALUS LAXMANNII JACQUIN (LEGUMINOSAE)
IN NORTH AMERICA
R. C. Barnehyi and S. L. Welsh-
Key words: Astragalus laxmannii, nomenclature, North America.
In a recent article Podlech (1993) proposed
lectotypes for two names that have impHca-
tions in the flora of North America, i.e., A. lax-
DUDUiii lacquin and A. adsiirgens Pallas. Both
names have been used in the literature of
American Astragalus in application to the
one species that has been generally accepted
in modern times as A. adsiirgens sens. lat.
(Bameby 1964). Podlech's typifications may be
summarized:
Astragalus laxmannii Jacquin, Hort. Vindob. 3: 22, Tab.
.34. 1776.
Lectotypus (Podlech, Sendtnera 1: 270. 1993): "Planta
culta in Horto Vindobonensi e seminibus a Laxmann e
Sibiria (Samen von Pallas erhalten, siehe Pallas, Sp. Astra-
gal, p. 39. 1800). Specimen a Jacquin missum (BM!)."
Astragalus adsurgens Pallas, Sp. Astragal. 40. 1800.
"Crescit hie Astragalus tantum in regionibus Trans-
Baicalensibus, cum A. Laxmanni promiscue, frequens ad
Selengam, Ononem, circa Tarei-noor, et usque in Mongo-
liae desertum."
Type (Podlech 1993): "Transbaicalia, ad Selengam, Pal-
las (BM!); Onenem circa Tareinoor, Pallas (BM!); Syn-
bi'pen." "Lectot>'pus: ad Selengam, Pallas (BM!)."
Following examination of the proposals by
Podlech, we obtained pertinent specimens on
loan from The Natural History Museum (BM)
in London, through the courtesy of A. R. Vick-
ery. There are 7 pertinent sheets at BM, 6
from the Pallas herbarium and 1 from the
Jacquin herbarium, none of them annotated
by Podlech. The sheet from the Jacquin col-
lection is labeled "Astragalus Laxmannii. Jack.
Hort.VB." and has a notation on the back side,
"Herbar NJ Jacquin." It bears a single plant
with a branched caudex, several stems, and
inflorescences with withered flowers and early
fruit. The plant fits well within the characteri-
zation of A. adsiirgens var adsurgens as de-
scribed by Barneby (1964). It is certainly the
plant chosen as lectotype by Podlech.
Among the 6 specimens from the Pallas
herbarium, 2 bear the designation Astragalus
adsurgens and the additional notation, 'lax-
mannii. " The other 4 are annotated A. laxman-
nii. One of the specimens labeled A. adsur-
gens has 2 notations, 1 at the top, "ad [Tarei-
noor, crossed out] Selengam," and 1 below the
specimen, "Specimen drawn in plate 31. Pall."
This specimen (49221 BM) is the undoubted
lectotype for A. adsurgens. It is mounted with
at least 3 other fragments of the same species.
Sheet 49227 (BM!), bearing a "Type Specimen"
label and with the name A. adsurgens, is likely
a paratype. One (49222 BM!) of the 4 sheets,
all bearing the name laxmannii, also has a
notation, "ad Selengam," and another, "ad
Tareinoor." They are possible paratypes of A.
adsurgens and are mounted with 2 other frag-
ments. Sheets 49223, 49224, 49225 (all BM!)
are all A. laxmannii (as annotated), but appar-
ently they are nomenclaturally irrelevant.
Two infraspecific taxa have been recog-
nized within A. adsurgens in the flora of North
America. Their names require nomenclatural
realignment within A. laxmannii, as follows:
Astragalus laxmannii var. robustior (Hooker) Barneby &
Welsh, comb, nov., based on A. adsurgens var. robustior
Hooker, Fl. Bor.-Amer. 1: 149. 1831.
Astragalus nitidus var. robustior (Hooker) M.E. Jones,
Contr. W. Bot. 10: 64. 1902.
Astragalus adsurgens ssp. robustior (Hooker) Welsh,
Iowa State J. Sci. 37: 357. 1963.
Astragalus laxmannii var. tananaicus (Hulten) Barneby
& Welsh, comb, nov., based on A. tananaicus Hulten, Fl.
Alaska & Yukon 1763. 1959, a substitute for A. viciifolius
Hulten, Ark. Bot. 33B: 1, fig. 1. 1947 (non A. viciaefolius
DC. 1802).
A. adsurgens var. tananaicus (Hulten) Barneby, Mem.
New York Bot. Card. 13: 616. 1964.
'The New York Botanical Garden, Bron.x, NY 10458-5126.
-Herbarium. M. L. Bean Life Science Museum, Brighani Young University, Prove, UT 84602.
85
86 Great Basin Naturalist [Volume 56
References Pall.\s, E S. 1800. Astragalus adswgens Pallas. Species
Astragalorum. Godofiedi Martini, Lipsiae.
Barneby, R. C. 1964. Atlas of North American species of PoDLECH, D. 1993. Miscellaneous notes on Astragalus.
Astragalus. Memoirs of the New York Botanical Gar- Sendtnera 1: 270.
den 13: 1-1188.
Jacquin, N. J. 1776. Astragalus laxmannii Jacquin. Hortus Received 19 July 1995
botanicus vindobonensis 3: 22. Tab. 37. Accepted 5 September 1995
Great Basin Naturalist 56(1), © 1996, pp. 87-89
INTERMOUNTAIN MOVEMENT BY MEXICAN SPOTTED OWLS
{STRIX OCCIDENTALIS LUCIDA)
R. J. Gutierrez^'^, Mark E. Seamans^, and M. Zachariah Peeiy^
Key icorch: Strix occidentalis, Spotted Oivl, dispersal.
The Mexican Spotted Owl {Strix occiden-
talis liicida) is a threatened subspecies in the
United States (USDI 1993). Both the Mexican
and Cahfornia (S. o. occidentalis) Spotted Owl
subspecies are distributed as fragmented pop-
ulations across their respective ranges (USDI
1993, LaHaye et al. 1994). However, it is not
known whether these distributional patterns
represent metapopulations or are the result of
isolation events because no cases of interpop-
ulation (i.e., inteniiountain) dispersal have been
published. A true metapopulation structure
would depend on dispersal among populations
(Levins 1970, Gutien-ez and Hairison in press).
In the course of extensive banding of juve-
nile (n = 95), subadult (n = 21), and adult {n
= 57) Mexican Spotted Owls in the Tularosa
Mountains, New Mexico, we recorded 3 cases
of owl movement among mountain ranges. We
report herein the circumstances of these
movements.
Our study area is in west central New Mex-
ico in the Tularosa Mountains (Fig. 1). We
attempted to capture and color mark every
Spotted Owl during 1991-1995 in a 323-km'2
study area (approximately 70% of the Tularosa
Mountain range) using the methods of Forsman
(1983). In 1994 we established random sample
quadrats to estimate owl densities in areas
surrounding the Tularosa Mountains.
The following movements were recorded:
1. We banded an adult female owl on 24
May 1994. This bird was paired with an adult
male. A female was heard vocalizing from this
territory as late as 13 luly 1994. This female
was found dead near Deming, New Mexico, on
19 January 1995. The bird was autopsied by a
veterinarian in Las Cruces, New Mexico, who
said probable cause of death was electrocution.
which was consistent with circumstances lead-
ing to the bird's discoveiy (i.e., found below a
power pole where an electrical transformer
short had occurred). Although the bird was 68
g lighter in weight when recovered than when
banded, it was in good condition (i.e., no indi-
cation of stai"vation or poor health).
The bird was recovered approximately 187
km south southeast of its banding location
(Fig. 1). Of particular interest was the fact that
the bird probably crossed several mountain
ranges before it entered treeless Chihauhuan
desert grassland where it was recovered. The
nearest suitable owl habitat (e.g., mixed-conifer
or pine-oak forest [Pinus ponderosa/Quercus
spp.]) was in the Animas Mountains, a straight-
line distance of approximately 80 km. The
mountain range nearest (approximately 20 km)
the bird's final location was the Florida Moun-
tains. The highest peak in these mountains is a
prominent landmark (maximum elevation
2224 m) in the desert, but it contains no suit-
able owl habitat (Fig. 1).
We surveyed this bird's territory in early
spring 1995. The male from 1994 was still pres-
ent at the historical location, but we could not
detect a female. However, by June we obsei-ved
an adult female roosting with this male. There-
fore, the female recovered at Deming apparent-
ly left her mate, a relatively uncommon event
among tenitorial Spotted Owls (Gutierrez et al.
1995).
2. In 1993 we banded a juvenile female owl
that we recaptured 56 km west northwest of
its natal site in 1994 on Escudilla Mountain,
Arizona (Fig. 1). This mountain is part of the
San Francisco Mountain Range. This female
was paired at the time of capture and had no
young.
' Department of Wildlife, Humboldt State Universit\', Areata, CA 95521.
-Send reprint requests to Department of Wildlife. Humboldt State University, Areata, C\ 95521.
87
Great Basin Naturalist
[Volume 56
Shortest distance and direction
between owl capture sites and
relocation
FtT^TT ^
Fig. 1. Shortest distance and direction between banding location and final location of dispersing Mexican Spotted
Owls in New Mexico. Shaded area represents all forested/woodland areas whether or not diey are suitable habitat for
Spotted Owls. Numbered lines correspond to nunil>ers in text and do not inipK actual dispersal route of tlie liird.
1996]
Notes
89
3. In 1992 we bunded a jnvenile female owl
whieh we reeaptnred in 1994 in the Mogollon
Mountains, New Mexieo, 22 km south of its
natal site (Fig. 1). This female was paired at
the time of capture and had no young.
Considering that no examples of intermoim-
tain movements ha\'e been recorded among
more extensively studied California Spotted
Owl populations (LaHaye et al. 1992, 1994),
these observations are notable. For example,
between 1987 and 1995, approximately 750
juvenile and adult California Spotted Owls were
banded in the San Bernardino, San Jacinto,
Palomar, and San Gabriel mountain ranges with
no subsequent recoveries in another mountain
range (LaHaye et al. 1994).
Our observation of female-only emigration
out of the Tularosa Mountains is consistent
with the general obsei-vation of female-biased
dispersal in birds (Greenwood 1980). Further,
during our study we relocated a total of 10 dis-
persing juveniles in subsequent years. Of
these, 8 (5 males, 3 females) dispersed within
the Tularosa Mountains. The 5 females dis-
persed an average of 21.8 km (range = 7.75-
56.32 km, s — 20.0) while the 5 males dis-
persed an average of 5.8 km (range = 2.04-
12.58, s = 4.0). Thus, these females dispersed
farther than males (Mann-Whitney U one-
tailed test, ^-value = -2.194, P = 0.0158),
which also supports the idea of female-biased
dispersal in Mexican Spotted Owls.
These intermountain movements also are
consistent with a metapopulation structure
(Levins et al. 1970, Gutierrez and Harrison in
press). In addition, while Spotted Owls are
known to be obligate dispersers (Gutierrez et
al. 1995), the long-distance movement by an
adult female does not fit the general model of
Spotted Owl dispersal (Gutierrez et al. 1985)
in which juveniles are the more likely long-
distance dispersers. However, dispersal car-
ries risks, such as predation, starvation, and
accidents while traveling in unfamiliar habi-
tats. Even though the adult we banded acci-
dentally died, it is possible that adult birds,
which have greater experience, may have a
higher probability of success when crossing
desert grasslands or otherwise unsuitable
habitats in the Southwest than juveniles, who
have little hunting and predator-avoidance
experience. Thus, while studies of juvenile
Spotted Owl dispersal are essential to the study
of metapopulation dynamics (Gutierrez and
Harrison in press), the role of dispersing
adults in maintaining metapopulation struc-
ture should be considered carefully.
Acknowledgments
We thank the following field assistants: V
Baxter, J. Bamesberger, D. Juliano, E. Gunder-
shaug, B. Kwasny, W Michael, W. Moore, and
M. Stauber D. Kristan and G. deSobrino read
the manuscript. The Rocky Mountain Forest
and Range Experiment Station provided fund-
ing for the project (Contract #53-82FT-4-07 to
RJG).
Literature Cited
FORSMAN, E. D. 1983. Methods and materials for locating
and capturing Spotted Owls. USDA Forest Service,
General Technical Report PNW-I62. Pacific North-
west Forest and Range E.xperiment Station, Port-
land, OR.
Greenwood, P J. 1980. Mating systems, philopatiy and
dispersal in birds and mammals. Animal Behavior
28: 1140-1162.
Gutierrez, R. J., and S. Harrison. In press. Applications
of metapopulation theory to Spotted Owl manage-
ment: a histor)' and critique. In D. R. McGullough,
editor, Metapopulations: wildlife management and
conservation. Island Press, Covelo, CA.
Gutierrez, R. J., A. B. Franklin, and W. S. LaHaye.
1995. Spotted Owl. In: A. Poole and F Gill, editors,
The birds of North America, No. 179. The Academy
of Natural Sciences, Philadelphia, PA, and The
American Ornithologists' Union, Washington, DC.
Gutierrez, R. }., A. Fr/\nklin, W. LaHaye, V J. Meretsky,
AND J. P Ward, Jr. 1985. Juvenile Spotted Owl dis-
persal in northwestern Galifomia: preliminary results.
Pages 60-63 in R. J. Gutierrez and A. B. Garey, edi-
tors. Ecology and management of the Spotted Owl
in the Pacific Northwest. USDA Forest Service,
General Technical Report PNW-185. Pacific North-
west Forest and Range Experiment Station, Port-
land, OR.
LaHaye, W. S., R. J. Gutierrez, and H. R. AKgAKAYA. 1994.
Spotted Owl metapopulation dynamics in southern
California. Journal of Animal Ecology 63: 775-785.
LaHaye, W. S., R. J. Gutierrez, and D.' R. Call. 1992.
Demography of an insular population of Spotted
Owls {Sitrix occidentalis occidcntalis). Pages 803-814
in D. R. McGullough and R. H. Barrett, editors.
Wildlife 2001: populations. Elsevier, New York.
Levins, R. 1970. Extinction. Lectures on Mathematics in
the Life Sciences 2: 75-107.
USDI. 1993. Final nile to list the Mexican Spotted Owl as
a threatened species. Federal Register Volume 58,
Number 49: 14248-14271.
Received 25 July 1995
Accepted 2 October 1995
Great Basin Naturalist 56(1), © 1996, pp. 90-92
LIMBER PINE AND BEARS
Heniy E. McCutchen^
Key words: Umber pine, black bears, food habits, Rocky Mountains.
Limber pine [Pinus flexilis) is not consid-
ered a fall food for black bears {Ursus ameri-
canus) or grizzly bears {Ursus arctos) in the
Rocky Mountain region of the United States.
Previous studies have found that other nut-
bearing plant species such as whitebark pine
{P. alhicaulis) and Gambel oak {Quercus gam-
belii) are preferred over limber pine by bears
(Kendall 1983, Mace and Jonkel 1986, Beck
1991). However, these studies have been con-
ducted only in areas where limber pine is in
sympatry with other hard-mast species.
During a study of black bears from 1984 to
1992 (McCutchen 1993) in Rocky Mountain
National Park, it became apparent that bears
utilized limber pine some years. This paper
reports on that use of limber pine and dis-
cusses the implications.
Rocky Mountain National Park, encompas-
sing 107,000 ha, contains elevations among the
highest in the continental U.S., ranging from
2440 m to 4345 m. Nearly 1/3 of the area is
alpine tundra above a 3200-m timberline.
Below timberline, on the upper slopes, is a
subalpine zone of Englemann spruce {Picea
englemannii) and subalpine fir {Abies lasio-
carpa). Extensive stands of lodgepole pine
{Pinus contorta) and scattered stands of limber
pine intermixed with other species are on
middle slopes. At lower elevations Douglas-fir
{Pseudotsuga menziesii) and ponderosa pine
{Pinus ponderosa) are common. Limber pine is
not found west of the Continental Divide in
the park, and the species makes up only about
1% of the forest cover (Hess 1991).
Between 1984 and 1991 I captured 40 indi-
vidual black bears in and adjacent to Rocky
Mountain National Park with culvert traps,
with Aldrich foot snares, or at denning sites.
Twenty-six bears were radiocollared (Telonics,
Mesa, Arizona). Between 1984 and 1991, 9 sub-
adult and adult bears (4 females, 5 males) were
captured and radiocollared on the east side of
the park. Radiolocations were primarily col-
lected by triangulation from automobile or by
hiking. I occasionally used snow tracking to
determine bear activities. I used a vegetation
type map (Hess 1991) to determine the pro-
portion of radiolocations in stands containing
limber pine. Bear scats were collected and
analyzed at the Composition Analysis Labora-
tory, Fort Collins, Colorado, on a gross and
microhistological scale (Sparks and Malacheck
1968).
Two female bears (2 and 3) were monitored
intensively in 1985 and 1986 (McCutchen
1989). Bear 3, a 3-year-old, was captured on
6 August 1985. In 1985 bear 3 spent a signifi-
cant amount of time in limber pine stands in
fall. During the summer, from 6 August to 3
September, she stayed below 3047 m. We
located her 11 times, and none of these loca-
tions were in limber pine. In fall, after annual
plant senescence began to occur, she made a
migration to near timberline and sta\'ed above
3047 m for the next month. From 3 September
to 15 October we located her 14 times; 12 of
these locations were in stands containing lim-
ber pine. On 23 September and again on 11
October I tracked her in the snow and disco\'-
ered that she had been feeding on nuts of lim-
ber pine cones cached in red squirrel {Tamias-
ciurius hudsonicus) middens. At each feeding
site the area was littered with cone cores and
scales, indicating that she spent considerable
time removing nuts from cones. At 2 bed sites,
4 scats were found that consisted almost
entirely of limber pine nut shells. She was
radiolocated in stands containing limber pine
until 15 October, when she moved and
denned on 17 October
The amount of time spent feeding in limber
pine stands was high when calculated in rela-
tion to die amount of time I estimate she was
iNatioiial Biological Sunoy, Colorado Plateau Held Station, Northern Arizona Universit\\ Box 5(iI4. Klafistafl, AZ 860n.
90
1996]
Notes
91
Tabi.K 1. Radiotixes ot black hears in and out ol linil)L'r pine stands in Koek) Mountain National Park, 1985-1990.
Year
1985
1986
1987
1988
1989
1990
Bear #
Out
In
%
Out
In
9f
Out
In
%
Out
In
%
Out
In
%
Out
In
%
2
30
I
3
39
0
0
11
0
0
8
1
11
10
2
17
11
I
8
3
13
12
48
47
0
0
5
1
17
—
—
—
15
3
17
12
1
8
12
—
—
—
—
—
—
8
1
11
8
0
0
17
3
15
10
2
17
Total
43
13
86
0
24
2
16
1
42
8
33
4
out of the den during the year. Her emergence
date from the den in 1985 was unknown be-
cause she had not yet been captured. However,
she emerged from the den in 1986 on 9 May.
Assuming she emerged in 1985 about the
same time (9 May) and denned on 17 October,
she was out for about 160 d. During 1985 she
fed in hmber pine areas from 16 September to
15 October, a period of 30 d, or 19% of her
active time during the year.
Bear 2 did not utihze hmber pine to the
same extent as bear 3 in 1985. Ahhough home
ranges of both were adjacent (McCutchen
1989), bear 2 was located in a Hmber pine area
only once out of 30 radiolocations. Bears 2 and
3 were again intensively radiomonitored in
1986 but were not observed to use limber pine
(Table 1).
Another obsei^vation of bear use of limber
pine habitat was made in 1991. A radiocollared
3-year-old male moved into bear 3's home
range during emigration from his natal range
about 20 km to the south. On 6 December he
was tracked in the snow and was found to
have dug up squirrel caches of limber pine
cones and nuts. He was radiolocated in limber
pine areas until 17 December.
Further analysis of radiolocations from
bears on the east side of the park indicated
that 4 of 9 (bears 2, 3, 12, 24) had been located
in limber pine habitat at least once, 3 of these
several times (Table 1). Percentage of time indi-
vidual bears were found in limber pine stands
varied from 0% to 48%. Of 272 total radioloca-
tions, bears were found in forest types con-
taining limber pine 28 times, or 10.3%.
The importance of limber pine for bears in
the park during the 1985 radiotracking opera-
tion was reinforced by 14 scat samples col-
lected during that year. Four of these (29%)
consisted almost entirely of limber pine seeds.
A review of the literature on bear research
north and south of the park in the Rocky
Mountain region suggests that limber pine is
not important if other hard-mast species are
present. Black and grizzly bears fed on white-
bark pine but not limber pine in Yellowstone
National Park (Kendall 1983) and in northern
Montana (Mace and Jonkel 1986). Aune and
Kasworm (1989) found essentially no grizzly
use of limber pine in 10 yr of study in the
Montana Front Range. In Montana, Idaho, and
most of Wyoming, whitebark pine is either the
sole hard-mast species or is more common
than limber pine. In south central Colorado,
Beck (1991) found that black bears made long-
distance movements to feed on acorns of
Gambel oak but not on limber pine. However,
there are areas in Colorado and Wyoming
where limber pine is the sole hard-mast-pro-
ducing species present and may be important
to bears (Fig. 1).
Bear preference for whitebark pine and
Gambel oak over limber pine is probably
related to several factors. Limber pine seeds
are smaller than the other two, producing 10.8
X 10^ seeds/kg as compared to whitebark pine
at 5.7 X 103 seeds/kg (McCaughey and Schmidt
1990) and Gambel oak at 1.3 X 10^ /kg (Haiper
et al. 1985). Limber pine generally produces
large seed crops at wide and irregular intervals
with small amounts produced nearly every
season. Whitebark pine seeds are produced at
frequent and regular intervals (Harlow et al.
1979) with good crops produced at intervals of
3-5 yr (McCaughey and Schmidt 1990). In
south central Colorado, Beck (1991) found
Gambel oak production to be quite regular
with only 1 massive acorn crop failure in 10 yr.
From the limited number of obsei"vations of
limber pine use by bears in Rocky Mountain
National Park, I suggest that if limber pine is
the only hard mast available during certain
years, perhaps years of limited production of
other foods, it may be an important food
source for the survival of bears. This hypothe-
sis needs to be tested by further research.
92
Great Basin Naturalist
[Volume 56
I :::J Limber pine only
1 I Otiier species preferred
Fig. 1. Distribution map of limber pine in the Rocky
Mountain region in relation to other hard-mast bear foods
based on Little (1971). Limber pine only (shaded fill) des-
ignates areas where limber pine occurs exclusive of any
other hard-mast species. In this area limber pine may be
important to bears as a food source. Other species pre-
ferred (dotted fill) designates areas of limber pine distrib-
ution where other hard-mast species, whitebark pine to
the north and Gambel oak to the south, dominate and are
preferred by bears over limber pine. (Note: State and
county boundaiies are shown to locate limber pine range;
GNP'== Glacier National I'ark. YNP = Yellowstone
National Park, RMP = Rocky Mountain National Park,
MP = Monarch Pass.)
Acknowledgments
Funding for this research was provided by
the National Park Service. I thank David
Stevens, Robert Schiller, and the staff of Rocky
Mountain National Park for their support on
this project.
Literature Cited
Aune, K., and W. Kasvvorm. 1989. Final report East Front
grizzly studies. Montana Department of Fish, Wildlife
and Parks, Helena. 332 pp.
Bec:k, T. 1991. Black bears of west-central Colorado.
Technical Publication 39. Colorado Division of
Wildlife, Fort Collins, CO. 86 pp.
Harlow, W. M., E. S. Harrar, and F M. White. 1979.
Textbook of dendrology. 6th edition. McGraw-Hill,
New York. 510 pp.
Harper, K. T, E J. W.^gstaff, and L. Kunzler. 1985.
Biology and management of the Gambel oak vegeta-
tive type; a literature review. General Technical
Report I NT- 179. U.S. Department of Agriculture,
Forest Service, Intermountain Forest and Range
Experiment Station, Ogden, UT. 31 pp.
Hess, K. 1991. Description and e\aluation of co\'er t>pes
in the Rock^ Mountain National Park. Final report
to Rocky Mountain National Park, Colorado. Januan-
1991. 195 pp.
Kendall, K. C. 1983. Use of pine nuts by grizzly and black
bears in the Yellowstone area. International Confer-
ence of Bear Research and Management 5: 166-173.
Little, E. L., Jr. 1971. Atlas of United States trees. Vol-
ume 1. Conifers and important hardwoods. Miscella-
neous Publication 1146. U.S. Department of Agricul-
ture, Forest Sei-vice, Washington, DC. 200 pp.
Mace, R. D., and C. J. Jonkel. 1986. Local food habits of
the grizzly bear in Montana. International Conference
on Bear Research and Management 6; 105-110.
McCaughey, W. W, and \V. C. Schmidt. 1990. Autecolog>
of whitebark pine. Pages 85-96 in W. C. Schmidt
and K. J. McDonald, editors. Proceedings of a sym-
posium on whitebark pine ecos\stems; ecology' and
management of a high mountain resource, Bozeman,
MX 29-31 March 1989. General Technical Report
INT-270. U.S. Department of Agriculture, Forest Ser-
vice, Intermountain Forest and Range Reseaich Sta-
tion, Ogden, UT. 386 pp.
McCuTCHEN, H. E. 1989. Cnptic beha\'ior of black bears
(Ursits americanus) in Rocky Mountain National
Park, Colorado. International Conference on Bear
Research and Management 8; 65-72.
. 1993. Ecolog>' of a high mountain black bear pop-
ulation in relation to land use at Rocky Mountain
NP Park Science 13; 25-27.
Sharks, D. R., and J. C. M.-\lachek. 1968. Estimating per-
centage diy weight in diets using a microscopic tech-
niciue. Journal of Range Management 21: 264-265.
Received 4 January 1995
Accepted 14 August 1995
Great Basin Naturalist 56(1), © 1996, p. 93
BOOK REVIEW
Utah Wildflowers: A Field Guide to Northern
and Central Mountains and Valleys.
Richard J. Shaw. Utah State University
Press, Logan, UT. 1995. $12.95 softback. '
Wildflovver books belong to a genre of pub-
lications specifically designed for people who
wish to see and identify pretty flowers. The
wildflowers of the region covered by this
handsomely designed book are certainly wor-
thy of such a publication. It is conveniently
sized for carrying into the field and presents
species by flower color, as in many other
books of this kind. This enables the user to
find potential identities of plants encountered
in the field.
As in practically all other wildflower books,
the writer confronts the enigma of presenting
an overall view of the plant or emphasizing the
flowers alone. It is the impossibility again of
having a wide-angle telephoto lens. The images
are clear and shaip, and if the user is able to
make the comparison of flowers alone, then
the book will be very useful as an identifica-
tion tool. The author of the book also had to
make arbitrary decisions on which examples to
treat. There are more than a thousand species of
flowering plants in the region covered by this
book, which treats some 92 of them. Those
presented are, however, beautiful.
This book should be enjoyed for more than
its usefiilness in identification. It can be viewed
in those times of year, and in those places,
where wildflowers are not flowering. The pho-
tos will add chanii and understanding by them-
selves.
The author and the press responsible for
production of this book should be compli-
mented.
Stanley L. Welsh
Life Science Museum
Brigham Young University
Provo, UT 84602
93
INFORMATION FOR AUTHORS
The Great Basin Naturalist welcomes previously
uiipuhlished manuscripts pertaining to the biologi-
cal natural history of western North America.
Preference will he given to concise manuscripts of
up to 12,000 words. Simple species lists are dis-
couraged.
SUBMIT MANUSCRIPTS to Richard W. Baumann,
Editor, Great Basin Naturalist, 290 MLBM, PO Box
20200, Brigham Young University, Provo, UT
84602-0200. A cover letter accompanying the man-
uscript must include phone number(s) of the author
submitting the manuscript, and FAX number and
E-mail address when applicable; the letter must
also provide information describing the extent to
which data, text, or illustrations have been used in
other papers or books that are published, in press,
submitted, or soon to be submitted elsewhere.
Authors should adhere to the following guidelines;
manuscripts not so prepared may be returned for
revision.
MANUSCRIPT PREPARATION. In general, the Great
Bosin Naturalist follows recommendations in
Scientific Style and Format: The CBE Manual for
Authors, Editors, and Publishers, 6th edition
(Council of Biology Editors, Inc., II South LaSalle
Street, Suite 1400, Chicago, IL 60603, USA; phone
312-201-0101; F.\x 312-201-0214). We do, however,
differ in our treatment of entries in Literature Cited.
Authors may consult Vol. 51, No. 2 of this journal
for specific instructions on format; these instruc-
tions. Guidelines for Manuscripts Submitted
TO THE Great Basin Naturalist, are printed at the
back of the issue. Also, check the most recent issue
of the Great Basin Naturalist for changes.
TYPE AND DOUBLE SPACE all materials, including
literature cited, table headings, and figure legends.
Avoid hyphenated words at the right-hand margins.
Underline words to be printed in italics. Use stan-
dard bond (22x28 cm), leaving 2.5-cm margins on
all sides.
SUBMIT 3 COPIES of the manuscript and the origi-
nal on a 5.25- or 3.5-inch disk utilizing WordPerfect
5.1 or above. Number all pages and assemble each
copy separately: title page, abstract and key words,
text, acknowledgments, literature cited, appendices,
tables, figure legends, figures.
TITLE PAGE includes an infomiative title no longer
than 15 words, names and addresses of authors, a
running head of fewer than 40 letters and spaces,
footnotes to indicate change of address and author
to whom correspondence should be addressed if
other than the first author.
ABSTRACT states the purpose, methods, results,
and conclusions of the research. It is followed by
6-12 key words, listed in order of decreasing
importance, to be used for indexing.
TEXT has centered main headings printed in all
capital letters; second-level headings are centered
in upper- and lowercase letters; third-level head-
ings begin paragraphs.
VOUCHER SPECIMENS. Authors are encouraged to
designate, properly prepare, label, and deposit
high-cjuality voucher specimens and cultures docu-
menting their research in an established permanent
collection, and to cite the repository in publication.
REFERENCES IN THE TEXT are cited by author and
date: e.g., Nfartin (1989) or (Martin 1989). Multiple
citations should be separated by commas and listed
in chronological order. Use "et al. " after name of
first author for citations having more than two
authors.
ACKNOWLEDGMENTS, under a centered main
heading, include special publication numbers when
appropriate.
LITERATURE CITED, also under a centered main
heading, lists references alphabetically in the fol-
lowing formats:
Mack, G. D., and L. D. Flake. 1980. Habitat rela-
tionships of waterfowl broods on South Dakota
stock ponds. Journal of Wildlife Management
44: 695-700.
Sousa, W. P 1985. Disturbance and patch dynamics
on rocky intertidal shores. Pages 101-124 in
S. T. A. Pickett and P S. White, eds.. The ecolo-
gy of natural disturbance and patch dynamics.
Academic Press, New York.
Coulson, R. N., and J. A. Witter. 1984. Forest ento-
mology: ecology and management. John Wiley
and Sons, Inc., New York. 669 pp.
TABLES are double spaced on separate sheets and
designed to fit the width of either a single column
or a page. Use lowercase letters to indicate foot-
notes.
PHOTOCOPIES OF FIGURES are submitted initially
with the manuscript; editors may suggest changes.
Lettering on figures should be large enough to
withstand reduction to one- or two-column width.
Originals must be no larger than 22x28 cm.
NOTES. If the manuscript would be more appro-
priate as a short communication or note, follow the
above instructions but do not include an abstract.
A CHARGE of $50 per page is made for articles
published; the rate for individual subscribers will
be $35 per page. However, manuscripts with com-
plex tables and/or numerous half-tones will be
assessed an additional charge. Reprints may be pur-
chased at the time of publication (an order form is
sent with the proofs).
FINAL CHECK:
• Cover letter explaining any duplication of
information and providing phone number(s),
FAX number, and E-mail address
• 3 copies of the manuscript and WordPerfect
diskette
• Conformity with instructions
• Photocopies of illustrations
(ISSN 001 7-3614)
GREAT BASIN NATURALIST Vol 56 no lJanuaryl996
CONTENTS
Articles
Temporal and spatial dishibution of highway mortality of mule deer on newly con-
structed roads at Jordanelle Reservoir, Utah Laura A. Romin
and John A. Bissonette 1
Exceptional fish yield in a mid-elevation Utah trout reservoir: effects of angling
regulations Wayne A. Wurtsbaugh, David Barnard,
and Thomas Pettengill 1 2
Consumption of diffuse knapweed by two species of polyphagous grasshoppers
(Orthoptera: Acrididae) in southern Idaho Dennis J. Fielding,
M. A. Brusven, and L. E Kish 22
Fire frequency and the vegetative mosaic of a spruce-fir forest in northern Utah
Linda Wadleigh and Michael J. Jenkins 28
Arizona distribution of three Sonoran Desert anurans: Bufo retiforrnis,
Gastrophryne olivacea, and Pternohijla fodiens Brian K. Sullivan,
Robert W. Bowker, Keith B. Malmos, and Erik W. A. Gergus 38
Habitat affinities of bats from northeastern Nevada Mark A. Ports
and Peter V Bradley 48
Nuptial, pre, and postnuptial activity of the thatching ant, Formica ohscuripes
Forel, in Colorado John R. Conway 54
Trachytes kaliszewskii n. sp. (Acari: Uropodina) from the Great Basin (Utah, USA),
with remarks on the habitats and distribution of the members of the genus
Trachytes Jerzy Bloszyk and Pawel Szymkowiak 59
Productivity, food habits, and associated variables of Barn Owls utilizing nest
boxes in north central Utah Sandra J. Looman, Dennis L. Shirley,
and Clayton M. White 73
Notes
Astragalus laxmannii Jacquin (Leguminosae) in North America
R. C. Barneby and S. L. Welsh 85
Intermountain movement by Mexican Spotted Owls {Strix occidentalis lucida)
R. J. Gutierrez, Mark E. Seamans, and M. Zachariah Peery 87
Limber pine and bears Henry E. McCutchen 90
Book Review
Utah wildflowers: a field guide to northern and central mountains and valleys
Richard J. Shaw Stanley L. Welsh 93
H E
^A!\>
GREAT BASIN
NATURALIST
VOLUME 56 N2 2 — APRIL 1996
BRIGHAM YOUNG UNIVERSITY
GREAT BASIN NATURALIST
Editor
Richard W. Baumann
290 MLBM
PO Box 20200
Brigham Young University
Provo, UT 84602-0200
801-378-5053
FAX 801-378-3733
Assistant Editor
Nathan M. Smith
190 MLBM
PO Box 26879
Brigham Young University
Provo, UT 84602-6879
801-378-6688
E-mail: NMS@HBLL1.BYU.EDU
Associate Editors
Michael A. Bovvers
Blandy Experimental Fann, University of
Virginia, Box 175, Boyce, VA 22620
J. R. Callahan
Museum of Southwestern Biology, University' of
New Mexico, Albuquerque, NM
Mailing address: Box 3140, Hemet, CA 92546
Jeffrey J. Johansen
Department of Biology, John Carroll University
University Heights, OH 441 18
Boris C. Kondratieff
Department of Entomology', Colorado State
Universitv, Fort Collins, CO 80523
Paul C. Marsh
Center for Environmental Studies, Arizona
State University, Tempe, AZ 85287
Stanley D. Smith
Department of Biology
University of Nevada-Las Vegas
Las Vegas, NV 89154-4004
Paul T. Tueller
Department of Environmental Resource Sciences
Universitv of Nevada-Reno, 1000 Vallev Road
Reno, NV 89512
Robert C. Whitmore
Division of Forestry, Box 6125, West Virginia
University, Morgantown, WV 26506-6125
Editorial Board. Jerran T. Flinders, Chairman, Botany and Range Science; Duke S. Rogers, Zoolog}';
Wilford M. Hess, Botany and Range Science; Richard R. Tolman, Zoology. All are at Brigham '\bung
University. Ex Officio Editorial Board members include Steven L. Taylor, College of Biology and Agriculture;
H. Duane Smith, Director, Monte L. Bean Life Science Museum; Richard W. Baumann, Editor, Great Basin
Naturalist.
The Great Basin Naturalist, founded in 1939, is published quarterly by Brigham Young University.
Unpublished manuscripts that Rniher our biological understanding of the Great Basin and suirounding areas
in western North America are accepted for publication.
Subscriptions. Annual subscriptions to the Great Basin Naturalist for 1996 are $25 for individual sub-
scribers ($30 outside the United States) and $50 for institutions. The price of single issues is $12. All back
issues are in print and available for sale. All matters pertaining to subscriptions, back issues, or other busi-
ness should be directed to the Editor, Great Basin Naturalist, 290 MLBM, PO Box 20200, Brigham Young
University, Provo, UT 84602-0200.
Scholarly Exchanges. Libraries or other organizations interested in obtaining the Great Basin
Naturalist through a continuing exchange of scholarlv publications should contact the Exchange Librarian,
6385 HBLL, PO Box 26889, Brigham \bung University, Provo, UT 84602-6889.
Editorial Production Staff
JoAnne Abel Technical Editor
Jan Spencer Assistant to the Editor
Copyright © 1996 by Brigham Young University
Official publication date: 29 April 1996
ISSN 0017-3614
4-96 750 17922
The Great Basin Naturalist
Published at Proxo, Utah, by
Brigham Young University
ISSN 0017-3614
Volume 56 30 April 1996 No. 2
Great Basin Naturalist 56(2), © 1996, pp. 95-118
SELECTING WILDERNESS AREAS TO CONSERVE
UTAH'S BIOLOGICAL DIVERSITY
Diane W. Davidson l, William D. Newmark-, Jack W. Sites, Jr.'^ Dennis K. Shiozawa'^,
Eric A. Rickart-, Kimball T. Harper"', and Robert R. Keiter^
Abstr.\c:t. — Congress is currently evaluating the wilderness status of Bureau of Land Management (BLM) public
lands in Utah. Wilderness areas play many important roles, and one critical role is the consei'vation of biological diver-
sity. We propose that objectives for conser\'ing biodiversity on BLM lands in Utah be to (1) ensure the long-term popu-
lation viability of native animal and plant species, (2) maintain the critical ecological and evolutionaiy processes upon
which these species depend, and (3) preserve the full range of commimities, successional stages, and environmental gra-
dients. To achieve these objectives, wilderness areas should be selected so as to protect large, contiguous areas, augment
existing protected areas, buffer wilderness areas with multiple-use public lands, interconnect existing protected areas
with dispersal and movement corridors, conserve entire watersheds and elevational gradients, protect native communi-
ties from invasions of e.xotic species, protect sites of maximum species diversity, protect sites with rare and endemic
species, and protect habitats of threatened and endangered species. We use a few comparatively well-studied ta.xa as
examples to highlight the importance of particular BLM lands.
Key words: wilderness, biodiversity, conservation. Utah, Bureau of Land Management, endemic species, exotic
species, cryptobiotic soils, plants, bees, vertebrates.
The Wilderness Act and Biodiversity historical value" (16 U.S. Code, § 1131 [c][4]).
Ecological concerns have also figured promi-
In the Wilderness Act of 1964, Congress nently in several congressional wilderness
endorsed the presentation of federal land in its bills for Bureau of Land Management (BLM)
natural state (16 U.S. Code, Sections 1131-36). public lands. Both the Alaska National Interest
Congress plainly anticipated that ecological Lands Conservation Act, 16 U.S. Code, § 3101
considerations w^ere an important dimension (b), and the California Desert Protection Act,
of the wilderness concept, since the act pro- 103 Public Law 433 Section 2 (b) (1) (B) (1994),
vides that wilderness may contain "ecological" expressly acknowledge that wilderness designa-
features of "scientific, educational, scenic, or tion is intended to protect important ecological
1 15fpartiiu-iit ot BinloKx', Unh'ersih- of Utah, Salt Lake Git)'. UT S4112.
^Utah Museum of Natural History, Universit\- of Ut;di, Salt Lake City, UT 84112.
■'Department of Zoology, Brigham Young University, Provo, UT 84602.
"^Department of Botan\', Brigham Young University, Provo. UT 84602.
^College of Law, Uni%ersit\ of Utah, Salt Lake Cit\', UT 84112.
95
96
Great Basin Naturalist
[Volume 56
values. Among the significant ecological func-
tions of wilderness areas is their role in con-
sei^ving biological diversity (biodiversity).
In Utah, undeveloped public lands admin-
istered by the BLM (Fig. 1) can potentially
play a key role in conserving the state's natural
heritage. The BLM is now pursuing an ecosys-
tem management policy designed to ensure
sustainable ecological processes and biological
diversity on lands under its jurisdiction (Depart-
ment of the Interior 1994). By using these
same criteria to designate wilderness areas.
Congress could not only advance the BLM's
ecosystem management goals but also reduce
conflict over the agency's multiple-use lands
(e.g., by diminishing the risk of future endan-
gered species listings and the accompanying
regulatoiy limitations). Over the long tenn, it is
both cheaper and easier to protect species in
aggregate in their intact, functioning ecosys-
tems than to conserve them individually in
fragmented and decimated populations under
the Endangered Species Act.
In short, the use of biological and ecologi-
cal criteria to designate BLM wilderness areas
in Utah is consistent with the legal concept of
wilderness and would help to avoid future
conflicts over resource management.
BioDivERSiTi' Defined
Biological diversity — the variety of life in a
given area — includes three hierarchical com-
ponents: genetic diversity, species diversity, and
ecosystem diversity (e.g.. National Research
Council 1978, Wilson 1988, Reid and Miller
1989, Raven 1992). Cenetic diversity refers to
the variety of genes within species. Depletion
of genetic diversity during population bottle-
necks, or because of inbreeding within frag-
mented and isolated populations, can threaten
a species' sundval by reducing the capacity of
organisms to adapt to changing environments
(Soule and Wilcox 1980, Frankel and Soule
1981). Species diversity, or the number of
species within a region (species richness), can
be divided into three major components
(Whittaker 1972): alpha diversity {a), the num-
ber of species in a homogeneous habitat; beta
diversity (/3), the rate of species-turnover
across habitats; and gamma diversity (y), the
total number of species observed in all habi-
tats within a region. Finally, ecosystem diver-
Fig. 1. Map of the state of Utah showing (in black) loca-
tions of all existing roadless areas proposed for BLM
wilderness status. The BLM formally studied a suliset of
these areas and recommended a portion ot studied lands
for wilderness status. Data are from a Department of Inte-
rior map of BLM Wilderness Study Areas, BLM Proposed
Wilderness, and the Utah Wilderness Coalition's BLM
Wilderness Proposal. County boundaries also are shown.
Isolated moiuitain ranges in Utah's western deserts are
identified as follows: a = Deep Creek; b = Fish Springs; c
— House range, and d = Newfoundland range (not for-
mally proposed or studied for wilderness designation). On
the Colorado Plateau, e = the Henn' Mountains.
sity consists of the xariety of major ecological
communities within areas that are heteroge-
neous in their physical attributes, for example,
in elevation or soil type.
Genetic, species, and ecosystem di\'ersity
all result from both interactions bet\\'een organ-
isms and their environments, and interactions
of organisms with one another. The physical
environment sets limits on wliich species can
inhabit an area, and interactions among those
species determine which are most abundant.
Strategies for preserving biodixcrsitx' must
therefore take note of all li\ ing things in the
landscape, and the linkages among them.
Finally, since different species specialize on
different stages of natural disturbance cycles,
it is important to presei-ve a range of commu-
nities and ecosystems representing all stages
in the disturbance cvcle.
1996]
WlLDKRNESS SELECTION FOR BlODIN KHSITY
97
Objectives
The success of conserving biological diver-
sit) within a s\ stem ot" protected areas can
only be assessed in relationship to a series of
selected objectives. We propose that the con-
sei-vation of Utah's biological diversity depends
on (1) ensuring die long-term viability of native
plant and animal populations, (2) maintaining
the criticd ecological and evolutionaiy processes
upon which these species depend, and (3) pro-
tecting the full range of communities, succes-
sional stages, and environmental gradients (e.g.,
lUCN 1978, MacKinnon et al. 1986, Noss 1992).
Both the size of the network of protected
areas and the selection of individual wilderness
areas should be guided by these 3 goals.
Although it is possible to presence a small sub-
set of species and genotypes in zoological and
botanical gardens, communities and species
interactions must be consei-ved in situ. Large
areas with minimal human intrusion, and with
natural processes reasonabK' intact, are critical
elements of an in situ conservation strategy;
tliey provide protection for fiagile habitats, such
as easily eroded soils, and preserve habitat for
reclusive species. Moreover, wilderness areas
offer natural ecosystems some protection from
the biological invasions that have devastated
many communities, especially plant communi-
ties, across Utah.
Here we describe a strategy, based upon
widely accepted principles of conservation
biology (see e.g., Primack 1993, Meffe and
Carroll 1994), for both selecting critical sites
for wilderness designation and determining
the amount of habitat that should be pre-
served as wilderness (see also Babbitt 1995).
Criteria for Selection
Viable Populations
Utah contains approximately 3000 indige-
nous plant species and varieties and about 584
vertebrate species. Viable populations for most
of these plants and animals can be ensured by
focusing, within ecological communities, on
species for which the risk of extinction is
greatest. Risk-prone species typically include
those with small populations, large home
range requirements, low reproductive poten-
tial, restricted geographic ranges, or large
temporal xariation in population size (Brown
1971, Willis 1974, Terborgh and Winter 1980,
Diamond 1984, Pimm et al. 1988, Belovsky et
al. 1994, Newmark 1995). Many top predators
have several of these traits. On BLM lands in
Utah, examples of such organisms are river
otter {Lutra (•anadi'iisi.s) and both Bald and
Golden Eagles {Haliaeefus lencoc('})lialu.s and
AqiiiJa chrijsaetos). Risk-prone plants include
Holmgren locoweed {Astragalus hobngrenio-
nim) and Jones cycladenia {Cijclaclenia huinilis
var. jonesii), which have highly specific sub-
strate recjuirements .
Viability of populations depends on both
the level of risk one is willing to accept, and the
time frame over which one wishes to consene
the population (Shaffer 1981, Schonewald-Cox
1983, Soule 1987). In general, both survival
time and the likelihood of population persis-
tence increase with population size. A level of
risk and persistence that is commonly pro-
posed as a management goal is a 99% chance
of sui-vival for 1000 years (e.g., Belovsky 1987,
Armbruster and Lande 1993).
For large carnivores, the minimum viable
population necessary to ensure a 99% chance
of survival for 1000 vears is estimated to be
approximately 10,000-100,000 individuals (Be-
lovskv 1987). In habitat area, this is equivalent
to 100,000-1,000,000 km2, or 2.5-25 million
acres. Although this area requirement may
seem remarkably large, documented losses of
mammalian species from among the largest of
North American national parks (e.g., the
10,328-km^ Yellowstone-Grand Teton park
assemblage) during the last 90 years make
clear the importance of protecting large areas
(Newmark 1987, 1995).
Maintenance of Ecological
and Evolutionary Processes
In selecting wilderness areas, one must
take care to ensure the maintenance of the
ecological and evolutionary processes upon
which all plant and animal species depend
(Pickett and Thompson 1978, Kushlan 1979).
Among the most important of these processes
are natural disturbance and recovery cycles.
Ideally, criteria for the selection of wilderness
areas should include information on fre-
quency, size, and longevity of natural distur-
bances. Protected areas should be large
enough to contain minimum critical areas of
the entire range of recovery stages for each
community type (Pickett and Thompson
1978). In western North America, natural dis-
turbance regimes can encompass tens of thou-
sands to millions of acres, as witnessed by the
recent and extensive wildfires in Yellowstone
National Park (Christensen et al. 1989).
98
Great Basin Natueulist
[Volume 56
Two other critical ecological processes are
migration and dispersal of terrestrial organ-
isms across landscapes, and of aquatic species
within watersheds. The selection of wilder-
ness areas requires that attention be given to
ensuring that migratoiT pathways are open to
organisms migrating seasonally along eleva-
tional gradients. Of particular importance is
the need to maintain winter ranges and migra-
tory routes of large mammals such as mule deer
{Odocoileus hemioniis), elk {Cervus elaphiis),
and moose {Alces dices).
Interactions among competitors, and be-
tween predators and prey, are integral aspects
of natural ecosystems and should be pre-
served. For example, in the southwestern
deserts of the United States, the direct and
indirect effects of seed predation on plant
community structure have been documented
in long-term experiments manipulating densi-
ties of rodent and ant grani\'ores (Daxidson et
al. 19S4, Samson et al. 1992). These effects
include transformation ol a shrubland into a
grassland biome (Brown and Heske 1990).
Special care must be taken to consei^ve popu-
lations of predators with large area require-
ments, because extinctions of these species
can alter whole communities (e.g., by leading
to outbreak densities of prey, which then over-
exploit their plant resources). Some of the
strongest e\'idence for such "trophic cascades"
comes from the Greater Yellowstone Ecosys-
tem, where intensive browsing by elk has
greatly altered many riparian zones by the re-
moval of willows (genus Salix), and has elimi-
nated aspen seedlings {Popiihis fremiiloides)
recruiting from seeds and rhizomes shortly
after the extensive 1988 fires. Huge contem-
porary elk herds, numbering ~ 40,000 individ-
uals in the park, and 20,000 in the northern
herd alone, are likely the result of reductions
in the full complement of large predators (Kav
1990, Wagner et al. 1995). Gonsidcrable evi'-
dence also suggests that deer and elk herds in
Utah average significant!)' larger at present
than during any extended period in the histor-
ical past (Durrant 1950, Julander 1962, Haiper
1986).
Strategies for Selecting
Wilderness Areas
Landscape-wide Priorities
Given the large area requirements of many
extinction-prone Utah species, it is important
to protect large, contiguous land blocks. In
designating wilderness areas, high priority
should be given to lands whose selection
would enlarge and connect existing protected
areas (e.g., national parks, wildlife refuges, and
Forest Service wilderness areas) and thus
enhance the viability of animal and plant pop-
ulations (Newmark 1985, Salwasser et al. 1987,
Noss 1992, Grumbine 1994). By themselves,
BLM wilderness areas in Utah clearly cannot
satisfy the huge area requirements noted above
as requisite for maintaining viable populations
of large carnivores. However, when linked to
other public lands (e.g., Utahs national parks,
and wilderness areas in other states), BLM
wilderness in Utah can be a key component of
strategies for long-term presei"vation of biolog-
ical diversity.
Other high-priority areas are those which,
alone or together with other protected areas,
encompass entire watersheds. In addition to
affording direct benefits to humans, watershed
protection is the most effective means of con-
sei"ving the aquatic and riparian communities
that account for a disproportionate fraction of
both species diversity and endangered and
threatened species in arid western North
America (Miller 1961, Minckley and Deacon
1968, 1990, Holden et al. 1974, Johnson et al.
1977, Cross 1985, Knopf 1985, Moyle and
Williams 1990). Moreover, since populations
of riparian species are usually isolated from
similar communities in other drainage systems,
species losses from these environments are
not easily remedied b>' natural recolonization.
A 3rd priority in selecting wilderness sites
is land that fomis or helps to complete the pro-
tection of entire elevational gradients, for
example, in isolated mountain ranges of the
Great Basin. Scant attention paid to consemng
these gradients in the past is evident in the
restriction of most national parks and wilder-
ness areas in western North America to higher
elevation sites. Designation of wilderness in
comparatively low elevation BLM lands would
afford protection to regions of greatest species
richness for man\' organisms (e.g., mammals,
birds, amphibians, insects, and trees) whose
diversity generally declines with elevation
throughout much of western North America
(Harris 1984, Ste^'ens 1992).
Optimal Design Goals 1
If BLM wilderness areas are to contribute
substantialK' to the preser\'ation of biodiver-
sitv in Utah, then site selection must take into
1996]
Wilderness Selection for BiODiVERsm-
99
Buffer Zone
account tlie 3 general goals outlined above. H ^'-^ wilderness
Ideally, BLM wilderness lands should form an <&% ^ . o .., ,.
fijif;: Forest Service Wilderness
interconnected core zone of roadless lands
w hen combined with otlier federal wilderness H National Park service
areas, national and state parks, and wildlife
refuges (Fig. 2). Special attention should be Public Lands
given to linking roadless lands so as to pre-
clude further fragmentation of natural habitat.
Inagmentation, or the transformation of an
unbroken block of natural habitat into a num-
ber of smaller patches separated by altered
habitats, reduces population sizes, increases
their isolation, and threatens their long-term
viability. It is one of the greatest threats to bio-
logical dixersit)' worldwide (Wilcox and Mur-
phy 1985, Wilcove et al. 1986, Saunders et al. <
1991). Across diverse habitats, there are numer-
ous examples of species extinctions precipi- 'KWi;;-^-;.— core zone
tated by both natural and human-induced
habitat fragmentation (e.g., Brown 1971, Ter-
borgh and Winter 1980, Diamond 1984, Fig. 2. An example of a preferred arrangement of
Heaney 1984, Patterson 1984, Newmark 1987, wilderness and multiple-use federal and state lands to
1991, 1995, Case and Codv 1988, Soule et al. conserve biological diversity. Wilderness areas adminis-
irvoo n 1 L ^ inm\ ' teied bv the Bureau of Land Management, Forest Service,
1988, Bolger et al. 1991 . m ^- i d i c j u- i i ^^/i uf c
\ 1 rr National rark bervice, and Fish and Wikllile Service
Adjacent multiple-use lands can buffer ,]^o^,i,i fo^m a contiguous core zone in which the most
human impacts on biological diversity within extinction-prone species in Utah can be protected. Multi-
wilderness areas. Such lands can be expected pie-use lands can effectively buffer this core zone and
to pro\'ide marginal habitat for tlie manv species P'^^'de additional marginal habitat to species that are pri-
.1 . 1. • i 1 • -1 1. ■ I.- marilv restricted to roadless areas.
tliat are restricted prnnanly to more pristme
wilderness regions. Thus, proposed wilder-
ness areas surrounded bv public lands should
receive high priority for protection. their genes move about only through the pro-
cesses of seed dispersal and pollen transport.
Therefore, it is not surprising that many plants
have narrowly restricted ranges, are locally
adapted to conditions within those ranges, and
are isolated, often by great distances, from
other sites where similar conditions prevail.
Although locally endemic plants can often be
relatively abundant inside their ranges, their
populations are easily jeopardized by habitat
alteration (e.g., by all-terrain vehicles) within
their narrow distributions. Of Utah's approxi-
mately 2600 plant species and 400 named
varieties (Albee et al. 1988, Welsh et al. 1993),
about 180 (or 7% of species) are currently clas-
sified by federal or state agencies as endan-
gered, threatened, or sensitive. A majority of
these (133, or —74%) definitely or probably
occur on BLM lands (Atwood et al. 1991), and
a substantial subset of the classified species
are narrow endemics.
Shultz (1993) provides a useful summary of
endemism in the Utah flora. Approximately
240 species, or 10% of all Utah plant species, are
endemic to the state. This rate of endemism.
E.XAMPLES OF Rare and
Endemic Species
The design advocated above is based
largely on conservation strategies for preserv-
ing wide-ranging vertebrate species. Although
such strategies can help to ensure the long-
term viability of most species within a given
region, exclusive reliance on such approaches
may well overlook and endanger many locally
isolated, rare, and endemic plants and animals.
We cannot give a comprehensive treatment of
this subject here, but we discuss 3 ta.xonomic
groups of organisms for which especially high
rates of endemism or existing threats to iso-
lated populations present particular manage-
ment dilemmas that should be taken into
account in wilderness decisions. In most cases,
specific habitats must be protected to assure
the presei'vation of these species.
Plants of Special Concern
Unlike the wide-ranging animals discussed
above, plants occupy fixed positions; they and
100
Great Basin Natufl\list
[Volume 56
the percentage of the flora considered for hst-
ing as threatened or endangered, and the per-
centage of rare species in the flora are among
the highest in the continental United States.
The vast majority (86%) of Utah endemics reside
in arid and semiarid regions of the state, and
90% are edaphicalK' restricted to fine-textured
and/or high pH substrates (limestone, clay, silt,
mudstone, and shale) that magnify drought
stress. Plant distributions generally appear to
respond more to edaphic, topographic, and
geologic features of the environment when
drought is a factor (Stebbins 1952). Because
most endemics live in close proximity to mor-
phologically similar species (Albee et al. 1988),
these species appear to be mainly neoendemics
that have evolved since the last glacial maxi-
mum (18,000 yi's BP), or in the Bonneville basin
during the past 10,000 >ts.
Geographically, endemism of Utah plants is
highest in the Canyonlands Phytogeographic
Section of the Colorado Plateau Division of
the Intermountain Region (Cronquist et al.
1972, Fig. 3 modified from Shultz et al. 1987).
An unusual diversity of substrates occurs here,
and these substrates are more apt to be exposed,
rather than coxered with alhnium as in other
areas of semiarid Utah (Welsh et al. 1993). Thus,
fully 50% of Utah's 240 rare and endemic
plant species occur on the Colorado Plateau,
whereas just 15% occur in the Great Basin,
11% in the Mojave Desert, and 10% in the
Uinta Desert (Welsh 1978, Shultz 1993).
About half of Utah's endemics belong to just 5
genera that are both common and physiologi-
cally adapted to aridity (total Utah species and
percent endemics, in parentheses): Astragalus,
Fabaceae (114, 36.8%), Penstetnon, Scrophulari-
aceae (106, 26.4%), Cnjptantha, Boraginaceae
(61, 36.1%), Eriogoniim, Polygonaceae (60,
23.3%), and Erigeron, Asteraceae (54, 24.1%;
Welsh et al. 1975, Welsh 1978, Shultz 1993).
Because most of the state's endemic plants
are restricted to particular geologic formations,
and because multiple endemics often occur on
the same formation, groups of endemics gen-
erally can be protected simultaneously by safe-
guarding those soil formations and surround-
ing areas. Two regions where large nimibers of
endemics stand to benefit from wilderness
protection of BLM lands are the Uinta Basin
and the San Rafael Swell and surrounding San
Rafael Desert (Fig. 3, Table 1; M. Windham
personal communication). No fewer than 15
plant species are endemic to the region in and
around the proposed wilderness area (PWA)
near the White River south of Vernal (UWC
1990), and most of these are confined to the
Parachute and Evacuation Creek members of
the Green River Shale formation. Another
dozen endemics occur in a diversity of habi-
tats in and around the San Rafael Swell. Here
the most important habitat is a beige (rather
than red) Moenkopi formation, spatially iso-
lated from other Moenkopi outcrops and un-
usual in its soil chemistiy. A few endemics also
occur on the younger Carmel and Summer-
ville formations surrounding the core of the
swell, especially between Muddy Creek and
Crack Canyon (S. Welsh personal communica-
tion). Wilderness designation in these 2 regions
(the San Rafael PWA and the White River
PWA of the Uinta Basin [Fig. 3]; see UWC
1990) could afford significant protection to some
of Utah's endemic plants. South and east of
the San Rafael, in the Dirty Dexil PWA (UWC
1990), are the distinctixe flora of the Orange
Cliffs region (Fig. 3) and some additional nar-
row endemics deserving protection in the Main
and South forks of Happv Canvon (Shultz et
al. 1987).
The Moenkopi formation is also important
as a substrate for endemics elsewhere in semi-
arid Utah. Two federalK' listed endangered
species, Arctomecon limnilis (the dwarf bear-
claw poppy) and Pediocactiis sileri (a cactus),
and several other species are endemic to par-
ticular Moenkopi outcrops in southwestern
Utah. Wherever possible, the boundaries of
wilderness areas and other protected areas
should encompass these specialized habitats.
Bees and Wasps in the
San Rafael Desert
Because of their capacit) for directed mo\'e-
ments, animals are less likely than plants to
exhibit high rates of endemism. Nexertheless,
since insects often tend to be host- or habitat-
specific (e.g., in pollinators, herbixores, or sub-
strate-specific ground nesters), endemism can
often be high in insect taxa. Bees and wasps
(order H\menoptera) are examples of such
insects. Here, as elsewhere, bees and preda-
tor)' wasps are especialh' di\'erse in arid regions
(Michener 1979). The state supports a mini-
mum of 950 species of native bees (roughly
25% of the total number of species known
1996]
Wilderness Selection eor Biodin i.Ksin
101
SweU
San Rafael Desert
^Orange CUffe
Fig. 3. Satellite image of Utah showing the positions of the San Rafael Swell, the San Rafael Desert, and tlic Orange
Cliffs, all within the Canyonlands Phytogeographic Section, ontlined in bold. The arrow in the Uinta Basin shows the
approximate position of the White River PWA (Utah Wilderness Coalition 1990).
from America north of Mexico), and 50 of the
Utah species are currently inidescribed (T.
Griswold, K Parker, and V. Tepedino personal
communication). Many areas, especially in the
southern part of the state, have not been
explored intensively and undoubtedly harbor
many additional undescribed species.
Bees and plants often show comparable geo-
graphic patterns in diversity and endemism
(Neff and Simpson 1993), and many of the areas
currently under consideration for wilderness
designation in Utah are centers of endemism
for both groups. Although we lack extensive in-
formation on bees of the Canyonlands Section
(Fig. 3), where endemism is highest for plants
(see above), intensive collecting in that small
part known as the San Rafael Desert has
yielded a total of 316 species of bees, 42 of
102
Great Basin Natur.\list
[Volume 56
Table 1. Plants endeinic to the 2 areas with the higliest eiuleinisiii on Utah BLM lands.
Endemics of tlie sontheni Uinta Basi
Endei
the San Haiael Swel
Aqtiilegia barnebtji Miniz (Ranunciilaceae)
Asfragdiiis eqiiisolensis N'eese 6c Welsh (Fahaceae)
A. hainiltonii C. Porter
A. hitosii.s Jones
A. saiiriniLs Barnehx
Cirsiiiin hanwhiji Johnst. (^Asteraceae)
Cryptcmtlid hantchiji Johnst. ( Boraginaceae)
C grdliainii Johnst.
CtjDioptcri.s (liiclu'siicn.sis Jones (Apiaceae)
Pensteinoii floiccifiii Neese & \M'lsh
(Scrophulariaceae)
P. goodhcliii .\. Holmgren
P. grahainii Keck
SchoencniinlH' argilhicea (\\'elsh & Atwood)
Rollins (Brassicaceae)
S. .suff'ruti'.sceus (Rollins) Welsh 6c ChatterK'
Sclerocactiis glaitciis (K. Schnm.) L. Benson
Astragalus rafaclciisis Jones (Fabaceae)
Cnjpfaiitha crciitzfclclii Welsh ( Boraginaceae)
C. Johnstoiui Higgins
C. joiu'siaiui (Pa\son) Pa\son
Erigeroii inaquirci Cronquist (Asteraceae)
Loinatiiimjiinceiiiii Banieh\ 6c N. Holmgren (Apiaceae)
Lijgoclcsmia entrada Welsh 6c (iooilrieh (Asteraceae)
Pcdiocactiis dcspaiiiii Welsh 6c C.ot)drieh (Cactaceae)
Pcitsteinoit inarnisii (Keck) \. Holmgren (Scropluilariaceae)
ScliOi'iicraiidH' harncbt/i (Welsh 6c .\tA\ood) Rollins (Brassicaceae)
'ndinuiii dtninpsoiiii Atwood 6c Welsh (Portiilacaceae)
Tt>uiisciidi(i aprica Welsh 6c Re\eal (Asteraceae)
which are presentK undeseribed (T. Griswold,
E Paikei; and \^ Tepedino personal conniuuii-
cation). Thus, 33% of the state's total species
count, and 84% of Utah s undeseribed (but
catalogued) species, are endemic to a region
comprising just 2.0% ot the states land area.
Fin-thermore, a signiticant portion of this tauna
(24%) occin\s onI\' on the Colorado Plateau.
The remainder of the Cainonlands Ph\togeo-
graphic Section, in which the San Rafael Desert
is embedded, is likeK" to be equalK di\ erse
and to ha\ e as man> new species.
Other hymenopteran groups, such as the
aculeate \\asps, also are highly di\ erse in the
San Rafael Desert (T. Gris\\ old, E Parker, and
V Tepedino personal connnimication). For ex-
ample, with a total of 22 species there, the cir-
cinnglobal genus Fhildntluia is more di\"erse in
the San Rafael Desert than an\A\'here else in
North America, and probably the world. These
predatoiy "digger wasps " nest in the soil and
ma\ ha\ e di\"ersified in response to the \ aried
substrates present in this desert. ClearK, des-
ignation of wilderness in the San Rafael region
(see UWC 1990) could afford significant pro-
tection to an area of \er>- high endemism and
di\ersity for the order H\nienoptera.
Bees and wasps are among the most benefi-
cial insects. Predaton' and parasitic wasps help
to control populations of pest species (e.g.,
grasshoppers, aphids, etc.) below outbreak
densities. An estimated 67% of flowering plants
depend on insects (primariK' bees) for pollen
transfer and sexual reproduction (Axlerod
1960), and the welfare of nian\ plant species
in semiarid Utah assuredh depends on their
relationships with bees. Eor example, a rare
species of Pcrdita, found in Utah only at the
BeeHi\e Dome site southeast of St. George,
pollinates the rare and endangered dwarf bear-
claw popp\' (^. Tepedino personal commimica-
tion). Bees that ha\e specialized b\ collecting
pollen onl\ from flow ers of a particular plant
family, or exen from a single genus within a
famih; are termed oligoleges. Such bees tend
to be most common in arid regions (Neft and
Simpson 1993) and generalK" are regarded as
being closely adapted to the phenolog>' and
floral traits of the plants on \\ hich the)' spe-
cialize. Such adaptations tend to make them
superior pollinators. Scjuash bees and squash
flowers are examples of such a co-adapted pair
in the Americas (Tepedino 1981). Some oligo-
leges ma> one da\' proxe to be useful as crop
pollinators. The legume specialist Osniia san-
rafachic. a nati\ e ot the San Rafael Desert, has
been inxestigated as a potential pollinator of
alfalfa {Mcdicdgo sativa L.), an important for-
age crop (Parker 1985, 1986). Man> of the
species of the San Rafael Desert appear to be
oligoleges. A brief list of some of the unde-
seribed and recentK" described bee species
and their host plants is pro\ ided in Table 2.
These entries were chosen only to illustrate
the \ariety of plant taxa upon which nati\e
bees specialize.
Nati\ e and Endemic Fishes
Freshwater ecosxstems are natinal habitat
"islands ; as sutli. thcii- long-tcnn isolation b\
1996]
WiiJ^EKNEss Selection for Biodivkksitv
103
Tahi.K 2. Pollen piffereiices for represcnlativf oli.ujolL'ctif liccs in llic San lialacl Dl'S(m1 (data Cioni 'I'. Crisvvold, F Farkt-r
and V. Tepedinc) personal connnunication).
Plant family
Plant genus/species
Bee species
Asteraceae
Boraginaceae
Eiipliorbiaceae
Fabaceae
Loasaceae
Onagraceae
Papaveraceae
Polenioniaeeae
Scn)})liu!ariaceae
Hcliiiiitliiis (inoinolu.s
W'l/ctliid Kccil specie
intei'vening terrestrial liabitats, or by unsuitable
aquatie habitats, often promotes loeal speeial-
ization, evoliitionaiy diversification, and endem-
ism in aquatic organisms. Seven centers of
endemism are recognized for fishes of western
North America (Miller 1959), and Utah includes
substantial portions of 2 of these centers, the
Bonneville Basin and the Colorado River
Basin. Collectively, 28 fish species are native
to these basins (Smith 1978), and 27 are extant.
Because of their limited distributions, en-
demic species are easily endangered by both
habitat alterations and introductions of nonna-
tive competitors and predators. Seven species
and subspecies from the Bonneville and Col-
orado basins are now federally listed as endan-
gered (U.S. Fish and Wildlife Ser-vice 1993). A
further 11 species and subspecies are consid-
ered by fisheiy specialists to be endangered,
threatened, or of special concern in Utah (War-
ren and Burr 1994). The decline of native
fishes has been associated with both water-
shed development (e.g., reservoirs, irrigation
diversions, channelization, floodplain drainage)
and the introduction of alien species.
Conservation of endemic fish populations
has been especially successful when much of
the watershed has been protected (Williams
1991), but adherence to strict legal definitions
of wilderness often precludes such wide-
spread protection. In Utah, opportunities for
protecting entire watersheds are limited to
relatively small drainage systems extending
from stream headwaters in mountain ranges of
the Bonneville Basin to diy or saline lake beds
at lower elevations. A particularly important
case is in the Deep Creek Range, where the
Bonneville cutthroat ti'out {Oncorhtjnclms clarki
iitali), once thought to be extinct (Behnke
1992), survives in populations in Trout Creek
and Birch Creek within the Deep Creek PWA
(UWC 1990).
Where protection of whole watersheds is
not possible, wilderness that includes key habi-
tats may help to stabilize declining populations
of native fishes, preclude new listings and draft-
ings of recovery plans, and promote recoveries
and delistings. This should be the case most
often for fishes living in headwater streams
protected by natural and artificial downstream
barriers from unintended invasions of alien
cold-water species. For example, habitat in the
upper Book Cliffs-Desolation Canyon PWA
may support the Colorado River cutthroat trout
{Oncorhyncluis clarki plenriticu.s), considered
the rarest of the cutthroat taxa (Behnke and
Zani 1976) and federally listed as a categon^ 2
species (Kerchner 1995). Although the region
has not been surveyed for this subspecies,
native populations occur in streams entering
the Duschesne River from the north (Shiozawa
and Evans 1994) and have recently been found
in streams of the western Book Cliffs, closer to
Price and Soldier Summit (Shiozawa and Evans
unpublished data). Given these obsei-vations, it
is likely that streams flowing into the Book
Cliffs-Desolation Canyon PWA will also con-
tain this subspecies.
In relatively large downstream systems
(secondary and tertiaiy streams), key habitats
include floodplain wetlands, among the first
habitats to be lost due to human activities.
Although wetlands have been viewed tradi-
tionally either as breeding sources for insect
104
Great Basin Naturalist
[Volume 56
pests or as waterfowl production sites, periodic
or continuous connection to rivers renders
them important appendages to lotic systems.
Densities of aquatic invertebrates are signifi-
cantly higher in wetlands than in main river
channels, over 100-fold in some cases (Wolz
and Shiozawa 1995, Mabey and Shiozawa
unpublished data). Floodplain wetlands can
therefore serve as important nurseiy grounds
for laival and immature native fishes.
The loss of wetlands may be a significant
factor endangering sexeral native fishes in the
Colorado River (Tyus and Karp 1989). Fishes
native to the larger streams and rivers of the
Colorado River Basin are predominantly min-
nows (Cyprinidae) and suckers (Catostomidae)
that have evolved in isolation, are adapted to
unique local conditions of this drainage (e.g.,
heav\' silt loads and wide fluctuations in dis-
charge and temperature), and are the most
moiphologicalK' distinct fishes in North Amer-
ica (Hubbs 1940, 1941, Deacon and Minckley
1974, Minckley et al. 1986). Four of these
native species, the Colorado squawfish {Pty-
chocheilus lucius), the humpback chub {Gila
cijpha), the bonytail chub {Gila elegans), and
the razorback sucker {Xyrauchen texamis), are
now federally listed as endangered. The decline
of both the bluehead sucker {Catostomus [Pan-
tosfeii.s] discobolus) and the flannelmouth sucker
{Catostomus latipinnis) within the main stems
of the Colorado and Green rivers may result in
their listings as threatened, especially if popu-
lations in tributaiy streams are not stabilized.
Several of these species occur in areas under
consideration for wilderness status. Both the
Price River, in the Book Cliffs-Desolation Can-
yon PWA, and the San Rafael River, in the San
Rafael PWA, have populations of roundtail
chub, flannelmoudi sucker, and bluehead sucker
Bluehead sucker are also known from the
Dirty Devil and Muddy Creek drainages (Smith
1966), and both flannelmouth sucker and round-
tail chub are likely to occur there. Wilderness
designation could broaden the protected ranges
of several of these species by stabilizing wet-
land habitats in the Dirty Devil, San Rafael,
and Book Cliffs-Desolation Canyon PWAs.
Although the Virgin River drainage is also
part of the Colorado River Basin, it has a
unique fish fauna that appears to have evoKed
in isolation from populations in other parts of
the basin. The Virgin River spinedace {Lepi-
domeda mollispinus). the woundfin {Plagoptenis
argentissimus), and the Virgin River chub {Gila
robusta seminuda) are endemic to this system.
Two additional species, the flannelmouth sucker
and the desert sucker {Catostomus clarki),
have evolved very slender caudal peduncles,
possibly as a response to occasional high flows
in the Virgin River (Smith 1966).
The health of this unique fish fauna already
is cause for concern. Two of the endemics, the
woundfin and the Virgin River chub, are feder-
ally listed as endangered. Although the desert
sucker occurs in Arizona, Nevada, and New
Mexico, this species merits special concern in
Utah (Utah Division of Wildlife Resources
[UDWR] 1992), where it is limited to the Virgin
Ri\'er drainage. Loss of either this species or
the flannelmouth sucker from the Virgin River
system would eliminate only a subset of their
existing populations and is unlikely to move
either species to endangered status. However,
the uniqueness of these populations (Smith
1966) may warrant their designation as sepa-
rate subspecies. This, toge flier wifli the concern
now e\'idenced for the flannelmouth sucker
throughout its range, could easily translate into
candidacy for listing if existing populations are
not protected.
Concern for native fishes of the Virgin River
drainage has already constrained water devel-
opment ill Washington Count); Utali. An>' actions
that would help presene the integrit}' of ripar-
ian habitat and stream channels would also
reduce stress for these fishes. Since the integ-
rit\' of riparian habitats is best maintained over
large areas, wilderness designation in PWAs of
the Beaver Dam slope and the greater Zion
area would sei"ve this purpose.
Finally, protection of Utah s rare and en-
dangered fishes would likeK also afford signif-
icant protection to other aquatic organisms,
for example, Utah's diverse communities of
aquatic insects. Reciprocally, the maintenance
of high species diversity in stream insect com-
munities is critical to assuring a continuous
food supply to fishes in rivers with wide sea-
sonal and annual fluctuations in flow rates.
Mayflies (Ephemeroptera) are among the best-
studied stream insects in Utah, and 16-18
genera (22-24 species) are known from warm
water tributaries of the Colorado Rixer sxsteni
(G. Edmunds personal commimication). Con-
struction of reservoirs on these rivers has
iilready inundated many river miles and altered
flow rates, sediment loads, and downstream
1996]
Wilderness Selection for Biodiversity
105
teniperutures. Mayflies mid other aquatic insects
are highly sensitive to all these variables.
Unnatinalh' constant temperatures in tailwaters
beneath dams can lead to depauperate com-
munities of ma) flies and other stream insects,
for example, below Flaming Gorge Reservoir
(Edmunds 1994, 1995). (Four mayfly genera
from this area of extremely high natural diver-
sity have not been collected since the dam was
built.) Habitats rich in mayflies and other
aquatic insects, and most in need of protection
from future impoundments, include the Green
River from the Colorado border to Ouray,
Utah, and the Colorado River from the Colo-
rado border to Moab, Utah. Relatively warm
sections of the Duchesne, Uintah, White,
Escalante, Virgin, and Santa Clara rivers
would also be sensitive to manipulations of
stream flows.
Examples of Biologically
Important Sites on BLM Lands
The floras and faunas in different parts of
Utah have unique evolutionaiy histories deter-
mined by the geography and topography of
the lands they inhabit. In this section, we dis-
cuss 4 such sites in the context of important
scientific criteria (outlined above) for wilder-
ness site selection. We also review various sci-
entific and educational values of these same
sites.
Book Cliffs and the Tavaputs Plateau
For several reasons, the Book Cliffs and
Tavaputs Plateau areas, along both sides of the
Green River, are critical for the long-term con-
servation of biological diversity in Utah. This
region contains some of the largest remaining
roadless areas on BLM lands in Utah (Fig. 1)
and therefore provides important habitat for
sensitive species with large area requirements.
It includes broad elevational gradients with
the potential to protect a wide range of natural
communities and to maintain crucial routes
for seasonal wildlife migration between high
and low elevation. Furthermore, it constitutes
a vital dispersal coiridor linking the Uinta moun-
tains to the north and the Colorado Plateau to
the south.
Because of both the high habitat diversity
and the central location of the Book Cliffs-
Tavaputs region, the biota is unusually diverse
and compositionally unique, and includes many
species at their distributional limits. Among
reptiles and amphibians, for example, the Great
Basin spadefoot toad {Scaphiojms intennon-
Uinus), the western whiptail lizard {Cnemi-
dopJwrm fi^ris), and possibly the rubber boa
{Charina bottae) reach their eastern distribu-
tional limits here. Three additional species,
the longnose leopard lizard {Gmnhclia wis-
lizenii), the collared lizard {Crotaplujtus col-
laris), and possibly the plateau striped whip-
tail {Cnemidoplwrns velox) are represented
here by "edge" populations at the periphery of
their respective ranges. Other species, such as
the northern leopard frog {Rana pipiens), east-
ern fence lizard {Sceloporus undnlatiis). Great
Plains ratsnake {Elephe guttata), and the Utah
milk snake {Lampropeltis triangidum), have
their westernmost limits in this region (Steb-
bins 1985, unpublished BYU museum records).
While none of these species is federally listed
as threatened or endangered, a few are so
listed by the state (UDWR 1992). Moreover,
geographically peripheral populations such as
these are particularly important as dynamic
foci of evolutionaiy change (e.g.. Brown 1995,
Lesica and Allendorf 1995).
The Book Cliffs-Tavaputs region also sup-
ports a rich mammalian fiiuna. Although our
knowledge is far from complete, the area con-
tains at least 62 native species, including a rel-
atively stable population of black bear {Ursus
americanus; H. Black personal communication).
Recent fieldwork has resulted in records for 6
species previously unreported from the region
(D. Rogers personal communication); these
include Merriam's shrew {Sorex merriami),
dwarf shrew (S. nanus), water shrew (S. palus-
ths), big fi-ee-tailed bat {Nijctinomops macrotis),
northern flying squirrel {Glaucomys sabrimis),
and western jumping mouse {Zapiis princeps).
Of these species, S. merriami, S. nanus, and N.
macrotis appear to be rare throughout their
known distributions. More fieldwork is likely
to pioduce additional records for this region.
Isolated Desert Mountain Ranges
The isolated mountain ranges in Utah's Great
Basin and Colorado Deserts are extremely
important biologically because of their role in
maintaining critical ecological and evolution-
ary processes. Because of their broad eleva-
tional gradients, extending from high peaks to
desert valley floors, these ranges support a
wider variety of habitats and a greater diver-
sity of species than do areas of comparable
106
Great Basin Naturalist
[Volume 56
size but less elevational relief. This eharacter-
istic also enables them to support the seasonal
migrations of animals ranging from large ungu-
lates to small passerine birds. Furthermore,
these mountain ranges have outstanding sei-
entific value because they represent cool and
mesic habitat islands in an otherwise warm,
arid landscape. Their natural communities have
developed through intermittent periods of
extreme isolation (Grayson 1993). Coupled with
the great geological diversity of the region,
this isolation has led to the formation of
unique plant assemblages, often including rare
local endemics (Albee et al. 1988, Welsh et al.
1993). By illustrating how populations and
communities of habitat islands are modified
through colonization and extinction, these
mountain ranges have played a major role in
the development of theories of geographical
ecology and biogeography (Brown 1971, 1995,
Grayson 1993, E. Rickart in preparation).
Portions of several isolated mountain ranges
are represented within PWAs on BLM lands
(UWC 1990). Such ranges include the Henry
Mountains of the Colorado Plateau and the
Deep Creek, Fish Springs, House, and New-
foundland ranges of Utah's west deserts (Fig.
1). As the most isolated range in Utah, the
Newfoundland Mountains in Box Elder County
are especially distinctive. At 2129 m above sea
level, Desert Peak and a considerable area of
surrounding uplands would have existed as an
island throughout the histoiy of ancient Lake
Bonneville. Currently, the range forms a 154 +
km- island of arid to semiarid vegetation
immersed in a salt playa sea. No doubt salt
marshes have covered the present salt flats
periodically as the lake has advanced or
receded in response to glacial and interglacial
climates. The range has therefore been an eco-
logical island throughout nearly 2 million years
of Pleistocene and Quateniaiy time. Given such
long isolation, these mountains have much to
teach scientists about the persistence, local
extinction, vagility, and evolutionaiy dynamics
of a variety of animal and plant species that
either live there now or have lived there in the
past. In Utah and elsewhere in tlie inteniioun-
tain region, knowledge of these topics will be
important in the future as land managers tiy to
anticipate plant and animal responses to the
increasing fragmentation and isolation of nat-
ural habitats within the human-dominated
landscape (Brown 1995).
Mojave Desert in Southwestern Utah
Washington County includes Utah's only
representative of the Mojave Desert, a warm
desert commonly recognized by biogeogra-
phers as lying between the Great Basin Desert
to the north and the Sonoran Desert to the
south (Shreve 1942, Jaeger 1957, Rowlands et al.
1982, MacMahon 1986). The Mojave Desert is
physically part of the Basin and Range Geo-
logical Province, but it is characterized by rel-
atively low elevation over most of its area (600
to 1500 m above sea level) and by both limited
precipitation (100-275 mm annually in most
places) and warm summers (35°-40°C mean
maxima for July; see MacMahon 1986). The
uniqueness of the physical environment of the
Mojave is reflected in its biota. Characteristic
plants include the Joshua tree {Yucca brevifolia),
creosote bush {Larrea tridentoto), white bur-
sage {Ambrosia diimosa), brittle bush {Encelia
farinosa), and several species of saltbush {Atri-
plex). Of these, the Joshua tree can be consid-
ered endemic, and if the distribution of this
species is used to define the boundaries of the
Mojave Desert, then the desert covers a sub-
stantial portion of southeastern California, the
southern cone of Nevada, the northwestern
and west central parts of Arizona, and the ex-
treme southwestern corner of Utah.
Judicious designation of new wilderness areas
in this corner of the state could help to safe-
guard the many components of Utah's biologi-
cal diversity that are endemic to the Mojave
Desert and the associated Virgin Mountains of
northwestern Arizona and adjacent Nevada.
Figure 4 details land ownership in this region
of Washington County. Because so much of
this land is already in the public domain, there
is opportunity' for biodiversity conservation
with minimal disruption of economic activity.
Protected areas include Zion National Park, a
sul)stantial wilderness in the Pine Valley
Mountains of the Dixie National Forest (no. 1
in Fig. 4), the Upper Virgin River Desert W'ild-
life Management Area (or DWMA, a reserve
for the desert tortoise, Gophcrus a
:=:3 -2-. i a
- ^ a, a;
1; _£ .2 U
a _2 Z ^
OJ . o >
O -73 N 0
5 CQ
o ^ OJ <;
-f=
^
^ Q
in
'•^
^ u-
u
Hi
a 5^
>
;5
o 2
ii
"
5. ^
1^
n ,
t*-
Oj
E a
^
v:
(; w.
*;
O
c ji:
>.
2^
p
^ ,
_j
t£
t;
.5J (M
^ ~ — '
•£
p
« <
"3
— 1
^iS
S
,^
u
-r >
.^ — ,
-
5 C
-t-
=
-S
Ji t
=
>_/
_c
i- """
c
^
- C
-J
~ ir.
SI
_
^' >-
X
-
■-
St c<
i\
»*
_^-c
— Ph
1^
^
. "" r^
ir
f^
1
r-
■^
b ci
r^
'.fch
^r?
«
O
>
^^
rf^
s
2
^—
O "t-
-—
y?
^
E^z
^
1; 1^
(J
5
^
_aj
aj
^
"5
EC
sr
hr
j_
> .^
.^
Uh
^
ij .*-
108
Great Basin Naturalist
[Volume 56
too high in elevation and/or too far to the
northeast to include many Mojave Desert
species. The Upper Virgin River DWMA will
protect lower elevation communities and will
include some Mojave Desert taxa. However,
many Mojave Desert species in Utah do not
extend northeast of the Beaver Dam Moun-
tains, and existing protected areas on the
Beaver Dam slope are relatively small and iso-
lated from each other (Fig. 4). By virtue of both
size and location, 2 PWAs, the Beaver Dam
Wash and Joshua Tree units (nos. 3a and 3b,
respectively, in Fig. 4; see UWC 1990), could
make important contributions to biodiversity
consei-vation in Utah. Together these 2 units
cover a range of elevations, include several
distinctive plant communities not represented
in the Upper Virgin River DWMA, and are
close enough to one another and to the exist-
ing protected areas to serve as stepping stones
for animal movement.
We illustrate the conservation value of
these 2 PWAs dirough an example. The heipeto-
fauna of the Mojave Desert includes 3 anu-
rans, 1 tortoise, 16 lizards, 18 snakes, and about
28 additional species whose distributions are
peripheral but extend into this desert along
one of its edges (Stewart 1994). The portion of
this fauna ranging into Utah includes 2 anu-
rans, the turtle, and 13 squamates (5 lizards
and 8 snakes). Their distributions across exist-
ing or proposed protected areas are summa-
rized in Table 3. Of this total, the relict leop-
ard frog {Rana onca) apparently is extinct in
Utah (Platz 1984, Jennings and Hayes 1994)
and therefore absent from all existing and pro-
posed protected areas in Washington County
The other anuran confined to this part of Utah
is the southwestern toad {Bnfo microscaphus).
It is known to exist with certainty in several
areas and is likely widespread throughout the
region where appropriate acjuatic habitats
exist (Table 3).
The desert tortoise {Gophcnis agassizii) has
been studied extensively over the past decade
and intermittently for a much longer period of
time (Woodbury and Hardy 1948, Bur\' and
Germano 1994, Grover and DeRilco 1995).
While Utah populations have apparently de-
clined in the Beaver Dam slope area, they
persist at high densities north of St. George
(data summarized in Bury and Germano 1994)
and are now protected in the Virgin River
DWMA. Protection of the proposed Joshua
Tree and Beaver Dam Wash wilderness areas
would thus provide an economical way to aug-
ment consei-vation of tortoise populations con-
fined to the south-facing slopes of the Beaver
Dam Mountains.
Of the 13 squamate reptiles listed in Table 3,
nine are confined to either the Mojave habitats
proper (sites 3a, 3b, 4, and 5 in Fig. 4) or to
tliese sites plus the Upper Virgin River DWMA
(sites 2a and 2b in Fig. 4). Four species have
more extensixe distributions because they are
also recorded from Zion National Park. Among
the 9 squamates with restricted distributions,
the lizards Helodenna suspecfum and Xantusia
vigilis and the snakes Crotalus cerastes and
Leptotyphlops humilis may occur at all 5 Mojave
sites, although this needs to be confirmed
through additional fieldwork. Xantusia vigilis
also occurs further east in isolated populations
in Garfield and San Juan counties, and previ-
ous molecular studies by Bezy and Sites (1987)
show deep genetic divisions among many iso-
lated populations. Many of these isolates would
qualify as full species, following the criteria of
Davis and Nixon (1992), but the specific status
of the isolated Utah populations remains un-
known. The lizard Callisaurus draconoides
occurs with certainty in the upper Virgin River
DWMA (in Snow Canyon State Park), Beaver
Dam Wash PWA, and Lytic Ranch Preserve
(sites 2a, 3a, and 5 in Fig. 4). The iguana {Dip-
sosaiiriis dorsalis) is known confidenth- from
onh' the lower Beaver Dam Wash PWA,
although it may occur at low densities in the
other 3 Mojave sites. Among the snakes, Cro-
talus scutulatus is confined to the 4 strict
Mojave Desert areas, and C. mitchellii is known
with certaint)' from onh' the higher elevation
Moja\'e sites (3b and 4, although the other 2
locations are possible). Based on a new snake
record for Utah, Phyllorhynchus decurtatus is
known from a specimen (BYU 45605) taken on
11 July 1995, ca 1.5 mi N of the Utah-Arizona
border along the Beaver Dam slope road. Based
on this record, the species likeK occurs in the
Beaxer Dam Wash and Joshua Tree areas (3a
and 3b), w hich are similar in \egetative struc-
ture to the collecting site, and possibly at the
other Mojaxe Desert sites as well. Regardless
of exact distributions, all 9 squamate species
with the most restricted distributions would
benefit by wilderness designation of the pro-
posed Beaver Dam Wash and Joshua Tree units
{IJW'C 1990); and for 7 species (C. draconoides.
1996] Wilderness Selection for Biodiversity 109
Table 3. Distribution of amphibians and reptiles restricted to southwestern Utah, relative to existing protected areas
and Beaver Dam Wash and Joshua Tree units of proposed BLM wilderness iucluded in H.R. 1500. The areas numbered
are shown in Figure 4''. The proposed Red Mountain and Cottonwood Canyon wilderness areas (UWC 1990) are not
illustrated because thc\- are lariielx (Red Mountain) or entireK' (Cottonwood ('anyon) contained within the Upper Virgin
River DWMA.
Bea\'cr Dam
Dixie N.E
Upper Virgin
Wash
Joshua Tree
Bea\'cr Dam
L>tle
Zion National
Wilderness
RiN'cr DWMA
Wildcnicss
Wilderness
Wilderness
Ranch
Taxon
Park
(1)
(2A. 2B)
(3A)
(3B)
(4)
(5)
An LIRA
Rami onca
—
—
—
—
—
—
—
Biifo microscaplnis
+
-(?)
-(?)
+
-(?)
-(?)
+
Testudines
Gopheriis a^assizii
—
—
+
+
+
+
+
Squamata
Callisaunis
draconoides
—
—
+
+
V
9
+
Coleomjx variegatits
+
-(?)
+
+
+ (?)
+ (?)
+ (?)
Dipsosourus
dorsalis
—
—
—
+
?
?
?
Helodenna sitspectinn —
—
+
+
?
•?
4-
Xantusia vigilis
—
—
+ (?)
+
+
+■(?)
4-
Crotahis cerastes
—
—
+
+
+
+ (?)
4-
Crotalus mitchellii
—
—
—
?
+
+
?
Crotahis scutiihitus
—
—
—
+
+
+
4-
Leptotijphlups
hiimiUs
—
—
+
+
?
?
+(?)
Masticophis
flagellwn
+
-(?)
+
+
+
+
4-
PhijUorhijnchiis
decurtatus
—
—
—
+ (?)
+ (?)
+ (?)
+ (?)
Sonora
semiannulata
+
-(?)
+ (?)
+
+ (?)
+ (?)
+ (?)
Trimorphodnn
hisciitatits
+
— C^)
+
+
+
+
+
•'Distributions were inferred from localih' records available in research collections of California Academy of Sciences; M. L. Bean Lite Sciences Museum,
Brigham Young University, Prove. Utah; Museum of Vertebrate Zoology, University of California, Berkeley; Utah .Museum of Natural History, University of
Utah, Salt Lake City. Species listed as present ( + ) if they (1) e.xist as museum voucher specimens, (2) have been documented photographicalK' but not collected
because of threatened or endangered status, or (3) have been collected near a protected area and are known to occupy the appropriate habitat. For example,
Stewart (1994) summarized distributions of all Mojave Desert amphibians and reptiles on the basis of their occurrence in distinct habitat t>pes, and we used
these data as an indication of the likely presence of a species in an area if not actually documented. Doubts about any occurrences are indicated b\' (?).
D. dorsalis, the 3 species of Crotahis, L. species (Furlow and AiTnijo-Prewitt 1995, Lesica
Jmmilis, and P. decurtatus), diese 2 PWAs would and Allendorf 1995, Lomolino and Channell
constitute the largest blocks of protected area 1995). Designation of the Beaver Dam Wash
in the Utah portions of their distributions. and Joshua Tree PWAs as wilderness would
The biological significance of the Mojave provide an extremely economical, proactive
Desert region could be illustrated with com- conservation strategy for many species,
parable examples involving native birds, small
mammals, and vascular plants; literally scores IMPACT OF ROADS ON PLANT
of species are restricted to the low-elevation and Animal Communities
Joshua tree habitats on the southwestern slopes
of the Beaver Dam Mountains (see Behle et al. By definition under the 1964 Wilderness
1985, Albee et al. 1988, and Zeveloff 1988 for Act, wilderness areas must be large (at least
recent species compilations). Although most 5000 acres) and roadless. Because even some
areon tlie periphery of tlieir ranges, it is increas- remote and pristine areas contain primitive
ingly apparent that such peripheral popula- roads or tracks, roadlessness is often an issue
tions are critical to maintaining genetic diver- in debates over wilderness designation. Envi-
sity and to ensuring the long-term survival of ronmentalists tend to argue that the existence
110
Great Basin Naturalist
[Volume 56
of minor roads or dirt tracks is not contradic-
toiy to wilderness, but that no new roads should
be built. Wilderness opponents respond that
any road, no matter how primitive, disqualifies
PWAs for wilderness status. Decision makers
may be pressured to make exceptions to allow
new roads and water development within
wilderness boundaries. Here, we review the
objective evidence bearing on the importance
of roadlessness from a purely biological per-
spective. We deal with the effects of roads on
animals and plants independently.
Effects of Roads on Animals
Roads affect wildlife in many ways, both
direct and indirect. Among the more com-
monly reported adverse impacts of roads on
animal populations are road mortalities, animal
avoidance of roads, isolation of populations by
roads acting as barriers to animal movement,
reductions in natural habitats, increased poach-
ing, and elevated erosion leading to siltation of
aquatic habitats. On Utah BLM lands, large
mammals such as bighorn sheep [Ovis cana-
densis), black bear, and river otter are gener-
ally intolerant of human disturbance and activ-
ities. These and other mammals are known
also to avoid habitat adjacent to roads (Oxle>'
et al. 1974, Rost and Bailey 1979, Mader 1984,
Witmer and Calesta 1985, Van Dyke et al.
1986) and can therefore be displaced by the
presence of roads. Historically, humans in
western North America have also persecuted a
number of contemporaiy or former occupants
of BLM lands; such species include Golden
and Bald Eagles, gray wolf, and grizzly bear
(Bortolotti 1984, Mech 1995). In Utah, the in-
cidence of poaching is considerably higher in
regions adjacent to roads than in roadless areas
(W. Woody, UDWR, personal communication).
The negative effects of roads on wildlife can
generally be ameliorated by closing the roads
to traffic. Road mortality and the advance of
habitat alteration along roads should stop en-
tirely, and poaching should be sharply cur-
tailed. For larger animals, roads would likely
cease to act as barriers to animal movement
and gene flow. However, this might not be true
for some smaller species, whose moxements
are more restricted generally. Significant ero-
sion and siltation of aquatic habitats might be
reduced only slightly. Siltation can be an impor-
tant consideration, for example, on the Aquar-
ius Plateau, where reductions (by as much as
1/2) in the depths of some naturally shallow
lakes have already increased winter fish kills.
Finally, if efforts were made to reintroduce
some of the large mammals considered above,
these efforts might be greatly facilitated by the
protection of large blocks of roadless lands
that experience minimal human intrusion.
In siunmar)', if tra\'el on minor roads and
tracks were to be permanently restricted, most
but not all of the negative effects on wildlife
would likely be ameliorated. Similar reasoning
would suggest that the effects of any new un-
paved minor roads or tracks might be minimal
if the roads were used briefly and sporadically,
e.g., to cany communications equipment.
Effects of Roads on Plant Communities
The most compelling argument for large
roadless areas is probal)l> the protection of plant
communities from disturbances that can even-
tually transform whole ecosystems. Through
both direct and indirect effects, roads tend to
disiTipt nati\e communities of both microphytes
and macrophytes. Increased off-road vehicle
traffic in roaded areas directh' harms ciyptobi-
otic soil crusts, which play a key role in main-
taining healthy ecosystems in semiarid and
arid lands, and kills or injures plants and per-
haps soil-nesting insects like bees and wasps.
Indirect effects include the introduction of
nonnative pest plants, which have gradually
replaced many native species and drastically
altered features of certain habitats. The eco-
system-wide effects of these exotics are well
illustrated b>' Asian tamarisk {Tainarix cliincn-
sis), which has channelized rivers and streams
throughout the Colorado drainage and thereby
altered the characteristics (flow regimes, tem-
peratures, and sediment loads) of both aquatic
and riparian habitats to flie detiiment of num-
erous native fishes, insects, birds, mammals,
and plants (Loope et al. 1988, Sudbrock 1993).
Below, we elaborate on the direct and indirect
effects of roads on plant communities and on
the maintenance of both biodiversity and nat-
ural networks of interactions in Utah's native
ecosystems.
Threats to cryptobiotic soils. — Across
Utah's arid rangelands, a collection of cyano-
bacteria, algao, lichens, and mosses form micro-
phytic or cr> ptobiotic crusts on soil surfaces.
In pristine plant communities tliese cnists often
account for at least as much soil surface cover
as do \ ascular plants. The cnptoph>'tes provide
1996]
Wilderness Selection for Biodiversity
111
a number of \'akiable ecosystem services (re-
viewed in Harper and Marl^Ie 1988, West
1990, and Johansen 1993), including stabiliza-
tion of soils against wind and water erosion,
enhancement of water retention and infiltra-
tion (Brotherson and Rushforth 1983, Harpei"
and St. Clair 1985, Haiper and Marble 1988),
and nitrogen fixation by autotrophic bacteria,
including both free-living and symbiotic cyano-
bacteria (e.g., Snyder and WuUstein 1973, West
and Skujins 1977, Klubek and Skujins 1980,
Terry and Burns 1987). Their contribution to
the nitrogen economy of these arid ecosystems
is substantive. In southern Utah grasslands and
cold deserts dominated by pinyon pine and
juniper, nitrogen fixation by crusts is demon-
strably the dominant source of nitrogen for
vascular plants (Evans and Ehleringer 1993).
The greater soil moisture and fertility associ-
ated with biotic crusts have been shown to
result in higher tissue nutrient levels (Belnap
and Harper 1995 and references therein),
higher seedling sui^vivorship in associated vas-
cular plants (St. Clair et al. 1984, Harper and
St. Clair 1985, Belnap 1994), and greater (a)
floristic diversity (Kleiner and Harper 1972).
Herbivores and other consumers may benefit
indirectly from the enhanced nutrient status of
these ecosystems (Haiper and Pendleton 1993,
Belnap and Haiper 1995).
Growing recognition of the importance of
ciyptobiotic crusts to ecosystem processes has
led to concern about the impact of disturbance
by recreational users and nonnative grazers on
such surfaces (Anderson et al. 1982, Johansen
et al. 1984, Terry and Burns 1987, Cole 1991,
Evans and Ehleringer 1993, Belnap et al.
1994, Belnap 1995). On most semiarid Utah
lands, a single pass of an off-road vehicle will
reduce nitrogen fixation by cyanobacteria and
increase wind and water erosion of surface
soils (Williams et al. 1995). Estimates of time
to full recoveiy of disturbed biotic ciTists (includ-
ing niti-ogen-fixing capacity) range up to 50 years
in the Great Basin or 100 years on the Colorado
Plateau (J. Belnap personal communication).
The full biological and economic conse-
quences of disturbing biotic crusts remain to
be quantified. However, in semiarid ecosys-
tems where plant productivity is limited by
availability of water and nitrogen, even small
reductions in these resources can be expected
to diminish primary productivity to the detri-
ment of both the producers themselves and
the many consumers depending directly or in-
directly on thes(> pioducers for food. Haiper and
Pendleton (1993) have suggested that destruc-
tion of soil crusts, and associated changes in
forage quality, may be related to a decline in
the health of desert tortoise populations in
southwestern Utah (Grover and DeFalco 1995).
If that suggestion is supported by empirical
evidence in the Riture, then destniction of cmsts
may account in part for the ~$1() million cost
(to date, T. Esque personal communication) of
the Desert Tortoise Recovery Program.
Roads as corridors for invasions of
introduced species. — Possibly the greatest
adverse impact of roads on biological commu-
nities in Utah is the aggravation of invasions of
aggressive weeds along road corridors, where
disturbance from road construction has elimi-
nated native competitors. These introduced
plants now form the dominant cover on many
arid and semiarid landscapes in western North
America and are widespread in Utah (Mack
1981, Morrow and Stahlman 1984, Young et
al. 1987, papers in McArthur et al. 1990 and
Monsen and Kitchen 1994). Habitat degrada-
tion by nonnative, congregating grazers un-
doubtedly aided the initial spread of brome
grasses (genus Bromus) and other European or
Asian annuals into native habitats, including
grasslands previously dominated by caespitose
or tussock grasses (Young and Evans 1971,
Loope 1976, Mack 1981, 1989, Billings 1990,
1994). Brome grasses (red brome [B. riibens],
Japanese brome [B. japoniciis], downy brome
[B. mollis], ripgut brome [B. diandrus], and
especially cheat grass [B. tectonim]) have gready
increased fire frequenc)' (from an average of
60-110 yr to <5 yr in sagebrush steppe), as
well as altered the pattern and dynamics of
fires (e.g., Whisenant 1990). Invaded lands suf-
fer declining productivity (Stewart and Young
1939) and watershed damage (Buckhouse 1985)
and become drastically depleted in both native
plant species and cnptobiotic soil crusts (Young
and Evans 1978, Whisenant 1990, Billings 1990,
1994, Rosentreter 1994; Fig. 5). Treatments to
restore these lands often involve introductions
of still other exotics (e.g., Agropyron cristatum,
Kochia prostrata; see contributions to McAi-thur
et al. 1990 and Monsen and Kitchen 1994).
The influx of invading weedy annuals has
profound effects on genetic, species, and eco-
system diversity, although such effects remain
poorly documented. In some parts of Utah,
112
Great Basin Naturalist
[Volume 56
Harner Quadrats (1.0 YdO
124 Quadrats
Total species
No. of natives
No. aliens/quadrat
Fig. 5. Relationship of both total species richness, and
numbers of native species per quadrat, to the number of
individuals of introduced species per quadrat; plotted
from raw data in Harner and Harper (1973). Data are
from sagebrush-grasslands on private and BLM foothill
lands in Salt Lake, Davis, and Tooele counties.
brome grasses form virtual monocultures, en-
tirely replacing native communities, especially
in wet years (e.g., Pellant and Hall 1994, and
authors' observations). In other western states
brome grass invasions threaten state or feder-
ally listed plant species (Rosentreter 1994,
California Native Plant Society, personal com-
munication). Effects of habitat conversion
radiate upward through the food chain, and
adverse effects have been documented on
pronghorn {Antilocapra americana) and deer
(Pellant 1990, Roberts 1994), small vertebrate
prey of eagles and other raptors (Kochert and
Pellant 1986, Nydegger and Smith 1986), native
birds (Dobler 1994), and insects (Fielding and
Brusven 1994). As summarized by Billings
(1994), exotic annual grasses could constitute
a genuine threat to the existence of large inte-
grated ecosystems that have existed since the
Pleistocene in the relatively arid lands between
the Rocky Moimtains and Sierra Nevada. These
operational ecosystems could disappear o\ er large
areas of thousands of square kilometers.
A very high priority for future ecological
work in Utah will be to determine the extent
to which the remote BLM lands being consid-
ered for wilderness status might serve as ref-
uges for native flora and fauna. Seeds of brome
grass, dispersed by animal vectors, certainly
travel over long distances and into wilderness
areas. However, lanre roadless areas with low
circumference-to-area ratios might protect arid
and semiarid western ecosystems against whole-
sale habitat conversion. Exotic weeds tend to
invade native plant communities mainly along
roadsides, railroad right-of-ways, and other
highly disturbed sites (Forcella and Harvey
1983, Hunter 1990, literature cited in Billings
1990 and 1994; see also Bergelson et al. 1993).
Favorably wet drainage ditches provide inroads
to new habitat, and invaders spread outward
from the ditches during particularly wet years.
vUthough systematic suneys of nonnatives do
not presently exist for PWAs (and are sorely
needed), there is evidence that invasions of
exotic weeds may be prevented by restricting
access on existing roads. Thus, of the replicate
roadsides studied by Hunter (1990), introduced
species (including not only brome grasses but
Erodium cicutarium, Salsohi spp., and Sisijm-
J)riiiin altissiinuin) dominated all but the one
that had been closed to traffic and left undis-
turbed for many years prior to censusing.
The effects of roads on plant communities
appear to differ importantly from those on ani-
mal communities. Construction of new roads,
especially those with drainage ditches, may
hasten long-term and permanent changes to
local floras, and these changes may eventually
have markedly adverse effects on whole eco-
systems. Existing dirt tracks are probably less
threatening to plant communities; although
moisture conditions on the tracks may be as
favorable here as in drainage ditches, soil com-
paction appears to retard growth of most plants.
Given the costliness of aggressive fire sup-
pression (e.g., Vail 1994) and habitat restoration
measures (see I'eports in McArthur et al. 1990
and Monsen and Kitchen 1994), the most eco-
nomical strategy for prexenting the spread of
introduced grasses to areas that are still rela-
tively pristine ma>' be to maintain their road-
less character This also would proxide oppor-
tunities for investigating the effects of roads
(or lack thereof) on the advance of exotic
plants on arid lands in Utah.
Conclusions
Wilderness serves man\ purposes, and its
designation inxoK es man\' and \aried consid-
erations. The technical issues and evidence
presented here demonstrate that BLM w ilder-
ness lands can play a major and perhaps pre-
dominant role in safeguarding genetic, species,
and ecosx'stem dixersitx' across much of arid
1996]
Wilderness Selection fok Biodiversi-h-
113
Utah. Over the lont:; term, large, eontii^iuous
networks of wilderness and other protected
lands can provide sanctuaiy for populations of
animals with large area requirements, and can
help maintain natural processes and interac-
tions that sustain healthy biotic communities.
In many situations, wilderness designation can
pro\'ide low-cost protection for rare and en-
dangered species. BLM lands in geographi-
cally diverse regions of Utah all offer unique
ecological, scientific, and educational values.
To an extent so far unmeasured, wilderness
lands may protect native ecosystems from
wholesale transformation by invasions of exotic
species. Clearly, if biological considerations
are taken into account in wilderness decisions,
wilderness can play a critical role in the long-
term presei^vation of Utah's biological heritage.
Acknowledgments
Jayne Belnap, Phyllis Coley, Donald Duff
Sharon Emerson, Donald Feener, Bruce How-
lett, Thomas Kursar, and Samuel Rushforth
contributed to an earlier, nontechnical version
of this paper. In addition to these individuals
and the authors, Patricia Berger, Lynn Bohs,
Rex Gates, Steven Clark, Susan Fairbanks, Jer-
ran Flinders, Sarah George, James Harris,
Richard Hildreth, Carl Marti, Brian Maurer,
Norman Negus, Duke Rogers, Jon Seger, John
Sperry, Richard Tolman, Delbert Wiens,
Michael Windham, and Samuel Zeveloff en-
dorsed the earlier paper. Klancy de Nevers
and Jon Seger rescued corrupted computer
files. T. Griswold, F Parker, and V. Tepedino
allowed us to use their unpublished data (Table
2 and section on bees). J. Belnap and L. Shultz
commented on a previous draft of the manu-
script, and Garla Garrison helped us obtain
BLM maps.
Literature Cited
Albee, B. J., L. M. Schultz, and S. Goodrich. 1988.
Atlas of the vascular plants of Utah. Occasional Pub-
lications No. 7, Utah Museum of Natural History,
University of Utah, Salt Lake City. 670 pp.
Anderson, D. C., K. T. Harper, and S. R. Rushforth.
1982. Recoveiy of ci"\i)toganiic soil crusts from graz-
ing on Utah winter ranges. Journal of Range Man-
agement .35: 35.5-3.59.
Armbruster, E, and R. L\nde. 1993. A population viability
analysis for Mrican elephant {Loxodonta africana):
How big should reserves be? Conservation Biology
7: 602-610.
Aiwoon, !).. J. lloMAND, R. Bolander, B. Kiunkun, D. E.
House, L. Armstkonc, K. Thorne, and L.
England. 1991. Utah threatened, endangered, and
.sensitive plant field guide.
A.Xl.EROD, D. I. i960. The evolution of flowering plants.
Pages 227-305 in S. Tax, editor. Evolution after Dar-
win. Volume 1, The evolution ol" life. University of
Chicago Press, Chicago.
BABHrrr, B. 1995. Science: opening tlie iic^xt cluiptcr of
consei-vation history. Science 267: 1954-19.56.
Behle, W. H., E. D. Sorensen, and C. M. White. 1985.
Utah birds: a revised checklist. Occasional Publica-
tions No. 4, Utah Mu,seum of Natural Histon', Uni-
versity of Utah, Salt Lake City
Behnke, R. J. 1992. Native trout of western North Amer-
ica. American Fisheries Society Monograph 6. 275
pp.
Behnke R. J., and M. Zarn. 1976. Biology and manage-
ment of threatened and endangered western trouts.
USDA Forest Sei-vice, Technical Report RM-GTR-
28. Fort Collins, CO.
Belnap, J. 1994. Potential value of cyanobacterial inocula-
tion in revegetation efforts. Pages 179-185 in S. B.
Monsen and S. G. Kitchen, editors, Proceedings —
Ecology and Management of Annual Rangelands.
Report INT-GTR-313, Ogden, UT
. 1995. Soil surface disturbances; their role in
accelerating desertification. Environmental Moni-
toring and Assessment 37: 1-17.
Belnap, J., and K. T. Harper. 1995. The influence of ciyp-
tobiotic soil crusts on elemental content of tissue in
two desert seed plants. Arid Soil Research and
Rehabilitation 9: 107-115.
Belnap, J., K. T Harper, and S. D. Warren. 1994. Sur-
face disturbance of ciyptobiotic soil crusts: nitroge-
nase activity, chlorophyll content, and chlorophyll
degradation. Arid Soil Research and Reliabilitation
8: 1-8.
Belovsky, G. E. 1987. Extinction models and mammalian
persistence. Pages 35-57 in M. E. Soule, editor,
Viable populations for consei"vation. Cambridge Uni-
versity Press, Cambridge, UK.
Belovsky, G. E., J. A. Bissonette, R. D. Dueser, T. C.
Edwards, Jr., C. M. Luecke, M. E. Ritchie, J. B.
Slade, and E H. W.-^gner. 1994. Management of
small populations: concepts affecting the recovery of
endangered species. Wildlife Societ\' Bulletin 22:
307-316.
Bergelson, J., J. A. Newman, and E. M. Floresrou.x.
1993. Rates of weed spread in spatially heteroge-
neous environments. Ecology 74: 999-1011.
Bezy, R. L., and J. W. Sites, Jr. 1987. A preliminaiy stud\
of allozyme evolution in the lizard family Xantusi-
idae (Reptilia, Sauria). Heipetologica 43: 280-292.
Billings, W. D. 1990. Bromus tectoruni, a biotic cause of
ecosystem impoverishment in the Great Basin.
Pages 301-322 in G. M. Woodell, editor. The earth in
transition: patterns and processes of biotic impover-
ishment. Cambridge University Press, Cambridge,
UK.
. 1994. Ecological impacts of cheatgrass and resul-
tant fire on ecosystems in the western Great Basin.
Pages 22-30 in S. B. Monsen and S. G. Kitclien,
editors, Proceedings — Ecology and Management of
Annual Rangelands, USDA Forest Service, Technical
Report INT-GTR-313. Ogden, UT
114
Great Basin Natupialist
[Volume 56
BoLGER, D. T, A. C. Alberts, and M. E. Soule. 1991.
Occurrence patterns of bird species in habitat frag-
ments: sampling, extinction, and nested species sub-
sets. American Naturalist 1.37: 1.55-166.
BORTOLOTTI, G. R. 1984. Trap and poison mortalit>' of
Golden and Bald Eagles. Journal of Wildlife Man-
agement 48: 117.3-1179.
Brotherson, J. D., AND S. R. Rlshforth. 1983. Influ-
ence of cr}'ptogamic ciiists on moisture relationships
of soils in Navajo National Monument, Arizona.
Great Basin Natiu'alist 43: 73-78.
Brown, J. H. 1971. Mammals on mountaintops: nonequi-
librium insular biogeographv'. American Naturalist
105: 467-478.
. 1995. Macroecolog>'. Universit\ of Cliicago Press,
Chicago.
Brown, J. H., and E. J. Heske. 1990. Control of a desert
grassland transition by a keystone rodent guild. Sci-
ence 250: 1705-1707.
Buckhouse, J. C. 1985. Effects of fire on rangeland water-
sheds. Pages 58-60 in K. Sanders and J. Durham, edi-
tors. Proceedings of a Symposium: Rangeland Fire
Effects. U.S. Department of the Interior, Bureau of
Land Management, Boise, ID.
BiRY. R. B., .\SD D. J. Germano, editors. 1994. Biology-
of North American tortoises. USDI Fish and Wild-
life Research Report ISSN 1040-2411, No. 13. 204
pp.
C.-^se, T. J., and M. L. Cody. 1988. Testing theories of
island biogeography. American Scientist 7.5: 402—411.
Christensen, N. L., et al. 1989. Inteqireting the Yellow-
stone fires of 1988. Bioscience 39: 678-685.
Cole, D. N. 1991. Trampling disturbance and recovery of
cryptogamic soil crusts in Grand Canyon National
Park. Great Basin Naturalist 50: 321-325.
Cronouist, a., a. H. Holmgren, N. H. Holmgren, and
J. L. Reveal. 1972. Intermountain flora, \blume 1.
Hafiier Publishers, New York.
Cross, S. P 1985. Responses of small manmials to forest
perturbations. Pages 269-275 in Riparian ecosys-
tems and their management: reconciling conflicting
uses. USDA Forest Sewice, Technical Report RM-
GTR-120. Fort Collins, CO.
Davidson, D. W, R. S. Inouye, and J. H. Brown. 1984.
Granivory in a desert ecosystem: experimental evi-
dence for indirect facilitation of ants by rodents.
Ecology' 65: 1780-1786.
D.A\ is, J. I., AND K. C. NI.XON. 1992. Populations, genetic
variation, and the delimitation of phylogenetic
species. Systematic Biolog\- 41: 421—435.
De.\CON, J. E., AND W. L. MiNCKLEY. 1974. Desert fishes.
Pages 385-488 in G. W Brown, editor Desert biol-
ogy. Volume II. Academic Press, New York.
DEPARTMENT OF THE INTERIOR (BUREAU OF L.\ND MaN./VGE-
MENt). 1994. Ecosystem management in the BLM:
from concept to commitment. U.S. Government
Printing Office (573-183), VV^ishington, DC. 16 pp.
Diamond, J. M. 1984. "Normal" extinctions of isolated
populations. Pages 191-246 in M. H. Nitecki, cditoi;
Extinctions. University of Chicago Press, Chicago.
Dobler, E C. 1994. Washington state shrub-steppe eco-
system studies with emphasis on the relationship
between nongame birds and shrub and grass coxcr
densities. Pages 149-161 in S. B. Monsen and S. C;.
Kitchen, editors, Procecding.s — Ecology and Man-
agement of Annual Rangelands, USDA Forest Ser-
vice, Technical Report INT-GTR-313. Ogden, VT.
Durr.\nt, S. D. 1950. Mammals of Utah: taxonomy and
distribution. Museum of Natural Histoiy, University
of Kansas Publications 6.
Edmunds, G. F 1994. Tail water insects. The Landing Net
(Sinnmer): 6-7.
. 1995. Tail water insects; re\isited. The Landing
Net (Fall): 4, 11.
Evans, R. D., and J. R. Ehleringer. 1993. A break in the
nitrogen cycle in aridlands? E\idence from S'^N of
soils. Oecologia 94: 314-317.
Fielding, D. J., and M. A. Brusven. 1994. Grasshopper
community responses to shrub loss, annual grass-
lands, and crested wheatgrass seedings: manage-
ment implications. Pages 162-166 //) S. B. Monsen
and S. G. Kitchen, editors. Proceedings — Ecology
and Management of Annual Rangelands, USDA For-
est Senice, Technical Report INT-GTR-313. Ogden,
UT
Forcella, F, and S. J. Har\ey. 1983. Eurasian weed
infestation in western Montana in relation to vegeta-
tion and disturbance. Madroiio 30: 102-109.
Frankel, O. H., and M. E. Soule. 1981. Consen ation
and e\oIution. Cambridge Universit\' Press, Cam-
bridge, UK.
FuRLOW, F B., AND T. Armijo-Prewitt. 1995. Peripheral
populations and range collapse. Conservation Biol-
ogy 9: 1.345.
Gr.\YSON, D. K. 1993. The desert's past: a natural prehis-
ton' of the Great Basin. Smithsonian Institution
Press, Washington, DC.
Grover, M. C, and L. a. DeFalco. 1995. Desert tortoise
{Gophenis agassizii): status-of-knowledge outline
with references. USDA Forest Service. Technical
Report INT-GTR-316. Ogden, UT 134 pp.
Grumbine, R. E. 1994. What is ecosystem management':'
Consei-vation Biologv' 8: 27-38.
Harner, R. F, and K. T. Harper. 1973. Mineral composi-
tion of grassland species of the eastern Great Basin
in relation to stand producti\it\'. Canadian Journal of
Botany 51: 2037-2046.
Harper, K. T. 1986. Historical environments. Pages 51-63
()! W. L. D Azevedo, editor. Handbook of North
American Indians, \blume 11: Great Basin. Smith-
sonian Institution, Washington, DC.
Harper, K. T, and J. R. M.\rble. 1988. A role for non\ as-
cular plants in management of arid and semiarid
rangelands. Pages 13.5-169 i)i P. T. Tneller, editor.
Vegetation science applications for rangeland analy-
sis and management. Khiwer .\cademic Publisher,
Dordrecht.
Harper, K. T, and R. L. Pendleton. 1993. Cyanobacte-
ria and cyanolichens: Can they enhance a\ailabilit\'
of essential minerals for higher plants? Great Basin
Naturalist 53: 59-72.
Harper, K. T, and L. L. St Cl.\ir. 1985. Ci-yptogamic
soil cnists on arid and semiarid rangelands in Utah:
effects on seedling establishment and soil stability.
Final Report Buieau of Land Management, Utah
State Office, Salt Lake Cit\.
Hahkis, L. D. 1984. The fragmented forest: island bio-
geography theory and the preser\'ation of biotic
di\ersity. l'ni\'ersit\' of Chicago Press, Chicago.
Heanky. 1-. H. 19S4. Mammalian species richness on
islands on the Simda Shelf Southeast Asia. Oecolo-
gia (Berlin) 61: 11-17.
H()Li:)F\. P B., R. Stone, \\'. White, G. Somermlle, D.
Di KK \\. Gkhvais, and S. Gloss. 1974. Threatened
1996]
Wll.nKKNKSS SkI.KC'I'ION I'OH B|()I)I\ khsht
115
fishes of I'tali. Procet'dintis ol tlu' I'tali Academy ol
Science, Arts, and Ix-tters 51: 4(i-(i5.
Ik BBS, C. L. 1940. Speciation of I'ishes. Anicrican Natu-
ralist 74: 198-211.
, 1941. The relation of hydrological conditions to
speciation in fishes. Pages 182-195 in J. (J. Need-
liani, editor, .\ S\nip()siinn on Hydrol)iolog\. Uni\er-
sit\ ol \\'isconsin Press, Madison.
Ill \II:k, H. 1990. Recent increases in Hroiiius on [\\r
i\e\ada Test Site. Pages 22-25 in E. D. McArtlun;
E. M. Ronine>, S. D. Smith, R T. Tueller, editors.
Proceedings — S\inposiuni on Cheatgrass Invasion,
Shrub Die-Off, and Other Aspects of Shrub Riology
and Management, USDA Forest Ser\ice, Technical
Report INT-GTR-276. Ogden, UT
lUCN. 1978. Categories, objectives and criteria lor pro-
tected areas. Merges, Switzerland.
J.\EGER, E. C. 1957. The North American deserts. Stan-
ford University Press, Palo Alto, CA. 308 pp.
Jennings, M. R., and M. R Hayes. 1994. Decline of native
ranid frogs in the desert Southwest. Pages 183-211
in R R. Rrovvn and J. W. Wright, editors, Heipetol-
ogy of the North American deserts. Southwestern
Herpetological Society Special Publication No. 5.
Van Nuys, CA. 311 pp.
JOHANSEN, J. R. 1993. Ciyptogamic crusts ot semiarid and
arid lands of North America. Journal of Ph\cology
29; 140-147.
JOHANSEN, J. R., L. L. St. Claih, R. L. Webb, and G. T.
Nebeker. 1984. Recovery patterns of cryptogamic
soil crusts in desert rangelands following fire distur-
bance. Bryologist 87: 238-243.
Johnson, R. R., L. T Haight, and J. M. Simpson. 1977.
Endangered species vs. endangered habitats: a con-
cept. Pages 68-79 in Importance, presei'vation and
management of riparian habitat. USDA Forest Ser-
vice, Technical Report RM-GTR-43. Fort Collins, CO.
Ji lander, O. 1962. Range management in relation to
mule deer habitat and herd productivity in Utah.
Journal of Range Management 15: 278-281.
K.\v, C. E. 1990. Yellowstone s northern elk herd: a critical
evaluation of the "natural regulation paradigm.
Unpublished doctoral dissertation, Utah State Uni-
versity, Logan.
Kershner, J. L. 1995. Ronneville cutthroat trout. Pages
28-35 in M. K. Young, editor, Consenation assess-
ment for inland cutthroat trout. USDA Forest Ser-
vice, Technical Report RM-GTR-256. Fort Collins,
CO.
Kleiner, E. F, .\nd K. T Harper. 1972. Environment and
community organization in grasslands of Canyon-
lands National Park. Ecolog\- 53: 299-309.
Klubek, B., and J. Ski JINS. 1980. Heterotrophic nitrogen
fi.xation in arid soils crusts. Soil Biology and Bio-
chemistiy 12: 229-236.
Knopf, F L. 1985. Significance of riparian vegetation to
breeding birds across an altitudinal cline. Pages
105-111 in Riparian ecosystems and their manage-
ment: reconciling conflicting uses. USDA Forest Ser-
vice, Technical Report RM-GTR-120. Fort Collins,
CO.
Kochert, M. N., and M. Pellant. 1986. Multiple use in
the Snake River Birds of Prev Area. Rangelands 8:
217-220.
Kushlan, J. A. 1979. Design and management of conti-
nental wildlife resei-ves: lessons from the Everglades.
Biological Conservation 15: 281-290.
Lesiga, P, and E W. Ai.LENDORE 1995. WIk-h are periph-
eral populations valuable for conservation? Conser-
\ation Biology 9: 753-760.
LoMOLiNO, M. v., and R. CllANNEl.L. 1995. Splendid iso-
lation: patterns of the geographic range colla])se in
endangered mannnals. |onrna! of .Mainmalogv 76:
335-347.
LooPE, L. L.. P C;. Sanghe/., P W. Tarr, VV. L. L(X)PE,
AND R. L. Anderson. 1988. Biological invasions of
arid land nature reserves. Biological Conservation
44:95-118.
LooPE, W. L. 1976. Relationships of \egetation to envi-
ronment, Canyonlands National Park, Utah. Unpub-
lished doctoral dissertation, Utah State Uni\t'rsit\,
Logan.
NfvGk, R. N. 1981. The invasion oi' Bronnis tectorum L.
into western North America: an ecological chronicle.
Agro-Ecosystems 7: 145-165
. 1989. Temperate grasslands vulnerable to plant
invasions: characteristics and conseciucnccs. Pages
155-179 ()i J. A. Drake, H. A. Mooncy, F di Castri,
R. H. Groves, F J. Kruger, M. Rejmanek, and M.
Williamson, editors. Biological invasions. John Wiley
and Sons, New York.
MacKinnon, J., K. MacKinnon, G. Child, and J. Tiior-
SELL. 1986. Managing protected areas in the Tropics.
lUCN, Gland, Switzerland.
MacMahon, J. A. 1986. Deserts. Audubon Society Nature
Guides. Alfred Knopf New York. 638 pp.
Mader, H. J. 1984. Animal habitat isolation by roads and
agricultural fields. Biological Conservation 29:
81-96.
McArthur, E. D., E. M. Romney, S. D. Smith, R T
TuELLER, editors. 1990. Proceedings — Symposium
on Cheatgrass Invasion, Shrub Die-Off and Other
Aspects of Shrub Biology and Management. USDA
Forest Service, Technical Report INT-GTR-276.
Ogden, UT
Mech, L. D. 1995. The challenge and opportunitv- of
recovering wolf populations. Conservation Biologv
9: 270-274.
Meffe, G. K., and C. R. Carroll. 1994. Principles of con-
sei^vation biology. Sinauer, Sunderland, fvIA. 600 pp.
MiCHENER, C. D. 1979. Biogeography of the bees. Annals
of the Missouri Botanical Garden 66: 277-347.
Miller, R. R. 1959. Origin and affinities of the freshwater
fish faima of western North America. Pages 187-222
in C. L. Ilubbs, editor. Zoogeographv'. .\j\AS Publi-
cation 51.
. 1961. Man and the changing fish fauna of the
American Southwest. Papers, Michigan .'\cademy of
Science, Arts, and Letters 46: 365-404.
MiNCKLEY, W. L., AND J. E. Deacon. 1968. Southwestern
fishes and the enigma of "endangered species. Sci-
ence 159: 1424-1432.
MiNCKLEY, W. L. AND J. E. DEACON, EDITORS. 1990. BatUes
against extinction: native fish management in the
American West. University of Arizona Press, Tucson.
MiNCKLEY, W L., D. A. Hendrickson, and C. E. Bond.
1986. Geography of western North American fresh-
water fishes: description and relationships to intra-
continental tectonism. Pages 519-613 in C. H.
Hocutt and E. O. Wiley, editors. The zoogeography
of North American freshwater fishes. John Wiley &
Sons, New York.
MoNSEN, S. B., and S. G. Kitchen, editors. 1994. Pro-
ceedings— Ecology and Management of Annual
116
Great Basin Naturalist
[Volume 56
Rangelands. USDA Forest Service, Technical Report
INT-GTR-313. Ogden, UT.
Morrow, L. A., and R W. Stahlman. 1984. The histor\'
and distribution of downy brome {Bromiis tectonim)
in North America. Weed Science 32 (Supplement 1):
2-7.
MoYLE, R B., AND J. E. Williams. 1990. Biodiversity loss
in the temperate zone: decline of the native fish
fauna of California. Conservation Biology 4: 275-2S4.
National Research Council. 1978. Conservation of
germplasm resources: an imperative. National Acad-
emy of Sciences, Washington, DC.
Neff, J. L., AND B. B. Simpson. 1993. Bees, pollination
systems and plant diversity. Pages 143-167 /)! J.
LaSalle and I. E. Gauld, editors, Hymenoptera and
biodiversity. C.A.B. International, Wallingford, UK.
Newmark, W D. 1985. Legal and biotic boundaries of
western North American national parks: a problem
of congruence. Biological Consewation 33: 197-208.
. 1987. A land-bridge island perspective on mam-
malian extinctions in western North American parks.
Nature 325: 430-432.
. 1991. Tropical forest fragmentation and the local
e.xtinction of understor\' birds in the Eastern Usam-
bara Mountains, Tanzania. Conservation Biologv 5:
67-78.
. 1995. Extinction of mammal populations in west-
ern North American national parks. Conservation
Biology 9: 512-526.
Noss, R. E 1992. The wildlands project: land consei^vation
strategy. Pages 10-25 in The wildlands project. Wild
Earth Special Issue, Cenozoic Society.
Nydegger, N. C, and G. W. Smith. 1986.' Pages 152-156
in D. E. McArthur and B. L. Welch, editors. Pro-
ceedings— Symposium on the Biology o{ Artemisia
and Chnjsothainnus, USDA Forest Service, Techni-
cal Report INT-GTR-200. Ogden, UT.
OxLEY, D. J., M. B. Fenton, and G. R. C.\r.\iodv. 1974.
The effects of roads on populations of small mam-
mals. Journal of Applied Ecology 11: 51-59.
Parker, E D. 1985. A candidate legume pollinator Osinia
sanrafaclae Parker journal of Apiculture Research
24: 132-136.
. 1986. Field studies with Osmia sanrafaelae Parker,
a pollinator of alfalfa (Hymenoptera: Megachilidae).
Journal of Economic Entomology 79: 384—386.
Patterson, B. D. 1984. Mammalian extinction and bio-
geography in the southern Rocky Mountains. Pages
247-293 in M. H. Nitecki, editor. Extinctions. Uni-
versity of Chicago Press, Chicago.
Pellant, M. 1990. The cheatgrass-wildfire cycle — are
diere any solutions? Pages 11-18 in E. D. McArthur,
E. M. Romney, S. D. Smith, P T. Tueller, editors.
Proceedings — Symposium on Cheatgrass Invasion,
Shrub Die-Off and Other Aspects of Shrub Biology
and Management, USDA Forest Service, Technical
Report INT-GTR-276. Ogden, UT.
Pellant, M., and C. Hall. 1994. Distribution of two
exotic grasses on intermountain rangelands. Pages
109-112 in S. B. Monsen and S. G. Kitchen, editors.
Proceedings — Ecology and Management of Annual
Rangelands, USDA Forest Senice, Technical Report
INT-GTR-313. Ogden, UT.
Pickett, S. T. A., and J. N. Thompson. 1978. Patch
dynamics and the design of nature reserves. Biologi-
cal Consei^vation 13: 27-37.
Pimm, S. L., H. L. Jones, and J. Diamond. 1988. On the
risk of extinction. American Naturalist 132: 757-785.
Pl.\tz, J. 1984. Status report for Rana onca Cope. Pre-
pared for Office of Endangered Species, U.S. Fish
and Wildlife Sei^vice, Albuquerque, NM. 27 pp.
Primac:k, R. B. 1993. Essentials of conservation biology
Sinauer, Sunderland, MA. 561 pp.
FIaven, P R. 1992. The nature and value of biodiversity.
Pages 1-18 in Global biodiversity strategy: guide-
lines for action to save, study, and use earth s biotic
wealth sustainably and equitablv. WRI, IL'CN,
UNEP
Reid, W v., and K. R. Miller. 1989. Keeping options
alive: the scientific basis of consenang biodiversity.
World Resources Institute, Washington, DC.
RiCKART, E. A. In preparation. Elevational diversity gradi-
ents, biogeography, and the structure of montane
mammal communities in the intermountain region.
Journal of Biogeography: In preparation.
Roberts, T. C, Jr. 1994. Resource impacts of cheatgrass
and wildfires on public lands and livestock grazing.
Pages 167-169 in S. B. Monsen and S. G. Kitchen,
editors. Proceedings — Ecology and Management of
Annual Rangelands, USDA Forest Service, Technical
Report INT-GTR-313. Ogden, UT.
Rosentreter, R. 1994. Displacement of rare plants by
e.xotic grasses. Pages 170-175 in S. B. Monsen and
S. G. Kitchen, editors. Proceedings — Ecology and
Management of Annual Rangelands, USDA Forest
Service, Technical Report INT-GTR-313. Ogden,
UT.
RosT, G. R., and J. A. Bailey. 1979. Distribution of mule
deer and elk in relation to roads. Journal of Wildlife
Management 43: 634-641.
Rowlands, P, H. Johnson, E. Ritter, and A. Endo.
1982. The Mojave Desert. Pages 103-162 in G. L.
Bender, editor. Reference handbook on the deserts
of North America. Greenwood Press, Westwood, CT.
594 pp.
Salwasser, H., C. Schonewald-Co.\, and R. Baker.
1987. The role of interagency cooperation in manag-
ing viable populations. Pages 159-173 in M. E.
Soule, editor, Viable populations for consen'ation.
Cambridge University' Press, Cambridge, UK.
Samson, D. A., T. E. Philippi, and D. W. Da\ idson. 1992.
Granivon' and competition as determinants of annual
plant diversity in the Chihuahuan Desert. Oikos 65:
61-80.
Saunders, D. A., R. J. Hobbs, and C. R. M.\rgules. 1991.
Biological consequences of ecosystem fragmenta-
tion: a review. Consenation Biolog\ 5: 18-32.
Schonewald-Co.x, C. M. 1983. Guidelines to manage-
ment: a beginning attempt. Pages 414—445 in C. M.
Schonewald, S. M. Chambers, B. MacBnde, and, L.
Thomas, editors. Genetics and consenation. Benja-
min Cummings, Menio Park, CA.
Shaffer, M. L. 1981. Minimum population size for
species conservation. Bioscience 31: 131-134
Shiozavva, D. K., and R. P Evans. 1994. The use of DNA
to identify geographical isolation in trout stocks.
Pages 125-131 in R. Barnhart, B. Shake, and R. H.
Hamre, editors. Wild trout \. wild trout in the 21st
century.
SiiREVE, E 1942. The desert vegetation of North America.
Botanical Reviews 8: 195-246.
ShuLTZ, L. M. 1993. Patterns of endemisni in the Utah
flora. Pages 249-263 in R. Si\inski and K. Lightfoot,
1996]
Wilderness Selection for Biodiversiit
117
editors. Southwestern rare and endangered plants.
New Mexico Department of ForestiT and Resources
Conservation Division, Miscellaneous Publication
No. 2. Santa Fe, NM.
Shultz, L. M., E. E. Neely, and J. S. Tuhy. 1987. Flora of
the Orange Cliffs of Utah. Great Basin Naturalist 47:
287-298.'
Smith, G. R. 1966. Distrilnition and evolution of the
North American catostomid fishes of the subgenus
Pantosteus, genus Catostoiitiis. Miscellaneous Publi-
cations of the Museum of Zoology, University of
Michigan, No. 129.
. 1978. Biogeography of intermountain fishes. Inter-
mountain Biogeography: a Symposium. Great Basin
Naturalist Memoirs 2: 17-42.
Snyder, J. M., ,^nd L. H. VVlllstein. 1973. The role of
desert cryptogams in nitrogen fixation. American
Midland Naturalist 90: 257-265.
SoULE, M. E., EDITOR. 1987. Viable populations for con-
sen'ation. Cambridge LIniversitv Press, Cambridge,
UK. '
SouLE, M. E., D. T. Bolger, A. C. Alberts, R. Sauvajot,
J. Wright, M. Sorice, and S. Hill. 1988. Recon-
structed dynamics of rapid extinctions of chaparral-
requiring birds on urban habitat islands. Consen'a-
tion Biolog)' 2: 75-92.
SouLE, M. E., and B. a. Wilco.x. 1980. Conservation
biology: an evolutionaiy-ecological perspective. Sin-
auer, Sunderland, MA.
St. Clair, L. L., B. L. Webb, J. R. Johansen, and G. T.
Nebeker. 1984. Cryptogamic soil crusts: enhance-
ment of seedling establishment in disturbed and
undisturbed areas. Reclamation and Revegetation
Research 3: 129-136.
Stebbins, G. L., Jr. 1952. Aridity as a stimulus to plant
evolution. American Naturalist 86: 33-44.
Stebbins, R. C. 1985. A field guide to western reptiles
and amphibians. Houghton Mifflin Co., Boston, MA.
Stevens, C. C. 1992. The elevational gradient in altitudi-
nal range: an extension of Rapoport's latitudinal rule
to altitude. American Naturalist 140: 893-911.
Stewart, G. R. 1994. An overview of the Mohave Desert
and its herpetofauna. Pages 55-69 in P. R. Brown
and J. W Wright, editors, Herpetology of the North
American deserts. Southwestern Heipetological Soci-
ety Special Publication No. 5. Van Nuys, CA. 311 pp.
Stewart, G., and A. E. Young. 1939. The hazard of basing
permanent grazing capacity on Bromus tectonim.
American Society of Agronomy Journal 31: 1002-1015.
Sudbrock, a. 1993. Tlimarisk control. I. Fighting back: an
overview of the invasion, and a low-impact way of
fighting it. Restoration and Management Notes 11:
31-34.
Tepedino, V. J. 1981. The pollination efficiency of the
squash bee {Pepomipis pniinosa) and the honeybee
{Apis mellifera) on summer scjuash {Citcurbifa pepo).
Journal of the Kansas Entomological Societv 54:
359-377.
Terborgh, J., AND B. Winter. 1980. Some causes of
extinction. Pages 119-134 ;'/! M. E. Soule and B. A.
Wilcox, editors, Conservation biology. Sinauer, Sun-
derland, MA.
Terry, R. E., and S. J. Burns. 1987. Nitrogen fi.xation in
cryptogamic soil crusts as affected by disturbance.
Pages 369-372 in Proceedings — Pinyon-Juniper
Conference, USDA Forest Service, Technical Report
INT-GTR-215. Ogden, UT
Tyus, H. M., AM) C. A. K\HH 1989. Habitat use and
streamtlow needs ol rare and endangered fishes,
Yampa River, Colorado. U.S. Fish and Wildlife Ser-
vice Biological Report 89(14). 27 pp.
U.S. Fish and Wildlife Servige. 1993. Endangered and
threatened wildlife and plants, 50 CFR 17.11 and
17.12, August 23, 1993. Division of Endangered
Species, U.S. Fish and Wildlife Semce, Washington,
DC.
Utah Division of Wildlife Resourges. 1992. Native
Utah wildlife species of special concern. Draft docu-
ment. Utah Division of Wildlife Resources, Salt Lake
City. 31 pp.
Utah Wilderness Coalition (UWC). 1990. Wilderness
at the edge. Peregrine Smith Books, Layton, UT.
Vail, D. 1994. Symposium introduction: management of
semiarid rangelands — impacts of annual weeds on
resource values. Pages 3-4 in S. B. Monsen and S. G.
Kitchen, editors, Proceedings — Ecology and Man-
agement of Annual Rangelands, USDA Forest Ser-
vice, Technical Report INT-GTR-313. Ogden, UT
Van Dyke, E G., R. H. Brogke, H. G. Shaw, B. B. Agker-
man, T. E Hemker, and E G. Lindzey. 1986. Reac-
tions of mountain lions to logging and human activ-
ity Journal of Wildlife Management 50: 95-102.
Wagner, E H., R. Foresta, R. B. Gill, D. R. MgCul-
LOUGH, M. R. Pelton, W. E Porter, and gonsulta-
TION ON LAW and poligy BY J. L. S.\x. 1995. Wildlife
policies in the U.S. National Parks. Island Press. 242
pp.
Warren, M. L., and B. M. Burr. 1994. Status of fieshwater
fishes of the United States: ovendew of an imperiled
fauna. Fisheries 19: 6-18.
Welsh, S. L. 1978. Endangered and dueatened plants of
Utali, a reevaluation. Great Basin Naturalist 38: 1-18.
Welsh, S. L., N. D. Atwood, S. Goodrigh, and L. C.
Higgins. 1993. A Utah flora. 2nd edition. Print Ser-
vices, Brigham Young University, Provo, UT.
Welsh, S. L., N. D. Atwood, and J. L. Reveal. 1975.
Endangered, threatened, extinct, endemic and rare
or restricted LJtah vascular plants. Great Basin Natu-
ralist 35: 327-326.
West, N. E. 1990. Stiiicture and hmction of microphytic
soil crusts in wildland ecosystems of arid to semiarid
regions. Advances in Ecological Research 20:
179-223.
West, N. E., and J. Skujins. 1977. The nitrogen cycle in
North American cold-winter semi-desert ecosys-
tems. Oecologia Plantarum 12: 45—53.
Whisenant, S. G. 1990. Changing fire frequencies on
Idaho's Snake River Plains: ecological and manage-
ment implications. Pages 4-10 in E. D. McArthur,
E. M. Romney, S. D. Smith, P T. Tueller, editors.
Proceedings — Symposium on Cheatgrass Invasion,
Shrub Die-Off, and Other Aspects of Shrub Biology
and Management, USDA Forest Service, Technical
Report INT-GTR-276. Ogden, UT
Whittaker, R. H. 1972. Evolution and measurement of
species diversity. Ta.xon 21: 213-251.
Wilgove, D. S., C. H. MgLellan, and A. E Dobson.
1986. Habitat fragmentation in the temperate zone.
Pages 237-256 in M. E. Soule, editor, Consei-vation
biology: the science of scarcity and diversity. Sinauer
Associates, Sunderland, MA.
WiLGO.X, B. A., AND D. D. Murphy. 1985. Conservation
strategy: the effects of fragmentation on extinction.
American Naturalist 125: 879-887.
118
Great Basin Naturalist
[Volume 56
Williams, J. D., J. E Dobrowolski, N. E. West, and D. A.
Gillette. 1995. Microphytic cnist influence on wind
erosion. Transactions of the American Society of
Agricultural Engineers 38: 131-137.
Williams, J. E. 1991. Preserves and refuges for native
western fishes; histoiy and management. Pages
171-189 in W. L. Minckley and J. E. Deacon, edi-
tors. Battle against extinction, native fish manage-
ment in the American West. University of Arizona
Press, Tucson. 517 pp.
Willis, E. O. 1974. Populations and local extinctions of
birds on Barro Colorado Island, Panama. Ecological
Monographs 44: 153-169.
Wilson, E. O. 1988. Biodiversity. National Academy
Press, Washington, DC.
WiTMER, G. W, AND D. S. Calesta. 1985. Effect of forest
roads on habitat use by Roosevelt elk. Northwest
Science 59; 122-125.
WOLZ, E. R., and D. K. Shiozawa. 1995. Soft sediment
benthic macroinvertebrate communities of the Green
River at the Ouray National Wildlife Refuge, Uinta
County, Utah. Great Basin Naturalist 55; 213-224.
Woodbury, A. M., and R. Hardy. 1948. Studies on the
desert tortoise, Gopherus agassizii. Ecological Mono-
graphs 18; 145-200.
Young, J. A., and R. A. Evans. 1971. Invasion of medusa-
head into the Great Basin. Weed Science 18: 89-97.
. 1978. Population dynamics after wildfires in sage-
brush grasslands. Journal of Range Management 31:
283-289.
Young, J. A., R. A. Evans, and B. L. K.ay 1987. Cheatgrass.
Rangelands 9; 266-270.
Zeveloff, S. I. 1988. Mammals of the Intemiountain West.
University of Utah Press, Salt Lake Cit>.
Received 27 September 1995
Accepted 21 March 1996
CrcMt Basin Naturalist 56(2), © 1996, pp. 119-12«
NUTRIENT DISTRIBUTION IN QUERCUS GAMBELIl
STANDS IN CENTRAL UTAH
A. R. Tiedeinann' and VV. P. Claiy-
Abstract. — Gambel oak {Qucrciis gambchi Nutt.) is increasingly recognized as a valuable fuelwood throughout Ari-
zona, Colorado, New Mexico, and Utah. Knowledge of the distribution of nutrients among biotic and abiotic coinpo-
iicnts is an important step in developing prescriptions for managing these stands for sustainable productivity.
Eight Q. gamhclii stands were sampled for concentrations (%) and accumulations (kg ha"l) of total nitrogen (N),
phosphorus (P), sulfur (S), calcium (Ca), magnesium (Mg), potassiimi (K), and sodiimi (Na) among aboveground and
helowground biomass components and the upper 30 cm of soil. Highest concentrations of N, R and S occuned in oak
Itaxes, underston' leaves, and the forest floor layer. Generally, highest concentrations of Ca, Mg, K, and Na occurred in
the soil.
The greatest proportion of the total capital of individual nutrients was contained in the soil (82%-99%). Above-
ground components of li\e biomass, standing and down-dead, and forest floor contained 10%, 14%, and 8%, respec-
tixely, of total capitals of N, V, and S. The forest floor had the largest accumulation (63%) of total nutrients (N, R S, Ca,
Mg, K, and Na) of live and dead aboveground components. Nutrient accumulation in live biomass was heavily weighted
to the belowground component. The dense system of roots, rhizomes, and lignotubers comprising 56%' of total biomass
contained 62% of the total accumulation of nutrients in live biomass.
Low levels of total P in the soil and accumulation of 14% of the ecosystem total of P in aboveground biomass compo-
nents suggest the need for a better understanding of the role of P in productivity of these stands in development of pre-
scriptions for management of residues after harvest.
Key words: nutrient cycling, soil nutrients, nitrogen, phosphorus, sulfur cations, Quercus gambelii, Utah.
Gambel oak {Quercus ^^amhelii Nutt.) is
found as a small shruh or large tree on about
3.8 million ha in Colorado, Arizona, New
Mexico, and Utah. It is a clonal species that
sprouts readily after harvest or other distur-
bance from a dense belowground system of
lignotubers and rhizomes (Tiedemann et al.
1987). The lignotubers are similar to those
found on Eucalyptus (Carrodus and Blake
1970). Rhizomes (belowground stems) are also
common in oaks (Muller 1951).
With increasing demands for fuelwood
throughout its range, Q. gambelii is coming
under close scrutiny for its initial value as a
fuelwood source and for continued fuelwood
production potential (Wagstaff 1984, Claiy and
Tiedemann 1992). The density of the wood, its
superior heat-yielding qualities compared with
softwoods (Barger and Ffolliott 1972), and its
sprouting nature (Tiedemann et al. 1987) make
this species ideal for fuelwood management.
In the development of management strate-
gies for sustainable productivity' of Q. gambelii,
an important step is to determine the manner
in which nutrients are distributed among the
abiotic and biotic components of the system.
This information will help develop manage-
ment guidelines so that harvest activities do
not deplete nutrients to the extent that future
site productivity may be jeopardized.
Our objectives were to determine the con-
centrations and total amounts of major plant
nutrients — nitrogen (N), phosphorus (P), sul-
fur (S), calcium (Ca), potassium (K), magne-
sium (Mg), and sodium (Na) — in live and dead
Q. gambelii biomass components and in soil,
understory, and forest floor of a representative
portion of the Q. gambelii ecosystem in central
Utah; and to relate findings to similar studies
in other hardwood stands. This study was a
companion to a study of biomass distrilnition
(Clan' and Tiedemann 1986).
Study Areas and Methods
Eight Q. gambelii stands (plots) were selected
near Ephraim in central Utah. The stands were
on slopes with gradients from 5% to 40%. Soils
'Pacific Northwest Research Station, 1401 Gekeler, La Grande, OR 97850.
2lnterinountain Research Station, 316 East Mvrtle Street, Boise, ID 83702.
119
120
Great Basin Naturalist
[Volume 56
are Typic Calcixerolls formed on alluvium and
colluvium derived from limestone, sandstone,
and shale (Swenson et al. 1981). Soils are cob-
bly loams in the surface 50 cm and very stony
clay loams in the substratum to depths of 150
cm. Elevations of the 8 stands range from
2089 to 2480 m. Average annual precipitation
ranges from 36 to 51 cm, and the annual frost-
free period is 90 to 110 d (Swenson et al.
1981).
Plot sizes varied in approximately inverse
proportion to tree stem density (Clary and
Tiedemann 1986). We attempted to obtain a
sample of the range of stand densities and
stem heights. A 3 X 3-m plot was used for the
densest stand (34,444 stems/ha), a 10 X 10-m
plot for the least dense stand (5000 stems/ha).
Mean ages of stems ranged from 37 to 109 yr
(Claiy and Tiedemann 1986).
At each plot, all live stems were counted
and numbered, and 5 were selected at random
for measurement of height, diameter, biomass,
and nutrient concentration. Sample stems
were cut about 4 cm above the ground, parti-
tioned into 60-cm sections, and weighed in
the field. Live and dead branches and leaves
were removed. A 10-cm portion of each bole
section was placed in a plastic bag, sealed, and
returned to the laboratoiy for determination of
moisture content and nutrient concentrations.
Live branches, dead branches, and leaves from
each tree were bagged, retuined to the labora-
toiy, and oven-dried at 70 °C to constant weight.
After weighing, a sample was taken from each
component for analysis of nutrient concentra-
tion. Standing dead trees were counted on
each plot, and 5 were randomly selected to be
cut and weighed in the field. A section was
taken from each including any attached
branches for determination of moisture and
nutrient concentrations.
Understoiy biomass — including Q. gainbc-
lii < 1 m, other shrubs, herbaceous plants, for-
est floor and down and dead oak — was sam-
pled on three 1-m- subplots randomly located
within each plot, except plot 8, where only 1
subplot was sampled. Plot 8 was sampled at a
different time from plots 1-7, with the main
objective of excavation to determine charac-
teristics of the underground system (Tiede-
mann et al. 1987). We inadvertently collected
only 1 subplot for determination of understoiy
biomass, forest floor, down and dead oak, and
soil. On all subplots, forest floor was collected
to mineral soil. No separation into litter (L),
fermentation (F), and humus (H) layers was
made. Hence, the forest floor includes plant
detritus accumulated above mineral soil in-
cluding down and dead oak <0.5 cm. All sam-
ples were oven-dried at 70 °C and weighed to
determine mass per unit area (kg ha~^) of the
forest floor. Weight of down and dead oak >0.5
cm was assigned to the categoiy of down and
dead oak trees. A small sample of each compo-
nent from each l-m^ plot was used for nutri-
ent analysis. Forest floor samples contained
some soil as a result of wind deposition and
the fact that sampling results in collection of a
small amount of soil from the forest floor/soil
interface. Therefore, weights of forest floor
samples were adjusted for content of soil by
determining weight loss on combustion of
small samples in a muffle furnace at 900 °C.
Combustion of organic materials results in a
small amount of mineral ash residue of 5 g per
100 g of forest floor (Tiedemann 1987b). We
adjusted forest floor weights by this amount.
Soil volume weight (bulk density) was
determined by collecting a 15- to 20-cm-diam-
eter sample to a depth of 30 cm at each of the
subplots after vegetation was harvested and
the forest floor sampled. This was the maxi-
mum depth feasible to collect without using
mechanized digging apparatus because of the
increased rocks, cobbles, roots, and rhizomes
at greater depths. The soil hole was lined with
plastic and the xolume determined by measur-
ing the quantit)' of water to the nearest 10 niL
required to fill the hole. Soil was oven-dried at
70 °C, weighed, and retained for nutrient
analysis. This method of bulk density determi-
nation compares favorably with the paraffin
clod technique (Howard and Singer 1981).
One plot (plot 8) was hydraulicalK' exca-
xated to a depth of 1 m by use of a hydraulic
pump capable of supplying 114 L/min (Tiede-
mann et al. 1987). All roots, rhizomes, and lig-
notubers were removed and transported to the
laboratoiy for drying, dissecting, weighing,
and nutrient analysis. Weight of roots at
depths > 1 m was estimated from taper-weight
relationships established for the first 1 m of
vertical roots. A composite sample of the roots
(<1.0 cm, 1.0-2.5 cm, and >2.5 cm) and rhi-
zomes was taken for nutrient analysis. The
proportion of each component in the sample
was weighted on the basis of its proportion of
total weight.
1996]
Nutrient Distribition in Quercus gambelii
121
Eacli 10-cni bole portion was separated
into 8 equal radial segments. One of these
from each portion \\'as further separated into
heartwood, sapwood, and bark. Samples from
each radial segment were then composited for
each tree prior to analysis. All vegetation sam-
ples were ground to 0.25-mm fineness in
preparation for analysis of nutrient concentra-
tion. Soil samples were sieved through a 2-
nnn mesh screen and ground to 0.125-mm
fineness prior to analysis.
All samples were analyzed for total N by
Kjeldalil digestion followed by titrimetric deter-
mination of distilled ammonium (Bremner
1965); for total P by sulfuric acid-selenium
digestion (Parkinson and Allen 1975) followed
b> molybdenum blue detemiination of P (Olsen
and Dean 1965); for total S by the procedure
of Tiedemann and Anderson (1971); and for
total cations Ca, Mg, Na, and K by atomic
absolution spectroscopy (Jones and Isaac 1969)
on the sulfuric acid-selenium digest used for
total P
Mass per unit area (kg ha"^) of individual
plot values for each individual biomass com-
ponent of trees (leaves, live branches, standing
dead, etc.) from the study of Claiy and Tiede-
mann (1986) were used to convert concentra-
tions of individual nutrients to mass per unit
area (kg ha'^). In the biomass determination
(Clary and Tiedemann 1986), stems were not
partitioned into bark, heartwood, and sapwood.
We determined the percentage by weight of
these 3 components for each bole and con-
verted weights to kg \ra~^ for each plot using
values from Clary and Tiedemann (1986). These
values were then multiplied by concentrations
of individual nutrients for determination of
mass per unit area (kg ha~^) content of nutri-
ents. Mass per unit area (kg ha"^) values for
understory vegetation, down-dead oak, and
the forest floor were multiplied by concentra-
tion values for individual nutrients to deter-
mine mass per unit area of each nutrient. Bulk
density of the upper 30 cm of soil (minus par-
ticles >2 mm) was used to develop mass per
unit area (kg har^) values for soil so we could
convert nutrient concentration values to mass
of indi\'idual nutrients per hectare. Mass per
unit area values of Quercus roots, rhizomes,
and lignotubers in the upper 1 m of the exca-
vated plot plus the extrapolation of larger
(>2.5 cm) vertical roots to their extinction
point was used to convert concentration values
of nutrients to a kg ha"^ basis. Extrapolation
was based on application of taper-weight rela-
tionships for each root.
For purposes of data presentation, nutrient
contents (kg ha~l) of individual aboveground
biomass components were grouped into three
categories: (1) aboveground live overstory and
understory vegetation; (2) standing and down-
dead that includes standing dead trees, dead
branches on live trees, down and dead trees,
and dead branches on the ground >0.5 cm;
and (3) the forest floor that includes all plant
detritus above mineral soil except for Quercus
branches >0.5 cm.
Analysis of variance in a randomized com-
plete block design with the 8 individual plots
as blocks was used to determine differences in
concentration among aboveground biomass
components for each nutrient constituent
(Steel and Torrie 1960). Biomass component
was the main effect term in the analysis. Val-
ues for the 5 individual trees and for the 3 for-
est floor and understory subplots in each of
the 8 plots (blocks) were pooled, and the means
were used in the analysis of variance. Statistical
comparison with underground biomass com-
ponents was not possible because this was de-
termined on only 1 plot. Where the F-test was
significant, differences among individual bio-
mass components were determined using the
LSD test (Carmer and Swanson 1971). Signifi-
cant differences are expressed at P < 0.01. No
statistical tests were applied to kg har^ nutri-
ent content data because individual compo-
nents were summed to provide more inclusive
groupings. For example, live aboveground bio-
mass includes oak leaves, live branches, heart-
wood, sapwood, bark, and understory leaves
and stems.
Results and Discussion
Nutrient Concentrations
There were no significant differences in con-
centrations of nutrients in biomass (F < 0.01)
among plots (blocks) for any nutrient con-
stituent except Ca. Differences among bio-
mass components were highly significant for
eveiy nutrient constituent.
Nitrogen concentrations in the forest floor
and in Quercus leaves were significantly higher
than in any other component (Table 1). Under-
story leaves were significantly lower in N con-
centration than the forest floor or Quercus
122 Great Basin Naturalist [Volume 56
T.ABLE I. Concentration (percent) of nutrient constituents in biotic and abiotic components of Qiiercus gambelii
Nutrient
Leaves
Live
branches
Heartwood
N
LSD 0.01
1.57
= 0.08
0.56
0.15
P
LSD 0.01
0.21
- 0.024
0.03
0.003
S
LSD 0.01
0.08
= 0.014
0.03
0.03
Ca
LSD 0.01
0.91
= 0.30
0.90
0.17
Mg
LSD 0.01
0.35
= 0.29
0.16
0.02
K
LSD 0.01
0.68
= 0.18
0.36
0.33
Na
LSD 0.01
0.04
= 0.007
0.01
0.01
Sapwood
Bark
Dead Standing
liranches dead trees
0.27
0.02
0.02
0.17
0.04
0.15
0.002
0.62
0.02
0.04
1.55
0.20
0.32
0.01
0.55
0.02
0.04
0.98
0.14
0.26
0.01
0.35
0.01
0.04
1.00
0.08
0.21
0.01
"Comparisons anions ahoveiiiouiHl 1)Ioiikiss coiiipiiiUTits miK.
leaves. We did not obsei^ve increases in N con-
centration of the forest floor that usually
accompany decomposition, mineralization, and
leaching of other constituents from the fallen
overstory leaves (Bocock 1963, Gosz et al.
1973). In a litter bag study Klemmedson (1992)
measured a 60% increase in N concentration
in Q. gambelii leaves in the litter layer over a
750-d time span. Differences between our
observations and those of Klemmedson were
probably because we report comparisons be-
tween Quercus leaves and the entire forest
floor, whereas his comparisons were for the lit-
ter layer only Lowest concentrations of N were
observed in the heartwood. Standing dead and
down-dead trees were both higher in N con-
centrations than were heartwood and sapwood
of living stems. This probably resulted from
selective decomposition and loss of other ele-
ments causing an increase in the concentration
of N in standing dead and down-dead trees.
Concentration of N in the upper 30 cm ot
soil (0.42) was greater than would be expected
for this site. According to Jenny (1941), the
normal range of soil N for semiarid sites is
0.10%-0.25% for the surlace 10 cm. The high
content of N in these soils can prol)ably be
attributed to 2 principal factors: (1) the high
clay content is conducive to retention of high
levels of organic N (Klenmiedson and Jenn\'
1966, Millar et al. 1966); and (2) the extraordi-
nary accumulation of forest floor (37,348 kg
ha~l) at this site (Claiy and Tiedemann 1986)
provides a continuous supply of N to the soil
through decomposition and leaching.
Leaves of understory plants (0.27%), Quer-
cus leaves (0.21%), and forest floor (0.12%) had
highest concentrations of R Differences among
these 3 components were significant. Reduced
concentration of P in the forest floor compared
to Quercus leaves corresponded to obsei^va-
tions of Klemmedson (1992). Concentration of
P in Q. gambelii leaves at the surface of the
forest floor began to decrease shortly after
deposit and declined steadily for 500 d to
about 60% of original concentration. Concen-
tration then leveled off for the remaining 250
d of the experiment. Our lowest levels of P
occurred in the heartwood (0.003%). Although
there were some significant diiferences among
other biomass components, the acttial differ-
ences were slight and probably of little biolog-
ical significance. Total P in soil (0.02%) was
substantially below normal levels, which are
0.09%-0.13% for soils of the United States
(Parker et al. 1946).
Concentrations of S were greatest in forest
floor (0.12%) and understory leaves (0.11%),
and there was no significant difference between
these 2 components. However, S concentra-
tion in both was significantly higher than in
Quercus leaves. Lowest S concentrations in
abo\ eground components were in the sapwood
and heartwood. Our comparisons of N and S
1996]
Nutrient Distribution in Quercus gamkfaai
123
(■ci)s\steni,s in cfiital Utah.
Understoiy Understory Down-dead I'orfst
leaves stems trees lloor
Roots and
rlii/.onies
Lif^notuhers Soil
1.46
0.54
0.43
1.66
0.44
0.33
0.42
0.27
0.0.5
0.01
0.12
0.03
0.02
0.02
0.11
0.04
0.06
0.12
0.04
0.03
0.04
0.98
0.61
0,76
2.67
0.97
1.15
1.29
0.40
0.1"
0.11
1.15
0.14
0.09
1.92
1.14
0.64
0.0'
0.43
0.21
0.14
0.87
0.008
0.02
0.005
0.06
0.02
0.008
0.08
levels in the forest floor with Quercus leaves
presented an anomaly. We would expect S
comparisons between forest floor and Quercus
leaves to be similar to those for N, because S
is a companion to N in several amino acids
(Allaway and Thompson 1966, Coleman 1966).
Klemmedson's (1992) observations bear this
out because both N and S concentrations in
Quercus leaves increased about 60% over a
750-d period after deposition at the surface of
the forest floor. However, when we compared
Quercus leaves and the entire forest floor, it
appeared that N and S responded differently
over the long periods required for develop-
ment of the forest floor. Nitrogen concentra-
tion tended to remain constant and S concen-
tration increased over time. Mineralization of
S in deeper layers of the forest floor may pro-
ceed more slowly than mineralization of N,
thereby resulting in an increase in S concen-
tration. Products of decomposition for N may
also be more mobile than those for S.
Total S concentration in soil (0.04%) was in
the middle of the range reported for U.S. soils,
0.01-0.06 (Burns 1968). The ratio of N:S of
10:1 in soil indicates that the S level is great
enough that N will be efficiently utilized for
the formation of plant proteins (Black 1968,
Burns 1968).
Concentrations of the 4 measured cations,
Ca, Mg, K, and Na, were generally higher in
the soil than in any plant component. Excep-
tions were higher concentrations of Ca in the
forest floor and in the bark of Quercus trees
and K in understoiy leaves.
Calcium concentrations in the forest floor
layer were more than 2.5 times greater than
Quercus leaves. The content of Ca in bark was
nearly 10 times greater than heartwood or sap-
wood. Quercus leaves, live branches, dead
branches, standing dead trees, and down-dead
trees were all comparable in Ca concentration.
Magnesium concentrations in biomass com-
ponents were highest in the forest floor
layer — approximately 3 times greater than in
Quercus and understoiy leaves. In contrast to
Ca patterns, Mg concentrations in live branches
and standing dead and down-dead trees were
significantly lower than in Quercus leaves.
Understoiy leaves were significantly higher
in K concentration (1.14%) than were Quercus
leaves (0.68%) or understory stems (0.64%).
Potassium concentrations were about equal for
live branches, heartwood, and bark, and about
half the concentration found in Quercus leaves.
Concentration of K in forest floor was substan-
tially lower than in Quercus leaves and may
reflect the ease with which K is leached from
the forest floor relative to the other cations
(Attiwill 1968).
Highest concentrations of Na occurred in
Quercus leaves and in the forest floor Differ-
ences among other biomass components were
124
Great Basin Naturalist
[Volume 56
minor, even though some were statisticalK'
significant.
Comparisons of cation knels in Quercus
leaves with levels in the forest floor were vari-
able between our study and results of the lit-
ter bag study of Klemmedson (1992). We
showed significantly greater Ca and Mg in the
forest floor than in Quercus leaves. Klemmed-
son (1992) found similar increases in Ca in
Quercus leaves over 750 d. However, Mg con-
centration in his study declined to about 80%
of the level in fresh leaves over the 750-d
study. Differences in K concentration that we
found between Quercus leaves and the forest
floor were not nearly as great as the decline in
K concentration over time in the litter layer
measured by Klemmedson (1992). Potassium
concentration in Quercus leaves declined
about 70% in 500 d and then stabilized to the
end of the 750-d study. Differences between
Klemmedson's observations and ours were
probably a result of the fact that he studied
changes in nutrient concentration in the litter
layer and our comparisons were with the
entire forest floor
There is little information on the concen-
trations of nutrients in biomass components in
western hardwood stands. There are 2 appar-
ent reasons for this. Compared with the east-
ern United States, the area occupied by stands
of hardwood species in the West is minor
Therefore, until recently, western hardwoods
have not been viewed as an economically im-
portant resource; rather, they were considered
weed species because they were assumed to
compete with marketable coniferous trees or
with understory forage-producing species.
With emerging demands for fuelwood and
new markets for unique woods for furniture,
there is increased awareness of the value of
western hardwoods and, especially, Q. gainbelii
(Wagstaff 1984, Claiy and Tiedemann 1992).
Nutrient concentrations of leaves agreed
closely with those reported by Klemmedson
(1992) for Q. gambelii in northern Arizona.
Bartos and Johnston (1978) determined the
concentrations and proportions of indi\ idual
nutrients in the various components of 3
clones o( Populus tronuloides Vlich.x. (cjuaking
aspen) trees in Utah and Wyoming but did not
consider the forest floor, understoiy, and down-
dead components of the nutrient pool. Con-
centrations of N in the various tree compo-
nents of Q. grnnhclii and P. freiiiuloides were
comparable except for higher concentrations
of N (2.5%) in leaves of P. tremuloides; concen-
trations of F, K, and Ca were similar for all tree
components. Sodium concentrations were gen-
erally greater in Q. gambelii than in P. tremu-
loides. Concentrations of N, I^ and S in live
aboveground biomass of Q. gambelii were
comparable to those reported for Q. robur in
Russia (Rodin and Bazilevich 1967) and in
Belgium (Duvigneaud and Denaeyer-De Smet
1970). Concentrations of N in forest floor and
dead branches also were comparable to values
for southern and eastern U.S. Quercus stands
(Lang and Forman 1978). Concentrations of
cations in our study did not agree as well with
those presented in the literature as for N, I^
and S. For example, Q. gambelii forest floor
concentrations of K and Mg were 3 and 8
times greater than those reported for Q. robur
Calcium concentrations in Q. gambelii were
substantialK' greater than those observed in
other studies in forest floor live branches,
dead branches, standing dead trees, and
down-dead trees.
Distribution of Nutrient Capital
Among Components
Comparisons of nutrient distribution be-
tween above- and belowground components
must be considered from the perspective that
our soil sampling was restricted to the upper
30 cm because of rock and the massixe under-
ground structures of Q. gainbelii. The actual
zone of rooting and nutrient acquisition was
undoubtedly much greater than the area we
sampled. Therefore, our estimates of the pro-
portions of nutrients in aboveground compo-
nents were likely to be higher than if the
entire rooting zone had been sampled. Also,
the kg ha~^ estimates were for the area of the
actual clone sampled. Clones ofQ. gambelii do
not occupy the entire area of the sites on
which they occur Most studies take into
account the high- and l()w-densit\ areas of
tree occupancy in determining nutrient distri-
bution. Therefore, in making projections to an
areal basis, the actual area occupied b>' Q.
gambelii clones must be considered.
The greatest proportion of total nutrient
capital sampled was contained in the soil
(Table 2). Of the total capitals of individual
nutrients, 82%-99% were contained in the
soil. Aboveground accumulations of indixidual
nutrients in Ii\c' biomass, standinii and down-
1996]
Nutrient DisiiiiBi iion in Qiercvs gambelii
125
Tablk 2. Distiil)iiti()n of nutrients anions liiomass, foR'st lloor, standinii plus dowii-ilcad, and soil components of {).
iciinhi'Iii stands.
Live-'
Standing
Live''
abo\e-
plus
Total
below-
Total'l
ground
down-
Forest''
above-
groniid
live
Total'
Nutrient
biomass
dead
floor
ground
biomass
biomass
Soil-'
capital
Nitrogen (kg ha"')
245
140
654
1039
270
515
9500
10810
% of total
abovegroiuid
24
13
63
% of total capital
10
2
88
Phosphorus (kg ha"^)
19
4
48
71
19
38
410
500
% of total
alioxeground
27
5
68
% of total capital
14
4
82
Sulfur (kg ha-i)
19
13
46
78
22
41
946
1046
% of total
aboveground
24
17
59
% of total capital
8
2
90
Calcium (kg ha-^)
334
303
1167
1804
924
1258
28844
31571
% of total
abovegroimd
18
17
65
% of total capital
6
3
91
Magnesium (kg ha"l)
62
35
381
478
63
125
42485
43023
% of total
abo\'egroimd
13
7
80
% of total capital
1
<1
99
Potassium (kg ha~l)
201
72
144
417
116
317
20268
20801
% of total
aboveground
48
18
34
% of total capital
2
<1
98
Sodium (kg ha"')
7
4
20
31
7
14
1765
1804
% of total
aboveground
22
13
65
% of total capital
2
<1
98
Total (kg ha-l)
887
571
2460
3918
1421
2308
% of total
abovegroimd
23
14
63
9( of total in
living biomass
3S
62
"Includes living aboveground overstory and understory vegetation.
"Includes all forest floor layers above mineral soil.
'Includes roots, rhizomes, and lignotubers in the upper 100 cm of soil.
"Standing crop plus belowground biomass.
■^Upper 30 cm of soil.
■Standing crop plus standing and down-dead plus forest floor plus belowground biomass plus soil.
dead, and forest floor ranged from 31 kg ha~^
for Na to 1804 kg ha-l foj. q.^ Proportions of
total capitals of N, E and S in aboveground
components were highest with 10%, 14%, and
8%, respectively. The proportion of N (the most
widely reported nutrient) in aboveground
components (10%) was comparable to that
described for other semiarid and temperate
forest and woodland ecosystems (Klennnedson
1975, Brown 1977, Tiedemann 1987a).
The forest floor was the most important
aboveground reservoir of nutrients with 63%
of the total accumulation above ground. Accu-
mulations of individual nutrients in the forest
floor ranged from 20 to 1167 kg ha'^ and con-
stituted 34%-80% of the aboxeground capitals.
126
Great Basin Naturalist
[Volume 56
Total nutrient content of the forest floor in our
Q. gambelii clones (2460 kg ha"^) substantially
exceeded the range described by Lang and
Forman (1978) in their summary for U.S.
Quercus forests (206 kg har^ [Yount 1975] to
1462 kg ha-l [Gosz et al. 1976]). Greater accu-
mulation of Ca in the forest floor layer (1167
kg ha~^) compared with that reported b>' other
obsei-vers (98-400 kg ha~^; Lang and Forman
1978) accounted for much of the difference in
total accumulation of nutrient elements in Q.
gambelii compared with other Quercus stands.
Also, forest floor biomass accumulation in our
Q. gambelii stands (37,348 kg ha"^; Claiy and
Tiedemann 1986) was near the upper limit
(46,800 kg ha"l) of that presented for U.S.
Quercus forests (Lang and Forman 1978).
The massive belowground system of ligno-
tubers, rhizomes, and roots comprised 56% of
the total biomass of Q. gambelii (Clary and
Tiedemann 1986) and contained <\% to 4% of
the total of the capitals of individual nutrients.
However, relative to the total nutrient accu-
mulation in live biomass, the live belowground
component was an important storage area con-
taining 37%-74% of the individual nutrient
accumulations. The proportion of total nutri-
ents in belowground biomass (61%) substan-
tially exceeded the range for deciduous forests
worldwide (30%-40%) summarized by Rodin
and Bazilevich (1967). This finding supported
the conclusions of Chattaway (1958), Robbins et
al. (1966), and Blake and Canodus (1970) that
storage of nutrients is an important fonction of
belowground components such as lignotubers.
Total content of nutrients in the entire
organic component (total live and dead above-
ground and belowground biomass) of our Q.
gambelii stands (5339 kg ha~l) was in the mid-
dle of the range for deciduous forests world-
wide (2000-7500 kg ha-l) summarized by
Rodin and Bazilevich (1967). Similarly, total
nutrient content of live biomass (2308 kg ha"^)
was comparable to values for oak forests in
Russia (2600-3400 kg ha-l; Ro^Iji-^ .j,^j Bazile-
vich 1967).
Worldwide, leaves usualh' constitute 8%'-10%
of the store of mineral elements in plant bio-
mass (Rodin and Bazilevich 1967). Mineral
element accumulation in Q. gambelii leaves
and understoiy leaves (245 kg ha-l; not shown
in Table 2) comprised 11% of the total mineral
content of live biomass and was within the rel-
atively constant, narrow range of 200-300 kg
ha-l normally found in leaves reported by
Rodin and Bazilevich (1967).
Conclusions
Gambel oak appears to be unique from
other deciduous forests in the accumulation of
nutrients in the forest floor and in below-
ground biomass components. Both were major
areas of nutrient accumulation. The leaves, in
contrast, were a minor storage area.
Accumulation of nutrients in aboveground
living and dead components expressed as a
proportion of total site nutrients was similar to
that reported for other semiarid and temper-
ate forest habitats. The quantity of N, the most
commonly measured nutrient stored in the
forest floor, also agreed well with this litera-
ture. It should be noted that had we been able
to sample a larger proportion of the total root-
ing zone, the proportion of the total nutrient
capital aboveground would likely have been
smaller.
Low levels of P in the upper 30 cm of soil
suggest that this element may limit productiv-
ity of Q. gambelii. Because of potential limita-
tions in the soil, accumulation of 14% (71 kg
ha-l) Qf jIjp jqj-^i ecosystem P in aboveground
living and dead components, we suggest cau-
tion in the wa)' the forest floor and residues
are managed. Fuelwood harvest followed by
removal of residues by broadcast burning could
cause large losses of P, depending on degree
of consumption of organic matter and fire
temperatures (Covington and DeBano 1988,
DeBano 1988). This loss may reach 60% (of 71
kg ha~l) if fuels are totalK' consumed (Raison
et al. 1985). However, such losses need to be
weighed against changes in P availability that
result from burning. In his summaiy of plant-
and litter-contained nutrients, DeBano (1988)
indicated that fire-induced increases in P avail-
abilit>' decline and reach pre-fire levels within
1 yr DeBano and Klopatek (1988) showed that
inorganic P is released In prescribed burning
but is quickly immobilized and ma\' not be
readily availalile for plant growth.
Although there are also substantial accu-
mulations of N and S in aboN egroimd biomass
and these are sensitive to losses from vola-
tilization (Knight 1966, Tiedemann 1987b),
they are not limiting in the soil and quantities
are likcK sutticient to replenish losses.
1996]
Nutrient Distribution in Quercus gambeui
127
Fertilizer amendment with P may warrant
consideration as a means of impro\in^ {). gain-
helii productivity after liarvest. This decision,
however, should be based on soil tests to
determine the a\ ailabilit\ of E
Literature Cited
Aluawav, W. H., and J. E Thompson. 1966. Sulfur in the
nutrition of plants and animals. Soil Science 101:
240-247.
Attiwill, P M. 1968. The loss of elements from decom-
posing litter Ecology 49: 142-145.
Barger, R. L., and R E Fkolliott. 1972. The physical
characteristics and utilization of major woodland
tree species in Arizona. USDA Forest Service
Research Paper RM-83. 80 pp. Rocky Mountain For-
est and Range E.xperiment Station, Fort Collins, CO.
Bartos, D. L., and R. S. Johnston. 1978. Biomass and
nutrient content of quaking aspen at two sites in the
western United States. Forest Science 24: 273-280.
Black, C. A. 1968. Soil-plant relationships. 2nd edition.
John Wiley and Sons, Inc., New York. 792 pp.
Blake, T. J., and B. B. Carrodus. 1970. Studies on the
lignotuhers oi Eucalyptus obliqua tHeri. II. Endoge-
nous inhibitor levels correlated with apical domi-
nance. New Phytologist 69: 1073-1079.
Bocock, K. L. 1963. Changes in the amounts of diy matter
nitrogen, carbon, and energy in decomposing wood-
land leaf litter in relation to the activities of soil
fauna. Journal of Ecology 52: 27.3-284.
Bremner, J. M. 1965. Total nitrogen. Pages 595-624 in
C. A. Black, D. D. Evans, L. E. Ensminger et al.,
editors. Methods of soil analysis. Part 2, Chemical
and microbiological properties. Agronomy 9, Ameri-
can Societ\' of Agronomy, Inc., Madison, WI.
Brown, T. H. 1977. Nutrient distribution of two eastern
Washington forested sites. Unpublished masters the-
sis, Washington State University, Pullman. Ill pp.
Burns, G. R. 1968. O.xidation of sulfur in soils. Sulphur
Institute Technical Bulletin 1. 41 pp.
Carmer, S. C, and M. R. Swanson. 1971. Detection of
differences between means: a Monte Carlo study of
five pairwise multiple comparison procedures.
Agronomy Journal 63: 940-945.
Carrodus, D. D., and T J. Blake. 1970. Studies on the lig-
notuber oi Eucalyptus obliqua EHeri. I. The nature
of the lignotuber New Ph\tologist 69: 1069-1072.
Chattawav; M. M. 1958. Bud development and lignotu-
ber formation in eucalypts. Australian Journal of
Botany 6: 103-115.
Cl.\rv, W. P, and a. R. Tiedemann. 1986. Distribution of
biomass within small tree and shrub form Quercus
f^ainhclii. Forest Science 32: 232-242.
. 1992. Ecology and values of Cambel oak wood-
lands. Pages 87-95 in P E Ffolliott et al., technical
coordinators. Ecology and management of oak and
associated woodlands: perspectives in the south-
western United States and northern Mexico, 27-30
April 1992, Sierra Vista, AZ. USDA Forest Sewice,
General Technical Report RM-218. Rocky Mountain
Forest and Range Experiment Station, Fort Collins,
CO.
Coleman, R. 1966. The importance of sulfur as a plant
nutrient in world crop production. Soil Science 101:
230-240.
CoviNCTON, W. W, AND L. E DkBano. 1988. Effects of
fire on pinyon-jimiper soils. Pages 78-86 in J. S.
Krammes, technical coordinator. Effects of fire man-
agement of southwestern natural resources. USDA
Forest Service, General Technical Report RM-191.
Rocky Moimtain Forest and Range Experiment Sta-
tion, Fort Collins, CO.
DeBano, L. E 1988. Effects of fire on the soil resource in
Arizona chaparral. Pages 65-77 in J. S. Krammes,
technical coordinator. Effects of fire management of
southwestern natural resoiux'es. USDA Forest Service,
General Technical Report RM-191. Rocky Mountain
Forest and Range Experiment Station, Fort Collins,
CO.
DeBano, L. E, and J. M. Klopatek. 1988. Phosphorus
dynamics of pinyon-juniper soils following simulated
bmning. Soil Science Society of America Journal 52:
271-277.
Duvigneaud, R, and S. Denaever-De Smet. 1970. Bio-
logical cycling of minerals in temperate deciduous
forests. Pages 199-225 in D. E. Reichle, editor,
Analysis of temperate forest ecosystems. Springer-
Verlag, Berlin, Heidelberg, Germany.
Gosz, J. R., G. E. Likens, and F H. Bormann. 1973.
Nutrient release from decomposing leaf and branch
litter in the Hubbard Brook Forest, New Hamp-
shire. Ecological Monograph 43: 17.3-191.
. 1976. Organic matter and nutrient dynamics of
the forest and forest floor in the Hubbard Brook For-
est. Oecologia (Berlin) 22: 305-320.
Howard, R. E, and M. J. Singer. 1981. Measuring forest
soil bulk density using irregular hole, paraffin clod,
and air permeability. Forest Science 27: 316-322.
Jenny, H. 1941. Factors of soil formation. McGraw-Hill
Book Co, Inc., New York. 281 pp.
Jones, J. G., Jr., and R. A. Isa,\c. 1969. Comparative ele-
mental analysis of plant tissue by spark emission and
atomic absorption spectroscopy. Agronomy Journal
61: 393-394.
Klemmedson, J. O. 1975. Nitrogen and carbon regimes in
an ecosystem of young dense ponderosa pine in Ari-
zona. Forest Science 21: 163-168.
. 1992. Decomposition and nutrient release from
mixtures of Cambel oak and ponderosa pine leaf lit-
ter Forest Ecology' and Management 47: 349-361.
Klemmedson, J. O., and H. Jenny. 1966. Nitrogen avail-
ability in California soils in relation to precipitation
and parent material. Soil Science 102: 215-222.
Knight, H. 1966. Loss of nitrogen from the forest floor by
burning. Forestry Chronicle (June): 149-152.
L\NG, G. E., AND R. T. T. FoRMAN. 1978. Detrital dynamics
in a mature oak forest; Hutcheson Memorial Forest,
New Jersey. Ecology 59: 580-595.
Millar, C. E., L. M. Turk, and H. O. Foth. 1966. Funda-
mentals of soil science. John Wiley and Sons, Inc.,
New York. 491pp.
Muller, C. H. 1951. The significance of vegetation repro-
duction in Quercus. Madrono 11: 129-137.
Olsen, S. R., and L. a. Dean. 1965. Phosphorus. Pages
10.35-1049 in C. A. Black, D. D. Evans, L. E. Ens-
minger et al, editors. Methods of soil analysis. Part
2, Chemical and microbiological properties. Agron-
omy 9, American Societ>' of Agronomy, Inc., Madi-
son, WI.
128
Great Basin Naturalist
[Volume 56
Parker, E W, J. R. Adams, K. G. Clark et al. 1946. Eer-
tilizers and lime in the United States: resources, pro-
duction, marketing, and use. USDA Miscellaneous
Publication 586. Washington, DC. 94 pp.
Parkinson, J. S., and S. E. Allen. 1975. A wet oxidation
procedure suitable for the determination of nitrogen
and mineral elements in biological material. Com-
munications in Soil Science and Plant Analysis 6;
1-11.
Raison, R. J., R K. Khanna, and R V. Woods. 1985.
Mechanisms of element transfer to the atmosphere
during vegetation fires. Canadian Joinuial of Forest
Research 15: 132-140.
RoBBiNS, W W, T. E. Weier, and C. R. Stocking. 1966.
Botany: an introduction to plant science. John Wiley
and Sons, Inc. 614 pp.
Rodin, L. E., and N. I. Bazilevich. 1967. Production and
mineral cycling in teiTestrial vegetation. [English
translation by G. E. Eogg, editor] Oliver and Boyd.
Edinburgh-London. 288 pp.
Steel, R. G. D., and J. H. Torrie. 1960. Principles and
procedures of statistics. McGraw-Hill, New York.
481 pp.
SwENSON, J. L., D. Beckstrand, D. T. Erickson et al.
1981. Soil sui-vey of Sanpete Valley area, Utah. USDA
Soil Consei"vation Sei'vice, and U.S. Department of
the Interior Bureau of Land Management, in coop-
eration with Utah Agriculture E.xperiment Station
and Utah State Division of Wildlife Resources. 179
pp.
Tiedemann, a. R. 1987a. Nutrient accumulations in
pinyon-juniper ecosystems — managing for future
site productivity. Pages 352-359 in R. L. Everett,
compiler. Proceedings — Pinyon Juniper Conference,
Reno, NV, 13-16 Januaiy 1986. USDA Eorest Ser-
vice, General Technical Report INT-215. Intermoun-
tain Research Station, Ogden, L^T.
. 1987b. Combustion losses of sulfur from forest
foliage and litter Forest Science 33: 216-223.
Tiedemann, A. R., and T. D. Anderson. 1971. Rapid analy-
sis of total sulfur in soils and plant material. Plant
and Soil 35: 197- 200.
Tiedemann, A. R., W. P Cl^ry, and R. J. Barbour. 1987.
Underground systems of Gambel oak {Qiiercus gam-
belii) in central LItah. American Journal of Botany
47: 1065-1071.
Wagstafe E J. 1984. Economic considerations in use and
management of Gambel oak for fuelwood. USDA For-
est Sendee, General Technical Report INT-165. 8 pp.
Intermountain Forest and Range E.xperiment Sta-
tion, Ogden, UT.
YOUNT, J. D. 1975. Forest-floor nutrient d\'namics in
soutliem Appalachian hardwood and white pine plan-
tation ecosystems. Pages 598-608 in F. G. Howell,
J. B. Gently, and M. H. Smith, editors. Mineral cycling
in southeasteiTi ecosystems. Energy Research and
Development Administration Symposium series.
Technical Information Center Springfield, VA.
Received 17 October 1995
Accepted 11 March 1996
Great Basin Naturalist 56(2), © UMi p|i. 129-134
COMPARISON OF TWO ROADSIDE SURVEY PROCEDURES
FOR DWARF MISTLETOES ON THE SAWTOOTH
NATIONAL FOREST IDAHO
RolH'rt L. Mathiasm', jaiiirs T. Iloirman-, Jolin C. Cuyon'^ and Linda L. Wadlci^Ir'
Abstract. — Two roadside surveys were condueted tor dwarl inistlctoes parasitizing lodgepole pine and Donglas-lir
on the Sawtooth National Forest, Idaho. One sur\e\ used xariahle-radins plots loeated less than 150 in From roads. The
2nd survey used variable-radius plots established at 2()()-in intewals along 16()0-m transects run peipendieular to the
same roads. Estimates of the incidence (percentage of trees infected and percentage of plots infested) and severit\- (aver-
age dwarf mistletoe rating) for both lodgepole pine and Douglas-fir dwari' mistletoes were not significantly different for
the 2 suney methods. These findings are further evidence that roadside-plot surveys and transect-plot surveys con-
ducted awa\- from roads pro\'ide similar estimates of the incidence of dwarf mistletoes for large forested areas.
Key ivords: dwaij mistletoes, surveys, lodgepole pine, Douglas-fir.
Dwarf mistletoes {Arccuilxohiwn spp.) are
damaging disease agents in many western
forests (Hawksworth and Wiens 1995). In the
Intermountain West lodgepole pine {Pinus con-
toiia Dongl. ex Loud.) and Douglas-fir [Pseudo-
tsiigo inenziesii [Mirb.] Franco) are the most
commonly infected trees (Hawksworth and
Wiens 1972, 1995, Hoffman 1979). Each of
these hosts is parasitized by a different dwarf
mistletoe: lodgepole pine dwarf mistletoe (A.
americanwn Nutt. ex Engelm.) and Douglas-
fir dwarf mistletoe (A. douglasii Engelm.).
Severe infection by these parasites is often
associated with tree mortality, reduced growth
and cone production, tree deformity, and pre-
disposition to attack by other diseases and/or
insects (Hawksworth and Wiens 1995). There-
fore, resource managers in many private, state,
and federal land-management agencies imple-
ment management activities designed to reduce
the damage associated with dwarf mistletoes.
Because information on the incidence and
severity of these pathogens is required by
resource managers for making decisions regard-
ing dwarf mistletoe management, surveys are
commonly conducted in designated manage-
ment units (stands) and over larger areas, such
as national forests.
Surveys of dwarf mistletoe infection over
large areas frequently combine roadside recon-
naissance information with data collected using
variable-radius or fixed-area plots located near
roads (roadside-plot surveys) for estimating
the incidence (percent of trees or plots in-
fected) and severity (intensity of infection in
individual trees; Hawksworth 1956, 1958,
Hawksworth and Lusher 1956, Andrews and
Daniels 1960, Graham 1960, 1964, Dooling
1978, Hoffman 1979, Johnson et al. 1980,
Johnson et al. 1981, Hoffman and Hobbs 1985,
Merrill et al. 1985, Maffei and Beatty 1988).
Roadside reconnaissance surveys consist of
driving roads at slow speed and recording \isual
estimates of dwarf mistletoe infection within a
short distance from the roadside, usualK 20 m.
Dwarf mistletoe incidence is estimated b\
determining the ratio of the number of kilo-
meters surveyed adjacent to infected trees to
the total kilometers surveyed adjacent to stands
predominated by host trees (Dooling 1978).
Roadside-plot surveys involve locating plots
near roads at specific intervals and collecting
tree data including species, diameter, height,
age, and mistletoe severity on each plot. Dwarf
mistletoe incidence has typically been repre-
sented by the percentage of plots infested
with mistletoe, rather than the percentage of
trees infected in all plots (Dooling 1978).
Roadside surveys have the benefit of allow-
ing large areas to be surveyed rapidly and
^Idaho Department of Lands, Box 670, Coeur d'Alene, ID 83816.
^Forest Pest Management, USDA Forest Service, 1750 Front Street, Boise, ID 83702.
^Forest Pest Management, USDA Forest Service, 4746 South 1900 East, Ogden, UT 84401.
■tUSDA Forest Service, 524 25th Street, Ogden, UT 84403.
129
130
Great Basin \atlii\li.st
[Volume 56
inexpensive!)'. In addition, roadside sunexs
concentrate efforts in areas that are accessible
and more likely to be considered for manage-
ment actions. Concerns about the relialiilit\ of
roadside survey methods are primarily related
to the bias that may be encountered by sam-
pling mistletoe incidence and se\'erity near
roads because roads are t\picall\' constructed
according to topographic features (in drainages
or along ridgetops) rather than randomly or
systematically located throughout the suney
area. Since there is exidence that d\\ arf mistle-
toe distiibution is related to topograph)" ( Hawks -
worth 1959, 1968), these concerns need to be
considered when conducting dwarf mistletoe
suneys oxer large forested areas.
Because few sun e)s ha\e compared data
collected from roadside reconnaissance or
roadside-plot sun'e)'s with data collected from
more intensixe, random or s) stematic sun e) s
for dwarf mistletoes over large areas (Haw ks-
worth 1956, 1958, Johnson et al. 1981, Merrill
et al. 1985), we initiated this study to compare
dwarf mistletoe incidence and sexerity esti-
mates obtained from roadside-plot surve)s
xvith those from transect-plot surx'e)S that
sampled areas at greater distances from roads.
We sune)ed 3 distiicts of the Saxxtooth National
Forest, Idaho, because this national forest is
representatix'e of forests in the Intennountain
West xvhere lodgepole pine and Douglas-fir
are the predominant tree species and dxxarf
mistletoes are common (Hoffman 1979, Hoff-
man and Hobbs 1985).
Methods
We used a roadside-plot sun e) and a tran-
sect-plot suney to collect dxx^arf mistletoe in-
cidence and sexerit)- data in 3 adjacent distiicts
(Ketchum Ranger District, Fairfield Ranger
District, and Saxvtooth National Recreation
Area) of the Sawtooth National Forest, Idaho,
in 1990. We surveyed each district b)' arbitrar-
ily selecting a major road S) stem in each toxxn-
ship containing > 10 sections of f ederall) man-
aged land. ToxxTiships xvitli no roads or with fexv
roads were not sampled. Road systems were
chosen before fieldxvork began, and adjustments
were made in the field onK xvhen selected I'oad
systems xvere closed or impassable.
Roadside-plot Suney
Field crexvs arbitraril) chose a starting ref-
erence point on each selected road sxstem.
Starting reference points xxere landmarks that
could easily be relocated such as a bridge,
stream crossing, or road junction. Crexxs droxe
a distance of 800 m from the starting reference
point toxvard the center of each toxxnship.
Thex' then selected a compass bearing peipen-
dicular to the right-hand side of the road and
located an end point 120 m from the road.
Three 20 basal area factor x ariable-radius plots
(iDoint samples; Aver)' and Burkhart 1983) xvere
established 40 m from this end point at com-
pass bearings of 240°, 120°, and 0° from the
compass bearing used to locate the end point.
Crexvs then droxe another 800 m doxxn the
road and established a 2nd cluster of 3 xari-
able-radius plots using the same procedure.
For each plot tree the folloxving information
xvas recorded: plot number, species, diameter
at 1.37 m aboxeground (nearest 0.25 cm), sta-
tus (live or dead), and dxvarf mistletoe rating
(DMR, 6-class system; Haxvksxvorth 1977). If a
plot did not contain trees, it xvas recorded as
nonstocked.
Transect-plot Suney
A 1600-m (approximately 1-mi) transect
perpendicular to the road xxas run along the
same compass bearing used for establishing
the 1st set of roadside plots (800 m from the
starting reference point) in each toxxiiship sur-
X exed. A 20 basal area factor xariable-radius
plot xxas located exen* 200 m along each tran-
sect for a total of 8 plots. Infonnation recorded
for plot trees xvas the same as aboxe.
Analyses
The incidence of each species of dxxarf
mistletoe (percentage of trees infected) xx'as
calculated for each set of roadside plots (up to
6 plots) and each set of transect plots (up to 8
plots) for each toxvnship. Incidence xvas calcu-
lated on a per-hectare basis b) multipKing b)'
pei-hectare conxersion factors based on 2.54-
cm-diameter classes for 20 basal area factor
xariable-radius plots (Axen and Burkhart 1983).
^\'eighted dxxarf mistletoe ratings xvere calcu-
lated b) multipKing the DMR of each tree by
the per-hectare conx ersion factors also. These
xveighted xalues xvere used to calculate the
mean percentage of trees infected and mean
dxvarf mistletoe rating for each suney proce-
dure in each toxxnship on a per-hectare basis.
These x'alues xx'ere then used to calculate the
1996]
Roadside Suhneys eoh Dwarf Mistletoes
131
percentage of trees infected and a mean DMR
for each tree species and survey method.
Data from townships wliere the surveys chd
not sample at least 3 Douglas-lir or lodgepole
pine for each of the survey procedures were
not included in the analyses. Only living trees
were used in the anaKses for calculating mean
DMR because it was not aK\a\s possible to
accurateh' assign a DMR to dead trees. Inci-
dence xalues were calculated for 9 townships
for lodgepole pine and for 17 townships for
Douglas-fir. The roadside-plot sur\'ey sampled
a total of 206 lodgepole pine and 357 Douglas-
fir in 46 and 75 plots, respectixeK. The tran-
sect-plot survey sampled 171 lodgepole pine
and 342 Douglas-fir in 42 and 87 plots,
respecti\ el\. A one-way analysis of variance
(ANOVA, P > 0.05) was used to determine if
the mean values for incidence and severity were
significantK different between the 2 survey
procediues. Percentages were conxerted using
arcsin transformations before ANOVA analyses
were perfomied (Snedecor and Cochran 1989).
To compare our results with those of other
dwarf mistletoe sui-veys, we determined inci-
dence of both dwarf mistletoes for both suiA/ey
procedures by calculating the percentage of
plots infested. If a plot had at least 1 infected
tree, it was considered infested. This method
of reporting dwarf mistletoe incidence has
been applied in the majority of roadside-plot
surxeys conducted for dwarf mistletoes in the
western United States.
Results AND Disc
L >>|()N
Mean diameters for trees sampled using
each survey method were approximately the
same for lodgepole pine and Douglas-fir
(Table 1). Sampled tree diameters were clearly
skewed toward larger trees (Table 1) because
both sur\e\' methods used variable-radius
plots that sample large trees more often than
small trees (Avery and Burkhart 1983).
Because both survey methods sampled trees
in the same way, the suney results should be
comparable. However, it is probable that the
percentage of infected trees and mean DMR
would have been lower for both lodgepole
pine and Douglas-fir had more small trees
been sampled because small trees are typically
less often and less severelv infected (Parmeter
1978).
Estimates of incidence for Douglas-lir dwaif
mistletoe using the 2 survey methods were
within 3% of each other based on the percent-
age of trees infected (Table 2). Estimates of
Douglas-fir dwarf mistletoe severity were sim-
ilar also. The differences between Douglas-fir
dwarf mistletoe incidence and severity for the
2 sui-vey methods were not statistically signifi-
cant. The differences between estimates of the
incidence and severity of lodgepole pine
dwarf mistletoe for the 2 sunex' methods were
larger than for Douglas-fir dwarf mistletoe
(Table 3). However, the differences were not
significant. Therefore, the 2 survey methods
Table L Distribution of lodgepole pine and Douglas-fir sampled by diameter classes for the roadside-plot and transect-
plot sur\ e\s on the Saw tooth \ational Forest, Idaho.
Lod,£
;epole
pine
Douglas-
fir
Roadside
-plot
Transect-plot
Roadside
-plot
Transect-plot
Diameter
Mean
Mean
Mean
Mean
class
diameter
diameter
diameter
diameter
(cm)
(cm)
N
(cm)
N
(cm)
N
(cm)
N
2-13
9.1
47
S.6
39
10.6
17
9.1
10
14-25
19.6
108
19.8
92
20.6
90
20.1
108
26-38
29.2
39
30.0
34
31.8
98
32.0
99
39-51
41.9
7
42.9
5
43.7
85
44.7
60
52-64
60.7
5
51.1
1
56.6
32
57.4
24
>64
— •'
—
—
—
96.5
35
89.4
41
Total
20.8
206
20,1
171
39.5
.357
40.9
342
••No trees sampled in this size tla
132
Great Basin Naturalist
[Volume 56
Table 2. Incidence and seventy of Doiiglas-fir dwarf mistletoe estimated from roadside-plot and transect-plot sui-veys
on the Sawtooth National Forest, Idaho.
Incidence
Se\erit\
Snrvey
method
Mean
percent
infected''
95% mean
confidence limit
Mean DM Rl'
95% mean
confidence limit
Roadside-plot
Transect-plot
28.4^-
25.8
11.0-15.8
10.0-41.5
0.9^-
0.8
0.2-1.5
0.2-1.4
''Based on the percentage of individual trees infrctecl im a per-lifctart- l)asis
''Dwaii mistletoe rating (Hawksworth 1977)
^'Means in this cohiinn are not significantK- different; one-way A.NO\'A. P > 0.05.
Table 3. Incidence and severit)' of lodgepole pine dwarf mistletoe estimated from roadside-plot and transect-plot sur-
ie\'s on the Sawtooth National Forest, Idaho.
Incidence
Se\'e
■it\'
Sin-vey
method
Mean
percent
infected''
95% mean
confidence limit
Mean DMR'^
95% mean
confidence limit
Roadside-plot
Transect-plot
48.5^-
55.7
29.4-67.5
35.3-76.1
1.2''
1.6
0.6-1.8
1.1-2.1
ctare basis
'^ Dwarf mistletoe rating (Hawksworth 1977)
''Means in this column are not significantly different; one-way ANOVA, P > 0.0.5.
provided equivalent estimates of dwarf mistle-
toe incidence, based on the percentage of
trees infected, and severity for both dwarf
mistletoes.
Dwarf mistletoe incidence based on the
percentage of plots infested is presented in
Table 4. Both survey methods provided esti-
mates that were within 2% of each other for
both dwarf mistletoes. Calculating dwarf mistle-
toe incidence based on the percentage of plots
infested greatly increases the estimates of
dwarf mistletoe incidence when compared to
the incidence based on the percentage of trees
infected because it requires only 1 infected ti^ee
for a plot to be treated as infested.
Lodgepole pine dwarf mistletoe is one of
the most widely distributed dwarf mistletoes
in the western United States (Hawksworth
and Wiens 1995). The incidence of this mistle-
toe, based on the percentage of plots infested,
has varied between approximately 40% and
70% for the majority of national forests sur-
veyed, and averages about 50% (Hawksworth
1958, Graham 1960, 1964, Johnson et al. 1980,
1981, Hoffinan and Hobbs 1985). The incidence
of lodgepole pine dwarf mistletoe, based on
the percentage of plots infested estimated from
our surveys in the Sawtooth National Forest
(approximately 80%), is higher than for most
national forests surveyed thus far. An earlier
dwarf mistletoe suivey of the Sawtootli National
Forest (Hoffman and Hobbs 1985) reported
the incidence of lodgepole pine dwarf mistle-
toe as 71%. However, that survey did not in-
clude the Sawtooth National Recreation Area,
the district in which we detected a veiy high
incidence of lodgepole pine dwarf mistletoe
(83%). Therefore, the Sawtooth National For-
est probably does have a higher incidence of
lodgepole pine dwarf mistletoe than many
other western national forests.
An earlier estimate of the incidence of
Douglas-fir dwarf mistletoe, based on the per-
centage of plots infested, for the Sawtooth
National Forest was 53% (Hoffman 1979).
Although that sune)' sampled onl\ the south-
ern districts of the Sawtooth National Forest
and did not include the districts we surveyed,
our estimate for Douglas-fir dwarf mistletoe,
based on the percentage of plots infested, is
approximately the same (almost 50%).
Our findings provide additional evidence
that estimates of incidence and severity of
dwarf mistletoes irsing roadside-plot surveys
1996]
Roadside Surveys for Dwarf Mistletoes
133
Table 4. Incidence of Donglas-fir and lodf^epole pine
dwarf mistletoes based on the percentage of plots infested
estimated from roadside-plot and transect-plot surveys on
the Sawtooth National Forest, Idaho.
Douglas-fir
dwarf mistletoe
Lodgepole pine
dwarf mistletoe
Surve\ method Plots
Percent
infested
Plots
Percent
infested
Roadside-plot
75
47
46
80
Transect-plot
87
48
42
78
approximate those of similar surve>'s con-
ducted away from roads. Hawksworth (1956)
reported similar results based on a more in-
tensive comparison of roadside-plot and tran-
sect-plot surveys for dwarf mistletoes on the
Mescalero Apache Indian Reservation, New
Mexico. Partridge and Canfield (1980) com-
pared the incidence of several forest pests in
southern Idaho estimated using roadside-plot
surveys and plots randomly located in areas
without roads. They reported no discernible
differences between the incidence of the pests
detected (including dwarf mistletoes) for the 2
sur\'ey procedures. Because this study and oth-
ers indicate that roadside-plot surveys provide
similar estimates of dwarf mistletoe incidence
to surveys conducted away from roads, we
recommend that resource managers continue
to use roadside-plot surveys for estimating
dwarf mistletoe incidence for national forests
or other large forested areas. However, because
these surveys sample only a small fraction of
the survey area, they will provide only rough
estimates of the incidence and severity of
dwarf mistletoes.
Acknowledgments
We appreciate the field assistance provided
by Al Dymerski, Valerie DeBlander, Carl
Koprowski, and Lia Spiegel. Reviews of the
original manuscript by Ralph Williams, Greg
Filip, and Catherine Parks are appreciated
also.
Literature Cited
Andrews, S. R., and J. E Daniels. 1960. A survey of
dwarfmistletoes in Arizona and New Mexico. USDA
Forest Service, Rocky Mountain Forest and Range
Experiment Station, Paper 49. 17 pp.
Avery, T. E., and H. E. Burkhart. 1983. Forest measure-
ments. McGraw-Hill Book Company, New York. 331
pp.
DooLLNC, O. J. 1978. Survey methoils to determme the
distribution and intensity of dwarf mistletoe. Pages
36-44 in Proceedings of the Symposium on Dwarf
Mistletoe Control Through Forest .Management.
USDA Forest Service, General Technical Report
PSVV-31.
Ghaham, D R 1960. Surveys expose dwarfmistletoe prob-
lem in Inland Empire. Western Conservation J(jur-
nal 17; 56-58.
• 1964. Dwarfmistletoe survey in western Mon-
tana. USDA Forest Sei-vice Research Note I NT- 14.
7 pp.
Hawksworth, E G. 1956. Region 3 dwarhnistlet(je survey,
progress report on the 1954-55 held work [mimeo-
graphed]. USDA Forest Service Special Report,
Rocky Mountain Forest and Range Experiment Sta-
tion, Fort Collins, CO. 5 pp.
. 1958. Survey of lodgepole pine dwarftnistletoe on
the Roosevelt, Medicine Bow, and Bighorn National
Forests. USDA Forest Sei-vice, Rocky Moimtain For-
est and Range Experiment Station, Paper 35. 13 pp.
. 19.59. Distribution of dwarfmistletoes in relation
to topography on the Mescalero Apache Reserva-
tion, New Mexico. Journal of Forestn' 57: 919-922.
. 1968. Ponderosa pine dwarf mistletoe in relation
to topography and soils on the Manitou Experimen-
tal Forest, Colorado. USDA Forest Service Research
Note RM-107. 4 pp.
. 1977. The 6-class dwarf mistletoe rating system.
USDA Forest Sendee Research Note RM-48. 7 pp.
Hawksworth, F G., and A. A. Lusher. 1956. Dwarf-
mistletoe sui^vey of the Mescalero Apache Indian
Reservation, New Mexico. Journal of Forestr)' 54:
384-390.
Hawksworth, F G., and D. Wiens. 1972. Biologx- and
classification of dwarf mistletoes {Arceuthohium).
USDA Forest Service Agricultural Handbook 401.
234 pp.
. 1995. Dwarf mistletoes: biology, patholog\', and
systematics. USDA Forest Service Agricultural Hand-
book 709. 410 pp.
Hoffman, J. T. 1979. Dwarf misdetoe loss assessment sur-
ve\- in Region 4, 1978. USDA Forest Senice, Inter-
mountain Region, Forest Pest Management Report
R.4.79.4. 12 pp.
Hoffman, J. T, and L. Hobbs. 1985. Lodgepole pine
dwarf mistletoe survey in the Intermountain Region.
Plant Disease 69: 429^31.
JcjHNSON, D. W., E G. Hawksworth, and D. B. Drum-
mond. 1980. 1979 dwarf mistletoe loss assessment
survey on national forest lands in Colorado. USDA
Forest Service, Forest Pest Management, Methods
Application Group Report 80-6. 18 pp.
Johnson, D. W., F G. Hawksworth, and D. B. Drum-
mond. 1981. Yield loss of lodgepole pine stands to
dwarf mistletoe in Colorado and Wyoming national
forests. Plant Disease 65: 437-438.
Maffei, H. M., and J. S. BE.^rrt'. 1988. Changes in the
incidence of dwarf mistletoe over 30 years in the
Southwest. Pages 80-90 in Proceedings of the 36th
Western International Forest Disease Work Confer-
ence, Park City, UT, 19-23 September 1988.
Merrill, L. M., E G. Hawksworth., and D. W. John-
son. 1985. Evaluation of a roadside survey proce-
dure for dwarf mistletoe on ponderosa pine in Col-
orado. Plant Disease 69: 572-573.
134
Great Basin Naturalist
[Volume 56
Parmeter, J. R. 1978. Forest stand cKnaiiiics and ecologi-
cal factors in relation to dwarf mistletoe spread,
impact, and control. Pages 16-30 in Proceedings of
the Symposium on Dwarf Mistletoe Control Through
Forest Management. USDA Forest Sewice, General
Technical Report PSW-31.
Partridge, A. D., and E. R. Canfield. 1980. Frequency
and damage by torest-tree pests in southern Idalio.
Universit)' ol Idaho, Moscow; Forestry, W'ildland,
and Range Experiment Station Note 34. 7 pp.
Snedecor, G. W, and W. G. Cochran. 1989. Statistical
methods. Iowa State University Press, Ames.
Received 13 February 1995
Accepted 23 October 1995
Great Basin Naturalist 56(2), © 1996, pp. 135-141
EFFECTS OF DOUGLAS-FIR FOLIAGE AGE CLASS
ON WESTERN SPRUCE BUDWORM OVIPOSITION CHOICE
AND LARVAL PERFORMANCE
KimherK' A. Dockls', Karen M. Clanc>-, Kathryn J. Lcyva'^,
David Greenherg\ and Peter W. Price'^
Abstract. — The western spruce hudworm {Churistoiiciira occidcnfali.s Freeman) prefers to feed on llusliin^ hiids
and current-year needles ot Douglas-tir {Pseudotsuga menziesii [Mirh.] Franco). Budworin lan'ae will not t\picail\- con-
sume older age classes of needles unless all current-year foliage is depleted. We tested the following null hypotheses:
(1) budworm lanae can feed on foliage with a wide range of ((ualities (i.e., cuirent-year versus 1-, 2-, or 3-vear-old needles)
without measurable effects on fitness; and (2) budworm adults do not show any oviposition preference linked to the age
of the foliage they fed on as larvae. We used both laborator\ and field experiments. There was strong evidence to sup-
port rejection of hypothesis 1. Budworm larvae had greater survival from the 4th instar to pupal stage when they fed on
current-year foliage (43%-52% sui-vival) versus older age classes of foliage (0-25% survival). Pupae from current-year
foliage were also heavier than pupae from > 1-year-old foliage. There was weak evidence to support rejecting hypothesis
2; budworm adults that had fed on current-year or 3-year-old foliage as larvae preferred to oxijiosit on current-year
foliage. Similar conclusions were drawn from the laboratoiy and field experiments.
Keij words: Choristoneura occidentalis, western spruce hiidwonii. oviposition preference, needle age, foliar qualitij.
eruptive species.
The western spruce budworm {Choris-
toneura occidentalis Freeman) is a major defo-
liator of Douglas -fir {Pseudotsuga menziesii
[Mirb.] Franco) trees in western North Amer-
ica (Fellin and Dewey 1982, Wulf and Gates
1987, Clancy et al. 1988). Budworm lan^ae pre-
fer to feed on the flushing buds and current-
year needles of their host trees. However, if all
current-year foliage is depleted, larvae will
feed on older needles (Fellin and Dewey 1982,
Talerico 1983, Blake and Wagner 1986). Previ-
ous experiments by Talerico (1983) and Blake
and Wagner (1986) show older foliage is sub-
optimal, resulting in reduced fecundit\; higher
mortality rates, and impaired dexelopment.
When budworm larvae are forced to feed on
only mature foliage, they have reduced growth,
lower pupal weights, and decreased sunaval,
or they may not sui-vive at all (Blake and Wag-
ner 1986).
Variations in host foliage quality may influ-
ence the feeding and oviposition behavior of
the western spruce budworm (Clancy et al.
1988). Differences in levels of foliar nutrients,
water content, needle toughness, etc., between
cunent-year and older (> 1-year-old) age classes
of needles impact the budworm's fecimdit);
growth rate, and survivorship (Mattson and
Scriber 1987, Clancy et al. 1988, Clancv- 1991b,
1991c), and may influence female oviposition
choices.
However the budwonn s oligophagous feed-
ing behavior and eruptive population dynamics
suggest it is unlikely that there is a tight link-
age between female oviposition preference
and larval performance (Price et al. 1990).
Female moths do not determine where their
offspring will feed. Budworm adults lay eggs
on mature foliage in late summer (Furniss and
Carolin 1977, Brookes et al. 1987); upon hatch-
ing, the 1st instars (which do not feed) dis-
perse to sheltered locations (e.g., beneath bark
scales), where they spin a hibernaculum and
ovei-winter When larvae emerge from their
hibernacula the following spring, the>- disperse
again (typically on silken threads) to find appro-
priate food sources. The budworm's life his-
tory suggests that neither adults nor larvae
actively select host foliage based on differ-
ences in nutritional quality among individual
host trees. Instead, larvae passively disperse
from their ovenvintering sites and ma> land
'Dt-partiiient ol Forestn-, Northern Arizona Universih; Box 1.5018, Flagstaff, A2. 86011.
2Rock>- Mountain Forest and Range Experiment Station, USDA Forest Senice Rcsearcli. 2.500 S. I'inc Kn
^Department of Biological Sciences, Northern Arizona University', Box .5640, Flagstaflf, .^Z 8601 1.
li:)rivc. Flagstaff, . 1/86001 .
135
136
Great Basin Naturalist
[Volume 56
on acceptable food sources. Once larvae are
on a host tree, they search for expanding cur-
rent-year l>uds and needles. If suitable foliage
is not available, larvae can disperse horizon-
tally or vertically within and between tree
crowns and stands, but dispersal invariably
results in significant losses; whether dispers-
ing larvae live or die depends largely on
whether they find hospitable sites (Brookes et
al. 1987). Therefore, the ability to utilize a
broad range of foliage qualities would be
advantageous for budworm sui"vival.
This study was designed to compare results
from laboratory and field tests of the null
hypotheses that (1) budworm larvae can feed
on foliage with a wide range of qualities (i.e.,
current-year versus 1-, 2-, or 3-year-old nee-
dles) without measurable effects on fitness;
and (2) budworm adults do not show any
oviposition preference linked to the age of the
foliage they fed on as lai"vae. Furthermore, we
wanted to determine if conclusions drawn
from laboratory versus field experiments were
similar. This is important because many previ-
ous studies conducted with budworms and
other forest defoliators have used clipped
foliage without knowing the effects this may
have on foliar nutrition or host defenses. By
conducting parallel experiments using intact
and excised foliage from the same trees, we
were able to evaluate the importance of
changes in foliar quality that may be associ-
ated with bagging lan'ae on intact branches in
the field versus feeding lai^vae excised foliage
in the laboratoiy
Study Area and Organisms
The study area is located at Little Springs
(elevation 2560 m), 16 km north of Flagstaff,
Arizona, within the Coconino National Forest.
The site is a high-elevation, mixed-conifer for-
est, with Douglas-fir as the primaiy host species
and with a recent history of western spruce
budworm infestation.
The western spruce budworm has a imix'ol-
tine life cycle. Adults are present from Jidy to
August, with mating typically occurring within
24 h of eclosion. Eggs are laid soon after mat-
ing; females lay between 25 and 40 eggs per
egg mass (Brookes et al. 1987). Eggs hatch in
about 10 d; after dispersing to sheltered loca-
tions and spinning a hibernaculum, larvae molt
into 2nd instars and ovei'winter. In spring, the
2nd instars emerge fi-om diapause, feed through
the 6th instar, and pupate in late June or early
July. Adults eclose within 10 d. Our laboratoiy
population of nondiapausing western spruce
budworm differs in tliat there is no ovei'win-
tering stage.
Methods
Field Experiment
To determine larval performance in the
field, we selected and tagged 50 Douglas-fir
trees of various sizes and ages on 1-2 June
1993. All tagged trees had abundant foliage in
the lower crown. Sleeve bags made of fine
mesh screen were placed over 4 branches on
each of the 50 trees, and each bagged branch
was randomly assigned to a foliage age class
(current-year, 1-year-, 2-year-, or 3-year-old
needles). We removed by hand all needles that
were not of the appropriate age class. Any wild
budworms present on the bagged branches
were also removed.
On 4 and 8 June, two 4th or early 5th instar
budworm larvae from oin- laboratory culture
were placed on foliage inside each bagged
branch (a total of 400 lan'ae were used); this
constituted the parental (P^) generation. We
have established that budworm larvae from
our laboratory culture have rates of sur\'ival
and reproduction equivalent to wild bud-
worms when reared on Douglas -fir foliage in
the field (Le\'va et al. 1995). Bags were closed
with string or duct tape at each end. We exam-
ined the bagged branches on 20 June to deter-
mine if sufficient foliage remained for comple-
tion of larval development. Pupae were not
obsei"ved at this time.
Budworm lanae remained in the field imtil
about half of them had pupated, and then the
bagged branches were clipped, placed inside
large plastic bags, and transported to the labo-
ratoiy. Pupae were weighed (to the nearest 0. 1
mg), sorted into trays according to treatment
(foliage age class) and sex, and then refriger-
ated at 10 °C until we obtained 10 males and
10 females from the same treatment. Larvae
that had not pupated were placed in labeled
petri dislies lined with moist filter paper. Dou-
glas-fir foliage of the appropriate age class was
provided for them to feed on until they
pupated; this foliage was collected at random
from tagged Douglas-fir trees at the study site.
Foliage was replaced ever\ 2-3 d to ensure
1996]
Effects of Fouack Ack Class on Budworms
137
freshness. Petri dishes were ehecked each Mon-
day, Wednesday, and Friday to remove and
weigh new pupae (tliest^ were also sorted and
refrigerated).
When 10 pairs of male and female pupae
were a\'ailable from a treatment, they were
placed in a l)rown paper mating hag; oviposi-
tion preference tests for both field and lahora-
tor\' experiments were conducted in the labo-
iator\. 13i"o\\'n paper bags provided appropri-
ate lighting conditions both for mating (which
occurs from 2000 to 2300 h in nature), when
only safety lights were on in the laboratory at
night, and for oviposition (which normallv
occurs the da>' following mating), when all lights
were on during the day (i.e., bags are not
opaque). Bags were checked every other day
until 5 or 6 moths emerged; then liranches of
freshly clipped Douglas-fir foliage were added
for oviposition substrate. Once foliage was
added, moths were allowed to mate and oviposit
for 7-8 d. After oviposition occurred, Douglas-
fir branches were removed and inspected for
egg masses. These Fj egg masses were sorted
according to treatment (foliage age class) to de-
temiine if female moths showed a preference for
ovipositing on a particular age class of foliage.
The Fj egg masses collected were surface-
sterilized with formalin and placed into labeled
cups containing an artificial diet nutritionally
similar to Douglas-fir foliage (Clancy 1991a).
Fj lai-vae were reared on the diet until the 4th
or early 5th instar stage, after which they were
placed in labeled petri dishes lined with moist
filter paper. Douglas-fir foliage of the same
age class that their parents consumed was pro-
vided for them to feed on until they pupated;
this foliage was collected at random fiom tagged
Douglas-fir trees at the study site. These lar-
vae were not placed in the field because it was
too late in the season for conditions suitable
for budwomi dexelopment. Foliage was replaced
eveiy other day to ensure freshness. Fj larvae
were reared on foliage within petri dishes
until they pupated; pupae were handled in the
same manner as in the first generation.
Laboratoiy Experiment
This study was conducted to determine if
laboratory experiments using excised foliage
would yield results similar to those from field
experiments using intact foliage. The experi-
ment was started 24 June 1993. Douglas-fir
foliage used in this experiment was collected
from the same 50 trees we used for the field
experiment. Four hundred 4th instar bud-
worms were placed on excised foliage in petri
dishes lined with moist filter paper, 2 larvae
per dish. Fift\' petri dishes were used per
foliage age class treatment (current-year, 1-
year-, 2-year-, and 3-year-old needles), corre-
sponding to the 50 trees used in the field
experiment. Needles of the appropriate age
class were left attached to the stem to prevent
desiccation of foliage. Foliage was replaced
ever>' 2-3 d to ensure freshness. Petri dishes
were labeled according to the tree number
and foliage age class. If a single larva or both
lai-vae in each petri dish died before pupation,
they were replaced with new lanae from our
lal:)oratoiy culture. Othenvdse, we used the same
procedures for the laboratory experiment as
for the field experiment.
Results
Effects of Foliage Age Class on
Pj Survival and Pupal Weight
Budworm larvae that consumed current-
year needles of Douglas-fir in the field experi-
ment had higher sumval rates from 4th instar
to pupal stage (43% sui-vival) compared to lar-
vae that fed on 1-year-old (2% survival), 2-
year-old (1% survival), or 3-year-old (0% sur-
vival) needles (Fig. lA) {y} =^130.19, df = 3, P
< 0.001, n = 400). We believe that many of
the larvae bagged on the branches with only
> 1-year-old needles to feed on escaped from
the mesh bag enclosures, so it may be more
appropriate to refer to this response as "per-
cent lai-vae accounted for" rather than "percent
lai-val sui-vival.' Lai-vae from older foliage age
class treatments were more likely to escape
because budwonn lan'ae tend to disperse when
suitable food is not available, and our bags
were not so tightly sealed that lanae could not
wriggle out through small openings along the
seams or closures at the ends.
The age class of foliage ingested had a simi-
lar effect on sunival from 4th instar to pupal
stage in the laboratory experiment (Fig. IB)
(X2 = 59.46, df = 3, P < 0.001, n = 727).
Approximately 52% of larvae that consumed
current-year needles survived. Survixal was
25% for larvae consuming 1-year-old needles,
18% for larvae feeding on 2-year-old needles,
and 20% for larvae eating 3-year-old needles.
138
Great Basin Naturalist
[Volume 56
Current-yr 1-yr-old 2-yr-old 3-yr-old
Current-yr 1-yr-old
2-yr-old 3-yr-old
Foliage Age Class
Fig. 1. Percentage of Pj 4th instar western spruce bud-
worms surviving to the pupal stage when reared on cur-
rent-, 1-, 2-, and 3-year-old Douglas-fir needles for the (A)
field experiment and (B) laboratory experiment. %- tests
showed that sui-vival varied among the foliage age classes
for both the field (P < 0.001) and laboratoiy (P < 0.001)
experiments. Numbers above the bars indicate sample
sizes, i.e., number of budworm lai^vae used per treatment.
Foliage age class did not have a significant
effect on pupal masses for the field experi-
ment (F = 1.97, df = 2,41, P = 0.152; Fig. 2).
This inability to detect differences among
foliage age classes can be attributed to the
veiy small sample sizes (n = 0-2) for > 1 -year-
old foliage. As expected, female pupae were
heavier than male pupae (F = 20.39, dl =
1,41, P < 0.001).
There were detectable differences in pupal
masses among different foliage age classes for
the laboratory experiment (F = 36.47, df =
3,182, P < Oi)01; Fig. 3). Larvae consuming
current-year foliage became nuich heavier
pupae than larvae feeding on > 1 -year-old
foliage. Once again, female pupae were bigger
than male pupae (F = 14.70, df = 1,182, P <
0.001).
Effects of Foliage Age Class on
Oviposition Preference of Fy Females
Sample sizes for the field experiment were
not large enough for data analysis (n = 2 egg
masses), but a contingency table analysis of
data from the laboratory experiment indicated
03
Q.
3
Q.
UJ
'\'a et al. 1995). Nonethe-
less, the distribution of Fj egg masses indi-
cated that moths reared as Uwvae on current-
year or 3-year-old foliage laid more of their F^
egg masses on current->'ear needles than on
older age classes of needles.
Effects of Foliage Age Class on
F| Sin\i\al and Pupal Weight
Only 2 Fj egg masses were produced from
the field experiment. This precluded analyz-
ing data on sur\ i\ al or pupal weights for this
experiment. For the laborator) experiment, we
found a significant difference in survival from
4th instar to pupal stage between larvae reared
on cinrent-\'ear (83.3% sundval) versus 3-year-
old (3().87r surNixal) needles {%- = 11.78, df =
1996]
Effects of Foliage Ace Class on Budworms
139
a
3
Q.
UJ
W
CN
+1
Current yr 1 -yrold 2vr-old 3-vr-old
Foliage Age Class
Fig. 3. Mean (± 2 s, or =95% confidence interval) Pj
male (■) and female (D) pupal weight for larvae reared
from the 4th instar to pupation on foliage of different age
classes in the laboratory experiment. ANOVA tests
showed that foliage age class had a significant effect on
pupal weight {P < 0.001), as did sex {P < 0.001).
1, P = 0.0006, n = 44). Sample sizes were 0
for 1-year- and 2-year-old needles. This result
was consistent with results for the Pj genera-
tion in that survival was higher for larvae
reared on current-year foliage than on 3-year-
old needles.
However, F^ pupal masses were equivalent
for pupae from current-year and 3-year-old
foliage (F = 1.14, df = 1,19, P = 0.299). As
before, female pupae were larger than male
pupae (F = 6.01, df = 1,19, P = 0.024).
Discussion
Although the western spruce budworms
life history and population dynamics suggest
that larvae shoidd be able to utilize a broad
range of foliage qualities, our results confirm
previous studies that indicate the budworm is
not well adapted to feeding on 1 -year-old or
older needles (Ttilerico 1983, Blake and Wag-
ner 1986). Thus, we must reject hypothesis 1
and conclude that the budworm cannot feed
on foliage with as wide a range in qualities as
is found in current-year versus > 1-year-old
needles without measurable effects on fitness.
We found that whether budworm larvae were
feeding on bagged foliage in the field or on
excised foliage in the laboratory, larval survival
(Fig. 1) and pupal masses (Figs. 2, 3) declined
for larvae feeding on > 1 -year-old needles
compared to larvae feeding on current-year
foliage. The same patterns in relation to the
effects of foliage age on sm-vival were evident
for both the P^ and Fj generations of the labo-
ratory experiment. It is well established that
the nutritional quality of Douglas-fir needles
declines rapidly as current-vear needles age
(Clancy et al. 1988, 1995). Furthermore,
Clancy et al. (1995) point out that
the general pattern for l-ycar or older needles of
conifers is typically an extension of the seasonal
trends for nutrient concentration changes in cur-
rent-year needles.
Needle toughness and fiber content also
increase as foliage matures, thus making older
needles less suitable food for the budworm.
On the other hand, the fact that budworm lar-
vae could survive at all when reared on older
age classes of needles may indicate that their
nutritional niche is indeed broad, as suggested
b>' Price et al. (1990) and Lex^a et al. (1995).
We fotmd less-convincing evidence to sup-
port rejection of hypothesis 2, but nonetheless
we conclude that budworm adults may show
an oviposition preference that is linked to the
age of the foliage they fed on as lanae (Table
1). Budworm females that fed as larvae on cm-
rent-year foliage laid more egg masses on cur-
rent-year needles than on older needles; females
reared on 3-year-old foliage also laid more egg
masses on current-year needles. This result is
suiprising because the budworm typically ovi-
posits on mature foliage (Brookes et al. 1987,
Price et al. 1990). However, many egg masses
from our experiment were very small and arc
most likely aberrant (Leyva et al. 1995); thus,
we suspect this may not represent normal
oviposition behavior for the budworm. If we
remove these veiy small egg masses from the
data set, most egg masses were laid on 3-year-
old needles, indicating the budworm does
indeed prefer to oviposit on mature foliage.
An alternative explanation ma\- be that female
moths distribute egg masses randomly across
age classes of needles available. Current-year
needles represent a small proportion of total
needles present under natural conditions, and
they may be nearly absent when defoliation is
heavy. Thus, our result could be an artifact of
providing an atypical distribution of needle
age classes for oviposition substrate.
140
Great Basin Naturalist
[Volume 56
Table 1. Distribution of Fj egg masses of the western spruce budworm laid on current-, 1-, 2-, or .3-year-old needles
of Douglas -fir, in relation to the age class of the foliage on which the Pj moths were reared'*.
Age class of foliage
on which Fj egg
masses were laid
Age class of foliage on which
Pj female was reared''
Cmrent-vear
l-\ ear-old
.'3-\ear-old
Current-year
1-year-old
2-year-old
3-year-old
27
6
5
11
14
0
1
0
''Data are the niimber of Fj egg niaises lioni the laboiatoiy experiment. The distiiluitioii ol egg masst.s \\ as examined for ri)w-column dependence nsing a 2 x 2
contingency talile (current-year versus > 1-year-old foliage; older foliage age classes weie combined in order to meet requirements for minimum expected cell
frequencies); Yates-corrected X" = .5.72.5, df = 1, P = 0.017, n = 64.
''Pi budwonu larvae were reared on foliage of different age classes from the 4th instar to pupation.
Conclusions from the field versus labora-
tory experiments were similar in regard to the
effects of foliage age class on larval survival
and pupal masses. This indicates that foliar qual-
ity does not change dramaticall)' when foliage
is excised (at least not over a 2-3 d period),
nor does intact foliage on bagged branches
change markedly in temis of a local or systemic-
induced response to budworm defoliation.
It is noteworthy that pupae from the cuirent-
year foliage treatment in the field experiment
were heavier than equivalent pupae from the
laboratory experiment (Figs. 2, 3). This differ-
ence may well be related to the 15-19 d delay
between the start of the field experiment (ini-
tiated on 4 and 8 June 1993) and the begin-
ning of the laboratory study (initiated on 24
June 1993), with the concomitant decline in
nutritional quality of the expanding current-
year foliage. Alternatively, current-year needles
remaining on bagged branches could have
acted as a local nutrient sink in the absence of
competing older needles, which were removed.
Thus, lai-vae feeding on bagged branches with
current-year foliage may have benefited from
this improved nutritional quality, whereas lar-
vae that were fed clipped foliage in the labora-
tory experiment would not have received this
nutritional boost.
Survival rates were higher overall in the
laboratoiy experiment than in the field experi-
ment (Fig. 1). We attribute this difference to a
combination of factors. For example, lai^vae on
bagged branches in the field were exposed to
some predation since the bags were not per-
fect barriers. Also, some budworm lai-vae un-
doubtedly escaped from the bags, and weather
could have played a role in the lower survival
of lai-vae in the field.
Acknowledgments
We thank Daniel Huebner and Kenneth
Dodds for assistance in collecting data and
conducting experiments. We also thank D.
Leatherman and an anonymous reviewer for
their comments on an earlier draft of this
paper. Our western spruce budworm colony
was started with egg masses obtained from the
Canadian Forest Service s Pest Management
Institute in Sault Ste. Marie, Ontario. This re-
search was funded in part by the Rock>' Moun-
tain Forest and Range Experiment Station,
Cooperative Agreement 28-C3-708 between
KMC and PWE
Literature Cited
Blake, E. A., and M. R. Wagner. 1986. Foliage age as a
factor in food utilization by the western spruce bud-
worm, Chorisfoneuro occidenfalis. Great Basin Natu-
ralist 46: 169-172.
Brookes, M. H., J. J. Colbert, R. G. Mitchell, and R. W.
Stark, editors. 1987. Western spruce budworm.
USDA Forest Service Technical Bulletin 1694.
Clancy, K. M. 1991a. Multiple-generation bioassa> for
investigating western spruce budworm (Lepidoptera;
Tortricidae) nutritional ecology. En\ ironmental Ento-
molog)' 20; 136.3-1374.
. 19911). Douglas-fir nutrients and terpenes as
potential factors intluencing western spruce budwomi
defoliation. Pages 123-133 in Y. N. Bai-anchikov, W. J.
Mattson, F Hain, and T. L. Payne, editors. Forest
insect guilds: patterns of interaction with host trees.
USDA Forest Sen ice, General Technical Report NE-
1,53.
. 1991c. Western spruce biidworni response to dif-
ferent moisture levels in artificial diets. Forest Ecol-
ogy and Management 39: 223-235.
Clancy, K. M., M. R. W.agner, and R B. Reich. 1995.
Ecophysiology and insect herbivory. Pages 125-180
in W K. Smith and T. M. Hinckley, editors, Eco-
physiology of coniferous forests. Academic Press,
New York.
19961
Effects ok Foliage Ace Class on Budworms
141
Clancy, K. NL, M. R. \V.u;m:i;, am) R. W. Tims. 198' plants; water,
nitrogen, fiber, and mineral considerations. Pages
105-146 in F. Slansky, Jr., and J. Rodriguez, editors,
Nutritional ecology of insects, mites, spiders, and
related inxertebrates. John VV'ilev and Sons, New
York.
l^RicE, R VV, N. Cobb, T R Crak., c. U. J-imn andls, j. K.
ITAMI, S. MOPPER, AND R. W. Preszler. 1990. Insect
herbivore population dynamics on trees and shrubs;
new apiiroaches relevant to latent and eruptive
species and life table development. Pages 1-38 i»
E. A. Rernays, editor, Insect-plant interactions. Vol-
ume 2. CRC Press, Roca Raton, FL.
Talerico, R. L. 1983. Summaiy of life histoiy and hosts (jf
the spruce budworms. Pages 1-4 in Proceedings;
Forest Defoliator- Host Interactions; a Comparison
Between Gypsy Moth and Spruce Rudworm, USDA
Forest Service, General Technical ReiMirt \E-85.
Wl'LF, N. W, and R. G. C.\tes. 1987. Site and stand char-
acteristics. Pages 89-115 in M. H. Brooks, R. VV.
Campbell, J. J. Colbert, R. G. Mitchell, and R. W.
Stark, editors. Western spruce budworni. USDA
Forest Service Technical Bulletin 1694.
Received 20 October 1995
Accepted 27 February 1996
Great Basin Naturalist 56(2), © 1996, pp. 142-149
TRYPANOPLASMA ATRARIA SR N. (KINETOPLASTIDA: BODONIDAE)
IN FISHES FROM THE SEVIER RIVER DRAINAGE, UTAH
J. Stephen Cranne\'' and Ricliard A. Heckmann^
Abstract. — A total of 181 fishes l)elon^ing to 10 species were captured near Richfield, Utah, and examined for para-
sites. A new species of hemotlagellate, Tnjpanoplasma afraria sp. n., was ohsei-ved in 3 species: Utah chub {Gila atraria
[Girard]), redside shiner (Richardsoniits balteatus [Richardson]), and speckled dace (Rhinichthijs oscuhis [Girard]).
Seven other species of fishes examined in the study area were negative for T. atraria sp. n. The salmonid leech, Piscicola
salmositica (Meyer), collected in the same area harbored developmental stages of Tnjpanoplasma, suggesting a possible
leech vector for the hemoHagellate. Characteristics of Tnjpanoplasma atraria sp. n. place it near T salmositica, but the
new species is twice as large.
Key words: Tnpanoplasma atraria n. sp., blood parasites, Gila atrari;i,_//s/i parasites.
Tnjpanoplasvm is a biflagellated protozoan
found in the blood of freshwater fishes in the
United States. It has caused significant mor-
tahty in rainbow trout {Oncorhiinchus inykiss
[Walbauni]) and king sahnon (O. t.shauytscha
[Walbaum]) under hatchen' conditions (Becker
and Katz 1966, Wales and Wolf 1995). This
genus has also been described from the blood
of marine fish (Strout 1965). Another name for
the blood biflagellate of salmonids described
above is Cryptobia. There are differing opinions
on the use of the two genera, Cryptobia and
Tryp(inoplas)na, but these differences have been
recently clarified by Lom and Dykova (1992).
The genus Cryptobia was first proposed by
Leidy (1846) for biflagellated protozoans occur-
ring as parasites in the seminal vesicles of snails.
Chalachnikow (1888) was the first to record
the parasite in the blood of fishes, observing it
in freshwater loaches in Russia. Uaveran and
Mesnil (1901) established the genus Trypano-
plasma for a biflagellated blood parasite from
freshwater fishes in France. In 1909, Crawley
stated that Cryptobia from snails and Trypano-
plasina iiom fishes were moiphologically iden-
tical, and that Cryptobia had taxonomic prior-
ity. In defending the creation of the genus Try-
panoplastna, Laveran and Mesnil (1912) argued
that morphological similarities were not suffi-
cient criteria for maintaining a single genus
when strong biological differences, such as
method of infection, were evident. The para-
sites in snails were transferred directh dminir
copulation, while a leech vector was necessar\^
to transfer the flagellate from the blood of one
fish to another Putz (1970) submitted that
comparative biological studies between simi-
lar morphological types are necessary for a
correct ta.xonomic classification. Use of the
genus Cryptobia has, in most cases, emerged
as the popular choice, and Trypanoplasina is
generally recognized as a synonym. Recently,
Lom and Dykova (1992) used Trypanoplasma
for biflagellated blood-inhibiting parasites of
fishes in which a leech vector is involved. Thus,
we adopted the classification scheme used by
Lom and Dykova (1992).
Four species of Trypanoplasma from the
blood of fresh\\'ater fishes ha\'e been reported
in North America. Mavor (1915) found T. borreli
in a moribund white sucker {Catostoinus com-
inersoni [Lacepe]) from Lake Hiu-on. The iden-
tification of T. borreli was based on similarities
with the species initialK described by Laxeran
and Mesnil (1901). Katz (1951) recorded C.
( = Trypanoplas)na) .sabuositica from silver
salmon (O. kisiitcli [Walbaum]) and C. ( = Try-
panoplas)na) lynclii from cottids in the state of
Washington. Subsequent transmission studies
showed C. hjnchi to be a synonym of C. salmo-
sitica (Becker and Katz i965a). Laird (1961)
described C. ( = Trypan()plasma) <;urni'yonim
from northern pike {Esox luciiis [Linnaeus])
and from 2 salmonids: lake whitefish {Core-
' 61: 1242-1250.
. 1984. Detection of infection and susceptibilib,' of
different Pacific salmon stocks {Oncorhynchiis spp.)
to the haemoflagellate Cryptohia sahnositica. Journal
of Parasitolog)- 70: 273-278.
BuRRESON, E. M. 1982. The life cycle of Trypanoplasma
hiillocki (Zoomastigophorea: Kinetoplastida). Journal
of Protozoology 29: 72-77.
Chalachnikow, A. P 1888. Recherches sur les parasites
du sang. Arkliives Veterinan' Science, St. Petersburg
1: 65.
Cope, O. B. 1958. Incidence of external parasites on cut-
throat trout in Yellowstone Lake. Proceedings of the
Utah Academy of Science, Arts, and Letters 35:
95-100.
Crawley, H. 1909. Studies on blood and blood parasites.
II. The priority of Cryptohia Leidy, 1846, over Try-
panoplasma Laveran and Mesnil, 1901. U.S. Depart-
ment of Agriculture, Bulletin of the Bureau of Ani-
mal Industry 119: 16-20.
Heckmann, R. A. 1971. Parasites of cutthroat trout from
'i'ellowstone Lake, Wyoming. Progressive Fish Cid-
turist 33: 103-106.
Hoffman, G. L. 1967. Parasites of North American fresh-
water fishes. University of California Press, Berkeley
and Los Angeles. 486 pp.
Humason, G. L. 1967. Animal tissue technicjues. Freeman
and Compan\', San Francisco. 596 pp.
K-VTZ, M. 1951. Two new hemoflagellates (genus Cryptohia)
from some western Washington teleosts. Journal of
Parasitolog>' 37: 245-250.
Khan, R. \. 1991. Further obsenations on Cryptohia dahli
(Mastigophorea: Kinetoplastida) parasitizing marine
fish. Journal of Protozoologx 38: 326-329.
Khan, R. A., and E. R. Noble. 1972. Ta.\onomy, preva-
lence, and specificity of Cryptohia dahli (Mobius)
(Mastigophora: Bodonidae) in liunpfish, Cyclopterus
himptis. Journal of the Fisheries Research Board of
Canada 29: 1291-1294.
L\1RD, M. 1961, Parasites from northern Canada. II.
Haematozoa of fishes. Canadian Journal of Zoolog\-
39:541-548.
Laveran, A., and F Mesnil. 1901. Sur les flagelles a mem-
l)rane oudulaute des poissons (genres Trypanosoma
1996]
Tnri'ANOPLASMA ATllMilA BLOOD PaRASITE OF FiSH
149
Fniln- ct Tn/i>(i)ioplasiiui n. gen.). Tran.saction.s of the
French Acaclcnu' of Science 133: 670-675.
. 1912. Tnpano.sonie.s et Tnpanosomiases. 2n(l edi-
tion. Xhusson et Cie, Editenrs, Pari.s. 999 pp.
Leidy, E 1846. Descriptions of a new genns and species of
Entozoa. Proceedings of the National Academy of
Science, l^hiladelpliia 3: 100-101.
LOM, J. 1979. BiologN of the trypanosonies and trxpano-
plasnis of fish. Pages 269-337 in W. II. H. Eninsden
and D. A. Evans, editors. Biology of the Kinetoplas-
tida. Volmne 2. .'\cademic Press, New York.
LoM, J., ,\ND I. DYKt)\A. 1992. Protozoan parasites of fishes.
Dexelopnients in acjiiacnhnre and fisheries science,
N'oknne 26. Elsevier Pnhlishers, Anistersckun. 31.5 pp.
M.A\OK, J. W. 1915. On the occnrrence of a tn'panoplasm,
probal)l\ Tnjpanophisina boireli Laveran et Mesnil,
in the blood of the common sncker, Catostornus com-
iiicrsonii. Jom^nal of Parasitology' 2: 1—6.
McD.'VNiEL, D. W. 1970. Personal communication. U.S.
Department of Interior Fisheries. Spring\'ille, Utah.
Noble, E. R. 1968. The flagellate Cnjptobia in two
species of deep sea fishes from the eastern Pacific.
Journal of Parasitolog>' 54: 720-724.
PUTZ, R. E. 1970. Biological studies on the hemoflagel-
lates (Kinetoplastida: Cnptobiidae) Cnjptohki catarac-
tae sp. n. and Cnjptobia saJinositica Katz, 1951. Un-
published doctoral dissertation, Fordham University,
New York. 98 pp.
. 1972a. Cnjptobia catamctae sp.n. (Kinetoplastida:
Cryptobiidae), a hemoflagellate of some cyprinid
fishes of West Virginia. Proceedings ul tlit
Helminthological Society of Washington 39: 18-22.
1972b. Biological studies on the heni()flagellate>
Cnjptobia cafaractae and Cnjptobia salmositica.
Technical Papers, Bureau of Sport Fisheries and
Wildlife 63: 1-25.
Stkout, R. G. 1965. A new hemoflagellate (genus Cnjpto-
bia) from marine fishes of northern .New England.
Journal of Parasitology 51: 654-659.
SWEZV, (). 1919. The occurrence q{ Tnj])anoplasma as an
ectoparasite. Transactions of the American .Micro-
scopical Society 38: 20-2-1.
W.ALES, J. H., AND H. WoLE 1955. Three protozoan dis-
eases of trout in California. California Fish and
Game 41: 183-187.
Wenkich, D. II. 1931. A tnpanoplasm on the gills of caip
from the Schnvlkill Hiver |ourn:il oii^nasilologv 18;
133.
Woo, P T. K. 1987. Cryptobia and ciyptobiosis in fishes.
Advances in Parasitology 26; 199-237.
Woo, P T. K., A\n Wkiinekt, S. D. 1983. Direct transmis-
sion of a hemoflagellate, Cnjptobia sabnositica
(Kinetoplastida: Bodonina), between rainbow trout
under laboraton conditions. Journal of Protozoology
30: 334-337.
Received 1 September 1995
Accepted 19 Januanj 1996
Great Basin Naturalist 56(2). © 1996, pp. 150-156
GEOGRAPHICAL REVIEW OF THE HISTORICAL AND
CURRENT STATUS OF OSPREYS {PANDION HALIAETUS) IN UTAH
Clark S. Monson^
Abstract. — Small numbers of Ospreys {Pandion haliaetits) are known to have nested historically in Utali. A precise
baseline figure is unavailable, but the 19th-century Osprey population in Utah probably consisted of at least 15 breeding
pairs scattered in 4 geographic regions. Human persecution is believed to have caused the abandonment of nesting ter-
ritories along the Wasatch Front and in the western Uinta Mountains b\' 1900 and 1960, respectively. Osprey popula-
tions in the southern plateaus and Green River areas, however, began increasing in the late 1970s. Sexeral recent nesting
attempts and numerous summer sightings at nontraditional and abandoned historical sites in Utah suggest the Osprey is
also expanding its range in Utah. High productivib.' for local pairs and long-range dispersal from more northerly Osprey
populations are discussed as sources for the current surge in Utah's Osprey population, which now consists of approxi-
mately 35 breeding pairs.
Key words: Osprey. Pandion haliaetus, raptor Flaming Gorge Reservoir, dispersal.
The Osprey {Pandion haliaetus) is one of the
most wideh' distributed species of raptors dur-
ing the breeding period. The extent of its cos-
mopohtan range is exceeded by only 2 other
raptors: the Peregrine Falcon {Falco peregri-
mis) (Cade 1982, del Hoyo et al. 1994) and
Barn Owl {Tyto alba) (Marti 1985, Eckert and
Karalus 1987). Despite the Ospreys broad
geographic distribution, local populations occur
in fragmented and low densities in much of
the species' range (Bent 1937, Palmer 1988, del
Hoyo et al. 1994). This scenario holds true for
most of the intermountain region of the west-
ern United States (Kenny 1986, Johnsgard
1990). In Utah, Osprey distribution has been
particularly limited. Recently, however, several
personal summer obsen^ations of Ospreys over
140 km from known breeding pairs prompted
an investigation into the possible occurrence of
Ospreys at other nontraditional Utah localities.
A sui^vey of indi\'iduals from the U.S. Forest
Sei^vice, Utah Division of Wildlife Resources,
Utah State Parks, and other persons familiar
with Osprey ecology was conducted dining
1994-95. The survey revealed man\' Osprey
sightings and several nesting attempts between
1 June and 15 August at numerous lakes, res-
ei"voirs, and rivers from nearly eveiy region of
the state since 1990. These sightings represent
the first widespread effort by Ospre>'s to expand
their range in Utah. This paper reviews histor-
ical Osprey breeding territories in Utah, sub-
secj[uent population declines, and current
Ospre>' population and range expansion in
Utah. '
Geographic History of the
Osprey in Utah
Nesting Ospreys have been reported from
4 geographical areas of Utah (Fig. 1): the
Wasatch Front, Uinta Mountains, southern pla-
teaus, and Green River (Table 1). Accounts of
early ornithologists, naturalists, and egg col-
lectors indicate the Osprey was a regular sum-
mer resident and breeder in Utali. Allen (1872)
found them along the Great Salt Lake marshes
west of Ogden, and Henshaw (1874) saw them
at Utah Lake near Proxo. Neither discussed
nest obsenations in these areas, but R. G. Bee
(unpublished ornithological notes) mentioned
that Ospreys formerly nested along the shores
and tributaries of Utah Lake (Fig. 1; Table 1,
region A).
Other records were for the Uinta Nh)imtains
(Fig. 1; Table 1, region B). J. D. Da>'nes (unpub-
lished ornithological notes) described the
repeated use of an Osprex nest from 1915 to
1938 on the Weber River, 20 km east of Oak-
ley, Sunniiit Count)'. Also, Ha\\\ard (1931)
^Department ofGeograpli), Brinhain Voiini; Uiiivursity, Frovo, UT 84(302-.5.526.
150
1996]
Osim{i:y Status in Utah
151
Fig. 1. Known historical distribution of nesting Ospre>s in Utah: A, Wasatch Front: B. western Uinta Mountains:
C, southern plateaus; D, Green River.
surveyed (8 July-21 August 1930) birds in the
western Uintas where Summit, Duchesne, and
Wasatch counties converge. While not giving
actual locations, he said a few Ospreys nested
in the Mirror and Tryol [sic] lakes region. Bee
and Hutchings (1942) specifically cite Mirror
and Trial lakes as having Osprey nests. Twomey
(1942: 382) visited an occupied nest bet\veen
16 and 20 July 1932 at the north end of Mirror
Lake, "Wisatch County." This nest was actually
152
Great Basin Naturalist
[Volume 56
Historical pairs
Cu
rent pairs
Unknown
1
5-8
0
2-4
8-11
6-8
20-25
Table 1. Nesting populations of Ospreys in Utali.
Region
A: Wasatch Front
B; Uinta Mountins
C; Southern plateaus
D: Green River
in Duchesne County and perhaps the same
nest Hayward et al. (1976) referred to when
they hsted Wasatch County as a former nest-
ing area. R. G. Bee (unpubhshed ornithological
notes) also cited single pairs at Fish, Scout, and
Lily lakes in the western Uintas. On 23 May
1945, Bee recorded that a game warden in
Duchesne informed him of 2 pairs at Moon
Lake and another pair at an unidentified Uinta
lake.
Other early observers of Osprey nests in
Utah include Wolfe and Cottam (Hayward et
al. 1976), who, along with Bee and Hutchings
(1942), saw Ospreys nest at Fish Lake (not the
Uinta Mountains lake with the same name),
Sevier County, beginning in 1928 (Fig. 1; Table
1, region C). On 18 July 1936, R. G. Bee (un-
published ornithological notes) visited the Fish
Lake nest. A local rancher told him Ospreys
had used that particular nesting site for at
least 20 years.
Behle et al. (1958) noted a pair of Ospreys
in southwestern Utah at Navajo Lake, Kane
County, on 17 and 18 June 1950 (Fig. 1; Table
1, region C). This particular territory (and an
additional site at nearby Panguitch Lake,
Garfield Count)') has been used regularly since
Behle's discovery (Eyre and Paul 1973, Salt
Lake Tribune, 13 August 1978, Walters 1981,
Anonymous 1989).
Ospreys also nested along the Green River,
northeastern Utah (Fig. 1; Table 1, region D).
On 23 and 24 July 1959, C. M. White and C.
Bosley (White and Behle 1960) located 2 nests
along this river in Horseshoe Canyon, Daggett
County. Both nests contained 2 young. An
additional nest was discovered on the Green
River in Uintah County by M. Horton in June
1974 (Behle 1981). White (1969) suggested the
total population of Ospreys nesting along the
Green River probably consisted of 6-8 pairs.
Status
Historical Events
Although numerous records of Ospreys nest-
ing in Utah exist, these birds have apparently
undergone 2 separate declines. The 1st decline
involved Ospreys nesting along the Wasatch
Front (Fig. 1). During winter 1848-49, depreda-
tions upon livestock, poultry, and grain led to a
much-publicized contest to kill the 'wasters and
destroyers ' (Arrington 1958: 51). Hundreds of
mammalian predators and thousands of raptors
were killed (Arrington 1958). Ospreys would
have been on their southeiTi wintering grounds
during this assault on local predators, but the
incident suggests that early pioneers in Utah
treated all carnivores and birds of prey with
contempt. Other similar hunts followed, and 40
years later the Utah Legislatiu'e implemented a
law awarding bounties for the killing of preda-
tors (Rawley 1985). Rewards were available for
several species of fish-eating birds including
Ospreys. The destiiiction that this bounty in-
flicted upon fish-eating birds in the name of
"consenation" was significant and is vi\'idly
described by Pritchett et al. (1981).
The attitudes of early residents toward pred-
ators, coupled with laws encouraging their
destruction, may have led to the Ospreys
extirpation from the Wasatch Front (Fig. 1;
Table 1, region A) around the turn of the cen-
tury. In 1935, R. G. Bee (unpublished ornitho-
logical notes) speculated that himian persecu-
tion caused the abandonment of Osprey nests
near Utah Lake. Bee did not record when these
Ospreys disappeared, but his manner of re-
flection on their absence suggests the loss
occurred well before his 1935 notation.
The 2nd period of Osprey decline occurred
in the western Uinta Mountains (Fig. 1). The
Uinta Mountain nests that Da\nes, Bee, Hay-
ward, and Twomey reported \\'ere observed
before, but apparently not after, the 1950s and
1960s when Osprey colonies along the eastern
seaboard were decimated b\' organochlorine
compounds (Palmer 1988, Poole 1989). .\ldiough
the impact of synthetic agricultural biocides
upon Ospreys in Utah is unknown, Ospreys in
other areas of the \\ estern United States were
generally less affected In enxironmental con-
taminants than eastern populations (Poole
1989, White 1994).
Another possible reason for the decline of
Ospreys in the Uinta Moimtains is indiscrimi-
1996]
OsPREY Status in Utah
153
nate shooting resulting from a hostile attitude
by local residents (Bee unpublished ornitho-
logical notes, Twomey 1942). Man> Ospreys
were formerly shot at northern Utah fish
hatcheries during spring migrations (White
1969, Ha\'\vard et al. 1976), and some of these
casualties could ha\e been local breeders.
HayAvard et al. (1976: 66) recorded the
Osprey was "formerly a sparse but regular
summer resident in Utah; now greatK' reduced
in numbers and considered to be rare and
endangered." They preface their discussion of
birds in Utah by stating they have included all
records concerning rare species in the state.
However, they cited Ospreys only in the west-
ern Uintas and records for Fish Lake in Sevier
County. They did not include information on
the 1 or 2 pairs nesting in the Navajo Lake-
Panguitch Lake area, southern Utah, or the
pairs at Flaming Gorge Resen^oir, northeast-
ern Utah. In the same year (1976) that Hay-
ward et al. (1976: 66) described the Osprey as
"gi-eatly reduced," more Osprevs (6 pairs) nested
at Flaming Gorge Reservoir (Wagner 1977,
Salt Lake Tribune, 13 August 1978) than had
been recorded in any particular year in the
western Uinta Mountains.
Current Events
Flaming Gorge Dam on the Green River
was completed in 1964 and created a narrow,
150-km-long reservoir on the Utah-Wyoming
border The Osprey population here remained
relatively stable until the late 1970s and 1980s
when an increase was noted (Behle 1981).
Crawley and White (1989) found 21 pairs and
1 trio of Ospreys at Flaming Gorge in 1989. Of
these, 15 pairs succeeded in fledging 37 young.
Osprey numbers at Fish Lake in Sevier
County increased from 2 pairs in 1989 (Anony-
mous 1989) to 6 in 1993 (B. Lowiy, U.S. Forest
Service, personal communication). Addition-
ally, 1 or 2 pairs now nest 3 km away at John-
son Valley Reservoir (E Wagner personal com-
munication). Other current Osprey nest sites
at traditional waters in Utah include 2 pairs
in the Panguitch Lake-Navajo Lake area of
southern Utah (Anonymous 1989).
In 1990 a pair of Ospreys nested at Tropic
Reservoir, Garfield County (Sorensen 1990).
This site is 20 km east of region C (Fig. 1, Table
1) and should be regarded as a geographical
extension of that area. In 1994 a pair of Ospreys
constructed a nest near the Midway fish
hatcheiy, Wisatch County (Fig. 2A). In 1995 a
2nd pair built a nest 2 km away at Deer Creek
Reservoir on a 5-m-high artificial platform
erected for Ospreys (Fig. 2B). Deer Creek
Reservoir and the adjacent Midway fish hatch-
en' have been fre(juented by Ospreys during
spring migrations for many years (Behle and
Perry 1975). Additional Osprey nesting attempts
in 1995 include 1 pair at Jordanelle Resen'oir,
Wasatch C^ounty (Fig. 2C), and another pair
near Highland, Utah County (Fig. 2D). Incu-
bation behavior at the latter site was observed
for approximately 2 wk before strong winds
destroyed the nest. This site was possibK- the
first Osprey nest along the Wasatch P>()nt in
80-100 yr
The origin of Ospreys colonizing new waters
in Utah is currently unknown, but their reluc-
tance to disperse more than 125 km from their
natal sites is well documented (Henny 1986,
Poole 1989). Reproduction for nests at Flam-
ing Gorge Reservoir is generalK' high (Craw-
ley and White 1989), and considering the
Ospreys pronounced philopatr>', one might
expect that Ospreys at new locations in Utah
derive from this local population. While high
productivity has augmented the Osprey popu-
lation on Flaming Gorge, the frequency with
which Ospreys are being witnessed in Utah is
too great to be the sole result of dispersal from
tliat resei"voir Moreover, if Flaming Gorge were
the primaiy source of Ospreys pioneering new
waters in Utah, one would expect lakes and
rivers near that resei"voir to be the initial areas
of range expansion. This has not been the case.
A more plausible source of Ospreys attempt-
ing to colonize nontraditional (and abandoned
historical) waters in Utah is fi^om spring migi-ants
stopping short of their natal territories farther
north. Osprey populations in Idaho and Wyo-
ming number in the hundreds of pairs (Henny
1986, Poole 1989), and Osprey counts made at
several migration points in the West ha\e bur-
geoned since 1983 (Hoffman et al. 1992). Fur-
thermore, migrating subadult Ospreys are knowTi
to linger sometimes and even remain at pro-
ductive foraging sites south of their traditional
breeding grounds (Swenson 1981, Poole 1989).
These lingering individuals may represent
young adults without an established histon- of
breeding elsewhere.
If more nordierly populations constitute the
primary source of Ospreys currently pioneer-
ing nontraditional waters in Utah, this long-
154
Great Basin Naturalist
[Volume 56
Fig. 2. A. Ospiev nest, Midway fish hatchery; B, ()spre>' nesting plationn and nest. Deer Creek Resenoir; C, Osprey
>st, Jordanelle Reservoir; D, Osprey nest and inenbating adnlt near Highhmd, Utali.
1996]
OsPHEY Status in Utah
155
distance dispersal is a recent phenomenon and
possibK indicates a satnrated hreedint!; popn-
lation in the northern Intennonntain West. A
cnrrent, qnantitati\e evahiation ot Osprc)' pop-
nlations in Idaho and Wyominu and extensive
handine; efforts in these states conld help
determine if this speculation is correct. Until
snch a project is nndertaken, the ori,'
le\'els, the difference in feeding rates between species was proportionally higher (10%). At high turbidity levels (> 20
NTU) trout predation rates were relatively insensitive to prey size. However, shiner continued to consume more, larger
pre>' at the highest turbidibi' levels. These results indicate that Lahontan redside shiner may be superior to Lahontan
cutthroat trout as zooplankton predators at high turbidity levels, and may explain the recent success of shiner in Summit
Lake.
Key words: Daphnia, Laliontan cutthroat front, Oncorlniichus clarki henshawi, Laliontan redside shiner Hichardso-
nius egregius, plankfivorij, predation, size selectivity, turbidity.
The Lahontan cutthroat tiout {Oncorhynchus
clarki henshawi) is an inland subspecies endemic
to die physiographic Laliontan basin in noilhern
Nevada, eastern California, and southern Ore-
gon. These ti^out were once widespread through-
out the basins of Pleistocene Lake Lahontan
(USFWS 1995). Currently, they occupy < 1%
of their former lacustrine range and 11% of
their former stream habitat within the native
range (USFWS 1995). Listed as endangered in
1970, the fish was subsequently reclassified as
threatened in 1975. This facilitated manage-
ment and peniiitted regulated angling (USFWS
1995).
Summit Lake is located in the Summit Lake
Paiute Indian Reservation in northwestern
Humboldt County Nevada (41 °N latitude
119°W longitude), at an elevation of 1828 m.
Formed by a landslide about 20,000 years ago,
Summit Lake is relatively shallow (maximum
depth 12 m) and has historically been subject
to high turbidity levels during summer months
from suspended algae and silt (LaRivers 1962).
It contains the most secure remaining lacus-
trine population of Lahontan cutthroat trout,
and no other salmonids occur in the basin
(Cowan and Blake 1989, Valeska 1989). Other
lacustrine populations are either maintained
by artificial stocking or are subject to higher
levels of harvest and disturbance. Conserx'ation
of this population is compelling, and it has
been identified as important for recover)' of
the subspecies (USFWS 1995).
Cutthroat trout spawning runs at Summit
Lake have generally declined since the late
1970s (Cowan and Blake 1989). Collection of
roe during the 1960s and 1970s and excessive
loss of spawning habitat in Mahogany Creek
from livestock overgrazing (Cowan and Blake
1989, Vinyard and Winzeler 1993) ha\e been
blamed. However, coinciding with the decline
in trout, Lahontan redside shiner {Richardso-
nius egregius) also increased in abundance in
the lake, suggesting a competition effect.
Redside shiner are native to the Great
Basin, but they do not occur naturally in Sum-
mit Lake. Origins of the present shiner popu-
lation in the lake are unknown, but the>' have
been used frequently as live bait. Lahontan
redside shiner feed on drift in streams and are
zooplanktivorous in lakes (Vinyard and Winzeler
1993). Laboratory observations suggest they
1 Dcpartim-nt of Biology, Universih' of Nevada. Reno, Nevada 895.57-001.5.
157
158
Great Basin Naturalist
[Volume 56
may also prey on larval trout (Vinyard and
Winzeler 1993). Analysis of stomach contents
suggests that Lahontan cutthroat trout and
Lahontan redside shiner probably consume
similar foods both in Summit Lake and in
Mahogany Creek, the primaiy spawning tribu-
taiy for trout from Summit Lake (Vinyard and
Winzeler 1993). Both species consume drift in
the stream, and mostly amphipods in Summit
Lake (Cowan and Blake 1989). In contrast, simi-
larly large Lahontan cutthroat trout in Pyramid
Lake are piscivorous (USFWS 1995). Because
most fish species depend on vision to locate
prey (Hobson 1979, Guthrie 1986), it is possi-
ble that high turbidit)' in Summit Lake limits
the visibility of prey and impedes the ability of
trout to catch redside shiner and other large
prey.
Our experiments compared the relationships
of feeding rate, turbidit)', and prey size for
Lahontan cutthroat trout and Lahontan red-
side shiner, with the primary focus being to
examine the relative performance of both
species under various turbidity le\'els.
Methods and Materials
Lahontan redside shiner were captured fiom
Mahogany Creek, Humboldt County, Nevada,
and transported to the University of Nevada.
Lahontan cutthroat trout from the current
Pyramid Lake stock were acquired from the
Lahontan National Fish Hatcheiy, Gardner-
ville, Nevada. Although the historical origins
of the existing Pyramid Lake stock are mixed.
Summit Lake fish were heavily planted into
Pyramid Lake for a number of years, and they
likely constitute the dominant component of
the population (USFWS 1995). Fish were
housed in 19-L tanks and acclimated to local
water conditions for at least 3 wk prior to
experiments.
Experiments were conducted in a secluded
section of a greenhouse at the University of
Nevada. The experimental protocol was simi-
lar to that emploved b\' Vinvard and Winzeler
(1993) and Li et'al. (1985). Visual isolaticm of
experimental tanks was ensured by opaque
black polyethylene sheeting (10 mil, 2.5 m
high), which enclosed all sides of the experi-
mental area and controlled external light.
Temperatures ranged between 12 °C and 17 °('
during the experiments, and diel variation
never exceeded 4°C, a range easih' tolerated
b\' both species. Lighting was provided by a
bank of three 56-watt fluorescent tubes con-
trolled by an automatic timer (10L:14D). Light
intensity' at the water surface averaged 93 jnE
m~- S~^. An airstone in the center of each of
four 38- L aquaria provided aeration and kept
turbidity in suspension. Turbidity (nephelo-
metric turbidity units, NTU) was measured
with an HF Instruments Model DRT 15 tur-
bidimeter Six turbidity levels (3.5, 6, 10, 20,
22, and 25 NTU) were produced using sus-
pensions of bentonite. Bentonite concentra-
tions (mg/L) were significantly correlated vdth
measured turbidity (NTU = 2.583 + 0.162 B,
r- = 0.99). This material is nontoxic and
remains in suspension for long periods.
Feeding rates were determined for fish ex-
posed to single-sized groups ofDaphnia magna
at each turbidit\' level. Laborator)'-reared D.
magna were sorted into 3 size groups using a
dissecting microscope: 1.7 mm, 2.2 mm, and
3.0 mm (top of head to base of tail spine, ± 0.3
mm). Before each feeding trial, a single fish
was placed into each experimental tank and
allowed to acclimate for 24 h. A group of 200
Daphnia were introduced into the tank and
the fish allowed to feed for 2 h. Fish were then
removed and the water and remaining prey
siphoned through a 363-micron mesh net. Prey
retained on the net were counted to deter-
mine consumption rates. This procedure was
repeated for each of the 3 prey size classes
and 6 turbidity levels with 4 fish from each
species, yielding a total of 144 feeding trials.
Fish used in the feeding trials ranged from 70
mm to 93 mm SL. Analysis of variance and lin-
ear regression were used to assess the effects
of fish species, prey size, and turbidity level
on predation rates.
Results
An analysis of overall predation rates for both
fish species consuming all prey sizes (Figs, la,
lb) indicates that feeding rates varied inversely
with turbidit)' (multiple regression, F = 1894,
P < 0.001) and between fish species (F =
28.4, P < 0.001), and that larger prey gener-
ally were consumed at greater rates (F = 38.3,
P < 0.001). Significant results were observed
for both the species* NTU and species '^daph-
nia size interaction terms, indicating that the 2
(ish species differ in their responses to these 2
\ariables. Lahontan redside shiner consumed
1996]
TiHiJii:)! IT Effects on Fish Feeding
159
significantly more prey than l^ahontan cut-
throat trout. At the lowest turhidit) level (3.5
NTU), approximately 909^ of all prey were
consumed l)> both fish species. Ilcmever, even
small increases in turbidity reduced predation
rates. This decrease in predation with tmbid-
ity was strongK linear and there was no indi-
cation of a minimum xalue ha\'ini2; been reached
b> 25 NTU. At that turbidity level, predation
rates declined bv approximately 80% for trout
(Fig. la) and by 60% to 80% for shiner (Fig.
lb), depending on prey size. Predation rates
for trout were significantly affected by prey
size and turbidit\' (multiple regression F =
2.67, P = 0.009 for prey size; F = 35.1, P <
0.001 for turbidit>). Similar results were
obsened for shiner (multiple regression F =
6.54, P < 0.001 for prev size; F = 27.15, P <
0.001 for turbidit> ).
At higher turbiditx levels, differences in
performance of the 2 fish species became most
apparent. At turbidity levels of 20 NTU or more,
prey of all sizes were consumed at virtually
equal rates by Lahontan cutthroat trout (Fig.
la). In contrast, Lahontan redside shiner showed
increasing predation on 3-mm prey relative to
the smaller sizes at high turbidity levels (Fig.
lb), and shiner showed the greatest differ-
ences in predation rates between prey of dif-
ferent size at the highest turbidity levels.
Lahontan cutthroat trout exhibited the oppo-
site trend, with greater differences in preda-
tion rates between prey of different sizes at
low turbidity levels.
Discussion
Foraging behaxior and efficiency are affected
by local visibility. Many workers have demon-
strated reduced effectiveness by visual preda-
tors at elevated turbiditv (Vinyard and O Brien
1976, Li et al. 1985, Barrett et al. 1992, Gregon-
and Northcote 1993). Sigler et al. (1984) found
that chronic high turbidity impedes growth
and increases mortality of steelhead (O. mykiss)
and coho salmon (O. kisiitch). Evidence sug-
gests that high turbidit\' or low light intensity
reduces predator selectivity because relative
differences in prey-detection distance for dif-
ferent sizes of prev are reduced (Vinvard and
O'Brien 1976, Gregoiy and Northcote 1993).
Gregory and Northcote (1993) observed log-
linear declines in reactive distance with
increased turbidity in chinook salmon (O.
tshaii'ijtscha).
Our results demonstrate that turbidity
reduces predation rates for all prey sizes for
both Lahontan redside shiner and Lahontan
cutthroat trout. Larger prey were generally
consumed with greater frequency, although
this frequency \'aries with turbiditx' and fish
species. The effect of prey size was most con-
sistent for Lahontan redside shiner. These fish
consumed more large (3.0 nun) pre\ at all tnr-
bidit> levels than did Lahontan cutthroat trout
(Figs, la, lb). In contrast, prey size had little
effect on the relative numbers of prey of each
size consumed by trout at turbiditv levels of
20 NTU or above'(Fig. la).
Redside shiner also consumed more pre\ oi
all 3 sizes combined over all turbidity levels.
For all prey sizes combined, shiner consimied
approximately 3% more prey than Lahontan
cutthroat trout at low turbidity levels and
approximately 10% more at high levels (Figs.
la, lb). Angradi and Griffith (1990) found pre-
dation by rainbow trout (O. tnyki.s.s) to be more
selective for large prey in clear water, \\hereas
selectivity was reduced in elevated turbidit\'.
Similar effects on prey selection under reduced
visibility conditions have been obser\ed in
bluegill sunfish {Lepoinis macrochirus). Under
low-light conditions bluegill sunfish consumed
fewer zooplankton but proportionally more
large individuals (Miner and Stein 1993).
Neither trout nor shiner have been shown
explicitly to possess adaptations that might en-
hance their effectiveness as foragers in turbid
waters. However, fish that feed nocturnally,
such as walleye {Stizostedion vitreiim), may
perform equally well in either clear or turbid
waters (Vandenbyllaardt et al. 1991). \\yie\'e
have higher densities of retinal cells and also
develop scotopic vision earlier in life in
comparison to salmonids (Vandenbyllaardt
et al. 1991, Borgstrom et al. 1992, Ilurber
and Rylander 1992). Such species-specific fac-
tors may contribute to differences in \ isual
performance.
Behavioral responses offish to turbidit> ma\
also affect their feeding abilities or rates. In
laboratoiy experiments, golden shiner {Nuteini-
gonus cnjsoleucas) showed increased flight
responses with increased turbidity (Chiasson
1993). Juvenile chinook salmon apparenth
experienced reduced predation from piscivo-
rous birds and fishes at elevated turbidit> le\'-
els (Gregory 1993). During our experiments,
redside shiner were observed to search faster
160
Great Basin Naturalist
[Volume 56
100
3.0 mm
LAHONTAN CUTTHROAT TROUT r^ i
(Oncorhynchus clarki henshawi)
10 15 20
TURBIDITY (NTU)
25
30
100
80
z
liJ
2 60
I-
§40
oc
UJ
°-20
2 mm
LAHONTAN REDSIDE SHINER
(RIchardsonius egreglus)
10 15 20
TURBIDITY (NTU)
25
30
Fig. 1. Mean percent prey consumed in relation to turbiditx. Upper pane! (a) shows results from feeding trials with
Lahontan cutthroat trout {Oncorlujnchus clarki henshawi), and lower panel (b) shows results from Lahontan redside
shiner (Ricliarclsoniiis egrcgiits). Four fish of each species were exposed to prey of a single size for 2-h feeding trials.
DapJuiia mapui prey sizes are as indicated. Vertical bars indicate 1 standard deviation.
and more widely at higher turbidity. Elevated
turbidity may have provided greater visual
i.solation and promoted greater mobility b>'
predators as suggested by Confer et al. (1978)
and Gradall and Swenson (1982). Increased
activity may have compensated for reduced
visual effectiveness, resulting in larger search
volumes for shiner than for trout. In a study of
brook trout {Salvclinus fontinalis) and creek
chub {Seinotihis atroinaciilatiis), Ciradall and
Swenson (1982) found creek chub to be less
affected by turbidity than brook trout. They
suggested such differential effects may explain
local disparities in fish density.
High turbidit}' in Summit Lake may decrease
reactive distance and search \ olinne unequally
for shiner and trout. This ma\' differentially
reduce the probabilitx of successful prey cap-
ture and could produce altered pre\' selection
patterns under different turbidity' conditions.
Although our results are generally similar to
those shown for other fishes (Vinyard and
O'Brien 1976, Berg and Northcote 1985, Li
et al. 1985), we document higliK significant
differences between potentially competing
fish species. Because Lahontan cutthroat trout
and Lahontan redside shiner consume the same
prey in Siunmit Lake, competition tor food
1996J
TiKBiDiiT Effects on Fish Ffkdinc
Ibl
may exist. Our results sut^^est that iu elevated
turbidit)' eouditious Lahontau redside shiner
nia> be a better competitor for food than
Lahontan cutthroat trout. A factor contribut-
inu; to the success of Lahontan redside shiner
in Summit Lake ma> be that their predation
rates are higher than those of cutthroat trout
at elevated turbidity levels.
Ackno\vled(;ments
We thank Larn' Marchant of the Lahontan
National Fish Hatchery and Alice Winzeler for
providing fish, and Louis Christensen for
assistance in setting up the experimental appa-
ratus. We thank R. S. Gregory and 2 anony-
mous reviewers for helpful suggestions for this
manuscript. The University of Nevada Depart-
ment of Biolog\' undergraduate tliesis commit-
tee facilitated completion of this project.
Literature Cited
Angradi, T. R., and J. S. Griffith. 1990. Diel feeding
chronology and diet selection of rainbow trout
{Oncorhynchiis mijkiss) in the Henry's Fork of the
Snake River, Idaho. Canadian Journal of Fisheries
and Aquatic Sciences 47: 199-209.
Barrett, J. C, G. D. Grossman, and J. Rosenfeld. 1992.
Turbidity-induced changes in reactive distance of
rainbow trout. Transactions of the American Fish-
eries Society 121: 437-443.
Berg, L., and T. G. Northcote. 1985. Changes in territo-
rial, gill-flaring, and feeding behavior in juvenile
coho salmon (Oncorhijnchits kisiitch) lollowing short
term pulses of suspended sediment. Canadian Jour-
nal of Fisheries and Aquatic Sciences 42: 1410-1417.
Borgstrom, R., a. Brabrand, and J. T. Solheim. 1992.
Effects of siltation on resource utilization and dynam-
ics of allopatric brown trout, Salmo tnittci, in a reser-
voir Environmental Biology of Fishes .34: 247-255.
Chiasson, a. 1993. The effect of suspended sediments on
ninespine stickleback, Piingitiiis pungitiii.s, and golden
shiner, Noteinigonits chrysoleiiccis, in a current of
varying velocitv. Environmental Biologv of Fishes
37; 283-295.
Confer, J. L., G. L. Howick, M. H. Corzette, S. L. K.\mer,
S. Fitzgibbon, and R. Landesberg. 1978. Visual
predation by planktivores. Oikos 31: 27-37.
Cowan, W, and R. Blake. 1989. Fisheries management
services contract #CTH50913089, annual report.
Report to Summit Lake Paiute Tribe. 31 pp.
Gradall, K. S., and VV. A. Swenson. 1982. Responses of
brook trout and creek chubs to turbidity. Transac-
tions of the American Fisheries Society 111:
392-395.
Gregory, R. S. 1993. Effects of turbidity on the predator
avoidance behavior of juvenile chinook salmon
(Oncorhynclni.s Ishawytsclm). Canadian Journal of
Fisheries and Aejuatic Sciences 50: 241-24(i.
Gregory, R. S., and T. G. Northcote. 1993. Surface,
planktonic, and benthic foraging by juvenile C:hi-
nook salmon (Oncorhynchus tsliauylscha) in turbid
laboratory conditions. Canadian Journal of Fisheries
and Aquatic Sciences 50: 2.3.3-240.
Guthrie, D. M. 1986. Role of vision in fish behavior.
Pages 75-113 in T. J. Pitcher, editor, The behavior of
teleost fishes. Groom Helm, [..ondon.
HoBSON, E. S. 1979. Interactions between piscivorous
fishes and their prey Pages 231-242 /;i R. H. Stoud
and H. Clepper, editors, Predator-i)re\' systems in
fisheries management. Sport Fishing Institute, Wash-
ington, DC.
HUBER, R., AND M. K. Rylandeh. 1992. yuautitative his-
tological study of the optic nerve in species of min-
nows (Cyprinidae. Teleostei) inhabiting clear and
turbid water Brain Behavior and Evolution 40:
2.50-255.
LaRivers, I. 1962. Fishes and fisheries of Nevada. .Nevada
State Fish and Game Commission, Reno. 782 pp.
Li, K. T, J. K. Wetterer, and N. G. Hairston, Jr. 1985.
Fish size, visual resolution, and pre\- selectivit^'.
Ecology 66: 1729-1735.
Miner, J. G., .\nd R. A. Stein. 1993. Interactiv e inlluences
of turbiditv' and light on larval bluegill (Lepomis
inacrachirus) foraging. Canadian Journal of Fisheries
and Aquatic Sciences 50: 781-788.
Sigler, E W. and J. W. SiGLER. 1987. Fishes of the Great
Basin: a natural histon; University of Nevada Press,
Reno. 425 pp.
SiGLER, J. W, T C. BjORNN, AND E H. EVEREST 1984. Effects
of chronic tiu^bidity on density- and growth of steel-
heads and coho salmon. Transactions of the .Ameri-
can Fisheries Society 113: 142-150.
U.S. Fish and Wildlife Service. 1995. Lahontan cut-
throat trout, Oncorhynchus chirki henshaui. recov-
eiy plan. Portland, OR.
Valeska, J. R 1989. Sunnnit Lake lacustrine studv tech-
niques. Unpublished manuscript. Summit Liike Paiute
Tribe, Winnemucca, NV
Vandenbylla.\rdt L., E J. Ward, C. R. Brakev eli. and
D. B. McIntyre. 1991. Relationships between tur-
bidity, piscivoiy, and development of the retina in
juvenile walleyes. Transactions of the .American
Fisheries Society 120: 382-390.
Vinyard, G. L., and W. J. O'Brien. 1976. Effects of light
and turbidity on the reactive-distance of bluegill
{Lepomis inacrochirns). Canadian Journal of Fish-
eries and Aquatic Sciences 33: 284.5-2849.
Vinyard, G. L, and A. L. Winzeler. 1993. Results of
investigations at Summit Lake. Report to Summit
Lake Paiute Tribe. 62 pp.
Received 12 June 1995
Accepted 19 January 1996
Great Basin Naturalist 56(2), © 1996, pp. 162-166
POGONOMYRMEX OWYHEEI NEST SITE DENSITY AND SIZE ON
A MINIMALLY IMPACTED SITE IN CENTRAL OREGON
Peter T. Soule' and Paul A. Knapp-
Abstract. — Little is known about the basic characteristics of the western hanester ant {Pogonoimjrmex oictjheei) in
the absence of anthropogenic disturbances. We examined the role of P. oictjhcei as an agent of disturbance in an area of
semiarid \ egetation in central Oregon known as the Horse Ridge Research Natural Area (HRRNA) that has been largely
free of livestock grazing and other significant anthropogenic influences for over 23 yr. We determined densit\' and size
characteristics of nest sites and estimated total area cleared by P. owyheei activities on HRRNA. From random sampling
of twenty-five 0.04-ha plots we found a mean nest density /standard eiror of 1.6 (±0.16) nests/0.04 ha. Mean area cleared
per nest site was 4.8 m-, which results in an estimated banen area of 46,080 m- on the 240-ha HRRNA. Comparing our
findings to others on P. owyheei and P. occidentalism we foimd nest densit} and mean cleared area to be in the middle
range of reported obsenations under a \ariet> of land-use influences. The literature suggests that moderate disturbance
ma\' increase nest site densit>', but little relationship exists between distinbance histon' and mean size of nest sites.
Key words: Pogonom\rmex ow>heei, western harvester (uits. nest density, nest size, vegetation clearing.
Western hai^vester ants are a major compo-
nent of arid rangeland ecosystems in the United
States. Because of the combined effects of seed
predation, seed dispersal, and vegetation re-
moval, har\'ester ants are "keystone species,"
meaning their effects on vegetation structure
and dynamics exceed expectations given their
density and biomass (Holldobler and Wilson
1990: 616). The most visible impact of har-
vester ant activities is vegetation clearing
around their nest sites. Although the size of
the cleared area, or disc, varies, Pogonomijnnex
harvester ants have the capacity to cut annual
plants surrounding their nest sites at rates of
over 200 million plants/ha/yr (Clark and
Comanor 1975). While much of the plant bio-
mass cut is not consumed by the ants, it
reduces the total volume availal)le for con-
sumption b)' livestock and other grazers (Willard
and Crowell 1965). Range managers have
viewed Pogunonnjnnex as pests that need to be
controlled, giving the ant both economic and
ecological importance in arid rangelands (Wight
and Nichols 1966, Cole 1968).
Because of the paucity of undistiubed areas
in the semiarid West, little is known about the
basic characteristic of P. owyheei nest sites in
the absence of anthropogenic disturbances.
The primaiy objectives of this stud) are (1) to
determine the densit\ and size characteristics
of P. owyheei nest sites and (2) to estimate the
total area denuded b\' clearing and foraging
acti\'ities of P. owyheei within a largely undis-
turbed semiarid ecosNstem.
Study Area
The Horse Ridge Research Natural Area
(HRRNA) is a 240-ha exclosure 31 km south-
east of Bend, Oregon, managed by the Prine-
\'ille District, Bureau of Land Management
(BLM). The natural area was established in
1967, and a surrounding fence was completed
in 1974. The exclosure ranges from 1250 to
1430 m elevation over rolling topography of
Columbia Basalts (Anonymous 1972). Direct
human impacts on the site are minimal as
there is only occasional use by hunters and
naturalists, and fire suppression is not active
(HaKorson 1991, R. Halvorson personal com-
munication 1995). The fence has kept the area
free of livestock grazing since 1974, but before
its establishment the area apparently received
minimal domestic animal grazing pressure
because of a lack of a permanent water source
to attract animals (Anon\'mous 1972) and the
distance from well-traveled public roads
(Baldwin 1974). AdditionalK, the abundance
on HRRNA of threadlea\ ed sedge (Carex fiU-
Jolia). a species that has been shown to decline
'ncp;irtniciiliir(;ciii;rapli> .mrI I'l.iiiiiiim. Aiip.ihuliiaii State L iii\rrsit\, lioone. NC 28fi07
^Dcpartiiii'iil 111 f ;ccii;rapliy, Ceoiyia Slalr UiuMMsit), Allantu. C..\ :H)M)^.
162
199(i]
r. OwiiiEEi Nest Densih and Size
163
because of o\'ergrazing; in the central Oregon
sagebrush steppe, and the absence of clieat-
grass {Broinus tectonim) suggest a minimally
disturbed site (Anon>'nious 1972, personal ob-
servation 1995).
Vegetation on IIHRNA is classified as the
western juniper/big sagebrush/threadleaved
sedge community {Jiiniperiis occidentalisi
Artemisia tridciitata/Carcx filijolia) (Franklin
and D\'rness 1988). Less common but present
species are bluebunch wheatgrass {Agropyron
spicatwn), Idaho fescue {Festuca idahoensis),
junegrass {Koeleria cristata), and horsebrush
{Tctradijniia glahruia) (Anonymous 1972).
HRRNA climate is dominated by winter
precipitation. Over half the annual 31 cm falls
as snow. Mean temperatures at Bend range
from -0.6°C in Januaiy to 17.7°C in Juh' (Karl
et al. 1990).
Soils on our study plots are entirely within
the Stookmoor-Wesbutte complex soil series
(USDA-NRCS in press). This soil series is
found on approximately 85% of HRRNA. A
t> pical soil profile is represented by a surface
layer of mixed ash and loamy material approxi-
mately 15 cm thick, and a pale brown, sandy
loam subsoil 46 cm deep overlying bedrock.
Percentage of organic matter in the topsoil is
l%-2% and 0.5%-2% in the subsoil (USDA-
NRCS in press).
Besides P. owyheei, there is disturbance
pressure on HRRNA from grazing activities of
herbivores and granivores such as Rock-)' Moun-
tain mule deer {Odocoileus hemioniis Jitnniomis),
badger {Taxidea taxus), and cottontail rabbits
{Syhilagus niif(dli) (Gashwiler 1972, personal
obsei-vation 1995). BLM records on HRRNA
report no outbreaks of intense herbivory or
episodes of pathogens causing severe plant
losses in the last 20 years (R. Halvorson per-
sonal communication 1995).
Methods
In roughly the center of HRRNA, a 19.6-ha
permanent grid was established by (kislnviler
(1977) for use in an ecological study in 1972.
Stations on the 12x12 grid are marked !)>• re-
bar stakes and spaced 40.2 m apart. Using this
grid, we randomly selected 25 stations and
established 0.04-ha circular plots from the lebar-
marked center points for a total sample area of
1 lia. We tallied and measured each acti\'c P.
owyheei nest site within each plot. We placed
line transects over the center of each nest site
and measured the cleared disc area in north-
south antl east-west directions. The edge of
each disc was determined i)y the intersection
of any perennial with the N-S or E-W transect
lines.
Results
There were 40 active P. owyheei nest sites
in our 1-ha sample. We found nest sites on 23
of the 25 circular plots, and the maxinumi
number of nest sites was 3 pei- 0.04 ha. Mean
nest density/standard error was 1.6 (±0.16)
nests/0.04 ha. Characteristics of the cleared
discs are shown in Table 1. Assuming a circu-
lar shape, the mean area cleared per nest site
is 4.8 m^. Factoring in the nest density results
in an estimated barren area of 192 m^/ha, or
1.92% of the total land area of the permanent
grid. If the influences oi P. owyheei are consis-
tent throughout the 240-ha HRRNA, then ant
foraging and plant cutting surrounding a total
of 9600 nest sites should leave approximateK
46,080 m- of barren land on the 240-ha site.
DlSCUSSI(3N
The premise of this article is to pro\ ide
information on P. owyheei nest site densitx
Table 1. Characteristics of P owyheei nest sites on HRRNA.
Standard
Discs
Mean
Median
M,
aximum
M
inimum
de\iation
.V
(cm)
(cm)
(cm)
(cm)
(cm)
N-S diameter
241.1
207.5
740
60
144.4
40
E-W diameter
254.6
220.0
670
68
1.56.S
40
164
Great Basin Naturalist
[Volume 56
Table 2. Pogonomyrtnex oivyheei and P. occidentalis nest
ren area due to P. ouyheci and P. occidentalis aeti\ ities repor
site densities and mean size of nest site and estimated Iiar-
ted in the literature.
Source
Pugonomynnex Nest site
State species densit\7h;i
Nest site
\ sr/,e 111 Ill-
Estimated
barren Stiidv' site
area % disturbance
Dominant
\egetation
SiiarpandBan(19«))
Idah(
ouyheei-'
Sharp and Barr (I960) Idaho owijheei^
Sharp and Barr (1960) Idaho owyheei'^
Willard and Crowell (1965) Oregon oivyheei
Wight and Nichols (1966) Wyoming occidentalism
Rogers and Lavigne (1974) Colorado occidentalis
Rogers etal. (1972)
Rogers et al. (1972)
Rogers et al. (1972)
Colorado occidentalis
Colorado occidentalis
Colorado occidentalis
Clark and Comanor (1975) Nevada occidentalis
Sneva (1979)
Sneva (1979)
Sneva (1979)
Oregon
Oregon
vyheei
vyheei
Oregon oicyhcei
Coffin and Lauenrotli (198S) Colorado occidentalis
CofFin and Lauenroth (1990) Colorado occidentalis
Nowaket al. (1990) Idaho owyheei
Nowak et al. (1990)
Idahi
owyheei
40
0.8
6.0
"misused/
depleted"
Atriplex nuttallii /
Halogeton glomeratus
9
1.3
3.7
"vigorous stand"
Atriplex nuttallii
12
nr''
nr
not discussed
Ariplex confertifoliei
49-74
22.5
11-17
not discussed
Bromiis tectorwn
nr
65.7
m-
lightK' grazed'^
Atriplex nuttallii
23
1.2
0.3
ungrazed for
30 years
Buchloe dactyloides 1
Bouteloua gracilis
28
0.7
nr
lightly grazed
Buchloe dactyloides 1
Bouteloua gracilis
31
0.4
nr
moderate grazing
Buchloe dactyloides 1
Bouteloua gracilis
3
0.6
0.02
lieaNy grazing
Buchloe dactyloides /
Bouteloua gracilis
30-13
2.4-15.9
nr
varied — lightly
grazed / recent
burns
Artemisia tridentata /
Agropyron desertorum
32
9.3
3.0
grazed pasture/
no intensit\-
specified
Artemisia tridentata 1
Agropyron spicatum 1
Stipa thurheriana 1
80
0.9
0.7
lightly grazed/
brush control
10 \T prior to
stud)' killed 95%
of plants
Artemisia tridentata 1
Agropyron spicatum 1
Stipa thurheriana 1
Bromus tectorwn
57
1.5
0.8
lightly grazed/
brush control
22 yr prior to study
killed 95% of plants
Artemisia tridentata 1
Agropyron spicatum 1
Stipa thurheriana 1
Bromus tectorum
25
1.4
nr
moderateK grazed
Bouteloua gracilis
31
1.2
nr
lightK grazed
Bouteloua gracilis
nr
3.5
nr
no grazing or
fire in 30+ yr
Artemisia tridentata 1
Oryzopsis hymenoides
11 r
5.3
nr
burned 5 yr
prior to sample,
then ungrazed
Artemisia tridentata 1
Onjzopsis hymenoides
"Identified as occidciilalis. Imt in the kiumii ranne of i)» |//i( ci
''Not reporteci
^All references to grazing refer to grazing of cattle or other li\'estocl<.
"P. owyheei was considered to be part of P occidentalis until 19.50.
and cleared disc size in an undisturbed area.
Much of the information on areas cleared b\'
Pogonomijrmex hai^vester ants relates to stud\'
sites with vaiying degrees of disturbance his-
toiy However, few studies examine the role of
P. owyheei and P. oceidentalis as agents of
plant removal in undisturbed environments.
In our study we briefly compare results of
plant removal in undisturbed areas with those
results presented elsewhere.
Our nest site density of 40/ha is in the
approximate middle range of reported obser-
vations under a variety of land-use influences
(Table 2). Disturbance may sene to inciease the
nest site densities at any given site up to a point.
For example, Rogers and Lavigne (1974: 995)
found an increase in nest site density under
"light" and "moderate" grazing, but shaiply re-
duced densities under "heavy" grazing. Findings
of Sharp and Barr (1960) and Sne\ a (1979) also
suggest increases in nest site densit\' are asso-
ciated with distiubance (Table 2). Across the
range of P. owyheei and P. oecidentalis, nest
site densities are likeK controlled b\ a suite of
factors (soils, \egetation composition, climate,
disturbance histor}) acting synergistically.
Increases in nest site density in grazed areas
probabK result from alterations of the d\nam-
ics of competition between plant species that
in turn modifv seed densit\ distributions
1996]
P. OwYHEEi Nest Density and Size
165
(HolldoMer and Wilson 1990). On their study-
site in southern Arizona, tor example, Da\id-
son et al. (1984) found that haivester ant popu-
lations began to decrease approxiniateK 2 > r
after rodent populations were intentionally
reduced. Da\idson et al. (1984: 1780) con-
eluded that rodent removal led to a "differen-
tial increase ' in large-seeded annuals because
of the cessation of granivory, and this in turn
precipitated the competiti\e displacement of
small-seeded species that were the ant s pri-
man' food source.
Although other studies have used larger
sample sizes to determine nest density (e.g..
Coffin and Lauenroth [1988] used a 2.5-ha
sample), we believe our nest site density is a
reasonable estimate for HRRNA because (1)
the study site is consistent in regard to soils
and \'egetation, and has only minor topo-
graphic variability; (2) our standard error per
sample for nest density is small, suggesting lit-
tle variability within our study area; and (3)
research from studies on other Pogonotnyrmex
species has shown that soil texture can affect
nesting location (e.g., Johnson 1992, DeMers
1993), and that a uniform dispersion of ant
colonies develops regardless of spatial scale
examined (Wiernasz and Cole 1995). There
appears to be little relationship between dis-
turbance history and mean size of nest sites
{Table 2). Sneva (1979) has speculated that
while there may be great variability in nest
site density and disc area, the potential avail-
able forage per nest site generally remains
consistent, suggesting that vegetation cover
and species composition can affect disc size.
Soil characteristics also impact disc size, with
a tendency for colonies to expand horizontally
in shallow soils (Sneva 1979). Therefore, disc
size may be largely linked to the amount of
vegetation cover, plant species composition,
and soil depth, and less influenced by distur-
bance than is nest density.
Literature Cited
Anonymous. 1972. Horse Ridge Re.searcli Natural Area.
Federal Research Natural Areas in Oregon and
Washington. Pacific Northwest Forest and Range
Experiment Station, Portland, OR.
Baldwin, E. M. 1974. Evaluation of Horse Ridge Natural
Area, Deschutes County, region for eligibility for
registered natural landmark designation. Supple-
mental Report, University of Oregon, Eugene.
Clark, W. H., and P L. Comanor. 1975. Removal of
annual plants from the desert ecosystem by western
harvester ants, Pononoimjrrnex occideutalis. En\ iron-
mental Entomology 4: 52-56.
Coi-i in, D. P, and W. K. LAL'liNKOTli. 1988. The ell'ects of
disturbance size and frcciuenc)' on a sliortgrass plant
conunuiiity. Ecology 69: 1609-1617.
• 1990. Vegetation associated with nest sites ol'
western harvester ants {Po^oiiomi/nncx accUlcntalis
Cresson) in a semiarid grassland. American Midland
Naturalist 123; 226-2.35.
Cole, A. C, Jh. 1968. Po^,()ii(>inijnii('x harxesler ants: a
study of the genus in North America. The University
oi ieunessee Press, Kno.wille. 222 pp.
Davidson, D. W, R. S. iNouvt:, and J. H. Brown. 1984.
Granivoiy in a desert ecosystem: experimental evi-
dence for indirect facilitation of ants b\ rodents.
Ecology 65: 1780-1786.
DeMers, M. N. 1993. Roadside ditches as corridors for
range expansion of the western harvester ant (fogo-
nomi/nncx occidcnidlis Cresson). LaiidscaiK' Ecolog\'
8: 93-102.
Fr\nklin, J. F, AND C. T. DVRNESS. 1988. Natural vegeta-
tion of Oregon and Wiishington. Oregon State Uni-
versity Press, Cor\'allis. 452 pp.
Gashwiler, J. S. 1972. List of birds, manunals, and plants
found on Horse Ridge. Letter (12/27) to Paul W.
Arrasmisth, Prineville, OR, BLM.
. 1977. Bird populations in four \egetational t>pes
in central Oregon. Special Scientific Report — Wildlife
No. 205. United States Department of Agriculture,
Fish and Wildlife Service, Washington, DC.
Halvorson, R. 1991. Natural ignition (fire) in Horse Ridge
ACEC/RNA, concerns and challenges. Correspon-
dence (9/16) to ADM-Resomce Serx'ices, Princv ille.
OR, BLM.
HOLLDOBLER, B., AND E. O. WiLsoN. 1990. The ants. The
Belknap Press, Cambridge, MA. 732 pp.
Johnson, R. A. 1992. Soil texture as an influence on the
distribution of the desert seed-hanester ants Pofioii-
omynnex rugosus and Messor pergandei. Oecologia
89: 118-124.
K.\RL, T R., C. N. Williams, Jr., F T Qhnlan, and T A.
BODEN. 1990. United States Historical Climatolog\'
Network (HCN) serial temperature and precipita-
tion data. Carbon Dioxide Information Analysis
Center, Oak Ridge, TN. 379 pp.
NowAK, R. S., C. L. NowAK, T. DeRocher, N. Cole, and
M. A. Jones. 1990. Prevalence of Orysopi.s hijincnokh's
near harvester ant mounds: indirect facilitation b\
ants. Oikos 58: 190-198.
Rogers, L. E., and R. J. Lavigne. 1974. Environmental
effects of western hai-vester ants in the sliortgrass
plains ecosystem. Environmental Entomolog\- 3:
994-997.
Rogers, L. E., R, J. Lwigne, and J. L. Miller. 1972.
Bioenergetics of the western harvester ant in the
shortgrass plains ecosystem. Environmental Ento-
mology 1: 763-768.
Sharp, L. E., and W F Barr. 1960. Prcliminan investiga-
tions of hanester ants on southern Idaho range-
lands. Journal of Range Management 13: 131-134.
Sneva, F A. 1979. The western harvester ants: their den-
sity and hill size in relation to herbaceous productiv-
ity and big sagebrush coven Journal of Range Man-
agement 32: 46-47.
USDA-Natural Resources Conservation Service. In
press. Upper Deschutes River Area, Oregon Soil Sur-
vey. USDA-NRCS, Washington, DC.
166
Great Basin Naturalist [Volume 56
WIERNASZ, D. C, AND B. J. CoLE. 1995. Spatial distribu- Willard. J. R. AND H. H. CRO^VELL. 1965. Biological
tion of Pogonomyrrnex occidentdis: recnutment. mor- activities of tl.e harvester .ni, Togonomynnex ou..jhe.^
tality and overdispersion. Journal of Animal Ecology in central Oregon. Journal of Economic Entomology
64: 519-527.
58: 484-489.
Wight J. R., and J. T Nichols. 1966. Effects of harvester . , ,o m 7 loo^t
* 1 „„ p J,c.o„ of a saUb,.^ c„,„„,u„«y, ,o,„™l r'^'^S* 'Z
of Range Management 19: 69-/1. ^^
Great Basin Xatiiralist 56(2), © 1996, pp. 167-171
FIELD MEASUREMENTS OF ALKALINITY FROM LAKES
IN THE UINTA MOUNTAINS, UTAH, 1956-1991
Dennis I). Austin'
Abstract. — Data follcitid hom alpine lakes in tl:r L inla Monntains dnrinu; I'islieiy siint-ys In' tlie Utali Division of
Wildlife Resources indicate alkalinit\ has decreased in some drainages since tlie mid 195()s. implications for continued
monitoring, as well as environmental and recreational values, are discussed.
Key uords: alkdliiiify. acid pirripitdtioii, dlpiiic lakes, tcaler (jiialitij, I'iiili.
Alpine lakes in the Uinta Mountains have
the lowest total alkalinity of all suriaee waters
in Utah (EPA 1982). The low alkalinity is due
to the Piecanibrian roek geologie origin eom-
posed primarily of nietamorphic quartzite,
ph\llite, and diamictite. Because of low alkalin-
it>; these lakes are sensitive to acid precipita-
tion, which may affect long-term water quality,
fish, and other a(|uatic organisms. The Utah
Division of Wildlife Resources has measured
alkalinity in many of these lakes since 1956.
The purpose of this paper is to document the
changes in alkalinit\' between 1956 and 1991
in the Uinta Mountains by drainage.
Methods
Water from lakes in the Uinta Mountains,
Utah, was sampled and measured for alkalinity
from 1956 to 1991 b>' the Utah Division of
Wildlife Resources in conjunction with the
fisheries sun^eys. Data were collected during
summers on selected lakes within 16 of the 18
major drainages (Fig. 1) and during 3 desig-
nated sampling periods: mid 1950s-early 1960s
(period 1), 1970s-early 1980s (period 2), and
mid 1980s-early 1990s (period 3). All alkalin-
it>' data were collected in the field using col-
orimetric methods and converted to mg/L. In
period 1, tests were made using methylpuqDle
indicator and titrating with 0.02 N sulfuric
acid. Alkalinity' titrations were made at stepwise
increments of 5.0 mg/L (per drop). In periods
2 and 3, tests were made with Hach (Hach
Company, PO Box 389, Loveland, CO 80539)
Water Ecology Kits, model AL-36B. Alkalinity
titrations were made at increments of 6.8
mg/L (per drop) in all drainages, except during
period 2 in Rock Creek, Duchesne, and Provo
River drainages, and during period 3 in Hock
Creek, Burnt Creek, and Sheep-Carter C'reek
drainages when the increments were 17.1
mg/L (per drop). The effects of 3 weaknesses
in the available data — the lack of data sets
from all drainages during all 3 periods, the 3
levels of sensitivity in the alkalinity measure-
ments, and the differences in sample sizes —
are unknown and suggest interpretive caution
of the results. The significance lc\el was set at
P < 0.05 for the 3 comparisons of statistical
testing.
To test for changes over all drainages, mean
alkalinity among drainages was compared
between periods using ANOVA for imequal
sample sizes (Sokal and Rohlf 1981).
To test for changes in alkalinit\ within drain-
ages, data were compared between periods. T
tests of the mean were used when data from 2
periods were available, and ANOVAs for un-
equal sample sizes were used when 3 periods
of data were available.
To test for changes in alkalinit)' within
drainages for the same sampled lakes, I com-
pared data between periods. T tests for paired
comparisons were used when 2 periods of data
were available, and ANOVAs for equal sample
sizes when 3 periods of data were a\ ailable.
Results
Mean alkalinity among drainages signifi-
cand\' decreased (F < 0.05) between all 3
' Dupiirtimiit of Rangf Scit-nce. i:tali State Uni\'ersit>', Logan, UT 84322-,5230. Present addres.s: 43 .South 700 East. H>Tuin. L'T 84319.
167
168
Great Basin Naturalist
[Volume 56
SALT LAKE CITY
US 189 SOUTH
US 80 NORTH
U S 40^
U S 30
KAMAS
• DUCHESNE RIVER
TABIONA^y J ROCK CREEKS! BEAR RIVErIN^
UTAH
WYOMING
U-150
EVANSTON
DUCHESNE
LAKE FORK ^BLACKS FORK J
YELLOWSTONE RIVER .SMITHS FORK>
*•>:.*•;•:• %•;;.•.*. :.\\\\\\v
gSWIFT CREEK| HENRYS FORK
^DRY GUlCHll;.- ^^^^^^^^^^^^
:••.•.*••-••-••••• EEAVER CREEK ^ (
\'.V UINTA RIVER,*, j^——- '
' • « % <
WHITEROCKS/
KAPOINT/
•WHITEROCKS RIVER ^""^^^ ^^^^
'/ ^ SHEEP CREEK
ASHLEY CREEK
CARTER 5
^5^ CREEK 1^
•ft
'MOUNTAIN VIEW
LONETREE
RED( CLOUD
LOOPl ROAD
U.S. HIGHWAY 30 EAST
MANILA
I
VERNAL
U.S. HIGHWAY 40 EAST
U-44
DUTCH JOHN
FLAMING GORGE RESERVOIR!
\
I
NORTH
Fig. 1. General location and major clrainatjes in the Uinta Mountains. Utah.
1996]
Uinta Lakes Alk.\linity
169
periods (Table 1) from 33 niti/L in period 1, to
23 mg/L in period 2, to 17 nie;/L in period 3.
(Standard de\iations for all means listed in
Tiible 1 are available from the anthor)
Alkalinity within individnal drainai^es sit:;-
nifieantK deereased in the Dnehesne Kixcr
and l^r()\() Hi\er drainages between all 3 peri-
ods. Alkalinit>' decreased between periods 1 or
2, and period 3 in the Rock Creek, Weber River,
and \Vhiteroeks River drainages. No change in
alkalinity was fonnd between periods 1, 2, and
3 in the Bear River drainage. Similarly, no
change in alkalinity between periods 2 and 3
was fonnd from Beaver Creek, Blacks Fork,
Smiths Fork, and Henns Fork drainages. Due
to lack of data, no additional comparisons could
be made.
Changes in alkalinity within drainages, where
data from the same lakes were a\'ailable be-
tween 2 periods, were variable. Alkalinity did
not change in die Rock Creek drainage between
periods 1 and 3, or in the Weber River drain-
age between periods 1 and 2. Alkalinity also
showed no change in the Rock Creek, Bear
Rixer, Blacks Fork, Smiths Fork, or Whiterocks
River drainages between periods 2 and 3.
However, alkalinity decreased in the Duchesne
River and Provo River drainages between
periods 1 and 2, 2 and 3, and 1 and 3. Alkalin-
ity also decreased in the Weber River and
Bear River drainages between periods 1 and 3,
but increased in the Henrys Fork and Beaver
Creek drainages between periods 2 and 3. No
additional comparisons could be made.
Alkalinity in the Duchesne River and Provo
River drainages where the same lakes were
sampled during all 3 periods was significantly
(F < 0.05) different between all 3 periods for
both drainages. Mean alkalinity values using
the combined data from these 2 drainages
decreased from 37 mg/L in period 1 to 22 in
period 2, to 6 in period 3. No comparisons
over the 3 periods could be made from the
other drainages.
Discussion
Negative effects of acid precipitation on
aquatic ecosystems have been well documented,
particularly in Europe and eastern North
America (Haines 1981a). Acid precipitation
can have negative impacts on water chemistiy
and quality, algae, bacteria, invertel)rates,
amphibians, fish, waterfowl, and aquatic vege-
tation (Rough and Wilson 1977, Fl^A 1979,
Haines 1981b, Kretser et al. 1983), and result
in a general reduction in biodiversity (Fryer
1980).
Alkalinity and pll are directly related in
maintaining acjuatic ecosystems; and as alka-
linity decreases, lakes become increasingly sus-
ceptible to acidification (Haines 1981a). Acidi-
fication rates were reported by Dillian et al.
(1987) for 2 Canadian lakes as 2 ue(|/L/yr be-
tween 1979 and 1985 with a 3-fold decrease in
alkalinity accompanied b\' a 0.2 pH decrease.
Decreases in alkalinity have been reported
in Colorado. In die Colorado Rockies, 64 lakes
were compared between 1938-1960 and 1979
with a mean decrease between periods (1938-
1960 vs. 1979) of 17% alkalinity (Lewis 1982).
In the Mt. Zirkel Wilderness Area, Turk and
Campbell (1987) reported an approximate loss
of buffering capacity of < 10% in most lakes
they sui-veyed.
Data from this study indicate a 50% de-
crease in alkalinity since the 1950s, with the
rate of decrease about 0.5 mg/L/yr in the Uinta
Mountains. At this rate of decrease, studies
extended for only a few years would likely
show no change in alkalinit\.
Contrary to our results, 2 previous studies
conducted in Utah indicated no effects of acidi-
fication. In a snowmelt stud\' of the Wasatch
Mountains, Messer et al. (1982) reported a
mean snowmelt pH of 6.17 and concluded that
enough buffering capacity was retained in the
snowpack to neutralize acid ecjuivalents from
air pollution. In a report from the Utah Techni-
cal Advisory Committee (m acid deposition,
Ellis (1986) concluded that although lakes and
streams in the Uinta Mountains are ven' sen-
sitive to acidification, no evidence was found
that demonstrated acidification had occurred.
The lack of acidification was based primarih
on data collected in the Mirror Lake water-
shed during 1983-1986. Both studies sug-
gested windblown particulates from the Great
Salt Lake Desert were sufficient to buffer acid
deposition.
Decreased alkalinity from alpine lakes sam-
pled by the Utah Division of Wildlife Resources
in the Uinta Mountains over 35 years indi-
cated a slow decline in alkalinity; particularly
in the Provo River and Duchesne River drain-
ages. Unaltered, this decline may eventually
result in deterioration of the aquatic ecosys-
tem and, sul)sequently, recreational values.
170
Great Basin Naturalist
[Volume 56
Table 1. Mean total alkalinib.' (mg/L) by drainage from alpine lakes in the Uinta Mountains, Utah.
C
Dnibine
d data from all
sampled lake
Phase n = IS
Period 1
Period 2
Period 3
Drainage
Year
n
mg/L
Year
11
mg/L
Year
n
mg/L
Rock Creek
1956
9
311"
1973
3
34"
1983
54
21b
Duchesne River
1956
30
35''
1979
1
241'
1985
34
8^
Provo River
1956
23
36"
1979
20
231'
1986
54
8^-
Weber River
1956
27
35"
1983
3
22"1'
1987
16
12''
Bear Ri\'er
1956
5
30
1982
26
17
1989
30
21
Blacks Fork
ND^
—
—
1982
22
26
1989
21
25
Smiths Fork
ND
—
—
1983
24
16
1990
20
15
Henr\'s Fork
ND
—
—
1984
22
15
1990
21
24
Beaver Creek
ND
—
—
1984
23
20
1991
17
29
Biunt Fork
ND
—
—
ND
—
—
1984
11
17
Sheep/Carter
ND
—
—
ND
—
—
1984
32
19
Creeks
Ashley Creek
ND
—
—
ND
—
—
1988
21
13
Whiterocks River
ND
—
—
1976
2
34"
1985
43
14''
Uinta River
ND
—
—
ND
—
—
ND
—
—
Di-y Gulch
ND
—
—
ND
—
—
1987
12
14
Yellowstone River
ND
—
—
ND
—
—
1986
24
14
Lake Fork River
ND
—
—
ND
—
—
ND
—
—
Swift Creek
ND
—
—
ND
—
—
1987
17
14
Total/Mean
—
94
33''
—
152
23'-
—
427
17^-
In umbers witli difierent letters across rows wen
^NDC = No data from common lakes available.
3ND = No alkalinit\' data collected.
sitjnilkantK dittercnt. F < 0.0.5.
Additional sampling is essential to monitor
and document alkalinit\' and potential acidifi-
cation of Uinta Mountain lakes.
Acknowledgments
This report was funded, in part, by the Utah
State Division of Wildlife Resources, Pittman-
Robertson, Federal Aid Project W-105-R. Spe-
cial thanks is given to Jeriy D. Weichman for
his careful reviews of this manuscript.
Literature Cited
DiLLiAN, D. J., R. A. Reid, and E. de Grosbois. 1987.
The rate of acidification of aquatic ecosystems in
Ontario, Canada. Nature 329: 45-49.
Ellis, M. T. 1986. Acid deposition in Utah. Utah Depart-
ment of Health, Salt Lake Cit>'.
EPA. 1979. Research summary: acid rain. U.S. En\iron-
mental Protection Agency Publication 600/8-79-028.
Washington, DC.
EPA. 1982. Total alkalinity' of smface waters. U.S. En\ iron-
mental Protection Agenc\ Publication 6()0/D-82-
333. Corvallis, OH.
Fryer, G. 1980. Acidity and species diversity in iresh
water crustacean faunas. Freshwater Biolog\ 10:
41-45.
Haines, T. A. 1981a. Acidic precipitation and its conse-
quences for acjuatic ecosystems: a review. Transac-
tions of the American Fisheries Society 110:
669-707.
. 1981b. Waterfowl and their habitat: threatened by-
acid rain? Pages 177-190 //; 4th International Water-
fowl Symposium.
KrETSER, W a., J. R. COLQLHOUN, .\ND M. H. Pfeiffer.
1983. Acid rain and the Adirondack sportfishen.
Conservationist 37: 22-29.
Lewis, W M., Jr. 1982. Changes in pH and buffering
capacity of lakes in the Colorado Rockies. Limnology
and Oceanography 27: 167-172.
Messer, J. J., L. Slezak, and C. I. Life 1982. Potential
for acid snowmelt in the Wasatch Mountains. Water
Quality Series UWRL/Q-82/06. Water Research
LaboratoiT, Utali State Uni\ersity, Logan.
PouGH, F H., AND R. E. Wilson. 1977. Acid precipitation
and reproduction success of AmhijsfoiiKi salaman-
ders. Water, Air, and Soil Pollution 7: 307-316.
SoiCAL, R. R., AND F J. ROHLF 1981. Biometn. 2nd edi-
tion. W. H. Freeman and Company, New York.
Turk, J. T., and D. H. Campbell. 1987. Estimates of acid-
ification of lakes in the Mt. Zirkel Wilderness .\rea,
Colorado. Water Resources Research 23: 1757-1761.
Received 22 March 1995
Accepted 30 Decouher 1995
1996]
Uinta Lakes Aljwvlinitv
171
Tablk 1. Contiiuifd.
Dat;
from the
saiiic lakes
Data ti
oni the same lakes sampled during 2
periods
saTiipl
L'd during
all 3 jieriotls
1
2
1
3
2
3
1
2 3
11
inii/L
niU/L
/(
ni,ti/L
ni.iVl.
n
niK/L
mg/L
II
mg/L
mg/L iiiK/L
NDC-
—
—
8
30
24
3
34
17
NDC
6
35''
231'
21
36"
5''
6
23"
61'
6
33"
23'- 6'-
14
37''
211'
14
38"
71,
15
22"
5I'
10
41"
21'' 6^-
2
33
7
8
33^'
10''
NDC
—
—
NDC
NDC
—
—
5
30"
I3I'
17
20
20
NDC
_ _
NDC
—
—
NDC
—
—
18
24
26
NDC
NDC
—
—
NDC
—
■ —
13
11
12
NDC
NDC
—
—
NDC
—
—
14
12"
171,
NDC
NDC
—
—
NDC
—
—
17
22"
29''
NDC
NDC
—
—
NDC
—
—
NDC
—
NDC
NDC
—
—
NDC
—
—
NDC
—
—
NDC
—
— —
NDC
NDC
NDC
_
_
NDC
NDC
—
—
NDC
—
—
2
34
14
NDC
NDC
—
—
NDC
—
—
NDC
—
—
NDC
NDC
—
—
NDC
—
—
NDC
—
NDC
NDC
—
—
NDC
—
—
NDC
—
—
NDC
NDC
—
—
NDC
—
—
NDC
—
—
NDC
NDC
—
—
NDC
—
—
NDC
—
—
NDC
22
35^'
171-
56
33"
12''
88
23
17
16
37"
22'^ 6^
Great Basin Naturalist 56(2), © 1996, pp. 172-176
DENSITY, BIOMASS, AND DIVERSITY OF GRASSHOPPERS
(ORTHOPTERA: ACRIDIDAE) IN A CALIFORNIA NATIVE GRASSLAND
Eric E. Porterl-, Richard A. Redakl, and H. Elizabedi Braker^
Abstract. — A native California perennial grassland \\'as sampled for grasshopper populations. The grassland is man-
aged for the presei-vation of the native perennial bunchgrass, Nassella pulchra Hitch. Grasshopper densit)', biomass,
diversity, and richness were measured from July 1993 to October 1994. Average density of all grasshoppers was 2.30
hoppers/m- (0.66 s) for 1994 (June through August). Overall forage consumed for 1994 was 140 kg/ha, suggesting diat
grasshopper populations e.xist at economically damaging levels. Grasshoppers do not appear in the grasslands until late
spring, after annual grasses have set seed. Biomass of grasshoppers peaks in July when adults are predominant. Both
grasshopper density and biomass were higher in 1993 than in 1994, and a total of 5 species were found throughout the
stud). Mclanophis sangiiinipes Fabricus dominated the acridid communities and accoimted for more than 95% of the
indi\iduals.
Key words: Nassella pulchra, Melanoplus sanguinipes, Califonuii native grassland, density, diversity, grasshopper
herbivory, Acrididae.
California's native perennial bunchgrass
communities have been reduced to less than
1% of their original range (Heady 1977), with
much of this loss attributable directly to the
development of agricultural and urban areas
(Huenneke 1989). Additionally, most undevel-
oped patches of native grasslands have con-
verted to grasslands dominated by annual
grasses native to the Mediterranean region
(Jackson 1985). Factors leading to the success
of these Mediterranean species are not com-
pletely understood; however, heavy grazing
pressure has been implicated as a major foctor
that favors these more ruderal annual species
(Burcham 1957). In their pristine state, before
the arrival of European settlers, California's
grasslands had light grazing pressure (Wagner
1989). Removal of major anthropogenic distur-
bances such as grazing and fire does not lead
to the recoveiy of native perennial grasslands
(White 1967, Keeley 1981). Most investigators
now agree that the annual grass species should
be considered naturalized, and a return to the
pristine disturbance pattern will not lead to
reestablishment of native grasslands (Headv
1977).
Joern (1989) suggests that through differen-
tial herbivoiy upon the perennial grasses (rela-
tive to annuals), grasshoppers may have con-
tributed to the establishment of exotic annual
grasses in California's native grasslands. Grass-
hopper herbivory is presumed to be greatest
in simimer months when annual grasses already
have set their seed and prior to germination in
the fall (Joern 1989). Therefore, only perennial
grasses and summer forbs are susceptible to
damage by grasshopper herbivoiy. Fintheniiore,
many grasshopper species exhibit preference
for perennial grasses in the field (Capinera and
Sechrist 1982). Joern (1989) suggests that this
phenology-based, selective damage could re-
duce the competitive abilit\' of native perennial
grasses against naturalized annuals.
There are few data available to support or
refute Joern's (1989) hypothesis beyond basic
surveys of grasshoppers throughout the state
(Strohecker et al. 1968). No population or
communit\ -level studies are axailable for Cali-
fornia s grasshoppers in California natixe peren-
nial grasslands (e.g., population densit>', species
abundance, and biomass estimations). The ob-
jective of this study was to describe the grass-
hopper communitx' found in a representatixe
remnant stand of nati\e perennial grassland
over a period of 2 seasons. These data will
provide information necessary to understand
the role of grasshoppers in California's grass-
lands and shoidd lead to more informed deci-
sions for grassland conservation managers.
'Department of Entomology, University of Clalifornia at Riverside. Ri\erside, C;.\ 92.521.
^Please address all correspondence to this author
■■^Departnu'Tit of Biolc)t;\, Occidental C^ollcKc, Los .\ngcles, C:A 90(1.| 1
172
1996]
GllASSllOlTKUS OF A CaLIFOKNIA GRASSLAND
173
Methods ani:> Matkhials
Study Area
The stucK" WHS condiieted in the Santa Rosa
Phitean Ee()l(),uical Reserve (SRPER), located
10 km west of Mnrrieta, C'ahfornia. The site is
acti\el\' managed b\ The Natuie C()nser\'ane>'
for the restoration and preser\'ation of its rare
habitats. Tlie resene covers 2800 ha and con-
tains abont 1200 ha of nati\e perennial grass-
lands amongst oak woodlands, coastal sage
scrnb, and chaparral. Six sites were established
within the perennial grasslands. These sites
were burned in June ot 1992 as a management
practice to retard annual grass establishment.
Grazing has been excluded from all sites since
at least 1990 (R. Wells, SRPER reserve man-
ager, personal communication).
Purple needle grass {Nassella pulchra A.
Hitch.) is the most abundant native grass in
the reserve. Common exotic annual grasses
include slender wild oats {Avena barbata Link)
and red brome {Brotmis laevipes rubens Labill).
Common forbs include annual bursage {AinJ)ro-
sia acanthicarpa Hook.), doveweed [Ereino-
carpus setigenis H.), and filaree {Erodium
cicutarhnn EHer.; Lathrop and Thorne 1985).
Grasshopper Sampling
Six tiansects were arbitrarily placed through-
out the perennial grassland areas representing
maximum topographic and vegetational het-
erogeneity. Each transect measured 200 m
long b)' 20 m wide. Grasshopper density was
determined with twenty 0.25-m^ hoops
(Onsager and Henry 1977, Thompson 1987).
Hoops were placed along each transect at 10-
m intervals. Density was determined monthly
beginning in July 1993. Grasshopper days
(GHD) and forage consumption estimates
were determined following Onsager (1984).
GHD is a measure of total grasshoppers found
per m- for a given year. Forage consumption is
an estimate of the yearly forage consumption
of grasshoppers based on estimated daily con-
sumption (0.65 times body weight) and GHD.
Biomass-days were calculated using the same
formula for GHD replacing grasshoppers/m-
with g/m-. When possible (density >0.5 grass-
hoppers/m-), 100 individual grasshoppers
were collected from each site and frozen
immediately. These collections were taken
directly following density counts and were
made in August and October 1993; June, JuK,
and August 1994. Grasshoppers were identi-
fied to species and weighed to the nearest mg.
Identifications provided the proportion of
adults (p.,) in each sample. Gixen total deiisit)'
(d), adult densit\ (dj was calculated with the
following formula:
da = (d) * (Pa).
Species diversity was measured using the
Shannon-Weiner index (Pielou 1977). Feeding
category designations follow Capinera and
Sechrist (1982) and Otte (1981; graminixorons,
forbivorous, or mi.xed). Identification of nymphid
stages is difficult, and damaging feeding does
not occur until the 4th instar (Onsager 1984).
Therefore, where possible, adult grasshopper
data are analyzed separately from total grass-
hopper data.
Results
Species of grasshoppers collected arc listed
with subfamily and known feeding prefer-
ences (Table 1). Average grasshopper density
for the 1994 season (June-August) was 2.30
grasshoppers/m^ (Table 1). Density measure-
ment began too late in the season to estimate
an average for 1993. A total 198 GHD were
determined for 1994, leading to an estimated
140 kg/ha of forage consumed. Density esti-
mates of zero grasshoppers/m^ were found from
November 1993 through May 1994. Densit>'
peaked in June for 1994 at 2.9 grasshoppers/
m^. This peak in density was dominated by
immature stages (Fig. lA). Density measures
were higher in 1993 than in 1994 for all paired
sample dates in July, August, and October (f =
4.69, df = 1, 20; F = 0.041). Biomass peaked
in July when most grasshoppers were in the
adult stage (Fig. IB). Biomass days for 1994
totaled 13.2 g-d/m^ (Table 1). Peak biomass
(August) was higher in 1993 than in 1994 (t =
2.43; F = 0.036).
The Shannon-Weiner diversity index, includ-
ing adults and nymphs combined, averaged
0.140 over the 5 collection dates (Fig. 2A). The
peak in adult diversity, in June 1994, repre-
sents only 6 individuals of 2 species. Combined
adult and nvniph species richness averaged
3.4 for the sampled dates. Highest species
richness was found in August. In total, 5 grass-
hopper species were found in these sites for
the collection dates.
174
Great Basin Naturalist
[Volume 56
Table 1. Species of grasshoppers collected in 1993-94 on the SRPER with known feeding t\pes (Capinera and
Sechrist 1982, Otte 1981) and appearance, % composition, GHD, hioniass-days, average density; and weight as calcn-
lated for the sampling period.
GHD^
A\erage
A\erage
Sub-
Feeding
%
(grasshopper
Bioniass-
densitv
adult
Species
famih''
type''
Appearance
composition
days/m-)
days^'
(#/m2)^-
weight (nig)
Cammila peUucida
Sciidder
O
G
Mav-Aug
0.2
0.36
0.076
<().01
72
Mchiiioplits aridiis
Scudder
M
n/a
Jun-Aug
0.3
0.50
0.031
<0.01
90
Melanopltis sansitiiiipes
Fabriciis
M
M
Mav-Oct
97.5
193.47
12.754
2.24
132
Menncria bivittata
Ser\ille
G
G
Mav-Oct
1.6
3.18
0.375
0.03
81
Psolessa fcxcina Scudder
G
G
Jun-Oct
0.5
0.89
0.121
0.01
61
Total.
198
13.240
2.30
131
''O = Oedipodiiiae, .\l = .Melanoplinae, G = Goniphocerinat-
"G = grass, M = nii.xed, N/A = not available
^GHD, biomass-da\'s. and average densit\' for 1994 season onl\
Discussion
Grasshopper populations in the SRPER
appear in late May or early June. Grasshopper
biomass peaks, and the most severe herbivoiy
occurs, in July and August. By this time,
annual grasses have already died and their
seeds are buried and protected from above-
ground herbivory (Savelle and Heady 1970).
Grasshopper densities decrease dramatically
after August, and few are present by October.
Annual grasses are triggered to germinate
after the first fall rains (Heady 1958). By the
time these rains arrive, grasshopper densities
are near 0; therefore, both the mature annual
grasses and dieir seedlings escape serious grass-
hopper herbivory.
One species, Melanopliis sanguinipes,
accounted for over 95% of grasshoppers found
in SRPER (Table 1). This species commonly
damages crops and rangelands throughout
North America (Hewitt 1977). Intense out-
breaks are common and can remove up to 92%
of aboveground vegetation (Nerney 1966, Hil-
bert and Logan 1981). Melanopliis sanguinipes
is classified as a mixed feeder and may prefer
grasses or forbs depending upon the area sam-
pled. If M. sanguinipes feeds extensively on
grasses within the SRPER, given its phenol-
ogy, it will damage perennial grasses more
than annual grasses. Overall forage consinnp-
tion was 140 kg/ha in 1994, which is an eco-
nomically damaging level according to Onsager
(1984). However, using M. sanguinipes in a
shortgrass prairie communit\, Quinn et al.
(1993) found significant reductions in grass
biomass only at grasshopper densities equiva-
lent to 845 GHD or greater. We found only
198 GHD in 1994, suggesting that densities
during these years may not greatly affect grass-
land plant community dynamics according to
Quinn (1993).
The climate of Galifornia s grasslands makes
comparison with other North American stud-
ies difficult. Most comparable studies examine
tallgrass, mixed-grass, and shortgrass prairies
east of the Sierra Nevada. The dn; hot summers
that characterize Galifornia s Mediterranean
climate severeK' limit growth (Risser et al.
1981). Regrowth following summer herbivory
is similarly limited. Glearly, only perennial
grasses are susceptible to herbi\'ory at this
time, and the actual effect of such herbivoiy in
CalifoiTiia grasslands is undocumented. There-
fore, specific studies on the effect of grass-
hopper herbivorx on nati\e Galifornia grass-
lands must be conducted to predict the level
and t>'pe of infestation, if an\, that ma>' fa\or
annual grasses o\'er nati\e perennial species.
Galifornia's grasshopper fauna is rich, com-
pared to other North American regions, with
over 120 species (Joern 1989). About half of
these grasshopper species arc considered
rangeland species. Southern Galifornia has the
richest grasshopper fauna of any region in Gal-
ifornia (Strohecker 1968). Still, richness for
the SRPER \\'as vcr\' low^ compared to other
studies of North American grasslands. Few
studies report dixersit)' indexes, but richness
has been measured for other natixe North
American grasslands. Richness is nearlv^ alwavs
1996]
GaVSSHOFPERS OF A CALIFORNIA GlUSSLAND
175
a Nymphs
■ Adults
■ Adults
D Adults and Nymphs
I '-' I ■ I ■ 1 1
Fig. 1. (A) Grasshopper density for all sampling dates
and (B) biomass for dates collected (a = proportion of
adults vs. nymphs unknown for July 1993).
higher than the vahie of 5 species determined
here (e.g., Joern 1982, Evans 1988). We feel it
is hkely that the grasshopper community is
particularly species depauperate due to isola-
tion of the SRPER grasslands (MacArthur and
Wilson 1967). Frequent burning of the grass-
lands may also help explain the low diversity
found in the SRPER. On the other hand,
burned sites contain more even species com-
positions than unburned grasslands and con-
tain species not found in unbunied sites (Porter
1995). Therefore, we feel that to presence the
diversity of not only grasshoppers but presum-
ably many arthropods, birds, plants, and other
taxa, it may be necessary to preserve larger
tracts of native grasslands. Furthermore, the
effects of grasshopper herbivoiy in these grass-
lands must be accounted for in a well-rounded
consenation effort.
Acknowledgments
We thank James Bethke, Brian Cabrera,
Miriam Cooperband, Kim Hammond, Mari-
anne van Laarhoven, Carl Matthies, and James
Nichols for participating in the grasshopper
collections. Helpful reviews by James Bethke,
Fig. 2. Diversity (A) and species richness (B) for all
dates in wliich collections were made.
Timothy Paine, Anthony Joern, and an anony-
mous reviewer are also greatly appreciated.
We also thank Robin Wells and Cedra Shapiro
of The Nature Conservancy for their assis-
tance as well as the use of the reserve. This
work was partially supported by Academic
Senate grants to R. A. Redak.
Literature Cited
BuRCHAM, L. T. 1957. California rangi-laiid. California
Division of Forestry, Sacramento.
Capinera, J. L., AND T. S. Sechrist. 1982. Grasshoppers
(Acrididae) of Colorado: identification, biolo,g\ and
management. Colorado State Universitv- Experiment
Station, Fort Collins. Bulletin 5.S4S.
Evans, E. W. 1988. Community' dynamics of prairie grass-
hoppers subjected to periodic fire: predictable tra-
jectories or random walks in time? Gikos 52: 28.'3-292.
Heady, H. F 1958. Vegetational changes in the annual
grassland t>'pe. Ecology 39: 402—416.
^. 1977. Valley grassland. Pages 491-574 //! M. G.
Barbour and J. Major, editors, Terrestrial vegetation
of California. Wiley-lnterscience, New York.
Hewitt, G. B. 1977. Review of forage losses caused by
rangeland grasshoppers. USDA Miscellaneous Publi-
cation 1348. 24 pp.
HiLBERT, D. VV., AND J. A. LoGAN. 1981. A review of the
population biolog>- of the migratory grasshopper,
176
Great Basin Naturalist
[Volume 56
Melanophis sanguinipes. Colorado State University
Experiment Station, Fort Collins. Bulletin 577S.
HUENNEKE, L. E 1989. Distributions and regional pat-
terns of California grasslands. Pages 1-12 in L. E
Heuenneke and H. Mooney, editors. Grassland stnie-
ture and function. Kluwer Academic Publishers,
Dordrecht.
Jackson, L. E. 1985. Ecological origins of California's Medi-
terranean grasses. Journal of Biogeography 12:
349-361.
JOERN, A. 1982. Distributions, densities and relative abun-
dances of grasshoppers (Orthoptera: Acrididae) in a
sandhills prairie. Prairie Naturalist 14: 37-45.
. 1989. Insect herbivorv' in the transition to Califor-
nia annual grasslands: did grasshoppers deliver the
coup de grass? Pages 117-134 in L. E Huenneke
and H. Mooney, editors, Grassland structure and
function. Kluwer Academic Publishers, Dordrecht.
Keeled; J. E. 1981. Reproductive cycles and fire regimes.
Pages 231-277 in H. A. Mooney et al., editors. Pro-
ceedings of the Conference Eire Regimes and Eco-
system Properties. USDA Eorest Service, General
Technical Report WO-26.
Lathrop, E. W, and R. F. Thorne. 1985. A flora of the
Santa Rosa Plateau. Southern California Botanists,
Special Publication 1.
MacArthur, R. H., and E. O. Wilson. 1967. The theory
of island biogeography. Princeton University Press,
Princeton, NJ.
Nerney, N. J. 1966. Interrelated effects of grasshoppers
and management practices on shortgrass rangeland.
USDA Agricultural Research Sei^vice Special Report
Z192: 1-14.
Onsager, J. A. 1984. A method for estimating economic
injury levels for control of rangeland grasshoppers
with malathion and carbaryl. Journal of Range Man-
agement 37: 200-203.
Onsager, J. A., and J. E. Henry. 1977. A method for esti-
mating the densit\- of rangeland grasshoppers (Orfiop-
tera: Acrididae) in experimental plots. Acrida: 6:
231-237.
Otte, D. 1981. The Nortli American grasshoppers. Volumes
1 and 2. Hai^vard University Press.
PlELOU, E. C. 1977. Mathematical ecology. Wiley, New
York and London.
Porter, E. E. 1995. The grasshoppers of a California
native grassland: a description of the community and
its ecological importance. Unpublished master's the-
sis. University of California, Riverside.
QuiNN, M. A., ET AL. 1993. Effect of grasshopper (Ortliop-
tera: Acrididae) density and plant composition on
growth and destruction of grasses. Environmental
Entomology 22: 993-1002.
RISSER, P G., ET AL. 1981. The true prairie ecosystem.
Hutchinson Ross Publishing Co., Stroudsburg, PA.
557 pp.
Sa\elle, G. D., and H. E Heady. 1970. Mediterranean
annual species: their responses to defoliation. Procla-
mations of the 11th International Grassland Congress
548-551.
Strohecker, H. E, et al. 1968. The grasshoppers of Cal-
ifornia (Orthoptera: Acridoidea). Bulletin of the Cali-
fornia Insect Survey No. 10. Uni\ersity of California
Press.
Thompson, D. C. 1987. Sampling rangeland grasshoppers.
Pages 219-233 in J. L. Capinera, editor. Integrated
pest management on rangelands: a shortgrass prairie
perspective. Westview Press.
Wagner, E H. 1989. Grazers, past and present. Pages
151-162 in L. E Huenneke and H. Mooney, editors.
Grassland stnicture and function. Kluwer Academic
Publishers, Dordrecht.
White, K. L. 1967. Native bunchgrass (Stipa pulchra)
on Hastings Reser\'ation, California. Ecology 48:
949-955.
Received 10 ]ulij 1995
Accepted 19 October 1995
Great Basin Naturalist 56(2), © 1996, pp. 177-179
SUMMER NOCTURNAL ROOST SITES OF BLUE GROUSE
IN NORTHEASTERN OREGON
Kenneth J. Popper', Eric C. Pelrenl-, and John A. Crawford'
Key words: Blue Gnntsf, DL'iidragapiis ohsciinis, nocturnal, Oregon, roost.
Avian habitat studies frequently focus on
diurnal habitat use because of ease of obsei'va-
tion and high le\'els of activity associated with
breeding and foraging. Nocturnal habitat use
may be critical for all birds but has received
far less attention. Thus, there is a need to bet-
ter understand nocturnal habitat use, espe-
cially by crepuscular and diurnal birds, and
factors that may contribute to this use.
Blue Grouse {Dendra^opiis obscunis) are
associated primarily with true fir {Abies spp.)
and Douglas-fir {Pseudotsuga menziesii) forests
in mountainous regions of western North Amer-
ica (Johnsgard 1983). Breeding season habitat
associations often include nonforested and
shrub or steppe regions. These birds are diur-
nal with increased activity in the morning and
evening hours. Pekins et al. (1991) determined
that both diurnal and nocturnal winter roosts
of Blue Grouse were located in conifers. Blue
Grouse shifted from eating conifer needles in
winter to groimd-layer vegetation in summer
and fall in northeastern Oregon (Crawford et
al. 1986). Blue Grouse summer habitat studies
have dealt with diurnal activities (Mussehl
1963, Bendell and Elliot 1966, Zwickel 1975),
but nocturnal obsen^ations are minimal. John-
son (1929) witnessed a brood fly into a tree,
apparently to roost overnight, and Blackford
(1958, 1963) observed > 3 adult males flying
into "roost trees" in spring, where they pre-
sumably stayed overnight. Blackford (1963)
also obser\'ed a male displaying on the ground
approximately 1 h after dark. Zwickel (1992)
suggested that ground roosting may occur,
particularly on breeding ranges where trees
are unavailable or before chicks are able to fly.
In the course of monitoring radio-equipped
Blue Grouse during summer, we identified 20
independent nocturnal roost sites. Our objec-
tive here is to describe these roost sites.
Study Area and Methods
The study area is located in northeastern
Oregon, 30 km north of Enteiprise in the Wal-
lowa-Whitman National Forest in Wallowa
County. Elevation ranges from 900 to 1500 m,
with ridge slopes as great as 35°. North-facing
slopes are dominated by stands of Douglas-fir
and ponderosa pine {Pinus pondero.sa), and
common shrubs are mallow ninebark [Phy.so-
carpiis malvaceiis), snowberry {Symphoricar-
pos albus), and big huckleberry {Vaccinium
membranacenm). Bunchgrass meadows, pre-
dominantly bluebunch wheatgrass {Agropyron
spicatiim) and Idaho fescue {Festucu idahoen-
sis), occur on south-facing slopes. Cattle graze
parts of the area during summer months,
resulting in variable grass cover.
Grouse were captured in walk-in traps and
fitted with poncho- or necklace-moimted radio
transmitters, 15 to 18 g (Advanced Telemetr>
Systems, Inc., Isanti, MN, and Telemetiy Sys-
tems, Inc., Mequon, WI), from June through
August 1993. Radio-equipped juvenile birds
were > 500 g, capable of flight, and > 1 mon of
age. Each radio-equipped bird was located at
night once between 5 July and 3 August 1993.
In addition to radio telemetry, a spotlight was
used to verify- the location of the bird. The exact
roost site was identified by the presence of fi-esh
fecal droppings. When 2 or more grouse were
observed roosting together (<10 m apart) only
1 roost site was counted for use in analyses to
ensure independence of locations.
1 Department of Fisheries ujui Wildlife, Nasli Hall 104, Oregon State University, Corvallis, OR 97331.
"Address all correspondence to Eric C. Pelren.
177
178
Great Basin Naturalist
[Volume 56
Results and Discussion
Twenty-five radio-equipped Blue Grouse
and 38 birds without radios were located at 20
independent nocturnal roost sites (Table 1).
The radio-equipped birds consisted of 12
adults and 13 juveniles; sexes and ages of the
other birds were unknown. All roost sites were
on the ground. Males usually roosted alone,
whereas hens and juveniles frequently roosted
together. Sixteen of 20 independent roosts,
including birds of all sex and age groups, were
in grass of a relatively consistent height; the
others were in forbs {n = 2) and shrubs {n =
2). Twenty-three of 25 radio-equipped birds
were within 50 m of potentially useful roost
trees. An adult female and a juvenile female
roosted 75 and 100 m from trees, respectively,
both easy flight distances for grouse. Adult
males usually roosted closer to trees than
other birds.
During daytime, radio-equipped birds
were seldom located in trees (<1% of 614
obsei-vations, July-August 1991 through 1993;
E. Pelren unpublished data). However, almost
all birds flushed during the day landed in
trees, and conifer needles were found in crops
of birds taken from the study area in August
and September 1981 and 1982 (Crawford et al.
1986). Crawford et al. also found plants such
as prickly lettuce [Lactuca serriola), yellow
salsify [Tragopogon diibius), wild buckwheat
(Eriogoniun spp.), and snowbeny {Syniphori-
carpos albiis), as well as short-homed grasshop-
pers {Acrididae) in at least 30% of 145 Blue
Grouse crops in this area. Douglas-fir needles
were found in only 16% of the crops. This
greater use of ground-cover forage and inver-
tebrates corresponded with observed diurnal
and nocturnal use of ground habitat by Blue
Grouse in summer. Blackford (1963) suggested
that selection of roosting sites may result from
foliage preference and feeding habits. Motion
sensors on grouse transmitters indicated that
some birds continued foraging on moonlit
nights, which implied that benefits of feeding
outweighed energy loss associated with move-
ment or increased risk of predation.
Pekins et al. (1991) suggested Blue Grouse
selection of conifers as roosts in winter may be
based primarily on thermal properties of the
sites. Higher temperatures during summer
make thermal considerations less relevant to
survival than during winter. The lowest tem-
perature we noted at a nocturnal roost site was
4°C, well above the lower critical temperature
of-10°C to -15 °C (Pekins 1988).
Hines (1986) found that 96% of juvenile
and adult Blue Grouse mortalities were the
result of predation. In winter. Blue Grouse in
trees may be less conspicuous or available to
predators than those on the ground (Bergerud
and Gratson 1988), and Pekins (1988) obsewed
snow roosting only occasionally, after heavy
snowstorms. However, lack of snow and in-
creased presence of grasses, forbs, and shrubs
in summer, along with cryptic coloration of
Blue Grouse, provide ground-layer camou-
flage superior to that available in winter. Food
availabilit)' ma\' outweigh any increased risk of
predation and account for use of nocturnal
ground roosts by Blue Grouse in summer where
selection of ground roosts occurs.
Table 1. Characteristics of 20 Blue Grouse nocturnal roost sites, northeastern Oregon, JuK-August 1993.
Adult male
Adult female
[uxenile male
|u\enile female
No. of roost sites
6
6
3(8a)
5
No. of other birds
1
16b
7
9^
Plant cover at roost
Grass
4
6
3(8*)
3
Forb
1
0
0
1
Shrub
1
0
0
1
Plant height (m) at roost
Median
0.50
0.45
0.50"
0.75
Range
0.25-1.20
0.25-1.00
0.30-0.75^'
0.30-1. .30
Distance (m) to potential
roost tree
Median
4.5
37.5
50.0^'
20.0
Range
1.0-40.0
15.0-75.0
3.0-75.0^'
5.0-100.0
•'Includes data for 5 radio-eiiiiippcd juvenile males that were with radio-equipped adult or ju\ enile fe
"Does not include 2 radio-ecinipjied jusenile males that were with radio-e(]nipped adult lemales-
"•Does not include.'} radio-equi|)iic-d juM-nile m.ilcs ih.it were \mIIi radio-cciuipped jumiuIc lemales.
1996]
Notes
179
Acknowledgments
This research was conducted as part of a
Bkie Grouse winter ecology study funded by
the U.S. Forest Service and Oregon Depart-
ment of Fish and Wildlife. We thank R. L.
Jarvis for assistance during the development
of this paper. This is Oregon Agricultural
E.xperiment Station Technical Paper 10,673.
Literature Cited
Bendell, J. F, AND P VV. Elliot. 1966. Habitat selection
in Blue Grouse. Condor 68; 431-466.
Bergerud, a. T, .\.\d M. W. Gratson. 1988. Sui-vival and
breeding strategies of grouse. Pages 47.3-.57.5 /';( A. T.
Bergeiiid and M. W. Gratson, editors. Adaptive strate-
gies and population ecology of Northern Grouse.
Uni\'ersity of Minnesota Press, Minneapolis.
Bl.\ckford, J. L. 19.58. Territoriality and breeding beliav-
ior of a population of Blue Grouse in Montana. Con-
dor 60; 145-1.58.
. 1963. Fiu^ther obsei"vations on the breeding be-
ha\ior of a Blue Grouse population in Montana.
Condor 65; 485-513.
Cr.\\vford, J. A., W. V. Dyke, S. M. Meyers, and T. E
Haensly. 1986. Ftill diet of Blue Grouse in Oregon.
Great Basin Naturalist 46: 123-127.
HiNES, J. E. 1986. Recruitment of young in a declining
population of Blue Grouse. Unpublished dissertation,
University of Alberta, Edmonton. 256 pp.
JoHNSCARD, i^ A. 1983. The grouse of the world. Univer-
sity of Nebraska Press, Lincoln. 413 pp.
Johnson, R. A. 1929. Sununer notes on the Sooty Grouse
of Mount Rainier Auk 46; 291-293.
MUSSEHL, T. W. 1963. Blue Grouse brood cover selection
and land-u.se implications. Journal oi' Wildlife Man-
agement 27; .547-555.
Pkkins, P J. 1988. Winter ecological energetics of Blue
Grouse. Unpublishi'd dissertation, Utah State Univer-
sity, Logan. 141 pp.
Pekins, P J., E G. LiNDZEY, AND J. A. Gessa.man. 1991.
Physical characteristics of Blue Grouse winter use-
trees and roost sites. Great Basin Naturalist 51;
244-248.
ZwiCKEL, E C. 1975. Nesting i)arameters ol Blue Grouse
and their relexance to poiiiilatious. (londor 77:
423-430.
. 1992. Blue Grouse. In: A. Poole, P Stetteuheim,
and E Gill, editors. The birds of Nordi .'\merica. No.
15. The Academ>' of Natural Sciences, Philadelphia.
The American Ornithologists Union, Washington,
DC. 28 pp.
Received 4 Augmt 1995
Accepted 23 October 1995
Great Basin Naturalist 56(2), © 1996, pp. 180-182
OOCHORISTICA SCELOPORI (CESTODA: LINSTOWIIDAE) IN
A GRASSLAND POPULATION OF THE BUNCH GRASS LIZARD,
SCELOPORUS SCALARIS (PHRYNOSOMATIDAE), FROM ARIZONA
Stephen R. Goldberg^, Charles R. Bursey^, Chris T. McAllister3,
Hobart M. Smith^, and Quynh A. Truong^
Key words: Sceloporus scalaris, bunch gross lizard, Phnjnosomatklae, Oochoristica scelopori, Cestoda, Arizona.
The bunch grass Hzard {Sceloporus scalaris
Wiegmann, 1828) is known from tlie Huachuca,
Dragoon, Santa Rita, and Chiricahua moun-
tains of Arizona, the Animas Mountains of New
Mexico, and in the Sierra Mache Occidental
and Sierra del Nido of Mexico, usually above
1830 m, but a few isolated valley populations
occur as low as 1200 m (Stebbins 1985). To our
knowledge, the only report of helminths of this
species was a study of a high-elevation (2438-
2560 m) Chiricahua Mountain population of
Sceloporus scalaris slevini by Goldberg and
Bursey (1992a). The puipose of our note is to
report on a helminthological examination of a
low-elevation (ca 1524 m) grassland popula-
tion of S. scalaris slevini Smith, 1937 from Ari-
zona, and to compare our findings with those
of Goldberg and Bursey (1992a).
We examined 51 S. scalaris slevini (mean
snout-vent length 51 ± 3.4 mm [s], range
40-55 mm) collected (mostly b\' hand, a few
by dust shot) on the Sonoita Plain, elevation ca
1524 m (3r39'N, lir32'W), in the vicinity of
Elgin, Santa Cruz County, Arizona. Specimens
were deposited in the University of Colorado,
Museum of Natural Historv, Boulder, Colorado
as UCM 57259-57282; 57284-57286; 57289-
57292; 57295-57298; 57300-57305; 57307-57310;
57313-57316; 57318-57319. UCM 57318-57319
were collected 20 August 1989; others were
collected 12-19 July 1990.
The abdomen was opened, and the esopha-
gus, stomach, and small and large intestines
were removed from the carcass. Each organ
was slit longitudinally and examined under a
dissecting microscope. The liver and body
cavitv were also examined. Each helminth was
identified using a glycerol wet mount. Repre-
sentative cestodes were stained with hema-
toxylin and mounted in balsam for further ex-
amination. Voucher specimens were deposited
in the U.S. National Parasite Collection,
Beltsville, Maryland 20705 (USNPC 85053).
Terminology' use is in accordance with Margo-
hsetal. (1982).
Only 1 helminth was found, the cestode
Oochoristica scelopori Voge and Fox 1950.
Prevalence of infection was 10% (5 of 51);
mean intensity = 1.2 ± 0.45 [.s], range 1-2.
In the only other investigation of helminths
of S. scalaris, Goldberg and Bursey (1992a)
reported finding tetrathyridia of the cestode
Mesocestoides sp. (prevalence 8%) and lan'ae
of the nematode Physaloptera sp. (prevalence
3%). That study was done on a coniferous for-
est high-elevation population (approximately
2500 m) in the Chiricahua Mountains, whereas
the current study considered a low-elevation
population (ca 1524 m) on the Sonoita Plain,
located ca 126 km SE of the Chiricahua Moun-
tains study site. Although both populations
harbored mutually exclusive helminth faunas,
additional work on larger S. scalaris samples
from these sites will be required to determine
the constancy of these differences.
Oochoristica scelopori is a common cestode
of North American lizards and has been found
in 14 other North American phrxnosomatid
lizards (Table 1). In addition, Anuein (1951)
and Telford (1964) reported finding O. scelo-
pori in the xantusiids, Xantusia henshawi, X.
riversiana, and X. vigilis. Measurements of
various structiues of these cestodes were strik-
ingly different from the measurements as given
'Department of Biology; Whittier College, Whittier, CA 90608. Address correspondence to this autlior
^Department of Biology, Pennsylvania State University, Slienango VUlley Campus. Sharon, PA 16146.
•'Department of Biology, Te.\as VVesleyan University, 1201 Wesleyan, Fori Worth, T.\ 76105-1536.
-•EPG Biology, University of Colorado, Boulder, CO 80,309-0334.
180
1996]
Notes
181
Table 1. Definiti\t^ hosts oi'OocIioristicd scclopori in North America.
Host
L<)L'aht\
Prevalence
Keierence
Crotaph ijtus colhiris
Cahlornia
Gamhelki wislizenii
California
Sceloponis clarkii
Arizona
S. graciosus
California
California
Idaho
Idalio
Utah
S. jarrovii
Arizona
Arizona
Arizona
S. imigisfer
Arizona
Texas
S. occidentalis
California
California
Idaho
Oregon
Utah
S. olicaccus
Texas
S. orcutti
California
S. poinsettii
Texas
S. scalaris
Arizona
S. iindiihitiis
Arizona
Uma inornata
California
U. nofata
California
Urosauriis graciosus
California
100% (1/1)
40% (2/5)
5% (1/20)
not given
10% (7/71)
22% (2/9)
1% (1/118)
5% (1/22)
10% (47/489)
3% (1/31)
5% (15/302)
(?/3)
6% (1/17)
20% (13/65)
23% (27/116)
11% (2/19)
33% (20/60)
9% (1/11)
3% (2/61)
22% (16/74)
30% (3/10)
10% (5/51)
6% (3/48)
7% (1/15)
42% (10/24)
6% (2/34)
Telford 1970
Telford 1970
Goldberg etal. 1994
Voge and l-bx 1950
Telford 1970
VVaitz 1961
Lyon 1986
Pearce and Tanner 1973
Goldberg and Bnrsey 1990
Goldberg and Bursey 1992b
Goldberg et al. 1995a
Walker and Mathias 1973
Goldberg et al. 1995b
Voge and Fox 1950
Telford 1970
Lyon 1986
White and Knapp 1979
Pearce and Tanner 1973
Goldberg et al. 1995b
Goldberg and Bursey 1991
Goldberg et al. 1993
this paper
Goldberg et al. 1994
Telford 1970
Telford 1970
Telford 1970
in the original description of O. scelopori by
Voge and Fox (1950). Amrein (1951) reported
the average length of 25 mature cestodes from
X. henshawi and X. vigilis to be 15.82 mm; the
cestodes from X. riversiana measured 33—37
mm. Telford (1964) indicated his cestode spec-
imens from .xantusiid lizards were less than 45
mm. Both Amrein and Telford identified these
cestodes as O. scelopori. Bursey and Goldberg
(1992) found Amrein's measurements of ces-
todes from X. henshawi and X. vigilis to approx-
imate the measurements of O. bezyi, whereas
Telford's measurements of cestodes from X.
riversiana approximated measurements of O.
islandensis and suggested that X. henshawi, X.
riversiana, and X. vigilis be removed from the
host list of O. scelopori, leaving only phrynoso-
matid lizards as hosts for O. scelopori.
Literature Cited
Amrein, Y. U. 1951. The intestinal entozoa of the night
lizards of California and their mode of transmission.
Unpublished doctoral dissertation, University of Cali-
fornia, Los Angeles. 162 pp.
Bursey, C. R., and S. R. Goldberg. 1992. Oochoristica
islandensis n. sp. (Cestoda: Linstowiidae) from the
island night lizard, Xanfttsia riversiana (Sauria: Xan-
tusiidae). Transactions of the American Microscopi-
cal Societ\' 111: 302-313.
Goldberg, S. R., and C. R. Bursey. 1990. Gastrointestinal
helminths of the Yarrow spiny lizard, Sceloporus jar-
rovii jarrovii Cope. American Midland Naturalist
124: 360-365.
. 1991. Intestinal helminths of the granite spiny
lizard [Sceloporus orcutti). Journal of Wildlife Dis-
eases 27: 355-357.
. 1992a. Helminths of the bunch grass lizard,
Sceloporus scalaris slevini (Iguanidae). Journal of the
Helminthological Societ\' of Washington 59: 130-131.
. 1992b. Prevalence of the nematode Spauligodon
giganticus (Oxyurida: Phaiyngodonidae) in neonatal
Yarrow's spiny lizards, Sceloporus jarrovii (Sauria:
Iguanidae). Journal of Parasitolog\' 78: 539-541.
Goldberg, S. R., C. R. Bursey, and R. L. Bezy. 1995a.
Helminths of isolated montane populations of Yan'ow's
spiny lizard, Sceloporus jarrovii (Phnnosomatidae).
Southwestern Naturalist 40: 330-333.
Goldberg, S. R., C. R. Bursey, .\nd C. T. McAllister.
1995b. Gastrointestinal helminths of nine species of
Sceloporus lizards (Phr>nosomatidae) from Texas.
Journal of the Helminthological Society of Washing-
ton 62: 188-196.
Goldberg, S. R., C. R. Bursey, and R. Tawtl. 1993. Gas-
trointestinal helminths of the crevice spiny lizard,
Sceloporus poinsettii (Phr\nosomatidae). Journal of
the Helminthological Society of Wishington 60:
263-265.
. 1994. Gastrointestinal helminths of Scel(>i)orus
lizards from Arizona. Journal of the Helminthologi-
cal Society of Washington 61: 73-83.
Lyon, R. E. 1986. Helminth parasites of six lizard species
from southern Idaho. Proceedings of the Helmintho-
logical Societ>- of Washington 53: 291-293.
182
Great Basin Naturalist
[Volume 56
Margolis, L., G. W. Esch, J. C. Holmes, A. M. Kuris, and
G. A. SCHAD. 1982. The use of ecological terms in
parasitology (report of an ad hoc committee of the
American Society' of Parasitologists). Journal of Para-
sitology 68: 131-133.
Pearce, R. C., and W. W. Tanner. 1973. Helminths of
Sceloporus lizards in the Great Basin and Upper
Colorado Plateau of Utah. Great Basin Naturalist 33:
1-18.
Stebbins, R. C. 1985. A field guide to western reptiles
and amphibians. Houghton Mifflin Companx', Boston.
336 pp.
Telford, S. R. 1964. A comparative study of endopara-
sitism among some southern California lizard popu-
lations. Unpublished doctoral dissertation. Univer-
sity of California, Los Angeles. 260 pp.
. 1970. A comparative study of endoparasitism
among some southern California lizard populations.
American Midland Naturalist 83: 516-554.
VoGE, M., and W. Fox. 1950. A new anoplocephalid ces-
tode, Oochoristica scelopori n. sp., from the Pacific
fence lizard, Sceloporus occidentalis occidentalis.
Transactions of the American Microscopical Society
69: 236-242.
VVaitz, J. A. 1961. Parasites of Idaho reptiles. Journal of
Parasitology 47: 51.
Walker, K. A., and D. V. Matthlas. 1973. Helminths of
some northern Arizona lizards. Proceedings of the
Helmin thole )gical Society- of Washington 40; 168-169.
White II, R. L., xsd S. E. Knapp. 1979. Helminth para-
sites of sceloporine (Iguanidae) lizards from central
Oregon. Proceedings of the Helminthological Soci-
ety of Wiishington 46: 270-272.
Recewed 25 July 1995
Accepted 31 October 1995
Great Basin Naturalist 56(2), © 1996, pp. 1S3-185
POCKET GOPHERS DAiMAGE SALTCEDAR
{TAMARIX RAMOSISSIMA) ROOTS
Sara j. Maniiiiiu', iiriaii L. Casliorc', and Joseph M. Szewczak^
Kcij uord.s: sdltcidar. Tamari.x VMwosissitmi. pockti ^(>i)lier. Thoiiioiins hottac, tuiiiari.sk. Owens Valley, imusivc i)lanl,
exotic plant.
Saltcedar {Tamarix ramosissima Ledeb.,
Tamaricaceae) is an invasive, exotic woody
slirul) natixe to Asia (Bauni 1978, Hickman
1993) that has colonized extensive areas
tliroiighout the western United States (Robin-
son 1965, Brotherson and Winkel 1986).
Saltcedar possesses many characteristics that
render it a nuisance plant (Brotherson and
Winkel 1986), and because it has been viewed
as a threat to native vegetation communities,
researchers have examined its ecology (Car-
man and Brotherson 1982, Brotherson and
Winkel 1986, Shafroth et al. 1995), water con-
sumption (Robinson 1958, van Hylckama 1970,
Davenport et al. 1982, Bureau of Reclamation
1992), and cost of control efforts (Brotherson
and Field 1987, Neill 1990, Barrows 1993). It
is known to inhibit flows in creeks and springs
(Robinson 1965, Rowlands 1990); thus, its
spread has been detrimental not only to native
vegetation but also to native wetland and
aquatic fauna (Neill 1983).
Altliougli efforts are under wa\' in the United
States to develop biocontrol agents using
insects that occur on saltcedar in its native
range (DeLoach 1990), to date there have
been no reports of native herbivores, insects,
or diseases causing saltcedar mortality. Herein
we report the first known mortality caused by
native mammals on saltcedar.
Our discovery occurred in Owens Valley,
California. Water has been exported from
Owens Valley — located in the rain shadow
created by the Sierra Nevada range directly to
its west — since 1913. Alteration of natural
water flows created conditions favorable to the
spread of saltcedar (Cashore 1985, Babb
1987).
During the winter of 1995, when foliage
was absent from saltcedar, we obsei-ved that a
few plants within a young, even-aged stand
were dead. Some of the plants were leaning
over, supported by neighboring plants. Upon
inspection, we obsened that dead plant tap-
roots had been gnawed apart approximateK' 10
cm beneath the soil surface. Teeth marks were
clearly visible on the tapered stumps. In addi-
tion, prolific gopher tunneling was exident
within and around the saltcedar stand, and
excavated dirt mounds were located near the
dead saltcedar. Examination of growth rings of
plants within the stand showed the saltcedar
plants to be 7 years old in 1995.
In early April 1995, when saltcedar w as just
beginning to break bud, we revisited the site
to quantify the extent of animal damage and to
capture and identify the species tunneling at
the site. We examined plants by working from
one end of the stand toward the center Ever>'
saltcedar plant in approximately 1/2 of the
stand was sampled, for a total of 545 plants.
Height was measured, and then plants were
tugged to detect the degree of below-ground
damage. If tugged plants freely exited the soil
and had no attached live roots, the damage
was scored as fatal. All of these plants appeared
dead, no resprouting was evident, and each
had a chewed taproot stump, the diameter of
which was measured and recorded. If tugged
plants could be pulled from the ground easily,
but still had live laterals above the chewed
taproot, they were noted as sustaining severe
damage. In these instances, diameter of the
largest chewed root was measured. Tpically,
these plants had many dead, but a few li\ ing,
branches. If tugged plants felt loose, but could
'liiNci County Water Department, 163 May Street, Bishop CA 93514.
-Deep Springs College, Dyer NV 89(110.
183
184
Great Basin Naturalist
[Volume 56
not be easily pulled from the soil, they were
scored as sustaining minimum damage. If
tugged plants were tightly rooted in the soil,
we assumed no root damage. The majority of
branches on plants in both these categories
appeared alive.
Results of gopher damage are listed in
Table 1. Nearly 23% of the plants sampled had
experienced some degree of gopher damage;
of these, 7.0% were dead as a result of gophers,
5.3% had been severely affected, and 10.6%
had been minimally affected. The diameter of
gopher-chewed roots ranged from 11 mm to
55 mm and averaged 27.7 mm.
Gopher damage appeared to affect plant
height; analysis of variance revealed significant
height differences between plants in the 4
categories of damage (F = 4.463, P = 0.004,
df = 3). However, saltcedar plants not dam-
aged by gophers tended to be only slightly
taller than plants sustaining gopher damage
(Table 1), suggesting that gopher damage had
been relatively recent.
The study area was searched for evidence
of active gopher mounds. Early in the evening,
7 active mounds were excavated, and Sherman
live-traps baited with seeds and fresh plant
material were placed at the tunnel level. These
traps were then covered with soil, using local
materials to prevent cave-ins at the trap en-
trance. Trapping was done under the provision
of a scientific collector's permit issued by the
California Department of Fish and Game. Traps
were checked the following morning shortly
after sunrise.
From tlie 7 traps set in active gopher tunnels,
1 valley pocket gopher {Thomoniys hottae)
(Ingles 1965) was captured. Two other traps
were found packed with soil, presumably by
gophers. The 4 remaining traps showed no
obvious sign of gopher activity.
These data are the first reported evidence
of a native species, Thomotnijs hottae, inducing
mortalit}^ in the exotic Tamarix ramosissima.
The proximity of a saltcedar stand to gopher
habitat may increase its susceptibility to gopher
damage. At our site, gopher mounds appeared
more extensive in the alkali meadow immedi-
ately adjacent to the saltcedar stand than in
the stand itself We subsequentK' made obser-
vations at other even-aged stands of saltcedar
that occur adjacent to alkali meadows at other
locations in Owens Valley and in Deep Springs
Table 1. E.xtent of gopher damage within a stand of
saltcedar plants in the Owens Valley.
Gopher
damage
# plants
% of Avg. ht.
total (cm)
None
420
77.0
128.9
33.3
Minimum
58
10.6
120.3
30.5
Severe
29
5.3
116.2
32.6
Fatal
38
7.0
113.4
25.3
All totii
545
100.0
126.:
32.8
Valley. Again we found gopher damage, so the
phenomenon is not isolated to this single stand.
In general, the influence of fossorial ani-
mals on plant communities has received rela-
tively little research attention (Andersen 1987).
Although gophers may kill or slow the growth
of saltcedar, their long-term effects on stand
size and vigor or on saltcedar establishment
in the meadow remain unknown. Other re-
searchers have found that pocket gophers
cause significant woody plant mortality in a
variety of plant communities (Crouch 1971,
Marsh and Steele 1992, Cox and Hunt 1994,
Ferguson and Adams 1994), and Huntly and
Inouye (1988) and Cantor and Whitman (1989)
reported that tree encroachment into mead-
ows was significantly slowed when gophers
were present in meadows. However, given the
vigorous growth of saltcedar in general,
gopher damage may merely thin the stand,
allowing the remaining individuals to continue
unabated.
Literature Cited
Andersen, D. C. 1987. Below-ground herbi\or>- in nat-
ural commimities: a review emphasizing fossorial
animals. Quarterly Review of BiologN' 62: 261-286.
Babb, D. E. 1987. Report on the saltcedar control stud>-.
Unpublished report prepared for the Invo/Los Ange-
les Technical Group. In\ o Coimt) Water Department,
Bishop, CA.
Barrows, C. W. 1993. Tamarisk control II: a success stoiy
Restoration and Management Notes 11: 35-38.
Baum, B. R. 1978. The genus Tamarix. Israel Academ\ of
Science and Humanities. 209 pp.
Brotherson, J. D., and D. Field. 1987. Tamarix: impacts
of a successful weed. Rangelands 9: 110-112.
Brotherson, J. D., and V. Winkel. 1986. Habitat rela-
tionships of saltcedar {Tamarix ramosissima) in cen-
tral Utah. Great Basin Naturalist 46: 53,5-541.
Bl real oe RECLANl.vriON. 1992. Vegetation management
study: Lower Colorado Ri\er Phase I report. Lower
Colorado Region, Boulder Cit\-, NV. 103 pp.
Cantor, L. F, and T. G. Wihtnl\n. 1989. Importance of
belowgroimd herbi\'ory: pocket gophers may limit
aspen to rock outcrop refugia. Ecology 70: 962-970.
1996]
Notes
185
Carman, J. G., and J. D. Brotfiehson. 1982. Comparison
of sites infested and not infested with saltcedar
{Tainarix pcntaiidra) and Russian olive (Elea^iuis
angustifolia). Weed Seienee 30: 360-364.
Cashore, B. 1985. Saltcedar in the Owens Valley. Unpub-
lished report, Inyo Coinit\ Water Department,
Bishop, CA. 32 pp.
Cox, G. W, AND J. Hunt. 1994. Pocket gopher herhivory
and mortality of oeotillo on stream terrace, bajada,
and hillside sites in the Colorado Desert, southern
California. Southwestern Naturalist 39; 364-370.
Crouch, G. L. 1971. Susceptibility' of ponderosa, Jeffrey,
and lodgepole pines to pocket gophers. Northwest
Science 45: 252-256.
Davenport, D. C, P E. Martin, and R. M. Hagan. 1982.
Evapotranspiration from riparian vegetation: water
relations and irreco\erable losses for saltcedar Joiu-
nal of Soil and Water Conservation 37: 233-236.
DeLoach, C. J. 1990. Prospects for biological control of
saltcedar (Tamarix spp.) in riparian habitats of the
southwestern United States. Pages 307-314 in E. S.
Delfosse, editor. Proceedings of the 7th Interna-
tional Symposium on Biological Control of Weeds,
6-11 March 1988, Rome, Italy
Ferguson, D. E., and D. L. Adams. 1994. Effects of
pocket gophers, bracken fern, and western cone-
flower on survival and growth of planted conifers.
Northwest Science 68: 241-249.
Hickman, J. C, editor 1993. The Jepson manual: higher
plants of California. University of California Press,
Berkeley 1400 pp.
HUNTLY, N., and R. Inouye. 1988. Pocket gophers in
ecosystems: patterns and mechanisms. Bioscience
38: 786-793.
Ingles, L. G. 1965. Mammals of the Pacific states. Stan-
ford University Press, Stanford, CA. 508 pp.
Marsh, R. E., and R. W Steele. 1992. Pocket gophers.
Pages 205-230 in H. C. Black, editor, Silvicultural
approaches to animal damage management in Pacific
Nortliwest forests. USDA Forest Service, General
'i'echnical Report PNW-GTR-287.
Nkill, W. M. 1983. The tamarisk invasion of desert ripar-
ian areas. Desert Protective Council, Educational
Bulletin 83-4.
• 1990. Control of tamarisk by cut-stump herbicide
treatments. Pages 91-98 in M. R. Kunzmann, R. R.
Johnson and P S. Bennett, editors, Tamarisk Control
in Southwestern United States: Proceedings of the
Tamarisk Conference, 2-3 September 1987. Special
Ik-port 9, University' of Arizona, Cooperative National
Park Resources Studies Unit, Tucson.
Robinson, T. W 1958. Phreatophytes. U.S. Geological Sur-
vey Water Supply Paper 1423. 84 pp.
. 1965. Introduction, spread, and aerial extent of
saltcedar (Taniarix) in the western states. U.S. Geo-
logical Survey Professional Paper 491-A.
Rowlands, R G. 1990. Histoiy and treatment of the
saltcedar problem in Death Valley National Monu-
ment. Pages 46-56 mi M. R. Kunzmann, R. R. John-
son, and P S. Bennett, editors. Tamarisk Control in
Southwestern United States: Proceedings of the
Tamarisk Conference, 2-3 September 1987. Special
Report 9, University of Arizona, Cooperative National
Park Resources Studies Unit, Tucson.
Shafroth, P B., J. M. Friedman, and L. S. Ischinger.
1995. Effects of salinity' on establishment of Popiihis
freinonfii (cottonwood) and Tamarix ramosissima
(saltcedar) in southwestern United States. Great Basin
Naturalist 55: 58-65.
van Hylckama, T. E. a. 1970. Water use by salt cedar
Water Resources Research 6: 728-735.
Received 11 June 1995
Accepted 19 January 1996
Great Basin Naturalist 56(2), © 1996, pp. 186-187
SALTCEDAR {TAMARIX RAMOSISSIMA), AN UNCOMMON HOST FOR
DESERT MISTLETOE {PHORADENDRON CALIFORNICUM)
Sandra L. Haigh^
Key words: Phoradendron californicuni, Taniarix ramosissima, inisflctoc, saltcedai; host, parasite.
The genus Tatnorix (saltcedar) contains
approximately 54 species of phreatophytic
plants whose origins are in Europe, Asia, and
Africa. Several members of the genus were
introduced into the United States in the early
1800s, mainly as ornamental plants. Approxi-
mately 8 species have since escaped cultiva-
tion and have become naturalized to varying
degrees (Baum 1967). Tainarix ramosissima
Ledeb. has become established in ripaiian areas
throughout the West and Southwest, where it
has proven to be an aggressive invader that
eventually displaces native vegetation.
Desert mistletoe {Phoradendron californiciim
Nutt.) is a native parasitic plant that grows on
several species of riparian plant hosts. Its range
includes southern Nevada, southwestern Utah,
southeastern California, southwestern Arizona,
and northern Baja California, Sonora, and
Sinaloa (Benson and Darrow 1981). Previously
published information on hosts for desert
mistletoe include Blumer (1910), Shreve and
Wiggins (1964), Walters (1976), Daniel and
Buttenvick (1992), and Overton (1992), none
of whom mentions T. ramosissima. Holland et
al. (1977) and Benson and Darrow (1981) state
that "saltcedar" and "the introduced tamarisks"
are possible hosts, while Munz and Keck (1965)
and McDougall (1973) list Tamarix but men-
tion no particular species. Cohan et al. (1978)
state that P. californiciim does not occur in
saltcedar This paper describes 2 occurrences
of P. californiciim on T. ramosissima in south-
ern Nevada.
I found the 1st parasite and host specimen
on 27 June 1995 at Hiko Springs in Clark
County, Nevada, approximately 11 km west of
Laugh'lin along State Highway 163 (3,894,000
N 711,650 E) at an elevation of 605 m (Fig. 1).
A 2nd specimen was found on this host tree on
16 October 1995. Voucher specimens from 1
parasite and host are deposited in the Depart-
ment of Biological Sciences herbarium. Uni-
versity of Nevada, Las Vegas, accession num-
ber 38971.
The host tree was growing in a canyon
approximately 2 m from a small, flowing stream
on quartz monzonite-derived soil. The first
mistletoe clump measured 33 cm long X 32 cm
high X 14 cm wide and was growing on the
southwest side of a branch 2.1 m above the
ground. The branch to which the mistletoe
was attached measured 5.2 cm in diameter
and 16.2 cm in circumference. The length of
the branch from trunk to point of mistletoe
attachment was 2.1 m. The trunk base of the
5-m-high saltcedar measured 8 cm in diameter
and 29 cm in circumference, which would
indicate an age of approximately 24 yr (based
on average value of California and Arizona
sites as reported by Smith 1989). The 2nd
mistletoe also faced southwest and was located
on the main trunk of the tree .9 m above the
ground. It was a newly sprouted plant that con-
sisted of only 12 stems, the longest of which
measured 4 cm. Both mistletoes and the host
tree appeared to be healthy, actively growing
specimens. The parasites were young plants
and were a more vivid green than other mis-
tletoes in the area. Sex of the mistletoes could
not be determined.
Other hosts for P. californiciim at this site
include catclaw acacia [Acacia greggii), honey
mesciuite (Prosopis glandidosa), and creosote
bush {Larrea tridentata). Although many other
Tamarix trees occur here, none ha\'e been in-
fected by mistletoe. Desert mistletoe is usually
spread from host to host by birds, which ingest
the seeds and later defecate them onto a branch.
Two bird species that occur frequently at this
'Departnient of BioloKical Sciences. 4505 Manlaml Parkuay, Box 454004. Las Veijas, NV 89154-4004.
186
1996]
Notes
187
Fig. 1. Parasite Fhoradendron calijornicwn growing on
host plant Tainarix rainosissima.
site and have been seen feeding on misdetoe
and perching in saltcedar are die Phainopepla
[Phainopepla nitens) and Northern Mocking-
bird {Miiniis polyglotfos) (personal obsei-vation).
Acknowledgments
I wish to thank Wesley Niles for help with
identification of specimens and review of the
manuscript, and Delbert Wiens who provided
information on mistletoe hosts. This project
was funded by a research grant provided by
the Harry Reid Center for Environmental
Studies, Las Vegas, Nevada.
LiTERAiuHK Cited
Halm, B. 1{. 1967. Introduci'd and iiatiiralizrd tamarisks
in the United States and Canada Clliiuaricaeeae).
Baileya 15: 19-25.
Bi'NSON, L., .AND R. A. Dahhow. 1981. Trees and shnihs of
the sonthvvesteni deserts. University of Arizona Press,
Tucson. 416 pp.
Blumer, J. C. 1910. Mistletoe in tlie Sonthwest. Plant
World 13; 240-246.
Cohan, D. R., B. W. Anderson, .•vnd R. II Ohmaut. 1978.
Avian population responses to salt cedar along the
Lower C:olorado River. Pages 371-382 in R. R. John-
son and J. E McCormick, editors, Strategies for pro-
tection and management of floodplain wetlands and
other riparian ecosystems. USDA Forest Service,
General Technical Report WO-12.
Daniel, T. E, and M. L. Buttervvick. 1992. Flora of .south
mountains of south-central Arizona. Desert Plants
10(3): 99-119.
Holland, J. S., R. K. Grater, and D. H. Huntzinger.
1977. Flowering plants of the Lake Mead region.
Southwest Parks and Monuments Association. Popu-
lar Series No. 23. 49 pp.
McDouGALL, W. B. 1973. Seed plants of northern Ari-
zona. The Museum of Northern Arizona, Flagstaff.
594 pp.
MUNZ, P A., and D. D. Keck. 1965. A California flora.
University' of California Press, Berkeley. 1681 pp.
0\ ERTON, J. M. 1992. Host specialization in desert mistle-
toe Phoradendnm califonucuiii. Bulletin of the Eco-
logical Society of America 73: 293.
Shreve, F, and I. L. Wiggins. 1964. Vegetation and flora
of the Sonoran Desert. Volume I. Stanford Univer-
sity Press, Palo Alto, CA. 840 pp.
Smith, S. D. 1989. The ecology of saltcedar (Tamahx chi-
nensis) in Death Valley National Moniunent and
Lake Mead National Recreation Area: an assessment
of techniques and monitoring for saltcedar control in
the park system. Contribution CPSU/UNLV 041/03,
National Park Service/University of Ne\ada, Las
Vegas. 65 pp.
Walters, J. W. 1976. A guide to misdetoes of Arizona and
New Mexico. USDA Forest Service, Southwestern
Region, Forest Insect and Disease Management. 7
pp.
Received 6 November 1995
Accepted 4 March 1996
Great Basin Naturalist 56(2), © 1996, pp. 188-189
BOOK REVIEW
Wild Plants of the Pueblo Province. Explor-
ing Ancient and Enduring Uses. William
W. Dunniire and Gail D. Tieniey. Foreword
by Gary Paul Nabhan. Museum of New
Mexico'Press, Santa Fe, NM. 1995. 290 pp.
$19.95, softback.
This book immediately appears field wor-
thy and feels good in the hands. And that's
simply judging the book by its cover! Once
opened there is much to praise about this text.
The authors have succeeded in putting together
a wondei-fully interesting and well-wiitten field
guide for the lay person as well as a useful ref-
erence for serious students and professionals
interested in ethnobotany of the Southwest.
Within 9 chapters of text, an illustrated section
involving about 73 plants, and an extensive
chart summarizing plant uses, the reader learns
of the ecology, representative flora, ethnobotany,
and cultural history of the Pueblo Province.
The original intent of the book was to provide
a guide to commonly seen plants of Bandelier
National Monument and the Pajarito Plateau
in central New Mexico, and a discussion of the
plants' prehistoric and recent uses. The authors
have surpassed this goal.
The 9 chapters reveal a cohesive and inter-
esting histoiy of the people, plants, and land
itself Ample information provides the reader
insight as to how these elements interact and
what the consequences of those interactions
have been and continue to be. It is easy not
only to move through the spatial and geo-
graphical regions, but to enjoy a voyage in
time as well and feel as if you were there. Line
drawings, photographs, and maps lend addi-
tional interest to the text. Although there is a
great deal of information given about vegeta-
tive zones, human history, and other topics,
the authors have retained the importance of
plants by referencing particular species wher-
ever appropriate. The chapter on indicator
species is particularly interesting and useful.
This is a subject that few field guides address,
and yet it is so easily applied and can be
obsen'ed in the field when adequate informa-
tion is provided. The main focus of the text is
the center section that includes photographs
and descriptions of 73 plants. Line drawings
accompany each plant treated. The technical
descriptions are somewhat brief, but the illus-
trations provide enough detail that field iden-
tification can be made easily in most cases.
Perhaps one of the most valuable sections is
the annotated plant list included at the end of
the book. In an easy-to-read fonnat, a great deal
of infomiation is concisely summarized for over
300 plants. The chart is subdivided into 7 gen-
eral categories of plant use (i.e., food and bev-
erage, medicine, constiiiction, etc.), with infor-
mation given on how each plant is used by
specific pueblos. The chart is well referenced
and includes original citations for every use.
A brief, yet well-organized analysis of the
changes in plant utilization that occurred with
the Spanish colonization in the Southwest is
provided in chapter 3. The authors take a very
complex histoiy and present it in the context
of plant ecology. It provides an informative
view of the ecological consec|uences of the
collision of cultures. Gontemporary culture,
plant use, and ecological modification are also
included in this text. Two chapters provide
insightful information on current cultural and
ecological issues. Throughout the text, and
reflected in the annotated plant list as well,
the authors have attempted to treat religious
and ceremonial plant uses with appropriate
respect. An added benefit of the book is the
authors' personal association with indi\'iduals
in different Pueblo tribes. Their sense of respect
and honor for these cultures is felt throughout
the book.
Our only complaint relating to this text is
the lack of references citing specific informa-
tion. It is quite difficult to identif\' references
for much of the information included within
the text, with the exception of the chapter dis-
cussing indicator species. A bibliography with
188
1996]
Book Review
189
145 references is included at the end of the
book, but it is difficult to relate these refer-
ences to particular chapters and specific infor-
mation. This omission weakens the usefulness
of the book as a reference lor serious students.
It may be that the authors consciously omitted
citations in an effort to allow the text to flow
more easily, but it is a constant frustration
when one is interested in identifying sources.
A list of suggested reading is included at the
end of each chapter, but no reference is given
to original sources that support specific facts.
In the preface, the authors do mention many
sources that contribute in a general way.
Overall, this book is one that should be
included in a field book box, on the bookcase
as a reference for plants and their uses by cul-
tures of the Southwest, and in a travel file as it
gives suggestions for specific hikes located in
the Pueblo Province. For anyone interested in
plant ecology, taxonomy, ethnobotany, cultural
anthropology, or simply those with a general
love for the Southwest, this book is highly rec-
ommended. It is well written, informative, and
aesthetically delightful.
Renee Van Buren
Kimberly Hamblin Hart
Department of Botany and Range Science
Brigham Young University
Provo, UT 84602
University of Nevada, Reno, Department of Anthropology,
Historic Preservation, Biological Resources Research Center, Divi-
sion OF Continuing Education
The University of Nevada, Reno, offers continuing education training courses
in heritage resources management. Courses are designed for professionals in cul-
tural and natural heritage management positions in the public and private sectors.
The program is conducted in cooperation with the Advisory Council on Historic
Preservation, the Bureau of Land Management, the National Park Service, and the
U.S. Forest Service. The following information is offered on one of the upcoming
courses.
Ecosystem Management
30-31 May 1996
Reno, Nevada
9:00 a.m. -4:00 p.m.
Fee: $250
Registration deadline: 2 May 1996
Instructor: Peter F. Brussard, Ph.D., is chairman of the biology department at
the University of Nevada, Reno, and director of the Nevada Biodiversity Initiative,
housed in the Biological Resources Research Center. Brussard is a founding mem-
ber and past president of the Society of Conservation Biology and recognized as a
leading authority in conservation biology and ecology.
Ecosystem management is only recently beginning to be understood and used.
This course will address the scientific basis for ecosystem management as well as
the steps for managing areas so that biological diversity and ecosystem services
remain conserved while human needs are also met. Ecosystem management focuses
on systems as a whole rather than simply on the parts and involves the public in
setting management goals. It represents a shift from linear comprehensive manage-
ment to adaptive management.
For further information, phone 1-702-784-4046 or fax 1-702-784-4801;
to register, call 1-800-233-8929.
INFORMATION FOR AUTHORS
The Great Basin Naturalist welcomes previously
unpublished manuscripts pertaining to the biologi-
cal natural history of western North America.
Preference will be given to concise manuscripts of
up to 12,000 words. Simple species lists are dis-
couraged.
SUBMIT MANUSCRIPTS to Richard VV. Baumann,
Editor, Great Basin Naturalist, 290 MLBM, PO Box
20200, Brigham Young University, Provo, UT
84602-0200. A cover letter accompanying the man-
uscript must include phone number(s) of the author
submitting the manuscript, and FAX number and
E-mail address when applicable; the letter must
also provide information describing the extent to
which data, text, or illustrations have been used in
other papers or books that are published, in press,
submitted, or soon to be submitted elsewhere.
Authors should adhere to the following guidelines;
manuscripts not so prepared may be returned for
revision.
MANUSCRIPT PREPARATION. In general, the Great
Basin Naturalist follows recommendations in
Scientific Style and Format: The CBE Manual for
Authors, Editors, and Publishers, 6th edition
(Council of Biology Editors, Inc., II South LaSalle
Street, Suite 1400,' Chicago, IL 60603, USA; phone
312-201-0101; FAX 312-201-0214). We do, however,
differ in our ti'eatment of entries in Literature Cited.
Authors may consult Vol. 51, No. 2 of this journal
for specific instructions on format; these instruc-
tions. Guidelines for Manuscripts Submitted
TO THE Great Basin Naturalist, are printed at the
back of the issue. Also, check the most recent issue
of the Great Basin Naturalist for changes.
TYPE AND DOUBLE SPACE all materials, including
literature cited, table headings, and figure legends.
Avoid hyphenated words at the right-hand margins.
Underline words to be printed in italics. Use stan-
dard bond (22x28 cm), leaving 2.5-cm margins on
all sides.
SUBMIT 3 COPIES of the manuscript and the origi-
nal on a 5.25- or 3.5-inch disk utilizing WordPerfect
5.1 or above. Number all pages and assemble each
copy separately: title page, abstract and key words,
text, acknowledgments, literature cited, appendices,
tables, figure legends, figures.
TITLE PAGE includes an infomiative title no longer
than 15 words, names and addresses of authors, a
running head of fewer than 40 letters and spaces,
footnotes to indicate change of address and author
to whom correspondence should be addressed if
other than the first author.
ABSTRACT states the puipose, methods, results,
and conclusions of the research. It is followed by
6-12 key words, listed in order of decreasing
importance, to be used for indexing.
TEXT has centered main headings printed in all
capital letters; second-level headings are centered
in upper- and lowercase letters; third-level head-
ings begin paragraphs.
VOUCHER SPECIMENS. Authors are encouraged to
designate, properly prepare, label, and deposit
high-(iuality voucher specimens and cultures docu-
menting their research in an established permanent
collection, and to cite the repository in publication.
REFERENCES IN THE TEXT are cited by author and
date: e.g., Martin (1989) or (Martin 1989). Multiple
citations should be separated by commas and listed
in chronological order. Use "et al." after name of
first author for citations having more than two
authors.
ACKNOWLEDGMENTS, under a centered main
heading, include special publication numbers when
appropriate.
LITERATURE CITED, also under a centered main
heading, lists references alphabetically in the fol-
lowing formats:
Mack, G. D., and L. D. Flake. 1980. Habitat rela-
tionships of waterfowl broods on South Dakota
stock ponds. Journal of Wildlife Management
44: 695-700.
Sousa, W P 1985. Disturbance and patch dynamics
on rocky intertidal shores. Pages 101-124 in
S. T. A. Pickett and P S. White, eds.. The ecolo-
gy of natural disturbance and patch dynamics.
Academic Press, New York.
Coulson, R. N., and J. A. Witter. 1984. Forest ento-
mology: ecology and management. John Wiley
and Sons, Inc., New York. 669 pp.
TABLES are double spaced on separate sheets and
designed to fit the width of either a single column
or a page. Use lowercase letters to indicate foot-
notes.
PHOTOCOPIES OF FIGURES are submitted initially
with the manuscript; editors may suggest changes.
Lettering on figures should be large enough to
withstand reduction to one- or two-column width.
Originals must be no larger than 22x28 cm.
NOTES. If the manuscript would be more appro-
priate as a short communication or note, follow the
above instructions but do not include an abstract.
A CHARGE of $50 per page is made for articles
published; the rate for individual subscribers will
be $35 per page. However, manuscripts with com-
plex tables and/or numerous half-tones will be
assessed an additional charge. Reprints may be pur-
chased at the time of publication (an order form is
sent with the proofs).
FINAL CHECK:
• Cover letter explaining any duplication of
information and providing phone number(s),
FAX number, and E-mail address
• 3 copies of the manuscript and WordPerfect
diskette
• Conformity with instructions
• Photocopies of illustrations
GREAT BASIN NATURALIST
(ISSN 001 7-361 4)
Vol. 56, No. 2, April 1996
CONTENTS
Articles
Selecting wilderness areas to consei^ve Utah's biological diversity
Diane W. Davidson, William D. Newmark, Jack W. Sites, Jr.,
Dennis K. Shiozawa, Eric A. Rickart, Kimball T. HaqDcr, and Robert B. Keiter 95
Nutrient distribution in Quercus gambelii stands in central Utah
A. R. Tiedemann and W. E Clar\' 1 1 9
Comparsion of two roadside survey procedures for dwarf mistletoes on the Saw-
tooth National Forest, Idaho Robert L. Mathiasen, James T. Hoffman,
John C. Guyon, and Linda L. Wadleigh 1 29
Effects of Douglas-fir foliage age class on western spruce budworm oviposition
choice and lai^val performance Kimberly A. Dodds, Karen M. Clancy,
Kathryn J. Le>'\'a, David Greenberg, and Peter W. Price 1 35
Trypanoplasma atraria sp. n. (Kinetoplastida: Bodonidae) in fishes from the Sevier
River drainage, Utah J. Stephen Cranney and Richard A. Heckmann 1 42
Geographical review of the historical and current status of Ospreys {Pandion
haliaetus) in Utah Glark S. Monson 1 50
Effects of turbidity on feeding rates of Lahontan cutthroat trout {Oncorhynchus
clarki henshawi) and Lahontan redside shiner {Richardsonius egregius) . . .
Gary L. Vinyard and Andy C. Yuan 1 57
Pogonomyrmex owyheei nest site density and size on a minimally impacted site
in central Oregon Peter T. Soule and Paul A. Knapp 1 62
Field measurements of alkalinity from lakes in the Uinta Mountains, Utah,
1956-1991 Dennis D. Austin 1 67
Density, biomass, and diversity of grasshoppers (Orthoptera: Acrididae) in a Cal-
ifornia native grassland Eric E. Porter, Richard A. Redak,
and H. Elizabeth Braker 1 72
Notes
Summer nocturnal roost sites of Blue Grouse in northeastern Oregon
Kenneth J. Popper, Eric C. Pelren, and John A. Crawford 177
Oochoristica scelopori (Cestoda: Linstowiidae) in a grassland population of the
bunch grass lizard, Sceloporus scalaris (Phrynosomatidae), from Arizona . . .
Stephen R. Goldberg, Charles R. Bursey, Chris T. McAllister,
Hobart M. Smith, and Quynh A. Truong 1 80
Pocket gophers damage saltcedar {Tamarix ramosissinia) roots . . . Sara J. Manning,
Brian L. Cashore, and Joseph M. Szewczak 1 83
Saltcedar {Tamarix ramosissima), an uncommon host for desert mistletoe {Phora-
dendron calif ornicum) Sandra L. Haigh 1 86
Book Review
Wild plants of the Pueblo Province. Exploring ancient and enduring uses.
William W. Dunmire ami Gail D. Tienwy Renee Van Buren
and Kimberly Hamblin Hart 1 88
H E
Sep U 5 1996
HARVARD
UNIVERSITY
GREAT BASIN
NATURALIST
VOLUME 56 Ne 3 — JULY 1996
BRIGHAM YOUNG UNIVERSITY
GREAT BASIN NATURALIST
Editor Assistant Editor
Richard W. Baumann Nathan M. Smith
290 MLBM 190 MLBM
PO Box 20200 PO Box 26879
Brigham Young University Brigham Young University
Provo, UT 84602-0200 Provo, UT 84602-6879
801-378-5053 801-378-6688
FAX 801-378-3733 E-mail: NMS@HBLL1.BYU.EDU
Associate Editors
Michael A. Bovvers Paul C. Marsh
Blandy Experimental Farm, University of Center for Environmental Studies, Arizona
Virginia, Box 175, Boyce, VA 22620 State University, Tempe, AZ 85287
J. R. Callahan Stanley D. Smith
Museum of Southwestern Biology, University of Department of Biology
New Mexico, Albuquerque, NM University of Nevada-Las Vegas
Mailing address: Box 3140, Hemet, CA 92546 Las Vegas, NV 89154-4004
Jeffrey J. Johansen Paul T. Tueller
Department of Biology, John CaiToll University Department of Environmental Resource Sciences
University Heights, OH 44118 University of Nevada-Reno, 1000 Vallev Road
BOBISCKONDRATIEFF Reno. NV 89512
Department of Entomology, Colorado State Robert C. Whitmore
University, Fort Collins, CO 80523 Division of Forestry, Box 6125, West Virginia
University, Morgantown, WV 26506-6125
Editorial Board. Jerran T. Flinders, Chairman, Botany and Range Science; Duke S. Rogers, Zoology;
Wilford M. Hess, Botany and Range Science; Richard R. Tolman, Zoology. All are at Brigham Young
University. Ex Officio Editorial Board members include Steven L. Taylor, College of Biology and Agriculture;
H. Duane Smith, Director Monte L. Bean Life Science Museum; Richard W. Baumann, Editor, Great Basin
Naturalist.
The Great Basin Natwalist, founded in 1939, is published quarterly by Brigham Young University.
Unpublished manuscripts that further our biological understanding of the Great Basin and surrounding areas
in western North America are accepted for publication.
Subscriptions. Annual subscriptions to the Great Basin Naturalist for 1996 are $25 for individual sub-
scribers ($30 outside the United States) and $50 for institutions. The price of single issues is $12. All back
issues are in print and available for sale. All matters pertaining to subscriptions, back issues, or other busi-
ness should be directed to the Editor Great Basin Naturalist, 290 MLBM, PO Box 20200, Brigham Young
University, Provo, UT 84602-0200.
Scholarly Exchanges. Libraries or other organizations interested in obtaining the Great Basin
Naturalist through a continuing exchange of scholarly publications should contact the Exchange Librarian,
6385 HBLL, PO Box 26889, Brigham Young University, Provo, UT 84602-6889.
Editorial Production Staff
JoAnne Abel Technical Editor
Jan Spencer Assistant to the Editor
Copyright © 1996 hv Brigham Young University ISSN 0017-3614
Official publication date: 26 July 1996 7-96 750 19016
The Great Basin Naturalist
Published at Pkovo, Utah, by
Brigham Young University
ISSN 0017-3614
Volume 56 31 July 1996 No. 3
Great Basin Naturalist 56(3), © 1996, pp. 191-196
BIOGEOGRAPHIC SIGNIFICANCE OF LOW-ELEVATION
RECORDS FOR NEOTOMA CINEREA FROM THE NORTHERN
BONNEVILLE BASIN, UTAH
Donald K. Grayson^ Stephanie D. Livingston^, Eric Rickart'^, and Monson W. Shaver III^
Abstract. — The existence of low-elevation populations of Neotoma cinerea in the northern Bonneville Basin shows
either that these mammals can survive many thousands of years in xeric habitats or that the\' can mo\'e across xeric low-
lands far more readih' than has been appreciated, or both. Current models of Great Basin small mammal biogeography
are far too static to encompass properly the interaction of the wide range of geographical and biological variabilit>' that
has produced the modem distribution of those mammals that have, for several decades, been treated as "montane" within
the Great Basin.
Key words: Great Basin, biogeography, island biogeography, Neotoma cinerea, mammals.
Ever since J. H. Brown's insightful analyses coherence that has been assigned to it. Here,
of Great Basin small mammal biogeography we add to that growing body and call for a
(Brown 1971, 1978, see also Lomolino et al. more dynamic view of Great Basin small mani-
1989), biogeographers have treated the bushy- mal historic biogeography.
tailed woodrat (Neotoma cinerea) as a member
of an assemblage of small mammals that is cur- Neotoma cinerea ON
rently isolated on Great Basin mountains. The Homestead Knoll, Utah
composition of this assemblage is of particular
importance because it has been used to gener- Located a few km west and south of Great
ate and test hypotheses about the past and Salt Lake in north central Utah, the Lakeside
fiiture of Great Basin "montane" mammals (e.g.. Mountains are formed fiom a complex of north-
Grayson 1987, 1993, Patterson 1990, Cutler trending hills, ridges, knolls, and small moun-
1991, McDonald and Brown 1992, Murphy and tains (Fig. 1). The northwestern-most spur of
Weiss 1992, Grayson and Livingston 1993). this complex is Homestead Knoll, a low (maxi-
However, there is a growing body of data that mum elevation 1615 m), rocky promontory
suggests that this group of mammals lacks the that is devoid of active springs and permanent
i Burke .Vlemorial Museum. Box .353010, Universih' of Washington, Seattle. WA 9819.5.
^Quaternary Sciences Center, Desert Research Institute, Box 60220, Reno, NV 8950(1
3Utah Museum of Natural History, Universit\' of Utah, Salt Lake City, UT 84112.
■»Utah Geological Sui^ey 2363 South FooothiU Drive, Salt Lake City, UT 84109.
191
192
Great Basin Naturalist
[Volume 56
Fig. 1. Location of Homestead Cave within the northern Bonneville Basin.
streams, and that is separated from other parts
of the Lakeside group by valleys whose maxi-
mum elevations do not exceed 1465 m.
The barren playa of Pleistocene Lake Bon-
neville is located to the immediate west and
northwest of Homestead Knoll. Vegetation of
the knoll is dominated by shrubs and grasses,
aldiough there are a few scattered Utali junipers
{Juniperiis osteospenua) on its highest reaches.
Most prominent among the shrubs are
Atriplex confertijolia, Tetradijinia spinosa, and
Tetradymia glabrata. Artemisia tridentata is
present along seasonally moist drainages, while
Aiii'inisia spincscens, ChrysntJiainniis sp., and
Sarcuhatus vcnniciilatiis are present but un-
common above the flanks of die knoll. Artemisia
nova occurs on those flanks as does Ceratoidcs
lanata, while S. vermieiilatiis becomes increas-
ingly abundant as the \alley bottoms are
approached. We made no attempt to identify
the grasses that form the understoiy beneath
the shrubs, but cheatgrass {Bromiis teetormn
and, perhaps, B. ruhe)is) is extremely abundant
on the flats beneath the knoll.
Homestead Knoll is dotted b\' a number of
caves, one of which. Homestead Cave, sits on
the northwestern edge of the knoll at an eleva-
tion of 1406 m (Fig. 2). Approximately 11 m
wide and 6 m high at its mouth, this 25-m-
deep cave has, since 1992, been the focus of
interdisciplinaiy paleoecological work funded
by the Department of Defense. With D. B.
Madsen of the Utah Geological Surve\', 3
auUiors of this paper (DKG, SDL, and MWS)
have been involved with the excavation and
analysis of a deep sequence of vertebrate
remains from this site. To provide background
data for the anaKsis of the mammalian compo-
nent of the excax ated fauna, we conducted a
brief (270 trap-night) small mammal suney in
the vicinit\' of Homestead Ca\'e in June 1995.
With 1 exception, the residts of this sui^vey
were quite predictable. Trapping success was
low, with 3 species — Dipodomijs ordii (3 indi-
viduals), Peromyseus immieulatiis (11 individu-
als), and Neotoma lepida (6 iudi\ iduals) — com-
prising nearly the entire trapped assemblage.
The 1 exception, howexer, was remarkable: we
1996]
LOVV-ELEVAIION NeOTOMA CINEREA
193
Fig. 2. Location of Homestead Cave (white anow) on Homestead Knoll; the prominent tenaces represent Proxo, post-
Provo regressive, and Stansbuiy beaclies left by the waters of Pleistocene Lake Boniiex ille.
took a single Neotoma cinerea from the back of
Homestead Cave itself
Because this individual was live-trapped
and released, we cannot report its age or sex or
provide standard measurements. Even though
we do not have a voucher specimen, we do
have an excellent videotape of the animal (taken
by MWS and on file at the Utah Geological
Sui-vey), and there is no doubt as to the identi-
fication of the individual.
Vegetation in the immediate vicinity of
Homestead Cave departs from the Homestead
Knoll vegetation that we have described in
only 1 major way: the mouth of the cave sup-
ports a luxuriant growth of Rihes cercum
immediately beneath the dripline. It would be
surprising if this shrub were not heavily uti-
lized by both Neotoma cinerea, taken at the
rear of the cave, and Neotoma lepida, taken at
the front.
Other Low-elevation Northern
Bonneville Basin Records
FOR Neotoma cinerea
Our discover)' of Neotoma cinerea on Home-
stead Knoll led us to search the mammal col-
lection at the Utah Museum of Natural His-
toiy, University' of Utah, for additional records
of this species from other low-elevation set-
tings in the northern Bonneville Basin. We
were (juite successful in this search:
(a) Locomotive Springs: The only pre\ iousK
published low-elevation record for Ne()t())iia
cinerea for the northern Bonneville Basin was
provided by Durrant (1952:348; UU 5048) as
having been taken in October 1947 fi-om "State-
house, Locomotive Springs, 5500 ft. [1676 in]."
However, we are unable to determine the loca-
tion of "Statehouse" and are othenvise hesitant
to accept this record because of the substantial
difference between the actual elevation of
Locomotive Springs (1283 m) and the reported
elevation of "Statehouse" (1676 m). Given the
well-watered nature of Locomoti\e Springs,
the record might be accurate, but it is in need
of verification. Locomotive Springs is approxi-
mately 60 km north of Homestead Knoll.
(b) Lakeside Mountains: A\\ adult male Neo-
toma cinerea (UU 14374) was collected "5 mi.
E Lakeside, 4600 ft. [1402 m]" in June 1957.
This distance and direction fi-om Lakeside, how-
ever, describe a point in the Great Salt Lake. If
194
Great Basin Naturalist
[Volume 56
the actual direction were southeast, the speci-
men could have come fi-om Cave Ridge on the
eastern edge of the Lakeside Mountains,
approximately 10 km east of Homestead Knoll.
(c) Newfoundland Mountains: A series of
three juvenile Neotoina cinerea (UU 9995,
9996, 9998) were collected in June 1951 from
an unspecified site at the north end of the
Newfoundland Mountains. The collector s field
notes do not provide the elevation of the site
but do indicate that the specimens came from
an area of granite cliffs with a plant communitv'
that included Juniperus and Tetrodyinia. The
north end of the Ne\\^oundland Mountains is
approximately 40 km west-northwest of Home-
stead Knoll.
(d) Cedar Mountains: There are records for
Neotoina cinerea from 2 separate locations in
the southern Cedar Mountains: 4 from the
Cane Springs area (elevation 1768 m; UU
26340, 27297, 27299, and 27301-2, collected
between October 1952 and Januaiy 1953), and
1 from the "south end Cedar Mtn., 4850 ft.
[1478 m]." This last specimen is reported to
have been caught in a garage, suggesting that it
may have come from near Dugway. Although
these specimens come from no closer than 95
km to the south of Homestead Cave, we men-
tion them because they establish the likelihood
that Neotoma cinerea occurs in suitable habitat
throughout the Cedar Range.
BlOGEOGRAPHIC CONSIDERATIONS
Although Neotoma cinerea has frequently
been treated as being isolated on Great Basin
mountains (Brown 1971, 1978, Grayson 1993),
these records demonstrate that bushy-tailed
woodrats can and do exist at low elevations in
arid contexts within at least the northern Bon-
neville Basin. How, one must wonder, did
Neotoma cinerea come to occupy such arid,
low-elevation settings as the Newfoundland
Mountains (maximum elevation 2130 m) and
isolated knolls on the Lakeside Mountains
(maximum elevation 2020 m)?
It is well established that during the late
Pleistocene, bushy-tailed woodrats were far
more widely distributed within the Great Basin
than they are today, occupying low-elevation
settings where they are no longer found (Gray-
son 1988, 1993). As a result, it is reasonable to
speculate that these animals were also wide-
spread in this part of the northern Bonneville
Basin during those years. We can, however, do
much more than speculate about the histoiy of
N. cinerea in the Homestead Knoll area.
Widi a maximum elevation of 1615 m. Home-
stead Knoll was covered by the waters of Pleis-
tocene Lake Bonneville 14,500 years B.E, when
Pleistocene Lake Bonneville was at its high
(see Figure 2). Obviously, Homestead Knoll
must have received its woodrats after this time,
but when this occurred is not clear. Between
14,500 and at least 14,200 years B.P, when Lake
Bonneville stood at the Provo level, Home-
stead Knoll was an island of approximately 770
acres. Not until Lake Bonneville fell to a local
elevation of 1463 m did this island become
connected to the main body of the Lakeside
Mountains. Once this occurred. Homestead
Knoll became part of the faunal mainland and
would have been open to overland colonization
by terrestrial mammals.
Unfortunately, we do not know when the
lake fell to this level. However, we do have
direct evidence from Homestead Cave con-
cerning the regional history of Neotoma cinerea.
E.xcavations in this cave have provided a rich,
stratified sequence of vertebrate remains, the
mammals of which are being identified and
analyzed by one of the authors (DKG). To date,
a substantial sample of mammal specimens
from the 4 lowest Homestead Cave strata has
been identified (37,381 specimens).
All 4 assemblages contain both N. cinerea
and N. lepida, but the ratio of N. cinerea to N.
lepida xaries dramatically through time. In
stratum I, which dates to between ca 11,300
and 10,000 years B.P, bush\'-tailed woodrats
make up 99.38% of the Neotoma fauna. In sub-
sequent strata, however, they decline steadily
in abundance; by stratum IV (ca 8200-7200
years B.P), N. cinerea comprises only 4.74% of
the Neotoma assemblage (Fig. 3). Similarh; N.
cinerea contributes 23.97% of die total number
of identified mammalian specimens in stratum
I, a number that declines to 1.01% in stratum
IV (Fig. 4).
The Homestead Cave fauna thus documents
that N. cinerea was present in the Homestead
Knoll area by 11,300 years B.P and remained a
common species in the small mammal fauna
through much of the EarK Ilolocene. After ca
8200 years B.P, however, N. lepida became the
ovenvhelmingly dominant member of the genus,
and N. cinerea became localK rare. Since mam-
mals from later strata within Homestead Cave
1996]
Low-elevation Neotoma cim:ria
195
N = 1933
00
99.38
-
N = 1564
90^
86.13
80-
70-
60-
N = 564
50-
46.81
40-
30-
s
o
o
20-
10-
§
O
o
s
5
N = 4392 g
4.74 °
■D
-f
■0
|!-:;;ii.K ■.;?;!;] "D
1
II
III
IV
ST
RATL
M
Fig. 3. Changing contribution of N. cinerea to the
Neotoma {N. cinerea phis N. lepida) fauna. Homestead
Cave strata I-IV (N = total number of Neotoma specimens
identified to the species level, including those identified as
N. of cinerea and N. cf lepida).
have not yet been completely identified, we do
not know whether N. cinerea sui-vived the veiy
xeric Middle Holocene (ca 7500-5000 years
B.E) here.
Currently there are 2 options for explaining
the modern existence of N. cinerea on Home-
stead Knoll. First, animals living here today
may be direct descendants of the initial woodrat
colonizers of the knoll, colonizers that arrived
sometime between 14,500 and 11,300 years B.E
If so, the population has survived even though
its numbers dropped precipitously toward the
end of the Early Holocene (ca 8200-7200 years
B.E), and presumably fell even further during
the heart of the Middle Holocene. Assuming
that N. cinerea does not now survive in the val-
leys diat separate Homestead Knoll fi-om nearby
uplands, and that it has not been able to sur-
vive in those valleys since at least 7000 years
B.E, then this population has existed on an iso-
lated upland a few thousand acres in extent for
a minimum of some 7 millennia.
The other, and certainly more likely, option
is that Neotoma cinerea has not been isolated
on Homestead Knoll for this entire period of
time, that populations on the knoll have been
augmented by immigrants from elsewhere, and
that any local extinctions of N. cinerea on the
knoll have been followed by recolonizations
from nearby populations. Indeed, it is even
possible that the current representatives of the
species colonized Homestead Knoll during the
mid-1980s, a time of extraordinarily high pre-
cipitation in the northern Great Basin (Amow
and Stephens 1990).
30-
N - 1912
25-
23.97
20-
15-
10-
I
5-
i
?
'u
1347
83
N -
1 1
264
,09
N - 208 I
1.01 w
Fig. 4. Changing contribution of Neotoma cinerea to the
total number of identified mammalian specimens (\1SP)
per stratum at Homestead Cave, strata I-IV.
Implications
The discovery o( Neotoma cinerea on Home-
stead Knoll does not simply represent an unex-
pected natural historical tidbit. Our discoven-
documents either that populations of Neotoma
cinerea within the Great Basin can find suffi-
cient refuge in low-elevation, xeric habitats to
survive for many thousands of years, or that
this species can move across xeric lowlands far
more readily than has been appreciated, or
both. Indeed, insofar as bushy-tailed woodrats
are more effective colonizers than has been
realized, an effective parallel may exist in the
yellow-nosed cotton rat {Sigmodon ochrogna-
thus), a "montane" mammal of the Southwest
that has apparently expanded its range across
low-elevation valleys dining the past 50 years
(Davis and Dunford 1987; see also Davis and
Callahan [1992] on Microtus mexicanus).
Elsewhere, Grayson and Livingston (1993)
have noted that Sylvilagus nuttallii can cross
valley bottoms in at least parts of the Great
Basin. Now, it seems that N. cinerea can sur-
vive in habitats that are anything but montane.
This fact leads us to suggest that the nested-
ness of Great Basin mammal faunas (sensu Fat-
terson and Atmar 1986, Fatterson 1987, 1990)
might reflect a combination of extinction histo-
ries and colonization abilities. In addition, die
Homestead Knoll record for N. cinerea takes
its place alongside other recent data docu-
menting that current models of Great Basin
small mammal biogeography are far too static
to encompass properly the wide range of geo-
graphical and biological variability that has
produced the modern distribution of those
196
Great Basin Natur.\list
[Volume 56
mammals that, for several decades, have been
treated as "montane" within the Great Basin
(e.g., Gravson 1993, Grayson and Livingston
1993, Lawlor 1995, Rickart 1995). In the South-
west, modern montane mammal distributions
have clearly been determined by a complex
combination of Holocene extinctions and
colonizations (e.g., Davis and Dunford 1987,
Lomohno et al. 1989, Davis and Ctilldian 1992).
It now appears that the situation in the Great
Basin is quite similar
Acknowledgments
The research reported here was supported
by a grant fi-om the U.S. Department of Defense
Legacy Program (Project #0304843028X728,
"Paleoenvironmental Change on Hill Air Force
Base and Dugway Proving Grounds"). Our
thanks to D. B. Madsen for assistance at all
stages of this project and to R. S. Thompson for
confinning plant identifications. We also thank
R. Davis, D. B. Madsen, B. D. Patterson, and
T A. Vaughan for helpful comments on a draft
of this paper
Literature Gited
Arnow, T, and D. Stephens. 1990. Hydrologic character-
istics of the Great Salt Lake, Utah: 1847-1986. U.S.
Geological Siii"vey Water-SuppIy Paper 2332.
Brown, J. H. 1971. Mammals on mountaintops: nonequi-
libriiiin insular biogeography. American Naturalist
10.5:467-178.
. 1978. The theoiy of insular biogeograph\- and the
distribution ot boreal birds and mammals. Pages
209-227 in K. T Harper and J. L. Reveal, editors,
Intermountain biogeography: a symposium. Great
Basin Naturalist Memoirs 2.
Cutler, A. 1991. Nested faunas and extinction in frag-
mented habitats. Consei^vation Biology 5: 49(i-.50.5.
Davis, R., and J. R. Callahan. 1992. Post-Pleistocene dis-
persal in the Mexican vole (Microfiis inexicanus): an
example of an apparent trend in the distribution of
southwestern mammals. (Jreat Basin Naturalist .52:
262-268.
Davis, R., and C. Dunfokd. 1987. An example of contem-
porary colonization of montane islands by small, non-
flying mammals in the American Southwest. Ameri-
can Naturalist 129: 398-406.
Durrant, S. D. 1952. Mammals of Utah: taxonom\' and
distribution. Universit> of Kansas Publications,
Museum of Natural Histoiy 6.
Grayson, D. K. 1987. The biogeographic histoiy of small
mammals in the Great Basin: observations on the last
20,000 years. Journal of Mammalogy 68: 359-375.
. 1988. Danger Ca\'e, Last Supper Cave, Hanging
Rock Shelter; the faimas. American Museum of Nat-
ural Histoiy Anthropological Papers 66.
. 1993. The deserts' past: a natural prehiston' of the
Great Basin. Smithsonian Institution Press, Washing-
ton, DC.
Gr.\yson, D. K., and S. D. Livinc.ston. 1993. Missing
mammals on Great Basin mountains: Holocene extinc-
tions and inadequate knowledge. Conservation Biol-
ogy' 7: 527-532.
Lawlor, T. 1995. Biogeography of Great Basin mannnals:
paradigm lost. Unpublished manuscript on file at the
Department of Biological Sciences, Hiunboldt State
University, Aicata, CA.
LoMOLiNO, M. v., J. H. Brown, and R. Dams. 1989. Island
biogeography of montane forest mammals in the
American Southwest. Ecology 70; 180-194.
McDonald, K. A., and J. H. Brown. 1992. Using mon-
tane mammals to model extinctions due to global
change. Consen'ation Biolog\' 6; 409—115.
Mlhphy, D. D., and Weiss, S. B. 1992. Effects of climate
change on biological diversity in western North
America: species losses and mechanisms. Pages
355-368 in R. L. Peters and T. E. Lovejoy, editors.
Global warming and biological di\'ersit\-. Yale Univ er-
sity Press, New Haven, CT.
PATfERSON, B. D. 1987. The principle of nested subsets
and its implications for biological consenation. Con-
sen ation Biologv' 1: 32.3-^334.
. 1990. On the temporal development of nested
subset patterns of species composition. Oikos 59:
330-342.
Patterson, B. D., and W. Atmar. 1986. Nested subsets
and the structure of insular mammalian faimas and
archipelagos. Biological lournal of the Linnean Soci-
ety' 28; 65-82.
Rickart, E. A. 1995. Ele\ational diversity gradients, bio-
geography', and the structure of montane mammal
communities in the intermountain region. Unpub-
lished manuscript on file at the Utah Museum of
Natural Histon, Unixersity of Utah, Salt Lake City.
Received 15 Novcmlycr 1995
Accepted 20 March 1996
(Jrt-at Basin Naturalist 56(3), © 199(i, pp. 197-204
SYNOPSIS OF THE MOSSES OF WYOMING
PM. Eckel'
Abstkact. — A it'\ isrd list ol the mosses oi tlic Stale ol Wyoming is prescntccl. Ik'corck'd are 315 species and \arieties.
Kci/ words: hn/dplit/tcs. \Vy()inin<:,. rortcr. Rocky Moiinlai)t.s. citcrklisl. flora.
Publication of the mosses of Wyoming began
with A\en Nelson (1900) listing 119 species.
His collections were made essentially in the
Laramie and Medicine Bow ranges of Carbon
and Albany counties, and his specimens are
now at the Rock)' Mountain Herbarium (RM).
Nelson s specimens were determined in large
part by Professor John M. Holzinger, of the
State Normal School at Winona, Minnesota.
Aven Nelson s and Elias Nelson's Wyoming col-
lections were distributed under printed labels
as "Plants of Wyoming from the Rocky Moun-
tain Herbarium" and "Plants of Yellowstone
National Park from the Rock)' Mountain Herb-
ariiun,' citing Holzinger and a few others as
determiners. However, because they were
issued without serial numbers, showing only
the collectors' numbers together with other
data, they do not constitute true exsiccatae
(Sayre 1971).
In 1935 Cedric Lambert Porter published a
valuable checklist of the mosses of the State of
Wyoming, citing 215 species and varieties. In
this publication Porter mentioned a paper by
Dwight C. Smiley on the mosses of Yellow-
stone National Park, which Porter cited again
two years later as "A Key to the Mosses of Yel-
lowstone National Park ' [unpublished] (Porter
1937). Porter included Smiley's names but
apparently did not examine his specimens.
The substance of a dissertation written by
Porter at the University of Washington in 1937
was the development of a useful key to the
hepatic and bryophyte taxa of Wyoming. It
includes additional county records, a few addi-
tional species, and references to Yellowstone
National Park, again apparently citing Smiley's
unpublished material. On a recent visit to the
Yellowstone National Park Herbarium (YE LLC),
I was able to locate and borrow for stud) some
of Mr Smile\' s specimens. Several were deter-
mined 1)\ R. S. Williams.
No checklist focusing on the moss flora of
Yellowstone National Park has as vet been pub-
lished. The 357 or so moss specimens curated
at YE LLC all derive from the park. The fol-
lowing are collectors and dates of collecting
activity: Dwight Smiley, 1932; H. S. Conard,
1948; Winona Welch, 1951; Eula Whitehouse,
1951; EKa Lawton, 1953. No biographical data
exist on Mr. Smiley at YELLO (Whipple per-
sonal commimication). I have no other record
that these collectors worked in Wyoming, but
my use of various herbaria for this checklist is
not exliaustive. Incidentally, there are 11 pack-
ets of livei^worts at YELLO, and no representa-
tion of Spljagnuin.
Recent collectors in the state include the
late Fredrick Hermann, Holmes Rolston, and
William Weber, all of whose Wyoming speci-
mens are distributed in various herbaria, espe-
cially COLO. Steven Churchill (1979, 1982),
John Spence (1985), and Alvin L. Medina (1994)
collected additional taxa. Two taxa listed here
as new to the state were recently collected by
Joseph Elliott {ScorpidiiDn scorpioides and
Cinclidium stygium). General references to ex-
tensive collections made in the state are given
by Lawton (1971).
The purpose of this paper is to present an
up-to-date list of the mosses of the State of
Wyoming, incorporating reports published
since Porter's manuscripts as well as additional
unpublished infonnation. There are 315 species
and varieties in the present list. Some idea of
the degree of representation of the flora of
Wyoming comprising this list may be inferred
from a glance at similar checklists for other
'Clinton Herliarium. BufSilo \Iiis£'uni of Science, BnflTalo, NY 14214.
197
198
Great Basin Natumlist
[Volume 56
ai-eas. It is probabK to be expected that the
arid intermountain states, such as Utah and
Nevada, will have a more depauperate flora;
Spence (1988) recorded 342 species for the
entire Intermountain West. The following
counts are for species and varieties (except
Utah which is species only): Arizona 381
(Johnsen, no date), Colorado 292 (Weber 1973),
Idaho 257 (McCleaiy and Green 1971), Mon-
tana 358 (Eversman and Shaip 1980), Nevada
165 (Lawton 1958, Lavin 1981), Oregon 441
(Christ)- et al. 1982), Utiili 256 (Flowers 1973).
There appears to be no checklist for the state
of Washington. New \brk State, which is said
to have a diverse moss flora, has 503 species
and varieties (Ketchledge 1980). A striking
comparison is to the oceanic island of New-
foundland, which boasts a moss flora of 445
species (Brassard 1983) and which is geoph>s-
iographicalK rather plain compared to tlie geo-
moiphic extremes and di\ersit>' of \\yoming in
tlie Central Rock> Mountains.
The following checklist is based largely on a
rexiew of specimens housed in the Rock>
Mountain Herbarium. Additional herbaria were
contacted in instances of taxa reported in die
literature but witli no representation at RM, and
nimierous new records ha\e been added from
field collections b>" m>self and others. I have
attempted to cite at least one reliable specimen
of each taxon by giving the abbreviation of tlie
herbarium at which the specimen is located.
Additional tiixa are added from Porter's 1937
dissertation if die\' did not occur in his previ-
ous publication. If a herbarium designation is
noted below for a ta.xon, no further reference
to the literature is given. Although sexeral ref-
erences to the same species throughout the lit-
eratiue ma\ ha\e been cited, only one citation
is presented for names for which specimens
ha\e not been seen.
Nomenclature other than Spliagninn follow s
Anderson et al. (1990). Sphagninn nomencla-
ture follows Anderson (1990).
Families are gi\en alphabetically, genera
alphabeticalK widiin families, and species iilpha-
beticalK within genera.
Checklist of Mosses of Wyomlxg
Amblysteciaceae
CaUu'rTa (RM. WTU)
Palustriclla decipiens (De Not.) Ochyra (RM. WTU)
Pseudoculliergon turgescens (T. Jens.) Loeske (COLO, RM)
Sanionia uncinata (Hedw.) Loeske (NY^ RM)
Sanncnthypnuni sannentosum (Wahlenb.) Tuom. &: T. Kop.
(RING. COLO, RM)
Scorpidiuni scorpioides (Hedw.) Limpr. (BUE)
Wanistoifia cxannulaia (Schimp. in BSG) Loeske
\ar. exannulata (CSU. RM)
Warnstoifia jhiitans (Hedw.) Loeske (COLO, RM)
.\\DREAEACE.\E
Andrcaca ntpesfris Hedw. (YELLO)
AUL.\C0M-\UCE.\E
Aulacomnium androgynwn (Hedw.) Schwaegr. (TENN)
Aulacouiniuni pahistre (Hedw.) Schwaegr. (BUE RM>
TENN)
Bartfl\miace.\.e
Anacolia nienziesii (Turn.) Par (WTU)
Bartramia ithyphylla Brid. (RM)
Philonotis fontana (Hedw.) Brid. \-dv. fontana (BUE R.M)
\ar americana (Dism.) Flow. (BUF RM)
\ar cacspitosa (Jun) Schimp. (BUE RM)
\ ar putnila (Tum.) Brid. (RM)
Amblystegium serpens (Hedw.) Schimp. \ar serpens (NY.
RM)
va.r.juratzka)utin (Schimp.) Ran & Hen-. (RM)
Amblystegium cariuin (Hedw.) Lindb. (NY. RM)
BtvU:inTHECI.\CE.\E
Brachythccium acutum (Mitt.) Sull. (RM)
Brachythcciuni (dhican.s (Hedw.) Schimp. in BSG (RM)
19961
Mosses of Wyoming
199
Brdclu/tluciiiin colliniim (Sclik'icli. c.v (J. MiR'll.) Scliiiiip.
//( BS(;(BUFNY, RM)
Bracltythecititn enjthrorrhizon Sciiimp. in BSU (RM)
Bracliijtlu'ciitmfendh'h (Sull.) Jaeg. (RM)
Braclujtlu'ciumfri^idum (C. Miiell.) Besch. (NY, US)
Brachythecitim leiij'-^ii Grout (NY, RM, US)
Brachytlu'ciwn nelsonii Cioiif "''SU, NY, RM)
Brachjtheciwn ocdipodiwn (Mitt I Tae?. (COLO, RM, US)
Brachijthcciuin rivularc Sdiinip. in BSG (NY, RM, US)
Bracliythcciiitn ndalndum (lli'tlw.) Schimp. in BSG (NY)
Brachytlteciuni salebrosuin (Wch. & Mohr) Scluin]). in
BSG (BUR RM, US)
Brachytlu'ciwn starkci (Brid.) Schimp. in BSG (Porter 1935)
Brachythcciian twnidwn (C.J. Hartin.) Kincll). (COLO, NY,
RM, US)
Bnichythcciwn velutinum (Hedw.) Schimp. in BSG
var. velutinum (Spence 1985)
var. venustum (De Not.) Arc. (BUR RM)
CiniphyUum cirroswn (Schvvaegr. in Schiiltes) Grout (NY)
Eurhynchium ureganum (Sull.) Jaeg. (Spence 1985)
Eurhynchium pulchcUum (Hedw.) Jenn. (RM)
Homidothccium acneuin (.Mitt.) Lawt. (RM, US)
Honmlotheciwn nevadense (Lesq.) Ren. & Card. (US)
Homcdothecium pinnatifidum (Sull. & Lesq.) Lawt. (BUF,
RM)
Tonientypnwn nitens (Hedw.) Loeske (COLO, RM)
BRVACEAE
Brywn alsis australasiac (Grev. & Hook.) Robins.
(BING, BUF)
PFERVGY-NANDRACEAE
Hetcrocladium diinori>lium (Brid.) Schimp. in BSG (BUF)
Myurclla julacca (Schwaegr) Schimji. in BSG (RM)
RllVTIDIACEAE
Rlujtidiuin rugosiim (Hedw.) Kiiidb. (WTU)
Selk;eriaceae
Blindia acuta (Hedw.) Brnch & Schimp. in BSG (COLO,
CSU, RM)
Seligcria campylopoda Kindb. in Macoun & Kindb. (ALT\,
BUERM)
Sphagnaceae
Si)hagnum angustifolium (C. Jens, t'.v Russ.) C. Jens, in Tolf
(COLO, RM)
Sphagnutn annulatum H. Lindb. ex Wanist. (BING, RM)
Sphagnum contoiium Scliultz (BING)
Sphagnuni find)riatum W'ils. in Wils. & Hook. f. (COLO,
RM)
Sphag)iU)nfuscum (Schimp.) Klinggr. (RM)
Sphagnum platyphyllum (Lindb. ex Braithw.) Sull. t'.v
Wamst. (RM)
Sphagnum russowii Wamst. (RM)
Sphagnum squarrosum Crome (RM)
Sphagnum subsecundum Nees in Sturm (BING)
Sphagnum teres (Schimp.) Aongstr in Hartni. (BING)
Sphagnum uarnstoifii Russ. (RM, COLO)
Spl.achn.\ceae
SjylachiHun splmcricum Hedw. (Y'ELLO)
Tayloria acuminata Hornsch. (Cram and Anderson 1981)
Tayloria ligulata (Dicks.) Lindb. (COLO, RM)
Tayloria serrata (Hedw.) Bruch & Schinip. in BSG (NY)
Tetr.\phii:)aceae
Tetraphis pcllucida Hedw. (RM)
Thuiduceae
Abictinclla ahiclina (Hedw.) Eleisch. (BUE RM)
Tl.MMI.ACEAE
Timmia austriaca Hedw. (BUE RM)
202
Great Basin Naturalist
[Volume 56
Timmia megapolitana Hedw. var. megapoUtana (NY, RM)
var. bavarica (Hessl.) Brid. (BUF; RM)
Problematic Taxa
Brachythecimn campestre (C. Muell.)
Schimp. in BSG. Old Faithful, Yellowstone
National Park, Smiley, according to Porter
(1934, 1935). Porter's citation is apparently not
based on a specimen but on Dwight Smiley's
checklist. Since this is primarily a taxon of the
eastern United States, it should probably be
excluded from the state flora. Specimen not
seen.
Brachythecium oxycladon (Brid.) Jaeg. This
taxon, typical of the eastern United States, is
based by Porter (1934, 1935) on a citation by
Smiley. No coiTcsponding specimens were seen.
Funaria flavicans Michx. No specimens were
seen of this taxon reported by Porter (1935). As
it appears to be a species of the eastern region
of the United States (Crum and Anderson
1981), it is of doubtful occurrence in Wyoming.
Mniiim honiiim Hedw. (Porter 1935). This is
a taxon of the eastern montane region of North
America and the Piedmont (Crum and Ander-
son 1981) and not likely to occur in Wyoming.
No specimen seen.
Platydictya confervoides (Brid.) Ciimi. Cited
by Porter (1935) as a doubtful determination; it
cannot be located in the herbaria consulted.
Rocomitriumfosciculare (Hedw.) Brid. Porter
(1935) did not see a specimen of this species,
reported by Nelson (1900). Spence's citation
for Teton County (1985) refers to Porter's
doubtful citation. No specimens were seen by
the present author
Excluded Taxa
Brachythecium calcarewn Kindb. A specimen
of Smiley's o{ Brachytheciwn flexicaule, now B.
calcarewn, from Yellowstone National Park and
cited by Porter (1935) was determined as a
depauperate specimen o( B. fiigichnn.
Brachythecium glareosum (Br) B. & S. Lake,
Yellowstone National Park (Smiley) (Porter
1934, 1935). Specimens of Smiley and other
collectors at YELLO were variously Brachy-
thecium salehrosum, B. leihergii, B. albicans.
and B. frigidum.
BreuteUa mohriana (C. Muell.) Broth., Car-
bon Co. (Porter 1937). E.xcluded from North
America (Anderson et al. 1990).
Brotherella recurvans (Michx.) Fleisch.,
Lincoln Gulch, Albany County (Aven Nelson
2628). "The material is scanty, and Prof.
Holzinger, who identified it, expressed a doubt
as to the correctness of the determination'
(Porter 1935). The specimen with Holzinger's
opinion is at RM.
Bryum canariense Brid. (Porter 1935). This
is a species of the West Coast and not likely to
occur in Wyoming. The specimen cited by
Porter (Nelson 7814) curated at RM and US
seems to be Bryum caespiticium Hedw.
Encalypta streptocarpa Hedw. (Porter 1930,
1935). Excluded from North America (Ander-
son et al. 1990).
Gymnostomwn calcarewn Nees & Homsch.
in Nees et al. (Porter 1937. "Washakie Co.").
The specimen from WTU of Porter, Sept. 9,
1935, No. 2094, "On limestone boulders in a
shady canyon' from the Ten Sleep Canyon in
the Big Horn Mts, Washakie, Co., has been
determined to be Gymnostomwn aeruginoswn
by R. Zander. The specimen in section shows a
ventral costal epidermis, two stereid bands, a
central strand in the stem. The capsules were
young and so rather ovoid.
Homalothecium lutescens (Hedw.) Robins.
Based on a citation by Porter (1937) and proba-
bly the specimen: Yellowstone National Park
Nelson No. 6041 (RM) appears to me to be
Homalothecium aeneum instead.
Hypnum callichrown Funck. ex Brid., Evan-
ston, Uinta County (Aven Nelson 4128, in part:
"The identity of this plant is doubtful, " Porter
1935; Uinta Co., Porter 1937). The specimen
Nelson 4128 appears to be Hypnum lindbergii.
Macrocoma sullivantii (C. Muell.) Grout
(BUF). This record is due to a labeling error
(Vitt 1981; D. Vitt, in litt). in the Orthotri-
chaceae Boreali — Americanae Exsiccatae Fas-
ciculus HI Nos. 21-30. The label issued with
this species name, number 30, should be num-
ber 27, Orthotrichum rupestre. The Macro-
coma specimen originated in North Carolina,
the Oiihotrichuiu from Yellowstone National
Park (J. A. Christie, in litt.).
Meesia triquetra (Richt.) Aongstr. reported
by Cooper and Andrus (1994) is Oncophorus
wahlenhergii.
Orthotrichwn spcciosum Nees in Sturm
(Porter 1935). Specimens at YELLO and RM
were determinable as O. laevigatutn.
Orthotrichum pallens Bruch ex Brid. var.
parvum Vent., "Yellowstone National Park"
1996]
Mosses of Wyoming
203
(Flowers 1973). Excluded from North America
(Anderson et al. 1990).
Physcomitrium pyrifonne (Hedw.) Hampe,
cited by Porter (1937) for Crook Co., is proba-
bly a specimen collected by Marion Ownbey
from that count>' (No. 556a, TENN) and deter-
mined by Porter as P. turhinatwn, but which,
upon examination, is P. hookeri.
P](i(i.ion}nhiiu ajfinc (Bland, ex Fimck) T. Kop.
(Porter 1935). Excluded from North America
(Anderson et al. 1990).
PlagiotJwciiitn cavifoUiim (Brid.) Iwats.
Porter (1935) based this name on a specimen of
Elias Nelson (5242), which is Isoptcrygiopsis
pulchella.
Sphagnmn capillifolium (Ehrh.) Hedw.
(Porter 1935). Porters specimens 1198 and
1199 collected in 1932 identified as S. capilli-
folium had been redetermined by R. Andrus as
S. nissowii (Andrus in lift.).
Sphagnum )najus (Russ.) C. Jens. (Porter
1935, CiTun 1984). Taxa collected from Wyom-
ing and identified as S. majus have all been S.
annulatum, according to Andrus {in lift.), who
states that the nearest sites would be in British
Columbia, central Alberta, and Minnesota.
Sphagnum palustre L. (COLO, RM). This
species has been found only along the West
Coast by Andrus {in lift).
Sphagnum recurvum R Beauv. (COLO, RM).
Material of this species from the interior of the
United States is referable to S. angustijolium,
according to Andrus {in lift.). Sphagnum recur-
vum is an eastern coastal plain species.
Tortula princeps De Not. Reports by Porter
from Carbon and Crook counties (1935) were
based on Nelson 2818 and 5034 at RM, and a
specimen (RM) by Hennann (No. 17844), which
were redetermined by R. Zander as Tortula
ruralis.
Weissia controversa Hedw. (Porter 1935). All
citations for this species appear to be based on
specimens of Dwight C. Smiley, deposited at
YELLO. All 3 specimens seen were Dicrano-
weisia crispula.
Acknowledgments
The author gratefulK' acknowledges the help
of the late Mason Hale for his early support of
this project, Holmes Rolston, and the late Fred
Hermann. The assistance of Ronald Hartman
of the Rock\' Mountain Herbarium, who pro-
vided me with a copy of Porter's dissertation
and access to field collections, and Jennifer
Wipple of Yellowstone is acknowledged. Taxo-
nomic assistance was provided by Richard
Andrus, Teny Mcintosh, Howard Crinii, John
Spence, Ronald Pursell, Norton Miller, Dale
Vitt, and Richard Zander. The treatment of
Sphagnum was considcral)ly improved by infor-
mation provided by Dr Richard Andrus. Valu-
able specimens were sent to me by Janice R
McKee, Nancy Kastning-Culp, William Buck,
Holmes Rolston, William Reese, and Joseph
Elliott. This project was sponsored b\' the Sav-
age Fund of the Buffalo Museum of Science
and is dedicated to the memory of the late
Elva Lawton.
Literature Cited
Anderson, L. E. 1990. A checklist of Sphagnum in North
America north ofMe.xico. Bi^ologist 93: 500-501.
Anderson, L. E., H. A. Crum, and W. R. Buck. 1990. List
of the mosses of North America north of Mexico.
Biyologist 93: 448-199.
Brassard, Guy R. 1983. Checklist of the mosses of the
island of Newfoimclland, Canada. Biyologist 86: 54—63.
Christy, J. A., J. H. Lvford, and D. H. Wagner. 1982.
Checklist of Oregon mosses. Biyologist 85: 22-36.
Churchill, S. E 1979. Mosses of the Great Plains III.
Additions to Nebraska and the Black Mills of St^uth
Dakota and Wyoming. Biyologist 82: 72-75.
. 1982. Mosses of the Great Plains VIII. Additions.
Biyologist 85: 218-221.
Cooper, David J., and Richard E. Andrus. 1994. Pat-
terns of vegetation and water chemistry in peatlands
of the west-central Wind River Range, W\'oming,
U.S.A. Canadian Jonrnal of Botany 72: 1586-1597.
Crum, H. A. 1984 Sphagnopsida, Sphagnaceae. North
American flora, series II; part 11. New York Botanical
Garden. 180 pp.
Crum, H. A., and L. E. Anderson. 1981. Mosses of east-
em Nortli America. Columbia University Press, NY.
EvERSMAN, S., AND A. J. Sharr 1980. First checklist of
Montana mosses. Proceedings of the Montana Acad-
emy of Science 39: 12-24.
Flowers, S. A. Holmgren, editor. 1973. Mosses, Utah
and the West. Brigham Young University Press,
Provo, UT
HoRTON, D. G. 1983. A revision of the Encahptaceae. II.
Journal of the Hattori Botanical Laboratoiy 54:
353-532.
Ireland, R. R. 1982. Moss flora of tlie Maritime Provinces.
National Museums of Canada, Ottawa.
JOHNSEN, Ardith B. [Undated]. Keys to the mosses of Ari-
zona. Research Paper 14. Museum of Northern Ari-
zona, Flagstaff.
Ketchledge, E. H. 1980. Revised checkhst of the mosses
of New York State. In: R. S. Mitchell, editor. Flora of
New York State. New Y'ork State Museum Bulletin
440. Albany, NY
Lavin, M. 1981. New records for the moss flora of Nevada.
Bryologist 84: 93-94.
Lawton, E. 1958. Mosses of Nevada. Bryologist 61:
314-334.
204
Great Basin Naturalist
[Volume 56
. 1971. Moss flora of the Pacific Nortliwest. Journal
of the Hattori Botanical Laljoraton, Nichinan, Japan.
McCleary, J. A., AND V. V. Gree.x. 1971. A checklist of
Idalio mosses. Biyologist 74: 175-180.
Medina, Alvix L. 1994. Lichens and biyophytes of the
Rochelle Hills, Campbell County, Wyoming. Evansia
1: 121-130.
Nelson, Aven. 1900. The ci-yptogams of Wyoming. 10th
annual report of the W\oming E.xperiment Station,
Laramie. 38 pp.
Porter, C. L. 1930. Fruiting plants of Encalypta contorta.
Bi-yologist 34; 93.
. 1934. The moss genus Brachythecium in Wyoming.
University of Wyoming Publications in Science.
Botany 1(9): 235-241.
. 1935. Bnophytes of Wyoming. Part II. Hepaticae
(concluded) and Musci. Bryologist 38: 101-114
. 1937. The biyophytes of Wyoming. Unpublished
doctoral dissertation. University of Washington,
Seatde.
Sayre, G. 1971. Ciyptogamae e.xsiccatae, an annotated bib-
liograph) of published e.xsiccatae of Algae, Lichenes,
Hepaticae, and Musci. Memoirs of the New York
Botanical Garden 19(2): 175-176; 214-215.
Shaw, A. J. 1981. A taxonomic revision of the propagulifer-
ous species of Pohlia (Musci) in North America. Jour-
nal of the Hattori Botanical Laboratoiy 50; 1-81.
Spence, J. R. 1985. Checklist of the mosses of Grand Teton
National Park and Teton County, Wyoming. Great
Basin Naturalist 45; 124-126.
. 1988. Checklist of the mosses of the Intermoun-
tain West, USA. Great Basin Naturalist 48; 394-101.
ViTT, D. H. 1973. A revision of the genus Orthotrichtim.
Br>'ophytorum Bibliotheca, J. Cramer Verlag, Vaduz.
, editor. 1981. Orthotrichaceae Boreali — Ameri-
eanae exsiccatae fasciculus III, Nos. 21-30. Univer-
sity of Alberta, Edmonton, Canada.
. 1991. Rediscovei-y of Orfhotrichwn holziiii'eri: its
moipholog)' and habitat in western North America.
Biyologist 94: 77-79.
Weber, W. 1973. Guide to the mosses of Colorado. Insti-
tute of Airtic and Alpine Research Occasional Paper
6. Universit}' of Colorado, Den\'er
Wynne, E E. 1943. Range extensions of mosses in western
North America. Biyologist 46: 149-155.
Received 21 December 1995
Accepted 27 March 1996
Great Basin Naturalist 56(3), © 1996, pp. 205-210
VARIATION IN BITTERBRUSH {PURSHIA TRIDENTATA PURSH)
CRUDE PROTEIN IN SOUTHWESTERN MONTANA
Carl L. Wanibolti, W. Wyatt Fraas^, and Michael R. Krisina^
Abstract. — The objective of this study was to clcterniine iCcrudc protein \aries sij^iiilicaiitlN duriiiti; late suiimier and
midwinter among stands of hitterbrush {Piirshia trklentata Piush) in southwestern Montana. A secondaiy objective was
to determine if leaves, when present, contribute significant additional protein in the region. Nine sites with different
en\ironniental conditions and witliin a radius of 14.5 km were studied. Bitterl)rush leaves and leaders collected in
August 1990 and 1991 and FebruaiT 1991 were used for crude protein and leaf-to-leader ratio determinations. Crude
protein difTered (F < 0.001) among sites for both leaves and leaders on individual collection dates. Crude protein in
lea\es was nearly twice the level found in leaders. Because few leaves were present in Februaiy, they increased crude
protein in total foliage by only 0..3% over twigs alone. Feliruan' crude protein levels averaged 6.8% for total foliage,
which is below the estimated requirement for wintering deer.
Kcij words: Purshia tridentata, hitterhrush, crude protein, winter range, big game nutrition, Montana.
Protein is one of the most important nutri-
ents for wintering ungulates (Dietz 1972).
Welch et al. (1983) estimated that winter crude
protein levels o{ Purshia trklentata Pursh (hit-
terbrush) are not high enough to meet ungu-
late requirements, but postulated that protein
content might vaiy with populations of bitter-
brush. Differences in hitterbrush protein con-
tent between sites have been noted (Giunta et
al. 1978), although not between local habitat
types (Morton 1976). Slausen and Ward (1986)
found no difference in crude protein among 3
Colorado accessions in a common garden, but
Welch et al. (1983) found differences in a com-
mon garden test with plants from a wider geo-
graphical area. No differences in nutrient con-
tent have been found at varying browse levels
of hitterbrush plants (Dietz et al. 1962, Shep-
herd 1971). Crude protein levels were higher
when winter leaves were present (Dietz et al.
1962), but winter leaf presence varies between
populations of hitterbrush (Welch et al. 1983).
Our objective was to determine if crude pro-
tein varies significantly during late summer and
midwinter among stands of hitterbrush in soutli-
western Montana. Secondarily, we wished to
determine if hitterbrush leaves in our region
contribute significant additional crude protein
quantities when present.
Methods
Study Sites
Nine study sites were chosen primarily to
represent hitterbrush stands from a range of
environmental conditions (Table 1). This in-
cluded burned sites and hitterbrush sites pro-
tected from browsing. All study sites were
located within a radius of 14.5 km near Butte
and Anaconda in southwestern Montana. Long-
term climatic records were available for the
general study area from the Anaconda weather
station at 1700 m elevation. Annual precipita-
tion at Anaconda averages 340 mm, with 47%
received between April and July (NOAA 1991).
Vegetation types at all but 3 sites (burn,
unburn, and High Rye) were serai stages of the
bitterbrush-bluebunch wheatgrass {Agropijron
spicatum Pursh) habitat type (Mueggler and
Stewart 1980). The dominant shrub was hitter-
brush, but understoiy vegetation was regressed
(Fraas et al. 1992) on the other 6 sites from the
described potential climax composition (Youtie
et al. 1988).
The Butte site at Maude S Canyon, near
Butte, Montana, was selected because it re-
ceives no ungulate browsing. The plant com-
munity consisted of bitterbmsh, Centaurea mac-
ulosa Lam. (spotted knapweed), Ribes cereum
Dougl. (squaw currant), and Rosa woodsii
Lindl. (Woods rose).
'Department of Animal and Range Sciences, Montana State University, Bozeman, MT 59717
2Montana Fish, Wildlife, and Parks. Butte, MT 59701.
205
206
Great Basin Naturalist
[Volume 56
Table 1. Topographic characteristics of the 9 study sites.
Data from the last 4 sites were obtained from Guenther
(1989).
Site
Elevation
Slope
Aspect
(m)
(%)
(degrees)
Butte
1730
26
234
Cattle e.xclosiue
1830
16
188
Cattle + deer
1820
10
190
Bum
2010
21
220
Unburn
2010
24
180
Powerline
1640
16
85
Willow Creek
1780
31
110
Railroad Gulch
1650
32
115
High R\e
1940
38
120
At Diy Cottonwood Creek in the Deerlodge
district of the Deerlodge National Forest, a
livestock e.xclosiue with deer-only use was
studied and known as the cattle exclosure site.
Near the exclosure, a bitterbrush stand was
studied and known as the cattle + deer site
because it sustained both cattle and mule deer
browsing. These 2 sites have a scattered over-
story of Pseudotsuga menziesii [Mirb.] Franco
(Douglas-fir). A high number of native peren-
nial forbs occurred in the understory on these
sites.
Two sites were selected to gauge the impacts
of burning bitterbrush in southwestern Mon-
tana. The 2 sites (bum, unburn) were situated
on either side of the burn line on the south
flank of Steep Mountain, 8 km northwest of
Butte, in the Butte District of the Deerlodge
National Forest. The plant community on these
2 sites was a bitterbrush-mountain big sage-
brush {Artemisia thdentata Nutt. ssp. vaseijana
[Rydb.] Beetle)-bluebunch wheatgrass associa-
tion intermediate to the big sagebrush-blue-
bunch wheatgrass and bitterbrush-bluebunch
wheatgrass habitat types of Mueggler and
Stewart (1980). The prescribed burn was con-
ducted 3 November 1981 after a year's rest fi'om
livestock grazing to increase fuel loads. Live-
stock use resumed 15 September 1982. When
sampled for protein content, bitterbrush on the
burned site was significantly lower in canopy
cover (F > 0.01), flower production (P > O.lj,
and seed production (F > 0.1) than on the un-
bunied site (Fraas et al. 1992).
Four sites were located on the Moimt Hag-
gin Wildlife Management Area (MHWMA),
owned and managed In Montana Fish, Wildlife,
and Parks. The Powerline site was on a slope
50 m above a perennial stream on the north-
east edge of the MHWMA big game winter
range. The plant community consisted of bitter-
biTish and spotted knapweed. The WiUow Creek
site was near the top of a grassy ridge 150 m
above Willow Creek. This site supported a rel-
atively large amount o{Elijmus cinereus Scribn.
& Merr. (basin wild lye), along with other
perennial grasses and bitterbrush. This area
was used as winter range b\' mule deer, elk,
and moose. The Railroad Gulch site was also
on the deer and elk winter range. This site
occupied a midslope position 30 m above an
intermittent stream, where the plant commu-
nity consisted of bitterbrush and spotted knap-
weed. The High Rye site was 1500 m higher in
elevation than the other MHWMA sites and
appeared to receive the greatest snowpack.
The plant community on die High Rye site was
typical of the bitterbrush-rough fescue {Fes-
tiica scahreUa Torrey ex Hook.) habitat type
(Mueggler and Stewart 1980) with those species
currently dominant. Guenther (1989) found the
least amount of big game use at this location
among the 4 MHWMA sites. The MHWMA
study sites received insignificant levels of live-
stock grazing.
Sampling and Analysis
Leaves and leaders (current-year stem
growth minus leaves) were collected at each
study site from 10 randomly selected plants for
crude protein analysis on each of the sampling
dates. The same plants were sampled to deter-
mine leaf-to-leader ratios. Material was col-
lected the 1st week of August prior to or at
seed set in 1990 and 1991. This was estimated
to be the period of minimum soluble carbohy-
drate content for bitterbrush plants (Menke
and Trlica 1981). Material was also collected on
12 Februaiy 1991, when mule deer were con-
centrated on these sites. Plant material was
oven-dried at 60 °C for 48 h and weight of diy
matter determined. Leaves were separated
from leaders and weighed separateK' to deter-
mine leaf-to-leader ratios on a percent dry
matter basis. Leaves and leaders were then
ground to approximateK 1 mm diameter in a
grinder (Janke & Kunkel kg, t\pe AlO). Kjel-
dahl (nitrogen) analyses were used to arrive at
crude protein contents. Winter crude protein
values were calculated with a weighted aver-
age of winter leaf and leader protein levels.
This allowed comparison with other studies
(Dietz et al. 1962, Welch et al. 1983).
1996]
Variation in Buterbrusii Crude Protein
207
Soil samples were obtained at a depth of 15
cm below the surface fiom a soil pit in each
study plot. Because soils at most sites con-
tained a large rock fiaction, it was necessan' to
sample at the relatixeK shallow depth of 15 cm
to standardize sampling. The Montana State
Unixersity Soil Test Laboratory performed
organic matter determinations and total Kjel-
dahl nitrogen analyses on all non-MHWMA
samples. Texture was determined by both the
h> drometer and Bouyucous mechanical analy-
sis methods. Soil pH was determined in 1 part
soil to 2 parts water extractions. Topographic
information was also recorded at each site.
Aspect was determined by taking a compass
bearing from the major slope. Slope was mea-
sured widi a clinometer. Elevation was deter-
mined from USGS topographic maps. The
information from MHWMA sites was derived
from Guenther (1989).
A one-way ANOVA, with site as the factor,
was conducted for each sampling date and pro-
tein source combination (Snedecor and Coch-
ran 1989). This was done with the knowledge
that protein sources (leaves or leaders) con-
tained veiy different levels of crude protein
within each sampling date. Site was also the
factor in an ANOVA for percent leafiness for
the Febiaiaiy 1991 sample. The least significant
difference (LSD) method (F < 0.05) protected
by a prior F-test (F < 0.05) was used for com-
paring treatment means (Snedecor and Cochran
1989).
Results and Discussion
Crude protein levels differed (F < 0.001)
among sites within each protein source and
collection date combination (Fig. 1). Thus, we
rejected the hypothesis that crude protein val-
ues are equal during August and Februaiy
among local stands of bitterbnjsh. Crude pro-
tein in the leaves, when averaged over all sites,
varied with a 13% to 10% decline from August
1990 to FebiTjaiy 1991 and subsequent increase
to 15% by August 1991. Crude protein in the
leaders for these 3 dates was 7.1%, 6.5%, and
7.2%, respectively, when averaged over all
sites. These crude protein levels generally
agreed with previous reports for bitterbrush
throughout its range (Dietz et al. 1962, Bay-
oumi and Smith 1976, Morton 1976, Tiede-
mann 1983, Welch et al. 1983).
Protein levels also differed (F < 0.001)
among the 3 collection dates (Fig. 1). When all
sites were pooled, August leaf protein increased
11% between years (F < 0.05) and February
leaf protein decreased 21% from August levels
(F < 0.001). Leader crude protein did not vary
significantly between years but was higher in
August 1991 than during the previous Febni-
aiy (F < 0.05).
The unbrowsed Butte site rated highest in
crude protein (Fig. 1) for 3 of the 6 measure-
ments, although none was significantly higher
than the next lower site. When the Butte site
was compared to the aggregated crude protein
levels of the other 8 sites, it was significantly (F
< 0.05) higher for both leaves and leaders in
August 1990, but did not differ from browsed
sites in Febiaiar)' or August 1991. Thus, it does
not appear that browsing affects crude protein
levels.
Protein values for the 4 MHWMA sites were
lower for August 1990 leaves (F < 0.07) and
leaders (F < 0.01) than for other sites and col-
lectively rated lowest for 4 of the 6 measure-
ments. These site differences were not expected
fi-om Morton's (1976) work, but were supported
by that of Giunta et al. (1978) and Welch et al.
(1983).
Bitterbrush crude protein levels on the deer
+ cattle site were 1% higher (F < 0.05) than
on the adjacent cattle exclosure site for August
1990 leaves (Fig. 1). Other protein levels did
not differ significantly between these 2 sites.
Although a difference in use might thus seem
to affect protein levels on these sites, the
unbrowsed Butte site had higher protein levels
than browsed sites in August 1990 (F < 0.05)
and no difference in February or August 1991.
Related to these site and possible popula-
tion (Alderfer 1977) differences are soil differ-
ences. Soil samples from shrub interspaces
(Table 2) contained 49% more soil nitrogen at
the Butte site than at the burn and unburn
sites and 78% more than at the cattle exclosure
and deer + cattle sites. Bayoumi and Smith
(1976) found a positive response of bitterbrush
protein levels to fertilization with nitrogen,
although Tiedemann (1983) found slighdy neg-
ative to no response to fertilization. However,
most desert shmbs accumulate nutrients under
their canopies, and the surrounding interspaces
have low nutrient content (Garcia-Moya and
McKell 1970, Tiedemann and Klemmedson
1973), conditions that we did not sample.
208
Great Basin Naturalist
[Volume 56
Leaves
Rajiroad Gulch
Willow. Creek
Powerlirle
Un&urn/
BurrT
CAitjepfid Deer
jCatlle Exclosure
Aug. 90
Aug.
Leaders
ajkoa^ Gulch
ow Creek
rliPie
Burn V
Cqltle^hd Deer
_Cattle Exclosure
BuUe_Z
Aug. 90
Feb. 91
Aug. 91
Fig. 1. Average percent crude protein in I)itterhnish Iea\es and leaders found in August 1990, Februaiy 1991, and
Augirst 1991. Protein values within eacli protein source and collection date with similar lowercase letters are not signifi-
cantK' diflercnt (LSD, P > 0.05). Insufficient leaf material was available for statistical analysis in Febniar>' 1991.
1996]
Vauiatiun in Buterbuush Crude Piu)ii:inj
209
Tablk 2. Soil charactt-ristics for stuch' areas, including pll, organic matter (OM), total Kjeldahl nitrogen (N), percent
sand, silt, and clay, and textiiral class. Soil nitrogen was sampled at onl\' 5 sites. Data from the last 4 sites were oI)tained
from Cinenther (1989).
Site
pll
OM
N
Sand
Silt
C:lay
'le.xtural class
(%)
(%)
(%)
(%)
(%)
Butte
5.6
2,6
0.11
63
24
13
sandy loam
Cattle e.xclosure
5.7
1.0
0.06
80
12
8
loamy sand
Cattle + deer
5.7
1.0
0.06
80
12
8
loamy sand
Burn
6.3
1.1
0.07
67
23
10
sandy loam
Unburn
6.3
1.1
0.07
67
23
10
sand}' loam
Powerline
5.8
1.4
—
65
15
20
sandy loam
Willow Creek
5.2
3.6
—
65
18
17
sandy loam
Railroad Gulch
5.7
1.0
—
72
18
10
sandy loam
High R> e
6.7
2.8
—
69
15
16
sand) loam
Protein levels at our study sites were therefore
not necessarily related to soil nitrogen levels.
Although most leaves had fallen by Febru-
aiy, all sites contained plants that had retained
some leaves at that time. Bitterbrush phenol-
ogy seems to vaiy more by season and climate
than b\' ecot>pe (Shaw and Monsen 1983). Most
leaves are deciduous, dropping in response to
moisture stress in late summer or fall (Shaw
and Monsen 1983), but some small leaves over-
winter on some populations (Alderfer 1977).
Dietz et al. (1962) alluded to the high protein
level of leaves in winter but did not quantify
those levels. Welch et al. (1983) reported that
winter leafiness (presumably, weight of leaves
compared with weight of stems) of plants from
Idaho, Colorado, Utah, and California ranged
from 5.9% to 15.5%, while combined leaf and
leader crude protein ranged from 5.9% to
7.9%. These ranges are similar to values found
for these Montana sites: leafiness (percent
weight of leaves per weight of stems) of 1.5% to
15.8% and combined cnide protein of 6.1% to
7.6% (Table 3). Because so few leaves were
present in Febiaiaiy (Table 3), crude protein in
total foliage increased by only 0.3% over twigs
alone for all sites.
Although we concluded that leaves contain
significantly more crude protein than leaders
on our study sites, leaf scarcity during winter
in our region prevents total (leaf and leader)
crude protein from meeting deer requirements
(Welch et al. 1983). The February crude pro-
tein levels for total foliage averaged 6.8% across
sites, which were below the estimated neces-
saiy direshold of 8.9% for wintering deer (Welch
et al. 1983). However, September through
November protein levels might have been
higher, as many plants retained leaves through
that period.
Guenther (1989) reported that deer pellets
from the MHWMA sites contained large
amounts of Rocky Mountain juniper {Jiiniperus
scopiilonou Sarg.) and Oregon grape {Bcrheris
repens Lindl.). Protein values for small winter
samples of Oregon grape and juniper from the
Willow Creek site were 8.4% and 6.9%, re-
spectively. These values are below those re-
ported by Welch et al. (1983) and, like bitter-
brush, are also below what they considered to
be the necessary threshold of 8.9% crude pro-
tein for wintering deer. Hamlin and Mackie
(1989) suggested that mule deer have more
need for high-quality forage in the fall, while
building energy reserves, than in the winter.
Bitterbrush in southwestern Montana may
supply this needed level of nutrients in the fall,
as we observed delayed leaf-fall on wind-pro-
tected bitterbiTish plants in late November 1990,
but we did not sample plants at that time.
Restoration efforts for ungulate winter
ranges capable of maintaining bitterbrush ma\
benefit through consideration of our results.
We have found that bitterbiiish populations of
even a localized ecotype, such as we studied,
should not be expected to attain the same lev-
els of crude protein over different environmen-
tal conditions that will var>' between sites.
Revegetation of bitterbrush ranges will involve
consideration for obtaining the best possible
plant materials. Our evidence indicates that
plant characteristics, other than protein con-
tent, should likely be of primar\' concern as
protein can be expected to van' b\' site condi-
tions regardless of plant material. However, it
appears that consideration should be made of
bitterbmsh genot>q3es that maintain a high per-
centage of leaves into the winter These geno-
types may provide a higher level of crude pro-
tein that is desirable for \vintering ungulates.
210
Great Basin Naturalist
[Volume 56
Table 3. Winter cnide protein content (percent) of bit-
terbrush leaves and leaders combined and percent leafi-
ness (percent weight of leaves per weight of stems) for
study sites sampled Februaiy 1991. Column entries with
similar letters are not significantly different (LSD, P <
0.05).
Site
Crude protein
Leafiness
{%)
(%)
Butte
7.fr'
I3.(t'1'
Cattle e.xclosure
6.4"!
6.5'l
Catde + deer
6.6^'
9.9^-
Bum
6.7l'^-
9.6«l
Unburn
7.6«
10.5'^^
Powerline
6.1^
8.5"'
Willow Creek
7.lah
8.1"!
Railroad Gulch
7.2^>
15.8"
High Rye
6. Id
1.5^
Literature Cited
Alderfer, J. M. 1977. A ta.\on()mic study of bitterbrush
{Piirshia tridentata [Pursh] DC.) in Oregon. Unpub-
lished master's thesis, Oregon State University, Cor-
vallis.
B.4,YOUMi, M. A., AND A. D. Smith. 1976. Response of big
game winter range vegetation to fertilization. Journal
of Range Management 29: 44-48.
DiETZ, D. R. 1972. Nutritive value of shrubs. Pages
289-302 in C. M. McKell, J. P Blaisdell. and J. R.
Goodin, editors, Wildland shrubs — tlieir biology and
utilization. USDA Forest Sei-vice, General Technical
Report INT-1, Ogden, UT.
DiETZ, D. R., R. H. Udall, and L. E. Yeager. 1962.
Chemical composition and digestibility' by mule deer
of selected forage species. Cache la Poudre Range,
Colorado. Colorado Game and Fish Department
Technical Publication 14.
Fraas, W. W, C. L. Wambolt, and M. R. Frisina. 1992.
Prescribed fire effects on a bitterlinish-mountain big
sagebrush-bluebimch wheatgrass community. Pages
212-216 in W. P Claiy, E. D. McArthur, D. Bedunah,
and C. L. Wambolt, compilers, Proceedings of the
Symposium on Ecolog\' and Management ot Riparian
Shnib Communities. USDA Forest Senice, General
Technical Report INT-289, Ogden, UT.
Garcia-Moya, E., and C. M. McKell. 1970. Contribution
of shrubs to the nitrogen economy of a desert-wash
plant community. Ecology 51: 81-88.
Giunta, B. C, R. Stevens, K. R. Jorcensen, and A. P
Plummer. 1978. Antelope bitterbrush: an important
wildland shrub. Utah Division of Wildlife Research
Publication 78-12.
GUENTHER, G. E. 1989. Ecological relationships of bitter-
brush communities on the Mount Haggin Wildlife
Management Area. Montana Department of Fish,
Wildlife, and Parks and Montana State University
Department of Animal and Range Sciences, Boze-
Hamlin, K. L., and R. J. Mackie. 1989. Mule deer in the
Missouri River Breaks, Montana. Montana Depart-
ment of Fish, Wildlife, and Parks.
Menke, J. W, and M. J. Trlica. 1981. Carbohydrate
reserve, phenology, and growth cycles of nine Col-
orado range species. Journal of Range Management
34: 269-277.
Morton, M. A. 1976. Forage relationships of mule deer in
the Bridger Mountains, Montana; nutritional values
of important mule deer winter forage plants in the
Bridger Mountains, Montana. Unpublished master's
thesis, Montana State University, Bozeman.
Mueggler, W, and W. L. Stewart. 1980. Grassland and
shnibland habitat types of western Montana. USDA
Forest Service, Intermountain Forest and Range
Experiment Station, General Technical Report INT-
66, Ogden, UT.
NOAA. 1991. Climatological data, Montana, 84-94(1-13).
National Climatic Data Center, Asheville, NC.
Shaw, N. L., and S. B. Monsen. 1983. Phenology and
growth habits of nine antelope bitterbrush, desert
bitterbrush, Stansbur\' cliffrose, and Apache-plume
accessions. Pages 55-69 in A. R. Tiedemann, and
K. L. Johnson, compilers. Proceedings of the Research
and Management of Bitterbrush and Cliffrose in
Western North America. USDA Forest Sen'ice, Gen-
eral Technical Report INT-152.
Shepherd, H. R. 1971. Effects of clipping on key browse
species in southwestern Colorado. Colorado Game,
Fish, and Parks Di\ ision, GFP-R-T-28, Denver, CO.
Slausen, W L., and R. T. Ward. 1986. Ecogenetic pat-
terns of four shrub species in semi-arid communities
of northwestern Colorado. Southwestern Naturalist
31: 319-329.
Snedecor, G. W, and W G. Cochr.\n. 1989. Statistical
methods. Iowa State Universit>- Press, Ames.
Tiedemann, A. R. 1983. Response of bitterbrush and asso-
ciated plant species to broadcast nitrogen, phospho-
rus, and sulfur fertilization. Pages 240-253 in A. R.
Tiedemann, and K. L. Johnson, compilers. Proceed-
ings of the Research and Management of Bitterbrush
and Cliffrose in Western North America. USDA For-
est Sei-vice, General Technical Report INT-152.
Tiedemann, A. R., and J. O. Klemmedson. 1973. Effect of
mesquite on physical and chemical properties of die
soil. Journal of Range Management 26: 27-29.
Welch, B. L., S. B. Monsen, .and N. L. Sh.aw. 1983.
Nutriti\e \alue of antelope and desert bitterbrush,
Stansbur\ cliffrose, and Apache-plume. Pages
173-185 in A. R. Tiedemann, and K. L. Johnson,
compilers. Proceedings of the Research and Manage-
ment of Cliffrose in Western North America. USDA
Forest Service, General Technical Report INT-152.
YouTiE, B. A., B. Griefith, .\nd J. M. Peek. 1988. Succes-
sional patterns in bitterbrush habitat t\pes in north-
central Washington. Journal of Range Management
41: 122-126.
Received 1 November 1995
Accepted 1 Maij 1996
CiL-at Basin Naturalist 56(3), © 199fi, pp. 211-224
DAM-FORMING CACTI AND NITROGEN ENRICHMENT IN A
PINON-JUNIPER WOODLAND IN NORTHWESTERN ARIZONA
M()ll\ rhoinas Ihsrll' and Charles C. (irier^
Abstract. — In a pinon-junipcr woodland in nortliwi'stcrn Aii/ona, Lonncctcd basal cladodcs of a prickK pt-ar cactus
{Opuntio Jittoralis var. martiniaiw) ionn check dams (hat cause deposition ol N-rich detritus in interspaces otherwise
lackinu litter Seventy-eight percent of connected hasal eladodes measured in transects grew at an angle (w ith respect to
till' slope contour) < 45° — an orientation facilitating tleposition of flood-horne itehris.
Soil total N was significantly greater {F < 0.01) and organic C was greater, hut not significantly, a!)o\'e cactus dams
compared to helow cactus dams. Soil total N and organic C both above and below cactus dams were significantly greater
{P — 0.0001) compared to adjacent interspaces. Soil total N and organic C above cactus dams were equal to areas
beneath canopies (tree and shrub combined). Net NO3 (0-5 em depth) above cactus dams was significantly greater (P =
0.0001) than below cactus dams, at interspaces, and beneath canopies. Net NH4 (0-5 cm soil depth) above cactus dams
was significantly greater {P < 0.01) than below cactus dams and interspaces, and was greater (but not significantly) than
beneath canopies. At 5-10 cm soil depth, differences in net NH4 and net NO3 between sampling locations were not
significant except for the difference in net NO3 above and below cactus dams {P < 0.05). The litter layer above cactus
dams had twice as much total N (P < 0.01) as the litter layer beneath canopies (tree and shrub combined); differences in
net mineralized N were not significant between litter layers. Over the course of a single rainy season, detritus depth
behind cactus dams increased up to 23 cm, with a mean increase of 4.3 cm (sj — 0.625, P = 0.0001).
Key words: prickhj pear cactus, nitrogen enrichment, growth habit, soil characteristics, check dams, detriltis, runoff,
bulk density, total nitrogen, organic carbon, mineral nitrogen, pinon-jtmiper woodlands, islands of fertility.
The growtli habit of Opuntia littoralis var.
martiniana (L. Benson) L. Benson consists of
connected basal eladodes growing across wood-
land slopes roughly along the contour Clad-
odes in contact with the ground sprout adven-
titious roots and become anchored. Sequentially
anchored eladodes fiuiction as check dams dur-
ing runoff events, causing deposition of flood-
borne detritus including surface soil, animal
feces, and litter of piiion pines, juniper, and
oak.
Piiion-juniper woodlands occupy at least 17
X 10*^ ha in the western U.S., with widespread
distribution in Colorado, New Mexico, Arizona,
eastern California, Nevada, and Utah (West
1988). These woodlands fall between mesic
conditions that support closed-forest canopies
and arid conditions in which plants are widely
spaced. Compared with forests of wetter envi-
ronments, pinon-juniper woodlands have low
biomass, leaf area, and primary productivity
(Crier et al. 1992). Woodland structure varies
but can generally be described as single trees
and shrubs and clumps of trees and shrubs sur-
rounded by a network of interspaces (Lanner
1981). Litter occurs in patches due to the non-
contiguous canopy cover, and soil N distribu-
tion corresponds to litter and canopy distribu-
tion (DeBano and Klopatek 1987, Tiedemann
1987). In mixed-species stands, patches may be
mosaics of different litter components.
Interspace and canopy area soils usually dif-
fer in characteristics such as concentrations of
nutrients, pH, bulk density, soil water, and in
numbers and species of resident microorgan-
isms and microarthropods (Everett and Shar-
row 1985, Klopatek 1987, Klopatek and Klo-
patek 1987), although there are exceptions to
this generalization (DeBano et al. 1987). Soil
organic matter and nutrients are concentrated
near the soil surface (West and Klemmedson
1978, Lyons and Gifford 1980, DeBano and
Klopatek 1987), and runoff from storms can
cany considerable amoimts of detritus rich in
organic matter and N (Fletcher et al. 1978).
Objectives of this study were (1) to charac-
terize the angle of growth (relative to slope con-
tour) of connected basal eladodes of Opuntia
littoralis van martiniana, (2) to compare litter
and soil properties above and below cactus
'Department of Forest Resources, College of Natural Resources, Utah State University, Logan, UT 84321.
-Department of Forest Science, Colorado State Universit\; Fort Collins, CO 8052.3.
211
212
Great Basin Naturalist
[Volume 56
Table 1. Sites of measurement of angle (relative to the slope contour) of connected basal cladodes. (Samples for soil
comparisons were taken only in the Hualapai Mountains [see Table 2].)
Transect
Soil
Piiion pine
length
parent
Elevation
Aspect
Slope
cover
Location
(m)
material
(m)
(%)
(%)
Cerbat Mountains''
401
granite
1930
S-SSE
10-25
10-30
Hualapai Mountains''
110
granite
1524
N
15-45
30-40
Music Mountains^
302
vesicular
basalt
1712
E
30
40-70
«27 km NW of Kingman, AZ (Iat.35''27', long.ll4°09'; T24NR18\VS23nw).
'n2 km SE of Kingman, AZ {lat..35''08', long.ll3°5.5'; T20NR16WSlsw).
^^^53 bii NE of Kingman, AZ (kit..35°41', long.ll3°49'; T27NR36WS.36ne).
dams, and (3) to compare litter and soil proper-
ties above and below cactus dams with inter-
spaces and areas beneath canopies.
Methods
Two distinct physiographic provinces come
together in northwestern Arizona: southeast,
west, and north of Kingman, Arizona, is the
Basin and Range Province, characterized by
north-trending fault-block mountain ranges
separated by broad desert valleys; the Col-
orado Plateau lies to the east. This area is the
interface of 3 deserts as well as a physiographic
interface. North of Kingman is the Great Basin
Desert, west is the Mojave Desert, and south-
west is the Sonoran Desert. The climate of
northwestern Arizona is semiarid (Sellers and
Hill 1974). Precipitation is bimodal, occuning
mostly in winter and summer months, with
more rainfall during winter than summer.
Summer rain sometimes occurs as intense
thundershowers (Sellers and Hill 1974).
We first observed dam-forming cacti in the
Hualapai Mountains (rising to over 2438 m,
12 km southeast of Kingman, Arizona) in the
course of data collection for studies of piiion-
juniper woodland productivity'. We subse(iuently
visited 2 nearby ranges (the Cerbat Mountains
[over 2133 m at highest point] 29 km northwest
of Kingman, and the Music Mountains [over
2011 m] 53 km northeast of Kingman) and
found dam-forming cacti in these locations. To
characterize die angle of growth of connected
basal cladodes of prickly pear cacti (our 1st ol)-
jective), we took angle measurements in July
1991 on all cacti intercepting straight-line tran-
sects in the 3 mountain ranges (Table 1). Start-
ing points of line transects were randomK' lo-
cated, and direction of transects was along ran-
dom azimuths. A total of 233 angle measure-
ments were recorded. Sequentially connected
Fig. 1. Mcasurcineut of angle ol growth ol couuectetl basal cladodes with respect to the slofX' contour. Point of origin
ndicated In' solid cladode.
1996]
Dam-fohming Cacti and Nithogen Enrichment
213
basal cladodes with series ranging from 0.4 ni
to 2.5 m in length were nieasnred with an
engineer's acljnstable triangle as shown in Fig-
ure 1: a direction of growth parallel to slope
contour was 0° while a direction of growth per-
pendicular to slope contour either upslope or
downslopc, was 90°.
Soil and Litter Sampling
Site description. — We restricted litter and
soil sampling to 1 of the 3 transect locations
(the Hualapai Mountains, 12 km southeast of
Kingman [Table 2, Fig. 2]), to minimize con-
foimding factors such as different soil types,
site histories, and land-management practices.
About 40% of the study site is open interspaces
(combined data [unpublished] from eighteen 2
X 2-m plots using Daubenmire s [1968] cover-
age classes, and from 12 permanently marked
25-m-long line transects using methods de-
scribed in Meeuwig and Budy [1981]). Inter-
spaces are mostly bare soil and rock surface,
with 3% grass cover (mostly Boiiteloua gracilis
[H.B.K.] Lag. ex Steudel and B. curtipendiila
[Michx.] Torr.) and traces of litter, herbs, and
ciyptogams. Shrubs, mostly scrub oak {Qiier-
ciis iurbinella Greene), cover about 30% of the
study area. Pinon pines {Piniis monophylla war.
falhix [Little] Silba) cover about 36% of the
area and Jiiniperiis osteosperma (Torr) Little
about 4%. The added cover of vegetation com-
ponents is greater than total vegetation cover
due to the presence of different vertical layers
of shrub and tree canopies and aggregation of
vegetation in clumps. Trees ranged in age from
seedlings to about 260 w (estimated fiom annuiil
ring counts of cores [unpublished data]). Age
estimates are approximate due to occurrence
of false rings in wood of pifion pines and
junipers.
Size range of soil surface patches covered
by cacti and associated litter accumulations
was estimated by measuring eveiy cactus dam
on a 25 X 25-m plot. We recorded length, widtli,
and circumference for each cactus dam and
associated litter accumulation (32 total). The
area of soil surface covered by cactus dams and
litter was calculated as the area of a circle plus
1/2 the difference between the area of a rectan-
gle and a circle.
Soil and litter sampling approach. —
Sampling was stratified by woodland micro-
habitats: above cactus dams, below cactus dams,
interspaces, and beneath canopies. We took
Tahlk 2. (^Iiaractoristic'S of litter and soil saniplinf^ site
ill a pinoii-jiniipcr woodland in tlic Hualapai Mountains of
noitliwt'stern Arizona. Records (1967-1991) of licensed
livestock grazing show year-round grazing of cattle and
horses with year-to-year variation in season of heaviest use
and in number of animals (USDA BLM 1991).
Elevation: 1524 ni (5000 ft)
% slope-; 20-40
Aspect; north
Soil parent material; granite
Soil texture: sandy-loam
Soil classification'': Barkerville Series\
loamy, mixed, mesic, shallow\
Udorthentic liaplnstolls
Other soil characteristics'":
Al horizon 10 cm deep, 39% coarse frag.
pH surface soil interspace — 6.5
pH surface soil under canopy — 8.0
non-calcareous throughout
Species and % cover"^:
Pimi.s monophylla suhsp.jallax
Jiinipcrus osteosperma
Qtiereiis fitrbinella
Yueca hacatta
Opimtia littoralis var. iiuiiiiniana
RJiiis trilohata
Ceanothiis greggii
Canotiu holocantha
Bouteloua gracilis
Gutierrizia sarothrae
36.0%
«f
5.7
4.0%
«f
1.6
30.0%
Sy
6.2
4.0%
Sy
LI
L9%
Sy
1.1
0.7%
AY
0.4
0.4%
«r
0.2
0.3%
«r
0.3
2.9%
«r
1.3
<1%
"Richmond and Richardson (1974).
''Unpublished data, this study.
'Two methods were used to estimate cover: For all species, estimates were
made on eighteen 2-m2 plots according to coverage class ratings (Daubenmire
1968). Tree and shrub cover were estimated on 12 permanent 25-m line tran-
sects as % Cover = [(25*.3.14/Transect Length)] [Sum of crown diameters]
(Meeuwig and Budy 1981). Values reported here for trees and shrubs are aver-
ages of both methods, and standard eiTors are from pooled variances.
paired samples 10.2 cm above (litter present)
and below (little to no litter present) cactus
axes to compare soil properties above and
below cactus dams. We took additional sam-
ples from bare interspaces and from areas
beneath tree and shrub canopies to compare
these areas with the areas above and below
cactus dam. Interspaces were considered to lie
beyond the influence of canopies and associ-
ated litter and beyond the influence of cactus
dams and associated litter. Vegetation and litter
were scant to absent in interspaces. Beneath
tree and shrub canopy, sampling included
pinon pines, scrub oaks, junipers, and occa-
sionally mixed-species canopies roughly in pro-
portion to the presence of these components
(as estimated by percent canopy cover) on the
site (Table 2). The sampling location beneath
canopies was at 2/3 canopy radius out from the
stem or clump center Litter of Yucca haccata
214
Great Basin Naturalist
[Volume 56
Fig. 2. Soil and litter sampling area in the Hiialapai Mountains of northwestern Arizona. The contour intenal is 12.2 in
(40 ft). Enlarged from U.S. Geological Sunex', Rattlesnake Hill. Arizona Quadrangle.
Torr. and a few other species was occasionally
(though rarely) present in litter samples along
with litter of the dominant species. With the
exception of hulk density samples, soil and lit-
ter samples were composited within microhah-
itat strata by combining equal umubers of equal-
sized indix'idual samples. Compositing followed
guidelines in Peterson and Cabin (1986) and
was suitable for the present study since we
were not examining variation within nucrohab-
itats. As pointed out b\' Crepin and Johnson
(1993), composite sampling can be used in
1996]
Dam-I()Kmi\g Cacti and Nitrogen Enrichment
215
conjunction with stratification: i.e., the hmd-
scape can be cli\ ided into meaningful units and
good averages of soil properties obtained b\
compositing samples within each unit. All soil
and litter sampling was conducted in Jul\ 1991.
Bulk density. — Bulk density was deter-
mined by the exca\ation method (Blake and
Hartge 1986). T\vent>'-t\vo paired samples weie
taken 10.2 cm al)o\e an'' below cactus axes, 10
samples were taken fiom '"'^erspaces, and 10
were taken from beneath tree and shrub
canopies. Soil was e.\ca\"ated with a bulb planter
(diameter 5.5 cm at cutting edge), creating a
hole 7 cm deep. A thin, tough plastic bag was
placed in tlie hole, filled witli water, and then
emptied into a graduated c>'linder to deter-
mine hole \olume. Extracted soil was dried at
105° C and weighed, resulting in a weight-to-
\ olume measurement.
Total N, total organic C, .\nd soil tex-
ture.— Thirt)' pairs of soil cores (mineral soil
surface to 7 cm deep) were extracted with a
bulb planter (diameter 5.5 cm at the cutting
edge) adjacent to cactus axes (10.2 cm above
and below cactus axes), 30 li-om beneath cano-
pies, and 30 from interspaces. Samples were
taken near each of the 6 satellite plots estab-
lished for the net mineralization stud\ (see
below). Litter (all litter from surface to mineral
soil) was retained for detemiination of total N.
Samples were air-dried and stored in paper
wrappers. Soil samples originalK' taken for
determining bulk density (see above) were
added to these soil samples for a total of 51
samples from each side of cactus dams, 40
samples from beneath canopies, and 40 from
interspaces. One of the 22 paired bulk densit)
samples was lost and could not be included.
Samples were combined to create compos-
ites: above cactus dams 51 samples of soil were
composited to make 3 samples of soil, and 30
samples of litter w ere combined to make 3 lit-
ter samples. Below cactus dams (no litter pre-
sent) 51 samples of soil were composited to
make 3 soil samples. Beneath canopies 40 soil
samples were composited to make 3 samples of
soil, and 30 litter samples were composited to
make 3 litter samples. From interspace areas
(no litter present) 40 samples were composited
to make 3 samples of soil. AnaKsis was b>- Utah
State Universit}' Soils Testing Lab following
the Kjeldahl method (Bremner and Mulvane\
1982) to detennine percent total N, the Walk-
le\-Black method (Nelson and Sommers 1982)
for percent organic C, and methods described
b\ Gee and Bander (1986) for particle-size
analysis.
Net mineralized N.— The total amount of
N liberated from organic matter is "gross min-
eralization"; the quantity remaining after micro-
bial immobilization is "net mineralization" (Car-
lyle 1986). Net mineral N, the N available for
plant uptake, is an index of soil fertilitv'. To com-
pare soil N fertilit)' among woodland sites, net
mineral N was assessed by laboratorx' aerobic
incubations (Binkle\- and Vitousek 1989).
Seven pemianent plots were created on the
study site, the 1st plot serving as a central
point fiom which 6 satellite plots were created,
each 32 m fiom tlie central point at 60° inteniils
beginning with a random azimuth. Because of
topography, 1 plot was relocated 32 m from the
center of a satellite plot. From each plot center
8 cacti (0.5 to 5 m IroiP center) were selected
at 45° intervals beginnmg with a random
azimuth, for a total sample of 56 cacti.
Paired soil samples were taken 10.2 cm
from cactus axes on all 7 permanent plots be-
ginning li-om the easternmost cactus and mov-
ing clock-wise. Samples were composited com-
bining 4 individual samples into 1 composite
sample. Compositing and field processing (see
below) were perfomied immediatcK- upon the
extraction of 4 cores. For example, on the 1st
plot 4 cores 10.2 cm above cactus axes in the
90° -270° hemisphere of the plot were taken,
composited, and field processed before the
next 4 cores were drawn. This ensured pro-
cessing fresh soil. Fourteen composite sample
pairs were prepared.
At approximateh" midpoints of the six 32-m
lines creating satellite plots, 2 samples were
taken beneath canopies (piiion pines sampled
most heavily followed by scrub oak, mi.xed-
species canopies, and juniper) and 2 fiom inter-
spaces. Composites of 4 individual samples
were prepared and field processing completed
immediateK' as each set of 4 cores was drawoi.
Three composite samples were prepared.
Samples were taken with a 2-cm-diameter
soil corer to a depth of 10 cm. Preparation of
samples for anabsis followed methods outlined
in Vitousek et al. (1982): In the field cores were
divided into 3 components (litter layer, top 5
cm of mineral soil, and mineral soil between 5
and 10 cm soil depth) and composited. Com-
posite soil samples were sieved through a 2-
216
Great Basin Naturalist
[Volume 56
30-39 40-49 50-59 60-69 70-79 80-90
degrees
Fig. 3. Angle of growth of connected basal cladodes with
respect to slope contour Zero degrees is a direction of
growth parallel to the slope contoin-; 90 degrees is a direc-
tion of growth peipendicular to the slope.
Table 3. Size distribution of cactus dams and associated
litter accumulations on a 25 X 25-m plot. The area mea-
sured was the soil surface co\'ercd b\ cactus dams and
associated litter
Size class
(m2)
0.05
0.1-1.0
1.1-2.0
2.1-3.0
3.1-4.0
10.30
Number of
cactus dams
1
16
6
6
2
1
mm screen; litter was not sieved. Subsamples
were sealed in bags for detemiination of mois-
ture content, while a 2nd subsample of approx-
imately 10 g was placed in 100 ml 1 N KCl
adjusted with HCl to pH 2.5 with phenylmer-
curic acetate (PMA) added as a presei-vative.
Solutions were refrigerated, transported to the
laboratory, mixed frequently for 4 d, then
allowed to settle for 48 h. After settling, the
solution was removed with a pipette, and
NH4"^ and NO3 were determined at Bilby
Research Facilit>' at Northern Arizona Univer-
sity using methods described by Keeney and
Nelson (1982).
The remainder of composited field samples
(after removal of the above 2 subsamples) was
transported to the laboratory and incubated
aerobically following procedures in Vitousek et
al. (1982): Soils were wetted to approximately
field moisture capacity (assessed visually),
placed in plastic-covered cups, each of which
had a small air hole, and kept in a dark, moist
chamber at a constant temperature of 22° C.
During an 8-wk incubation period, samples
received distilled water (applied as a fine mist
to the surface with no mixing) as needed to
maintain an approximately constant moisture
content. So as not to disturb incubating sam-
ples, moisture content was assessed by visible
soil color easily observable through the clear
plastic incubation cups.
At the end of 8 wk, subsamples (approxi-
mately 10 g) of incubated samples weie taken
for determination of moisture content, and
subsamples of approximately 10 g were placed
in the KCl solution described above. These
solutions were shipped to the soils testing lab-
oratoiy at Utah State Universit)' for detemiina-
tion of NH4+ and NO3" (U.S. EPA 1983).
Change in deposit depth. — Depth of de-
posits above cactus dams (i.e., above con-
nected basal cladodes) was measured before
(Jul)') and after (September) the rainy season of
1991 on 6 of the 7 plots designated for net min-
eralization sampling (see above). Two sampling
points could not be relocated at the end of the
rainy season, making a total sample size of 46
cactus dams (i.e., 6 plots, 8 cacti per plot, minus
2). Depth was measured from base to top of
deposits in the area of greatest accumulation.
Statistical Analysis
A heterogeneity chi-square analysis followed
by a chi-square anabsis (Zar 1984) was per-
formed with the 3 data sets of angle of cactus
growth from the 3 mountain ranges.
Soil and litter analyses. — Tests of nor-
mality were performed for each data set (above
cactus dams, below cactus dams, interspaces,
and beneath canopies) of each soil and litter
characteristic sampled. A paired t test (»= =
0.05) was used to compare means of soil char-
acteristics abo\ e and below cactus dams, and
to compare the depth of deposits at cactus
dams l)efore and after the rainy season. An
anabsis of variance F-test (oc = 0.05) for unbal-
anced sample sizes (the GLM procedure in
SAS software [SAS 1985]) was used to compare
sample means of soil abo\ e and below cactus
dams with beneath canop\ and interspace sam-
ple means. Plots of residuals were generated to
assess equalit) of variance. Significant differ-
ences between means were separated and
lanktxi using a nniltiple comparison method
1996]
Dam-fokminc Cacti and Nitroc;en Enrichment
21'
Table 4. Results of paired t tests comparing sample means of soil characteristics above and below cactus dams, and
comparing depth of detritus above cactus dams between early July and mid-September A minus B refers to tlu' value
alio\e cactus tlams minus the \alue below.
N
Pail
ed /-test statistics
.Utnbute
Mean difference
■''■.V
/
P
A minus B
Hulk di'nsit> (g/ml)
(natural log)
0-7 cm depth
22
-0.222
0.117
1.901
0.0711
Total N (%)
0-7 cm depth
3
0.043
0.003
13.000
0.0059
Organic C (%)
0-7 cm depth
3
0.933
0.231
4.035
0.0563
Net mineralized N
(ug/g)
0-5 cm depth NH4
14
9.621
2.353
4.088
0.0013
0-5 cm depth NO3"
14
41.979
5.398
7.776
0.0001
5-10 cm depth NH4''
12
2.050
2.063
0.993
0.3418
5-10 cm depth NO3"
12
7.550
2.672
2.826
0.0165
Change in depth (cm)
of detritus
aIio\'e cactus dams
(natural log)
0-7 cm depth
46
1.240''
0.124
10.038
0.0001
'^Mean difference of September detritus depth ininu.s JiiK detritus depth.
(REGWF) cited as being compatible with the
overall analysis of variance F-test (SAS 1985).
A t test (oc = 0.05) was used to compare sam-
ple means of total N and net mineralized N
fi'om the litter above cactus dams with the lit-
ter layer beneath woodland canopies.
Results
The pattern of angle of connected basal
cladodes with respect to slope contour was
similar in the 3 mountain ranges sampled; data
were pooled based on results of a heterogene-
ity chi-square analysis. Analysis of pooled data
(X" = 85.4, P < 0.001) indicated that orienta-
tion of connected basal cladodes of Opuntia lit-
toralis var mai-tiniana was nonrandom: growth
was most frequently parallel to the woodland
slope contour (Fig. 3). The size range of cactus
dams and associated litter on a 25 X 25-m plot
at the Hualapai Mountains study area is given
in Table 3.
Soil and litter analyses. — The null hy-
pothesis for normality was not rejected for
most of the data sets; however, total N data
were nonnormal and were not normally dis-
tributed when transformed with standard
transformations. Therefore, results of total N
analyses should be interpreted with caution.
Residual plots indicated equality of \'ariance
assumptions were reasonable.
Bulk density above and below cactus dams
was not significantly different at P = 0.05 (Table
4). Bulk density was significantly lower (F =
0.0001) in soil deposits above cactus dams,
below cactus dams, and beneath tree canopies,
compared to soil from interspaces (Table 5,
Fig. 4). Soil above and below cactus dams was
also lower in bulk density than soil beneath
tree canopies, although this difference was not
significant at F = 0.05. There was little differ-
ence in soil texture among the 4 microhabitats
(Tible 5).
Soil total N above cactus dams was greater
(F < 0.01) than below cactus dams {Table 4).
Organic C was not significantly different (F =
0.05) above cactus dams compared to below
cactus dams. Soil total N and organic C were
2-3 times greater (F = 0.0001 in both cases) in
soil above and below cactus dams than in
interspace soil (Table 5, Fig. 4). Soil total N and
218
Great Basin Naturalist
[Volume 56
E
A. Bulk density
F = 9.29
p = 0.0001
r
T
I X'''-
above below interspaces
cactus dams cactus dams
beneath
canopies
B. Total
F = 35.77
p = 0.0001
""d&'T"
above below
cactus dams cactus dams
interspaces
beneath
canopies
C. Organic C
above
cactus dams
below
cactus dams
interspaces
beneath
canopies
Fig. 4. Comparisons of soil characteristics above cactus dams, helow cactus dams, beneath canopies (tree and shrub
combined), and in bare interspaces.
organic C above cactus dams were equal to
areas beneatli canopies. Below cactiis dams, soil
total N was significantly lower than beneath
canopies, and organic C was not significantly
different compared to beneath canopies. While
soil organic C and soil total N differed among
woodland locations, die C:N ratio was similar
between locations (Table 5).
Net mineral NH4^ and NOg" at 0-5 cm
depth were significantly greater (F = 0.001
and P = 0.0001) above cactus dams compared
to below (Table 4). At 5-10 cm depth net min-
eral NO3 was significanth' greater (F =
0.0165) abo\'e cactus dams compared to below.
Net mineral N in soil 0-5 cm deep above cac-
tus dams was over 3 times that in interspace
Dam-i'()kminc Cacti and Nithocen Emuchment
219
D. NH4"^ 0-5 cm depth
■3
16
14
12
10
8
6
4
2
0
F^T^^rn
F = 4.93
p = 0.0067
above
cactus dams
below
cactus dams
interspaces
beneath
canopies
3
E. NO" 0-5 cm depth
90
80
70 H
60
50
40
30 H
20
10
0
F = 11.21
p = 0.0001
p-SfWiife^'fe^^^
below
cactus dams
interspaces
beneath
canopies
F. NH4"^ 5-10 cm depth
O)
5 -
4 -
3 -
2 -
1
0
F = 0.74
p = 0.5374
above below interspaces
cactus dams cactus dams
beneath
canopies
G. NO3" 5-10 cm depth
16 n
14 -
12 -
10 -
8 -
6 -
4 -
2 -
0 -
,-^
iiii
isjiiiiiSiiiiiS
F = 0.97
p = 0.4219
0:
=3
r
CO
0
z
Bil
„._
I. :
.!
„
above below interspaces beneath
cactus dams cactus dams canopies
Fig. 4. Continued.
220
Great Basin NatuR'\list
[Volume 56
Table 5. Comparison of sample means of soil characteristics at 4 woodland microhahitats (above and below cactus
clams, interspaces, and beneath canopies) at the Hualapai Mountains site. Superscript letters separate means signifi-
cantly different at °<: = 0.05. For te.xture, s = sand, si = silt, and cl — cla\. Samples are composites except for bulk den-
sity'. N = sample size and is followed in parentheses by the number of individual samples that were composited.
Attribute
Location
Above cactus dams
Below cactus dams
Interspaces
Beneath canopies
sandy loam
sandv loam
sandv loam
sandv loam
s 65.0
s 63.0
s 63.7
s 62.7
si 25.0
si 26.0
si 27.6
si 28.3
cl 9.0
cl 11.0
cl 8.7
cl 9.0
n = 3 (51)
n = 3 (51)
n = 3 (40)
n = 3 (40)
X = 0.99b
X = 1.13b
X = 1.85"
X = 1.40b
±0.11
±0.92
±0.10
±0.16
n = 22
n =22
/) = 10
n = 10
.T = 0.16^'
X = 0.12b
.V = 0.06^'
.V = 0.17"
± 0.006
± 0.003
± 0.007
±0.16
J! =3(51)
n = 3 (51)
n = 3 (40)
n = 3 (40)
X = 3.9''
.V = 3.0''
X = 1.5b
.V = 3.6-'
±0.24
± 0.07
±0.04
±0.26
n = 3 (51)
n = 3 (51)
n = 3 (40)
n = 3 (40)
Soil te.xture
% separates
kilk density
(g/ml)
Total N (%)
0-7 cm depth
Organic C (%)
0-7 cm depth
C:N ratio
Net mineralized N
(iig/g)
0-5 cm depth
NO,"
5-10 cm depth
NH4^
Nor
24.4
25.0
25.0
21.2
X = 14.8"
X = 5.3b
.V = 2.0b
.V = 11.5"
±2.6
±1.6
±0.2
± 1.3
X = 88.7"
.V = 46.7bc
X = 27. l^'
X = 54.2b
±6.9
± 5.9
±2.6
±6.4
n = 14 (56)
n = 14 (56)
n = 3 (12)
n =3(12)
X = 5.3
.V = 3.2
x =0.9
X = 4.5
±1.1
±1.8
±0.97
±1.5
.f = 31.6"
X = 24.0b
X = 16,6
.V = 30.9
± 4.0
±4.4
± 3.0
±8.2
u = 14 (.56)
n = 14 (.56)
/! =3(12)
n = 3 (12)
soil and almost twice that in soil beneath tree
canopies (Table 5, Fig. 4). Net mineral N below
cactus dams was greater than in interspaces,
but the difference was not statistically signifi-
cant.
Litter accumulated at cactus dams had total
N (0.74%) over twice as high as litter beneatli
tree and shrub canopies (0.32%) {t = -8.4, P =
0.01). NH4^ and NO^" in the litter layer were
greater beneath canopies than above cactus
dams, but not significantly (Table 6, Fig. 5).
From early July to mid- September, depth of
detritus behind cactus dams increased signifi-
cantly (P = 0.0001) from -2 cm to -1-23 cm, widi
an average of 4-4.3 cm (.v^ 0.625; Fig. 6).
Discussion
The similarit) of soil te.xture abo\e cactus
dams, below cactus dams, beneath tree and
shrub canopies, and in interspaces agrees with
findings of Schlesinger et al. (19(S9) that desert
soils receiving overland flow and adjacent soils
deprived of overland flow were similar in fine
material or cla\ content. The effects of cactus
dams and associated litter and detritus deposits
on bulk density, total N, organic C, and net
mineralized N of nearby soil were expected
based on a mnnber of studies in shrub lands
and woodlands documenting islands of fertility,
i.e., localized areas of nutrient enrichment
1996]
Dam-fouminc Cacti and Nith(x;i:n Enhichmi<:nt
221
I'aiu.K 6. Comparison of total \ and ni-t niinerali/.fcl N in tlit' littt-r la\t'r hcMU'atli tanopics with liltiT acciininlations above
cattns dams. N = sample size and is followed in pari-ntheses by tlie miniber of individnal samples that were eomposited.
A1)()\(
Beueatl
cactus d
lUlS
canopies
.Attribute
Mean
.S^T
Mean
■""■.v
Total N (%)
0.737
n = 3 (30)
0.047
0.320
n = 3 (30)
0.015
Net mineralized N
(ug/g)
NH4
60.233
n = 2^' (12)
15.018
84.550
n = 121' (48)
11.250
NO3"
-2.108
n =2(12)
6.410
40.050
n = 12 (48)
.34.950
-8.4275
1.2929
1.1865
0.0095
0.2426
0.4398
''Three composites were prepared; however, initial (before incubation) net mineral N values were not obtained lor 1 sample
''Values before incubation were not obtained for 2 of the original 14 composited samples.
(Garcia-Moya and McKell 1970, Tiedemann
and Klemniedson 1973, Baith and Klemmed-
son 1978, Baith 1980, Doescher et al. 1984,
Exerett et al. 1986, Garner and Steinberger
1989, Schlesinger et al. 1990).
Deposits at cactus dams ofOpiintia littoraUs
var. inariiniana raised soil total N from 0.06%
(interspace soil) to 0.16% above connected
basal cladodes and to 0.12% below (Table 4).
Nitrogen enrichment and soil amelioration
associated with deposits at cactus dams may
increase cactus productivity. Nobel et al. (1987)
observed that while annual aboveground pro-
ductivity of prickly pear cacti can be higli imdcr
optimal conditions, cacti productivity is often
limited by low levels of soil N (Nobel et al.
Litter total N and net mineral N
z .g
120
100
80
60 -
40
20
-20
p = 0.0095
total N
mi cactus dams
canopies
p = 0.2426
^i^
net NH/
p = 0.4398
net NO,
Fig. 5. Nitrogen in the litter accumulated above cactus dams compared with the litter layer beneath canopie.s (trees and
shrubs combined).
222
Great Basin Naturalist
[Volume 56
Q.
September
mean
depth 1 1 .4 cm
July mean
depth 7.1 cm
deposition at individual cactus dams
Fig. 6. Deposition at cactus dams during 1 season of summer thundershovvers; depths of detritus accumulations at 46
cactus dams in July and in September.
1987, Nobel 1989). Increased productivity in
desert prickly pear cacti is positively coirelated
with both number of new cladodes produced
and cladode size (Nobel et al. 1987). VVe do not
know if similar patterns occur in woodland
species of prickly pear. Additionally, Nobel
(1988) describes a tendency for "daughter" clad-
odes to replicate the orientation of "mother
cladodes and points out that if a particular
direction of growth is favorable, it may be per-
petuated. This happens because favorably ori-
ented cladodes are expected to be more pro-
ductive than other cladodes and produce more
and larger similarly oriented cladodes. This may
be occurring in dam-forming cacti, but it was
not investigated in this study.
Cactus dams lower soil bulk density and
enrich patches of woodland interspace with
organic matter, total N, and net mineral N, sug-
gesting that they may play roles in nutrient
cycling and other ecosystem processes. Some
possible functions of cactus dams are to (1) in-
crease woodland detritus storage, (2) increase
the rate of N turnover, (3) mitigate nutrient loss
in interspace areas, (4) reduce soil erosion and
dampen effects of disturbances, (5) provide
seedbeds, and (6) provide habitat for other
organisms.
Acknowledgments
We thank Ron Lanner and Helga Van
Miegroet for advice, encouragement, and assis-
tance. The senior author thanks Chuck Crier
for accepting an unconventional student and
for sharing his abundant cajoleiy and scientific
acumen.
Literature Cited
Barth, R. C. 1980. Influence of pin\ on pine trees on soil
chemical and ph\sical properties. Soil Science Soci-
ety' of America Journal 44: 112-114.
Barth, R. C, and J. O. Klemmedson. 1978. Shrub-
induced spatial patterns of dn' matter, nitrogen, and
organic carbon. Soil Science Society of America Jour-
nal 42: 804-809.
BiNKLEY, D., and E Vitousek. 1989. Soil nutrient avail-
ability. Pages 75-96 in R. W. Pearc\-, J. R. Ehleringer,
H. A. Mooney, and P W. Rundel, editors, Plant physi-
ological ecolog>'. Field methods and instrumentation.
Chapman and Hall, London/New York.
Blake, G. R.. and K. H. Hartce. 1986. Bulk density
Pages 363-375 in A. Klute, editor. Methods of soil
analysis, part 1: physiciil and mineralogical methods.
2nd edition. Soil Science Society of America, Inc.,
Madison, WL
Bremner, J. M., AND C. S Ml LVANEV. 1982. Nitrogen —
total. Pages 595-694 in A. L. Page et al., editors.
Methods of soil analysis, part 2. 2nd edition. Agron-
omy Monographs 9. American Societ}' of Agronomy
and Soil Science Societ\' of America, Madison, Wl.
1996]
Dam-forming Cacti and Nitrogen Enrichment
223
Carlyle, J. C. 1986. Nitrogen cycling in lorestccl ecosys-
tems. Forestiy Abstracts 47: 30(5-336.
Crepin, J., A.ND R. L. J()iiN.S()N. 1993. Soil .sampling for en-
vironmental assessment. Pages 5-18 ;/i M. \\. Carter,
editor. Soil sampling and methods of analysis. Lewis
Pnblishers, Boca Raton/Ann Arbor/London/ Tokyo.
D.\UBENMIRE, R. 196S. Plant conminnities: a textbook of
plant s\necolog>. Harper &: Row, New York.
DeBano, L. E, and J. M. Keop.\tek. 1987. Effect of man-
agement on nutrient dynamics in sonthwesteni pinxon
juniper woodlands. Pages 157-160 in C. A. Troendle,
M. R. Kanfmann, R. II. Hamre, and R. P Winokur,
coordinators. Management of subalpine forests: build-
ing on 50 \'ears of research. General Technical Report
RM-149. Rock> Mountain Research Station, Fort
Collins, CO.
DeBano, L. E, H. M. Perry, and S. T Overby^ 1987. Effects
of fiielwood harvesting and slash burning on biomass
and nutrient relationships in a pinyon-juniper stand.
Pages 382-386 in R. L. Everett, compiler. Proceed-
ings— pinyon-juniper conference. General Technical
Report INT-215. Intermountain Research Station,
Ogden, UT
Doescher, R S., R. E Miller, and A. H. Winward. 1984.
Soil chemical patterns under eastern Oregon plant
communities dominated by big sagebrush. Soil Sci-
ence Society- of America Jovmial 48: 659-663.
Everett, R. L., and S. H. Sharrow. 1985. Soil water and
temperature in hai"vested and nonhai"vested pinyon-
juniper stands. Forest Service Research Paper INT-
342. Intermoimtain Research Station, Ogden, UT.
Everett, R. L., S. H. Sharrow, and D. Than. 1986. Soil
nutrient distribution imder and adjacent to singleleaf
pin>'on crowiis. Soil Science Society' of America Jour-
nal 50: 788-792.
Fletcher, J. E., D. L. Sorensen, and D. B. Porcella.
1978. Erosional transfer of nitrogen in desert ecosys-
tems. Pages 171-181 in N. E. West and J. Skujins,
editors. Nitrogen in desert ecosystems. Dowden,
Hutchinson & Ross, Inc., Stroudsburg, PA.
Garcia-Moya, E., and C. M. McKell. 1970. Contribution
of shrubs to the nitrogen economy of a desert-wash
plant community. Ecology 51: 81-88.
Garner, W, and Y. Steinberger. 1989. A proposed mech-
anism for the formation of 'Fertile Islands' in the
desert ecosystem. Journal of Arid Environment 16:
257-262.
Gee, G. W, and J. W. Balider. 1986. Particle-size analysis.
Pages 383-411 in A. Klute, editor. Methods of soil
analysis, part 1: physical and mineralogical methods.
2nd edition. Soil Science Society of America, Inc.,
Madison, WI.
Grier, C. C, K. J. Elliott, and D. G. McCullough.
1992. Biomass distribution and productivity of Finns
edulis-Juniperus monospenna woodlands of north-
central Arizona. Forest Ecologv and Management 50:
331-350.
Keeney, D. R., and D. W. Nelson. 1982. Nitrogen — inor-
ganic foiTns. Pages 643-698 in A. L. Page et al., edi-
tors. Methods of soil analysis, part 2. 2nd edition.
Agronomy Monographs 9. American Society of Agron-
omy and Soil Science Society of America, Madison,
WI.
Klopatek, J. M. 1987. Nutrient patterns and succession in
pinyon-juniper ecosystems of northern Arizona. Pages
391-396 in R. L. Everett, compiler, Proceedings —
pinyon-juniper conference. General Technical Report
INT-215. Inteimountain Research Station, Ogden, UT.
Klopaiek, C. C, and J. M. Klop.viek. 1987. Mycorrhizae,
microbes and nutrient cycling processes in pinyon-
juniper systems. Pages 360-364 in R. L. Everett,
compiler. Proceedings — pinyon-juniper conference.
General Technical Report INT-215. Intermountain
Research Station, Ogden, UT.
Lanner, R. M. 1981. The pinon pine: a natural and cul-
tural history. University of Nevada Press, Reno.
Lyon,s, S. M., and G. E GlKEORD. 1980. Impact of incre-
mental smface soil depths on plant production, tran-
spiration ratios, and nitrogen mineralization rates.
Journal of Range Management 33: 189-196.
Meeuwig, R. O., and J. D. Budy. 1981. Point and line
intersect sampling in pinyon-juniper woodlands.
General Technical Report INT-104. Intermountain
Research Station, Ogden, UT.
Nelson, D. W, and L. E. Sommers. 1982. Total carbon,
organic carbon, and organic matter. Pages 539-579 in
A. L. Page et al., editors. Methods of soil analysis,
part 2. 2nd edition. Agronomy Monographs 9. Ameri-
can Society of Agronomy and Soil Science Societ)' of
America, Madison, WI.
Nobel, P S. 1988. Environmental biology of agaves and
cacti. Cambridge University' Press, Cambridge.
. 1989. A nutrient inde.x quantifying productivity' of
agaves and cacti. Journal of Applied Ecology 26:
635-645.
Nobel, P S., C. E. Russell, P Felker, G. C. Medin.^, and
E. ACUNA. 1987. Nutrient relations and productivity
of prickly pear cacti. Agronomy Journal 79: 550-555.
Petersen, R. G., and L. D. Calvin. 1986. Sampling. Pages
33-51 in A. Klute, editor. Methods of soil analysis,
part 1: physical and mineralogical methods. 2nd edi-
tion. Soil Science Society of America, Inc., Madison,
WI.
Richmond, D. L., and M. L. Richardson. 1974. General
soil map and inteqoretations, Mohave County, Ari-
zona. U.S. Department of Agriculture, Soil Conserva-
tion Sei'vice.
SAS. 1985. SAS user's guide: statistics, version 5 edition.
SAS Institute, Caiy, NC.
SCHLESINGER, W H., P J. FONTEYN, AND W A. REINERS.
1989. Effects of overland flow on plant water rela-
tions, erosion, and soil water percolation on a Mojave
Desert landscape. Soil Science Society of America
Journal 53: 1567-1572.
SCHLESINGER, W H., J. E REYNOLDS, G. L. CUNNINGHA.M,
L. E HUENNEDE, W M J.'VRRELL, R. A. VIRGINIA, AND
W. G. Whitford. 1990. Biological feedbacks in global
desertification. Science 247: 1043-1048.
Sellers, W. D., and R. H. Hill, editors. 1974. Arizona
climate, 1931-1972. University of Arizona Press,
Tucson.
Tiedemann, a. R. 1987. Nutrient accumulations in pinyon-
juniper ecosystems — managing for future site pro-
ductivity. Pages 352-359 in R. L. Everett, compiler.
Proceedings — pinyon-juniper conference. General
Technical Report INT-215. Intermountain Research
Station, Ogden, UT
Tiedemann, A. R., and J. O. Klemmedson. 1973. Effect of
mesquite on physical and chemical properties of the
soil. Joumal of Range Management 26: 27-29.
224
Great Basin Naturalist
[Volume 56
USDA Bureau of Land Management. 199L Record of
licensed livestock, Hualapai Ph. allotment, 1967-199L
Kingman Resoince Area, Kingman, AZ.
U.S. EPA. 1983. Methods for chemical analysis of water
and wastes. EPA-600/4-79-020, revised March 1983,
Method 350.1, Method 353.2.
ViTOUSEK, P. M., J. R. Gosz, C. C. Grier, J. M. Melillo,
AND W. A. Reiners. 1982. A comparative analysis of
potential nitrification and nitrate mobility in forest
ecosystems. Ecological Monographs 52: 155-177.
West, N. E. 1988. Intermountain deserts, shiiib steppes,
and woodlands. Pages 209-230 in M. G. Barbour and
W. D. Billings, editors. North American terrestrial
vegetation. Cambridge University Press, Cambridge.
West, N. E., and J. O. Klemmedson. 1978. Stmctural dis-
tribution of nitrogen in desert ecosystems. Pages
1-16 in N. E. West and J. Skujins, editors. Nitrogen
in desert ecosystems. Dowden, Hutchinson & Ross,
Inc., Stroudsburg, PA.
Zar, J. H. 1984. Biostatistical analysis. 2nd edition. Pren-
tice-Hall, Inc., Englewood Cliffs, NJ.
Received 20 April 1995
Accepted 14 March 1996
Great Basin Naturalist 56(3), © 1996, pp. 225-236
DISTRIBUTION AND ECOLOGICAL CHARACTERISTICS OF LEWISIA
LONGIPETALA (PIPER) CLAY, A HIGH-ALTITUDE ENDEMIC PLANT
Anne S. Halford'-^ and Robert S. Nowak'-'^
Abstract. — Lewisia longipetala (Piper) Clay is a high-altitude endemic Iduiul in llic northern Sierra Nevada. The
characteristics of 12 sites with L. longipetala, which represent all known populations, were studied to define habitat
requirements of the species. Meso- and microscale characteristics of the habitat were examined, including characteris-
tics of the associated plant communit>'. Average plant size and plant density of L. longipetala were also determined for
each population. Similar measurements were made on 6 populations of Lewisia pijgniaeu (A. Cray) Robinson, a more
common Lewisia. Populations of L. longipetala that had larger plants and higher plant density were associated with gen-
tK' sloped, north-facing sites that were near large, persistent snowbanks and had low vegetative cover. Plant species
associated with populations of L. longipetala were similar among the 12 sites and were indicative of mcsic, rocky alpine
sites. These t\pes of plant commimities found near persistent snowbanks are often termed snow-bed vegetation. In con-
trast, L. pijgnmea was found to be less site specific. Lewisia pygmaea was foimd adjacent to or interspersed with L.
longipetala at 5 sites, but was found in areas associated with a higher percentage of herbaceous cover and a wider vari-
ety' of species. This integration of ecological and commmiity information for L. longi}H'tala populations contributes to the
interim management and long-term monitoring of this species by providing needed information concerning its habitat
and en\ ironmental specificity.
Key words: Lewisia longipetala, Lewisia pygmaea, site cluiracteristies, snow-bed vegetation, alpine, endemic, plant
size, plant density.
The recent implementation of programs to
preserve rare plant taxa indicates the elevated
concern for effective and long-term steward-
ship of sensitive species (Sutter 1986). One of
the initial steps toward the protection of rare
plants is to document their occurrences (Utter
and Hurst 1990). Mountain ranges are typi-
cally rich in endemics (Major 1989), and within
the Sierra Nevada they comprise a high per-
centage of the flora (Stebbins and Major 1965).
Factors that characterize the species habitat
are inferred from the species' geographic dis-
tribution and often suggest environmentally
imposed limitations on the distribution of sen-
sitive plant taxa (Baskin and Baskin 1988,
Hutchings 1991, Nelson and Haiper 1991). For
example, some limitations that influence en-
demic plants within alpine environments are
snowbank depth and duration (Komarkova 1975,
Webber et al. 1976) and levels of disturbance
to root systems fi-om needle ice (Fitzgerald et al.
1990). To help ensure the survival of rare plant
species, habitat and biological information
should be integrated with long-term monitor-
ing programs (Sutter 1986, Baskin and Baskin
1988, Hutchings 1991).
Species within the genus Lewisia (Portula-
caceae) are well known in horticulture (Elliot
1966, Mathew 1989). However, little informa-
tion exists regarding these species in their
native environments. Only 4 species within the
genus Lewisia have relatively wide distribu-
tions: Lewisia pygmaea (A. Gray) Robinson, L.
nevadensis (A. Gray) Robinson, L. triphyUa (S.
Watson) Robinson, and L. rediviva Fursh. The
remaining 15 species have considerably smaller
distributions, and 9 that occur in California are
listed by the U.S. Fish and Wildlife Ser\ace as
candidates for threatened or endangered status.
Leivisia longipetala (Piper) Clay is a federal
candidate 2 species, which implies that data on
identifiable threats are insufficient to support
federal listing as threatened or endangered
(Skinner and Pavlik 1994). Lewisia longipetala
is an endemic species with limited distribution
that the California Native Plant Society classi-
fies as a category 1 B species, which is a cate-
gory for rare, threatened, or endangered plants
within California. Lewisia longipetala popula-
tions are fairly remote, and most exist in U.S.
Forest Service wilderness areas. Although L.
'Department ot Environmental and Resource Sciences, Mail Stop 199, University of Nevada at Reno, Reno, NV 89.5.57.
^Present address: Bureau of Land Management. 785 N. Main St. Suite ?. Bishop, CA 93514.
■'Author to whom reprint requests should be submitted.
225
226
Great Basin Naturalist
[Volume 56
longipetala populations are not an immediate
management concern, one population (Basin
Peak) is on private land, and mining claims
within close proximity of the site pose a poten-
tial threat. FurtheiTnore, the potential also exists
for activation of mining claims within wilder-
ness areas as well as increased ski area devel-
opment within the vicinity of the other L. longi-
petala populations.
The first specimen of L. longipetala was col-
lected by J. G. Lemmon in 1875 in the moun-
tains west of TiTickee, California. In 1913, Piper
described L. longipetala as Oreohroma longi-
petalum, an intermediate between L. pygmaea
and L. oppositifolia (S. Watson) Robinson. Later
descriptions (Munz 1959) placed L. longipetala
as a subspecies of L. pygmaea. More recently,
L. longipetala was again recognized as a dis-
tinct species (Dempster 1993), a distinction
supported b>' moiphological as well as chromo-
somal differences between L. longipetala and
L. pygmaea (Stebbins 1968, Halford 1992).
Lewisia longipetala (Fig. 1) is an herbaceous
perennial with a basal tuft of green, linear
leaves. An individual plant produces numerous
scapes, 30-60 mm long, each bearing 1-3 pale
pink flowers with petals 11-20 nmi long. The
two sepals are distinctly fuchsia in color, 4-10
Fiy;. 1. Line drawing ni Lcicisia loii^iix-tdld (Piper) Cla\ slum iiiii lirow tli liahil.
I99(ij
Distribution and Ecolocy oi^ L. longipetala
227
mm long, and coiispicuoush i!;laii(liiIai"-(l(Mitate
(Elliot 1966, Mathew 19S9). In contrast, inllo-
rescences of L. i)i/30%) with west-, southwest-,
or soudieast-facing slopes. Regression analyses
between plant density and 9 site characteris-
tics did not yield 1 "best" model but rather 2
models that had similar adjusted R^ values
(Table 3). For both models, slope was a signifi-
cant dependent variable, and plant density was
inversely correlated with slope (i.e., as slope
increased, plant density decreased). Surface
water and surface rock cover were significant
dependent variables in 1 model, and plant
density was positively correlated with both of
these dependent variables. In the 2nd model,
total vegetative cover was a significant depen-
dent variable, and L. longipetala density was
inversely correlated with vegetative cover.
Populations with the largest plants generally
were also those with the highest plant density
(Table 1). Hc-gression analyses between clump
diameter and 9 site characteristics yielded a
single model that all forward and backward
stepwise regressions converged upon (Table 3).
Mean plant diameter from each population
was inversely correlated with distance from
the nearest uphill snowbank. The value of the
regression coefficient for surface litter cover
was significantly different from zero at the 6%
probability level rather than the 5% level, and
plant size was inversely correlated with the
amount of surface litter cover.
Classification and ordination of die floristic
data corroborated these results (Fig. 3). Four
site characteristics were found to be significantly
correlated with DCA axis 1 scores (Table 4),
and these site characteristics are shown in Fig.
3A as vectors that indicate the directional
increase of slope, surface rock cover, bare
ground cover, and total vegetative coven
rVVINSPAN classified the 12 populations into
3 groups (Fig. 3A), and the species groupings
associated with the TWINSPAN population
groupings are shown in Fig. 3B. The Basin
Peak populations had higher vegetative cover,
whereas the other populations had higher rock
cover (Fig. 3A). These populations that are
associated with increasing rock cover contain
species such as Antennaria media, Cassiope
nieHensiana, and Kalmia polifolia var. micropy-
Ua that are indicative of such environments
(Fig. 3B). L. longipetala populations at Granite
Chief, Top Lake, Mt. Price 2, Mt. Price 3, and
Table 1. Descriptive site attributes for 12 Lcicisia longipetala populations, ordered from north to south. Mean ± stan-
dard error ot plant diameter from 20 randomly selected plants, as well as plant density, is gi\en for each population.
Parent
Elevation
Slope
Plant diameter
Plant density
Population
material
(m)
Aspect
m
(cm)
(# per 0.5 ha)
Basin Peak 1
Basalt
2800
NNE
2-8
9.7 ± 0.3
185
Basin Peak 2
Basalt
2840
NNE
2-8
3.9 ±0.1
10
Pole Creek 1
Basalt
2733
NNE
2-6
13.0 ±0.9
>.500
Pole Creek 2
Basalt
2733
NNE
2-6
8.2 ± 0.4
>.500
Granite Chief
Granite
2800
N
>.30
6.5 ±0.1
135
Dick s Lake
Granite
3033
NNE
2-10
8.6 ±0.4
>.500
Top Lake
Granite
2866
W
>30
4.2 ± 0.2
12
Mt. Price 3
Granite
31.33
wsw
>30
6.3 ± 0.4
35
Mt. Price 1
Granite
3200
SSE
2-8
3.4 ±0.2
40
Keith's Dome
Granite
2800
NNE
2-8
10.8 ±0.4
>,500
Mt. Price 2
Granite
2966
ssw
>30
8.3 ±0.3
30
Pyramid Peak
Granite
2787
WNW
>30
4.1 ±0.2
25
230
Great Basin Naturalist
[Volume 56
Table 2. Mean percent cover for species found within Lewisia longipetahi populations and the number of L.
longipetala populations that contained that species. Species are listed from highest to lowest cover. Hickman (1993) was
used as the authorit\' for all species. Letter codes used in Figures .3 and 4 are gi\'cn in brackets for each species.
Species
Species
code
Mean
cover
#of
pop.
Carex scopuloriim Holm, var bracteosa (L. Bailey) E Hemi.
Antennaria media E. Greene
Juncus mertensiamis Bong.
Erigeron peregrinus (Pursh) E. Greene
Lupinus hreweri A. Gray
Lewisia pijgmaea (A. Gray) Robinson
Lewisia longipetala (Piper) Clay
Arnica mollis Hook.
Mimulus guttatus DG.
Salix artica Pallus
Aster alpigemis (Toirey & A. Gray) A. Gray ssp. andcrsonnii (A. Gray) M. Peck
Calijptridium umheUatum (Torrey) E. Greene
Phletim alpimim L.
Juncus drummondii E. Meyer
Sibhaldia procumbens L.
Dodccatheon alpinum (A. Gray) E. Greene
Cassiupe mertensiana (Bong.) Don
Kabnia poIifoUa Wangenh. ssp. microphijlla (Hook.) Galder & Roy Taylor
Ltjcopodiuin sp.
Mi)nubis primuloidcs Benth.
Foa wheeleri Vasey
Polygonum bistortoides Pursh
Eriogonum incamim (Torrey & A. Gray)
Penstemon rijdbergii Nelson ssp. oreocharis (E. Greene) N. Holmgren
PhyUodoce breweri (A. Gray) Ma.xim.
Anemone drummondii S. Watson
Poa secimda J.S. Presl ssp. secunda
Sedum roseum (L.) Scop. ssp. integrifolium (Raf ) Hulten
[Gascb]
6.2
9
[Anme]
5.1
9
Qume]
4.2
4
[Erpe]
2.9
2
[Lubr]
2.6
1
[Lepy]
1.9
5
[Lelo]
1.7
12
[Anno]
1.4
1
[Migu]
1.2
3
[Saar]
1.2
3
[Asala]
1.0
5
[Gaum]
0.9
4
[Phal]
0.9
2
[Judr]
0.8
3
[Sipr]
0.8
4
[Doal]
0.7
2
[Came]
0.7
2
[Kapom]
0.5
2
[Lycsp]
0.5
5
[Mipr]
0.5
3
[Powh]
0.3
1
[Pobi]
0.3
2
[Erin]
0.2
1
[Peryo]
0.2
1
[Phbr]
0.2
2
[Andr]
0.1
1
[Poses]
0.1
1
[Seroi]
0.1
1
Pyramid Peak were situated iu cracks on steep
granitic sU^bs, and one of the most common
species found associated with these sites was
rock sedum {Sedum roseum ssp. integrifolium).
NJMDS results differed slightly from the
TWINSPAN classification by the separation of
the Top Lake population from all other popula-
tions and by a change in association of the
Basin Peak 2 population from the group of 3
Basin Peak 1 plots to the group of 7 plots to the
right of Basin Peak 2. In the 2-dimensional
space defined by DCA axes 1 and 2, Basin
Peak 2 appears to be transitional in its floristic
composition between the Basin Peak 1 plots
and this group of 7 populations. Species that
contributed to these different classifications of
the Basin Peak populations were Phleum alpin-
um and Lupinus breweri: P. alpinwn was within
Table 3. Multiple regression results between each of 2 dependent xariables (plant density' and plant diameter") and the
set of 9 site characteristics for 12 Lewisia longipetala populations.
Dependent variable
Regression statistics
adj. R-
Variable
statistics
Model variables
CoefRcient
P
Slope
-18.5
<0.01
Surface water cover
11.2
<0.01
Sm-face lock co\er
6.9
0.01
Slope
-16.6
<0.01
Total vegetati\e co\'er
-6.5
<0.()I
Snowbank distance
-0.07
<0.01
Surface litter cover
-0.24
0.06
Plant density
Plant diameter
12
12
12
0.81 <0.01
0.81
0.53
<0.01
0.01
1996]
DlSIRlliU IKJN AND EcoUKiV OF L. LONGIPETALA
231
400
300
200
100
CN
<
A
Vegetation cover
Bareground
cover
ick's Lk Q^eith's Dome
PoleCkl O OPoleCk2
BasinPk2 0; ^^ O ^Pyramid Pk
Mt Price 3
Rock cover
>
100 200 300 400
DCA Axis 1
500 600
Fig. 3. A, Population ordinations generated by DCA for L. longipetala; B, species groupings associated with the popula-
tions. For both graphs, circled groups were determined from TWINSPAN dench-ograms; broken hues indicate NMDS
groupings. Letter codes for each species are given in Table 2.
232
Great Basin Naturalist
[Volume 56
Table 4. Conelation coefficients generated from a rota-
tional correlation program for all Lewisia lon^ipctala and
Lewisia pijginaea DCA axis 1 scores. Variables with an *
are significant at the 0.05 level.
Species
Vniables
Correlation coefficients
Le w is id longipetala
Elevation
0.24
*Slope
-0.58
Aspect
-0.49
Snowbank distance
0.41
*Bare ground cover
0.66
Litter cover
0.36
*Surface rock cover
0.71
Surface water cover
0.11
*Total vegetative cover
0.67
Lewisia pygimiea
Elevation
-0.53
Slope
0.17
Aspect
-0.14
Snowbank distance
0.38
*Bare ground cover
0.92
* Litter cover
0.92
Surface rock cover
0.34
Surface water cover
0.26
*Total vegetative cover
0.78
the sampled transects of both Basin Peak 1 and
2 populations but not within any of the other
populations; on the other hand, L. breweri was
only within the Basin Peak 1 populations. In
general, it should be noted that classification
inferences based solely on location within the
2-dimensional space defined by any 2 DCA
axes can be misleading: for example, Mt. Price
1 and Mt. Price 2 are close together in the 2-
dimensional space defined by DCA axes 1 and
2, but they do not classify into the same group
in either TWINSPAN or NMDS because they
are on different planes in the 3-dimensional
space defined by the addition of a 3rd axis.
Lewisia pygmaea
Lewisia pygmaea grew in areas where total
vegetative cover was greater than that where
L. longipetala was found. The strongest evi-
dence for this difference in site characteristics
was from the 5 sites where L. longipetala and
L. pygmaea coexisted in proximit)' to each other:
Basin Peak 1, Basin Peak 2, Granite Chief,
Dick's Lake, and Keith s Dome. Total vegeta-
tive cover for areas with L. pygmaea averaged
60.4 (s^: 5.9), which was 55% greater than the
mean cover of .39.0 {sj: 12.3) for areas with L.
longijH'tala: tliis difference was significant at P
< 0.10 (paired t test, 4 d.f, P = 0.067). This
large difference in vegetative cover persisted
even when all populations were considered:
for all the known L. longipetala populations,
mean vegetative cover was 31.8 (s^: 5.8); for
the 6 L. pygmaea populations used in this
study, mean vegetative cover was 53.7 {s^- 8.3).
Although this difference was significant (2-
sample t test, 16 d.f, P = 0.046), note that our
original selection of L. pygmaea populations
was not designed to be a random sample of all
L. pygmaea populations and thus extrapolation
to all L. pygmaea populations is not statistically
justified.
TWINSPAN results for L. pygmaea popula-
tions grouped the Basin Peak populations sep-
arately from other populations (Fig. 4A), but
the environmental site attributes that were sig-
nificantly correlated with DCA axis 1 scores
differed between L. pygmaea and L. longi-
petala (Table 4). Litter cover was a significant
site attribute for L. pygmaea, but slope and
surface rock cover were not. The vegetative
cover vector increased toward the Basin Peak
population, which suggested that these popula-
tions contained a greater herbaceous compo-
nent. The species indicative of such areas
include Erigeron peregrinus, Sali.x artica, and
Arniea mollis (Fig. 4B). The bare ground vector
also increased toward the Basin Peak stands.
Although the concomitant increases in bare
ground and vegetative cover may seem contra-
dictoiy, surface rock cover tended to decrease
toward Basin Peak. Thus, smface rock was
replaced by vegetation and bare ground (i.e.,
inorganic soil) along these vectors. High litter
cover was commonK' associated with the Gran-
ite Chief and Keith s Dome populations.
Piute Pass is an area historically thought to
contain L. longipetala. However, only L. pyg-
nmea indixiduals were verified at this site. The
area is south of Yosemite National Park, Cali-
fornia, which makes it the southernmost site
surveyed for L. longipetala. Environmental
attributes of Piute Pass are similar to those of
other L. longipetala and L. pygmaea popula-
tions, except for some differences in species
composition, nameK the relative preponder-
ance of Dodeeatlieon Jeffreyi.
Discussion
Site chaiacteristics that were most highly
associated with the occurrence of L. longi-
petala and also correlated with plant size and
1996]
DlSTKIBUTION AND EcXiUKiY OK L. LONGIPETALA
233
400
300
A
Litter cover Vegetation cover
Bareground
cover
-300
-TOO
00
200
300
400
500
DCA Axis 1
Fig. 4. A, Population ordinations generated by DCA for L. pygmaea; B, species groupings associated with the popula-
tions. For both graphs, circled groups were determined from TWINSPAN dendrograms. Letter codes for each species are
given in Table 2, except Doje = Dodecatheon jeffreiji.
234
Great Basin Naturalist
[Volume 56
density were proximity of snowbanks, steep-
ness of slope, slope aspect, and cover of vege-
tation, surface rock, and surface water. These
inferences are supported by inspections of the
site characteristics and by statistical analyses.
The L. longipetala populations with higher
densit\' were found on gently sloping sites with
a northern exposure that were close to snow-
banks and had low vegetative cover of all
species. For example, Pole Creek 1 and 2, Keitli s
Dome, and Dick's Lake have populations that
exceeded 500 individuals, whereas Basin Peak,
which overall had the most herbaceous cover
of any of the other populations, had much
lower plant density. Plant density of L. longi-
petala populations increased with increased
cover of surface water and rock, but decreased
with total vegetative cover and slope steep-
ness. Plant size, as measured by clump diame-
ter, increased with decreased distance from
snowbanks and decreased litter cover Further-
more, at Basin Peak 1 as well as other sites of
L. longipetala populations, plants that were
more distant fi'om snowbanks or that were on
south-facing slopes were more water stressed
(Halford 1992).
Site characteristics that are associated with
more vigorous L. longipetala populations are
indicative of areas that receive high snowpack
accumulations. In alpine environments plant
communities whose occurrences are influ-
enced by geomoiphological characteristics that
favor high snowpack accumulations are often
termed snow-bed vegetation (Billings and Bliss
1959, Kuramoto and Bliss 1970, Canaday and
Fonda 1974, Tomaselli 1991). Some species in
the Siena Nevada that Major and Taylor (1977)
commonly found associated with areas of high
snowpack accumulations and that often occur
in mesic depressions with low vegetative cover
are Fhijllodoce hreweri, Cassiopc inciiensiana,
Kaltnia polijolia van niicrophylla, Fhleiun
alpinum, Mimulus primuloides, and M. gutta-
tus. Additional species that occm- in mesic to
even hydric habitats include Antennaria media,
Sibbaldia procumhens, Dodecatheon alpinwn,
and Sedwn roseum (Major and Taylor 1977).
These species were tilso associated with L. longi-
petala populations. Conversely, species that are
more frequently associated with xeric sites,
such as Lupinus hreweri and Juncus drwn-
mondii (Chabot and Billings 1971, Nachlinger
1985), were less fre(|uently associated with L.
longipetala populations.
The restriction of some species to sites with
low vegetative cover may be related to reduced
interspecific competition (Ostler et al. 1982).
For example, competition partially accounts for
the reduced growth of Talinum calcaricwn, a
highly restricted rock outcrop species of the
Portulacaceae family, in herbaceous sites domi-
nated by Poa pratensis (Ware 1991). Viable
populations of the endangered Furbish's louse-
wort {Pedicularis furbishiae) occur on mesic,
rocky sites that experience intermediate distur-
bances from hydrological processes, which
remove potential competitors (Menges 1990).
Potentilla robbinsiana, an endemic from New
Hampshire's White Mountains, also requires
rocky mesic sites that are moderately dis-
turbed, in this case by frost heaving that limits
other species (Fitzgerald et al. 1990). The lower
densities and smaller L. longipetala plants in
areas \\dth high vegetative cover and high soil
organic matter (Halford 1992) suggest that
interspecific competition may also restrict this
species, but specific studies need to be con-
ducted to explicitly test this mechanism.
The environmental site characteristics of L.
pijgmaea are broader than those of L. longi-
petala. Populations of L. pygmaea have been
documented in dense herbaceous meadows,
cracks in steep rocks, and open gravely depres-
sions (Elliot 1966, Major and Taylor 1977). In
our study plants of L. pygmaea were found
adjacent to 3 and interspersed with 2 of the 12
L. longipetala populations, which suggests that
L. pygmaea and L. longipetala can grow in sim-
ilar environments. However, an important dif-
ference between the 2 species is that L. pyg-
maea was found in areas with more herbaceous
cover The less pronounced site specificity ex-
hibited by L. pygmaea parallels other widely
distributed, mesic alpine species, whereas the
relative restriction of L. longipetala to more
open sites is similar to other restricted plant
taxa (Fitzgerald et al. 1990, Menges 1990).
The potential threats to L. longipetala are
not imminent at this time but include both sto-
chastic and anthropogenic processes. Climatic
events such as periodic droughts that reduce
snowpack accmnulations as well as potential in-
creases in interspecific competition may signif-
icantly reduce die viability' of L. longipetala pop-
ulations, especially those that already have low
densities of individuals. Human activities may
also ha\'e significant impacts. For example, if
slopes above populations are altered by mining
1996]
DiS I lUBUTION AND EcOLOCiY OF L. LONGIPETALA
235
activity or ski area cle\ t^lopment, the displace-
ment of substrate could alter the topograph)
and hence hydrology of the site through changes
in snow acciunulation and melt water rimoff
To enhance the long-term viability of this
endemic species, primary management goals
should include (1) monitoring of L. lonciisis + , O. tuhcrctilata, and Thrassis iHin(lorae + ), 1 X^iok (Ixodes sculptu.s + ),
and an eyeworni (Nematoda: Hhalxliti.s ()rhitalis* + , also 1st records from Sciuridae); S. /;. ctulcmiciis was host to a louse
species {Neohaematopimis la('iiiisciilii.s + ). 5 flea ta.\a (RJuidmopsijUu sp. + , (). t. tubcrculata, Tlirassis f. fr(incm + , T. f.
b(irnesi + , and T. f. rockwoodi), and a mite {Aiidn)l(i('lai).s J(ilircnht)lzi + ). S))enni)pliilus hrimneus had fewer known
ectoparasite species than other congeners. Although all of their parasites had many other hosts, S. h. endeinicus and S. h.
bntnneiis shared only a single parasite species in common, whereas all but one of their eetoparasites also occurred on
the closely related Townsend's ground squirrel (S. townsendii). The proportion of parasitized individuals and the para-
site loads per individual were significantly lower in S. b. bninneus, which lives in small, isolated populations, than in S. b.
cndemlcus, which has larger, less fragmented populations, suggesting a relationship between host population structure,
parasite loads, and parasite species diversity. All but one of the flea species have been linked to plague transmission.
Keij words: ground squirrels, eetoparasites, Spermophilus brnnneus, Idaho.
Tlie Idaho ground squirrel {Spermophilus
bninneus) is one of die rarest and, until recently,
least known North American mammals (Sher-
man 1989, Yensen 1991, Yensen and Sherman
in press). This endemic species inhabits a 125
X 90-km area in west central Idaho, but it
actually occupies only a small fraction of this
limited range (Yensen 1991). Despite the
species' restricted geographic distribution,
there are 2 allopatric subspecies that are mor-
phologically and genetically differentiated and
possibly have reached species-level separation
(Yensen 1991, Gill and Yensen 1992, Gavin et
al. submitted).
Spermophilus b. brunneus occurs in montane
meadows surrounded by coniferous forests at
elevations of 1035 to 1550 m in Adams and Val-
ley counties (Yensen 1991). As of 1995, only 18
of the 28 known populations remained, and
only one of these contained >100 animals. The
majority of the sites were within an area of 22
X 9 km and totaled <300 ha of occupied habi-
tat (T A. Gavin, E W. Sherman, and E. Yensen
unpublished data).
Fire supression began in the area about 100
yr ago. Subsequent succession and expansion
of forests has filled in many of the natural
meadows in the range of S. /;. brunneus (Truksa
and Yensen 1990), eliminating habitat. The
remaining populations are presently isolated
from each other by the encroachment of coni-
fers into meadows and by competition with
Columbian ground squirrels (Yensen and Sher-
man in press). Today, there is apparently little
or no gene flow among populations. Allozyme
analyses of 55 protein loci in 12 populations
(Gavin et al. submitted) indicated that the pro-
portion of polymoiphic loci was 11.5%-19.2%
and heterozygosity values were 0.041-0.080.
Fj.f was 0.317, implying that there is genetic
differentiation among populations despite their
geographic proximity and the apparent recency
of their separation. In 1993 the total number of
individual S. b. brunneus was 1000-1200, but
the number fell to 600-800 in 1994 and 1995
(T A. Gavin, E W. Sherman, E. Yensen per-
sonal observation).
Spermophihis b. endemicus occurs in rolling
foothills at elexations of 670 to 975 m in Gem,
Payette, and Washington counties (Yensen 1991).
It is patchily distributed throughout its range
of 75 X 30 km. Although censuses of S. /;.
endemicus populations have not been made, its
total population is apparently much larger than
that of S. h. brunneus. The area occupied, esti-
mates of population densities, and the amount
iMuseum of Natural History, Albertson College, Caldwell, ID 83605.
^University of Idaho, Pamia Research and Extension Center, Pamia, ID 836fi().
■^Section of Neurobiology and Behavior. Cornell Universih', Ithaca, NY 14853.
237
238
Great Basin Naturalist
[Volume 56
of remaining habitat are more than 2 orders
of magnitude greater than for S. b. brimneus
(E. Yensen personal obsei'vation).
Parasites of S. brunneiis have not been pre-
viously sui^veyed. The only prior records (Baird
and Saunders 1992) were 2 flea species, Oro-
psijUa t. tuberciilata and Thrassis francisi rock-
woodi, collected from specimens now referred
to S. b. endemicus (Yensen 1991).
We were interested in how ectoparasite
diversit>' and density are affected by reduction
in size and isolation of host populations.
According to epidemiological models (Ander-
son and May 1979, May and Anderson 1979),
the number of contacts between hosts and in-
fective stages of parasites determines the rate
at which adult parasites are acquired. Mean
parasite load should equal growth rate of the
population divided by mortality from the dis-
ease. Thus, as population growth slows, para-
site load per individual should drop. At veiy
low host population densities, there may be too
few contacts even to maintain ectoparasite popu-
lations. Thus, we predicted that S. b. brimneus
should have fewer ectoparasite species and
fewer ectoparasites per individual than con-
generic, more widely distributed western
ground squirrels {Spennophilus spp.). We also
predicted that due to its fragmented popula-
tion structure and smaller population sizes, S. b.
brimneus should have fewer ectoparasite species
than S. b. endemicus.
Because of questions about the taxonomic
similarity of S. /;. brunneiis and S. b. endemi-
cus, we also wished to leani if they had similar
ectoparasites, and how similar their ectopara-
sites were to those of other western ground
squirrels. Further, because of the limited geo-
graphic range and low number of small popu-
lations, both subspecies of S. brunneiis would
be vulnerable to extirpation liy an epizootic
such as plague. Thus, it was important to learn
if their ectoparasites were species involved in
plague transmission.
Methods
From 1980 to 1990, specimens of S. brunneiis
were collected for a taxonomic study (Yensen
1991). To minimize negative impacts on small
populations, a mean of 0.5 individuals/site/yr of
S. brunneus was collected. Squirrels were killed
by shooting or by live-trapping and injecting
nembutol into the heart. ImmediateK post-
mortem, squirrels were placed individually in
plastic bags; fleas, ticks, lice, and larger mites
were collected with forceps or a camel's hair
brush moistened with 70% ethanol as they left
the host. Squinels were not examined under a
dissecting microscope, so smaller mites were
not collected; eyes were not examined for eye-
worms.
From 1987 to 1994, S. b. brunneus were
live -trapped for demographic and behavioral
studies (Sherman 1989, and ongoing). They
were hand-held and parasites were picked off
with forceps; because the animals were not
anesthetized, all of the smaller and some of the
larger ectoparasites may not have been seen.
Eyes were checked for eyeworms by pulling
back the upper lid; specimens were removed
from the cornea of the eye with a cotton swab
moistened with sterile water. All parasites were
placed in 70% ethanol. In addition, 21 S. b.
endemicus were live-trapped at Sand Hollow,
Payette Count)', Idaho, in 1994 and examined
for eyewomis.
Collected specimens of S. brunneus were
prepared as standard museum study skins and
skiills and deposited in the Albertson College
Museum of Natural Histoiy (ACMNH), Cald-
well, Idaho, and the National Museum of Nat-
ural Histoiy (USNM); diey are identified below
by museum number Specimens of ectopara-
sites were sent to appropriate specialists for
identification and deposited in the entomologi-
cal collections at the University of Idaho,
Moscow, and ACMNH. Differences in parasite
loads between individuals and subspecies were
analyzed with hand-calculated Mann-Whitney
C^-tests and chi-square tests, as appropriate.
Results
We examined 29 freshly collected individu-
als of S. b. brimneus and 53 of S. b. endemicus
for ectoparasites. These represent 43% of the
192 museum specimens of this species known
to us (Yensen 1991, plus 4 additiouiil specimens).
AdditionalK', we opportunisticalK' collected ecto-
parasitic arthropods from 12 lixe-trapped indi-
\ iduals of S. b. brunneus and eyeworms from
another 36; we examined 21 S. b. endemicus
for eyeworms.
We collected 6 ectoparasite species from
Spennophilus b. brunneus: 4 fleas, 1 tick, and 1
nematode (Table 1). We collected 7 taxa of
ectoparasites from S. 1). endemicus: 5 fleas, 1
louse, and 1 mite.
1996]
SpERMOPHILUS BRUNNEUS Ec rOPARASITES
239
Tahlk 1. Parasites (if S. hninneits that also occur on some other species of western j^round squin-els (subgenus Sper-
inopliilti.'i). S\ inhols: * = knowni priniaiA' host(s); + = records, possibly accidental on host; - = no records in references
bi'iow''.
Host
This
study
Lit
erature
reco
-ds'>
Parasite
Sbb
Sbe
Sto
Sec
Sbl
Sar
Seh-
Sri
Swa
Sp>'
\ACE
\c()l}(inntit(>f)iiiti.s lacriiisciiluti
-
+
+
+
-
+
+
+
-
+
V\a:\s
\c(>i)siilhi inopiiia
+
-
+
+
+
+
+
+
-
_
Oropst/Ild icialiocn.sis
+
-
+
*
*
*
*
+
_
*
O. t. tiiberculata
+
+
+
+
+
+
_
+
+
_
RhadiiiopsijUa s. scctilis
-
?
+
-
-
-
-
-
+
_
Thrassis f. barnesi
-
+
+
-
-
+
+
-
_
_
T. f. francisi
-
+
*
-
+
+
-
_
_
_
T.f. rockwoodi
-
+
+
-
*
-
+
_
_
_
T. p. pandorae
+
-
+
+
*
*
*
+
+
-
i'lCKS
Ixodes scidptus
+
-
+
+
+
+
+
+
-
-
Mrn:s
Aiidroladups jahrcnholzi
-
+
+
+
-
+
+
+
-
+
Nematoda
Rliahdiiis orhitalis
+
-
-
-
-
-
-
-
-
-
•'lYom records in Hulibard (1947), Burgess (1955), Stark (1970), Hilton and Mahrt (1971), Wliitaker and Wilson (1974), Holekamp (1983), Lewis at al. (1988), Baird
and Saunders (1992), Baird (unpul)lished), and this study.
''Host acronyms: Sbb = Spennophtlus h. bnmnetis, Sbe = S. b. endemictK:, Sto = S. "totvnsendii" (sensu latu), Sec = S. columbianus, Sbl = S. beklingi, Sar = S.
iinniittis, Sel = S. clcgans. Sri = S. richardsonii, Swa = S. washingtoni. Spy — S. pamjii.
'Confused in the literature with S. richardsonii. The records here are those that unambiguously refer to this species, and the total for S. richardsonii may include
a few parasites of this species.
The proportion of parasitized individuals in
the 2 subspecies was strikingly different. We
found ectoparasites on 37 of 53 (70%) S. h.
endemicus but on only 8 of 29 (28%) S. b. briin-
neus collected (x^ = 13.4, d.f = 1, P < 0.001).
Parasitized individuals of S. b. brunneus had
1-3 species of ectoparasites each {X = 1.75, n
= 8), and parasitized individuals of S. b.
endemicus had 1-4 species of ectoparasites (X
= 1.59, n = 37). This difference was not signif-
icant (Uj = 154, P > 0.5). However, there was
a significant difference between subspecies in
the parasite load of parasitized individuals.
Fleas were the only common group of ectopar-
asites of both ground squirrel taxa. There were
4.1 fleas per parasitized individual in S. b.
brunneus and 7.8 in S. b. endemicus (U^ =
95.5, P < 0.05).
Annotated List of Ectoparasites
In the ectoparasite species accounts below,
letters and numbers in brackets refer to the
number of male and female fleas, e.g., [1 m, 2 f],
or to conversions of original collecting data to
latitude, longitude, and metric units.
Anopleura: Haematopinidae
Neohaemafopinus laeviiisculus (Grube)
We found this louse on S. b. endemicus in
the following locations: 11 mi [18 km] N
Emmett, Gem Co., T8N, R2W, Sec. 13
[44°02'N, 116°31'W, 830 m elev], 21 February
1982 (ACMNH 222), 28 February 1982
(ACMNH 226, 227, 236, 237, 238); 0.1 mi E
Payette Co. line, 12.6 mi [20 km] N Emmett,
Gem Co., T8N, R2W, Sec. 12 [44°03'N,
116°32'W, 810 m elev], 28 February 1982
(ACMNH 224); Weiser Cove, Washington Co.
[44°13'N, 116°44'W 715 m elev], 7 March
1982 (ACMNH 228, 229, 230); lower Mann
Creek, 2.5 mi [4 km] N jet. Weiser River Road,
Washington Co. [44°16'N, 116°51'W, 720 m
elev], 14 March 1982 (ACMNH 231, 240, 242,
243, 244).
This louse occurs fi-om Eurasia east to Alaska
and the Northwest Territories, and south
through western United States to Mexico; it is
apparently a species complex (K. C. Emerson
personal communication). Lice of this complex
have been collected from many ground squir-
rels (Eurasian Spermophilus major, S. citellus,
S. pygmaeus, S. undulatus, and North American
240
Great Basin Naturalist
[Volume 56
S. beecheyi, S. armatus, S. beldingi, S. colum-
bianus, S. parryii, S. townsendii, S. washing-
toni, and Ammospennophilus leucunis), as well
as marmots [Marmota flaviventris), chipmunks
{Tamias niiniiniis), pocket mice {PerognatJiiis
parvus), and deer mice {Peromijscus manicida-
tus; Raybum et al. 1975, Shaw and Hood 1975,
records fiom National Museum of Natural His-
toiy). Although N. kieviiiscidus is the most com-
mon louse species taken fi-om ground squirrels
in Idaho (C. R. Baird personal communication,
K. C. Emerson personal communication), S. b.
endemicus is a new host record.
Siphonaptera: Hystrichopsyllidae
Neopsyllo inopina Rothschild
We collected 8 individuals of this flea from
S. b. brunneus in the following locations: Lick
Creek, Adams Co., T19N, R3W, Sec. 14
[44°59'N, 116°40'W, 1290 m elev.], 17 April
1983 (ACMNH 305 [1 m, 2 f], ACMNH 306
[1 f]); 1 mi [1.6 km] NE Bear Guard Station,
Adams Co. [45°05'N, 116°37'W, 1480 m],
2 June 1988 (ACMNH 518 [1 f]); and Price
Valley [45°01'N, 116°26°W, 1270 m elev.],
3 June 1981 (ACMNH 209 [1 m, 1 f], ACMNH
210 [If]).
This flea occurs from British Columbia
south to Oregon and Nevada and east to
Saskatchewan, North Dakota, and Utah (Lewis
et al. 1988). It has been collected from other
western ground squirrels of subgenus Sper-
mophdiis (Table 1) and from badger [Taxidea
taxiis) dens (Lewis et al. 1988, Baird and Saun-
ders 1992); S. b. brunneus is a new host record.
RJiadinopsylla sp.
We collected 1 female specimen of this flea
genus from S. b. endemicus. UnfortunateK', it
could not be identified to species. The locality'
was Diy Creek Road, Payette Co., 1.4 mi [2.2
km] E Litde Willow Creek, T9N, R2W, Sec. 18
[44°07'N, 116°37'W, 815 m elev.], 26 February
1983 (ACMNH 318 [1 f], reported in Baird and
Saunders 1992).
The flea is most likely R. s. secfdis, which
occurs in many western states on deer mice
{Peronujscus sp.) and ground squirrels, includ-
ing S. townsendii and S. washingtoni (Lewis et
al. 1988, Baird and Saunders 1992). Rhadino-
psylla are uncommon fleas and have popula-
tion peaks in the colder months (Lewis et al.
1988). This is the 1st record of any Rhadino-
psylla species Iroiii N. brunneus.
Siphonaptera: CeratophyUidae
Oropsylla idahoensis (Baker)
This flea species was collected on S. b.
brunneus at the following locations: Price Val-
ley [45°01'N, 116°26°W, 1270 m elev.], 3 June
1981 (ACMNH 209 [3 fj); and OX Ranch 1-2
km S, 1-2 km E Bear, Adams Co. [45°00'N,
116°39'W, 1340 m elev.] (live-trapping collec-
tions).
Oropsylla idahoensis occurs from Alaska to
New Mexico and is one of the most common
fleas of ground squirrels in the Rocky Moun-
tains and westward. Hosts include other west-
ern ground sc^uirrels of subgenus Sj}ermophihis
(Table 1), golden-mantled ground squirrels (S.
lateralis), and marmots {Marmota sp.; Lewis et
al. 1988, Baird and Saunders 1992); S. b. brun-
neus is a new host record.
Oropsylla tuberculata tuberculata (Baker)
This was the most common flea on both
S. b. brunneus and S. b. endemicus, occurring
at nearly all locations fi'om which we collected
ectoparasites. We found O. t. tuberculata on
S. b. brimneus at the following localities: Price
Vallev [45°01'N, 116°26°W, 1270 m elev.],
3 June 1981 (ACMNH 209 [1 m], ACMNH
210 [1 m]); MiH Creek summit, 5 km N Hornet
Guard Station, Adams Co., T18N, R3W, Sec.
25, 4500' elev. [44°53'N, 116°39'W, 1370 m], 2
June 1985 (ACMNH 510 [2 m, 3 f], ACMNH
512 [2 m, 3 f]); Lick Creek, Adams Co., T19N,
R3W, Sec. 14 [44°54'N, 116°40'W, 1290 m
elev.], 17 April 1983 (ACMNH 305 [4 m, 3 f],
ACMNH 306 [1 fj); Round Vallev, Vallev Co.
[44°21'N, 116°00'W, 1460 m elev.], 18 Mav
1985 (ACMNH 315 [1 f]).
Records from S. b. endemicus are as follows:
Sucker Cr 11 mi [18 km] N Emmett, Gem Co.,
T8N, R2W, Sec. 13 [44°02'N, 116°31'W, 830
m elev.], 21 FebruaiT 1982 (ACMNH 221, 222,
223), 28 Februarv 1982 (ACMNH 225, 226,
227), 3 May 1987 (ACMNH 544 [1 m]); 0.1 mi
E Payette Co. line, 12.6 mi [20 km] N Emmett,
Gem' Co., T8N, R2W, Sec. 12 [44°03'N,
116°32'W, 810 m], 28 Februaiy 1982 (ACMNH
224, 236, 237, 238; reported in Baird and
Saunders 1992); Diy Creek Road, 1.4 mi [2.2
km] E Little Willow Creek, Payette Co., T4N,
R2W, Sec. 18 [44°07'N, 116°37'W, 815 m
elcN.], 20 FebruaiT 1983 (ACMNH 318 [10 m,
13 f]), 26 Februaiy 1983 (ACMNH 317 [8 m,
3 f]); Weiser Coxe', \\ashington Co. [44°13'N,
116°44'W, 715 m ele\.], 7 March 1982
1996]
Sl'EliMOl'lUlA 'S BIWNNEUS ECTO PARASITES
241
(ACMNH 228, 229, 230); lower Mann Civck.
2.5 mi [4 km] N jet. Weiser Ki\ er Road, Wash-
ington Co. [44°13'N, llCrsrW, 720 m elev.],
14 Mareh 1982 (ACMNH 231, 232, 233, 240,
242, 243, 244); Washington Co., lower Mann
Creek, 3.3 mi [5.3 km] N jet. Weiser River
Road [44°17'N, 116°51'W,"730 m elev.], 14
Nlareh 1982 (ACMNH 239).
This is a \er)' common flea in most of the
western United States and western Canadian
provinces (Baird and Saunders 1992). Hosts
include other \\'estem ground squirrels of sub-
genus Spennophihis (Table 1), antelope ground
squirrels {A)ninospennophilus Icuciirus), wood-
rats {Neototna sp.), and badgers (Lewis et al.
1988, Baird and Saunders 1992). It was previ-
ously recorded from S. hriinneus by Baird and
Saunders (1992).
Thrassis pandorac pandorae Jellison
We iound 1 specimen of this flea on S. b.
briinnens at Lick Creek, Adams Co., T19N,
R3W, Sec. 14 [44°54'N, 116°40'W 1290 m
elev.], 17 April 1983 (ACMNH 305 [1 m]).
This flea is distributed from Washington to
California and east to Colorado (Stark 1970). It
is found most fi^equently on Sperrnophilus artna-
tiis, S. beldingu and S. elegans (= richardsonii in
Stark 1970), but also occurs on S. cohiinbiamis,
S. elegons (Table 1), and a variety of other
rodents, lagomorphs, and carnivores (Stark
1970). S. b. bninneus is a new host record.
Thrassis francisi barnesi Stark
We found this flea on S. b. endemicus at
Sucker Cr. 11 mi [18 km] N Emmett, Gem Co.,
T8N, R2W, Sec. 13 [44°02'N, 116°31'W, 830
m elev.], 31 May 1981 (ACMNH 220 [3 m,
4 f]), 3 May 1987 (ACMNH 540 [4 m, 3f],
ACMNH 541 [2 m, 1 f], ACMNH 542 [1 m, 6 f],
ACMNH 543 [4 m, 7 f], ACMNH 544 [1 m,
1 f], ACMNH 545 [2 m, 1 f], ACMNH 547
[4 m, 9 f], ACMNH 548 [1 f], ACMNH 549
[3 m, 7 f]); 7 mi [11 km] N Emmett, Gem Co.,
T7N, RIW Sec. 5 [43°58'N, 116°29'W, 920 m
elev], 23 May 1987 (ACMNH 546 [4 m, 2 f]);
Sand Hollow, 5.6 km N, 5.0 km E Payette,
Payette Co., T9N, R4W, Sec. 7 [44°08'N,
116°51'W, 750 m elev.], 30 March 1989 (USNM
565927 [3 m, 2 f]).
This flea occurs north of the Snake River in
western Idaho, and on both sides of the river
in eastern Idaho and south into central Utah
and eastern Nevada (Stark 1970). Its most
connnon hosts are S. annafiis and S. elegans,
rather than S. townsendii mollis, the usual host
oi' T. f francisi. Stark (1970) felt that host asso-
ciations ma\ separate the 2 subspecies of T.
francisi, although the 2 lleas appeared to inter-
grade in eastern Nevada. S. b. endemicus is a
new host record.
Thrassis francisi fnnicisi (Fox)
We collected 14 indi\iduals of this Ilea from
S. b. endemicus at 1 locality: Dry Creek Road,
1.4 mi [2.2 km] E Little Willow Creek, Pavette
Co., T4N, R2W, Sec. 18 [44°07'N, 116°37'W,
815 m elev.], 26 February 1983 (ACMNH 318
[1 m, 5 f], SM2 [2 m, 3 f]), 24 Februar^' 1986
(ACMNH 920 [2 m, If]).
This flea is known from the Great Basin
desert of eastern Oregon, Idaho south of the
Snake River, eastern Nevada, Utah, and parts
of Wyoming. It occins primarily on S. town-
sendii, but the white-tailed prairie dog {Cyno-
mys leucurus) is the usual host in Wyoming
(Stark 1970). There are incidental records from
several species of ground squirrels (Table 1),
marmots, and deer mice (Stark 1970). Our
records are the 1st from any host north of the
Snake River in Idaho (Stark 1970, Lewis et al.
1988, Baird and Saunders 1992); S. /;. endemi-
cus is a new host record.
Thrassis francisi rockwoodi Hubbard
Two males of this flea were collected from
S. b. endemicus at a single locality: Sucker
Creek, 11 mi [18 km] N Emmett, Gem Co.,
T8N, R2W, Sec. 13 [44°02'N, 116°31'W, 830
m elev.], 21 Februaiy 1982 (ACMNH 223), 28
February 1982 (ACMNH 227 [2 m]; reported
in Baird and Saunders 1992).
This subspecies has been recorded liom east-
ern Oregon, northwestern Nevada, and north-
ern California, where it occurs most commonly
on S. beldingi, although collections ha\'e been
made from S. townsendii (Stark 1970, Lewis
et al. 1988).
Acarina: I.xodidae
Ixodes sculptus Neumann
We collected specimens of this tick fiom S. b.
brunneus at 1 localitv: OX Ranch 1-2 km S,
1-2 km E Bear, Adams Co. [45°00'N, 116°39'W,
1340 m elev.] (live-trapping collections).
This widespread tick occurs from western
Canada south to California and Texas and east
across the Great Plains. It occurs on several
242
Great Basin Naturalist
[Volume 56
western ground squirrels of the subgenus
SpermopJuhis (Table 1), prairie clogs {Cynomijs
sp.), marmots, voles {Microtus sp.), pikas
[Ochotona sp.), gophers {Thomonit/s sp.), jump-
ing mice {Zapiis sp.), domestic animals, and
various carnivores (Doss et al. 1974). S. h.
hrimneus is a new host record.
Acarina: Laelapidae
Androlaelaps fahrenholzi (Berlese)
We collected 8 specimens of this mite from
S. b. endemicus at the following localities: Sucker
Cr 11 mi [18 km] N Emmett, Gem Co., T8N,
R2W, Sec. 13 [44°02'N, 116°31'W, 830 m
elev.], 21 February 1982 (ACMNH 227 [4 f, 2
deutonymphs]); lower Mann Creek, 2.5 mi [4
km] N jet. Weiser River Road, Washington Go.
[44°16'N, 116°5rW, 720 m elev.], 14 March
1982 (ACMNH 233 [2 f]).
This mite is widespread in Eurasia, North
America (Whitaker 1979), and Central America
(Strandtmann 1949). It occurs on a wide vari-
ety of mammals, including marsupials {Didel-
phis sp.), insectivores, bats, several families of
rodents, lagomorphs, carni\'ores, and birds
(Strandtmann 1949, Whitaker and Wilson 1974,
Raybum et al. 1975). Opossums, insectivores,
and rodents are the primaiy hosts, but A.
fahrenholzi has the least host specificity and
widest geographic range of any North Ameri-
can ectoparasitic mite (Whitaker 1979). These
are the 1st records from S. hrimneus.
Nematoda: Rhabditidae
Rliahditis {Pelodera) orhitalis Sudhaus and
Schulte
We obsen^ed this parasitic eyeworm only in
live-trapped S. h. hrunneiis from OX Ranch 1-2
km S, 1-2 km E Bear, Adams Co. [45°00'N,
116°39'W, 1340 m elev.].
All specimens were collected in April and
May 1990 to 1994. We found them in 1 eye or
both eyes of yearling and adult S. h. bninneiis.
The number per eye varied from 0 to 1272.
The museum specimens were not checked for
eyeworms. In 1994, T A. Gavin and E W Sher-
man examined 21 live-trapped S. h. endemicus
from Sand Hollow, Payette Co., and found no
eyeworms.
This eyeworm has been reported previously
from Eurasian and North American voles and
lemmings {Microtus spp., Lemmus trimucroiui-
tus, Dicrostomjx groenlandicus, Pitinn/s suhter-
raneus, Arvicola terrestris, and Clelhrio)U)nu/s
spp.), mice {Apodemus spp. and Mus muscii-
his), and rats {Rattus norregicus; Poinar 1965,
Kinsella 1967, Cliff et al. 1978, Hominick and
Aston 1981, Schulte 1989). S. h. hrunneus is a
new host record, the 1st record of any RJiabdi-
tis from Sciuridae, and also the 1st record of R.
orbitaUs from Idaho.
Epizootics
In 11 field seasons (April-June) of work
with S. b. hrunneus, we found only 2 dead indi-
viduals, and none were obsei^ved sick or dying.
While a number of populations have declined
(T. A. Gavin, P W Shemian, and E. Yensen per-
sonal obsei"vation), mortality occuiTcd while die
animals were in hibernation rather than during
the active season. The most serious population
declines were estimated to be around 50% in
1 yr, rather than the 95%-100% active season
mortality typically associated with plague
(Lechleitner et al. 1968, Payor 1985). Although
numbers of fleas on indixidual squirrels were
relatively low, especially in S. b. hrunneus, all
flea species we collected are important in
plague epidemiolog)' in other hosts (Pratt and
Stark 1973) and could potentially play a role in
an Idaho epizootic.
Discussion
Collections of ectoparasites from S. hrun-
neus have resulted in new state records for the
flea TJirassis francisi rockwoodi and the eye-
worm Rhahdiiis orhitaUs, plus 9 new host
records. Because there have been no previous
studies of S. hrunneus, the new records are
hardly surprising. However, the records of
Thrassis f. francisi and T. f rockwoodi on S. h.
endemicus were unexpected. Thrassis f har-
nesi occurs north of the Snake River in the
Snake Ri\'er Plain (Stark 1970) and is the sub-
species of Tlvassis francisi that would be ex-
pected to occur in the range of S. b. endemicus.
Instead, we found TJirassis f francisi. which is
common in S. toicnsendii mollis south of the
Snake River, and T f rockwoodi, for which the
nearest locality' is from Oregon across die Snake
River (Stark 1970), a major biogeographic bar-
rier in southern Idaho (Da\ is 1939). This inter-
esting situation merits fmtlier stud\'.
With the exception of eyeworms, ectopara-
sites of S. hrunneus are all known from multi-
ple other species of ground stjuirrels (Table 1).
Thus it is curious that S. h. hrunneus and S'. I).
1996]
SrEHMormus bhuxneus Ectopaiusites
243
endemicus shared only a single ectoparasite,
OropsijUci t. tiiberadata, a widespread Ilea tonnd
on at least 4 other species of ground s(|uirrels.
By contrast, die geographicalK and taxononii-
cally close (Nadler et al. 1984) S. tuwii.sendii has
all but one of die ectoparasite species found on
both S. ])riiii)U'ii.s subspecies. However, Sper-
mophdiis townsciidii is now recognized (Ih)ff-
niann et al. 1993) as a complex of 3 closely
related sibling species with different kary-
otypes, and it was not always clear to us from
the literature (Table 1) which parasites were
associated with which host. Consequently, we
have treated S. towiisendii as a single entity
herein.
There are several possible explanations for
the lack of shared ectoparasites between S. h.
brunneiis and S. b. endemicus: (1) they are geo-
graphically separated, and their ranges are
inhabited b)' different ectoparasites; (2) they
occur in different habitats and therefore have
different ectoparasites; (3) pelage differences
between them may be different "microhabitats"
for ectoparasites; (4) possibly the formerly
shared ectoparasites on one or the other sub-
species have been lost via a founder event, due
to population structure, or because of popula-
tion bottlenecks; and (5) we did not adequately
sample all ectoparasites on either subspecies.
Among these hypotheses, (5) is the least inter-
esting evolutionarily, and (4) is the most inter-
esting.
Most western ground squirrel species are
allopatric or parapatric; thus, there is little pos-
siliility of direct transmission of ectoparasites
among them. Historically, the 2 subspecies of
S. brunneiis were separated by 19 km, 250 m in
elevation, and a habitat change from arid shiTib-
steppe vegetation to montane meadows (Yensen
1991). At present, the nearest extant popula-
tions are separated by 48 km. Because S. town-
sendii is allopatiic to S. brunneiis, occurs in non-
montane habitats, and has all ectoparasites found
on both subspecies of S. brunneiis, differences
in geography (hypothesis 1) and habitats (2) are
unlikely to be the sole explanations for the dif-
ferences in ectoparasites between S. I), brun-
neiis and S. b. endemicus.
There are significant differences in pelage
length between S. b. brunneiis and S. b. endem-
icus (Yensen 1991). Interestingly, the pelage of
S. townsendii is intermediate in length between
the 2 S. brunneiis subspecies (E. Yensen un-
published data). There also appear to be differ-
ences in hair density and diameter, although
these were not quantified b>' Yensen (1991).
Possibly S. townsendii is inhabitable by the
entire set of ectoparasites, and each subspecies
ol S. brunneiis is a suitable host for about half
the set. Thus, pelage differences (hypothesis 3)
aic a possible explanation for the lack of over-
laj) in ectoparasite species between 2 veiy
close relatives, but it would not explain the dif-
ferences in parasite loads or the low percent-
ages of nonparasitized individuals.
Anderson and May (1979) argued that para-
site infestations should be sensitive to host
population structure (hypothesis 4). As popula-
tion size declines and populations become more
isolated, the probability of parasite species loss
should increase. Our data were consistent with
this pattern: the proportion of parasitized S. b.
brunneiis was significantly lower than that of
S. b. endemicus; the former has smaller, more
isolated populations.
The isolated S. b. brunneiis populations
would also retard exchange of ectoparasites
among populations. Thus, there might be sto-
chastic losses of parasite populations with low
probabilit)' of recolonization (Anderson and May
1979). The differences in incidence of parasites
between S. b. brunneus and S. b. endemicus
are consistent with this inteipretation.
The low density and wide dispersion of in-
dividuals within S. b. brunneus populations at a
site (E. Yensen and E W. Sherman personal
observation) may also retard direct transfer of
ectoparasites, and, consequently, S. /;. brunneus
populations may not be able to support large
ectoparasite populations. The low incidence of
parasitism in Idaho ground squirrels thus
appears to be related to population stioicture.
Because we did not examine ground squir-
rels under a microscope, we do not suppose
that all ectoparasites were collected (hypothe-
sis 5). However, there was no systematic bias
in the sampling that would account for the dif-
ferences in the proportion of parasitized ani-
mals and parasite load differences between S. b.
brunneus and S. b. endemicus. The low propor-
tion of parasitized S. /;. brunneus (28%) and S. b.
endemicus (70%) in this study may have been
partially because our collecting techniques
missed smaller ectoparasites. However, the same
techniciues were used for both subspecies;
therefore, the sampling differences between
them should reflect real differences in parasite
load. Thus, with the number of animals and
244
Great Basin Naturalist
[Volume 56
localities sampled, the low overlap in lists of
parasites is striking.
Further, the low proportion of S. briinneus
with ectoparasites (55%), especially in S. b.
brunneus, is atypical of Spennophilus. For ex-
ample, Hilton and Mahrt (1971) found that in
Alberta 100% of S. cohitnbiamis and S. frank-
linii and 92% of S. richardsonii had ectopara-
sites. We were collecting S. townsendii and S.
columbianus at the same time as S. brunneus
and were impressed by the much higher para-
site loads on those species.
Although we did not obsei^ve plague in S.
brunneus during this study, it does occur in
southwestern Idaho. Serum samples positive
for Yersina pestis, the plague bacterium, were
reported from S. townsendii during a major
ground squirrel die-off in 1941-42 in Ada,
Canyon, and Payette counties, immediately
south of the range of S. b. brunneus (Hubbard
1947, Link 1955). In 1975-1977, positive anti-
body titers to plague were found in 72%-91%
of badgers in the Snake River Birds of Prey
Area, 50 km south of the range of S. b. endemi-
cus (Messick et al. 1983). Badgers are impor-
tant predators of ground squirrels. Eight of 9
dead Townsend s ground squirrels examined
by Messick et al. (1983) were positive for Y.
pestis. The plague bacterium has been detected
in other species of Spenno))hilus in all 5 Idaho
counties where S. brunneus populations exist,
but until 1995 no S. brunneus had been exam-
ined (Idaho Department of Health and Welfare
personal obsenation). In April 1995, T. A. Gavin
found a dead S. b. brunneus at the OX Ranch
and sent it to the Wyoming State Veterinaiy
Laboratoiy (Laramie) where it was assigned case
#95W3914. The carcass was found to be nega-
tive for Y. pestis (E. Williams persontil comment).
Nonetheless, in tlie event of a plague epizootic,
local populations of S. brunneus could easily be
decimated. With only a small number of popu-
lations remaining, plague could jeopardize the
sui'vival of both subspecies of S. brunneus.
Note Added in Press
Six hibemacula of S. b. brunneus were exca-
vated in spring 1995 (Yensen and Sherman
unpublished data). Nests recovered from the
hibernacula were placed in plastic bags in the
field, taken to the laboratory, and then placed
in Berlese funnels; small invertebrates were
collected in 70% ethanol. Onl>' the fleas have
been identified to date, but we can now add
the following records:
Neopsylla inopina
Adams Co., 1.5 km N, 1.5 km E Bear Guard
Station, 28 April 1995 [6 m, 7 f]; Adams Co.,
Steve's Creek, 2 km S, 2 km E Bear, 15 April
1995 [8 m, 7 f]; Adams Co., mouth of Cold
Springs Creek, 14 May 1995 [1 m, 1 f].
Oropsijlla idahoensis
Adams Co., 1.5 km N, 1.5 km E Bear Guard
Station, 28 April 1995 [1 m, 2 f]; Adams Co.,
Steve's Creek, 2 km S, 2 km E Bear, 15 April
1995 [4 m, 2 f]; Adams Co., 3 km S Bear, 16
April 1995 [1 f].
Oropsijlla tuberculata tuberculata
Adams Co., 1.5 km N, 1.5 km E Bear Guard
Station, 28 April 1995 [18 m, 16 f]; Adams Co.,
Steve's Creek, 2 km S, 2 km E Bear, 15 April
1995 [20 m, 21 f]; Adams Co., moutli of Cold
Springs Creek, 14 May 1995 [3 f].
Thrassis pandorae pandorae
Adams Co., 1.5 km N, 1.5 km E Bear Guard
Station, 28 April 1995 [28 m, 31 f]; Adams Co.,
Steve's Creek, 2 km S, 2 km E Bear, 15 April
1995 [8 m, 15 f]; Adams Co., 3 km S Bear, 16
April 1995 [1 m].
Cat(dlagia sp., prob. descipiens
Adams Co., 1.5 km N, 1.5 km E Bear Guard
Station, 28 April 1995 [1 f].
Foxella ignota
Adams Co., Steve's Creek, 2 km S, 2 km E
Bear, 15 April 1995 [4 m, 3 f].
Spennopliilus b. brunneus is a new host
record for Catallagia sp. and Foxella ignota.
Catallagia deeipiens is widely distributed in
the western United States and is usualK found
on deer mice (Baird and Saunders 1992).
Foxella ignota is commonly foimd on pocket
gophers in the northern Rock^' Mountains
(Hubbard 1947).
These new records also indicate that differ-
ent sets of ectoparasites occur on S. b. brun-
neus and S. /;. endeniicus, thus corroborating
the earlier results. The same 4 flea species
were again found associated with S. b. brun-
neus, and neither Catallagia nor Foxella is
kiiown from S. b. endemicus.
1996]
SPERMOrillLUS BHl'S'NEUS ECTOPAIUSITES
245
Acknowledgments
We thank Elizabeth J. Dyni, Elizabeth
Domingiie, Thomas A. Cia\in, D. Brad llam-
mond, Da\id O Neill, Daniel A. Stevens, and
William E Lanrance for assistanee in the field.
Robert E. Lewis (Iowa State Universit)),
Richai'd B. Eads and Eduardo Campos (Centers
for Disease Control), and Elizabeth Doniinyue
(Cornell Universit) ) kindly identified the fleas;
Nixon Wilson (University of Northern Iowa),
J. E. Keirans (Rock>' Mountain Laboratory), and
JoAnn Tenorio (Bishop Museum) identified the
ticks and mites; K. C. Emerson (National Mu-
seum of Natiual History) identified the lice;
Susan E. Wade (Cornell University) identified
the nematode eyeworms; and Amy Doerger-
Fields (University of Wyoming) tested the spec-
imen for plague. Financial support was pro-
vided by the National Science Foundation
(DEB-9225081), National Geographic Society
(grant #3485-86 to P W Sherman and E.
Yensen), George C. ("Tim ) Hixon, and Univer-
sity of Idaho Agricultural Experiment Station.
John and Jeanne Dyer and Tim Hixon pro-
vided encouragement, housing, and access to
research sites. We thank Sherilyn Robison and
Rita Colwell for helpful discussions, and William
H. Clark, Eric Eldredge, James Munger, John
O. Whitaker, Jr., and an anonymous referee for
constructive comments on an earlier version of
the manuscript.
Literature Cited
Anderson, R. M., and R. M. May. 1979. Population biol-
ogy of infectious diseases: part I. Nature 280: 361-367.
Baird, C. R., and R. C. Saunders. 1992. An annotated
checklist of the fleas of Idaho. University of Idaho,
College of Agriculture, Research Bulletin 148. 34 pp.
Burgess, G. D. 195.5. Arthropod ectoparasites of Richard-
son's ground squirrel. Joiunal of Parasitology 35:
325-352.
Cliff, G. M., R. C. Anderson, and E E Mallory. 1978.
Dauerlai-vae of Pclodera stronglyoides (Schneider,
1860) (Nematoda: Rhabditidae) in the conjimctival
sacs of lemmings. Canadian Journal of Zoology 56:
2117-2121.
D.wis, W. B. 1939. The Recent mammals of Idaho. Ca\ton
Printers, Ltd., Caldwell, ID. 400 pp.
Doss, M. A., M. M. Farr, K. E Roach, and G. Anastos.
1974. Inde.x-catalogue of medical and veterinaiy zool-
ogy. Special Publication No. 3. Ticks and tick-borne
diseases. I. Genera and species of ticks, part 2. Gen-
era H-N. U.S. Department of Agriculture, Agricul-
tural Research Service, Washington, DC. 593 pp.
Gavin, T. A., P VV. Sherman, E. Yensen, and B. May Pop-
ulation stracture and gene flow among disjunct pop-
ulations of Idaho ground squirrels {Spermophilus
l)niniH'tis), with reference to otlu-r species of SjU'r-
iiiDpliilits. Submitted.
Cii.L, A. E., and E. Yensen. 1992. Biocliemical dilferenti-
ation in the Idaho ground s(|uirrel, Siwrinophilus
hninnciis (Rodentia: Sciuridae). (weat Basin Natural-
ist ,52: 1.5.5-159.
Hilton, D. E J., and J. L. Mahrt. 1971. Ectoparasites
from three species of Spennopliilus (Rodentia: Sciuri-
dae) in Alberta. C'anadian [ournal of ZoologN 49:
1,501-1504.
Hoffmann, R. S., G. G. Anderson, R. W. Th(jrinc;ton,
Jr., and L. R. Heaney 1993. Family Sciuridae. Pages
419-465 in D. E. Wilson and D. M. Reeder, editors,
Mammal species of the world. 2nd edition. Smith-
sonian Institution Press, Washington and London.
1206 pp.
Holekamp, K. E. 1983. Pro.ximal mechanisms of natal dis-
persal in Bclding's ground scjuirrel {Spenni)philu.s
heldmgi hcldin' colonies have
been found in cavities of both live and dead
trees in California (Rainey et al. in press).
Despite these records, a clear understanding
of siKer-haired bat roosts and roost habitat is
still lacking.
To better understand the roost requirements
of siKer-haired bats, we investigated roost
selection b\ the silver-haired bat in the Black
Hills of South Dakota. Although forests in this
region have been intensu ely managed for tim-
ber (Boldt and Van Deusen 1974), silver-haired
bats are relatively abundant compared to the 9
other bat species present in the region (Matt-
son 1994). Although Mattson (1994) captured
twice as many males as females, pregnant or
lactating females were not uncommon. Our goal
was to characterize roost selection by silver-
haired bats in terms of attributes potentially
affected by cunent forestry practices.
Study Area
Our study area is located in the southern
Black Hills of South Dakota near the town of
Custer (43°46'N, 103°35'W). Most of the study
area is in the Black Hills National Forest and
occurs at elevations from 1360 to 1985 m asl.
The topograph)- of the area varies from rolling
highlands with parklike valleys to narrow, steep
canyons with rock\' ridge tops. The climate of
the Black Hills differs from the surrounding
semiarid plains in that it is moister and less
subject to temperature extremes. Average maxi-
mum temperature at Custer in JuK is about 23°
C, while mean annual precipitation is 457 mm.
The forests of the area are dominated by pure
stands of ponderosa pine (Pimis ponderosa).
Small stands of quaking aspen (Populus tremu-
loides) precede ponderosa pine on disturbed
sites. Paper birch (Betida papyhfera) grows in
small clusters in more mesic sites, whereas
'Department of Zoology and Physiolog\, L'ni\ersit> of Wyoming, Laramie, V\T S2071-.3166.
^Present address: PIC Technologies, Inc., .309 South 4th Street, Suite 201, Laramie. \\T 82070.
241
248
Great Basin Naturalist
[Volume 56
Rocky Mountain juniper {Junipenis scopiiloruin)
grows on diy ridges.
The forests of the Black Hills have been
managed for timber production since logging
first began in the 1870s. During the past 100
yr, most areas have been cut once, and many
have experienced multiple partial cuts (Alexan-
der 1987). In all, nearly 12 X 10^^ m'^ of timber
has been removed. Only a few small scattered
stands of unharvested forest remain (Boldt and
Van Deusen 1974). Although clearcutting was
once the primary means of harvest, shelter-
wood cutting, a method using a series of cuts,
is now standard.
We delineated two 10.1 X 10.1-km study
sites in areas in which we located silver-haired
bat roosts. The Jewel Cave Study Site encom-
passes Jewel Cave National Monument and
adjacent areas of the Black Hills National For-
est. The Hazelrodt Study Site is located south-
east of Custer on national forest land and
Custer State Park. Much of the Hazelrodt
Study Site burned during a fire in 1990 that
covered over 5670 ha.
Materials and Methods
Capture and Tracking Techniques
Silver-haired bats were captured using mist
nets set above small ponds and streams be-
tween 25 June and 4 August 1994. We deter-
mined the sex and reproductive condition for
all captured bats using external features (Racey
1988). Bats were classified as adult or juvenile
based on fusion of the epiphyseal-diaphyseal
suture of the finger bones (Anthony 1988).
We attached 0.7-g radio transmitters (model
BD-2B, Holohil Systems Ltd., Woodlawn,
Ontario) to 4 adult males and 12 adult females.
After fur had been trimmed from the bats,
transmitters were attached to the area between
the shoulder blades using a cyanoaciy late -based
glue (Fing'rs, Camarillo, CA). Bats to which
transmitters were affixed weighed 11-14 g, so
that transmitters represented 5-6.4% of body
mass, slightly over the 5% maximum recom-
mended by Aldridge and Brigham (1988). We
did not use any other marking technique to
identify individuals.
Hand-held, 3-element yagi antennas and
portable receivers (model TR-2, Telonics, Mesa,
AZ) were used to track bats to roost trees. If
we were unable to determine where in the tree
the bat was roosting, or whether it was alone or
with others, we returned to the tree before
dusk to watch and count bats leaving the site.
We attempted to approach the tree quietly to
reduce disturbance. We used a bat detector
(Bat Box III, Stag Electronics, St. Agnes, Eng-
land) to listen for echolocation calls. These,
along with body size and flight pattern, were
used to confirm that bats in a given roost were
only silver-haired bats.
Roost Measurements
We located 18 roost trees in the Jewel Cave
Study Site and 21 in the Hazelrodt Study Site.
When possible, the type of roost (i.e., wood-
pecker cavity, crevice, loose bark, etc.) was
recorded. Each roost tree was classified as
being used by either a maternity aggregation
or solitary bats. Maternity roosts, located by
tracking pregnant and lactating females, always
contained 6 or more bats. Solitan' roosts con-
tained only a single bat and were located by
tracking males or females that did not appear
pregnant or lactating or were post-lactating.
We categorized the aspect of the roost exit as
northeast (0-89°), southeast (90-179°), south-
west (180-269°), or northwest (270-359°).
Each roost tree was identified to species
and its height and diameter at breast height
(dbh) measured. We placed each roost tree into
1 of 7 decay stages; decay stage 1 included live
trees with intact bark and branches, whereas
decay stage 7 included dead trees beginning to
decompose with broken tops and no loose bark
(Thomas et al. 1979).
Plot Measurements
Within a 5-m-radius (78-m-) circular plot
centered at each roost tree, we measured aver-
age tree size, total basal area, and snag densit)'.
Trees were defined as standing woody stems
>1.5 m in height and >10 cm dbh. We also
recorded whether disturbance b\' fire or log-
ging had taken place in each plot. Disturbance
by fire was considered to have occurred if there
was any charred woody material in the plot,
and disturbance by logging was noted if we
obserxed any saw cuts on wood\' material in
the plot.
To compare characteristics of roost site plots
with Hie sunounding areas, we located four 5-
m-radius neighborhood plots for each roost
plot and recorded the same information as for
roost plots. We located the center of the neigh-
borhood plots b\- pacing 100 m from the roost
1996]
Sil\eh-iiaihl;d Bat K(x:)sts
249
tree in each of the cardinal directions (north,
south, east, west) and then pacing an adchtional
30 ni in a randomly selected direction.
We measured elexation and distance to the
nearest source of Water loi' each loost tiee using
topographic maps (7.5 minute series, US(iS,
Denver, CO). For comparison, we randoniK
located a point in the Jewel C^ave StucK Site or
Hazelrodt Study Site lor each roost tree found
in that site. To examine roost site selection on a
larger scale, we calculated the number of snags
in all neighborhood plots to estimate snag den-
sity for the stud)' site generalK. This estimation
was made by dividing the total ninnber of snags
in the 156 neighborhood plots by their total
area. The fire in the Hazelrodt Study Site in-
flated snag densities in this area. To remove the
influence of fire, we calculated snag densities
witliin the study sites by removing the 77 neigh-
borhood plots that had been disturbed b\' fire.
Analysis
Chi-square tests for goodness-of-fit (Jelinski
1991) were used to compare obsei"ved with ex-
pected roost aspects and tree decay stages by
roost t)^pe (maternity vs. solitaiy). For the latter
test, because of small sample size, we pooled
the roost trees into 3 decay stage categories:
stage 1-3, stage 4, and stage 5-7.
To compare continuous attributes between
roost plots and neighborhood plots, we sub-
tracted attribute means for the 4 neighborhood
plots from corresponding means for the roost
plots. So, each roost plot was compared only to
its 4 neighborhood plots. We tested the null
hypotheses that the mean differences did not
differ from 0 using paired t tests. Chi-square
tests for homogeneity (Jelinski 1991) were used
to compare obsei-ved with expected distiu-bances
at roost plots. Expected disturbances were based
on the proportion of neighborhood plots that
had burned or been logged. We used 2-sample
t tests to compare the means for elevation and
distance to nearest water for roost sites and
random sites. To avoid type I errors that may
result fi-om using a number of inferential statis-
tical tests with the same predictor variable, we
arbitrarily set oc = 0.025.
Results
Roost Attributes
We radio-tracked 16 bats for a mean of 8 d
(range: 1-20) and located 39 roosts, all of
which occurred in trees. Nine adult females
were tracked to 10 trees that were used by
maternity aggregations averaging 22.2 ± 4.9
(.v^) indi\iduals (range: 6-55). Three other
females and 4 adult males were tracked to 25
roost tiees, none of which were used by mater-
nity aggregations. Three of the females that
originally used maternity aggregations were
lal(>r followed to 4 trees where they roosted
alone. Maternity roosts were found exclusively
in tree cavities, pinmarily those created l)y wood-
peckers (Picidae). Cavity opc-nings were 7.5-10
cm in diameter. Solitary bats roosted under
loose bark {n = 15), in a tree crack or crevice
{n = 5), or in a woodpecker cavity (n = 1). We
could not determine the specific roost location
for 8 trees. These trees were placed in the soli-
tary category because bats tracked to these 8
trees were always observed roosting alone at
other trees. Maternity roosts were 10.2 ± 1.5 m
(range: 3.1-13.8) aboveground. The height of
measured solitaiy roosts averaged 3.4 ± 0.5 m
(range: 0.9-8.9). Cavity openings of maternit>'
roosts and solitaiy bat roosts were found more
frequently on the south side of tree boles over
other aspects (x^ = 15.8, d.f = 3, P = 0.001).
Of 39 roost trees, 38 (97%) were ponderosa
pine and 1 (3%) was aspen. Of 508 trees on
neighborhood plots, 483 (95%) were ponderosa
pine and 25 (5%) were other species: aspen,
juniper, and paper birch. The 10 trees used by
maternity aggregations of silver-haired bats
ranged from decay stage 2 to 7 (median = 5).
The 29 trees used by solitar)' bats varied from
tree decay stage 3 to 7 (median = 4). Trees in
neighborhood plots ranged from decay stage 1
to 7 (median = 1). Bats in maternity aggrega-
tions selected roost trees in significantly differ-
ent decay stages than solitaiy roosting bats (x^
= 10.2, d.f = 2, P = 0.0062; Fig. 1). Roost trees
averaged 14.2 ± 0.9 m (range: 3.7-24.1) in total
height, and 39 ± 2 cm dbh (range: 13-63). They
averaged 17 ± 2 cm larger in dbh than neigh-
borhood trees. The 10 maternity roost trees
averaged 44 ± 4 cm dbh (range: 29-62), 24 ± 4
cm larger than neighborhood trees. The 29 soli-
taiy roost trees averaged 37 ± 2 cm dbh (range:
12-55), 15 ± 3 cm larger than neighborhood
trees. Matemitv and solitaiT roost trees did not
differ in diameter {t = 1.64,'? = 0.12).
The 9 bats found in maternity aggregations
returned to the same roost tree for a mean of
8 d (range: 1-21). We tracked 1 bat fi-om a tree
containing a maternity aggregation of 55 bats
250
Great Basin Naturalist
[Volume 56
1-3 4 5-7
Tree Decay Stage Categories
Available (n = 132) \^7/\ Aggregation (n = 10) ^^ Solitary (n=29)
Fig. 1. Percentages of trees in each tree decay stage categoi^ used by maternity aggregations and solitaiy roosting silver-
haired bats, and available trees in tlie Black Hills, South Dakota, June-August 1994.
to a 2nd tree with a maternity aggregation of
44 bats about 440 m away. The following eve-
ning no bats were observed exiting from the
1st roost tree, but it is not clear how many bats
from the 1st roost tree moved to the 2nd tree
with the bat we were tracking.
We tracked 10 bats that used solitaiy roosts
to a mean of 3 solitaiy roost trees (range: 1-6).
For the most part, these bats switched trees
daily. However, on 5 occasions solitary bats
used the same tree on consecutive days. Three
of the 7 solitaiy roosting bats that we followed
to multiple trees returned at least once to trees
they had used several days before. Solitary
roosting bats traveled a mean of 405 ±93.7 m
(n = 13) between successive roost trees. Radio-
tracked bats traveled a mean of 2060 ± 440 m
(n = 12) from the captme point to their first
roost tree, significantly farther (/ = 3.67, P =
0.004) than the distance between successive
roost trees.
Plot Attributes
Roost plots had 1.7 ± 0.6 more live trees {t
= 3.09, ? = 0.004) than neighborhood plots.
Live and dead trees on roost plots were 6.5 ±
1.7 cm larger in dbh on average than those on
neighborhood plots {t = 3.77, P = 0.0006).
Roost plots also had basal areas of both lixe
and dead standing trees that were 14.07 ± 3.46
cm^/m^ greater {t = 4.06, P = 0.0002) than
neighborhood plots. Neither fire disturbance
()^2 = 0 005, d.f = 1, P = 0.94) nor logging
disturbance (x^ = 2.72, d.f = 1, P = 0.099) dif-
fered between roost and neighborhood plots.
Maternity and solitan' plots did not differ in the
attributes studied (Table 1). Roost trees tended
to be located higher in ele\ation than random
points (/ = 1.67, P = 0.10). Roost sites were
significantlv farther from water than random
points {t = 2.78, P = 0.007).
Using all 156 neighborhood plots, we calcu-
lated snag densit)' for the area to be 117 snags/
ha. After removing 77 neighborhood plots that
were disturbed by fire, we recalculated snag
densities to be 21 snags/ha.
Discussion
Roosts used by maternity aggregations dif-
fered from those used b\' solitaiA sil\ er-haired
1996J
SlLVER-llAlHED BaI' RoOS TS
251
Table 1. Conipadson l)(>t\vfeii solitan aTuI iiiatfniit\ roosi plot attributes in tlit- Black Mills, South Dakota, Jiiiu'-
Aiigust 1994.
Attribute
Solitary
(;i = 29)
Maternitv
in = 10)
r \ali
Li\ e trees
(no./plot)
Snags
(no./plot)
Mean tree dbh
(cm)
Total basal area
(cni^/m-)
4.5 ± 0.6
5.2 ± 0.7
0.72
0.47
2.1 ±0.5
2.2 ± 0.5
0.15
0.88
26.7 ± 3.2
27.5 ± 1.9
0.22
0.83
17.8 ±1.3
25.3 ± 4.7
1.54
0.13
bats. Mateniih' aggregations always used a hol-
low ca\ity within a tree bole. Usually these
cavities were created by woodpeckers, likely
haiiy woodpeckers {Picoides villosus) or black-
backed woodpeckers {P. arcticus), based on the
size of the openings (Terres 1980). Although
rare in the Black Hills (Black Hills National
Forest 1989), Lewis' woodpeckers {Melanerpes
lewis), northern flickers {Colaptes aiiratiis), or
three-toed woodpeckers {Picoides tridactyhis)
may have excavated some of the cavities. Soli-
taiy roosts were located under loose bark or in
a natural crack or crevice in the tree bole. Only
once did a solitary bat use a woodpecker cavity.
Although silver-haired bats are cryptically col-
ored, they were never observed roosting openly
on a tree trunk or limb, or in foliage. This be-
havior differs from other cryptically colored,
tree-roosting bats (e.g., Lasiiirns spp.), which
tend to roost among tree foliage (Shump and
Shump 1982a, 1982b). Roosts required by ma-
ternity aggregations may limit silver-haired bat
abundance; clearly trees with cavities are less
available than are those without. Reproductive
females seem to require roosts that provide a
relatively enclosed and unexposed space for
protecting young from predators or maintain-
ing the necessary thermal environment.
Cavity openings of maternity roosts and
solitary bat roosts occurred more frequently
than expected on the south side of tree boles.
We hypothesize that these roosts are warmer
than sites facing north because of insolation
and that these differences result in energetic
savings, providing more energy for growth and
development (McNab 1982). Reller (1972) has
shown that several species of woodpeckers ori-
ent their nest cavity openings southwesterly for
warming by the sun and/or ventilation by the
wind. However, it is unclear whether bat use
of cavities with south-facing entrances reflects
the selections of bats or woodpeckers.
SiKer-haired bats roosted exclusively in trees
during the summer Although all but one of the
roosts were located in ponderosa pine trees,
the dominance of ponderosa pine in our study
area prevented us from testing for tree species
preference. The wide geographic distribution
of silver-haired bats relative to that of pon-
derosa pine and the use by silver-haired bats of
both coniferous and deciduous roost trees in
other parts of their range (Novakowski 1956,
Parsons et al. 1986, Barclay et al. 1988, Camp-
bell et al. in press, Rainey et al. in press) sug-
gest that these bats select for the structure of
the roost itself rather than for a particular tree
species. As for other tree-roosting bats (Tide-
mann and Flavel 1987), it is imlikely that tree
species is important to silver-haired bats ex-
cept that at the local level 1 species ma\ tend
to have preferred attributes.
Roost trees were standing, dead, and larger
than average in diameter. The single living tree
selected as a roost was dying (stage 2) and
missing its top; it also had many dead limbs
and several woodpecker holes high in the bole.
There was an obsei-ved difference in tree decay
stage between roost trees used b)' maternity
aggregations and solitary bats. Solitaiy roosting
bats frequently used trees in decay stage 4,
which are characterized by the presence of
loose bark. Alternatively, mateniit>' roosts tended
to be found in older, more decomposed trees
(decay stages 5-7), trees that are more com-
monly used by excavating woodpeckers (Thomas
et al. 1979). Although the importance of snags
as roost sites in other forest types remains in
question, large snags appear to be important
resources for silver-haired bats in ponderosa
pine forests.
252
Great Basin Naturalist
[Volume 56
Clearly, solitan' roosting silver-haired bats
switch roosts regularly. This lack of fidelity may
be related to the abundant nature of potential
roosts (Brigham 1991) or a predator-avoidance
strategy (Kunz 1982b). Because they will return
to roost trees used several days previously and
these roosts are often close together, solitary
bats may use a series of trees in the same area
and thus maintain a level of site familiarity.
Conversely, maternity aggregations tend to re-
main in the same roosts for longer periods.
This may be related to the less abundant nature
of tree cavities and the importance of retaining
roosts that are suitable for raising offspring. At
least some of the mateniit\' aggregations appear
to swatch roosts during the reproductix e period.
The reason for this is not clear, although it may
involve predator or ectoparasite avoidance
(Lewis 1995).
We expected bats to select roosts relatively
close to water bodies, minimizing energetic
costs of moving between roosting areas and
areas potentially used for drinking and forag-
ing. Although trees were al^undant in the study
sites, bats traveled an average of >2 km fi"om
point of capture to their 1st roost tree, and sig-
nificantly farther from water than expected
randomly. This seems to support other avail-
a])le evidence for insectivorous bats in that
roost site location is not strongly influenced by
commuting costs (Fenton et al. 1985, Brigham
1991). Roost sites located farther from water
tlian random points appear puzzling but may rep-
resent the large number of roost trees located
along hill or ridge tops, sites with potentially
higher snag densities.
Silver-haired bat roost trees were foimd at
sites that differed fi^om nearby areas in a num-
ber of attributes. Roost plots differed in having
more, large trees and hence a higher total basal
area than suiTounding plots. Roost trees located
in areas that are ideal for tree growth or are
logged infi-ecjuently might explain why the roost
plots have more, larger trees.
Undoubtedly, snags are important in pro-
viding roost sites for silver-haired bats in the
Black Hills. As suitable roosts are critical
resources for bat sui-vival (Kimz 1982b), snag
availability likely influences the distribution
and abimdance of this species. Forest stands
containing silver-haired bat roosts had snag
densities ol 21 snags/ha, a value much higher
than current management objectives. These
densities were even higher in the Hazelrodt
Study Site, an area with a large number of fire-
killed trees. How fire suppression and logging
practices have affected the number of snags in
the Black Hills remains unclear; however, early
photographs suggest that many forested areas
were more open with many standing dead
trees (Knight 1994). Because snags are used for
nests or roosts by a large number of vertebrate
species (Thomas et al. 1979), reduced snag
densities may increase interspecific competi-
tion. We hypothesize that forest management
practices that reduce snag densities will lead to
declines in local silver-haired bat populations.
Acknowledgments
Funding was proN'ided b>' the University of
W\'oming-National Park Senice Research Cen-
ter; the National Park Ser\'ice, Rocky Moun-
tain Region; and the National Biological Sur-
vey, Midcontinent Ecological Research Center.
We especially appreciate support from Kate
Cannon and the staff of Jewel Cave National
Monimient, from Mike Bogan of the National
Biological Sui^vey, and from Joel Tigner, Oscar
Maitinez, and Alice Lippacher of the Black Hills
National Forest. Mike Bjelland and Jay Grant
provided valuable field assistance. We thank
Thomas H. Kmiz and R. \hirk Brigham for com-
menting on an earlier draft of the manuscript.
Literature Cited
Aldrioge, H. D. J. N., and R. M. Bkicham. 1988. Load
earning and nianeu\erabilit> in an insectivorous hat;
a test of tlie .5% "rule of radio-telemetn; Jonrnal of
MammalogN' 69: 379-;382.
Alexander, R. R. 1987. SiKieultural systems, cutting
methods, and cultmal practices for Black Hills pon-
derosa pine. US DA Forest Senice, General Technical
Report RM-139. Rock\ Mountain Forest and Range
Experiment Station, Fort Collins, CO. 32 pp.
Anthony, E. L. R 1988. Age determination in hats. Pages
47-.58 in T. H. Kunz, editor. Ecological and hehav-
ioral methods for the study of bats. Smithsonian
Institution Press, Washington, DC.
Barbour, R. W., and W. H. Da\IS. 1969. Bats of .\nierica.
University^ of Kentucky, Le.xington. 286 pp.
Barclay, R. M. R., E A. Faure, and D. R. Farr. 1988.
Roosting behavior and roost selection by migrating
sil\ er-haired bats {Lasionycteri.s iioctivagans). Journal
of Manmialogy 69: 821-825.
Blvck Hills National Forest 1989. Black Hills National
Forest: checklist of birds. USDA Forest Senice.
Boldt, C. E., and J. L. Van Del sen. 1974. Silviculture of
ponderosa pine in the Black Hills: the status of our
knowledge. USDA Forest Service, Research Paper
RM-124. Rocky Mountain Forest and Range Experi-
ment Station, Fort Collins, CO.
1996]
Silm:k-iiaired Bat Roosts
253
Bhic.ham, R. M. 1991. Flexibility in tbniging and roostinj^
hehavioin- b\- the I)ig brown bat (Ei)t('siciis fiisctis).
Canadian Journal ofZoolou;\ 69: 117-121.
Campbell, L. A., J. C. IIai,li-:t, and M. A. O'C^onneli,. In
pre.ss. Consenation of bat.s in managed fore.sts: use ol
loost.s by Lasionyctcris n()rtiva' hoar\' and silver-
haired bats in Oregon. Murrelet 69: 21-24.
Hacev, P a. 1988. Reproductive assessment in bats. Pages
31-45 in T. H. Kunz, editor. Ecological and behav-
ioral methods for the study of bats. Smithsonian
Institution l^ress, Washington, DC.
Rainev, W. E., J. Sh'ehek, and R. M. Mii.i.ek. In press.
Colonial maternity roosts of the silver-haired bat
{Lasionycteris noctivagam) in northern California
forests. American Midland Naturalist.
RelleR, a. W. 1972. Aspects of behavioral ecology of lii'd-
headed and lied-bellied Woodpeckers. American
Midland Naturalist 88: 270-290.
Shump, K. a., and a. U. Shumr 1982a. Lasiurits borealis.
Mammalian Species 183. American Society of Mam-
malogists, New York. 6 pp.
. 19H2h. Las iunis cine re us. Mammalian Species 185.
American Society' of Mammalogists, New York. 5 pp.
Tehres, J. K. 1980. The Audubon Society: encyclopedia of
North American birds. Wings Books, Avenel, NJ.
1109 pp.
Thomas, D. W 1988. The distribution of bats in different
ages of Douglas-fir forests. Journal of Wildlife Man-
agement 52: 619-626.
Thomas, J. W, R. G. Anderson, C. Maser, and E. L. Bull.
1979. Snags. Pages 60-77 in J. W. Thomas, editor.
Wildlife habitats in managed forests; the Blue Moun-
tains of Oregon and Washington. US DA Forest Ser-
vice, Agricultural Handbook 553.
TiDEMANN, C. R., AND S. C. Fl.\\el. 1987. Factors affect-
ing the choice of diuiTial roost site by tree-hole bats
(Microchiroptera) in south-eastern Australia. Aus-
tralian Wildlife Research 14: 459-473.
Received 13 December 1995
Accepted 1 7 May 1996
Great Basin Naturalist 56(3), © 1996, pp. 25-J-260
PERCEPTIONS OF UTAH ALFALFA GROWERS ABOUT WILDLIFE
DAMAGE TO THEIR HAY CROPS: IMPLICATIONS FOR
MANAGING WILDLIFE ON PRIVATE LAND
Teny A. Messmer^ and Sue Schroeder-
Abstract. — We conducted a survey of Utali alfalfa (Mcdicago sativa) growers in 1993 to identify- wildlife damage
problems to hay crops. Such surveys can provide wildlife managers with important insights regarding landowners'
wildlife damage management concerns and needs. Pocket gophers (Thomonnjs spp.) and mule deer [Odecoileus
hcmkmus) were perceived by growers as causing the most damage. Respondents reported a total annual loss of $350,000
or $24.79/ha (2.8% of the total crop value) because of wildlife damage in alfalfa crops. Decreased hay quantity' was the
most fiequently cited problem caused by wildlife. Compensation and incentive programs were preferred over assistance
and information programs for managing wildlife damage in alfalfa crops.
Key words: wildlife damage perceptions, alfalfa growers, wildlife damage management, wildlife manageinent.
Alfalfa is an important livestock forage. In
1994 over 58 million tons of alfalfa ha\' were
harvested in the U.S. on 9,802,400 ha of pri-
vately owned land. This represents over 40% of
the hay hai^vested as livestock forage (National
Agricultural Statistics Sei"vice 1995).
Alfalfa hay is the most important cash crop
grown in Utah. In 1994 Utali farmers harvested
2,205,000 tons of alfalfa on 210,000 ha of pri-
vately owned land. This crop was worth $158
million (Gneiting 1994).
Rodents, lagomoiphs, ungulates, and water-
fowl can impact alfalfa production (Piper 1909,
Sauer 1978, Luce et al. 1981, Dunn et al. 1982,
Packam and ConnolK' 1992, Austin and Urness
1993, Conover 1994). Big game grazing of alfalfa
during the growing season creates conflicts be-
tween growers and wildlife managers (Austin
and Umess 1993).
Conflicts also may arise between landown-
ers and wildlife managers because of differing
perceptions about the extent of wildlife damage
in cultivated crops. Farmers ma\' feel that wild-
life managers are unaware of the extent of crop
losses caused by wildlife and hence are insen-
sitive to their needs (Decker et al. 1984,
Conover and Decker 1991). Crop owners con-
cerns about wildlife damage strongly affect
how the agricultiual conmnniity will respond
to environmental issues and whether federal or
state wildlife programs aimed at maintaining or
improving wildlife habitat on private property
will succeed (Conover 1994).
There is consensus among professionals
working for federal and state wildlife and agri-
cultural agencies that wildlife damage reduces
the profitability of U.S. agriculture (Conover
and Decker 1991). Professionals agree tliat wild-
life depredation has increased over time but
disagree over the seriousness of the impact.
Although the actual costs associated with wild-
life depredation are difficult to estimate and
can differ on each farm or ranch and crop t^pe
(Tebaldi and Anderson 1982, Austin and Umess
1987a, 1987b, 1989, 1993, Lewis and O'Brien
1990), landowners have demonstrated an abil-
ity' to accurately assess crop losses caused by
wildlife (Decker et al. 1984, Conover 1994,
Mch'or and Conover 1994a). Crop losses and
potential future losses caused h\\ or related to,
the presence of wildlife must be assessed to
determine if control is warranted (Rennison
and Buckle 1988).
Several Great Basin states including Utah,
Wyoming, Colorado, New Mexico, Nevada,
Idaho, and Arizona have enacted laws to com-
pensate crop owTiers for wildlife -caused dam-
age (Musgra\'e and Stein 1993). These actions
have been initiated largely in response to con-
stituent concerns oxer the economic impact of
depredating wildlife, particularh' big game, in
cultivated crops.
' l)(_ixntiiii-iil ol Hslu-i ii's ami WiUllilr, Utali StaU- Lni\ersit\. Lo.uan, UT S4322-.5210.
-OepailiiKMil of Foicsl Resources, Utah State Llniveisit\', Logan, UT 84322-.5215.
254
1996]
VViLDLiFK Damage to Alfalfa
255
Crop owTiers in lltali ma\' destroy depredat-
ing big game animals if the animals are not
removed by the Utah Division of WildHfe
Resources (UDWR) within 72 h of notification
(Chapter 183, Utah Code 1993a). Utah crop
owners also may receive monetary compensa-
tion for damage caused by big game animals
(Chapter 307, Utah Code 1994b) and ring-
necked pheasants {Pluisianiis colcliiciis: Chap-
ter 46, Utali Code 1971).
We surveyed Utah alfalfa growers to deter-
mine their perceptions regarding wildlife dam-
age to hay crops. Such surveys can provide
wildlife managers with important information
regarding landowner wildlife damage manage-
ment needs and concerns (Conover 1994).
Methods
We sunexed 334 alfalfa growers (4% of all
alfalfa growers in Utah) whose names were on
the Utah Department of Agriculture's (UDA)
1993 Hay List. The UDA maintains this list to
provide information to individuals who contact
the department about purchasing alfalfa hay in
Utah. The UDA updates this list each Januaiy
We included a 2-page wildlife damage sur-
vey in a UDA mailing sent to the growers. In
addition to the survey, growers received a cover
letter, the UDAs questionnaire, and a business
reply envelope. The cover letter stated that if
no response was received within 30 d, the
grower's name would be removed fi"om the hay
list. A follow-up letter was sent to nonrespon-
dents 3 wk after the initial mailing. Those fail-
ing to respond to the 2nd mailing were removed
from the hay list.
The sui-vey contained questions about the
growers' experiences witli wildlife in their alfalfa
crops. Growers were asked to identify wildlife
species causing damage to hay crops, type of
damage, their annual monetary loss from wild-
life damage, specific damage control techniques
employed on their fanii to control wildlife dam-
age, whether they received any type of damage
compensation or assistance, who they contacted
for assistance and information, and what type
of information and programs they found most
useful in managing wildlife damage. Further,
growers were asked to rate on a scale of 0 to 5
(0 = no cost through 5 = high cost) relative
losses caused by different wildlife species to
their alfalfa crops and the costs associated with
common management practices used on their
farms and ranches.
Responses were stratified and analyzed bv
the number of hectares in alfalfa (0— fO, 41-80,
81-200, 201-400, and >4()0) and type of oper-
ation (inigated or diyland). Levere's tests were
us(>d to determine ('(jualit}' of variances by types
and sizes of alfalfa operation (SPSS 1995).
We assumed that alfalfa growers on the hay
list have the same values and perceptions as
the population of Utah alfalfa growers, 'lb deter-
mine if the hay list was statistically representa-
tive of Utah alfalfa growers, we compared the
mean alfalfa farm size and regional distribu-
tions of farms on the hay list with acreage cate-
gories reported by the UDA for all Utah alfalfa
farms (Gneiting 1994) using a Kruskal-Wallis
one-way analysis of variance. Differences in
these tests were considered significant if P <
0.05.
Results
Alfalfa Production
One hundred sixty-four completed ques-
tionnaires (49.1%) were returned, of which 150
(91%) were useable for analysis. Sun'e)' respon-
dents reported growing 16,867 ha of alfalfa, of
which 14,391 ha (85%) was irrigated and 2486
ha (15%) was dryland alfalfa. Irrigated alfalfa
farms ranged in size from 5 to 1062 ha. Dr>'-
land alfalfa farms ranged in size from 3 to 320
ha. All farms were family owned and operated.
Since the UDA hay list is relatively dynam-
ic, it contains infoniiation regarding the grower's
mailing address, telephone number, and inter-
est in selling alfalfa hay, but not the size and
type of operation. Information on alfalfa opera-
tions was obtained through the survey; thus,
we were unable to determine if there were any
significant differences between respondents
and nonrespondents.
Although the responses received consti-
tuted 2% of all Utah alfalfli growers {N = 7600),
our sample was representative of the popula-
tion based on mean farm size {H = 7.0; 7 df; P
= 0.001) and regional distribution. Utah alfalfa
acreage percentages reported by the UDA for
northern, central, eastern, and southern regions
were 30%, 31%, 19%, and 20%, respectively
(Gneiting 1994). Regional alfalfa acreage per-
centages for our sample were northern 27%,
central 34%, eastern 21%, and southern 18%.
256
Great Basin Naturalist
[Volume 56
Wildlife Species Present in
Utah Alfalfa Fields
Respondents reported 20 different species
of wildlife were present in their alfalfa fields.
Pocket gophers and mule deer were the most
abundant, being reported present on 124
(82.7%) and 120 (80.0%) farms, respectively.
Other wildlife species reported by farmers as
common in alfalfa fields included jackrabbits
{Lepus spp.; n = 89, 59.3%), ground squirrels
{Spermophihis spp.; n = 83, 55.3%), prairie
dogs {Cynomys spp.; n = 69, 46.0%), waterfowl
{Anatidae\ n = 66, 44.0%), elk {Cervus elaphus;
n = 62, 41.3%), pronghorn {Antilocopro ameri-
cana; n = 54, 36.0%), and voles [Microtus spp.;
n = 50, 33.3%). Wildlife species reported by
farmers as being less common in alfalfa fields
included marmots {Marmoto flaviventtis), bad-
gers {Tax idea taxiis), red foxes {Vulpes vidpes),
sandhill cranes {Gnis canadensis), Canada geese
{Branta canadensis), cottontail rabbits {Syvda-
giis spp.), deer mice {Pewmyscus manicukitus),
raccoons {Procyon lotor), ring-necked pheas-
ants, and muski"ats {Ondatra zibethica).
Monetary Losses Caused by Wildlife
One hundred nine growers (72%) reported
losing $350,000 (a^ = $3242, .s- = 526) be-
cause of wildlife damage in their alfalfa fields.
Monetaiy losses averaged $24.79/ha.
The average dollar loss reported by respon-
dents who grew only iiri gated alfalfa was $3016
{n = 86, Sy = 554). Respondents who grew both
irrigated and dnland alfalfa reported an aver-
age loss of $4388 {n = 21, .sy = 1525). Those who
grew only dryland alfalfa reported an average
foss of $3750 (n = 2, 5- = 250).
The highest losses per/ha were reported by
respondents who grew both irrigated and dr>^-
land alfalfa ($42 ha). Respondents who grew
only irrigated or dryland alfalfa reported losses
per/ha of $19 and $28, respectiveK'.
Growers with irrigated alfalfa farms >200
ha in size reported significantly higher mone-
taiy losses than operations <200 ha in size (F
= 15.5; 1,103 df; P < 0.001). Although the
average monetary loss reported by larger alfalfa
farms was $5078 (n = 50) compared to $1639
for smaller farms {n = 55), the average loss
per/ha was higher on smaller ($37) tlian larger
farms ($21; F = 24.9; 1,103 df; P < 0.001).
Growers reported no significant difference in
damage losses by size for irrigated/drvland
alfalfa farms (F = 0.4; 1,26 df ■ P = 0.52).
Respondents with alfalfa farms >80 ha re-
ported that rodents (F = 7.9; 1,107 df; P =
0.006) and ungulates (F = 18.2; 1,107 df; P <
0.001) caused higher monetaiy losses when
compared to smaller farms (<80 ha). No signif-
icant diflferences in monetaiy losses due to water-
fowl were detected bv alfalfa farm size (F =
0.006; 1,107 df;P = 0.940).
Relative Costs of Wildlife
Damage in Alfalfa Fields
Respondents ranked on a scale of 0-5 (0 =
no cost through 5 = high cost) the relative
damage costs associated with common wildlife
species reported in their alfalfa fields as fol-
lows: mule deer (2.9), pocket gophers (2.4), elk
(1.6), prairie dogs (1.4), ground squirrels (1.4),
jackrabbits (1.3), waterfowl (1.0), pronghorn
(0.7), and meadow voles (0.9). Respondents
with irrigated alfalfa farms >200 ha reported
that elk (F = 7.9; 1,56 df; P = .007) and prong-
horn (F = 7.5; 1,48 df; P = .008) caused signif-
icantly greater cost-related problems than on
smaller farms (<200 ha). Respondents with
diyland alfalfa farms >200 ha reported greater
significant cost-related problems caused by
jackrabbits (F = 14.1; 1,20 df' P = 0.001) and
mule deer (F = 8.5; 1,28 df; P = 0.007) than on
smaller farms (<200 ha). Sun'ey respondents
indicated that alfalfa production problems dif-
fered b)' specific wildlife species (Talile 1).
Farm and Ranch Management
Practice Comparisons
Respondents ranked on a scale of 0—5 (0 =
no cost through 5 = high cost) the relatixe cost
of the 7 farm management practices as follows:
irrigation (3.8), fertilization (3.4), weed control
(2.9), insect control (2.6), fencing (2.3), big
game control (2.0), and rodent/rabbit control
(1.9). Fertilization, weed control, and irrigation
were used on 82%, 81%, and 80% of the farms,
respectively. Big game and rodent/rabbit con-
trol were used by 71% and 38% of the respon-
dents, respectively. Respondents also reported
employing several techni(|ues to control wildlife
damage in alfalfa fields (Table 2). Based on
sizes and types ol alfalfa operations, the only
significant cost differences reported In' man-
agement practices were for irrigation on farms
>200 ha (F = 5.0; 1,124 df; P = 0.03).
1996]
WiLDLiKK Damage to Alfalfa
257
Table L Percentage of all respondents {N = 150) reporting problems caused by a specific wildlife species in Utah alfalfa
fields in 1993 and a breakdown of that percentage into subcategories based on the most severe tvpe of problem caused.
cies
i^ejiorting
i\'rcent
age identilying ;
speci
iic problem
i-s most
severe
Wildlife spi"
Hay
Hay
Equipmen
Increased
causing dan
age
problems
(juality
quantity
damage
costs
Pocket gophers
68.7
14.0
20.7
26.0
8.0
Ground sciu
nel
33.3
4.0
10.7
15.3
3.3
\bles
10.7
2.7
6.7
1.3
0.0
Jackrabbits
32.8
2.7
28.7
0.7
0.7
Prairie tlogs
23.3
0.7
8.0
13.3
1.3
Elk
20.0
6.0
12.7
1.3
0.0
Mule deer
64.0
8.7
54.0
1.3
0.0
Antelope
9.3
1.3
8.0
0.0
0.0
Waterfowl
17.3
2.7
14.7
0.0
0.0
Wildlife Damage Management
Assistance Programs
Fourteen respondents (9%) reported receiv-
ing compensation for wildlife damage in their
alialfa fields. Of these, 12 received compensa-
tion for damage caused by mule deer. Another
48 (31%) indicated they received some type of
technical assistance to control wildlife damage.
Most of this assistance (75%) was provided to
control damage caused by mule deer.
One hundred twenty-two respondents (80%)
reported seeking either information or assis-
tance in dealing with wildlife depredation prob-
lems. Conservation officers were cited by 53
growers (43%) as being their primary contact
for infoniiation or assistance. Countv' agents and
UDWR biologists ranked 2nd (22%) and 3rd
(18%), respectively. Other sources of informa-
tion in order of decreasing importance were
other landowners (7%), farm and ranch stores
(5%), and UDA agricultural representatives (3%).
Respondents preferred compensation and
incentive programs (42%) to other types of pro-
grams to manage damage caused by wildlife in
alfalfa fields. Research (17%), field demonstra-
tions (13%), workshops (13%), facts sheets (13%),
and videos (14%) were rated nearly equal in
usefulness.
Discussion
Relationship of Perceived Damage
Costs to Wildlife Management
Surveys can be cost-effective means of assess-
ing the magnitude and economic impact of wild-
life depredation (Crabb et al. 1986). Unfortu-
nately, due to the cost and time associated with
conducting reliable surveys, many wildlife
agencies are unable to perform this work on a
regular basis. Our experience suggests that
wildlife agencies should consider using state
agriculture department hay lists to conduct
benchmark sui^veys to identify wildlife damage
management concerns and needs. Most states
maintain hay lists (R. Parker, personal commu-
nication, UDA, 1995).
Our results summarize perceived losses.
The relationship between perceived and actual
losses is unclear and probably difficult to esti-
mate (Conover 1994). This relationship depends
in part on how conspicuous the damage appears
and which wildlife species causes the damage
(Wakeley and Mitchell 1981, Decker et al. 1984,
Mclvor and Conover 1994b).
Most respondents reported problems with
pocket gophers and mule deer. Other species
commonly causing problems included jackrab-
bits, ground squirrels, prairie dogs, waterfowl,
elk, pronghorn, and meadow voles. Conover
(1994) also found that these species, in particu-
lar deer, were perceived to cause most damage
to agricultural crops in the U.S.
Based on statewide averages, in 1993 Utah
alfalfa growers harvested 10.5 tons/ha with a
market value of $71.66 a ton. Survey respon-
dents produced 177,104 tons of alfalfa on
16,867 ha having a total value of $12,691,000.
The $350,000 loss reported due to wildlife rep-
resents 2.8% of the crop value. Expanding this
to the total value of alfalfa produced in Utah
during 1993 results in a total perceived loss of
$4.4 million. This is 9 times the amount the
Utah State Legislature annually appropriates
($500,000) to reimljurse crop owner depreda-
tion claims and expenses (Chapter 307, Utah
Code 1994b).
258
Great Basin Naturalist
[Volume 56
Table 2. Percentage of all respondents {N = 150) using a specific technique to control damage caused by wildlife
species in Utah alfalfa fields in 1993 and a breakdowTi of that percentage into subcategories based on the most effective
technique used.
Percentage identif\ ing a specific techniiiue
as being most efPectixe
Using
Wildlife species damage control Shooting/ Poison
causing damage techniques (%) Trapping hunting baits Fumigants Cultural Fencing Hazing
Pocket gopher
Ground squinel
Voles
Jackiabbits
Prairie dogs
Elk
Mule deer
Antelope
Waterfowl
41.7
6.7
0.0
33.0
2.0
0.0
0.0
0.0
45.4
4.7
17.3
22.0
0.0
0.7
0.0
0.7
13.3
2.0
2.7
7.3
0.0
1.3
0.0
0.0
39.3
0.0
36.0
2.7
0.0
U.O
0.7
0.0
24.0
2.0
12.7
7.3
1.3
0.0
0.7
0.0
21.3
0.0
12.0
0.0
0.7
0.0
7.3
1.3
46.7
0.0
22.7
0.0
0.7
0.0
16.0
7.3
9.7
0.0
4.0
0.0
0.7
0.0
--> -
1.3
16.7
0.0
13.3
0.7
0.7
0.0
0.0
2.0
Utah Code authoiizes the UDWR to imme-
diately pay any approved damage claims
< $500. Claims or total amounts of claims sub-
mitted by a claimant in the fiscal year that are
>$50() are not paid until the total amount of
approved claims for the fiscal )'ear is deter-
mined. If the amount claimed exceeds the
appropriation, the per claimant amounts paid
in excess of $500 are prorated. The current
appropriation falls short of satisfying wildlife
damage compensation claims and expenses
(R. Valentine, personiil communication, UDWR,
1996).
If 13^'^ of Utah alfalfa growers {n = 1000)
submitted appro\ed claims of $500, their
claims would deplete the annual appropriation.
Although the alflilfa growers we suneyed pre-
ferred compensation and incentive payments
over other types of wildlife damage manage-
ment programs, only 9% had e\er receixed an\'
financial support.
In the United States, 2.1 million farmers
control 400 million ha of our 937 million ha
land base. Their actions largely influence the
qualit)' and quantit) of the existing wildlife
habitat base (Cierard 1995). Landowners per-
ceptions and concerns about wildlife damage
are important because they influence their atti-
tudes and behavior toward wildlife. Conover
(1994) suggested that wildlife damage has
reached levels that discourage prixate land-
owners from managing for wildlife on their
property'. Our results suggest that Utah alfalfa
growers also perceive wildlife damage in alfalfa
fields as a serious concern. Although wildlife
professionals working for federal or state w ild-
life and agricultural agencies believe that wild-
life damage has increased in the last 30 \ r, our
sun'ey results reinforce Conoxer and Decker's
(1991) suggestion that programs necessar\' to
adequately address crop owner conceiTis have
not yet been implemented.
Role of State Agencies in ResoKing
Wildlife Damage Management Concerns
State wildlife management agencies are
responsible for managing damage caused by
big game, upland game, and waterfowl (Mus-
grave and Stein 1993). State agriculture de-
partments administer and enforce pesticide
control legislation that regulates the safe and
proper use of pesticides for vertebrate pest
damage. Because of tliis role, agiiculture depart-
ments have jurisdiction oxer the control of
unprotected wildlife species (xertebrate pests).
In Utah these include pocket gophers, field
mice, muskrats, ground squirrels, jackrabbits,
raccoons, skunks, red fox, and coyotes.
The UDWR recognizes that prixate lands
xvithin Utah proxide habitat for xvildlife and
that under some circumstances xxildlife may
cause economic losses to the landoxxnier. With
this understanding, the UDWR cooperates xxidi
the UDA and the U.S. Department of Agricul-
ture Animal Plant Health and Inspection Ser-
vice/Animal Damage Control (ADC) program
to conduct predator, bird, and rodent control
actix ities and compensate landox\ners for cer-
tain losses caused by xxildlife using funds
appropriated by the legislatine.
In 1994 the Utah legislature enacted an
alteinatixe compensation program that alloxx's
landoxxners to receix e pennits to han est antler-
less animals as mitigation for damage caused
1996]
WlLDLII i: llWIACli TO Al.KALFA
259
b\- big game (Cliapter 176, Utah Code 1994a).
In 1995 the UDW'R Southern Region issued
>12()() mitigation permits, of which 50% were
filled, in 1996 both the number ol tags issued
and number ol animals har\ ested declined as
landowners lost interest in the program (N.
McKee, personal connnunication, UIDWR,
1996).
To better address landowners concerns
gi\'en fiscal and legal constraints, we suggest
that agencies and organizations responsible for
managing wildlife resources and w ildlife dam-
age on Utah agricultural lands collaborate to
develop strategies that allow profitable agricul-
ture and wildlife to coexist. Utah's posted hunt-
ing unit (PHU; Chapter 288, Utah Code 1993b)
and \\ ildlife habitat authorization (WHA) pro-
grams (Chapter 75, Utah Code 1995) may offer
additional mechanisms to achieve this goal.
The Posted Hunting Unit Program
The UDWR also recognizes that wildlife can
be a significant benefit to the landowner The
PHU program provides landowaiers with mon-
etary incentives, through an allocation of hunt-
ing pennits, to include wildlife (small game,
waterfowl, and big game) in farm and ranch
management plans. Landowners who partici-
pate in the program are required to improve
wildlife habitat but are ineligible to receive
compensation for crop losses caused by wildlife.
The most successfiil of UtiilVs PHU programs
involves big game animals. In 1994, 47 big
game PHU programs, encompassing over
400,000 ha of private land, proxided additional
economic returns for hundreds of landowners
and hunting experiences for thousands of
hunters. Current program guidelines limit par-
ticipation to landowners or landowner groups
who own at least 4000 ha (Chapter 288, Utah
Code 1993b). The size limitation was estab-
lished to create more manageable herd units.
In our sui-vey, respondents reported that big
game animals caused the greatest damage. We
suggest that big game PHU guidelines be
modified to accommodate farm or ranch units
<4000 ha in size. This modification would pro-
N'ide the stimulus necessary to alleviate many
crop owners' wildlife damage concerns and
provide an additional incentive to include wild-
life in farm and ranch management plans. In
addition, we suggest that big game PHU oper-
ators be encouraged to incoiporate provisions
in their wildlife management plans to compen-
sate smaller nonparticipating landowners adja-
cent to their operation for crop damage- caused
by big game animals.
The Wildlife Habitat
Authorization Program
The WHA program rc(|uires persons 14 yr
ol age or older to purchase a wildlife habitat
authorization prior to purchasing certain hunt-
ing or fishing licenses or permits. The funds
generated from this authorization arc placed
into a restricted account to be used lor wildlife
habitat improvements. Several odier Croat Basin
states operate similar programs designed to
generate funds to do habitat work.
We reconunend that state wildlife agencies
consider using habitat funds to implement and
evaluate enhancement projects and programs
on public and private land that are designed
specifically to reduce big game depredation on
pri\'ate land. Habitat funds could be used to es-
tablish big game lure crops, situate interceptor
strips, or modify migration corridors as a means
of abating localized depredation problems.
Wildlife Damage Education Needs
Crop owners also need additional informa-
tion on techniques used to manage wildlife
damage. Several respondents reported using
fumigants and poison baits to control damage
caused by ungulates, lagomorphs, and birds.
These practices are illegal, as no products are
currently registered in the U.S. to control dam-
age caused by these species.
We recommend that state wildlife agencies,
agriculture departments, and federal ADC pro-
grams cooperate in the development of public
outreach, extension education, and research
activities intended to inform crop owners about
techniques that can l)e used to manage wildlife
damage. These programs also should provide
information on consei^vation technologies, non-
lethal strategies, and opportunities that can be
used to control wildlife damage and benefit
wildlife resources while maintaining or enhanc-
ing agricultural profitability.
In conclusion, previous studies conducted
in the Great Basin focused on evaluating the
effects of big game depredation (Tebaldi and
Anderson 1982, Austin and Urness 1987a,
19871), 1989, 1993) and sandhill cranes (Mclvor
and Conover 1994b) on agricultmal production.
Our study adds to this research b\' providing
important insights regarding crop owners'
260
Great Basin Naturalist
[Volume 56
perceptions about wildlife damage and their
needs and preferences in managing damage.
Our results suggest that Utah alfalfa grow-
ers perceive wildlife damage as a serious con-
cern. This concern should be shared by wild-
life managers.
In addition to informing landowners of their
concern over wildlife damage, wildlife man-
agers should demonstrate it by addressing
potentials for increasing damage on private
lands when developing wildlife habitat man-
agement plans (Conover 1994). Wildlife man-
agers also should incoiporate strategies in man-
agement plans to benefit wildlife and reduce
depredation potentials on private land.
Acknowledgments
We acknowledge R. Parker for assistance in
distributing the sui"vey. We thank M. Conover,
S. Barras, and A. Hall for reviewing earlier
drafts of this manuscript.
Literature Cited
Austin, D. A., and R J. Urness. 1987a. Consumption of
fresh alfalfa hay by mule deer and elk. Great Basin
Naturalist 47: 100-102.
. 1987h. Guidelines for evaluating annual crop
losses due to depredating big game. Utah Division of
Wildlife Resources, Publication 87-5. 42 pp.
. 1989. Evaluating production losses from mule
deer depredation in apple orchards. Wildlife Society
Bulletin 17: 161-165.
. 1993. Evaluating production losses from mule deer
depredation in alfalfa fields. Wildlife Society Bulletin
21:397-401.
Conover, M. R. 1994. Perceptions of grass-roots leaders of
the agricultiual community about wildlife damage on
their farms and ranches. Wildlife Societv Bulletin 22:
94-100.
Conover, M. R., and D. J. Decker. 1991. Wildlife dam-
age to crops: perceptions of agricultural and wildlife
professionals in 1957 and 1987. Wildlife Society- Bul-
letin 19: 46-52.
Crabb, a. C, T. P Salmon, and R. E. Marsh. 1986. Sunex s
as an approach to gathering animal damage informa-
tion. Pages 2-4 in Vertebrate pest control and man-
agement materials. American Society for Testing and
Materials STP 974, Philadelphia, Pa! 12 pp.
Decker, D. J., G. E Mattfieed, and T. L. Brown. 1984.
Influence of deer damage on farmers' perceptions of
deer population trends: important implications for
managers. Proceedings of the First Eastern Wildlife
Damage Control Conference 1: 191-195.
Dunn, J. P, J. A. Chapman, and R. E. Marsh. 1982. Jack-
rabbits. Pages 124-145 in Wild manuuals of North
America: biology, management and economics. John
Hopkins University Press, Baltimore, MD. 1147 pp.
Ger\RD, R W. 1995. Agricultural practices, farm iDolic\, and
the conservation ol biological (li\t'rsit\. Biological
Science Report 4. U.S. Department of Interior,
National Biological Sei-vice, Washington, DC 28 pp.
Gneiting, D. J. 1994. Utah agricultural statistics. Utah
Agricultural Statistics Service and the Utah Depart-
ment of Agriculture, Salt Lake Cit\'. 138 pp.
Lewis, S. R., and J. M. O'Brien. 1990. Survey of rodent
and rabbit damage to alfalfa hay in Nevada. Pages
166-117 in Proceedings of the 14th Vertebrate Pest
Conference. University of California, Davis. 320 pp.
Luce, D. G., R. M. Case, and J. L. Stubbendieck. 1981.
Damage to alfalf;i fields by plains pocket gophers.
Journal of Wildlife Management 45: 258-260.
McIvoR, D. E., AND M. R. Conover. 1994a. Perceptions of
fanners and non-famiers toward management of prob-
lem wildlife. Wildlife Societ>' Bulletin 22: 211-219.
. 1994b. Impact of Greater Sandhill Cranes forag-
ing on corn and barley crops. Agriculture, Ecosys-
tems, and Environment 49: 233-237.
MusGRAVE, R. S., A.ND M. A. Stein. 1993. State wildlife laws
handbook. Center for W''ildlife Law, Institute of Public
Law, University of New Mexico, Albuquerque. 840 pp.
National Agricultural Statistics Service. 1995. 1994
crop statistics. U.S. Department of Agriculture, Wash-
ington, DC.
Pack-am, C. J., and G. Connolly. 1992. Control methods
research priorities for animal damage control. Pages
12-16 in Proceedings of the 15th Vertebrate Pest
Conference. University of California, Davis. 300 pp.
Piper, S. E. 1909. The Nevada mouse plague of 1907-08.
Farmers Bulletin 352: 1-23.
Rennison, B. D., and a. P Buckle. 1988. Methods for
estimating the losses caused in rice and other crops
by rodents. Pages 69-80 in Rodent pest management.
CRC Press Inc., Boca Raton, FL. 238 pp.
Sauer, W. C. 1978. Control of the Oregon ground squiirel.
Pages 99-109 in Proceedings of the 7th Vertebrate
Pest Conference. University of California, Davis. 323
pp.
Statistical Programs for Social Sciences. 1995. Micro-
soft Windows Release 5.0. Microsoft Corporation,
Redmond, WA.
Teb.aldi, a., and C. C. Anderson. 1982. Effects of deer
use on winter wheat and alfalfa production. WVoming
Fish and Game Department. Job Final Report FW-3-
R-26. 78 pp.
Utah Code. 1971. Chapter 46. Section 23-17-5. Damages
for destroyed crops — Limitations — Appeal.
. 1993a. Chapter 183. Section 23-16-3. Damage to
cultivated crops by big game animals — Notice to
di\ ision — Crop owmer authorized to kill animals.
. 1993b. Chapter 288. Section 23-23-1. Posted hunt-
ing units.
. 1994a. Chapter 176. Section 23-16-3.5. Damages
to livestock forage, fences, or irrigation equipment on
private land.
. 1994b. Chapter 307. Section 23-16-4. Damages for
destroyed crops — Limitations — .\ppraisal.
. 1995. Cliapter 75. Section 23-19-42. Wildlife habi-
tat autliorization.
Wakelev, J. S., AND R. C. Mitchell. 1981. Blackbird dam-
age to ripening field corn in PennsvKania. Wildlife
Society Bulletin 9: 52-55.
Received 11 Septonber 1995
Accepted 10 April 1996
Great Basin Naturalist 56(3), © I99(i, iiji, 261-266
SPATIAL RELATIONSHIPS AMONG YOUNG CERCOCARPUS LEDIFOLIUS
(CURLLEAF MOUNTAIN MAHOGANY)
Brad W. Scliultzl, Kohin J. Tausch^, ami Paul T lueller^
AusTKACT. — Tilis stiul\ anaK/.t'd spatial location patterns of Cercncarpiis ledifoliiis Nutt. (cnrllcai nionnlain
niali()gan>) plants, classified as current-year seedling, estahlislied seedling, juvenile, and inmiatuie indi\ iduals, at a cen-
tral Nevada stncK site. Most current-year seedlings were located in mahogany stands in wliicli large, niatiue individuals
had the greatest ahundance. These stands had greater litter cover and a thicker layer of litter than areas with few cur-
rent-\'ear seedlings. Most estahlislied young Cercocarpiis were located in adjacent Artemisia tridentata ssp. vaseyana
(mountain big sagebrush) communities, or in infrequent canopy gaps between relatively few large, mature Cercocarpus.
We discuss potential roles of plant litter, root growth characteristics, nurse plants, and herbivoiy in the establishment
and renewal oi Cercociirpiis connmmities.
Key words: Cercocarpus, litter iiuniiitaiii inalio^dny, seedlin 1 yvM' ol age; 2-7 mm hasal
diameter; smooth hark; ma\ he u]) to
30 cm tall; S or i.. .^' leaxcs.
Young plants >7 nun hasal diameter;
smooth hark; plants to HO cm tali.
^oung plants >1.25 cm hasal diameter;
smooth hark; plants to 1.5 m tall.
Cracked hark; 1.5-3.0 m tall; crown
broadened; ma\' be nuiltistemmed from
base; not suppressed by adjacent larger
mountain mahogany plants.
Cracked bark; wide full crown; few dead
branches; may have several stems from
base; >3 m tall.
Cracked bark; ma\' be multistemmed;
numerous dead branches; ma)' be >3 m
tall; frequentK suppressed by adjacent
larger mountain mahogany plants.
and had the same aspect throughout its length,
and (4) all transects located in the same stand
were 40 m or more apart. Table 3 describes the
elevation, slope, and aspect of each transect.
Cercocarpiis in the Shoshone Range are largely
restricted to the Fo.xmount soil seiies (Carol Jett
personal communication), which is a gravelly
loam (specifically, a Loamy- skeletal, mixed Topic
Cryboroll). This soil is well drained and moder-
ately permeable. Depth to a paralithic contact
averages 60-100 cm.
All transects were located such that 20 m
occurred in the Cercocarpiis stand and 20 m in
the adjacent Artemisia community. Each tran-
sect was divided into forty 1 X 1-m quadrats.
Every Cercocarpiis rooted in each quadrat was
classified by maturity class. For Cercocarpiis in
established seedling, juvenile, and immature
maturity classes, we determined whether the
plant was rooted under the protective canopy
of a live or dead shrub.
Distribution of current-year seedling, estab-
lished seedling, juvenile, and immature Cerco-
carpiis was summarized for 10 classification
categories (populations). These were (1) the
number of Cercocarpiis in current-year seed-
ling, established seedling, juvenile, and imma-
ture maturity classes rooted in either the Cer-
Elevation
Slope
Aspect
Transect
(ill)
(%)
(degrees)
1
2688
41
80
2
2688
41
80
3
2688
41
80
4
2400
29
290
5
2758
34
0
6
2758
34
0
FT
2758
25
168
cocarpiis connnunity or the adjacent Arteiimia
community, and (2) the number of established
seedling, juvenile, and inmiatiue Cercocarpiis
rooted under and not under the canopy of a
live or dead shrub. The Wilcoxon signed rank
test was used to determine if there was a sig-
nificant difference in the distribution of indi-
viduals in the Cercocarpiis and Artemisia com-
munities, respectively, for each maturity class.
The significance level is P < 0.05 unless other-
wise noted.
Results
Cunent-year Cercocaqms seedlings were not
distributed evenly between Cercocarpiis stands
and adjacent Artemisia communities (Table 4).
Significantly more current-year seedlings were
rooted in the Cercocarpiis community.
At least 81% of established seedling, juve-
nile, and immature Cercocarpiis were rooted
in the adjacent Artemisia community (Table 4).
For established seedling and juvenile maturity
classes the difference in spatial distribution was
significant; the significance level for immature
Cercocarpus was P < 0.06.
More established seedling, juvenile, and
immature Cercocarpiis were rooted under the
protective canopy of a live or dead shrub than
in the open (Table 5). Only 1 transect had more
plants without a protective canopy, but the sig-
nificance level was P < 0. 10.
Discussion
Spatial distribution of current-year Cerco-
carpiis seedlings and established young Cerco-
carpiis had an inverse relationship (Tables 1,
4). Current-year seedlings were most abundant
in Cercocarpiis stands dominated by large, ma-
ture Cercocarpiis and least abundant in ddyd-
cent Artemisia communities. Young, established
264
Great Basin Naturalist
[Volume 56
Table 4. Number of cunent-year seedling, establislied seedling, juxenile, and immature mahogany rooted in Cercocar-
piis (CER) stands dominated by mature individuals, and in adjacent Artem/sifl (ART) communities. Within each maturity
class, total values between community types with different letters are significantly different (P < 0.05).
Current-
■vear
Established
seedling
seedli
ng
Juveni
le
I nun
ature
Transect
CER
ART
CER
ART
CER
ART
CER
ART
1
20
0
1
11
1
5
0
1
2
72
15
1
15
3
3
0
5
3
75
53
0
16
0
6
5
6
4
31
39
0
2
0
7
0
4
5
.337
25
0
11
0
19
0
0
6
.506
28
1
11
0
4
0
0
7
33
0
1
9
0
2
0
5
Total
1074a
160b
4a
75b
4a
46b
5a
21ai
Percent
87
13
5
95
8
92
19
81
ISiRiiificantly different at P < 0.06.
Cercocarpus were virtually absent from mature
Cercocarpus stands but had a greater abun-
dance in adjacent Artemisia communities (Tables
1, 4). Young Cercocarpus were also abundant in
stands with low Cercocarpus crown cover or
relatively few large Cercocarpus (Table 1). The
low densitv' of current-year seedlings in adja-
cent Artemisia communities (Table 4) has 2
possible interpretations: (1) viable Cercocarpus
seeds were not dispersed into the Artemisia
community, or (2) germination of Cercocarpus
seed was impaired. Because data about seed
densities are lacking, a definitive conclusion
cannot be made. Cercocarpus seed, however, is
primarily wind dispersed (US DA 1948); there-
fore, it is unlikely that few seeds were present
in the Artemisia community, particularly since
all data were collected within 20 m of the Cer-
cocarpus stands. Most likely, over 85% fewer
Cercocarpus seedlings were in the Artemisia
community (Table 4) because seed germination
was substantially lower than in the Cercocar-
pus stands.
The inverse relationship for distribution of
current-year seedlings and established young
Cercocarpus indicates that locations with a
high abundance of current-year seedlings are
not necessarily locations with the best seedling
sui"vival. Populations peipetuate when seedlings
survive and advance into successively older
niatiuity classes, eventually producing new
seedlings. The pattern for spatial distribution
of current-year seedling, established seedling,
juvenile, and immature Cercocarpus deri\ed
from this sttidy and that conducted by Schult/
et al. (1990, 1991) indicates that 4 factors may
influence sui-vival of current-year seedlings as
well as plants in the youngest maturity classes:
(1) presence or thickness of plant litter, (2) root
growth characteristics, (3) presence of nurse
plants, and (4) herbivoiy.
Moderate levels of litter can favor seed ger-
mination and seedling establishment by de-
creasing soil temperature and increasing soil
moisture (Evans and Young 1970). Thick litter,
however, can reduce seedling establishment and
survival by preventing or restricting contact
between soil and seed or soil and root (Fowler
1986).
High litter cover (Table 1) and a thick la\er
of litter (personal obser\'ation) were common
in Cercocarpus stands in the Shoshone Range.
Litter cover and litter thickness were not mea-
sured in adjacent Artemisia communities; how-
ever, litter cover in high-ele\ation (>2200 m)
Artemisia communities ranges from 15% to
50% (Tueller and Eckert 1987). Extensive and
deep litter in Cercocarpus stands may promote
seed germination but decrease seedling sur-
vival because roots from Cercocarpus seedlings
seldom make contact with the mineral soil.
Less litter in the Artemisia community' may re-
duce Cercocarpus seed geniiination but enhance
stuAi\'al of seeds that germinate. Root growth
characteristics ma\ pla> an important role.
Rapid root giowth that current-year Cerco-
carpus seedlings experience (Dedy 1975) should
enhance sin-\ i\ orship of Cercocarpus seedlings
during seasonal drought, a common phenome-
non in the Great Basin. Root systems that
undergo rapid elongation should be able to fol-
low a retreating zone of soil moistiue (down-
ward) better than root systems that elongate
slowly. We excavated several Cercocarpus seed-
lings rooted in thick plant litter and found that
root growth was extensive (20+ cm) but not
1996]
Cercocahi'us Ki:c;i:Ni:iivrioN
265
Table 5. The number of established seecHin^, juvenile,
and ininiatiue Ccrrocarpiis rooted under and not under
another sliiiib or sluiib skeleton. Siuniliianee le\t'l is /' <
0.10.
Transeet
Hooted ui
uler
Not
rooted under
1
16
3
2
2.3
4
3
20
13
4
S
5
5
6
24
6
9
/
7
15
2
Total
y7a
58b
Peixi-ntaue
03
37
d()\\n\\'arcl toward or into the mineral soil.
Root growth was largely lateral. Following ger-
mination in early spring, available moistnre in
both mineral soil and plant litter is probably
high, sinee cool temperatures and abundant
precipitation are common (Houghton et al.
1975). Because moisture is not limiting early in
the growing season, root growth probably fol-
lows the path of least resistance. When thick
litter resides on top of mineral soil, the path of
least resistance would be laterally through the
litter, not downward through the mineral soil.
The loamy soil that Cercocorpiis stands inhabit
undoubtedly stores and retains more water
than plant litter does, and thus should desic-
cate more slowly. If thick plant litter prevents
or retards roots of current-year Cercocarpiis
seedlings from reaching or penetrating moist
mineral soil, seedling mortality should be high
when litter desiccates rapidly later in the sum-
mer. We obsen'ed high mortality for current-
year Cercocarpiis seedlings in August in Cerco-
carpiis stands with thick accumulations of lit-
ter. Less litter on Peavine Mountain (Table 1)
and in the Artemisia community (see Tueller
and Eckert 1987) may enable root systems of
Cercocarpiis seedlings at these locations to
grow downward into mineral soil immediately
following germination. This should increase
survivorship of current-year seedlings, which
may account (at least partially) for the greater
abundance of established seedling, juvenile,
and immature Cercocarpiis on sites with less
surface litter.
Herbivory may also play a role in seedling
survival. Current-year Cercocarpiis seedlings
have an average leaf surface area of only 4 cm^
(Dealy 1975), which herbivores can easily con-
sume. Herbivory can adversely affect estab-
lishment of woody species (Marquis 1974,
McAuliffc 1986), including Cercocar^ms (Scheldt
and Tisdale 1970). The presence of protective
ninsc plants, therefore, may be important for
regeneration ol Cercocarpiis seedlings.
Cercocarpiis stands in the Shoshone Range
had a mean shrub canopy cover of 11% (Schultz
et al. 1990). Total shrub canopy cover was not
measured in adjacent Artemisia communities;
however, it generally ranges from 41% to 50%
(Tueller and Eckert 1987). Thus, shrub cover
in adjacent Artemisia communities is 3.5 to 4
times greater than that in Cercocarpiis stands.
Since more established seedling, juvenile, and
immature Cercocarpiis were rooted imder a
shrub or shrub skeleton than not (Table 5), the
difference in shrub canopy cover between Cer-
cocarpiis stands and adjacent Artemisia com-
mimities may influence survival of cuiTcnt-year
seedlings, established seedlings, juvenile, and
immature Cercocarpiis. Artemisia and other
short-statured shrubs may serve as nurse plants
and protect small Cercocarpiis (including cur-
rent-year seedlings) from herbivores until their
photosynthetic surface is large enough to cope
with frequent browsing. Since shrub cover is
low in Cercocarpiis stands, more young Cerco-
carpiis are probably exposed to herbivores than
in Artemisia communities. This may help explain
the near absence of young Cercocarpiis in Cer-
cocarpus stands and their greater abundance in
adjacent Artemisia communities.
Conclusions
Abundance of current-year Cercocarpiis
seedlings is greatest in Cercocarpiis stands that
have high Cercocarpiis canopy cover, large mean
Cercocarpiis crown volume, and an extensive
layer of plant litter. These stand attributes also
result in a low density of plants in established
seedling, juvenile, and immature maturity
classes. Established young Cercocarpiis are
most abundant where gaps occur in the Cerco-
carpiis canopy, or in adjacent Artemisia com-
munities. Survival of current-year seedlings
appears best at locations that permit roots of
seedlings to make contact with mineral soil.
Survival of current-year seedlings and progres-
sion of individuals from established seedling
maturity class into successively older maturity
classes appear to be enhanced by the presence
of a shrub canopy that protects small Cercocar-
piis from herbivores.
266
Great Basin Naturalist
[Volume 56
Literature Cited
Austin, D. A., and E J. Urness. 1980. Response of curlleaf
mountain mahogany to piiining treatments in north-
em Utali. Journal of Range Management 33: 27.5-277.
Davis, J. N. 1976. Eeological investigations in Cercocarpus
ledifoliiis Nutt. communities of Utah. Unpubhshed
master's thesis, Brigham Young University', Prove, UT.
Davis, J. N., and J. D. Brotmerson. 1991. Ecological rela-
tionships of curlleaf mountain mahogany {Cercocar-
pus ledifolitis Nutt.) communities in Utah and impli-
cations for management. Great Basin Naturalist 51:
153-166.
Dealy, J. E. 1975. Ecology of curlleaf mountain mahogany
{Cercocarpus ledifolius Nutt.) in eastern Oregon and
adjacent areas. Unpublished doctoral dissertation,
Oregon State University, Coi^vallis.
Duncan, E. 1975. The ecology of curlleaf mountain mahog-
any in southwestern Montana with special reference
to mule deer Unpublished master's thesis, Montana
State University, Bozeman.
Evans, R. A., and J. A. Young. 1970. Plant litter and the
establishment of alien annual weeds in rangeland
communities. Weed Science 18: 697-703.
Fowler, N. L. 1986. Microsite requirements for germina-
tion and establishment of three grass species. Ameri-
can Midland Naturalist 115: 131-145.
Holmoren, R. C. 1954. A comparison of browse species
for the revegetation of big-game winter ranges of
southwestern Idaho. USDA, Intermountain Forest
and Range E.xperiment Station, Researcli Paper 33.
HosKiNS, L. W., and R D. Dalke. 1955. Winter browse on
the Pocatello big game range in southeastern Idaho.
Journal of Wildlife Management 19: 215-225.
Houghton, J. G., C. M. Sakamoto, and R. O. Gifford.
1975. Nevada's weatlier and climate. Nevada Bureau
of Mines and Geology, Special Publication 2.
LlACOS, L. G., and E. C. Nord. 1961. Cercocarpus seed dor-
mancy yields to acid and thiorea. Journal of Range
Management 14: 317-320.
Marquis, D. A. 1974. The impact of deer browsing on
Allegheny hardwood vegetation. USDA, Forest Ser-
vice, Research Paper NE-308.
McAuLlFFE, J. R. 1986. Herbivore limited establishment
of a Sonoran Desert tree, Cercidium microphijUum.
Ecology 67: 276-280.
Ormiston, J. H. 1978. Response of curlleaf mountain
mahogany to top piiming in southwest Montana. Pro-
ceedings of the First International Range Congress,
Denver, CO.
Plummer, A. R, R. L. Gensen, and H. D. Stapley. 1957.
Job completion report for game forage revegetation
project W-82-R-2. Utah State Department of Fish
and Game.
Plummer, A. R, D. R. Christensen, and S. B. Monsen.
1968. Restoring big game range in Utah. Utah Divi-
sion of Fish and Game, Publication 69-3.
Scheldt, R. S. 1969. Ecologx' and utilization of curl-leaf
mountain mahogany in Idaho. Unpublished master's
thesis. University' of Idaho, Moscow.
Scheldt, R. S., .4ND E. W. Tisdale. 1970. Ecology' and uti-
lization of curlleaf mountain mahogany in Idaho. Uni-
\'ersity of Idaho, Forest, Wildlife, and Range E.xperi-
ment Station, Note 15.
SCHULTZ, B. W 1987. Ecology of curlleaf mountain
mahogany {Cercocarpus ledifolius) in western and
central Nevada: population stnicture and dynamics.
Unpublished master's thesis. University of Nevada,
Reno. Ill pp.
Schultz, B. W, R. J. Tausch, and R T Tueller. 1991.
Size, age, and density relationships in curlleaf moun-
tain mahogany {Cercocarpus ledifolius) populations in
western and central Nevada: competitive implica-
tions. Great Basin Naturalist 51: 183-191.
Schultz, B. W, P T Tueller, and R J. Tausch. 1990.
Ecology of mountain mahogany {Cercocarpus ledi-
folius) in western and central Nevada: community
and population structure. Journal of Range Manage-
ment 43: 13-20.
Smith, A. D. 1950. Feeding deer on browse species during
winter. Journal of Range Management 3: 130-132.
Smith, A. D., and R. L. Hubrard. 1954. Preference rat-
ings for winter deer forages from northern Utah
ranges based on browsing time and forage consumed.
Journal of Range Management 7: 262-265.
Tueller, R T, and R. E. Eckert 1987. Big sagebrush
{Artemisia tridentata vaseyana) and longleaf snow-
beny {Symphoricarpos oreophilus) plant associations
in northeastern Nevada. Great Basin Naturalist 47:
117-131.
USDA. 1948. Cercocarpus H. B. K. mountain mahogany.
Pages 132-133 in Woody plant seed manual. USDA
Miscellaneous Publication 654.
Young, J. A., R. A. Evans, and D. L. Neal. 1978. Treat-
ment of curlleaf cercocaipus seeds to enhance germi-
nation. Joimial of \\'ildlife Management 42: 614—620.
Received 17 May 1995
Accepted 25 March 1996
Great Basin Naturalist 56(3), © 1996, pp. 267-271
POTENTIAL FOR CONTROLLING THE SPREAD OF
CENTAUREA MACULOSA WITH GRASS COMPEL ITION
John L. Liiul' Mountains. It is oiten found in plant communities dominated by Psc'uchmn'pu-ria spicatum or Fcstuca idahoeims,
hut it rareK invades roadsides dominated !)>• Bnnniis biennis Leyss. Ahove^round hiomass of the 3 j^rass species j^rown in
nii.xture with Cciitaiirea was compared to growdi in monoculture at a range of nitrogen input levels. The results suggest
that Brointi.s is capable of suppressing the gnw'th of Centaurca with the degree of sui:)pression increasing with increasing
nitrogen lexels. The 2 nati\e grasses had no impact on Centaurca under tlie controlled en\ ironment conditions of this
study.
Keij words: annpctition, weed control. Centaurca maculosa, Bromus inermis, Agropyron spicatimi, Festuca idahoen-
sis, exotic plants.
Centaurca maculosa Lam. (spotted knap-
weed) is a major weed associated with spring
wet-simimer diy areas of the northern Rocky
Mountains (Forcella and Harvey 198L Tyser
and Key 1988, Weaver et al 1989). Centaurca
dominates waste places, invades disturbed
rangeland, and sometimes invades undisturbed
range (Tyser and Key 1988). In contrast, it rarely
invades roadsides dominated by Bromus inermis
Leyss. (Weaver et ak 1989). This suggests that
it may be exchided fi-om waste pkices that are
planted to Bromus before Centaurca invades.
Alternatively, because planting exotics violates
the charge of national park managers, one may
ask whether Centaurca might also be excluded
from disturbed areas by planting native grasses
that naturally dominate either relatively dry
{Pseudorocgneria spicatum [Pursh] Scribner and
Smith = Agropyron spicatum) or more moist
{Festuca idahoensis Elmer) foothill habitats.
Weed suppression may be accomplished by
(1) preempting resources with more competi-
tive plant species or (2) using biocontrols or
herbicides that selectively increase weed mor-
talit\', decrease vigor, or prevent reproduction
(Lindquist et al. 1995). This study considers
management of Centaurca maculosa by compe-
tition rather than by common herbicide and
biocontrol methods. This approach deserves
attention because it may be less expensive and
more effective than herbicides in the long
term.
Our objective was to measure the competi-
tive ability of 3 grass species against Centaurca
in 2-way interaction experiments in sand cul-
ture. Mixture and monoculture treatments
were tested for 12 wk at 5 positions on a nitro-
gen gradient to determine whether competi-
tive relations were influenced by differences in
nitrogen availability. A plant's ability to com-
pete is related to its growth rate or ability to
gain biomass relative to associated species
(Haiper 1977). We compared aboveground bio-
mass of each species grown in mixture with
Centaurca to its growth in monoculture.
Materials and Methods
The rhizomatous exotic pasture grass Bromus
inermis Leyss. and 2 native bunchgrasses nor-
mally dominating relatively diy foothills {Pseu-
dorocgneria spicatum) or moister grasslands
immediately above and below the conifer zone
{Festuca idahoensis) were grown in 2-species
mixtures (replacement series) with C. maculosa.
Experiments consisted of 3 competition treat-
ments (monocultures of both grass and Centau-
rca, and 50:50 mixture) combined with 5 nitro-
gen addition treatments. Each treatment com-
bination had 10 replicates. Within each experi-
ment, pots were arranged in a completely ran-
domized design on a greenhouse bench and
rotated weekly to minimize position effects.
Each experiment was subject to different light
'Department of Biology and Department ot Plant, Soil, and Environmental Sciences, Montana State University, Bozeman, MT.5971"
-Present addres.s: Department of Agronomy, Universit\' of Nebraska. Lincoln, NE 68.58.3-091.'5.
267
268
Great Basin Naturalist
[Volume 56
conditions because of its position in the green-
house. A square planting pattern was used
with 4 plants spaced 5 cm apart. In each pot in
the mixture treatments, plants of the same
species were located on the diagonal.
Seeds were planted at a depth of 1.0 cm in
1000-cm'^ pots filled with coarse washed sand.
Pots were watered daily for 1 wk to allow
seedling establishment. Excess seedlings were
thinned and remaining seedlings allowed to
grow for an additional week prior to the addi-
tion of nutrients. The basic nutrient solution
was balanced with respect to all essential nutri-
ents but could be varied to allow the establish-
ment of nitrogen levels from 0%, 1%, 10%,
30%, and 100% of a standard level (Machlis and
Torrey 1956). Sufficient nutrient solution (200
ml) was applied to satinate the pot twice weekly
and water (200 ml) was added once each week.
Regular watering with nutrient solution and
alternate washing with tap water held the soil
solution near the applied level and pre\'ented
any concentration of the soil solution due to
evapoti-anspiration. Experiments were conducted
during March, April, and May 1988, when
greenhouse temperatures ranged from 14° to
32° C (25° C mean).
Twelve weeks after emergence, plants in
each pot were clipped at the soil surface, sepa-
rated by species, dried at 45° C for 5 d, and
weighed.
Nonlinear regression procedin-es (SAS 1988,
Gauss-Newton least squares estimation method)
were used to fit a rectangular hyperbola equa-
tion [1] (Cousens 1985) to mixture and mono-
culture data for each species:
/, • N
;i +
/. • N
[1]
calculated to determine whether competitive
relationships varied across relative nitrogen
levels. RCI is calculated as
RGI = (B,
I^mix)/B,
[2]
A,
where B,^^,^,-,^ and B,^^j^ are the aboveground diy
biomass (g plant"^) for a species grown in mono-
culture and mixture, respectively. A negative
RCI value indicates that the species performs
better in mixture than in monoculture. RCI may
be the best measure for determining species
displacement under competitive conditions
across a resource gradient (Grace 1995). Analy-
sis of variance was used to test for differences
in RCI within a species across nitrogen treat-
ments. Student's t was used to compare RCI
between species at each nitrogen addition level.
Results
A hvperbolic relationship between individ-
ual plant biomass and relative nitrogen level
was found in all mixtures and monocultures
(Figs, la-f ). Estimates of /,• (biomass at inter-
cept) differed between mixtures and monocul-
tures only for Centaurea grown in mixture with
Broiniis (Table 1). Estimates of A,- (maximum
biomass) differed for Bronuis and Centaurea
(Tiible 1).
Relative competition intensity was signifi-
cantly negative for Broimis at all nitrogen addi-
tion levels; it varied from negati\'e values at
low nitrogen to positixe \'alues at higher nitro-
gen levels for Centaurea in competition with
Bronius (Fig. 2). However, RCI did not differ
from zero in the experiments where P. spiea-
tum and E idahoensis were in competition w ith
Centaurea (data not shown).
Discussion
where B — aboveground dry biomass (g
plant~l), Aj = maximum aboveground biomass
of species / (g plant"^), N = relative nitrogen
addition level, and /, = biomass of species / as
relative nitrogen addition level approaches zero.
To determine the relative success of Centau-
rea in competition with each grass species,
estimates ol A,- and Ij were compared between
mixtures and monocultures using the extra
sum of sc^uares procedure (Ratkowsky 1983,
Lindcjuist et al. 1996). In addition, relative^
competition intensity (RCI; Grace 1995) was
Growth response of Bromus to nitrogen was
greater in mixture with Centaurea than in
monoculture, as indicated by the regression
lines (Fig. la) and the negative RCI values
across all nitrogen addition levels (Fig. 2). In
contrast, growth response of Centaurea was
lower in mixtme with Bromus than in mono-
culture (Fig. 1). The increase in Centaurea RCI
at high relative nitrogen le\el indicates that
Bronuis is a better competitor in the high
nitiogen treatments (Fig. 2). Results suggest
that Bnnnus is capable of suppressing the
1996]
CoMFEiiTivE Suppression of Centaurea
269
CQ
0.0
0.2
0.4
0.6
0.8
1.0
i3
CL
OS
E
o
03
CX3
O
"3
o
nj
S
d
0.0
0.2
0.4
0.6 0.8 1.0
1.5
Q.
o>
1.2
CO
CO
TO
F
0.9
o
CQ
0.6
E
3
^
0.3
o
a
to
0.0
CT3
1.5
Q.
-S
1.2
CO
CO
CO
F
0.9
O
CQ
nfi
cn
on
o
"3
0.3
o
(t!
0.0
0.0 0.2 0.4 0.6 0.8 1.0
O
0.0 0.2 0.4 0.6 0.8 1.0
10
C
O
-c
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0.0
0.2 0.4 0.6 0.8
1.0
ca
3.0
Q.
O)
2.5
c/)
CO
2.0
F
o
m
l.b
to
1.0
u
n
0,5
o
0.0
o
0.0 0.2 0.4 0.6 0.8 1.0
Relative Nitrogen Addition Level
Fig. 1. Plot of obsei-ved (o) and predicted (•) abovegroiind dn' biomass plant"' on relative nitrogen addition level
when grown in monoculture ( ) and mi.xture ( ): a, Bnvniis grown in monoculture and in mixture with Centaurea;
b, Centaurea grown in monoculture and in mixture with Bronius; c, Pscudoroegneria grown in monoculture and in mix-
ture with Centaurea; d, Centaurea grown in monoculture and in mixture with Pseudoroegneria; e, Centaurea grown in
monoculture and in mixture with Fe.stuea; f, Festuca grown in monoculture and in mixture with Centaurea.
270
Great Basin Naturalist
[Volume 56
Table 1. Estimates of parameter values followed by asymptotic standard errors for maximum aboveground biomass (g
plant~l) (A), biomass as relative nitrogen level approaches zero (/), and the coefficient of determination (r-) obtained
from fitting equation [1] to monoculture and mixture data of each species. Variation in / and A between competition
treatments was tested using the extra sum of squares principle, with P value indicating the significance level for the
comparison of parameter (Coeff ) values from the monoculture and mixture regressions.
Species
Coeff
Competition treatment
Monoculture
Mixture
P value
Broiiius
A
I
1-2
Centaiirea with
A
Bromiis
/
1-2
Pscudoroef^neria
A
/
r2
Centaiirea with
A
Pseiidoroegneria
I
r2
Festuca
A
I
r2
Centaiirea with
A
Festuca
I
1.884(0.11.5)
8.769(1.141)
0.90
1.306(0.082)
10.621(1.861)
0.85
1.289(0.237)
1.758(0.323)
0.80
0.636(0.048)
6.770(1.639)
0.77
0.595(0.029)
16.747(3.892)
0.83
1.262(0.114)
6.418(1.295)
0.77
3.819(0..361)
12.893(2.197)
0.84
0.438(0.0,56)
45.717(35.47)
0.14
0.827(0.087)
2.490(0.442)
0.82
0.805(0.107)
4.617(1.454)
0.60
0.491(0.056)
23.921(14.925)
0.39
1.522(0.172)
10.061(2.870)
0.66
< 0.000
0.283
< 0.000
0.023
0.081
0.410
0.307
0.665
0.303
0.857
0.514
0.535
growth of Centaiirea, the degree of suppres-
sion increasing with increasing nitrogen levels.
Growth response of Pseudoroegneria and
Festuca to nitrogen when growing in mixture
witli Centaiirea did not differ fi-om their response
in monoculture. Likewise, growth response of
Centaiirea did not differ between monoculture
and mixtures with Pseudoroegneria or Festuca.
Therefore, these results suggest that these
native grasses are not likely to increase or sup-
press growth of Centaiirea, regardless of nitro-
gen addition level. This result is contrary to
the obsened invasion of Centaiirea into com-
munities dominated by these grasses. One ex-
planation may be that disturbance (especially
grazing) in the field creates gaps in the grass
communit\' where Centaiirea can establish itself
even though it is not a superior competitor for
resources.
Competitive interactions were greater be-
tween each grass species and Centaiirea at the
high end of the nitrogen gradient. This may be
a function of rapid growth. Thus, in nitrogen-
rich environments fast-growing plants may
rapidly occupy space and usurp resources to
the exclusion of slow-growing species (Grime
1979, Radosevich and Holt 1984). Similar com-
petitive effects may be expected to occur on
other soil resource gradients, assuming adapta-
tions for ac(}uisition of nitrogen and other mobile
nutrients, as well as water, are similar (Grime
1979, Fitter and Hay 1987). In addition, one
may hypothesize, based on the resource ratio
theory (Tilman 1982), that Bronius is a superior
competitor for nutrients other than nitrogen
relative to Centaiirea. By increasing nitrogen,
both species should be limited by essential
nutrients other than nitrogen, and the species
with the lowest R* (the superior competitior)
for the other nutrients should displace the
species with the higher R* for the same nutri-
ents (Tilman 1990).
The abilit) of Bromiis to out-compete Cen-
taiirea in nutrient culture provides one expla-
nation for the obsei'ved population dynamics of
Centaiirea in the field. Roadsides seeded with
Bromiis are rareh in\'aded In Centaiirea (Weaver
et al. 1989). Both field and laboraton' observa-
tions suggest that disturbed sites seeded simul-
taneousK' with Centaiirea and the exotic, Bro-
miis, will be dominated b\ Bromiis. The effec-
tiveness of Bromiis in suppressing Centaiirea
may be increased with fertilization. Further-
more, it may be expected that established Bro-
miis plants will suppress the growth oi Centaii-
rea seedlings. The results of this study suggest
that at the seedling stage Bromus may be used
to competiti\cl\' exclude Centaiirea. This method
of weed management merits trial in the field.
On the other hand, the regional dominants.
1996]
CoMiuniTivK Sli'I'hkssion of Cemavrea
271
-1.5
0 0.01 0.1 0.3 1.0
Relative Nitrogen Addition Level
Fig. 2. Relative competition intensity (RCI) of Broinii.s and Ccntawea across 5 relative nitrogen addition levels. Let-
ters above bars indicate whether RCI varies (Duncan's multiple range test, F < 0.05) within species across nitrogen
le\el. An asterisk indicates that RCI differs (P < 0.05) among species at that nitrogen level.
Pseiidoroegneria and Festuca, probably would
not sufficienth' suppress Centaiirea to decrease
the potential for invasion.
Advantages of the competitive method over
herbicides and biocontrol treatments used to
manage Centaurea are its long duration and
low environmental impact. Given these advan-
tages, exclusion of Centaurea with Bromiis
merits trial in environments where the danger
of invasion exists.
Literature Cited
COLSEXS, R. 1985. A simple model relating yield loss to
weed density. Anuids of Applied Biology- 107: 239-252.
Fitter, A. H., .^nd R. K. M. Hay. 1987. Environmental
physiology of plants. 2nd edition. Academic Press
Inc., San Diego, CA. 423 pp.
FORCELLA, E, AND S. Harvey. 1981. Migration and distri-
bution of 100 alien weeds in NW USA, 1881-1980.
Herbarium, Montana State University, Bozeman. 117
pp.
Gr\ce, J. B. 1995. On the measurement of plant competi-
tion intensit}'. Ecology' 76; 305-308.
Grime, J. E 1979. Plant strategies and vegetation pro-
cesses. John Wiley and Sons, London. 222 pp.
Harper, J. L. 1977. Population biology of plants. Academic
Press Inc., San Diego, CA. 892 pp.
LlNDQUIST, J. L., B. D. M.\.\\\ ELL, D. D. BlULER, AND J. L.
GuN,soLUS. 1995. Modeling the population dynamics
and economics of velvetleaf (Abutilon flicophrasti)
control in a com {Zea //ia|/.s)-soybean (Glycine max)
rotation. Weed Science 43: 269-275.
LlNDQUIST, J. L., D. A. MoRTENSEN, S. A. Clay, R. Schmenk,
J. J. Kells, K. Howatt, and R Westr\. 1996. Stability
of corn-velvetleaf interference relationships. Weed
Science 44; 309-313.
Machlis, L., and J. ToRREY 1956. Plants in action. Free-
man, San Francisco, CA. 282 pp.
Radosevich, R. S., and J. S. Holt 1984. Weed ecology:
implication for vegetation management. John Wiley
& Sons Inc., New York. 265 pp.
Ratkowsky, D. a. 1983. Nonlinear regression modeling: a
unified practical approach. Marcel Dekker, Inc., New
York. 276 pp.
SAS. 1988. SAS/STAT user's guide, release 6.03. SAS
Institute, Caiy, NC. 1028 pp.
Tilman, D. 1982. Resource competition and comnumity
stiiicture. Princeton University Press, Princeton, NJ.
. 1990. Mechanisms of plant competition for nutri-
ents: the elements of a predictive theory of competi-
tion. Pages 117-141 in J. B. Grace and D. Tilman,
editors. Perspectives on plant competition. Academic
Press, Inc., New York.
Tyser, R., and C. Key 1988. Spotted knapweed in natural
area fescue grasslands; an ecological assessment.
Northwest Science 62; 151-160.
Weaver, T, D. Gustaf.son, J. Lichthardt, and B. Woods.
1989. Distribution of e.xotic plants in the Northern
Rocky Mountains by environmental type and distur-
bance condition. MSU Biology Report 41, Bozeman,
MT 91 pp.
Received 26 September 1995
Accepted 29 April 1996
Great Basin Naturalist 56(3), © 1996, pp. 272-275
INDICATORS OF RED SQUIRREL {TAMIASCIURUS HUDSONICUS)
ABUNDANCE IN THE WHITEBARK PINE ZONE
David J. Mattsonl and Daniel E Reinhart^
Abstract. — We investigated occupied squinel middens and s(|uinel sightings and vocalizations as indicators of red
squiiTel (Tamiasciunis hudsonicus) abundance in the high-elevation whitebark pine {Pimis alhicauUs) zone. Data were col-
lected 1984-1989 fioni line transects located on 2 study sites in the Yellowstone ecosystem. We evaluated the perfomiance
of each measure on the basis of precision and biological considerations. We concluded that, of the 3 measures, active mid-
dens were the best indicator of red sc|uirrel abundance. We also obsei^ved that the density of active middens dropped by
48%-66% between 1987 and 1989, following a severe drought and extensive wildfires that burned one of the study sites
during 1988.
Key words: transect. Fourier series, midden, voccdization, sigJitinf;, wildfire.
Whitebark pine {Piniis alhicauUs) seeds are
an important bear food that affects the sui^vival
and fecundity of grizzly bears {Ursiis arctos) in
the Yellowstone ecosystem. Use of pine seeds
by grizzlies is almost entirely contingent upon
the availabilit)' of cones cached in middens (i.e.,
larder hoards) by red squirrels {Tamiasciunis
hudsonicus). Management of whitebark pine
habitats for grizzlies has thus become contin-
gent upon management of red squirrel popula-
tions (Mattson and Reinhart 1994).
We studied red squirrels in the whitebark
pine zone using data collected from line tran-
sects. Because these data included counts of
middens, animals, and vocalizations, we were
able to evaluate the relative efficacy of these 3
indicators of sc|uinel presence. We were inter-
ested in identifying a "well-behaved" and rele-
vant indicator of density to facilitate our inves-
tigation of relationships between squirrel
abundance and environmental factors such as
midden use by grizzly bears. We were also
interested in providing managers with an
approach they could use to indicate squirrel
abundance, short of using intensive methods
that relied upon marked animals.
Study Area
Om- study area consisted of 2 sites, one
located on the Mt. Washburn massif in north
central Yellowstone National Park (44°47'N),
and the other near Cooke City, Montana,
immediately northeast of the park (45°00'N).
These sites spanned the whitebark pine zone,
from 2360 to 2870 m elevation. The whitebark
pine zone borders upper timberline and is
accordingly cold (average annual temperatures
<0°C), often wind\', and subject to deep (1-2 m)
winter snow accumulations (Weaver 1990).
Materl^ls and Methods
Broad study objectixes affected our transect
design. We mapped the study area by habitat
type-cover type strata based upon ground-
truthed interpretation of 1:20,000 aerial pho-
tography. The result was a fine-scale mosaic,
with individual map poKgons (forest stands)
sometimes as small as indi\ idual S(]uirrel terri-
tories. To minimize effects of edge between
different habitat types, we placed transects so
as to maximize the number of right-angle inter-
sections with stand l)oimdaries as well as the
amount of intersection with stand interiors.
Because of this consideration and because for-
est and meadow were \ariousl\ intermixed,
transect lines were of imequal length.
We surveyed transects in the same order
each year, beginning after 10 August and end-
ing prior to 28 September Two obseivers walked
pennanenth' marked transect lines, with one ob-
sener primariK responsible for obsen ations and
the other primarily responsible for recording
iNalKMKil Bi(,li,i;ii;il Scr-vicv, Dcparlinciit nl Fish ami Wikllilc Ucmhuvcs, liiiivc-rsit) cit Idalio, Moscow. ID 83843.
2U.S. NalioTial Park Srnacc, Bcsoiinc MaiiaKrinnit, Ycllousloiic \atioiial Park, WV 82190.
272
1996J
Red Squikhkl Indicatoks
273
data and keeping on line. At least one of llie
observers (the junior author) was the same
dining all years at both sites. Voealizations and
obsened scjuirrels or middens (both aeti\'e and
inactive, b\^ criteria of Finlc)' [1969]) were
recorded along with their estimated perpen-
dicular distances from the transect line. Indi-
\ idual cone caches were not considered to be a
"midden and were easily distinguished fiom
these larger, more peniianent features.
We used die computer program TRANSECT
(Burnham et al. 1980) to estimate densities.
Individual transects constituted sample units
for densit) calculations. As recommended by
Burnham et al. (1980), we used the Fourier
series, with 1-4 terms, to estimate distance-to-
line probability detection functions (g[x]). The
distance at which we specified the limits of
detectabilit}' lor oiu" measures (i.e., the cut-point)
exerted considerable influence on the fit of the
Fourier function to the observed detection dis-
tribution. Accordingly, we varied cut-points to
achieve the best fit to each year- or site-spe-
cific data set. Because data were collected
from only 35 transects on the Mt. Washburn
massif during 1984 and from 15 transects in
the Cooke City area during 1984 and 1989 (com-
pared to 57 and 21 transects, respectively, for
all other years), we also calculated densities
solely from these original 35 and 15 transects
for all years so as to allow comparison with
results from 1984.
Results
We sampled the study area 5 yr, 1984-1987
and 1989. During 1988 wildfires Immed 562,000
ha of the Yellowstone area, including 52% of
the Mt. Washburn transects (47% severely).
Transects on the Mt. Washburn area totaled
18.9 km during 1984 (mean transect length [A']
= 539 ± 245 m[s]) and 29.8 km during the
remaining 4 yr (X = 523 ± 258 m). Similarly,
during 1984 and 1989 transects on the Cooke
City area totaled 16.4 km (X = 1091 ± 427 m)
and 21.1 km during the remaining years (X =
1005 ± 405 m). We recorded 124 squirrel sight-
ings, 641 vocalizations, and 300 active middens
on die Mt. Washburn study site and 54 sight-
ings, 528 vocalizations, and 201 active middens
on the Cooke Cit>' study site during the 5 study
years. The small number of sightings from the
Cooke Cit>' site prevented us from estimating
annual densities from this measure for this
area.
Total distance-to-line frecjuencN' distribu-
tions for each of the 3 measures did not differ
between the Mt. Washburn and Cooke City
study sites (Mantel-Haenszel x^ for ordinal
categories, P = 0.51, 0.61, and 0.35 for active
middens, vocalizations, and sightings, respec-
tively). Perpendicular distributions of sightings
and active middens peaked in the nearest (<10
m) distance categoiy, although the distribution
of sightings more closely resembled a negative
exponential and the distribution of middens a
negative sigmoidal function. The majorit>' (65%
and 78%, respectively, by year and study site)
of both these distributions were adequately fit
ix^ test, P > 0.10) by a single-term Fourier
fimction. Distributions of vocalizations peaked
in the 2nd (11-20 m) distance categor)' and
were characteristically (94%) fit by a 2- or
higher-term Fourier function. In 3 (18%) in-
stances we could not achieve an adequate fit
by any model.
Relationships among annual density esti-
mates from the 3 measures were varied (Fig.
1). On the Mt. Washburn site, mean sighting
and vocalization densities were weakly corre-
lated (r = 0.722), but tended to have overlap-
ping 95% confidence inten'als. Onl\' 2 of 9 con-
fidence intei-vals for the obsewed estimates (all
years, for both the 1984 and inclusive samples)
did not contain the line describing perfect cor-
respondence (Fig. Id). In all but a single in-
stance (Cooke City, 1984), mean midden densi-
ties were greater than mean densities of the
other 2 measures and were more strongly cor-
related with sightings than vocalizations (r =
0.981 versus r = 0.831, respectively, for tran-
sects 1-57, Mt. Washburn; Fig. Ic). However,
in this case, only two of nine 95% confidence
intei-vals for midden and sighting densities in-
cluded the possibility of perfect correspondence.
Conclusions
From these results we concluded diat densi-
ties calculated from active middens were more
useful than densities calculated from the odier
2 measures for indicating red squirrel abun-
dance. Our conclusion followed from the greater
apparent detectabilit}' of middens compared to
the squinels themselves, the consistency with
which a single-term Fourier function described
274
(a) MT. WASHBURN
Great Basin Naturalist
(b) COOKE CITY
[Volume 56
1.20
.00 0.20 0.40 0.60 0.80 1.00 1.20
DENSITY OF VOCALIZATIONS (n/lia)
Q 0.80
> 0.60
0.00 0.20 0.40 0.60 0.80 1.00 1.20
DENSITY OF VOCALIZATIONS (n/lia)
(c) MT. WASHBURN
(d) MT. WASHBURN
0.80
> 0.60
0.00
0.00 0.20 0.40 0.60 0.80 1.00 1.20
DENSITY OF SIGHTINGS (n/ha)
0.80
N 0.60 -
> 0.40 -
0.00 0.20 0.40 0.60 0.80
DENSITY OF SIGHTINGS (ii/lia)
Fig. 1. Relation.ship.s heKveen annual e,stiniate.s of den.sity for acti\t^ middens compared to \ ocalization.s. (a) for Mt.
Wa.shbiirn and (h) for Cooke Cih; (c) sightings compared to active middens for Mt. Washburn, and (d) sightings compared
to vocahzations, Mt. Washbimi, 1984— I9(S7 and 19S9. Enor bars correspond to 95% confidence inten'als, soHd circles to
results from all transects, and open circles to results from the fewer transects established and first sun e>ed in 1984. Diag-
onals represent perfect correspondence between estimates.
the probability detection distiibiition for mid-
dens, and the resulting consistently smaller
standard errors for the density estimates. In
addition, scatter plots showed that active mid-
den densities tended to be >() when sighting
and vocalization deiTsities were not. B\ impli-
cation, vocalization and especially sighting
densities were more likely to underestimate
true squirrel densities; i.e., at the same time
that active middens clearly indicated the pres-
ence of squirrels, sightings and \ocalizations
could suggest there were none.
Because red squiiTel middens ai^e nonmobile,
often numerous, relatixeK' easily obsei^ved, and
t\picall> associated w ith onl\ one s(iuirrel (Kil-
ham 1954, M. Smith 1968, Wolff and Zasada
1975, Vahle and Patton 1983), the\- are logical
indicators ol s(iuirrel abundance. Furthermore,
the\' do not suffer from sampling problems
associated with w eadier, season, and time of da\'
1996]
Rkd SnuiUREL Indicators
275
T.XBLK 1. E,stiniated mean (n lur') and standard error (.s^) Tor dcMisities of active middens oti the Mt. Washburn and
tlookc City study sites, 1984-1987 and 1989, percent coefficient of'variations for animal variation 1984-1987, and percent
decline in densit>' i'rom 1987 to 1989. Results are given for the transects estahlished and sui-veyed during 1984 (1-.35 and
1-15) and for the larijer sample of transects sin\e\'ed during ail other vears, except for 1989, in the Cooke Cit\' area (1-57
and 1-21).
Mt. Wash!
)nrn
(^ooki
■ City
Tians. [
1-35
Trans. 1-57
Trans. 1
1-15
Trans. 1-21
Year
.Mean
[s^)
Mean (,v^)
Mean
(.v.v)
Mean (.s-)
1984
0.447
(0.083)
—
0.428
(0.077)
1985
0.557
(0.110)
0.632 (0.084)
0.682
(0.116)
0.635 (0.095)
1986
0.219
(0.042)
0.426 (0.078)
0.548
(0.126)
0.540 (0.098)
1987
0.453
(0.093)
0.838 (0.143)
0.544
(0.103)
0.790 (0.170)
1989
0.234
(0.062)
0.285 (0.098)
0.262
(0.137)
—
C\
1984-
-1987
35.4
32.6
18.9
19.2
% decline
1987-
-1989
48.3
66.0
51.8
—
to the same extent as do sightings and vt)cahza-
tions (cf. C. Smith 1968, Pauls 1978, Ferron et
al. 1986). These expectations were corroborated
b\' our analysis. Middens also have a direct tie
to management of resources, such as bears,
that are of common concern in this zone.
Densities of active middens in our study
area averaged between 0.2 and 0.8 ha~^ and
on both study sites were lowest during 1989,
following the drought and wildfires of 1988
(Table 1). Although annual variation tended to
be greater on the Mt. Washburn site compared
to the Cooke Cit>' site, this difference was not
statistically significant (d.f = 4/4, F = 1.31, P
> 0.5). Both sites exhibited similar annual pat-
terns of variation, including relatively low den-
sities during 1984 and 1986 and a substantial
decline in active midden densities between
1987 and 1989.
Acknowledgments
This study was funded by the U.S. National
Park Service through the Interagency Grizzly
Bear Study Team. M. Hubbard, D. Campopi-
ano, D. Dunbar III, and G. Green provided
invaluable field assistance. We also thank L. S.
Mills, J. Peek, R. G. Wright, D. Johnson, C.
Smith, C. Benkman, and an anonymous re-
viewer for their thoughtful reviews of this
manuscript and its earlier versions.
Literature Gited
BuRNHAM, K. E, D. R. Anderson, and J. L. Laake. 1980.
Estimation of density fiom line transect sampling of
biologiciil populations. Wildlife Monographs 72; 1-202.
Ferron, J., J. R Ouellet, and Y. Lemay. 1986. Spring and
summer time budgets and feeding behaviom- of tlie
red squinel {Tainia.sciiirws liud.sonicits). Canadian Jom-
nal of Zoology 64: 385-.391.
FiNLEY, R. B., Jr. 1969. Cone caches and middens (A'Taini-
ascitirus in the Rocky Mountain region. Uni\ersit> of
Kansas Museum of Natural Ilistoiy Miscellaneous
Publications 51: 233-273.
KiLHAM, L. 1954. Territorial beha\iour of red scjuirrel.
Journal of Mammalogx' 35: 252-253.
Mattson, D. J., and D. E Reinhart. 1994. Bear use of
whitebark pine seeds in Noitli America. Pages 212-220
in W. C. Schmidt and F-K. Holtmeier, compilers,
Proceedings — International Workshop on Subalpine
Stone Pines and Their Environments: the Status of
Oiu' Knowledge. U.S. Forest Sendee, General Tech-
nical Report INT-GTR-309.
Pauls, R. W. 1978. Behavioural strategies rele\ant to the
energy economy of the red squinel {Tamiasciurus
huchonicus). Canadian Journal of Zoolog\- 56:
1519-1525.
Smith, C. C. 1968. The adaptive nature of social organiza-
tion in the genus of tree squinels Tainiasciiinus. Eco-
logical Monographs 38: 31-63.
Smith, M. C. 1968. Red squinel responses to spnice cone
failure in interior Alaska. Journal of Wildlife Manage-
ment 32: 305-317.
Vahle, J. R., AND D. R. Patton. 1983. Red scjuinel co\er
requirements in Arizona mixed conifer forests. Jour-
nal of Forestiy 81: 14-15, 22.
Wea\'ER, T. 1990. Climates of subalpine pine woodlands.
Pages 72-79 in W. C. Schmidt and K. J. McDonald,
compilers, Proceedings — S\mposium on Whitebark
Pine Ecosystems: Ecology and Management of a
High-mountain Resource. U.S. Forest Sei"vice, Gen-
eral Technical Report INT-270.
Wolff, J. O., and J. C. Zasada. 1975. Red squirrel re-
sponse to clearcut and shelteiAvood systems in interior
Alaska. U.S. Forest Service, Research Note PNW-
255.
Received 9 October 1995
Accepted 8 May 1996
Great Basin Naturalist 56(3), © 1996, pp. 276-278
THERMAL CHARACTERISTICS OF MOUNTAIN LION DENS
Vernon C. Bleichl-, Becky M. Piercel-2, Jeffiey L. Davisl, and Vicki L. Davisl
Abstract. — We used radiotelemetn,^ and searched with a trained hound to locate the dens of 3 recentK- parturient
mountain hons [Felis concolor). These dens were located in dense riparian vegetation along the same stream in die bot-
tom ot a steep canyon. We monitored the circadian temperatin-es of 2 dens at 1-h intenals and compared tliem to ambient
temperatures recorded simultaneously. We found mountain lion dens to effectively moderate high ambient temperatures,
but these dens failed to pro\ide a themial advantage at the lowest ambient temperatures recorded in this in\estigation.
We conclude that mountain lion dens pro\ide effecti\ e protection fi-om thermal maxima for \oung, immobile kittens.
Key words: Felis concolor. mountain lion, temperature. California, den. behavior
Female mountain lions (Fclis concolor) select
protected locations in which to bear young
(Shaw 1989:7, Beier et al. 1995), but little infor-
mation is available on den site characteristics
for this elusive felid. Here, we describe some
characteristics of 3 dens used by different
females and their litters and quantify' the ther-
mal characteristics of 2 of those dens.
Description of Study Area
Our study area is located in Mono Co., Cali-
fornia, approximatelv 35 km NW of Bishop
(118°25'W, 37°20'N), Inyo Co., California.
This area is on the western edge of the Great
Basin, immediately east of the crest of the
Sierra Nevada. The dominant x'egetation t\pe
in the general area is sagebrush {A)ie)nisia tri-
dentata) scrub with pinyon pine {Finns niono-
phijUa) forest at higher elevations. Dense vege-
tation, dominated b>' willows {Salix spp.) and
wild rose {Rosa spp.), occurs along the major
water courses.
Methods
During August and September 1994 and
1995, telemetry indicated that several adult
females in our investigation of mountain lion
ecolog)' had restricted their daiK movements.
These females returned repeatedh' to the same
locations, suggesting that they had established
natal dens (Beier et al. 1995). We searched
these 3 areas and, after detecting vocalizations
of neonatal mountain lions, we used a trained
hound (Bruce 1918) to locate the dens and kit-
tens. We estimated the ages of these kittens
according to criteria summarized by Anderson
(1983:43) and Currier (1983).
We examined the thermal characteristics of
the dens b>' placing a recording thennograph
(model RTM, R\'an Instruments, Inc., Kirk-
land, WA) on the floor of each den and an
identical instrument on the ground < 100 m
away, on a north exposure supporting sage-
brush and pinyon pine. Because of the shinibs
and trees present on these north-facing slopes,
thermographs were not exposed directK* to the
sun for most of each dax; Hourlx' temperatures
were recorded at den 2 fiom 4 September to 4
October 1994, and at den 3 from 11 August to
16 September 1995; we did not haxe access to
thermographs during the period that den 1 was
actixe. We made ocular estimates of tree height
and canop)' closure, as well as horizontal cover,
at each den.
We used analysis of xariance and anahsis of
covariance to explore the effects of da\' and
time on temperature, simple linear regression
to examine the relationship between da>" of the
study and daiK' temperature, and t tests to
compare den temperatures with ambient tem-
peratures (Zar 1984).
Results
Three dens containing kittens were located
along the Owens Ri\ er: den 1 contained 1 male
and 1 female; den 2 contained 3 males and 1
female; den 3 contained 2 males and 1 female.
IC.ililoTnunrpartiiic-iilorKisharKlCainr, tOTW. l.iiK-St,, Hi>.li(.p, CA 9:5511.
-histitvitc <)( Antic liii)l,ii;\ and D.parlini-iil cif Hii)l(ii;> and Wildlilf. liiiver.sity of Alaska, Fairbanks. AK 99775.
276
1996]
Mountain Lion Di:ns
277
We estimated tlie kittens at dens 1 and 3 to he
< 20 da\s of age, and those at den 2 to he < 10
dax's old.
All 3 dens were located in dense groves of
willows that ranged in height to approxiniateK
4 ni. Wild rose was ahnndant at all 3 sites, and
each den was located < 50 ni from the river.
Canop\ closme at each den was nearly 100%,
and direct simlight did not reach the suhstrate
during am of our midda\' visits {n = 2, den 1; ;i
= 5, den 2; n = 3, den 3). Horizontal coxer at
each location was sufficientK dense that, e\'en
while standing, we were totally obscured from
each other s xiew at < 3 m. The substrate of all
3 dens was littered with deciduous leaves as
well as tree trunks, branches, twigs, and bark.
We were able to reach the kittens only b\'
craw ling into the dense vegetation present at
each site.
We found significant differences between
the dens in mean ambient temperature (-t j^,^ 9
= 13.18 ± 9.94 [sY C; .vj^,, 3 = 20.47 ± 10.61°
C; F = 202.584, df = 1, 1630, P < 0.001),
mean den temperature (-V^jg„ 0 = 6.01 ± 5.77°
C; .va^„ 3 = 15.22 ± 7.08° C; F = 807.949, df =
1, 1630, P < 0.001), and mean dail>' tempera-
ture differential (ambient temperature - den
temperature; .v j^,^ .7 = 7.16 ± 6.05° C; .v^j^,-, 3 =
5.25 ± 6.09° C; F = 40.224, df = 1, 1630, P <
0.001). At den 2, there was a significant effect
of day on ambient temperature (F = 3.814, df
= 30, 713, P < 0.001), den temperature (F =
3.191, df = 30, 713, P < 0.001), and tempera-
ture differential (F = 4.320, df = 30, 713, P <
0.001). At den 3, however, there was no such
effect on ambient temperature (F = 0.421, df
= 36, 851, P = 0.999), den temperature (F =
0.535, df = 36, 851, P =0.989), or temperature
differential (F = 0.488, df = 36, 851, P =
0.995). As the study progressed, there was a
significant decline at den 2 in ambient temper-
ature (r = -0.340, P < 0.001), den temperature
(r = -0.112, P < 0.001), and temperature dif-
ferential (r = -0.228, P < 0.001); lesser de-
clines in ambient temperature (r = -0.100, P
< 0.001), den temperature (r = -0.051, P =
0.001), and temperature differential (r = -0.048,
P < 0.001) occun-ed at den 3.
At both dens, there was significant diel vari-
ation in ambient temperature (den 2: F =
103.382, df = 23, 720, P < 0.001; den 3: F =
618.443, df = 23, 864) and den temperature
(den 2: F = 91.008, df = 23, 720, P < 0.001;
den 3: F = 431.275, df = 23. 864, P < 0.001).
When date was used as a covariate to control
lor dail\ solar radiation, the mean temperatine
differential also \aried on an hourly basis at
both dens (den 2: F = 112.271, df = 23, 719, P
< 0.001; den 3: F = 329.936, df = 23, 863, P
< 0.001). HourK' ambient temperatures were
greater than corresponding den temperatures
at both locations (den 2: t = 32.285, df = 743,
P < 0.001; den 3: t = 25.662, df = 887, P <
0.001); this difference was especially pro-
nounced at high ambient temperatures (>31°
C [HAT]; Fig. 1). At HAT, the temperature dif-
ferential at den 2 (x = 21.92 ± 4.49° C) was
>3 times that at moderate ambient tempera-
tures (< 31° C [MAT]; X = 6.03 ± 4.46° C), and
the temperature differential at den 3 {x =
13.56 ± 8.37° C) at HAT was nearly 5 times
that at MAT (x = 2.87 ± 3.50° C). At den 2, the
mean range of daih' ambient temperatures (x
= 28.96 ± 7.81° C) was nearly double that of
daily den temperatures (x = 15.79 ± 5.26° C)
(/ = 15.83, df = 30, P < 0.001). Similarly at
den 3 the mean range of daily ambient temper-
atures (.Y = 32.54 ± 3.71° C) was >1.5 times
that of daiK' den temperatures (x = 20.89 ±
3.50° C; t = 15.24, df = 36, P < 0.001). For
both locations combined, den temperatures
were less than ambient temperatures for all but
2 (<0.2%) of the paired hourly obsenations.
Discussion
These mountain lion dens effectiveh' mod-
erated high ambient temperatures, consistent
with the hypothesis of Shaw (1989) that dens
play an important role in protecting young,
defenseless kittens from thermal maxima. At
HAT, mean temperature differentials were 3-5
times greater than at MAT (Fig. 1). There were
significant effects of time of day (both dens)
and da)' length (den 2) on temperature differ-
ential and, hence, the moderating influence of
the dens. Nevertheless, den temperatures
were less variable than were ambient tempera-
tures. We found no evidence that these dens
provided a thermal adxantage (i.e., den tem-
peratures greater than ambient temperatures)
at the minimum ambient temperatures we
recorded; dens may, however, provide protec-
tion for kittens when temperatures fall below
those that we encountered.
Few descriptions of mountain lion dens are
available, but females max- select caves, rocky
areas, or dense thickets in which to bear young
278
Great Basin Naturalist
[Volume 56
10 20 30 40
AMBIENT TEMPERATURE (C)
50
Fig. 1. Mean temperature differential (ambient - den) is
more than 3 times greater at high ambient temperatures
(>31° C) than at moderate ambient temperatures (<31° C;
data fi^om dens 2 and 3 combined, F — I, 241.07, df = 1,
1630, P < 0.001). Mountain Hon dens in dense vegetation
effectively moderate extreme high temperatures and afford
young, helpless kittens protection from ambient ma.xima,
consistent with the hypothesis of Shaw (1989).
(Bruce 1918, Young and Goldman 1946, McBride
1976, Russell 1978, Shaw 1989). We hv^podie-
size that thermal characteristics var\' among
types of dens, and that mountain lions inhabit-
ing particular environments select den sites
based, in part, on the thermal advantage(s)
they provide.
In an area with a warm, Mediterranean cli-
mate, Beier et al. (1995) reported 2 dens that
were located in a small canyon with very heavy
cover of "brush," similar to those we investi-
gated. Dens located in thick, woody vegetation
may conceal young that are vulnerable to pre-
dation, but they also provide protection for kit-
tens from extreme temperatures associated
with direct insolation. Such locations provide
important thermal benefits for kittens at high
ambient temperatures, and a more stable ther-
mal environment than exists outside the den
throughout the range of ambient temperatures
we recorded. Movements bv kittens for ther-
moregulator)' puiposes miglit be lessened under
these circumstances. Fewer movements by kit-
tens may decrease the probabilit\' of discovery
by potential predators, thereb>' enhancing the
survival of young, defenseless mountain lions.
Acknowledgments
We thank S. Parmenter, D. Becker, and C.
Milliron for lending us the recording themio-
graphs, N. G. Andrew and E Beier for critical
comments on an early draft of the manuscript,
and M. W Oehler Sr, for assistance in the field.
Financial support was provided by the Mule
Deer Foundation, University of California White
Mountain Research Station, CalifoiTiia Depart-
ment of Fish and Game (CDFG), National Rifle
Association, Safari Club International, and the
Fish and Game Advisoiy Committees of Inyo
and Mono counties. This is a contribution fi-om
the CDFG Deer Herd Management Plan
Implementation Program.
Literature Cited
Anderson, A. E. 1983. A critical review of literature on
puma {Felis concolor). Colorado Division of Wildlife
Special Report 54: 1-91.
Beier, R, D. Choate, and R. H. Barretf. 1995. Move-
ment patterns of mountain lions during different
behaviors. Journal of Mammalog\' 76; 1056-1070.
Bruce, J. C. 1918. Lioness tracked to lair. California Fish
and Game 4: 1.52-153.
Currier, M. J. E 1983. Felis concolor Mammalian Species
200; 1-7.
McBride, R. T. 1976. The status and ecologx- of the moun-
tain lion Felis concolor stanleijana of the Te.\as-Mex-
ico border. Unpublished master s thesis, Sul Ross
State University, i\]pine, TX.
Russell, K. R. 1978. Mountain lion. Pages 207-225 in J. L.
Schmidt and D. L. Gilbert, editors. Big game of North
America: ecologx' and management. Stackpole Books,
Harrisburg, PA.
Sh.w, H. 1989. Soul among lions. Johnson Publisliing Com-
pany, Boulder, CO.
Young, S. R, and E. A. Goldman. 1946. The puma, myste-
rious American cat. American Wildlife Institute, Wash-
ington, DC.
Zar, J. H. 1984. Biostatistic;iI an;Jysis. Prentice-Hall, Engle-
wood Cliffs, NJ.
Received 30 December 1995
Accepted 5 April 1996
Great Basin Naturalist 56(3), © 1996, pp. 279-2S0
JAMES WILLIAM BEE
1913-1996
W'iliner \\. TaniUT^
Jiunt's W. Bee
James W. Bee, professor of zoolog)' and emer-
itus, University of Kansas, Lawrence, Kansas,
died at Seatde', Washington, 18 April 1996. He
was bora 25 September 1913 in Provo, Utah.
His family, including parents, Robert G. and
Mary Culbertson Bee, and brother and sister.
Max and Mary, were residents of Provo, Utah,
where they receixed their earK' education. It
was fiom this setting in Utah Valley that James
was introduced at an early age to the sciences
of archaeolog)' and ornithology by his father,
who loved natural histoiy and the little-known
histoiy of Utah Valley, its lake, and its early
inhabitants.
As a youth and young man, he accompanied
his father on man\' collecting trips that resulted
in assembling artifacts of the past. These
archaeological finds provided valuable infonua-
tion pertaining lo IncUan winter camps, sum-
mer camping areas, and burial groimds, and an
insight into tlie role of Utah Lake and the sur-
roimding mountains as providers of aliundant
fish and game.
Their travels near this lake and in the moun-
tains brought them in contact with numerous
birds. Each spring flocks of birds entered the
valley — some remained and others moved on.
This phenomenon stimulated a great interest,
so much so that James, i.is father and various
friends became amature ornithologists. Their
ornithological work encompassed life histoiy
studies, obsei^vation of arrivals in the spring,
and investigation of nests and nesting. Ulti-
mately, this interest in birds lead to the assem-
bly and preparation of eggs for those species
nesting in the valleys and mountains of central
Utah.
Thus was bora a naturalist whose contribu-
tions are invaluable and most of which could
not now be assembled. The archaeological col-
lections are presently at the Museum of Peo-
ples and Cultures, Brigham Young University.
James and his father contributed 812 sets of
bird eggs and 1 12 single eggs representing 234
species. James contributed 7918 mammal, 245
bird, and 504 amphibian and reptile specimens
to the M. L. Bean Life Science Museum, also at
Brigliam Young University'. In die Bean Museum
Libraiy are field records, 27 volumes from
James and 20 xolumes of his father s, all well
documented and done with great care. Tliese
were written in the field as the data were
obtained and represent field records of a time
when some pristine conditions still existed.
James entered Brigham Young Unixersity in
1932 and received his B.A. degree in 1937.
World War II interrupted his studies for the
M.A., but this he finished in 1947. As an under-
graduate, he became interested in and re-
searched mammals. Thus his master's research
'M. L. Bi'.m Museum, Brigham Young Universih, Provo, UT S4602.
279
280
Great Basin Naturalist
[Volume 56
was the mammals of Utah County. While in the
Armed Forces (1941-1946), he was trained as a
hospital administrator and sewed as a sergeant-
major, organizing 50 key men as a cadre to
establish a new hospital. He supei-vised several
new hospital departments and for a year and a
half sei-ved in field hospitals for airborne units
in India, Assam, Burma, and China. During
these years he met Annette E Malseed, R.N.
They were married 15 October 1945 in Kun-
ming, Yeaman, China.
In September 1948 James entered the Uni-
versity of Kansas to continue his research in
mammalogy, with a desire to complete his
study on the genus Microtis. He completed his
studies at KU and spent a summer at Friday
Harbor, Washington. He was a noted field
zoologist and spent many years collecting re-
search material and field data for the Museum
of Natural Histoiy at the University of Kansas.
Students doing research in vertebrate zoology
at Brigham Young University or at the Univer-
sity of Kansas will find numerous specimen
tags labeled "collected b>' James Bee." After 37
years he retired from KU and built a new
home on Lopez Island, Washington. James and
Annette were the parents of three children:
James Robert, Annette Christine Kenagy, and
Mary Pauline Bee Kaufman.
It was my pleasure to have spent several
summer field trips with James. A highlight was
the summer of 1939 when we studied the ver-
tebrates of western Utah County. During this
time we prepared and assembled museum
specimens; of importance to me was finding a
nesting colony of the western skink and secur-
ing additional specimens of Hypsigalena. We
both participated in the new discoveries, and it
was obvious that Jim was at his best in prepar-
ing precise field data. I learned much from him
that sinnmer and appreciated his dedication to
a complete understanding of the natinal world
we were investigating.
James had a very likeable personalit) that
was reflected in his family, which he held in
high esteem.
Publications
Bee, James W.
1957 Biological sui'vey of the Virgin Islands National
Park. Procedures, obsei^vations and recommen-
dations. 33 pages, 2 maps. Submitted to National
Park Senice.
1958 Birds foimd in the Aictic Slope of noitliem Alaska.
Universitv' of Kansas Publications, Museum of
Natiu-al History 10(5): 163-211. 2 plates, 1 figure
in text.
1970 Vasco M. Tanner — a diversified career. Great
Basin Naturalist 30: 216-217.
1994 Rough Grouse siting. The Trumpeter, San Juan
Audubon Societ>' 14(4):3.
Bee, James W., Dumitru Murariu, and Robert S.
hoffmana.
1980 Histolog\' and histochemistiy of specified integu-
mentaiy glands in eight species of North Ameri-
can shrews (Mammalia, Insectivora). Travau.v du
Museum d' Histoire Katurelle "Grigor Antipas "
(Bucharest, Romania) 22: 547-569.
Bee, James W, and E. R. Hall
1951 An instance of coyote-dog hybridization. Trans-
actions of the Kansas Academy of Science 54(1):
73-77. 4 figures. 1 table.
1954 Occurrence of the harbor porpoise at Point
Barrow, Alaska. Journal of Mammalog\- 35(1):
122-123.
1956 Mammals of northern Alaska on the Aictic Slope.
University of Kansas, Museum of Natural His-
tory, Miscellanceous Publications 8: 1-309. 4
plates, 122 figures, 51 tables.
1960 The red fig-eating bat, Steuoderma rufum Des-
marert, found alive in the West Indies. Mam-
malia, Museum d'Histoire Naturelle, Paris 14(1):
67-75. 2 figures in text.
Bee, James W, and Howard Le\ enson
Bald Eagle use of Kansas Ri\er riparian habitat
in northeastern Kansas. Kansas Ornithological
Society Bulletin 3(4): 28-33.
Great Basin Naturalist 56(3). © 1996, i^p. 281-282
BROOK STICKLEBACK (CULAEA INCONSTANS [KIRTLAND 1841]),
A NEW ADDITION TO THE UPPER COLORADO
RIVER BASIN FISH FAUNA
TiTiiotln Moddc' and (I. Bruce llaiucs'
Key words: hrook stichlclHich. raiiiic cxtciisioii. noiiiialirc.
Brook stickleback {Ciilaea incoii.stdiis) is a
small gasterosteid fish native to Arctic and
Atlantic drainages in North America. The
species natixe range extends west from Nova
Scotia to British Columbia and south horn the
Northwest Territories to southern Ohio drain-
ages, including the Mississippi-Missouri River
above the confluence of the Illinois River
(Scott and Grossman 1973). llubbs and Lagler
(1958) reported brook stickleback from the
Illinois River in Illinois and the Missouri River
in Kansas. Historical accounts exist of relictual
populations in the Platte River system, but
Cross (1967) noted its absence from Kansas.
An isolated, and presumably relict, population
occurs in the Canadian River drainage of New
Mexico (Koster 1957). Brook stickleback has
been collected outside its native range in
Alabama (Boschung 1992), Kentucky (Burr and
Warren 1986), Tennessee (Etnier and Starnes
1993), the Rio Crande River drainage in New
Mexico (Sublette et al. 1990), Colorado (Zuck-
erman and Behnke 1986), and the Klamath
River, California (Peter Moyle, University of
California, Davis, personal communication).
Between luly and October 1995 we col-
lected 5 brook stickleback from the middle
Green River, Uintah County, Utah, the 1st rec-
ord for the species in Utah (catalog number
LFL 24871, Lai-val Fish Laboratory, Colorado
State University). Brook stickleback was first
reported elsewhere in the upper Colorado
River drainage in 3 small tributaries of the Elk
River (South, Coleman, and Deep creeks) in
northwestern Colorado in 1993 (Jake Bennett,
Colorado Division of Wildlife, personal com-
munication).
One brook stickleback juvenile, 27 mm
total length (TL), was collected 18 July 1995 in
a (juatrefoil light trap at the outlet of Old
Charley Wash, river kilometer (RK) 402 on the
Green River (RK measured from the conflu-
ence of the Green and Coloiado rivers). Four
adult fish, (41, 46, 48, 54 mm TL) were col-
lected between 1 October and 12 October 1995
from Old Charley Wash, a wetland on the
Oura>' National Wildlife Refuge that connects
to the Green River during high spring flows.
Fish were collected when the wetland was
drained (Modde in press); all were found in
low or no velocity habitats.
Tyus et al. (1982) cited the establishment of
42 nonnative fishes in the upper Colorado River
compared to 13 native species. Brook stickle-
back is an additional transplanted species, prob-
ably the result of human introduction rather
than a natural range extension. Brook stickle-
back introductions elsewhere in the United
States were presumably through bait bucket
transfers or contaminated game fish stockings
(Zuckerman and Behnke 1986, Sublette et al.
1990).
Literature Cited
Boschung, H. T. 1992. Catalogue of freshwater and marine
fishes of Alabama. Alabama Museum of Natural His-
toi-v' Bulletin 14:1-266.
Burr, B. M., and M. L. Warren, Jr.. 1986. A distrilni-
tional atlas of Kentucky' fishes. Kentucky Nature Pre-
serves Commission Scientific and Technical .Series,
No. 4.
Cross, E B. 1967. Handbook of fishes of Kansas. State Bio-
logical Sun'ey and the University' of Kansas Museum
of Natural History, Lawi-ence.
Etnier, D. A., and W. C. Starnes. 1993. The fishes of Ten-
nessee. The University of Tennessee Press, Kno.wille.
HUBBS, C. L., and K. E L.'\gler. 1958. Eishes of the Creat
Lakes region. The Universit\- of Michigan Press,
Ann Arbor
'Colorado River Fish Project, U.S. Fish and Wildlife Service. 266 West 100 North, Suite #2, Vernal, UT 84078.
281
282
Great Basin Natur.\list
[Volume 56
KOSTER, W. J. 1957. Guide to the fishes of New Mexico.
University of New Mexico Press, Albuquerque.
MODDE, T. In press. Juvenile razorback sucker in a man-
aged wetland adjacent to the Green River. Great
Basin Naturalist.
Scott, W. B., .\nd E. J. Crossm.an. 1973. Fishes of Canada.
Fisheries Research Board of Canada Bulletin 184.
Sublette, J. E., M. D. Hatch, and M. Sublette. 1990.
The fishes of New Mexico. University- of New Mex-
ico Press, Albuquerque.
Tvus, H. M., B. D. BuRDiCK, R. A. Valdez, C. M. H.aynes,
T. A. Lytle, and C. R. Berry. 1982. Fishes of the
Upper Colorado River Basin: distiibution, abundance.
and status. Pages 12-70 in W. H. Miller, H. M. Tyus,
and C. A. Carlson, editors. Fishes of the Upper Col-
orado River System: present and future. American
Fisheries Society, Bethesda, MD.
ZUCKER\L\N, L. D., AND R. J. Behnke. 1986. Introduced
fish in the San Luis Valley, Colorado. Pages 435^52
171 R. H. Stroud, editor Role offish culture in fishery
management. American Fisheries Societv, Bethesda,
MD.
Received 4 January 1996
Accepted 12 April 1996
Great Basin Naturalist 56(3), © 1996, p. 282
ERRATA
Correction to:
Sutherland, Steven D., and Robert K. Vickeiy,
Jr. 1993. On the relative importance of
flower color, shape, and nectar rewards in
attracting pollinators to Mimuhis. Great
Basin Naturahst 53: 107-117.
The article states: "Hummingbirds are com-
monly said to have evolved a preference for
red or orange-red flowers, " citing (1) K. A.
Grant, 1966, A h\q3othesis concerning tlie preva-
lence of red coloration in California humming-
bird flowers, American Naturalist 100: 85-98;
and (2) K. A. Grant and V. Grant, 1968, Hum-
mingbirds and their flowers, Columbia Uni-
versity Press, New York, 115 pp., among other
references. In fact, the Grants point out just
the opposite, i.e., that experimental investiga-
tion shows that hummingbirds ha\ e not evolved
a preference for red or any other color. Actu-
ally, Sutherland and Vickeiy's article comes to
this conclusion also.
Great Basin Naturalist 56(3), O 1996, pp. 283-285
BOOK REVIEW
Snakes of Utah. Douglas C. Cox and Wilmcr
W. Tanner; Mark Fhilbrick, photography.
Monte L. Bean Life Science Museum, Brig-
ham Young Universit), Provo, UT. 1996.
$17.95 softcover.
Snakes of Utah, anticipated for some time,
is finalK' avaikible for distribution. This book-
let (92 total pages) includes all known species
and subspecies of snakes found in the state,
with brief descriptions, habits, and habitats,
along with colored photographs of each. While
most people will likeK shudder at the thought
of snakes, especially while viewing photographs,
the enthusiast will recognize the value of the
illustrations and other published information.
Generalh; the booklet is written in nonscien-
tific language, but it also includes some scien-
tific notations. For instance, scientific names
and autliorities of the 33 species and subspecies,
along with common names, are included for
each. Of interest (perhaps only to the special-
ist) is the fact that only 2 binomials are found
among all Utah snakes; 31 are trinomials. It
might be concluded that, because of subspeci-
ation, onh' 27 kinds of snakes are found in Utah.
To the general public, a night snake is a night
snake, a garter snake is a garter snake, and a
rattlesnake is a rattlesnake. Heipetologists have
named subspecies for practically all snakes,
compounding one's knowledge of these animals.
Technically, where closely related subspecies
show sympatric distribution, there should be
intergradation between the 2 t\'pes. Most indi-
viduals using this booklet will probably not
recognize differences between related sub-
species found especially in these sympatric
regions. If intergrades are not present, then
these should be elevated to species and not kept
as subspecies. Little infomiation is found in the
booklet on intergradation of characteristics.
An important contribution of this booklet is
the colored photographs. While not captioned,
most photographs are obxious because they are
shown on the page opposite the name and other
information on that snake. This publication
would be more useful if a caption were shown
by the other photographs throughout the text,
e.g., the photo opposite page 1 and those shown
on pages 3, 4, 5, 8. The herpetologist will
probably recognize these without caption, but,
as stated, it's likely these specialists will not be
the primaiy users of the te.xt. Identification of
snakes by these photographs may not be obvi-
ous to most readers. Most photos show colors
and patterns of snakes, but a few, such as the
full view of the Upper Basin garter snake on
page 59, do not show these identifiable fea-
tures. It's interesting that the only snake not
represented by a photo of the entire body is
the Sonoran lyre snake on page 67. One won-
ders why. Perhaps it's because this snake is
"considered to be rare. " However, the Dixie
College Natural Science Museum contains
records of 7 specimens, 2 having been found in
what is now considered "downtown" St. George,
1 specimen as recently as 1980. It seems likely
that widi a little eflFort, one of diese "rai'c " snakes
might have been found. The photo of the Utah
blind snake on page 17 is a surprise. Of the
several dozen blind snakes observed by this
writer, representing localities from the Red
Cliffs Recreation Area near Leeds, Washing-
ton County, to the extreme northwest corner
of Arizona, not 1 specimen even approached
this dark phase. They have all been a pale tan
color, frequently showing a suffusion of pink.
Another important contribution of this book-
let is the distribution maps included with each
species along with the general and sometimes
specific distribution of the snake within the
state. While it is difficult to show accurac\- on
a small map, some maps are erroneous. For
instance, the distribution of the Painted Desert
glossy snake is "in the extreme southeastern
sector of the state, adjacent to northeastern
Arizona" (page 40). The map, however, shows it
is found more south central than southeastern.
An inconsistency from text to map is also
obsen^ed with the California king snake (page
46). If this snake occurs "from the southwest
corner east to the Colorado River," wh\' does
283
284
Great Basin Naturalist
[Volume 56
the distributional map extend considerabh'
beyond the Colorado Ri\ er along the San Juan
River? Nothing in the text is speculative of a
range extension. The maps of the Utah moun-
tain king snake (page 48) and the Utah milk
snake (page 50) do not accurate!)' depict their
known distributions in Washington Count)'.
On page 60, of the western blackneck garter
snake, the text states "its northernmost habitat
is associated with streams ... in the regions of
southeastern Utah. The map shows its distri-
bution into east central Utah. Reference is
made to a ground snake having been collected
in Carbon Count)', far from its known range,
and this area is shown on the map. Might this
specimen have been one that escaped or was
released from captivity? (Reports haxe been
made of indixiduals transporting this snake
from the St. George area, where it is common,
to elsewhere in the state.) There is speculation
that the Utah blackhead snake "may occur fur-
ther north in Emery and Carbon Counties."
(The proposed expansion is not shown on the
map.) Wh)' might it not, then, be found in
Wa)'ne Coimt)' and perhaps even San Juan and
Grand counties? If the midget faded rattle-
snake is found at Flaming Gorge, why does the
map not show distribution in that area?
While it would add to the length of the text,
it ^vould ha\e been better had the authors given
complete distribution ranges for all species
and subspecies, rather than just a few. A snake
doesn t recognize a political boundan as being
its limits! However, it could l)e reasoned, if the
distribution extends to the Utah boundaiy the
occurrence of that snake would also be in the
neighboring state.
The full-page map of the state of Utah (page
11) is a good addition to the text. However,
with the number of snakes found only in Utah s
Mojave Desert, this feature might ha\'e been
identified along with the others. In the geo-
graphical and ecological descriptions of Utah
(pages 9-10), considerable discussion is gi\'en
about montane regions, some at high ekna-
tions, yet little is written about the low, hot
desert or the higher, cold desert, although the
authors admit to the richness of reptile fauna,
especially in the low, hot desert, the south-
western region of the state.
In addition to these other features. Snakes
of Utah includes both glossar)-, though not
inclusive of all technical words used in the
text, and index.
The writer wonders at the importance of
the full page of illustrations (page 13) showing
scalation witli so little reference to most of these
features in descriptions. Some of these features
are referenced; most are not.
While full pages of color separate groups of
snakes, does this mean that Joshua trees are
characteristic of the distribution of the Utah
blind snake? Although the illustration on page
18 may be t) pical of the habitat of the rubber
boa in Utah, and on page 72 of the habitat of
some of the rattlesnakes, does the illustration
on page 22 depict the typical distribution of
the colubrids? Perhaps these "division pages"
were added merely for color; nevertheless, they
are attractive.
The authors of the booklet include a number
of interesting anthropomorphisms, perhaps
intentionally. Some of these are noted: (1) In
the introduction, the statement is made {page
5) that "the snake employs rocks and biaish to
snag the skin and hold it while the snake crawls
out. One wonders if the snake does this inten-
tionall)'. (2) "Denning is a behavior pattern that
provides the snake with an opportunity to
come in contact with other snakes of the same
species" (page 6). (3) Of the rubber boa, "it will
often cling like a bracelet and seem to enjoy it
as much as the person" (page 20). (4) The
statement is made about the western )'ellow-
belly racer (page 28) that "it will attempt to bite
if it feels at all threatened." (5) Another exam-
ple is that rattlesnakes use the rattle "as a
warning de\'ice to intimidate other animals
that may harm the snake (page 75).
Miscellaneous errors or inconsistencies in
narratix e, grammatical or othenvise, are found.
The introduction, for instance, discusses tall
tales and folklore of the American West. This
booklet is, of course, about snakes of one region
of the American West, but tall tales and folk-
lore— even some of the same stories heard in
the American West — are repeated wherexer
snakes are found.
On pages 4 and 5 the statement is made
tliat "the mouth is the most unixersally used
weapon emplo)'ed by snakes in self-defense."
The emphasis is obvious because the accom-
pan)ing text is about self-defense, but snakes
use their mouths more often as a means of
obtaining food. Also, in the introduction, the
statement is made that "these studies and our
inu.sciiDi i)ro''s Monte L. Bean Life Science
Museum. The complete identification of the
museum should ha\'e been made when it was
lirst referenced on page 6. It could he pointed
out, too, that other schools and nmseums
might ha\'e the same purpose — to "lielp them
to understand about snakes.
While the following is not necessarily in
error, it reflects a writing style. On page 12 the
following statements are made: "These snakes
do not pose any threat to man but they do pro-
\ ide a mild venom to help immobilize their
prey. Their prey includes worms, insects, frogs,
lizards, and small mammals. " In writing, re-
peated words and phrases should be avoided
in consecutive sentences or within the same
sentence. It could better have been written, "to
help immobilize their prey, which includes
worms, insects. . . . "
In the introduction to the tropical wormlike
snakes, the statement is made that "they feed
on insects and worms, especially termites and
ants, found in the soil." The emphasis in this
statement suggests that termites and ants are
kinds of wonns. This should have been written,
"they feed on worms and insects, especially
termites and ants. " In reference to the Utah
blind snake the statement is made (also on
page 15) that Vasco M. Tanner "had seven
specimens to examine, and the name is based
on No. 662 in the BYU type collection." Name
is inappropriately used, although specimen
No. 662 might have been published as the
t\"pe specimen.
One of the most frequently made grammat-
ical errors in writing is the inconsistency of
singulars and plurals within a sentence. On
page 20, this type of error is made. The rubber
boa "is a delightful animal to have around
their wrist." Inasmuch as their is plural, the
plurality ofivhsts must also be used.
Reference is made twice (on pages 30 and
44) that the snakes occur on "the margins of
deciduous forests." Small groups of deciduous
trees may occur in riparian areas or where
trees are cultivated, but technically, deciduous
forests do not occur in the state of Utah.
The redundant statement is made about the
western leafnose snake that the rostral scale
"looks leaflike. '
An inconsistency is noted about the Utah
mountain king snake and the Utah milk snake.
Page 48 states: "if a specimen has a white nose,
it is most likely a mountain king snake. If, how-
ever, it has a black nose, it is probably a milk
snake. These characteristics are not completely
reliable" (emphasis added). Page 50 states that
"the milk snake differs in that it has a black
nose."
On pages 68 and 70 the habits ol the Mesa
Verde night snake and the desert night snake
are described as "nocturnal, secretive, and sel-
dom seen." Furthermore, it is stated that the
former "feeds primarily on the lizard Uta stans-
huriana imiformis and other small lizards,"
while the latter "feeds primarily on the side-
blotched lizard Uta stanshuriami stanshiiriami."
One wonders about this inasmuch as lizards
are primarily diurnal and snakes nocturnal. Of
course, snakes could feed at night while lizards
are inactive.
While reference is made in the booklet about
the influence of soil on the ground color of
some snakes, there is no mention of this occur-
ring in the Mojave Desert sidewinder (page
78). Of the hundreds of sidewinders obsei^ved
by the author in the past 50 years, the influ-
ence of soil color on the ground color of the
snake is most obvious.
Despite these criticisms. Snakes of Utah
should contribute importantly to our knowl-
edge of these reptiles within a limited political
region. As noted, the booklet is written for lay-
men, and its distribution is more appropriate
in national and state parks and monuments than
in the scientific community. It is a "must" for
backpackers, individuals, and families spend-
ing time in tlie out-of-doors where snakes might
be encountered. The authors, tlie photographer,
and the publisher are to be commended for
finally making this booklet available.
Andrew H. Barnum
Professor Emeritus
Dixie College
St. George, UT 84770
The Future of Arid Grasslands:
Identifying Issues, Finding Solutions
9-1 3 October 1 996, Tucson, Arizona
A solution-oriented conference for everyone interested in the
future of grasslands in the American Southwest and northern
Mexico. This four-day conference will focus on understanding
problems facing those grasslands and practical tools for grass-
land management, preservation, and restoration. Attendees will
be a mix of private and public land managers and owners, scien-
tists, representatives of nonprofit groups, and concerned citi-
zens. Two full days will be spent in the field studying examples
of grassland management in southern Arizona. The other two
days will include keynote speakers and panelists as well as
small-group discussion and information sessions. The final day
will focus on methods for preservation ranging from coordi-
nated monitoring systems, land use, and taxation tools to public
involvement techniques.
Most of the speakers and panelists will be invited, but abstracts are welcome
for a few open sessions dealing with grasslands management, interrelationships
between grasslands and humans or wildlife, and specific methods for preser-
vation, especially success stories.
Researchers are encouraged to submit abstracts for poster sessions, which will
be incorporated into the program featuring on-the-ground examples of problem
solving to protect or restore grasslands. Both successful and unsuccessful ex-
amples are sought to illustrate what has and has not worked — and why.
The conference is organized by the Audubon Research Ranch and is co-
sponsored by numerous government agencies, educational institutions, and non-
profit groups.
For more information:
Grasslands Conference, Tucson Audubon Society
300 E. University #128
Tucson, AZ 85705
or
University of Arizona Water Resources Center
(520) 792-9591
INFORMATION FOR AUTHORS
The Great Basin Naturalist welcomes previously
unpublished manuscripts pertaining to the hiologi-
cal natural history of western North America.
Preference will he given to concise manuscripts of
up to 12,000 words. Simple species lists are dis-
couraged.
SUBMIT MANUSCRIPTS to Richard W. Baumann,
Editor Great Basin Naturalist, 290 MLBM, PO Box
20200, Brigham Young University, Provo, UT
84602-0200. An accompanying cover letter must
include phone numher(s) of the author submitting
the manuscript, and FAX number and E-mail
address when applicable; the letter must also pro-
\ ide infomiation describing the extent to which data,
text, or illustrations have been used in other papers
or books that are published, in press, submitted, or
soon to be submitted elsewhere. Authors should
adhere to the following guidelines; manuscripts not
so prepared may be returned for revision.
MANUSCRIPT PREPARATION. In general, the Great
Basin Naturalist follows recommendations in
Scientific Style and Format: The CBE Manual for
Authors, Editors, and Publishers, 6th edition
(Council of Biology' Editors, Inc., 11 South LaSalle
Street, Suite 1400,' Chicago, IL 60603, USA; phone
312-201-0101; FAX 312-201-0214). We do, however,
differ in our treatment of entries in Literature Cited.
Authors may consult Vol. 51, No. 2 of this journal
for specific instructions on format; these instruc-
tions. Guidelines for Manuscripts Submitted
TO THE Great Basin Naturalist, are printed at the
back of the issue. Also, check the most recent issue
of the Great Basin Natwalist for changes.
TYPE AND DOUBLE SPACE all materials, including
literature cited, table headings, and figure legends.
Avoid hyphenated words at the right-hand margins.
Underline words to be printed in italics. Use stan-
dard bond (22x28 cm), leaving 2.5-cm margins on
all sides.
SUBMIT 3 COPIES of the manuscript (5 copies of
fish manuscripts) and the original on a 5.25- or 3.5-
inch disk utilizing WordPerfect 5.1 or above.
Number all pages and assemble each copy sepa-
rately: title page, abstract and key words, text,
acknowledgments, literature cited, appendices,
tables, figure legends, figures.
TITLE PAGE includes an informative title no longer
than 15 words, names and addresses of authors, a
running head of fewer than 40 letters and spaces,
footnotes to indicate change of address and author
to whom correspondence should be addressed if
other than the first author.
ABSTRACT states the purpose, methods, results,
and conclusions of the research. It is followed by
6-12 key words, listed in order of decreasing
importance, to be used for indexing.
TEXT has centered main headings printed in all
capital letters; second-level headings are centered
in upper- and lowercase letters; third-level head-
ings begin paragraphs.
VOUCHER SPECIMENS. Authors are encouraged to
designate, properly prepare, label, and deposit
high-quality voucher specimens and cultures docu-
menting their research in an established permanent
collection, and to cite the repositoiy in publication.
REFERENCES IN THE TEXT are cited by author and
date: e.g., Martin (1989) or (Martin 1989). Multiple
citations should be separated by commas and listed
in chronological order Use "et al. ' after name of
first author for citations having more than two
authors.
ACKNOWLEDGMENTS, under a centered main
heading, include special publication numbers when
appropriate.
LITERATURE CITED, also under a centered main
heading, lists references alphabetically in the fol-
lowing formats:
Mack, G. D., and L. D. Flake. 1980. Habitat rela-
tionships of waterfowl broods on South Dakota
stock ponds. Journal of Wildlife Management
44: 695-700.
Sousa, W R 1985. Disturbance and patch dynamics
on rocky intertidal shores. Pages 101-124 in
S. T. A. Pickett and P S. White, eds.. The ecolo-
gy of natural disturbance and patch dynamics.
Academic Press, New York.
Coulson, R. N., and J. A. Witter. 1984. Forest ento-
mology: ecology and management. John Wiley
and Sons, Inc., New York. 669 pp.
TABLES are double spaced on separate sheets and
designed to fit the width of either a single column
or a page. Use lowercase letters to indicate foot-
notes.
PHOTOCOPIES OF FIGURES are submitted initially
with the manuscript; editors may suggest changes.
Lettering on figures should be large enough to
withstand reduction to one- or two-column width.
Originals must be no larger than 22x28 cm.
NOTES. If the manuscript would be more appro-
priate as a short communication or note, follow the
above instructions but do not include an abstract.
A CHARGE of $50 per page is made for articles
published; the rate for individual subscribers will
be $35 per page. However, manuscripts with com-
plex tables and/or numerous half-tones will be
assessed an additional charge. Reprints may be pur-
chased at the time of publication (an order form is
sent with the proofs).
FINAL CHECK:
• Cover letter explaining any duplication of
information and providing phone number(s),
FAX number, and E-mail address
• 3 copies of the manuscript and WordPerfect
diskette
• Conformity with instructions
• Photocopies of illustrations
(ISSN 001 7-3614)
GREAT BASIN NATURALIST vc 56 no 3 ju.y 1995
CONTENTS
Articles
Biogeographic significance of low-elevation records for Neotoma cinerea from the
northern Bonneville Basin, Utah Donald K. Grayson,
Stephanie D. Livingston, Eric Rickart, and Monson W. Shaver III 191
Synopsis of the mosses of Wyoming EM. Eckel 197
Variation in bitterbrush {Purshia tridentata Pursh) crude protein in southwestern
Montana Carl L. Wambolt, W. Wyatt Fraas, and Michael R. Frisina 205
Dam-forming cacti and nitrogen enrichment in a piiion-juniper woodland in noilli-
western Arizona Molly Thomas Hysell and Charles C. Crier 211
Distribution and ecological characteristics of Lewisia longipetala (Piper) Clay, a
high-altitude endemic plant Anne S. Halford and Robert S. Nowak 225
Larger ectoparasites of the Idaho ground squirrel {Spermophilus brunneus)
Eric Yensen, Craig R. Baird, and Paul W. Sherman 237
Roost sites of the silver-haired bat [Lasionycteris noctivagans) in the Black Hills,
South Dakota . . . Todd A. Mattson, Steven W. Busldrk, and Nancy L. Stanton 247
Perceptions of Utah alfalfa growers about wildlife damage to their hay crops:
implications for managing wildlife on private land Terry A. Messmer
and Sue Schroeder 254
Spatial relationships among young Cercocarpus ledifolius (curlleaf mountain
mahogany) Brad W. Schultz, Robin J. Tausch, and Paul T Tueller 261
Potential for controlling die spread of Centaurea maculosa with grass competition
John L. Lindquist, Bi-uce D. Maxwell, and T. Weaver 267
Indicators of red squirrel {Tamiasciurus hudsonicus) abundance in the whitebark
pine zone David J. Mattson and Daniel E Reinhart 272
Thermal characteristics of mountain lion dens Vernon C. Bleich,
Becky M. Pierce, Jeffrey L. Davis, and Vicki L. Davis 276
James William Bee, 1913-1996 Wilmer W Tanner 279
Note
Brook stickleback {Culaea inconstans [Kirtland 1841]), a new addition to the Upper
Colorado River Basin fish fauna Timothy Modde and G. Bruce Haines 281
Errata 282
Book Review
Snakes of Utah Douglas C. Cox and Wilmer W Tamier .... Andrew H. Barnimi 283
H E l-IBRARY
^^N Qc ;997
GREAT BASII^
ARD
SITV
NATURALIST
VOLUME 56 N^ 4 — OCTOBER 1996
BRIGHAM YOUNG UNIVERSITY
GREAT BASIN NATURALIST
Editor
Richard W. Baumann
290 MLBM
PO Box 20200
Brigham Young University
Provo, UT 84602-0200
801-378-5053
FAX 801-378-3733
Assistant Editor
Nathan M. Smith
190 MLBM
PO Box 26879
Brigham Young University
Provo, UT 84602-6879
801-378-6688
E-mail: NMS@HBLL1.BYU.EDU
Associate Editors
Michael A. Bowers
Blandy Experimental Farm, Universit>' of
Virginia, Box 175, Boyce, VA 22620
J. R. Callahan
Museum of Southwestern Biology, University of
New Mexico, Albuquerque, NM
Mailing address: Box 3140, Hemet, CA 92546
Jeffrey J. Johansen
Department of Biology, John Carroll University
University Heights, OH 44118
Boris C. Kondratieff
Department of Entomology, Colorado State
University, Fort Collins, CO 80523
Paul C. Marsh
Center for Environmental Studies, Arizona
State University, Tempe, AZ 85287
Stanley D. Smith
Department of Biology
University of Nevada-Las Vegas
Las Vegas, NV 89154-4004
Paul T. Tueller
Department of Environmental Resource Sciences
University of Nevada-Reno, 1000 Valley Road
Reno, NV 89512
Robert C. Whitmore
Division of Forestry, Box 6125, West Virginia
University, Morgantown, \W 26506-6125
Editorial Board. Jerran T. Flinders, Chairman, Botany and Range Science; Duke S. Rogers, Zoology;
Wilford M. Hess, Botany and Range Science; Richard R. Tolman, Zoology All are at Brigham Young
University. Ex Officio Editorial Board members include Steven L. Taylor, College of Biology and Agriculture;
H. Duane Smith, Director, Monte L. Bean Life Science Museum; Richard W. Baumann, Editor, Great Btisin
Naturalist.
The Great Basin Naturalist, founded in 1939, is published quarterly by Brigham Young University
Unpublished manuscripts that fiirther our biological understanding of the Great Basin and surrounding areas
in western North America are accepted for publication.
Subscriptions. Annual subscriptions to the Great Basin Naturalist for 1996 are $25 for individual sub-
scribers ($30 outside the United States) and $50 for institutions. The price of single issues is $12. All back
issues are in print and available for sale. All matters pertaining to subscriptions, back issues, or other busi-
ness should be directed to the Editor, Great Basin Naturalist, 290 MLBM, PO Box 20200, Brigham Young
University Provo, UT 84602-0200.
Scholarly Exchanges. Libraries or other organizations interested in obtaining the Great Basin
Naturalist through a continuing exchange of scholarly publications should contact the E.xchange Librarian,
6385 HBLL, PO Box 26889, Brigham Young University, Provo, UT 84602-6889.
Editorial Production Staff
JoAnne Abel Technical Editor
Jan Spencer Assistant to the Editor
Copyright © 1996 by Brigham Young University
Official publication date: 21 Novemher 1996
ISSN 0017-3614
11-96 750 20300
The Great Basin Naturalist
Published at Provo, Utah, bv
Brk.ham Younc; University
ISSN 0017-3614
Volume 56 31 October 1996 No. 4
Great Basin Naturalist 56(4), © 1996, pp. 287-29.3
SPECIES-ENVIRONMENT RELATIONSHIPS AMONG FILTER-FEEDING
CADDISFLIES (TRICHOPTERA: HYDROPSYCHIDAE)
IN ROCKY MOUNTAIN STREAMS
Timothy B. Mihuc'l-, G.Wayne MinshalP, and Janet R. Mihuc^
Abstract. — Species-environment relationships were determined for filter-feeding macroinvertehrates from 55
Rocky Mountain stream sites to estahlish species distribution patterns. Species abundance and 20 environmental vari-
ables were measured at each site with species-environment relationships determined using canonical correspondence
analysis and stepwise multiple regression. Results suggest that the distribution of several taxa was strongly related to
upstream-downstream environmental gradients. Arctopstjche grandis abundance increased with stream size (width and
depth) and decreased with increasing turbulence (Reynolds number). Brachijcentrus abundance also increased with
stream size (depth). Hijdropsyche abundance increased with increasing baseflow. Parapsyche elsis abundance demon-
strated negative correlation with depth, Froude number and conductivity. Ta,xa followed previously reported patterns,
partitioning habitat according to stream size. Arctopsyche grandis, Brachyccntrus. and Hydwpsyche were found in
larger (3rd- to 6th-order) streams, while Parapsyche elsis was obsei^ved in small headwater (1st- and 2nd-order) streams.
Other filter-feeding taxa such as Sirnuliwn, Pisidiuin, and ostracods exhibited little or no apparent habitat partitioning
among stream sites.
Key words: species-environment relationships, filter feeders. Rocky Mountain streams.
Benthic macroinvertebrates adapted for re- Minshall 1990, Richardson and xMackay 1991).
moving particles fi-om suspension (filter feeders) Many studies have determined filterer associa-
are an important component of stream commu- tions with food resources and environmental
nities. Distribution patterns and habitat associ- factors such as water velocity or temperature
ations among filterers have been well docu- (e.g., Edington 1968, Wallace 1974, Haddock
mented, particularly for members of the Tri- 1977, Wiillace and Merritt 1980, Alstad 1982,
choptera family Hydropsychidae (e.g.. Decamps Hauer and Stanford 1982, Brims et al. 1987,
1968, Edington and Hildrew 1973, Gordon Osborne and Herricks 1987, Wetmore et. al.
and Wallace 1975, Wdlace and Merritt 1980, 1990, Voelz and Ward 1992). Few studies have
Ross and Wallace 1982, Tachet et al. 1992) and considered the entire filterer component foimd
for lake outlet communities (e.g., Robinson and in natural (unimpounded, unregulated) streams
'Stream Ecolog\ Center, Department of Biologieal Seienees. Idaho State Universit\-, Pocatello, ID 8.3209-8007.
^Present address: Louisiana Cooperative Fisheries and VVildhfe Researeh Unit, School of Forestn, Wikllilr and Fisheries, U)uisiana State University, Baton
Rouge, LA 70803.
-^Biolog\- Program, 104 Life Sciences Building, Louisiana State University, Baton Rouge, LA 7080.3.
287
288
Great Basin Naturalist
[Volume 56
and distribution patterns of filterer species
with respect to a wide range of environmental
variables (Edington and Hildrew 1973, Gordon
and Wallace 1975, Boon 1978, Ross and Wal-
lace 1982). Our objective was to assess the dis-
tribution patterns of filter feeders in unim-
pounded Rocky Mountain, USA, streams to
determine relationships with specific environ-
mental variables including flow parameters;
stream size, depth, and width; benthic organic
matter content; slope; water chemistry; peri-
phyton biomass; and temperature. While many
studies have considered current velocit>', tem-
perature, and food relationships, partitioning
of habitat by filter feeders in relation to other
environmental variables is poorly known.
Methods
Stream sites encompassed the Rocky Moun-
tain region from northern Wyoming to central
Idaho, including 22 streams in Yellowstone
National Park and 33 in central Idaho. Streams
ranged from 1st to 6th order in size (Table 1).
All sites were unimpounded and none were
located below lake outflows. Yellowstone sites
were sampled each August from 1988 to 1992.
All other sites were sampled between July and
September during the year(s) indicated in
Table 1. Sampling methods were routine meth-
ods used in stream ecology (e.g., Platts et al.
1983). Briefly, benthic organisms were sam-
pled using a surber net (250 micron mesh) in
riffle habitat at 5 transects located at 50-m inter-
vals along a stream reach (250 m total reach
length). Samples were taken to a depth of 10
cm. VIean densitv' for each filterer species with-
in each stream reach was used in statistical
analyses to determine relationships with physi-
cal variables. Physical environmental variables
measured at each stream reach included stream
order, slope, width, baseflow (1 transect), mean
depth (n = 100 random measurements), mean
water velocity {n = 100 random measm-ements),
mean embeddedness (n = 100 random mea-
surements), and mean substrate size (n = 100
random measurements). Reach-scale means for
all variables were used in statistical analyses.
Width/depth ratio and several hydraulic para-
meters (mean Froude number, mean Reynolds
number) were calculated from tliese measine-
ments. Annual stream temperature range was
estimated from annual maximum (estimated
as temperature at the time of sampling) and
minimum temperature (the freezing point of
water). Water chemistry variables included
hardness, alkalinity, pH, and specific conduc-
tance. Other biotic variables measured at each
stream reach were chlorophyll a (n = 5 per
site), ash-free diy mass (AFDM) of periphyton
(n = 5 per site), biomass/chlorophyll ratio of
periphyton (B/C), and benthic organic matter
content (BOM; n = 5 per site). This study did
not address food resources or food acquisition
among filter feeders; therefore sampling of
transported and benthic fine particulate mater-
ial was not included in sampling protocol.
Relationships between species and environ-
mental variables were detemiined using canon-
ical correspondence analysis (Ter Braak 1986)
and stepwise multiple regression. All compar-
isons were made on reach-scale data (reach
means for all variables). Comparisons reflect
spatial differences among sites sampled in 1
season (summer) to detemiine large-scale distri-
bution patterns of filter feeders in 1st- through
6th-order streams. Temporal patterns were not
considered here. Canonical correspondence
analysis (CCA) allows the investigator to inter-
pret multiple species responses along a gradi-
ent of multiple environmental variables. This
analysis provides a useful interpretation of
species -environment relationships through the
resulting ordination plot. Once species-envi-
ronment correlations were identified using
CCA, multiple regression analysis was used to
further discern relationships between species
abundance and environmental variables.
Results
In the canonical correspondence analysis
(Fig. 1) the first ordination axis (.v axis) ex-
plained 37.9% of the total species-environment
relationship and the second {y axis) an addi-
tional 30.7% (Table 2). Results indicate that
sexeral environmental variables were impor-
tant in explaining variation in species abim-
dance across sites (Fig. 1). Arctopsychc ^randis
and Hyclrop.syche abundance related directly
to increasing baseflow, width, and stream order
(Fig. 1). Parapsyche ekis abundance was inverse-
ly related to increasing baseflow, width, and
stream order. Brachycentrus abundance related
primariK' to depth, substrate size, Reynolds
number, and annual temperature range (Fig. 1).
Si)uiilitim, Pisidiiim, and Ostracoda abundance
1996] Filter-feeding Invehtebiutes in Rocky Mountain Streams 289
T.\HI,K 1. Suminan of tlic 55 stii streaiiis. Sites are ananj^ed by increasing; stream order and increasinjj; depth within
each order
Stream
Sample dates
Order
Avg
Avg
Basedow
Slope
depth (ni)
width (m)
(m/s)
(%)
Caclie. YNP
1988-1992
1
O.OK
0,704
0.003
12
E Blacktail Deer, YNP
1988-1992
1
0.13
0.665
0.048
4.7
Twin, YNP
1988-1992
1
0.13
0.643
0.06
10.7
W Blacktail Deer, YNP
1988-1992
1
0.17
0.550
0.043
3.8
Faiiy, YNP
1988-1992
1
0.23
0.307
0.066
1.0
Pioneer, ID
1990
2
0.05
0.342
0.13
6
Dunce, ID
1990,91
2
0.06
0.109
0.07
17
Goat, ID
1990,91
2
0.06
0.089
0.05
18
Cache, YNP
1988-1992
2
0.09
0.764
0.012
10.1
Packhorse, ID
1991
2
0.09
0.413
0.04
4
Castle, ID
1992
2
0.09
0.160
0.03
11.5
Yellow. ID
1992
2
0.09
0.220
0.03
8
Rose, YNP
1988-1992
2
0.10
0.416
0.027
7.8
Sliver ID
1991
2
0.10
0.243
0.04
5
EFWhinistick. ID
1991
2
0.10
0.460
0.02
2
Cache. YNP
1988-1992
2
0.11
0.832
0.012
8.8
Cliff, ID
1988,90.91
2
0.12
0.407
0.18
12
Amphitheater, YNP
1988-1992
2
0.13
1.11
0.146
4.9
Pony, ID
1992
2
0.13
0.380
0.08
13
Iron Springs, YNP
1988-1992
2
0.14
0.237
0.038
13.1
E McCall, ID
1991
2
0.14
0.196
0.05
2
Blacktail Deer, YNP
1988-1992
2
0.15
0.710
0.151
15.2
Fair>; YNP
1988-1992
2
0.18
0.395
0.083
0.26
WF Cave, ID
1990
3
0.05
0.124
0.01
6
Doe, ID
1990
3
0.10
0.316
0.02
16
SF Cache, YNP
1988-1992
3
0.16
1.70
0.195
3.0
Pioneer, ID
1990
3
0.16
0.612
0.31
6
Hellroaring, YNP
1988-1992
3
0.17
1.23
0.32
2.5
McCall, ID
1991
3
0.17
0.196
0.05
2
Pebble, YNP
1988-1992
3
0.18
1.10
0.592
2.5
Cougar, ID
1990,91
3
0.18
0.297
0.10
12
Cache, YNP
1988-1992
3
0.19
4.60
0.475
1.7
Lava, YNP
1988-1992
3
0.24
0.768
0.893
2.1
Iron Springs, YNP
1988-1992
3
0.27
0.587
0.520
1.1
Beaver, ID
1988
3
0.27
0.800
1.17
4
Cache, YNP
1988-1992
4
0.18
2.05
0.67
1.2
Ramey, ID
1988
4
0.18
0.630
0.74
3.5
Boulder, ID
1992
4
0.19
1.23
0.41
2
Hellroaring, YNP
1988-1992
4
0.20
2.61
0.43
1.8
McCall, ID
1991
4
0.22
0.240
0.13
2
Whimstick Main, ID
1991
4
0.23
0.800
0.10
1
WF Rapid, ID
1992
4
0.25
0.930
1.80
3
Lamar YNP
1988-1992
4
0.34
2.87
2.85
.97
Soda Butte, YNP
1988-1992
4
0.35
2.90
3.00
1.3
Indian, ID
1992
5
0.21
1.43
1.31
1.5
Pistol. ID
1992
5
0.33
1.70
1.80
1.8
Rush, ID
1988
5
0.35
1.51
1.61
1
Camas, ID
1992
5
0.38
2.10
2.92
1
Chamberlain, ID
1992
6
0.24
1.69
2.43
3.5
Big Ck @ Coxey, ID
1988
6
0.31
3.42
5.23
1.5
Rapid, ID
1992
6
0.37
1.48
1.11
2.5
Loon, ID
1992
6
0.37
2.91
3.29
1
Big Ck @ Gorge, ID
1988
6
0.37
4.32
8.83
1
Big Ck@ Rush, ID
1988
6
0.45
4.3
8.04
1.5
Salmon Ri\en ID
1992
6
0.48
1.40
5.47
1
290
Great Basin Naturalist
[Volume 56
^
CN
«:
Ostraco^SL \CHLa
P. elsis
Axis #1 (X Axis)
Fig. 1. Biplot results of canonical correspondence analysis. Environmental variables (circled) are listed in Table 3.
Species are plotted using species names. Positive abundance relationships with a given environmental xariable are indi-
cated by species that fall close in the ordination plot to the environmental variable. Species that fall on the opposite end
of the plot from an environmental variable exhibit a negative relationship with that variable. Species near the center of
the plot exhibit little relationship with environmental variables.
did not relate to any of the environmental vari-
ables in the ordination plot and are not consid-
ered further
Stepwise multiple regression results indi-
cate species-environment relationships similar
to those found in the ordination (Table 3). Arc-
topsijche grandis abundance was positively
correlated with stream depth and width and
negatively correlated with turbident flow
(Reynolds number). BracJjycentni.s abimdance
was positively correlated with stream depth
(Table 3). Hydropsyche abundance showed
positive correlation with baseflow and negatixe
correlation xvith water hardness and substrate
size. Parapsyche elsis abundance showed nega-
tive correlation with depth, surface turbulence
(Froude number), and specific conductance
(Table 3).
Discussion
Our results support the idea that macroin-
vertebrate species in streams respond to envi-
ronmental conditions in individualistic ways.
Each ta.xon was related to a different set of
enx'ironmental variables. General relationships
xvith environmental variables for A. grandis,
Brachycentrus, and Hydropsyche suggest that
these ta.xa are adapted for larger rixer systems
(3rd-6th order; Fig. 2). Brachycentrus and Hy-
dropsyche are usually found in loxxer reaches
in river systems (4th-6th order; Edington and
Hildrexv 1973, Boon 1978, Hauer and Stanford
1982, Ross and Wallace 1982, Wetmore et. al.
1990). A. grandis is most often found in mid-
reaches (3rd-5th order; Alstad 1980, Cuffney
and Minshall 1981, Hauer and Stanford 1982).'
1996]
Filter-feeding Invertebrates in Rocky Mountain Streams
291
Table 2. Ht'sulls of canonical correspondence analysis. Eigenvalues give the importance of an axis on a scale between
0 and 1. Total inertia is the total variance in the species data. The species-environment correlations scale the strength of
the relationship hetwcen species and en\ironment lor the axes.
Total
Axes
1
2
3
4
inertia
Eigenvalues
.444
.360
.174
.085
2.50
Species-en vironnieul correlations
.S19
.772
.640
.441
Cumulative percentage of \arianci':
of species data
17.8
32.1
39.1
42.5
of species-environment relationsh
P
.37.9
68.6
83.5
90.7
Sum of all canonical eigenvalues
1.172
Among the taxa adapted for large streams,
habitat partitioning is apparent in tliis study as
in others (Edington and Hildrew 1973, Boon
1978, Alstad 1980, Hauer and Stanford 1982,
Ross and Wallace 1982). Taxa exhibited reach-
scale macrohabitat preferences with Brachij-
centnis distribution related to stream depth,
Hydropsijche related primarily to stream flow,
and A. grandis to a combination of width,
depth, and turbulence.
P. elsis was prevalent in headwater stream
reaches (Fig. 2), a pattern found in several
other studies (Alstad 1980, Hauer and Stanford
1982). Distribution patterns for P. elsis were
explained by flow and stream-size variables.
Stream temperature may also be an important
variable explaining P. elsis distribution patterns
(Alstad 1980, Hauer and Stanford 1982). Annual
temperature was measured in this study based
only on yearly max/min readings, which may
not adequately reflect differences in tempera-
ture between headwater sites and downstream
locations, resulting in the lack of P. elsis pat-
terns explained by temperature in our analysis.
Also, previous studies that suggest a down-
stream temperature gradient as the explanation
for P. elsis distribution (Hauer and Stanford
1982) did not consider other variables (e.g.,
physical and hydrologic variables) that may con-
tribute to habitat selection by P. elsis. Multiple
factors are probably responsible for P. elsis high
abundance in headwater streams, including
temperature patterns and hydraulic conditions.
Our results agree with published distribution
patterns for all 4 taxa and provide evidence for
physical factors that are important in determin-
ing habitat selection for each taxon (Fig. 2).
Habitat preferences demonstrated in this study
are for distribution patteiTis among streams at
the reach scale. Data were collected within a
250-m reach at each site and expressed as
reach means for all variables in order to iden-
tify factors affecting large-scale (among site)
distribution patterns among taxa. Microhabitat
requirements are ultimately responsible for the
physiciil habitat selected by filter feeders (Smith-
Cuffney and Wallace 1987, Wetmore et. al.
1990), but reach-scale comparisons allow
broader scale distribution patterns to be stud-
ied. The reach-scale comparisons herein indi-
cate general conditions at each site in temis of
available macrohabitat. The trends observed in
the data indicate animal preferences for a
given reach and its associated habitat condi-
tions. Differences in reach-scale means among
variables may also reflect differences in gen-
eral microhabitat conditions available among
sites (e.g., slow- or fast-velocit\' microhabitats).
Reach-scale means, therefore, can ser\e as a
useful integrator of microhabitat conditions in
order to facilitate comparisons at larger scales.
Evolutionary patterns probably have led to
habitat partitioning based on current speed
and filtration rate among filter feeders in Rock>'
Mountain streams with some taxa adapted for
larger streams {Brachycentrus, Hydropsyche,
and A. grandis) and some for smaller systems
{P elsis; Alstad 1980, 1982). Filter feeders may
be a usefijl group to address habitat partition-
ing on large spatial scales in streams because
many filterer taxa appear to have partitioned
habitat at these scales. In this study, stream
size (width, depth) and hydraulic parameters
(l)aseflow, turbulence) were more important in
explaining species-environment relationships
than other variables such as water chemistry,
periphyton biomass, or benthic organic matter.
Our results provide support for the idea that
evolutionaiy divergence among benthic macro-
invertebrate filterers has resulted in habitat
partitioning according to stream size and hydro-
logic parameters in Rocky Mountain streams
292
Great Basin Naturalist
[Volume 56
Table 3. Summaiy of the stepwise multiple regression results of the 4 most abundant species (dependent variable)
versus the 20 environmental variables. Partial con-elation coefficients and p values (parentheses) are shown for each
variable. Variables included in the regression model for each species are shown (varial)le included if P < 0.05). Variable
acronyms in Figure 1 are shown in parentheses.
Environmental
Arctopsijche
Braclujcciitnis
Hiidropsijchc
Parap.sijche
variable
'.
1996]
Filter-feeding Invertebrates in Rocky Mountain Streams
293
Acknowledgments
Yellowstone National Park streams were
sampled as part of a study supported by the
University of Wyoming/National Park Service
Research Center. Idaho stream sampling was
supported by the U.S. Forest Service, Idaho
Division of Environmental Quality; and Idaho
State University. Numerous members of the
Stream Ecology Center at Idaho State Univer-
sity aided in the field sampling, including P
Dey, P Koestier III, D. E. Lawrence, M. J.
Mclntrye, J. N. Minshall, G. C. Mladenka, D. C.
Moser, J. S. Nelson, C. T. Robinson, R. L. Van-
note, and many others. C. T. Robinson did
much of the data compilation for the environ-
mental variables that were used in the analy-
ses. Data analysis and writing by TBM and
JRM were supported in part by Louisiana
State University.
Literature Cited
Alstad, D. N. 1980. Comparative biolog>' of the common
Utali Hydropsycliidae. American Midland Natiualist
103; 167-174. '
. 1982. Cunent speed and filtration rate link cad-
disfly phylogeny and distributional patterns on a
stream gradient. Science 216: 533-534.
Boon, E J. 1978. The pre-impoundment distribution of
certain Trichoptera larvae in the North Tyne river
system, with particular reference to current speed.
Hydrobiologia 57: 167-174.
Bruns, D. a., a. B. Hale, and G. W. Minshall. 1987.
Ecological correlates of species richness in three
guilds of lotic macroi7i\'ertebrates. Journal of Fresh-
water Ecology 4: 163-176.
CUFFNEY, T. E, and G. W. MiNSHALL. 1981. Life histoiy
and bionomics of Arctopsyche granclis (Trichoptera)
in a central Idaho stream. Holarctic Ecologv 4:
252-262.
Decamps, H. 1968. Vicariances ecologiques chez les Tri-
chopteres de Pyrenees. Annates de Limnologie 4:
1-50.
Edington, J. M. 1968. Habitat preferences in net spinning
caddis larvae with special reference to the influence
of water velocity. Journal of Animal Ecologv 37:
675-692.
Edington, J. M., and A. H. Hildrevv. 1973. Experimental
observations relating to the distribution of net-spin-
ning Trichoptera in streams. Verhandlungen der
Internationalen Vereinigung fuer Limnologie 18:
1549-1558.
Gordon, E., and J. B. Wallace. 1975. Distribution of the
family Hydropsycliidae (Trichoptera) in the Savannah
River basin of North Carolina and Georgia. Hydrobi-
ologia 46: 405-123.
Haddock, J. D. 1977. The effect of stream cunent velocity
on the habitat preference of a net-spinning caddis fly
larva, Ihjdropsijche oslari Banks. Pan-Pacific Ento-
mologist 53: 169-174.
llAi 1., C. A. S., J. A. Stanford, and E R. Hauer. 1992.
rhe distribution and abundance of organisms as a
consecjucnce of energy balances along n\ultiple envi-
ronmental gradients. Oikos 65: 377-390.
Hauer, E R., and J. A. Stanford. 1982. Ecological
responses of hydropsychid caddisflies to stream regu-
lation. Canadian Journal of Fisheries and Aquatic Sci-
ences 39: 1235-1242.
Heard, S. B., and J. S. Richardson. 1995. Shredder-col-
lector facilitation in stream detrital food webs: is
there enough evidence? Oikos 72: 359-.366.
OsRORNE, L. L., AND E. E. Herricks. 1987. Microhabitat
characteristics o'i Hyihopsijche and the importance of
body size. Journal of the North American Bentholog-
ical Society 6: 115-124.
Platts, W. S., W. E Megahan, and G. W. Minshall. 1983.
Methods for evaluating stream, riparian, and biotic
conditions. U.S. Forest Senice General Technical
Report INT- 138. 70 pp.
Richardson, J. S., and R. J. Mack.\v. 1991. Lake outlets
and the distribution of filter feeders: an assessment of
hypotheses. Oikos 62: 370-380.
Robinson, C. T, and G. W. Minshall. 1990. Longitudinal
development of macroinvertebrate commimities be-
low oligotrophic lake outlets. Great Basin Naturalist
50:303-311.
Ross, D. H., and J. B. Wall.'\ce. 1982. Factors influencing
the longitudinal distribution of larval Hydropsychi-
dae in a southern Appalachian stream system.
Hydrobiologia 96: 185-199.
S.MITH-CUFFNEY, E L., AND J. B. WALLACE. 1987. The
influence of microhabitat on availability of drifting
invertebrate prey to a net-spinning caddisfly. Fresh-
water Biology 17: 91-98.
Tachet, H., J. E Peirrot C. Rou.x, and M. Bournaud.
1992. Net-building behaviour of si.x Hydropsyche
species (Trichoptera) in relation to ciUTent velocit)' and
distribution along the Rhone River. Journal of the
North American Benthological Societ>' 11: 350-.365.
Ter Bra.\k, C. J. F. 1986. Canonical conespondence analy-
sis: a new eigenvector techni(iue for direct gradient
analysis. Ecology 67: 1167-1179.
VoELZ, N. J., AND J. V. Ward. 1992. Feeding habits and
food resources of filter-feeding Trichoptera in a regu-
lated mountain stream. Hydrobiologia 231: 187-196.
Wallace, J. B. 1974. Food partitioning in net-spinning Tri-
choptera lai-vae: Hydropsyche ventdaris, Cheumato-
psyche etrona, and Macronema zehrafum. Annals of
the Entomological Society of America 68: 463—472.
Wallace, J. B., and R. W Merritt. 1980. Filter-feeding
ecology of aciuatic insects. Annual Review of Ento-
mology^' 25: 103-132.
Wetmore, S. H., R. J. Mackay, and R. W Newbury. 1990.
Characterization of the hydraulic habitat of Brachy-
centrus occidentalism a filter feeding caddisfly. Journal
of the North American Benthological Societ)' 9;
157-169.
Received 6 December 1995
Accepted 27 June 1996
Great Basin Naturalist 56(4), © 1996, pp. 294-299
STEM GROWTH AND LONGEVITY DYNAMICS
FOR SAL/A AR7ZOMCA DORN
Vicki L. Taylor^ Kimball T. Harper^, and Leroy L. Mead^
Abstract. — Diameter-age relationships of Salix arizonica (Arizona willow) stems were investigated for 5 populations
on the Markagunt, Paunsaugunt, and Sevier plateaus in southern and central Utah. Of the 430 stems studied, none
exceeded 26 mm in diameter at ground level (estimated age of 19 \ r). Equations developed for predicting age from stem
diameters consistently accounted for over 90% of the obsei-\'ed variation. Slopes of predictive equations were homoge-
neous across die 3 sites considered in detail. At 2 sites 46% and 38% of the stems exceeded 10 mm (~7 yr old) diameter
at ground level. At a 3rd site, no stems survived to exceed that size. Stem-age profiles at specific sites may thus be usefld
for assessing the relatixe favorability of local enxironments for the species.
Key words: Arizona icilloic, Salix, stem diameter dendroehi
jnviology. sou
them Utah.
The puipose of this study was to assess stem
diameter-age relationships in Salix arizonica
(Arizona willow), a species so rare that routine
severance of stems for aging cannot be justi-
fied. Our objective was to develop a stem-age
prediction model based on stem basal diame-
ters. Ultimately, we desired to accurately esti-
mate stem age at a broad range of ecological
situations without sacrificing stems. We also
evaluate the possibilit}' of using stem-age pro-
files at an array of sites to determine their rela-
tive favorabilit)' for growth of S. arizonica.
Dendrochronology as a Tool
Growth rings of trees and shrubs have been
used for many decades for aging stems and
dating past climatic events (Douglas 1935,
Glock 1937). Growth rings are also used to
establish unique sequences of good and poor
years that permit dating nonliving tree frag-
ments used in prehistoric human structures
(Schulman 1956, Fritts 1971, Stockton and
Meko 1975, Harper 1979). Ring-width varia-
tions are often used to assess differences in the
favorability of various environments for the
growth of selected species (Ferguson and
Humphrey 1959, Fritts 1962, Stockton and
Fritts 1973, Fritts 1974). Although these stud-
ies have focused mainly on trees (Glock 1955,
Argeter and Glock 1965), some have dealt with
shrub species (Ferguson 1958, 1959, Ferguson
and Humphrey 1959, Brotherson et al. 1984,
1987). Shrub studies have detailed the effects
of variations in available moisture on plant
growth in specific habitats or provided infor-
mation for interpreting archaeological prob-
lems. Ring counts have also been used to pre-
dict stem diameter-age relationships in predic-
tive models for inteipreting site quality for var-
ious species or for clarification of successional
patterns in vegetation diat includes many woody
species (Brotherson et al. 1984, 1987).
The Species and Its Distribution
Salix arizonica is small. Rarely do stems ex-
ceed 1.0 m in height. The species occurs in such
dense carpets of other species (both vascular
and nonvascular) that reproduction via its tiny,
wind-dispersed seeds appears to be uncom-
mon. Accordingly, the species apparently per-
sists at occupied sites primarily by vegetative
reproduction. In the process, what appear to
be large clones (as much as 10 m across) may
develop.
Salix arizonica occurs in 2 disjunct locations
in the Intermountain West. The species was
first discovered on the White Mountains of
east central Arizona by Carl-Eric Granfelt in
1969 (Galeano-Fopp 1988). Robert Dom (1975)
used holotype specimens collected by Granfelt
to describe the species in 1975. In November
1992, unaware that the species occurred in
Utah, the U.S. Fish and Wildlife Service pro-
posed S. arizonica for listing as endangered with
'Departriiciil dl HotaiiN ami Haiiiic ScioiKv, HriKliaiii Vouiii; Uiii\cr.sity, Provo, UT 84(i()2.
294
1996]
Arizona Willow Stlm CtHowth
295
designation of critical habitat (Atwood 1995).
In June 1993 a prcvionsly inisidcntilicd herb-
arium specimen of S. arizonica was discovered;
it had l)een collected on the "Sevier Forest"
(now Dixie National Forest) in 1913. During
June 1994, S. arizonica was discovered on the
Markagimt I^lateau near Brianhead resort area.
Subsequent searching revealed a small popula-
tion on the Paunsaugunt Plateau and 2 more
farther north on the Sevier Plateau (Mead
1996). Following this "rediscovery" of S. ari-
zonica in Utah, USDA Forest Service, USDI
Fish and Wildlife Sei-vice, and USDI National
Park Service officials cooperated in developing
a conservation agreement and strategy that
outlines the "actions, costs and skills needed to
implement protective measures and research
studies needed for the species" (Atwood 1995).
As a result of the consei"vation agreement and
strategy, which documents long-term plans for
consei-vation of S. arizonica, the Fish and Wild-
life Service withdrew their proposed rule to
list the species as endangered (Arizona Willow
Interagency Technical Team 1995).
Although die species is locally abundant near
Brianhead, its total range is small in both Ari-
zona and Utah, and populations rarely include
more than a few score plants. This rarity seems
related to the plant's preference for an uncom-
mon habitat: it grows preferentially on igneous
soils in cold, wet sites. In addition, in the White
Mountains, management has favored conifers
that reduce flow in riparian systems, leading
to poor drainage as waterways become peat-
choked. Such environments become poorly aer-
ated and less suitable habitat for S. arizonica.
Heavy use by elk has also adversely affected
the species in Arizona (Arizona Willow Intera-
gency Technical Team 1995). This study has
been confined to the Utah populations of Salix
arizonica (Fig. 1), but we have attempted to
sample the full range of conditions associated
with the species in our study area.
Methods and Study Areas
The diameter-age data for S. arizonica were
collected from 3 populations: 2 on the Cedar
City Ranger District and another on the Powell
Ranger District, Dixie National Forest (Fig. 1).
The Rainbow Meadows, Lowder Creek, and
East Fork of the Sevier River populations were
chosen because they represent environmen-
tally intermediate (Lowder Creek) as well as
extreme environmental conditions for S. ari-
zonica in Utah. The Rainbow Meadows site
occurs on acid soils at near niitximal elevations
for the species, while the East Fork of the
Sevier River population occurs on alluvium
derived from calcareous substrates at the low-
est elevation known for the species.
Depth of p(>at la>'er was determined at each
site by digging pits to expose soil profiles (Mead
1996). At Lowder Creek, Sheepherder Camp,
and Sevenmile Creek, depth to water table was
determined by opening a hole approximately 1
m deep with a 1.27-cm-diameter pointed rod,
then inserting a ().64-cm-diameter wooden
dowel into the hole to measure depth to water.
This measurement was taken at each plant sam-
pled and an average value was computed for
each site. Depth to water table at Rainbow
Meadows was determined by measuring dis-
tance from soil surface to water table surface in
a soil pit (Mead 1996). Depth to water table
was determined at the East Fork site by mea-
suring distance from the soil surface to the sur-
face of water running in the creek. This mea-
surement was taken at each S. arizonica clone;
the mean distance is reported in Table 1. Depth
of peat layer and depth to water table are vari-
able among the study sites, with the Rainbow
Meadows site having the highest water table
and greatest peat depth (Table 1).
Two otlier populations of S. arizonica are con-
sidered in this report. Populations at Sheep-
herder Camp, Sevenmile Creek, and Lowder
Creek have been sampled to establish stem-
diameter profiles based on samples of many
randomly chosen stems (154, 104, and 130
stems, respectively, sampled at the 3 foregoing
sites). No stems were severed for aging at the
Sheepherder or Sevenmile sites.
The Rainbow Meadows site is approximately
1.6 km south and slightly east of Brianliead Peak
at approximately 3155 m elevation (37°40'N,
112°56'W). Soils are derived from tertiaiy vol-
canics with a histosol surface horizon (Mead
1996). The Lowder Cieek population is approx-
imately 4 km east and slightly south of Brian-
head Peak (37°41'N, 112°48'W). Soil at this
site is developed from tertiaiy volcanic mater-
ial below an alluvium surface layer (Mead
1996). The East Fork population, approximately
48 km from the Lowder population (37°26'N,
112°2rW), is at the lowest elevation knowai for
296
Great Basin Naturalist
[Volume 56
UTAH
Fig. 1. A, Rainbow Meadows; B, Lowder Creek; C, Sheepherder Camp; D, Sevenniile Creek; E, East Fork of die
Sevier River
this species in Utah. This population grows on
alluvium from the Claron Limestone Forma-
tion with an organic surface horizon (Mead
1996).
Commonly associated plants at the sites sam-
pled include Salix planifolia. Polygonum bistor-
toides, Aconitum cohimbimnim, Carex rnicro-
ptera. Geranium richardsonii, Geum macro-
phijllwn, and Pedicuhiris groenlamlica (Mead
1996). As Mead (1996) has shown, the relative
abundance of these species varies from site to
site depending on such variables as soil temper-
ature, depth to water table, and soil reaction.
Fifteen randomly chosen stems were sam-
pled at each site at the Rainbow and Lowder
locations. At each site 4 quadrants were estab-
lished around randomly chosen points. The
stem closest to the random point in each of 5
size-classes was collected in each of 3 quad-
rants (the right rear quadrant was not sampled).
Stems were severed at ground level using wire
cutters or a small hand saw. The diameter-
classes sampled were 0-5 mm, 5.1-10 mm,
10.1-15 mm, 15.1-20 mm, and >20 mm at
ground level. Thus, 3 stems per size-class were
sampled at each site. Due to the low densit)' of
S. arizonica at the East Fork site, quadrants
were not used. Stems were collected from all
S. arizonica clones inside a livestock- grazing
exclosure in the stiuK' area. No stems could be
found at this site for the >20 mm size-class, so
only 12 stems were sampled.
Stem samples were labeled, placed in indi-
\idual bags, and taken to the lab. Stem bases
1996]
Arizona Willow S ilm Growth
297
Taulk L Environmental eonditions at 5 Salix arizonka sites. Water table was taken at all plants sampled wherever
soil stoniness permitted insertion of flie dowel to water depth. At East Fork water depth was based on only 16 points
beeanse only 16 plants e.xist at that site. The measme ot varianee around water table mean depth is standard error.
Site
Ele\ation
Soilpll
Soil temp, (a)
Mean depth to
Peat depth
(m)
50-cm depth (°C)
water table (em)
(em)
Rainbow
3155
5.15
8.3° (September)
5.1 ±NA
32
Lowder
3139
5.79
10° (August)
45.5 ±1.81
0
Sheepherder
3130
5.72
6° (August)
44.4 ±1.60
44
Sevenmile
2789
6.38
10° (August)
10.5 ±1.12
0
East Fork
2536
7.61
16° (July)
46.5 ± 6.89
0
N.\ = not a\ailal)le.
Table 2. Regression ecjuations relating stem diameter to age of S'«/j.v arizonica stems taken from 3 different sites. The
regression equation for all sites combined is also shown. In the equation the independent variable, X, represents stem
diameter (in mm). The symbol Y represents estimated age of any given stem.
Site
No. of
stems
Equation
fi2
Signiiieance
lexel
Lowder
15
Y = -0.42 + 0.82X
.953
0.01
Rainbow
15
Y - -0.28 + 0.78X
.950
0.01
East Fork
12
Y = -1.40 + 0.71X
.910
0.01
All 3 sites combined
42
Y = -0.99 + 0,81X
.926
0,01
were sectioned diagonally and sanded with fine
sandpaper; growth rings were counted twice
(once by each of 2 observers) with the aid of a
stereoscopic microscope (Brodierson et al. 1987).
Diagonally cut surfaces permitted growth rings
to be identified widi greater confidence. Sanded
surfaces sometimes had to be polished with
immersion lens oil to enhance ring visibility.
Each growth ring was assumed to represent 1
yr's growth. Linear regression was used to
quantify stem diameter-age relationships.
Results
S. arizonica stems from the 3 sites at which
stems were cut and aged ranged in basal diam-
eter from 2 to 26 mm and in age from 1 to 19
yr. Stem diameters (mm) were plotted against
stem age (yr), and regi"ession equations were
generated (Table 2). Slopes for regression
equations from the 3 sites were tested for simi-
larity using methods described in Snedecor
and Cochran (1967) and were found not to dif-
fer significantly (F > 0.50). Thus, data from all
sites were pooled to produce a single equation
(Y = -0.99 + 0.8 IX) for subsequent use in
estimating age (Y) from diameter (X) (Fig 2).
As a further test of the validity of pooling
data from all sites, we used the individual esti-
mator equation developed for each site to pre-
dict age of willows collected from the other 2
sites (i.e.. Rainbow equation used to test Low-
der and East Fork samples, Lowder equation
used to test Rainbow and East Fork samples,
etc.). These analyses demonstrated that esti-
mated ages for any equation-test site combina-
tion were always strongly correlated with actual
age {R^ always > .90). In these analyses no
stems were found to differ fi^om predicted age
based on diameter by more than 3 yr, and most
stems (>90%) differed by less than 2 \t (Fig. 1).
An application of the age-estimator equa-
tion is shown in Figure 3. As part of the yearly
monitoring program, basal diameters of S. ari-
zonica were taken for a large sample of stems
at each of 3 sites: Sheepherder Camp, located
approximately 8 kin south of Brianhead Peak at
3130 m elevation (37°37'N, 112°56'W); Sev-
enmile Creek, 11 km north of Fish Lake in the
Fishlake National Forest, Loa Ranger District
at 2789 km elevation (38°39'N, 111° 40 'W);
and Lowder Creek (described above). At each
of these sites, the numbers of stems within each
diameter-class were tabulated and are reported
as percent of total stems in each size-class. The
results (Fig. 3) demonstrate large differences in
stem-diameter profiles among the 3 sites. At
Sheepherder Camp over 4% of the stems are
298
Great Basin Naturalist
[Volume 56
Y = -0.99 + 0.8 IX ^
R' = 0.926 ■ Qp, □
10 20
Basal Diameter (mm)
30
Fig. 2. Stem basal diameter-age relationships of S. ari-
zonica on the Markagunt and Paunsaugunt plateaus of
southern Utah.
larger than 20 mm diameter at ground level.
However, less than 1% of the stems at Lowder
Creek exceed that diameter, and at Sevenmile
Creek no stems have sui-vived to become 10
mm in diameter. These results suggest that
Sheepherder Camp is a more favorable site for
growth of the willow than either Lowder Creek
or Sevenmile Creek. Alternatively, the results
may indicate that willows are less severely
browsed by ungulate grazers at Sheepherder
Camp than at the other 2 sites. Since ungulate
exclosures were not erected at these sites until
fall 1994, data are cuirently too limited to dis-
tinguish between these alteiTiatives.
Discussion
The regression equation created from the
pooled data of all 3 sites should be useful for
predicting ages of S. arizonica from any known
Utah location using only basal stem diameters.
The equation should be useful for many pro-
jects in which stem age is desired but stems
cannot be sacrificed. For example, the ability to
estimate age of stems accurately from basal
diameter may permit scientists studying the
species to correlate stem ages and stem-age
profiles with site conditions without destroying
individual stems.
The results of this study demonstrate little
variation in stem growth rates for S. arizonica
over a wide range of ele\ations and parent
materials (Table 1). That result suggests that the
species occupies but a narrow range of habitat
100
% of All
Stems
Stem Diameter
Class
■ 0-5 mm
B 5.1-10 mm
m 10.1-15 mm
0 15.1-20 mm
20.1-25 mm
>25mm
D
Lowder
Sheepherder
Sevenmile
Site
Fig. 3. C'oinpar:ili\e stem diameter distrihutioiis foi' sites iur whicii a large, rantloiii in\eiit()r\ oi stem diameters was
available.
1996J
Arizona Willow Stlm CiuowTii
299
situations witliin its o\CM-all geographic range.
Occupied sites almost always appear to have
been modified by biological processes that
result in peat deposition and development of a
rooting zone that is somewhat isolated hom
the unaltered geologic substrata at the site.
Stem-age profiles should permit managers
to identify sites where performance (stem sur-
vival and/or reproduction b\' seed oi" rhizome)
of the willow is above or below regional aver-
ages. Such data would help managers deter-
mine whether growth and i^eproduction of the
species could be enhanced by reduction of use
b)- browsers. To assist managers with such deci-
sions, fenced areas that exclude domestic and
wild ungulate browsers have been erected at
Lowder Creek, Sheepherder Camp, and on the
East Fork of the Sevier. An additional exclo-
sure will be built at Sevenmile Creek in 1996.
The U.S. Forest Sei'vice intends to continue
monitoring Salix arizonica populations through-
out its range to learn about factors that influ-
ence growth, reproduction, and stem sui^vival.
Data from grazing exclosures will reveal the
extent to which browsing controls stem size
and longevity. The extent to which the abiotic
environment limits stem growth and seed pro-
duction can be more readily separated from
the effects of browsing now that animal exclo-
sures have been constructed.
Acknowledgments
We thank Ron Rodriguez for financial and
moral support for this project. Julie Tolman
provided invaluable assistance in the field and
lab. This work was completed in part with
hmds provided by the U.S. Forest Service.
Literature Cited
Argeter, S. R., and W. S. Clock. 1965. An annotated bib-
liography of ti-ee growtli and growth rings, 1950-1962.
University of Arizona Press, Tucson. 180 pp.
Arizona Willow Interagency Technical Team. 1995.
Arizona willow consei"vation agreement and strategy.
U.S. Forest Sei"vice, Intermountain Region, Ogden,
UT; U.S. Forest Service, Southwest Region, Albu-
querque, NM; National Park Sei-vice, Rocky Moun-
tain Region, Denver, CO; U.S. Fish and Wildlife Ser-
vice, Mountain-Prairie Region, Salt Lake City, UT;
U.S. Fish and Wildlife Sei-vice, Southwest Region,
Albuquerque, NM.
Atwood, D. 1995. Where have all the Arizona willows
gone? Sego Lily (Newsletter of the Utah Native Plant
Society) 18; 3.
Brothkhson, J. D., J. G. Carman, and L. A. S/yska. 1984.
Steni-dianieter age relationships of Tamxiiix ravwsui-
sima in central Utah. Journal oi' Range Management
37: 362-364.
BiurniERsoN, J. D., K. P Price, and L. ()"R()Urke. 1987.
Age in relationship to stem circumference and stem
diameter in cliffrose {Cowania mexicana van stam-
huriana) in central Utah. Great Basin Naturalist 47:
334-338.
DoRN, R. D. 1975. A systematic stud> oi' Salix section Cor-
datae in Nordi America. (Canadian Joinnal of Bolanv
53: 1491-1522.
Douglas, A. E. 1935. Cliiuatic cycles and tree growth I: a
study of the annual rings of trees in relation to cli-
mate and solar activity CJarnegie Institute of Wash-
ington, Publication 289. Volume I. Washington, DC.
Ferguson, C. W. 1958. (irowth rings in big sagebrush as a
possible aid in dating archaeological sites. Pages 210-
211 in A. E. Dittert, Jr., editor. Recent developments
in Navajo Project salvage archaeol()g\. El Palacio 65:
201-211.
. 1959. Growth rings in wood) shrubs as potential
aids in archaeological interpretation. Kiva 25: 24-.30.
Ferguson, C. W, and R. R. Himphrey. 1959. Growth rings
on big sagebnish reveal rainfall records. Progressive
Agriculture in Arizona 11; 3.
Fritts, H. C. 1962. The relation of growth ring widths in
American beech and white oak to xariations in cli-
mate. Tree-Ring Bulletin 25; 2-10.
. 1971. Dendroclimatology and dendroecologv'.
Quaternan' Research 1: 419-449.
. 1974. Relationships of ring widths in arid-site coni-
fers to variations in monthly temperature and precip-
itation. Ecological Monographs 44: 411—140.
Galeano-Popp, R. 1988. Salix arizonica Dom on the Apache-
Sitgreaves National Forest: inventon' and habitat
study. USDA Forest Service, Apache-Sitgreaves
National Forest, Contract 43-8173-8-687. 46 pp.
Clock, W. S. 1937. Principles and methods of tree-ring
imalysis. Carnegie Institute of Washington, Publication
486. Washington, DC. 100 pp.
. 1955. Tree growth II. Growth rings and climate.
Botanical Review 21; 73-188.
Harper, K. T. 1979. Dendrochronologv' — dating with tree
rings. In W M. Hess and R. T. Matheny editors. Sci-
ence and religion: toward a more useful dialogue.
Volume 1. Paladin House Publishers, Geneva, IL.
Mead, L. L. 1996. Habitat characteristics of Arizona willow
in southwestern Utah. Unpublished master's thesis,
Brigham Yoimg University, Provo, UT.
SCHULMAN, E. 1956. Dendroclimatic changes in semi-arid
America. University' of Aiizona Press, Tucson. 142 pp.
Snedegor, C. W, and W. G. Cochran. 1967. Statistical
methods. Iowa State University Press, Ames. 593 pp.
Stockton, C. W, and H. C. Fritts. 1973. Ix)ng-term recon-
struction of water level changes for Lake Athabaska
by analysis of tree rings. Water Resources Bulletin 9:
1006-27.
Stockton, C. W, and D. M. Meko. 1975. A long-term his-
tory of drought occun^ence in western United States
as inferred from tree rings. Weathenvise 28: 245-249.
Received 15 March 1996
Accepted 5 July 1996
Great Basin Naturalist 56(4), © 1996, pii. 300-307
SOURCES OF VARIATION IN COUNTS OF MERISTIC
FEATURES OF YELLOWSTONE CUTTHROAT TROUT
[ONCORHYNCHUS CLARKI BOUVIERI)
Carter G. Kru.se^ \Va\ ne A. Huljert^, and Frank J. Raliel-
Abstfl\ct. — We determined variability in counts of nieristic featines (pyloric caecae, vertebrae, pelvic fin rays, gill-
rakers, basibranchial teeUi, scales above the lateral line, and scales in the lateral series) of Yellowstone cutdiroat trout
{Oncorhijnchiis chirki buiivieri) by 3 independent readers, by the same reader on 3 different occasions, and among fish
from 12 sampling sites within a 650-kni- watershed. Genetic purit>- of the cutthroat trout was determined b\' elec-
trophoretic analysis. Significant differences in nieristic counts were obsen'ed among 3 readers and among sampling sites,
but not among 3 occasions b\ a single reader Scale counts were within the reported range for Yellowstone cutthroat trout,
but counts of other structures (pyloric caecae, gillrakers, \ertebrae) were as similar to rainbow trout as to Yellowstone cut-
throat trout. Meristic counts identified the fish as cutthroat trout; however, variation among readers and sampling sites, as
well as within the species, limits their use w hen identih ing geneticalK pure cutthroat trout or assessing possible integi"a-
tion with rainbow trout.
Key words: meristic counts, Yclloustonc cutthroat trout, lucristic variation, genetics, rainlnnc trout, conservation
biology.
H>briclizati()n of nati\'e cutthroat trout
{OncorJiynchiis clarki) with introduced rainbow
trout (O. mykiss) has contributed to the dechne
of cutthroat trout in the western United States
(Allendorf and Leary 1988, Gresswell 1988,
Behnke 1992). An important initial step toward
restoration or presentation of native cutthroat
trout populations is reliable identification of
geneticalK' pure populations (Rinne 1985, Lean
et al. 1989).
Meristic features, such as fin ra> or \ erte-
brae counts, ha\'c been used to identif)' hy-
bridization among species of trout. The tech-
nic^ue assumes that hybrids are intermediate to
parental taxa and haxe increased morphologi-
cal variance (Leaiy et al. 1985, 1991, Marnell
et al. 1987). This assumption is not alwa\s \ alid
and meristic comparisons can proxide mislead-
ing taxonomic information (Leaiy et al. 1984,
1985, Cunens et al. 1989). Enxironmental influ-
ences and ol)ser\ er error are 2 factors that can
lead to variation in meristic counts for a species
among sampling sites (CuiTcns et ill. 1989, Leaiy
et al. 1991, Hubert and Alexander 1995). Even
though more definitive biochemical methods
have been dc\ eloped (Lean et al. 1987, 1989,
Nielsen 1995), biologists continue to use meristic
features to assess genetic purity of cutthroat
trout populations (Loudenslager and Gall 1980,
Rinne 1985, Behnke 1992).
Protein electrophoresis is a reliable method
of determining genetic status of trout popula-
tions (Marnell et al. 1987, Leary et al. 1989,
Nielsen 1995). Electrophoresis provides data
on allelic frequencies at genetic loci for differ-
ent populations (Avise 1974). Hybridization
can be determined when allele frequencies un-
usual for a particular species are found at sev-
eral diagnostic loci that occur between taxa
(A>ala and Powell 1972, Leaiy et al. 1989). For
example, Yellowstone cutthroat trout (O. c. hou-
vicri) can be differentiated from rainbow ti^out
using alleles at 10 diagnostic loci (R. Leary, Uni-
\ersit\ of Nh)ntana, personal communication).
If this procedure is xalid, managers could
save considerable time and money using meris-
tic features instead of biochemical analysis to
assess genetic purit\- of cutthroat trout. How-
ever, unless xariation in meristic counts is min-
imal among readers or sampling sites, the use-
fulness of meristic features in adecjuateh' assess-
ing genetic purit>- will be limited. The objec-
ti\es of this stiid\ were to determine \ariabil-
it\ in counts of meristic featines (1) among
'us. Ci-olo.nicul Survi'v, Wyoiiiini; ( :ocip.MMlivi- Kisli .uul Wikllilc Ucsrarch liiit. I'liivrrsilv (il WytiMiiivi l,aianiic, WY S2()71-.31fi(i. iTlic Wvcmium,!; Coop-
erative Fish and Wildlife Research Unit is joiiilU sii|)porteil In the UMi\ersit\ olWyoriiini;. llic WVoinini; Came ,iml h'isli l)r|i.»lnuMit. Ocpailnu'iil ol'tlu- Inte-
rior, and Wildlife Manasenient Instilnte.)
^Department of Zooloj,", and Ph\si<)lo,i;\. I nixcrsitv of \\\(iininu, Laramie, WV N2()71-:iHi(v
300
1996]
Variation in Yellowstone Cunii hoax Trout
301
Shoshone National Forest Boundary
15
Greybull River
Wood
River
Cody
77 km/
Meeteetse
34 km
Fig. 1. Map of Wyoming showing the location of the Greybull River drainage. Sites where cuttiiroat trout were sam-
pled are numbered in reference to Table 1 .
independent readers, (2) among counts by a
single reader, and (3) among sampling sites
within a moderate-sized watershed (650 ki
ull River drainage, numlier of fish collected, and sample
sizes from each used for meristic counts and analysis. Genetic status indicated by pure Yellowstone cutthroat trout (P) or
potential finespotted cutthroat trout hybridization (FSC). Ninnber preceding the stream name conesponds to sites in
Figure 1.
Stream
Number offish
collected
Allozyme
analysis
Counted
In all
readers
Counted
by single
reader
1
Anderson
15
2
Brown
17
3
Chimney
16
4
Cow
16
5
Deer
16
6
Dundee
2
7
Eleanor
19
8
Francs Fork
9
Upper Grevbull
15
10
Lo\ver Grevbull
20
11
Jack
21
12
Mabel
2
13
MFWood
15
14
NF Pickett
15
Picket
17
16
Pinev
17
Red
4
18
SF Anderson
19
SFWood
18
20
Venus
16
21
VVarhouse
18
22
W Timber
23
Wood
21
15 (P)
20 (P)
19 (FSC)
15 (FSC)
19 (P)
15 (FSC)
20 (FSC)
5
10
14
16
15
11
16
7
20
10
2
3
4
4
10
14
18
trout, and rainbow trout have been stocked in
the system.
Methods
Twenty-three streams in the Greybull River
drainage were sampled with batteiy baclq^ack
electroshockers from June to September 1994.
Cutthroat trout were collected from 1 site
(12-20 fish) on each of 18 streams. For analysis
puqjoses the upper and lower Greybull River
sites were considered separately (Table 1). Fish
were collected from the midpoint of the length
of each stream in which cutthroat trout were
found. A sample of eye, liver, and muscle tissue
was removed from each fish, wrapped in alu-
minum foil, and frozen within 1 h in liquid
nitrogen. The remainder of each specimen was
preserved in 75% ethyl alcohol. Tissue samples
from each fish were individually identified.
Frozen tissue samples from 7 of the 18
streams were sent to the Wild Trout and
Salmon Genetics Lab (WTSCiL) at the Univer-
sit\' of Montana, Missoula, for genetic analysis.
The 7 sites were selected to represent fish dis-
tribution in the drainage (Tible 1, Fig. 1). Also,
they were close to locations where finespotted
cutthroat trout and rainbow trout had been pre-
viously introduced in the drainage (Wyoming
Game and Fish Department records). Protein
electrophoresis (Allendorf and Phelps 1980,
Leary et al. 1984, Perkins et al. 1993) was per-
foiTned to detect each specimen's genetic char-
acteristics at 45 loci in muscle, liver, or eye tis-
sue. Allele frequencies at 10 diagnostic loci
(Table 2) were evaluated to determine hxbridi-
zation with rainbow trout. Additionally, the
presence of the AK- 1*333 allele was evaluated
to detect possible finespotted cutthroat trout
hybridization.
Seven meristic features were counted on
the preserved cutthroat trout: (1) basibranchiiil
teeth, (2) anterior gillrakers (upper and lower
limb of the first branchial arch), (3) pelvic fin
rays, (4) scales in the lateral series, (5) scales
above the lateral line, (6) pyloric caecae, and
(7) vertebrae (Mamell et d. 1987, Behnke 1992).
Three independent readers (all fisheries biolo-
gists with training in anatomy and taxonomy
of salmonids) counted each meristic structure
on the same 50 cutthroat trout (> 150 mm
total length) chosen randoniK from 9 of the 18
1996]
Variation in Yellowstone Cutthroat Tkolt
303
Tahle 2. Alleles at the 10 diagnostic loci that distin-
guish Yellowstone cntthroat trout and rainbow trout along
with the tissue needed for each. The most coninion alk'le
existing at each loci is listed fust.
Char
icteristic alleles
Locus
Ysc:
KB'r
Tissue
SAAT-1*
165
100,0
Liver
CK-A2*
84
100
Muscle
CK-Cl*
38
100,150,38
Eye
mIDHP-1*
75
100
M uscle
sIDHP-1*
71
100,114,71,40
Liver
sMEP-1*
90,100
100
Muscle
sMEP-2*
110
100,75
Liver
PEPA-1*
101
100,115
Eye
PEPB*
135
100
Eye
PGM-1*
null
lOO.null
Muscle
streams (Table 1) 3 different times to assess
repeatabilit)' and \'ariation of connts within and
among indi\'idnal readers. One reader counted
tlie 7 meristic features on 125 additional cut-
throat trout to determine mean counts for each
structure and allow comparison among the 12
sampling sites where > 5 fish were counted
(Table 1). The initial count from this reader's
original 50 fish was also included in the analy-
sis, leading to a sample of 175 cutthroat trout.
All counts were done on the right side of
each cutthroat trout. Scales in the lateral series
were counted 2 scale rows above the lateral
line starting at the opercle opening and contin-
uing to the insertion of the caudal fin, while
scales above the lateral line were counted from
the anterior of the dorsal fin on a vertical diag-
onal down to the lateral line. Vertebrae were
counted during dissection of the fish. Pyloric
caecae were enumerated by stretching the
stomach and counting caeca ends. Meristic fea-
tures were counted under a dissecting micro-
scope using 30X magnification and reflected
light. Readers practiced the protocol and com-
pared results to resolve procedinal differences
before initiation of counts. All fish were counted
at similar times by each reader with several
different cutthroat trout counted between sub-
sequent counts.
Three-way analysis of variance (ANOVA)
was used to assess differences in counts of
meristic features among (1) readers, (2) read-
ings by individual readers, and (3) sampling
sites. The sampling site effect was then con-
trolled for and a 2-way ANOVA was used.
One-way ANOVA was used to compare counts
among readers and sampling sites. Tukey's
multiple comparison test was used to make
pairwise comparisons if significant differences
were found. Statistical analyses were per-
formed using SPSS/PC+ (SPSS Inc. 1991).
Significance was determined at P < 0.05 for all
tests.
Results and Discussion
Cutthroat trout were present in all 23 study
streams. Electrophoretic anaKsis of fish from 7
streams found no genes at diagnostic loci that
identify niinliow tiout (Table 2). Because genetic
samples were collected from sites most likely
to contain rainbow trout alleles (e.g., streams
stocked with rainbow trout), we considered all
trout in the drainage to be pure cutthroat trout.
The AK-1*333 allele is common among fine-
spotted cutthroat tiout in the Snake Rivei* drain-
age and was detected in 4 of the 7 samples
(Table 1). This allele, while not unicjue to fine-
spotted cutthroat trout, is rare in Yellowstone
cutthroat trout populations outside the Snake
River drainage; its presence indicates possible
integration with finespotted cutthroat trout. An
ANOVA showed no consistent difference in
counts for any of the 7 meristic features between
fish from sites potentially hybridized with fine-
spotted cutthroat trout and those considered
pure Yellowstone cutthroat trout. Additionally,
Behnke (1992) stated that meristic counts of
finespotted and Yellowstone cutthroat trout are
indistinguishable, and there is considerable
debate as to whether finespotted cutthroat
trout are a formal subspecies. Therefore, we
did not differentiate between finespotted and
Yellowstone cutthroat trout in our analysis.
No significant differences among counts by
the same reader for any meristic feature were
obsei-ved. All 3 readers had high agreement
among multiple counts for each structine
(Tiible 3).
Significant differences in mean counts among
different readers were observed for all struc-
tures except gillrakers (Tables 4, 5). All 3 read-
ers had significantly different mean counts of
pyloric caecae, pelvic fin rays, and scales above
the lateral line, while at least 1 reader was sig-
nificantly different from the other 2 readers in
mean counts of vertebrae, basibranchial teeth,
and scales in the lateral series. Hubert and
Alexander (1995) also found poor agreement
304
Great Basin Naturalist
[Volume 56
Table 3. Significance values for differences in mean meristic counts among 3 readers (RDR), 3 readings by individual
readers (RUN), and sampling site (SITE).
Main effects
Interactions
Structiue
RDR
RUN
SITE
RDRxRUN
RDR X SITE
RUN X SITE
RDR X RUN X SITE
Pvloric caecae
0.000
0.903
0.000
1.000
0.000
1.000
1.000
Vertebrae
0.000
0.819
0.061
0.757
0.047
0.997
1.000
Pelvic
fin ravs
0.000
0.996
0.012
0.794
0.000
1.000
1.000
Gillrakers
0.765
0.356
0.244
0.352
0.045
0.098
0.051
Basibranchial
teeth
0.448
0.945
0.000
0.952
0.323
1.000
1.000
Scales in
lateral series
0.000
0.939
0.000
0.989
0.000
1.000
1.000
Scales above
lateral line
0.000
0.986
0.000
1.000
0.000
1.000
1.000
Table 4. Significance values for the difference in mean meristic counts among 3 readers (READER) and among 3
readings by individual readers (RUN) at 5 sampling sites.
effects
Structure
Site
READER
RUN
Interaction
0.998
1.000
0.993
0.808
0.S60
1.000
0.932
0.972
0.999
0.998
0.812
0.984
0.618
0.561
0.887
0.918
0.886
0.969
0.849
0.969
0.802
0.924
0.628
0.882
0.880
0.924
0.621
0.435
1 .000
1,000
1.000
1. 000
0.815
0.992
0.871
0.492
0.981
0.881
0.938
0.880
0.878
0.995
0.683
0.902
0.975
0.907
0.889
0.990
0.907
0.886
0.951
0.932
0.860
0.818
0.431
0.535
0.879
0.905
0.975
0.999
0.886
0.973
0.888
0.843
0.712
0.815
0.885
0.885
0.644
0.694
Pvloric caecae
Vertebrae
Pelvic fin ravs
Gillrakers
Basibranchial teeth
Scales in lateral series
Scales above lateral line
Anderson
Brovvn
SFWood
Venus
Wood
Anderson
Brown
SF Wood
Venus
Wood
Anderson
Brown
SF Wood
Venus
Wood
Anderson
Brown
SFWood
Venus
Wood
Anderson
Bro\vn
SFWood
Venus
Wood
Anderson
Brown
SFWood
Venus
Wood
Anderson
Brovvn
SF Wood
Venus
Wood
0.0S3
0.000
0.108
0.227
0.000
0.019
0.000
0.153
0.016
0.226
0.000
0.005
0.000
0.003
0.000
0.596
0.737
0.001
0.400
0.055
0.728
0.000
0.142
0.064
0.090
0.001
0.000
0,000
0.000
0.000
0.000
0.000
0,000
0.000
o.ooo
I99(ij
Variation in Yellowstone Cutthroat Trout
305
Table 5. Variation in mean meristic counts and standard deviations (in paren theses) of 3 readers. Means not signifi-
cantly different indicated In hold (Tiike\'s /'< 0.05).
Reader
Structure
1
2
3
r
Pyloric caecae
32.7 (6.3)
36,9 (9.5)
41.0 (11.7)
< 0.0001
Vertehrae
60.5 (L6)
59.5 (2.0)
59.3 (1.2)
< 0.0001
1\'!\ ic tin ra>s
9.0 (0.4)
8.8 (0.4)
9.4 (0.6)
< 0.0001
Gillrakers
18.9 (L6)
18.8 (1.3)
19.3 (10.8)
0.83
Basihranchial
teeth
13.7 (4.2)
15.3 (4.3)
14.2 (4.2)
0.003
Scales in later
il series
178.0 (14)
187.5 (14)
187.4 (13)
<0.0001
Scales ahoNC 1
iteral lini'
44 (4.2)
56.4 (5.2)
42.5 (3.6)
<0.0001
Table 6. Mean meristic counts and standard deviations (in parentheses) for 175 fish 1)\ 1 reader with ranges among
the 12 sample sites with >5 fish counted. A probability (P) of <0.05 indicates significant differences among sites.
Structure
Grand mean (s)
Range in means
among sites
Pyloric caecae
Vertebrae
Pelvic fin rays
Gillrakers
Basibranchia! teeth
Scales in lateral series
Scales ab()\e lateral line
42.29 (10.89)
58.57(1.39)
9.23 (0.86)
18.80 (2.08)
13.96 (5.45)
182.70 (14.77)
40..39(3.51)
29.9-51.4
57.9-60.6
9.0-9.9
17.8-19.9
11.4-21.8
175.5-207.3
37.1-45.5
< 0.0001
0.()()()2
0.0001
0.0018
0.0025
< 0.0001
0.0001
among readers when counting meristic fea-
tures of rainbow trout.
Significant differences were observed in
counts of meristic features among fish fi-om 12
streams (Tables 3, 6). Meristic features may be
environmentally controlled within specific
areas or drainages (Barlow 1961, Rinne 1985,
Currens et al. 1989), but environmental vari-
ables measured at each sampling site (eleva-
tion, gradient, and stream size) were not corre-
lated with meristic counts in the Greybull River
drainage (Kruse 1995).
Researchers have used meristic counts with
varied success to identify subspecies of cut-
throat trout (Loudenslager and Kitchen 1979,
Loudenslager and Gall 1980, Marnell et al.
1987). Recent research has shown that meristic
comparisons can provide potentially mislead-
ing information (Busack and Gall 1981, Leaiy
et al. 1984, 1985) because meristic characteris-
tics are often specific to localized populations
(Behnke 1992) and are strongly influenced by
genetic variation (Leaiy et al. 1991).
Behnke (1992) described typical meristic
counts for Yellowstone cutthroat trout and
rainbow trout (Table 7). Mean counts of meris-
tic features of cutthroat trout from the Grey-
bull River drainage (Tables 5, 6) were within
ranges for Yellowstone cutthroat trout (Table
7); however, mean counts of pyloric caecae,
vertebrae, and gillrakers were also within typi-
cal ranges for rainbow trout. Variation and sim-
ilarit)' in counts of meristic features of Yellow-
stone cutthroat trout and rainbow trout make it
difficult to determine species or hybrids using
meristic counts alone. Only the presence of
basihranchial teeth provided a distinction be-
tween the 2 species.
Variations among readers, and among sam-
pling sites in a small geographic area, along
with relatively wide ranges in counts for Yel-
lowstone cutthroat trout and rainbow trout,
make it difficult to differentiate these 2 species
with certainty using commonly assessed meris-
tic features (Tible 7). Furthermore, it is unlikely
that Yellowstone cutthroat trout X rainbow
trout hybrids can be identified due to the
extensive variation in counts.
306
Great Basin Naturalist
[Volume 56
Table 7. Ranges of meristie counts among species (YSC = Yellowstone cutthroat trout and RBT = rainbow trout),
readers, and sampling sites.
YSC
/'
RBT'
Variation
Vuiation among
Variable
Typical
Overall
Typical
Overall
among readers'^
sampling sites'^
Pyloric caecae
35-43
25-50
37-55
30-70
33-41 (36.9)
30-51
Vertebrae
61-62
60-63
62-64
61-66
59-61 (59.8)
58-61
Pelvic fin ravs
9
9-10
not
reported
9 (9.0)
9-10
Gillrakers
19-20
17-23
19-21
17-24
18-21 (19.0)
18-20
Basibranchial teeth
present
present
14-16(14.4)
11-22
Scales in lateral series
165-180
150-200
125-150
120-160
179-188 (184)
176-207
Scales above lateral line
45-50
40-55
30-32
26-35
42-57 (47.6)
37-46
•'From Behnke (1992)
''Ranges are from the 9 readings taken for each strncture with means in parentheses (3 readings by 3 readers).
'^Ranges are from means for the 12 sampHng sites that had > 5 cutthroat trout (> 150 mm total length) counted (Table 6).
Acknowledgments
We thank C. Ewers, C. Griffith, K. Harris,
K. Krueger, and M. Wilhams for field assis-
tance; E Rosebeny and R. Ziibik for technical
support; K. Krueger and D. Simpkins for serv-
ing as readers; and R. Behnke, B. Shepard, R.
Wiley, and S. Yekel for critical review of the
manuscript. This project was supported by the
Wyoming Game and Fish Department and the
U.S. Forest Sei"vice.
Literature Cited
Allendorf, K W., a.nd R. E Leary. 1988. Conservation
and distribution of genetic variation in a polytypic
species, the cutthroat trout. Consei"vation Biology 2:
170-184.
Allendorf, E W, and S. R. Phelps. 1980. Loss of genetic
variation in a hatcheiy stock of cutthroat trout. Trans-
actions of the American Fisheiy Society 109;
537-545.
AviSE, J. C. 1974. Systemic value of electrophoretic data.
Systemic Zoology 23: 465—181.
Ayala, E J., AND J. R. Powell. 1972. Allozymes as diagnos-
tic characters of sibling species of Drosophila. Pro-
ceedings of the National Academy of Sciences, USA
69: 1094-1096.
Barlow, G. W. 1961. Causes and significance of moipho-
logical variation in fishes. Systematic Zoology 10:
105-117.
Behnke, R. J. 1992. Native trout of western North Amer-
ica. American Fisheries Society Monograph 6,
Betliesda, MD.
Bu-SACK, C. A., AND G. A. E. Gall. 1981. Introgressive
hybridization in a population of Paiute cutthroat trout
{Salmo clarki .seleniris). Canadian Joimial of Fisheries
and Aquatic Sciences 38: 939-951.
Currens, K. R, C. S. Shahpe, R. Hjort, G. B. Schreck,
AND H. W Ll. 1989. Effects of different feeding
regimes on the morphometries of chinook salmon
(Oncorhynchtts fsliaici/tsclia) and rainbow trout (O.
iwikiss). Gopeia 1989: 689-695.
Gresswell, R. E. 1988. Status and management of inte-
rior stocks of cutthroat trout. American Fisheries
Society Symposium 4.
Hubert, W. A., and C. B. Ale.v\nder. 1995. Observer
variation in counts of meristie traits affects fluctuat-
ing asymmetry. North American Journal of Fisheries
Management 15: 156-158.
Kruse, C. G. 1995. Genetic purity, habitat, and population
characteristics of Yellowstone cutthroat trout in the
Greybull River drainage, Wyoming. Unpublished
master's thesis, University' of Wyoming, Laramie.
Leary, R. E, F W Allendorf, and K. L. Knudsen. 1985.
Developmental instability and high meristie coimts
in interspecific hybrids of salmonid fishes. Evolution
.39: 1318-1326.
. 1989. Genetic divergence among Yellowstone cut-
tliroat trout populations in the Yellowstone River drain-
age, Montana: update. Population Genetics Labora-
ton Report 89/2, Unixersity of Montana, Missoula.
. 1991. Effects of rearing density' on meristics and
developmental stability of rainbow trout. Copeia 1991:
44-49.
Leary, R. E, E W Allendorf, S. R. Phelps, and K. L.
Knudsen. 1984. Introgression between westslope
cutthroat trout and rainbow trout in the Clark Fork
River drainage, Montana. Proceedings of the Mon-
tana Academy of Sciences 43: 1-18.
. 1987. Genetic divergence and identification of
seven cutthroat trout subspecies and rainbow trout.
Transactions of the .\merican Fisheries Society 116:
580-587.
LouDENSL.\(,ER, E. J., AND G. A. E. Gall. 1980. Geogi-aphic
patterns of protein variation and subspeciation in
cutthroat trout. Sdliiw chirki. S\stemic Zoology 29:
27-12.
LouDENSLAGER, E. J., AND R. M. KiTCHEN. 1979. Genetic
similarity between two forms of cutthroat trout,
Salmo clarki, in Wyoming. Copeia 4: 673—678.
\L\HNELL, L. F, R. J. Behnke, and F W Allendorf.
1987. Genetic identification of cuttliroat trout, Salmo
clarki, in Glacier National Park, Montana. Canadian
Journal of Fisheries and A(}uatic Sciences 44:
1830-1839.
Nielsen, J. D. 1995. Exolulion and the aciuatic ecosystem.
American Fishi-ries Socii'tx Symposium 17.
1996] Variation in Yellowstone CuitiihoatTrolit 307
Perkins, D. L., C. C. Kiuecer, and li. May. 1993. Her- SPSS Inc. 199L The SPSS guide to data analysis for
itage brook trout in nortlieastem USA: genetic vari- SPSS/PC-(-. 2nd edition. SPSS Incorporated, Cliicago,
ability within and among popnhition.s. Transactions of IE.
the American Fisheries Society 122: 515-531.
RiNNE, J. N. 1985. Variation in Apaelie trout popuhitions in Received 10 April i.996
the White Mountains. Arizona. North American Accepted 2fi Aii-imt 1996
h)unial ol Fisheries Management 5: 146-158.
Great Basin Naturalist 56(4), © 1996, pp. 308-318
STUDIES OX XEARCTIC XEGASTRIUS
(COLEOPTEK\: EL.\TERIDAE)
Samuel A. Wells^
Abstract. — New species descriptions of Xegastrius rupicola from California. Oregon, \\ashington, and British
Columbia: .V. stibicki from California. Mont;ma, imd British Columbia: .V. solox from .\rizona and New Mexico: and .V.
atrosus from Ontario and Quebec are gi\'en. Negastrius colon is returned to species status, and a neot>pe is designated
for A^. choris. Fleutiaiixellus extricatits is a new combination. A ke\' is proNided to Nearctic species of Xegastrius.
Key words: Xegastrius. Elateridae. holotijpe. parafype. neotype.
History
Xegastrius was established in the famiK
Elateridae b\ Thomson (1859) to distinguish
those species of Cn/ptohypmis Eschscholtz
ha\ing arcuate prosteniiil sutures from species
with straight or double sutin-es. Candeze (1860)
did not use Thomson s assignments and placed
all Negastrius species in Cryptohypnus. Honi s
(1891) monograph of the species of Cryptolujp-
nus of Boreal America rejected die name Xegas-
trius and included all North American forms
into 9 groups within the genus Cryptohypnus.
Horns ehoris group included \. delumhis
(Horn), xV. choris (Say), .V. exiguus (Randall),
and A", ornatus (LeConte), which were equi\ a-
lent to Thomsons Xegastrius. Schwarz (1906)
included Xegastrius and Cryptohypnus with
the genus Hypnoidus Stephens in the tribe
H>pnoidini. Leng (1920) also placed iill species,
e.\cept .V. e.xiguus. in the genus CryptoJnjpnus.
Using mesostenial chai-acters, Xakane and
Kisliii (1956) made die distinction bet\veen the
subfamilies Xegastriinae and H>politliinae
(which die\ s\"non\iiiized under the Cteniceri-
nae). Aniett (1963) recognized onl\- the genera
Xegastrius and Oedostethus LeConte in the
Xegastiiinae from Xoitli .America. Stibick (1971)
recognized or established Xeohypdonus Stibick,
Migiwa Kisliii, OedostetJuis, FleutiauxeUus
Mequignon, Xegastrius. Zorochrus Thomson,
and Paradonus Stibick from Xorth America.
He restricted Xegastrius to diose species with
coarse pronota, single prosternal sutures, and
species with the 2nd and 3rd antenual seg-
ments equi\alent in length. Later, Kishii (1976)
erected Microlu/pnus. to which Stibick (1991)
assigned the single Xorth .American species of
M. striatidus (LeConte).
Discussion of Char.\cters
With the exception of die Cardiophorinae,
the subfamiK" Xegastriinae is distinguished
from odier subfamilies of Elateridae b\' liaxing
the nieso- and metastenia adjacent and sepa-
rating the mesocoxal ca\it> from the niese-
pimeron and mesepistenium. The Xegastri-
inae is distinguished from the Cardiophorinae
by possession of a pointed prosternal process,
which is shortened and tianicated in the Car-
diophorinae. Within the Xegastriinae, Xegas-
trius is apparentK most closeh" related to the
genus Microhypnus, bodi genera haxing a
strigate and/or rugose pronotum. Following
Stibick s (1971) presumed natural afrinities, die
sister group of Xegastrius could be mi) of die
Xorth .American genera, e.xcept Paradonus,
w liicli is more closeK related to the Old World
species of Thurana Stibick and Optitarynus
Stibick, both of w liich are without e.xtemall>'
\isible el\ tnil striae. Zivochrus is distinguished
from Xegastrius b\ die double prosternal sutm-es
and/or b\" the arcuateK" extended pronotum
that projects o\er die head. In addition, the
pronotum in Zorochrus is more coiU'seK gi-anu-
late on the anterior half Fh'utiau.xeUus differs
from the other genera of Xegastiiinae b\ lia\ -
ing tlie 3rd antennal segment nearly twice as
long as the 2nd. The genus Xeohypdonus is
'Biosvs Inc.. 10150 Old Columbia Rixid, Columbia. MD 2104(>-17O4.
308
1996]
Studies ox Nearctic Negastrics
309
separated from Xi'fiastrin.s b\- a smooth to
sliglitK punctate pronotum that is often shiny
(^^'ells 1991). Oedostethiis is chstinguished from
Xegastriiis h\' ha\in'pe. Fall's description of
N. extricatus is clear on several points. The api-
cal 3 segments of the antennae extend past the
hind angles of the thorax whereas N. nadezha-
dae has the antennae not attaining the hind
angles. Fall also refers to an impressed vertex
that is absent in all Negastrius species. Several
specimens from Cornell (taken from the type
locality of F. extricatus), the USNM, Chicago
Field Museum, and the Canadian National
Collection, all of which were taken from Alaska,
fit Falls description on all points and have the
2nd segment of the antennae reduced; this is
indicative of the genus Fleutiaiixellus.
Acknowledgments
I thank the following institutions and indi-
viduals for providing material: G. E. Ball, Strick-
land Museum, University of Alberta; E. C.
Becker, Canadian National Collection; R. W.
Brooks, Snow Entomological Museum, Univer-
sity of Kansas; S. D. Cannings, University of
British Columbia; J. A. Chemsack, University
of California, Berkeley; D. A. Kavanaugh, Cali-
fornia Academy of Science; J. K. Liebherr,
Cornell University; K. C. McGiffin, Illinois
Natural History Sui-vey; C. A. Olson, University
of Arizona; E. Riley, Texas A & M University;
C. Salvino, Chicago Field Museum; S. R. Shaw,
Museum of Comparative Zoolog)', Hai-vai'd Uni-
versity; J. N. L. Stibick, USDA,' APHIS; C. A.
Triplehom, Ohio State University; and the late
D. R. Whitehead, National Museum of Natural
Histoiy, Smithsonian Institution.
I also thank the Department of Zoology at
Brigham Young University and the D Elden
Beck family for financial support, and Richard
W. Baumann, C. Riley Nelson, and Michael F
Whiting for assistance in collecting specimens
throughout the western United States and
Canada. Special appreciation is extended to
J. N. L. Stibick, E. C. Becker, Paul Johnson, and
Stephen L. Wood for editorial suggestions.
Literature Cited
Arnett, R. H., Jr. 196.3. Fascicle 46, Elateridae in: The
beetles of the United States. Catholic Uni\ersit>' of
America Press, Washington, DC.
Becker, E. C. 1977. New and noteworthy records of
Coleoptera in Canada (1). Annales de la Societe
Entomologique dn Quebec 22:14-17.
Blatchley, W. S. 1910. Coleoptera or beetles known to
occur in Indiana. Indianapolis. 1.386 pp.
318
Great Basin Naturalist
[Volume 56
Brooks, A. R. 1960. Adult Elateridae of southern Alberta,
Saskatchewan and Manitoba (Coleoptera). Canadian
Entomologist, Supplement 20: 5-63.
Candeze, E. 1860. Monographie des Elaterides III. Liege.
512 pp.
Dietrich, H. 1945. The Elateridae of New York State.
Cornell University Agricultural Experiment Station
Memoir 269. 79 pp.
DOLIN, V G. 1971. New species of click beetles (Cole-
optera: Elateridae) from the Soviet Union. Entomo-
logical Re\'iew 50: 362-370.
Fall, H. C. 1926. A list of the Coleoptera taken in Alaska
and adjacent parts of the Yukon Territoiy in the sum-
mer of 1924. Pan-Pacific Entomologist 2: 127-154,
191-208.
Fattig, P W. 1951. The Elateridae or click beetles of Geor-
gia. Emeiy University Museum Bulletin 10. 25 pp.
Horn, G. H. 1891. A monograph of the species of Crypto-
hypnus of boreal America. Transactions of the Ameri-
can Entomological Society 18: 1-29.
KiSHll, T. 1976. New Nega.strius with some notes. Bulletin
of Heian High School 20:17-46.
Lane, M. C. 1971. Family Elateridae in: M. H. Hatch, The
beetles of the Pacific Northwest, part V. University of
Washington Press, Seatde. 662 pp.
LeConte, J. L. 1853. Revision of the Elateridae of the
United States. Transactions of the American Philo-
sophical Societ>' 10: 405-508.
Leng, G. W. 1920. Catalogue of the Coleoptera of America,
north of Mexico. John D. Sherman, Jr, New York.
470 pp.
Nakane and Ki.shii. 19.56. On the subfamilies of Elateri-
dae from Japan. Kont\'u 24: 201-206.
R\NDALL, J. W. 1838. Description of new species of cole-
opterous insects inhabiting the State of Maine. Boston
Journal of Natural Histoi-y 2: 1-52.
Say, T 1839. Descriptions of new North American insects,
and obseiAations on some alread\- described. Trans-
actions of the American Philosophical Societv 20:
155-190.
SCHWARZ, O. 1906. Elateridae, Genera Insectorum. Riscicle
46.
SCHENKLING, S. 1925. Ill: Junk: Goleopteromm Catalogus
81:212-215. W. Junk,' Berlin.
Stibick, J. N. L. 1971. The generic classification of the
Negastriinae (Coleoptera: Elateridae). Pacific Insects
13: 371-390.
. 1991. North American Negastriinae (Coleoptera,
Elateridae): the Negastriinae of the northeastern
United States and adjacent Canada (Coleoptera: Ela-
teridae). Insecta Mundi 4 (1-2): 1-;31.
Thomson, G. G. 1859. Coleoptera Scandinaviae 1.
Wells, S. A. 1991. Two new species of Neohypdonus
(Coleoptera: Elateridae) from North America widi a
key to Nearctic species. Entomological News 102(2):
73-78.
Received 15 November 1994
Accepted 4 October 1996
\
Gveal Basin Naturalist 56(4), © 1996, pp. 319-325
BIGHORN SHEEP RESPONSE TO EPHEMERAL
HABITAT FIUGMENTATION BY CATTLE
J. A. Bissonette' and M(>laiiic" J. Steinkampl'2
ABSThL\CT. — Wt' studied Sfasoiial tattk' mazing as an agent of eplieuierai habitat liagineiitation on a newly reintro-
duced popidation of California bighorn sheep (Ovia canadensis californiana) in Big Cottonwood Canyon, Idaho,
1988-89. W'e evaluated the In pothesis that bighorn sheep avoid cattle. We dociunented sheep response to the pro.ximity
to cattle by direct observation. The core areas used by bighorn and distances to escajje tenain generally decreased as
cattle moved closer to sheep. Likewise, sheep moved from cattle as cattle approached them. Severity of response we
obsened is in marked contrast with that reported for established bighorn populations, suggesting that newK' reintro-
duced bighorn sheep are more highly sensitive to the presence of cattle.
Krij words: hifjiorn slicep, cattle, disturbance, Idaho, Ovis canadensis.
Prior to the 20th century, CaHfornia big-
horn sheep were abundant in montane regions
of the western United States (Van Dyke et al.
1986). However, since 1840 population num-
bers of bighorn sheep and their area of distrib-
ution have decreased (Cowan 1940, Buechner
1960). Disease, excessive hunting, activities
associated with mining, human disturbance,
and pressure from hvestock for resources and
space reportedly contributed to the extiipation
of the subspecies from most of its range (Smith
1954, Geist 1971, Graham 1971, Demarchi and
Mitchell 1973, Demarchi 1975, Trefethan
1975, Van Dyke 1978, Smith et al. 1988).
California bighorn sheep were once abun-
dant in parts of southwestern Idaho; the last
observations were recorded during the 1920s
(Hanna 1978). The Idaho Department of Fish
and Game (IFG) initiated reintroduction pro-
grams of returning California bighorn to parts
of their historic range in 1963. Thirty-eight
sheep from the Chilcotin River herd in British
Columbia were transplanted into the drainages
of the East Fork of the Owyhee River between
1963 and 1966 and have provided a base for
subsequent reintroductions. In 1967, 12 addi-
tional bighorn were reintroduced into the
nearby Little Jack's Creek drainage. Both pop-
ulations were allowed to expand until 1980
(Toweill 1985). From 1980 to 1989, >100 sheep
were relocated to 5 different regions in south-
ern Idaho.
Livestock pressures have been heavy on
rangelands in the western United States that
historically supported populations of bighorn
sheep (Mackie 1978). Seventy percent of the
public land area in the 11 westernmost states is
grazed at least seasonally. Within Idaho range-
land conditions varied. In 1986 surveys from
the Owyhee range in Idaho reported 57% of
the range in poor condition, 35% fair, and only
5% in good condition (Bureau of Land Man-
agement, Owyhee rangeland program sum-
mary, Burley District, ID, files, 16 pp., 1986);
while in 1982, 30% of the range was in poor
condition, 57% fair, and 18% in good condition
(Bureau of Land Management, Twin Falls, land
use decisions summaiy and rangeland program
summary, Burley District, ID, files, 26 pp.,
1982). Peiper (1988) reported that improve-
ment in range condition has been slow since
1973.
Bighorn sheep are more sensitive to land
uses associated with development than most
native ungulates (Andryk and Irby 1986). Addi-
tionally, bighorn sheep are comparatively less
abundant, react adversely to disturbance, and
occupy habitats sensitive to change (Van Dyke
et al. 1986). Livestock activities on these sites
can negatively affect sheep through resource
exploitation (i.e., forage, space, cover, water) or
behaviorally (Geist 1971). On shared ranges
social intolerance may impose greater limita-
tions on distribution and habitat use of bighorn
'U.S. National Biological Service, Utah Cooperative Fish and Wildlife Research Unit, Department of iMslieries and Wilclliic. College of Natural Resources,
Utah State University, Logan, UT 84321-,5290.
^Present address: U.S. Fish and Wildlife Service, Box 2676, Vero Beach, FL .32961.
319
320
Great Basin Naturalist
[Volume 56
than competition for forage; however, biologists
disagree whether livestock impact bighorn
sheep spatial boundaries, limiting distribution.
Wilson (1975) and Van Dyke et al. (1986) re-
ported that bighorn show aversions to cattle
and avoid them when unaccustomed to their
presence on the range (Drewek 1970, Kornet
1978), while others did not detect reactions
between sheep and cattle (King 1985, King
and Workman 1985). Analyses that test the
avoidance of livestock by bighorn sheep are
limited.
Habitat fragmentation theoiy has applica-
tion to seasonal livestock grazing. Habitat frag-
mentation may be permanent (e.g., subdivision
constmction) or ephemeral, as in seasonal lixe-
stock grazing. Effects of pemianent fragmenta-
tion on habitat use have received increasing
attention in recent > ears; ho\\ e\er, less is under-
stood about effects of seasonal fragmentation.
We postulated that areas used b\' bighorn sheep
are fragmented during spring and summer by
cattle on grazing allotments. An area ma\' appear
large but, due to fragmentation, ha\'e a much
smaller useable area. If bighorn sheep a\'oid
livestock, the area available to them is reduced
temporarily as lixestock graze seasonally in
sheep habitat, resulting in sheep exclusion fiom
areas of potential use. A population may be in-
fluenced as sheep are restricted to smaller
patclies of habitat and effects of densit>' depen-
dence are felt. In our study we wanted to
detennine whether avoidance occurs, assess its
effect on habitat use by sheep, and consider
how a\'oidance, if it occurs, might influence
future decisions for reintroductions.
Study Area
We conducted the studv' in Big Cottonwood
Canyon 16 km northwest of Oakle>' (Cassia
Co.), Idalio. The canyon is approximately 18 km
long, with Cottonwood Creek flowing to the
northeast through the canyon bottom. Eleva-
tion of the canyon floor increases gradually
from 1400 to 2100 m. Average elexation gain
from the canyon floor to the mesa top is 365 m.
Canyon walls are steep and characterized by a
combination of cliffs, boulder slopes, grass, and
shrub slopes. Woody vegetation includes four-
\ving salt brush {Atn})Icx cancsccns), spin\' hop-
sage [Grayia spinusa), low sage [Aiieiuisia arhii-
sciila), horse brush [Tetradymia canescens),
rabbit brush (Chn/sothamniis nauseosiis). blue-
bunch wheatgrass {Agropyron spicatuin), and
juniper {Jiinipenis occidentaUs).
Big Cottonwood Canyon lies within the Saw-
tooth National Forest and contains a cattle graz-
ing allotment that is leased fi-om late Ma\' until
earh' October This grazing allotment consists
of 5 pastures managed on a reverse-rotation
basis and supports 400 cows with calves. Mesas
south of the canyon contain another allotment
of 3 pastures; this allotment is managed on a
deferred-rotational system with 100 cows with
calves. Permit dates for the Big Hollow allot-
ment are late May to late October.
Methods
Thirt\'-seven California bighorn sheep (19
with radio-collars, 18 with patteni-coded col-
lars) were released into Big Cottonw ood Canyon
b\' the Idaho Department of Fish and Game
during December 1986, December 1987, and
Noxember 1988. Collars marked with different
designs in pennanent ink allowed us to distin-
guish between non-transmittered indi\'iduals.
The population at the beginning of our 1st
summer field season (1988) was 23, 13 from
the 1st reintroduction in 1986 and 10 fi"om the
2nd in 1987. Fourteen additional sheep were
released in November 1988.
We recorded daiK- locations of bigliom sheep
by visual obsenation fi-om May to September
1988 and June to September 1989. Telemetiy
was used onl>' to aid in locating radio-collared
bigliom sheep. We conducted weekly visual sur-
veys to locate an)' uncollared sheep not close
to collared indi\iduals. Sheep were viewed
fi-om > 500 m using a spotting scope to reduce
chance of detection and disturbance. If we
were detected and sheep moxement followed,
we disregarded subsecjuent obsenations of
those indi\ iduals for die remainder of the day.
Every effort was made to identify individuals
within groups. \\'e determined indi\iduals by
collar design or by telemetn' frequency. Loca-
tions were recorded in Universal Transverse
Mercator (UTM) coordinates. For each location
we recorded group size and composition.
We defined escape terrain as broken habitat
on which mountain sheep max* safel)' outma-
neuver or outdistance predators (Gionfriddo
and Krausman 1985). SpecificalK; escape ter-
rain ma\ be characterized b\- a ruggedness
index as defined by Beasom et al. (1983), and
terrain class and number of cliff faces > 120%
1996]
Sheep Response to Fragmentation
321
following Krausnian and Leopold (1986). For
every location we nieasnred distance to escape
terrain using a range Finder once sheep left the
area. We determined slope with a clinometer.
We located cattle by hiking a systematic route
on foot 3-4 times/wk. With the exception of
group composition, data lecorded for each cat-
tle location were identical to sheep locations.
We recorded cattle and sheep locations simul-
taneously iillowing sheep movements to he anal-
yzed in response to cattle mo\'ement for that
specific time. Data not taken during identical
time periods were not used in paired analyses.
Even though a controlled test was not possi-
ble, we wanted to obsene tlie response of sheep
when livestock were in proximity to sheep. On
14 August 1989, 5 cows were moved directly
into an area of continuous sheep use and held
continuousK' for 40 h. Cattle were kept within
approximately a 0.8-km- area by 2 cowboys.
Sheep response was observed and recorded.
Cattle were watered every 5 h by removing
them from the group one at a time and taking
them to a trough in the bordering pasture.
After 40 h all cattle were removed. We located
sheep daily for the next 10 d.
We combined individual bighorn sheep loca-
tions for each group for analysis with Program
Home Range (Samuel et al. 1985); thus, each
location represented a group of bighorn sheep,
not an individual. We used 95% haniionic mean
measures of activity to estimate home ranges
and core areas. We defined core areas as the
maximum area where the obsei-ved utilization
distribution as detemiined from the harmonic
mean values was greater than a uniform uti-
lization distribution (Samuel et al. 1985). Kol-
mogorov's test was used to determine if ob-
sei-ved use was significantly (P < 0.05) greater
than expected. All comparisons were consid-
ered significant at the 0.05 level. All data points
were plotted at a scale of 1:12,000.
We recognize that harmonic mean measures
have been criticized. Naef-Daenzer (1993) tested
the spatial resolution of the conventional har-
monic mean measure and a bivariate normal
kernel estimator with a new kernel estimator
he developed. The harmonic mean estimator
generalized the distributions of 2 parallel gra-
dients and estimated density at higher than
zero for areas containing no sample points.
Worton (1989, 1995) and Boulanger and White
(1990) have outlined some undesirable proper-
ties of harmonic mean measures that were
eliminated from kernel estimators using appro-
priate smoothing techni(iues. Specifically with
the harmonic measure, estimates of zero area
can occur, and isopleths may include areas
witli no sample points (Worton 1995). We had
no estimates of home range or core areas that
approached or even came close to zero. Addi-
tionally, the isopleths we generated were based
on tightly grouped locations of sheep, thus
avoiding the problem ol' areas with no sample
points. Finally, we did not employ interstudy
comparisons, thus avoiding the onerous prob-
lem of comparing between methods, thereby
reducing the effect of inherent bias.
We plotted mean monthly home ranges and
core areas of sheep and cattle and then over-
laid them to determine changes in size and
location between consecutive months. We
measured avoidance by quantifying changes in
size and location of bighorn sheep range and
core areas as cattle moved through bighorn
sheep habitat. Changes in location were deter-
mined from harmonic means. We compared
data collected during the 1st and 2nd field sea-
sons to determine whether range and core
areas were related to seasonal changes.
We calculated daily distances between big-
horn sheep and cattle using UTM location co-
ordinates. We defined consecutive locations as
locations taken 1 d apart. Only cattle and big-
horn sheep paired locations recorded at the
same time were analyzed. Simple linear regres-
sions were used to test for associations between
3 variables: distance (m) between cattle and big-
horn, distance sheep moved in response, and
distance fiom location of sheep to escape ter-
rain. First, we tested sheep response to prox-
imity of cattle; then we tested to determine
whether distance between sheep and escape
terrain was related to proximity of cattle.
Results
Response of Bighorn Sheep to Cattle
Sheep range size did not change signifi-
cantly in size or location (P < 0.05) from June
to July in 1988 or 1989. Cattle were in adjacent
pastures but because of topography were usu-
ally not visible to sheep or the observers. Dur-
ing August 1988, when cattle were moved to
an allotment adjacent to areas receiving high
sheep use, home range position shifted and
range size decreased (Table 1). In September
sheep expanded their range, coincident with
322
Great Basin Naturalist
[Volume 56
Table 1
Sp
atial responses
of bighorn sheep
in Little Cottonwood Canyon
Idaho, to the
pro.ximity of
cattle.
Range
Core area
Mean
chstanee (in)
size
Size
% use
% area^
c-s''
e-t^-
Sheep''
Date
(kni2)
(kni2)
6/88
13.4
4.3
61.4
32.1
4019
101
1616
7/88
13.7
4.7
53.9
27.3
4045
86
1246
8/88<^
5.0
1.5
59.0
42.9
2251
55
1046
6/89
13.4
4.7
57.5
40.0
4820
112
1698
7/89
13.5
1.5
67.0
40.0
5148
63
1008
8/89^
7.2
1.5
0.5'
55.6
40.0
3346
56
11
1276
■'Percent
>l total Ik
nie range ai
■a that c
ore an
a encompasses
''Mean daiK
distance sheep nii
>\ ed dun
ig the 1
lontli
'Wlean distance between cattle
and big
loni
'"Cattle plac
=d in allotments ck
«e to she
ep
'-'Mean distance of sheep to escape terrain
fpield experiment data
the movement of cattle during late August into
a pasture adjacent to a high use sheep area.
Sheep tended to concentrate into smaller core
areas in 1988 and 1989 as proximit)' to cattle
decreased.
No significant change (<3%) in core area of
bighorn sheep occurred between July and
August 1989 prior to moving cattle close to big-
horn. When cattle were moved puiposefully to
within 800 m, bighorn sheep responded by
immediately vacating the area and creating a
new distinct core area. Distances moved by
bighorn sheep directly after movement of cat-
tle into the sheep core area were 355% greater
than daily sheep movements during early August
(3000 vs. 845 m, respectively). Sheep remained
together and stayed within 35 m of escape ter-
rain for the following 9 d. This was the longest
time period during the study that sheep re-
mained within 35 m of escape tenain. Distances
between cattle and bighorn sheep remained
>4000 m for the following 5 d.
Response of Bighoni Sheep
Relative to Escape Terrain
As mean daily distance between cattle and
sheep decreased, the mean distance between
sheep and escape terrain tended to decrease.
Core-area size appeared to be directly related
(adjusted r^ = 0.81) to distance to escape ter-
rain (Fig. 1); the closer to escape terrain, the
tighter sheep grouped together A correlation
matrix, generated from these spatial data, adds
further corroboration for the association (Table
2). The mean daily distance that bighorn
moved dining the month was positively corre-
lated (r^ = 0.88) with increasing distance of
sheep to escape terrain.
Discussion
Hicks and Elder (1979) suggested that big-
horn sheep were more likely to move greater
distances when cattle were close, but were less
likely to relocate when cattle were distant. Our
data show increased movement by bighorn
sheep as cattle mo\'ed closer When we moved
cattle to within 800 m, bighorn left the area.
Sheep response to cattle was much more ex-
treme than at any other time or when com-
pared to their behavior when confronted by
humans at other times during the field season.
We were unable to differentiate between the
effect that cattle had and the potential effect of
the personnel involved. We do not doubt that
personnel moving the cattle had an effect. Fur-
thermore, the presence of both cattle and per-
sonnel close to sheep may well have augmented
bighoni response nonlinearly However, at other
times when we accidentally alerted sheep dur-
ing the study [n = 10), bighorn responded by
relocating much shorter distances (between
872 and 1190 m). Additionally, their response
was t>q3ically short-lived and they left the prox-
imity of escape cover by the next day or sooner.
Although lx)th the proximity of cattle and per-
sonnel influenced bighoni response, the impor-
tant point is that extreme proximity evoked a
higliK charged response. E\en without our
intentional moxement of cattle toward sheep,
their increasing affinity for escape cover as cat-
tle moved closer suggests strongly that live-
stock were perceived as a threat.
Escape terrain is an important component
of good sheep habitat (McQui\ e> 1978, Leslie
and Douglas 1979, Weyhausen 1980, Krausman
and Leopold 1986). We would lia\e predicted
1996]
Sheep Response to Fragmentation
323
0 50 100 150
DISTANCE TO ESCAPE TERRAIN (m)
Fig. 1. Relationship between size of core area of
l)ishorn sheep, Cottonwood Canyon, Idaho, and distance
to escape terrain, 1988-89.
that tighter grouping should result as sheep
moved farther from escape cover. However,
our data show the direct opposite result, sug-
gesting tliat when sheep move farther from
escape terrain, they do so under less threaten-
ing situations. Selective pressures under these
conditions appear not to result in tighter groups.
The response of bighorn sheep to cattle we
obsened is in contrast with bighorn sheep in
national parks. In some parks sheep approached
humans closely and were photographed from
car windows (Van Dyke et al. 1986). Smidi (1954)
reported sheep eating from his hand, whereas
others reported that sheep unaccustomed to
people or cattle fled at the sight of humans or
\ chicles >1600 m (Van Dyke et al. 1985). It
appears that newly reintroduced sheep are more
sensitive to disturbance, perhaps resulting from
recent transplant activities, and react differ-
endy dian do established, undisturbed j-jopula-
tious. Sheep reintrockiced into Big (Jottonwood
Canyon were net-gunned from helicopters,
blindfolded, and flown to a base. They then
had blood drawn, were given inoculations,
weighed, measured, placed into the l)ack of a
covered pickup with several conspecifics, and
then transported approximately 160 kni and kept
overnight in the vehicles. All were released the
following moniing into an area foreign to them.
As a result of exposure to such activities, any
disturbance may more likely be viewed as a
threat. In the Big Cottonwood Canyon popula-
tion, alert-alarm behavior appears to be rein-
forced yearly with each new group of reintro-
duced animals. Age may also play a part; 55%
of individuals released were <2 years of age.
Heightened sensitivity and subsequent fre-
quent reinforcement of alert behaviors appear
to characterize the population and may be a
general phenomenon for newly reintroduced
populations placed into new areas. Sensitivity of
these populations to disturbance may diminish
over time as populations become estal^lished.
Avoidance has implications for reintroduc-
tions of bighorn sheep. The total area of poten-
tial habitat may not be used by sheep if live-
stock are present. If cattle allotments remain in
use, it would appear wise to consider the possi-
bility of ephemeral fragmentation by cattle
when goals for desired bighorn population sizes
are developed. Goals should be consistent with
total useable habitat. Control of disturbance for
recently reintroduced populations of bighorn
sheep is certainly appropriate.
Table 2. Correlation matrix for home rantie, core area, and mean distance variables for bighorn sheep in Bi
wood Canyon, Idalio, 1988-89.
Cotton-
Range size
Core area
Mean
distance (m)
Size
% use
% area-*
c-sl>
e-f^-
Sheep''
Range size
1.0
0.694
0.234
-0.601
0.887
0.721
0.440
Size
1.0
-0.380
-0.704
0.38.5
0.916
0.765
% use
1.0
0.335
0.410
-0.144
-0.272
%area
1.0
-0.220
-0.458
-0.266
c-s
1.0
0.520
0.308
e-t
I.O
0.887
Sheep
1.0
^Percent of total home range area that core area enconipassi
"Mean distance between cattle and bighorn
■^Mean distance of sheep to escape terrain
324
Great Basin Natir.\i,ist
[\'olunie 56
Acknowledgments
We thank J. J. Beecham, \\; L. Bodie. T. C.
Edwards, D. G. Oman, H. G. Hudak, E R.
Krausman, R. B. Smith, D. E. Toweill, and E J.
Urness (deceased) for tlieir ad\ice and help
during the project. We also thank the Harold
Cranneys who allowed us to use dieir land and
utilities. We extend a special posthimius thanks
to D. Balph for his insight and wisdom. We iilso
thank the editor and associate editor of GBN
and an anon>nious reviewer for constructive
comments that helped us impro\'e the manu-
script. This stud\ was fimded b> the Idaho De-
partment of Fish imd Game dirough the Utah
Cooperatixe Fisheries and Wildlife Research
Unit (NBS) at Utah State Uni\ersit>-. We thank
the United States Forest Senice for pro\iding
aerial photographs and topographical maps.
The research proposal was e\aliiated b\" the
Animal Care Committee at Utah State for ad-
herence to established animal cai-e guidelines.
Data were collected follow ing acceptable field
methodolog) established b> the American
Society' of Mammalogists (1987). The United
States National Biological Senice, tlie Utah
Di\ision of \\ildlife Resources, the Wildlife
Management Institute, and Utah State Uni\er-
sit>- jointK support the Utiili Cooperati\ e Fish
and Wildlife Research Unit and make our
research possible. We tliank them.
Literature Cited
American Society of Mammalogists. 1987. Acceptable
field inetliods in maniiniilog\ : preliniiniin guidelines
approxed by tlie .\merican SocieU' of Mammalogists.
Jouniid of Maniinalog)' 65 (4. supplement). 18 pp.
Andryk, T. a., and L. R. Irby. 1986. Eviiluation of a moun-
tain sheep transplant in noitli-cenbiJ Mont;uia. Jour-
nal of Enxironmental Management 24: 337-346.
Be.\som, S. L., E. P Wiggers, .\nd J. R. Giardino. 1983. A
technique for assessing land surface ruggedness.
Journal of ^^'ildlife Miuiagement 47: 1163-1166.
BouL.\NCER, J. G., AND G. C. WHITE. 1990. A comparison
of home nmge estimators using Monte Carlo sinuila-
tion. Journal of Wildlife Management 54: 31t)-315.
Blechner, H. K. 1960. The bighorn sheep in tlie United
States, its past, present, and future. Wildlife Mono-
graph 4. 174 pp.
Cowan, I. M. 1940. Distribution and \ariation in the
native sheep ot North .\merica. American Midland
Natiuiilist 24: 505-580.
Demarc:hi, D. .\. 1975. Report and recommendations of
the workshop on California bighorn sheep. Pages
143-163 /'(I J. B. Trefctlien, editor. The wild sheep in
modem Xordi .-Vmerica. Winchester Press, New York.
Dem.vrcih, D. A., .\ND H. B. Mirhkll. 1973. The Chil-
cotin Ri\er bighorn population. C'anadian Field-Nat-
uralist 87: 433-454.
Dkewek, J., Jr. 1970. Population characteristics and be-
havior of inhoduced bighoni sheep in Owyhee Coimt>,
Idalio. Unpublished master's thesis, Universit\ of
Idalio, Moscow. 46 pp.
Geist, \'. 1971. Mountain sheep: a study in behavior and
evolution. Uni\ersit>- of Chicago Press, Chicago, IL.
383 pp.
GlONFRlDDO, J. P, AND P R. Kralsman. 1985. Summer
habitat use h\ mountain sheep. Journal of Wildlife
Management 50: 331-336.
GlullAM, H. 1971. Environmental analysis procedures for
bighoni in die S;ui Gabriel Mountiiins. Desert Bigliom
Council Transactions: 15: 38 — 15.
Hanna, P 1978. OwAhee connection. Idaho Wildlife 1(2):
7-9.
tllCKS, L. L., AND J. M. Elder. 1979. Human disturbance
of Sierra Nevada biglioni sheep. Journal of Wildlife
Management 43: 909-915.
Klng, M. M. 1985. Behavioral response of desert bighoni
sheep to himian harassment: a comparison of dis-
turbed and undisturbed populations. Unpublished
doctoral dissertation. Utah State Universitv. Logan.
137 pp.
Klng, M. M.. and G. W. Worknlxn. 1985. Response of
desert bighoni sheep to liimiim harassment: niiuiage-
nient implications. Transactions of the 51st North
American \\'ildlife and Natural Resources Confer-
ence 51: 7-1-85.
Kornet, C. A. 1978. Status and habitat use of C;ilifornia
bighoni sheep on Hart Mountain, Oregon. Unpub-
lished masters tliesis, Oregon State Universit\; Cor-
V iillis. 49 pp.
Kr\lsnl\n, P R.. and B. D. Leopold. 1986. Habitat com-
ponents for desert bighoni in die Harcjualiala Moun-
tains, Arizona. Journal of Wildlife Management 50:
504-508.
Leslie, D. M.. Jr., .\nd C. L. Doiglxs. 1979. Desert big-
honi sheep of the River Mountains. Nevada, ^^ildlife
Monogiaphs 66: 1-56.
McQuiVEV, R. P. 1978. The bighorn sheep of Nevada.
Nevada Department of Fish and Game, Biological
Bulletin 6. 81 pp.
M.VCKIE, R. J. 1978. Impacts of livestock giiizing on wild un-
gulates. Transactions of the Nordi .\nierican Wildlife
and Natural Resources Conference 43: 462—176.
Naef-Daenzer, B. 1993. .\ new transmitter for sniidl ani-
mals ;md enhanced methods of home-riuige aiiidv sis.
Jouniid of Wildlife Nhuiagement 57: 680-689.
Peiper, R. D. 1988. Grazing svstems and management.
Pages 9-24 in B. .\. Buchanan, editor, Rangelands.
University' of New Mexico Press, ,\lbuquerque.
Samuel, M. D., D. J. Pierce, .\nd E. O. Carton. 1985.
Identifying areas of concentrated use within the
home nmge. Jouniiil of AniuKd Ecologv 54: 711-719.
Smith, D. R. 1954. The bighorn sheep in kkilio: its status,
lite historv and management. Idaho Department of
Fish and CJanie. Wildlife Bulletin 1. 154 pp.
Smith, N., P R. Kiulsnlvn, and R. E. Kirby. 1988. Desert
bighoni sheep: a guide tt) selected management prac-
tices. U.S. Department oi Interior. Biological Report.
88. 35 pp.
Tow EILL, D. 1985. The Calitbrnia bigiiorn sheep in Idaho.
Pages 45-.56 in Proceedings of the CiJifoniia Bighorn
Workshop, Reno. N\'.
1996]
SiiKKp Kksi'onsk k) Fragmentation
325
I'RKFETHEN, J. B., !• DITOK. 1975. The wild sIk'C'p in North
America. Wincliester Press, NY. 302 pp.
\ \\ DvKi;, W. A. 1978. Population characteristics and
habitat utihzation of hi^horn sheep, Steens Moun-
tain, Oregon. Unpuhhslied master's thesis, Orej^on
State University, Corxalh's. 87 pp.
\ AN DvKE, W. A., A. Sands, J. Yoakum, A. Polenz, and J.
Bl^MSDELL. 1986. Wiklhle habitats in manaj^ed range-
land.s — the Great Basin oi southeast Oreuon — bij^honi
sheep. 2nd edition. Cleneral 'leclinical ]^e]K)rt PNVV-
159. 37 pp.
Weyhausen, J. D. 1980. Siena Nevada bitihoru slieep: his-
tory and population ecology. Unpublislied doctoral
dissertation. University of iMichigan, Ann Arbor 240
pp.
WIL.SON, L. O. 1975. Report and recommendations of the
Desert and Mexican bighorn sheep workshop. Pages
110-13! in J.B. Trefethan, edit(jr. The wild sheej) in
modern North America. Winchester Press, N.Y.
WoRTON, B. J. 1989. Kernel methods lor estimating the uti-
lization distribution in home-range studies. Kcology
70: 164-168.
• 1995. Using Monte Garlo simulation to evaluate
kernel-based home range estimators. Journal ol Wild-
lile Management 59: 794-800.
Received 23 October 1995
Accepted 3 April 1996
Great Basin Naturalist 56(4), © 1996, pp. 326-332
A FIELD STUDY OF THE NESTING ECOLOGY OF THE
THATCHING ANT FORMICA OBSCURIPES FOREL,
AT HIGH ALTITUDE IN COLORADO
John R. Conway^
Abstract. — ^A field study of the thatching ant, Fonnicu obsciiripes Forel, at 2560 m elevation in Colorado provided
information on mound density, composition, dimensions, and temperatures; worker longevity; and mite parasitization.
Density was 115 mounds/lia. Mounds had 1-52 entrances and Peromysciis fecal pellets in the thatch. Mounds conserved
heat and exhibited thermal stratification. Excavations of 4 nests revealed depths of 0.3 m to almost 1 m, novel myrme-
cophiles, and 0-198 wingless queens per nest. Marking experiments demonstrated that some workers overwinter and
live more than a year
Key words: Formica obscuripes, thatching ant, Colorado, ant mounds, mijrnwcophiles.
Formica obscuripes Forel is in the Formica
nifa-group (Weber 1935) and ranges from Indi-
ana and Michigan westward across the United
States and southern Canada. It is one of the
most abundant ants in western North America,
especially in semiarid sagebrush areas (Gregg
1963), and has been found at altitudes up to
3194 m (Wheeler and Wheeler 1986).
The objective of this field study was to com-
pare mound density, formation, composition,
dimensions, and temperatures, worker longevit>'
and parasitization, nest depths, mynnecophiles,
and the number of wingless queens per colony
of this species at high altitude in Colorado with
findings from lower altitude studies in Colo-
rado (Jones 1929, Gregg 1963, Windsor 1964),
Idaho (Cole 1932), Iowa (King and Sallee 1953,
1956), Michigan (Talbot 1972), Nevada (Clark
and Comanor 1972, Wheeler and Wheeler
1986), North Dakota (McCook 1884, Weber
1935), Oregon (Mclver and Loomis 1993,
Mclver and Steen 1994), Washington (Hender-
son and Akre 1986), and Canada (Bradley 1972,
1973a, 1973b). Although this species seems to
be most common at altitudes of 1524-2743 m
in tlie mountiiinous states (Gregg 1963, Wheeler
and Wheeler 1986), the highest previous study
site was at an elevation of 1550 m (Clark and
Comanor 1972). It is hypothesized that cli-
matic and vegetational changes associated with
higher altitude may alter the nest ecology of
this species.
Materials and Methods
The study site is in Gunnison County north
of Blue Mesa Resei^voir and west of Soap
Creek Co. Rd. in western Colorado at an alti-
tude of 2560 m. Field obsei^vations were con-
ducted 5-6 August 1990, 20 June -11 October
1992, 28 June-16 August 1993, 29 June-31
July and 14-16 August 1994, and 3, 29-31 July
and 15-16 August 1995. The area, dominated
by big sagebmsh {Arfemcsia tridcntata Nuttall)
and to a lesser extent by rubber rabbitbrush
{Clirysothammis miiiseosiis [Pallas] Britton), is
adjacent to a grove of quaking aspens {Popiihis
tremuloides Michaux).
The locations of 85 mounds were mapped
in a study area (64.6 m x 114 m) using a sur-
veyor's transect and compass in JuK' 1993 to
determine densit).
The diameters and heights of 97 mounds in
the stud\' area and sunounding area were mea-
sured. The number of entrances per mound
was determined by inserting sprinkler flags
into the active openings on each mound.
Mound temperatures were measured with a
Model 100-A VWR digital thenuometer probe.
Sixty-seven temperatiue measurements were
made on 34 mounds in the evenings (1915-2045
h) 2-14 July 1993 by inserting the probe
appro.ximately 15 cm into tlie top of each mound.
The temperatures of 4 of these mounds were
also recorded in the afternoon (1538-1600 h)
Inipartiiii-iit of Hiolug), Liiivt-rsily ofScruiitoii, Scnintoii, P.\ 1S.t1(I.
326
1996]
Nkstinc; Ecolcx;y of Thatching Ant
327
on 2 July 1993. In addition, hourly temptM-a-
tnres were reeorded at 4 loeations (air, liround,
mound top, and mound base) lor 3 diflerent-
si/ed mounds in July 1994 between 0700 and
ilOOO h tt) determine how moimd size aifeets
thermal d) namies. Temperatures were taken at
a mid-sized mound (height = 25.4 cm, average
diametei- = 1 m) on 16-18 July, at a large mound
(height = 49.5 cm, a\'erage diameter =1.21 m)
on 18 JuK; and at a small mound (height =
27.9 cm, average diameter = 0.51 m) on 17
July. The small and mid-sized mounds were
about 4.6 m apart and about 34-37 m from the
large mound. The probe was inserted approxi-
mately 15 cm into the top, base, and ground
adjacent to each mound. Temperatures were
also recorded in the shade about 15 cm above
the ground near each mound.
Hundreds of workers were marked on 8
nioimds and 5 plants in 1992-93 by applying
model aiiplane paint with a fine-tipped brush
and by spraying 5 mounds in 1994 with colored
acnlic enamel. Although many workers were
incapacitated or killed, especially by spraying,
most sun'i\ ed. Spraying was the most efficient
technique for marking large numbers of ants.
Four nests were excavated, 1 each on 6 August
1990, 27-28 June 1992, 12 July 1993, and 11-25
July 1994. The 1993 nest was poisoned with
IV2 cups Hi-Yield ant killer granules (Diazi-
non) wetted down with about 7.6 L of water
prior to excavation to investigate another tech-
nique for collecting queens and mynnecophiles.
Results
Nest Density
The extrapolated density for the 85 mounds
mapped in the 7364-m2 area was 115 mounds/
ha. The closest mounds were 2.36 m apart.
Mounds
Formation and composition. — Mounds are
composed of thatch and are usually dome
sliaped. Some moimds are exposed while oth-
crs are overgrown or shaded by low vegeta-
tion. Dead sagebrush protruded from or was
found on 63 of 98 mounds (64%). The largest
mound was built aroimd the base of a fence
post. No mounds were found inside the aspen
grove, but 2 were built around small aspen
trees on the forest edge.
Mound thatch consisted mainly of twigs but
also contained fecal pellets, probably from the
deer mouse {Pcwmijscus maniculatus [Wagner])
or vole {Micwlm sp.). Thatch (n = 58) from 1
mound consisted mainly of small twigs 4-89
nnn (mean = 24.19 nnn) long and 1-5 mm
(mean = 2.19 nmi) in diameter Workers were
observed carrying fecal pellets into or out of
mound entrances, but not on trails.
Dimensions and entrances. — The diame-
ters of 97 mounds ranged from 19 cm to 142 cm
(mean = 65 cm). Mound heights ranged from 6
cm to 58 cm (mean = 26 cm).
The number of entrances to 97 mounds
ranged from 1 to 52 per mound (mean = 12),
but their number, size, position, and activity
changed over time. For example, 1 mound had
10 or more entrances in August but only 2 in
October. Some entrances were larger than oth-
ers, and some surrounded plant stalks growing
out of mounds.
Temperatures. — Measurements of mound-
top and air temperatures in July 1993 demon-
strated that moimds are warmer than air tem-
peratures and that the differential is greater in
the evening than in the afternoon. Evening
temperatures {n = 67) for 34 mounds were
1.0°-15.5°C (mean = 8.6 °C) warmer than cor-
responding air temperatures. Afternoon tem-
peratures for 4 of these mounds were slightly
warmer (0.5°-0.9°C; mean = 0.7 °C) than cor-
responding air temperatures.
Hourly mound-top and mound-base tem-
peratures recorded in July 1994 were almost
always higher than ground temperatures, and
top temperatures were warmer than air tem-
peratures (Figs. 1-3). Differences in top and
air temperatures were greater in the evening
(1900-2000 h) for a large nest (8.9°-irC) and
mid-sized nest (6.8°-14.4°C) than their after-
noon (1500-1600 h) differences, 2.6°-6.3°C and
0.6°-8°C, respectively. On die otlier hand, hour-
ly top and air temperatures did not differ much
for the small nest in die evening (1.1°-3.3°C)
and in the afternoon (1.6°-2.2°C).
Average hourly top and base temperatures
were higher than average air temperatures for
die mid-sized and large mounds (Figs. 1-3). For
example, average top and base temperatures
were 6.2 °C and 3.1 °C higher than average air
temperatures for the large mound and 4.6 °C
and 0.5 °C higher for the mid-sized mound.
However, for the small mound the average top
temperature was actually 0.8 °C lower, whereas
the average base temperature was 2.7 °C higher
than the average air temperature.
328
Great Basin Naturalist
[Volume 56
MOUND #3 JULY 16-18, 1994
MOUND #14 JULY 18, 1994
35.0
35.0 ^
Time of Day
-•- Mound Top "♦■■ Mound Base
-•♦-• Ground -O- Air
Fig. 1. Average mound-top, mound-base, ground, and
air temperatures around a mid-sized Formica ohscuripes
mound from 0700 to 2000 h on 16-18 Jul\ 1994 at 2560 m
in Colorado.
10 11 12 13 14 15 16 17 18 19 20
Time of Day
-•- Mound Top -*^' Mound Base
-♦-■ Ground -O- Air
Fig. 2. Mound-top, mound-base, ground, and air tem-
peratures around a large Formica obsciiripes mound from
0800 to 2000 h on 18 Jul\ 1994 at 2560 m in Colorado.
Hourly top and base temperatures showed
thenual stratification. Average top temperatures
were 3.2 °C and 4.1 °C higher than average
base temperatures for the large and mid-sized
mounds, respectively. However, for the small
mound the stratification was reversed: average
top temperature was 3.5 °C lower than the
average base temperature.
The poor thermal regulation of smaller
mounds was also reflected by a greater fluctua-
tion of hourly top and base temperatures. Daily
ranges of top/base temperatures were 7.6/8.7 °C,
13.3/15.9 °C, and 13.8/26.3 °C for the large, mid-
sized, and small nests, respectively. Thus, larger
mounds exhibited less daily temperature fluc-
tuation than smaller mounds.
Worker Longevity
Most marking experiments {n = 14) indicated
that some workers live 19 to 44 d (mean = 31.6
d). However, 2 workers marked on a mound
between 7-9 July and 15-27 JuK' 1994, respec-
tively, were observed on 30 July 1995 on
another mound and on the original mound.
Thus, some workers overwinter and live more
than 1 yr
Mites
Mite infestation was not common. Orange,
spherical mites were noted on only 1 worker at
3 of the many mounds observed. The largest
number of mites obsened was 4-5 on the tho-
rax and gaster oi 1 worker.
Excavated Nests
Each oi the 4 nests excavated contained
numerous workers, larvae, and pupae, but the
nimiber of wingless queens per nest varied
greatK : 0, 1, 32, and 198. No winged reproduc-
ti\ es were found except a male in 1 nest. The
1996]
Nesting Ecology ok Tiiai-giiing Ant
329
MOUND #98 JULY 17, 1994
ers); and Lepidoptera
Table 1).
'E Noctuidae — lan'ae;
45.0 T
40.0 ..
Q 35.0
e
n 30.0
t
i 25.0
g
r 20.0
a
15.0
e
10.0
5.0
0.0
H 1 1 1 1 1 1 1 1 I ' '
7 8 9 10 11 12 13 14 15 16 17 18 19 20
Time of Day
*-• Mound Base
■O- Air
»- Mound Top
►-• Ground
Fig. 3. Mound-top, mound-base, ground, and air tem-
peratures around a small Formica obsciihpcs mound from
0800 to 2000 h on 17 Jul\ 1994 at 2560 m in Colorado.
depth.s of the nests were 0.3 m (estimated), 0.3
m, 0.64 m, and 0.97 m.
The nest excavated in 1993 contained the
following arthropods: pseudoscorpions, coUem-
bolans, beetles and beetle larvae (1 Ctenicera
sp. [E Elateridae] and 4 Eleodes sp. [E Tene-
brionidae]; Table 1).
The following insects were identified in the
1994 nest: CoUenibola (E Entomobryidae); Hom-
optera (E Cicadellidae — 1 inmiature, E Aphid-
idae — 2 immatures); Hemiptera (E Anthocori-
dae — 1 specimen); Coleoptera (E Curcnlion-
idae — 5 adults, E Scarabaeidae — 1 adult and
Cremastocheihis pupa and larval skin, probable
E Carabidae — 1 adult, probable E Anthri-
bidae — 2 larvae, E Tenebrionidae — unidenti-
fied lai-vae, probable Eleodes sp. lai-vae, and
Eleodes sp. pupae, E Cerambycidae — Lepturi-
nae, probable Leptiira sp. larva); Diptera
(probable E Asilidae — pupa); Hymenoptera (E
Formicidae — few Tapinoma sessile [Say] work-
DiscussioN .\ND Conclusions
The extrapolated densit\' of 115 mounds/ha
is about 1.8 times greater than the highest den-
sity reported: 64Aia of Jack pine in Manitoba
(Bradley 1973a).
Colonies are known to be polydomous and
to reproduce by budding (Herbers 1979). Some
primarv' mounds and small secondaiy mound-
lets along trails appeared and disappeared in
our study area over the years as previously
reported, and some may have moved. For ex-
ample, a primary mound that was active in
1990 was largely deserted by 1994 and com-
pletely abandoned in 1995. Colonies have
been reported to move at least 3 times during
their life and to move 18 m from their original
location, or 1.3-;33 m after transplantation (Brad-
ley 1972, 1973a). King and Sallee (1953, 1956)
noted desertions of many old nests and the
establishment of 1 or more new ones from each
of them.
All our mounds were in open sagebrush ex-
cept for 2 built around aspens at the forest edge.
Weber (1935) also noted that most mounds are
in the open, but did find some mounds par-
tiallv shaded and 1 enormous mound almost
completely shaded in an aspen grove.
In our study, 63 of 98 mounds (64%) showed
evidence of being built around sagebrush as
reported by Weber (1935), but a few were built
around other structures such as trees and a
fencepost. Weber noted that workers kill sage-
bmsh by chewing bark at the base and spraying
formic acid on the cambium. After 3 months,
the stem is removed to form a longitudinal pas-
sage in the center of the mound leading to the
main entrance.
Weber (1935) reported that mounds are
composed of slighdy longer twigs (1-12 cm)
than the ones we measured (0.4-8.9 cm), but
these slight differences may simply reflect the
availability of materials.
A new discoveiy was the presence of fecal
pellets of P. manicuhitits or Microtus (Clark
personal communication) on the surface and in
the thatch of Colorado mounds. Since workers
were never observed canying pellets to mounds,
their origin is unclear.
Although Clark and Comanor (1972), Tilbot
(1972), and Wheeler and Wheeler (1986)
330
Great Basin Naturalist
[Volume 56
Table 1. Arthropods in Formica obsctiripcs Forel nests
reported in die literature luid identified troni 2 excavated
nests near Soap Creek, Colorado (*).
*Collembola (unident.) Weber (1935)
*E Entomohnidae
*Honioptera
*F. Aphididae — 2 immatin-es
*E Cicadellidae — 1 innnature
*Heiniptera
*E Anthocoridae — 1 specimen
Diptera
*E Asilidae — pupa
E Milichiidae
PliyUonujza seciiriconiis Weber (1935)
E Leptidae — Ian ae Weber (1935)
E Anthoniyiidae — lanae \\'eber (1935)
E There\ idae — lai-\'ae Weber (1935)
E Phoridae — lana W'indsor (1964)
Lepidoptera
*E Noctuidae — lar\ae
Epizcitxis sp. — lar\ae \\'eber (1935)
H\nienoptera
E Eormicidae
Lasius hitipcs W;dsh \\eber (1935)
Lcptoflwrax hiiiicornis Emen Weber (1935)
*T(ipiiio)na sessile Say \\'eber (1935)
Th\ sanura — sil\ ertisli ^^■indsor (1964)
Coleoptera
Unident. beeUe pupa Windsor (1964)
E Elateridae
*Cteiucera sp. lana
Mehiiwtus sp. lanae Weber (1935)
E Tenebrionidae
I'nident. lanae and adults \\'indsor (1964)
*Eleo(Ies sp. — lanae and pupae
*Unident. lanae
E Carabidae
Ainara sp. — adult temale \\eber (1935)
*Prob. adult
Unident. adults Windsor (1964)
*E Anthribidae
*2 prob. lanae
*E Ceranib\ cidae
*Prob. Leptuni sp. — lana
*E Cmculionidae
*5 unident. adults
Coleoptera (continued)
E Scarabaeidae
*Unident. adult
*Crein' of
mound entrances changed o\"er time as re-
ported b\ Weber (1935). The number of en-
trances per mound in our stud); 1—52, is close
to the range of 3^50 per mound reported (Cole
1932. \\heeler and Wheeler 1986). In the earl>-
morning ants use openings in die sunlight; later
as the temperatine rises tiie\' use onK' shaded
entrances as reported b>" Weber (1935). Hen-
derson and Akre (1986) speculate entrances are
opened dining the da\ and closed widi diatch
at night to help control nest temperatures.
Our mounds. especialK mid-sized and large
ones, were geneialK warmer than groimd and
1996]
Nestinc: Ecoixxn of Tiiatchinc; Ant
331
air temperatures and exhibited Hiennal stratifi-
cation from top to base. Weber (1935) and
Andrews (1927) also noted that mounds are
w anner than the ground, and Andrews reported
that the upper parts are warmer tlian the lower
parts of" mounds. The differential between our
mound-top and air temperatures was greater in
the evening dimi in the iiftenioon. Sniiill mounds
showed a re\ ersal of diermal stratification and
greater liourK fluctuation of top and base tem-
peratures, whicli is indicati\ e of poorer ther-
mal regulation.
Marking experiments suggest that worker
longevit}' is often short but that some workers
oxenvinter and lixe more than a year. Little
information is available on the longevity of
worker ants and none was found for diis species.
Akhough maximum longevit>' is known to be 3
\ r for workers of some species, such as Aphae-
)u) area. Weber (1935), on the other hand,
noted that mites (Parasitidae, Tyroglyphidae,
Vropoda sp.) were common on workers and
sexuals, especially on the tibia-tarsal joint, and
estimated over 200 on 1 queen.
Excavated nests varied in depdi from 0.3 to
0.97 m, or less than the maximum depth of
1.37 m noted by McCook (1884) and 1.58 m
reported by Weber (1935). Weber speculated
that the water table (below 1.52 m) limits nest
depth.
Nests excavated from 27 June to 6 August
did not contain winged reproductives except a
male in 1 nest. This finding differs from Cole's
(1932) obsei-\ations of large numbers of winged
reproductives through June and July.
Many species of Formica are polygynous
(Kannowski 1963). The number of wingless
( [ueens per Colorado nest varied from 0 to 198
(Conway 1996). The latter number far exceeds
the 2 or more queens per colony reported by
Cole (1932).
The following arthropod groups found in
our excavated nests had not been reported
associated with this species: pseudoscorpions,
coUembolans (E Entomobiyidae), homopterans
(E Aphididae, E Cicadellidae), hemipterans (E
Anthocoridae), dipterans (E Asilidae), and cole-
opterans (E Carabidae, E Anthribidae, E Cur-
culionidae, E Elateridae — Ctenicera sp., E
Tenebrionidae — Eleode.s sp., and E Ceramby-
cidae — probable Leptura sp.; Table 1).
On the other hand, Windsor (1964) and
Henderson and Akre (1986) reported 3 major
groups not found in our limited sampling:
Arachnida (small spiders), Tlnsanura (silvcHlsh),
and Orthoptera (E Cryllidae). In addition,
Weber (1935), Windsor (1964), and MacKay
and MacKay (1984) noted many dipteran and
coleopteran families not in our nests and new
genera and species in a few of the same fami-
lies found in our nests (Table 1).
The relationship of these myrmecophiles
with the host colony is unclear. Larval and
adult coleopterans and noctuid lawae may use
the chambers for hibernation or development
(W^eber 1935). Staphylinid beetles may prey
upon brood or workers. Jones (1929) suggested
that lepidopteran, coleopteran, and dipteran
larvae are tolerated because they feed on
decaying vegetable matter in the nest. Cremas-
tochcihts is a well-known symbiont in the nests
of a number of ant species (Holldobler and
Wilson 1990). The scarab genus Euphoria may
be a symbiont treated with indifference liy the
host colony (Wheeler 1910). On the other hand,
ants are aggressive to other guests, such as the
m>rmecophilous cricket {Mijnnecophihi manni
Schimmer; Henderson and Akre 1986).
Weber (1935) reported 3 ant species in nests
(Table 1) and noted that Leptothorax hirticornis
Emery may prey upon brood or isolated work-
ers. Tapinoma sessile, one of the species in our
nests, often steals honeydew from thatching
ants throughout its territoiy, but seems to elicit
little defensive response (Mclver and Loomis
1993).
The high altitude of our study site did not
seem to significand)' alter nest dimensions and
ecology, but the work did provide new findings
on this species, such as the greatest mound
density per hectare, first report of probable P.
maniculatm fecal pellets associated widi mound
thatch, new information on the thermal prop-
erties of mounds, new^ information on worker
longevity', greatest number of wingless queens
reported in a nest, and possible new myrme-
cophiles.
Acknowledgments
Support for this research was pro\ided by
grants from die Howard Hughes Medical Insti-
tute through the Undergraduate Biological
332
Great Basin Naturalist
[Volume 56
Sciences Education Program in 1993-94 to the
following University of Scran ton students: John
Bridge, Tom Sabalaske, Anthony Musingo, and
Jeanne Rohan. I would also like to thank the
Systematic Entomolog) Laboraton' in Beltsxille,
Maiyland, for identification of myrmecophiles;
William Clark at the Onna J. Smith Museum of
Natural Histoiy, Albertson College of Idaho, in
Caldwell, Idaho, for identification of mam-
malian fecal pellets; and the Forest Sei-vice in
Gunnison, Colorado, for identification of plants.
LiTER.\TURE Cited
Andrews, E. A. 1927. Ant-mounds as to temperature and
sunshine. Journal of \Iorpholo^\ and Ph\ siologv' 44:
1-20.
Br.\dley, G. a. 1972. Transplanting Formica obsciiripes
and DoUchodenis taschenbcrgi (H\ nienoptera: Fonni-
cidae) colonies in Jack pine stands of southeastern
Manitoba. Canadian Entomologist 104: 245-249.
. 1973a. Interference between nest populations of
Formica obscuripes and Dolichoclcrus taschenbcrgi.
Canadian Entomologist 10,5: 1525-1528.
. 1973b. Effect ofFonnica obscuripes (H\inenoptera:
Fonnicidae) on tlie predator-pre\' relationship between
Hyperaspis congressis (Coleoptera: Coccinellidae) and
ToiimeycUa uumismaticiim (Homoptera: Coccidae).
Canadian Entomologist 105: 1113-1118.
Clark, W. H., and P. L. Com.\.\'OR. 1972. Flights of the
western thatching ant, Formica obscuripes Forel, in
Nevada. Great Basin Naturalist 32: 202-207.
Cole, A. C, Jr. 1932. The thatching ant, Formica obscur-
ipes Forel. Ps\'che 39: 30-33.
Conway, J. R. 1996. Nuptial, pre-, and postnuptial acti\ it\
of the thatching ant, Formica obscuripes Forel, in
Colorado. Great Basin Naturalist 56: 54—58.
Grecg, R. E. 1963. The ants of Colorado. University of
Colorado Press, Boulder 792 pp.
He\der,so\, G., and R. D. Akre. 1986. Biologv of die myr-
mecophilous cricket, Mynnccophila monni (Oitlioptera:
Gnllidae). Journal of the Kansas Entomological Soci-
ety' ,59: 4,54^67.
Herbers, J. M. 1979. The evolution of se.x-ratio strategies
in hvnienopteran societies. American Natiualist 114:
818-834.
Holldobler, B., and E. O. Wilson. 1990. The ants. Bel-
knap Press of Han ard University' Press, Cambridge,
MA. 732 pp.
Jones, C. R. 1929. Ants and their relation to aphids. Bul-
letin of the Colorado Agricultural E.\periment Station
.341: 1-96.
KiNt;, R. L., AND R. M. Sallee. 1953. On the duration of
nests oi Formica obscuripes Forel. Proceedings of the
Iowa Academ\' of Science 60: 656-659.
. 1956. On the half-life of nests of Formica obscur-
ipes Forel. Proceedings of the Iowa Academy of Sci-
ence 63: 721-723.
Kannowski, R J. 1963. The flight acti\ ities of formicine
ants. Symposia Genetica et Biologica Ittilia 12: 74—102.
MacKay, E. E., and W. P M.\(:K.\y. 1984. Biology of the
thatching ant Formica Jiaemorrlioidalis Emeiy (Hyme-
noptera: Formicidae). Pan-Pacific Entomologist 60:
79-87.
Mann, \V. M. 1911. On some north western ants and their
guests. Psyche 18: 102-109.
McCoOK, H. C. 1884. The mfous or thatching ant of Dakota
and Colorado. Proceedings of the Academy of Nat-
ural Sciences of Philadelphia, part 1: 57-65.
Mcl\ ER, J. D., AND C. LooMis. 1993. A size-distance rela-
tion in Homoptera-tending thatch ants (Fonnica
obscuripes, Formica planipilis). Insectes Sociau.x 40:
207-218.
McIvER, J. D., .^ND T. Steen. 1994. Use of a secondary
nest in Great Basin Desert thatch ants {Formica ob-
scuripes Forel). Great Basin Naturalist 54: 359-365.
Talbot, M. 1972. Flights and swamis of the ant Fonnica
obscuripes Forel. Joinnal of the Kansas Entomologi-
cal Society 45: 254-258.
Weber, N. A. 1935. The biology of the thatching ant
Formica obscuripes Forel in North Dakota. Ecological
Monographs 5: 16.5-206.
\\'heeler, W. M. 1910. Ants, their stioicture, development
and beliavior Columbia University Press, New York.
663 pp.
Wheeler, G. C., and J. N. Wheeler. 1986. The ants of
Nevada. Natural Histoiy Museum of Los Angeles
County, Los Angeles, CA. 138 pp.
Windsor, J. K. 1964. Three scarabaeid genera found in
nests of Formica obscuripes Forel in Colorado. Bul-
letin of the Southern California Academy of Sciences
63: 205-209.
Received 5 January 1996
Accepted 2S June 1996
Cri'at Basin Naturalist 56(4), © 199fi, pp. 333-3 10
GAS EXCHANGE, 8l^C, AND HETEROTKOPHY FOR CASTILLEJA
LINARIIFOLIA AND ORrHOCARPUS TOLMlEl FACULTATIVE ROOT
HEMIPARASITES ON ARTEMISIA TRIDENTATA
Lori A. Duchainu'' and James H. Klilfiiii^cr'
Abstract. — Gas e.xchange and carbon isotope ratios were measured on 2 Facultative heniiparasites, CasUllcja liiiari-
ifolia Benth. (Indiau paintbrush; Scropliulariaceae) and Orthocarpu.s toliniei II. & A. Oblmie owl clover; Scropluilari-
aci'ae), and their Ar/e//u'.s/V; trklentata L. (bit;; sauebrush; Asteraeeae) hosts. Photosynthetic rates differed j^reatly between
\ cars; rates in 1995 were more than double those in 1994, likely due to more precipitation and less water stress during
U)95. Despite this difference in precipitation, photos\nthetic rates for C liuanifoUa were not different from those of
(heir hosts for either year. However, carbon isotope ratios of C. linariifolia and O. tolm'wi were up to 3%o more negative
than those of their A. tridentata hosts. Using measmed 8'^C ratios in conjunction with S^^C ratios predicted from gas-
ixchange measurements, we calculated that C. linariifolia derived, on average, 40% of its leaf carbon heterotrophically.
(^outran' to current suggestions that high photosynthetic rates of heniiparasites are an indication of reduced heterotro-
ph\, C. linariifolia e.xhibited photosynthetic rates similar to autotrophic plants and used a substantial amount of host-
(iciixed carbon. Moreover, this evidence shows that manipulation of a heterotrophic carbon supply transcends obligate
lii'miparasites to include those plants whose parasitism is facultative.
Key words: hetcrotropluj, luiniparasitc, photofiijnilu'fiis, carbon isotope ratios, slirith ecolofiy.
Heniiparasites, chlorophyllous parasitic
plants, form an apoplastic continuum with host
x\ lem (Ra\'en 1983). R has been assumed that
these plants are largely autotrophic plants,
being parasitic only for water and minerals
(Smith et al. 1969). However, heniiparasites
may also gain carbon through the passive
uptake of dilute concentrations of organic car-
bon contained within host xyleni sap (Raven
1983). Early studies using radiocarbon labeling
demonstrated the transfer of solutes from host
to parasite (Hull and Leonard 1964, Govier et al.
1967), although it was not possible to quantify
tliis flux. Experiments of Govier et al. (1967), in
which [^"^C]urea or ^"^COq was fed to hosts,
showed the movement of ^'^C labeled coni-
poimds to all parts of the hemiparasite Odon-
tites venia (Scropliulariaceae). More recent stud-
ies used a carbon budget model and/or a h^'^C
method to (juantify the extent of heterotrophy
(Press et al. 1987a, Graves et al. 1989, Marshall
and Ehleringer 1990, Schulze et al. 1991, Mar-
shall et al. 1994, Richter et al. 1995). Using the
latter method, Press et al. (1987a) calculated tliat
28-35% of total carbon in Striga hennonthica
and Striga asiatica (Scropliulariaceae) is host-
derived carbon. There is also ample evidence
that hemiparasitic mistletoes utilize host-derived
carbon, although the values vaiy greatly, from
5% to over 60% (Marshall and Ehleringer
1990, Schulze et al. 1991, Marshall et al. 1994,
Richter et al. 1995). Despite the potential
importance of heterotrophy to carbon acquisi-
tion in parasitic plants, relatively few studies
have addressed this aspect of parasite-host
interactions. Moreover, none have evaluated
the exploitation of this carbon source by facul-
tative root heniiparasites.
Photosynthetic rates of heniiparasites fall
within the lower range reported for C3 plants
and are generally much lower than photo-
synthetic rates of the host. S. hennonthica has
a poorly developed palisade mesophyll, con-
tributing, in part, to photosynthetic rates as
low as 2.5 /zmol ni-2 s-1 (Shah et al. 1987).
Moreover, these rates are half those reported
for their Sorghum hosts (Press et al. 1987b).
Striga species are the most extensively studied
root heniiparasites because of their importance
as agricultural weeds in semiarid Africa, and
as obligate hemiparasites they require host
attachment for sui-vival. Similarly, low photo-
synthetic rates were found in facultative root
hemiparasites. Press et al. (1988) measured
iDqwrtiiK-iitofBioIogy, Universih' of Utah, Salt Lake Cih; UT84U2
333
334
Great Basin Naturalist
[Volume 56
light-saturated photosynthetic rates of 2.1 to
7.5 fimol m~^ s~^ for 8 facultative species of
Scrophulariaceae. However, 1 exception to this
trend of low photosynthetic rates is the Medi-
teiTanean facultative heniipai"asite Bartsia trixago
(Scrophulariaceae), which has CO2 assimilation
rates ranging from 12.4 to 18.8 jU-mol m"^ s"^
well within the range measured for potential
hosts (Press et al. 1993).
Castilleja and Orthocarpiis are facultative
hemiparasites, those with the abilit\' to sui-\'ive
in the absence of a host. It is this facultative
parasitism that distinguishes them from Striga.
The majority of Castilleja species are perennial,
while Orthocarpiis are annuals. Both occur
throughout the Intemiountain West most com-
monly in the pinyon-juniper, mountain brush,
and aspen-conifer zones (1140-3140 m eleva-
tion), with Orthocarpiis tohniei occurring only
at the higher elevations (2195-3265 m; Welsh
et al. 1987). Castilleja hnariifolia and Oiihocar-
piis tohniei parasitize a variety of host species
(Heckard 1962, Atsatt and Strong 1970). Artem-
isia tridentata is the common host for both
hemiparasites at the sites studied in the paper
Our overall objective was to investigate gas
exchange and heterotrophy characteristics for
facultative hemiparasites. We focused primarily
on the facultative root hemiparasite Castilleja
linariifolia infecting Artemisia tridentata hosts.
A secondaiy focus of this study was Orthocar-
piis tohniei, a closely related annual facultative
root hemiparasite, also infecting A. tridentata
hosts. We asked the following questions: Do C.
hnariifoha and O. tohniei exhibit gas-exchange
activities similar to those of their hosts? Does
C. hnariifoha utilize heteroti-ophic carbon? Does
hemiparasite infection impact water availability
and gas-exchange rates of A. tridentata hosts?
To evaluate these questions, we measured gas
exchange and analyzed carbon isotope compo-
sition for C hnariifoha, O tohniei, infected and
uninfected A. tridentata. In addition, predawn
water potentials (^pj) were measured for
infected and uninfected A. tridentata to exam-
ine the impact of hemiparasite infection on
host water availability.
Materials and Methods
Study Sites
This study was conducted at 2 sites in Utah
where the hemiparasites have different grow-
ing seasons. The first site, Tintic, is located just
off Mclntyre Road, approximately 12 km south
of Eureka, Utah (Juab Co.), at the Desert
Range Experimental Station operated by Utah
State University (latitude 39°51'N, longitude
112°12'W). The area is a sagebrush steppe
habitat at about 1525 m elevation where sage-
bi-ush is interspersed with herbaceous species
such as Erigeron, Castilleja, Astragalus, and
Phlox. The growing season for Castilleja at this
site begins in late April and ends in late June
to early July. The second site, Sti^awbeny Reser-
voir (Wasatch Co.), is about 130 km southeast
of Salt Lake City and approximately 800 m
north of Highway 40 along Coop Creek Road
(latitude 40°15'N, longitude 111°8'W). This
site lies in the southern tip of the Uinta
National Forest at about 2280 m elevation.
Sagebnish is the dominant shrub mixed with a
few herbaceous species such as Castilleja,
Oiihocarpiis, and Malta. The growing season
for C. hnariifoha at Strawbeny Reservoir begins
in early June and extends through August; O.
tohniei begins a few weeks later and extends
into September.
Twenty pairs of C. linariifolia and A. triden-
tata hosts were selected at each site. At Straw-
beny Reservoir an additional 20 pairs of O.
tohniei and A. tridentata hosts were selected.
In addition, 5 uninfected A. tridentata were
selected at both sites as hemiparasite-free
controls.
Gas Exchange
Photosynthesis and stomatal conductance
were measured with a portable gas-exchange
system (LI-6200, Licor Instruments, Lincoln,
NE, USA) twice during the C. hnariifolia grow-
ing season at the Tintic and Strawbeny Reser-
voir sites. Specific dates were chosen to corre-
spond with the early and late parts of the para-
site growing season. At both sites data were
collected during diurnal peak photosynthesis
(1000-1300 h MST) on 20 pairs of C. hnariifo-
lia and infected A. tridentata, and on an addi-
tional 5 uninfected A. tridentata. Dining the
late season at Strawbeiry Reservoir, measure-
ments were taken on an additional 20 pairs of
O. tohniei and infected A. tridentata. After gas-
exchange measurements were completed, foli-
age was removed for leaf-area measurements
using a leaf-area meter (LI-3100, Licor Instru-
ments, Lincoln, NE, USA).
mm
Heterotropiiv in Castilleja and ORTHOCARI'US
335
Water Potentials
Stems of approximate!) ecjual leut^th and
diameter were seleeted lor predawn water-
|i()tential (^p^) measurements using a pressure
homl) (PMS Instruments, Comillis, OR, USA)
lor 20 inleeted and 5 iminfeeted A. tridentata
at l)oth sites. These measurements were taken
approximately every 2 wk from May through
eaily July at the Tintic site and late June
through the end of August at the Strawberry
Hesenoir site.
Carbon Isotope Composition
Carbon isotope ratios (S^'^C) were analyzed
for the same plants used to measure gas
exchange. The foliage was dried for 24 h and
then finely ground with a mortar and pestle to
homogenize the tissue (Ehleringer and
Osmond 1989). Subsamples of 1-2 mg were
combusted to produce CO2, which was mea-
sured using an isotope ratio mass spectrometer
(delta-S, Finnigan MAT, Bremen, Germany).
Results are expressed using the S^'^C notation
(%o), which relates the isotopic composition of
the sample to the PDB standard as follows:
813c =
r5;^^i^^^-il*iooo%r, (1)
L ^staiulaid J
where R represents the ratio of ^■^C0.2/^'^C02
of the sample and standard, respectively
(Ehleringer and Osmond 1989). All isotope
ratio anah'ses were conducted at the Stable
Isotope Ratio Facility for Environmental
Research at the Universitv of Utah, Salt Lake
Cit>', Utah, USA.
Calculation of Heterotrophy
Heteroti'ophy was calculated using measured
and predicted b^-^C ratios. The predicted b^-^C
ratio (8pp), the carbon isotope composition of a
leai provided that all carbon is autotrophic,
was estimated with intercellular COo concen-
trations (cj) from gas-exchange measurements.
E(iuation 2 relates cj to the leaf carbon isotope
ratio as modeled by Farquhar et al. (1982):
K = 8a-a-(l3-il)(Ci/Ca),
(2)
where 8p is the h^^C of the plant (= 8 in this
study), 8^ is the approximate b^'^C of the air
(— 7.8%c), a and b are discrimination factors due
to diflhision (4.4%c) and carboxylation via RuBP
carb()x>lase (27%c), respectively, c., is the con-
centration of CO2 in air (ppm) and Cj was cal-
culated from gas-exchange measurements
described abovc>. Ileterotrophv (H) was calcu-
lated for the 1994 data (9 C. linariifolia, 5
infected and 5 iminfected A. tridentata) using
E(][uati()n 3:
5 2
jj _ "pp ~"in
Spp -^1.
(3)
where 8pp is the predicted S^'^C for the para-
site, 8^ is the h^-^C measured in the parasite
tissue, and 8], is the S^^C measured in the host
tissue (Press et al. 1987a).
Statistical Analysis
Analysis of variance was used to compare
yearly, seasonal, and plant means within a site
for all photosynthetic data, and yearly and sea-
sonal means for carbon isotope ratios (JMR
Version 3, SAS Institute Inc., Caiy, NC, USA).
The Tukey-Kramer Honestly Significant Dif-
ference test (HSD) was used to make specific
comparisons. In addition, for each hemipara-
site, carbon isotope ratios were compiled for
all seasons and sites, and differences between
hemiparasites and hosts were compared using
a t test. A paired t test was used to determine
differences between predicted and measured
b^'^C for each C. linariifolia, uninfected and
infected A. tridentata. Differences in ^p^j
water potential between infected and unin-
fected A. tridentata were determined b)- 1 tests
within each date.
Results
Analysis of annual trends in photosynthetic
rates for Strawbeny Resewoir (Fig. 1) revealed
that plants had significantly higher rates in
1995 than in 1994 for both parasite and host
(Tukey-Kramer, a = 0.05). For example, in
1995 photosynthetic rates for C. linariifolia and
infected A. tridentata were 18.3 ± 2.1 and 16.0
± 0.6 ^tmol m"2 s'^, respectively, more than
double those during the 1994 season. We also
found seasonal differences in photosynthetic
rates at StrawbeiTy Resenoir Both C. linariifo-
lia and infected A. tridentata at Strawberry
Reservoir experienced a significant decline in
photosynthetic rates late in the season, with
rates falling —6.7 and 8.6 ^tmol m'^ s'^ re-
spectively (Tukey-Kramer, a = 0.05). However,
336
Great Basin Naturalist
[Volume 56
Strawberry Reservoir, Utah
Tintic, Utah
o
E
CO
CO
(D
c
>>
CO
o
■I— «
o
sz
CL
%0^ %
Fig. 1. Mean pliotosynthetic rates for liosts and parasites. Sites and sample sizes are as follows; Uninfected A. triclen-
tata (Tintic: » = 3 for early season, n — 4 for late season; Strawberry Resenoir: n = 3 for late season), infected A. triden-
tata (Tintic: ii = 12 for early season, n = 7 for late season; Strawberiy Resei^voir: n = 7 for early season, n = 19 for late
season), C. linuriifoliu (Tintic: n = A for early season, /! = 6 for late season; Strawberiy Resewoir: n — 3 for early season,
n — 5 for late season), O. tohniei (Strawberiy Resenoir: n = 5 for late season). Data are shown for Strawberiy Resei^voir
(left panel) and Tintic (right panel) during the 1994 early season (open bars), 1995 early season (hatched bars), and 1995
late season (solid bars). Letters denote significant differences within each site. Error bars represent ± 1 s^.
photosvnthetic rates at Tintic showed no sea-
sonal differences (ANOVA, F = 1.88, F =
0.134; Fig. 1). In spite of annual and seasonal
differences in photosynthesis for parasite and
host plants, we found no difference in photo-
synthetic rates between C. linanifolia and
infected A. tndcntata. In contrast, O. tohniei
rates (14.0 ±1.1 jitmol m~- s"l) exceeded those
for infected A. tridentata (9.3 ± 0.4 ^tniol m~-
s~l; Tukey- Kramer, a = 0.05; Fig.l).
At both sites we found no significant differ-
ence in predawn water potentials (^p^)
between infected and uninfected A. tridentata
(F > 0.05 for all dates, t test), although there
was a general decline throughout the season
(Fig. 2). The range in ^ | was similar be-
tween sites; however, the values were slightly
more negative at Tintic.
Carbon isotope ratios differed between years
for infected and uninfected A. tridentafa, with
more negative values in 1995. However, 6^'^C
values for C. linariifolia did not differ between
years (Tukey- Kramer, a = 0.05; Table 1). Our
results showed a slight seasonal decline in
S^'^C values for parasites and hosts at Straw-
beriy Resen'oir, although only O. tohniei and
infected A. tridentata were significantly different
(Tukey-Kramer, a = 0.05; Table 1). This trend
in seasonal reduction was not evident for plants
at the Tintic site. Furthermore, we found that
hemiparasite S^'^C ratios were significantly more
negative than those of the hosts (C. hnariifoha,
t = 12.57, F < 0.001; O. tohniei, t = 11.94, F <
0.001). In 1994 C. hnariifoha b^'^C values
(-28.9 ± 0.34%<:i) were nearly 3%c more nega-
tive than those of the hosts (-26.2 ± 0.13%c),
while this difference naiTOwed in 1995 to ~2%c
at Tintic and ~l.5%c at Strawbeny Resei-voir.
Results from experiments in 1994 showed a
significant mean difference of 1.34%c between
predicted and measured 8^^C ratios for C.
hnariifoha (paired t test, f = 2.745, F < 0.05;
Table 2). Using this difference we calculated
that, on average, 40% of C. hnariifoha leaf car-
bon was host derived; individual plants ranged
from 16 to 60% (Table 3). C. hnariijoha hetero-
trophy is well within the range of xalues calcu-
lated for obligate hemiparasites. There was no
statistical difference lietween measured and
predicted 8^'^C values for either infected or
1996]
Heterothophy in Castilleja and Ortiiocarpus
337
-0.5
-0.5
1 1 1
Tintic, Utah 1995 o
1
infected
A. tridenlala
•
uninfected
-
A. tridentata
•
^^^^^
^~~'^'^->^
-
o ^^^^^
1 1 1
1 1 1
Strawberry Reservoir, Utah 1995
1
\ •
o \^
-
^\^
• "^~-
__•
6
1 1 t
1
Month
Fig. 2. Seasonal course of predawn water potentials tor
infected A. tridentata (open circles; n = 11) and unin-
fected A. tridentata (solid circles; n — 5). Data are pro-
vided for Tintic (upper panel) and Strawberw Reser\'oir
(lower panel) from May to late August of 1995. Error bars
represent ± 1 % •
uninfected A. tridentata, indicating no hetero-
trophic carbon gain as expected.
Discussion
Our results suggest that, with the exception
of photosynthesis, the hemiparasites in this
study behaved similarly to other hemiparasites.
Photosynthetic rates for hemiparasites in this
study were higher than rates for most other
hemiparasites and similar to those of their auto-
trophic host plants. We also found large differ-
ences between years, which likely reflected
differences in precipitation. In agreement with
other studies, hemiparasite S^'^C ratios were
more negative than those of the host (Press et al.
1987a, Marshall and Ehleringer 1990, Schulze
et al. 1991, Richter et al. 1995). Furthermore,
large differences in S^'^C ratios between the
parasite and host suggested that the hemipara-
site utilized a substantial amount of host-
derived carbon. Despite relatively high photo-
synthetic rates, heterotrophy estimates for C.
linariifolia range fi-om 16% to 69%.
We found large interannual differences in
photosyntlietic rates and carbon isotope ratios
for C. linariifolia and A. tridentata, which most
likely indicated a response to precipitation dif-
ferences. Climate records showed that the grow-
ing season at Strawbeny Resenoir in 1994 was
consideral)ly drier than in 1995; the spring
(March-May) of 1994 received only 96.3 mm
of precipitation, whereas precipitation in the
spring of 1995 totaled 216.4 mm (Utah Climate
Center, Meber station). Differences in precipi-
tation during the spring influence the amount
of soil water available to the plants. This water
supply can be indirectly assessed by measuring
the plant's water potential before the sun rises
and photosynthesis commences. Our "^Pp^j mea-
surements corroborated that 1994 was a drier
growing season; during 1994 the ^ | range for
A. tridentata (-1.7 to -3.2 MPa) was much
more negative than the ^ j range for A. tri-
dentata in 1995 at either Tintic (-0.7 to -1.1
MPa) or Strawberry Reservoir (-0.3 to -0.9
MPa). Photosynthetic rates doubled during
1995, presumably in response to this increased
precipitation. Interannual differences wei^e most
pronounced for C. linariijolia, which showed
photosynthetic rates 3-fold higher in 1995 rela-
tive to rates in 1994. Carbon isotope ratios for
autotrophic C3 plants represent an estimate of
long-term water-use efficiency (mmol C/mol
H2O; WUE), with more negative S^'^C ratios
reflecting a lower WUE (Ehleringer and
Osmond 1989). S^'^C ratios for infected and
uninfected A. tridentata were significantly more
negative during the wetter year, thus suggest-
ing they were less conservative in their water
use. Using S^'^C ratios as a measure of water-
use efficiency is inappropriate for hemipara-
sites because of the potentially confounding
effects of assimilating heterotrophic carbon.
Therefore, it follows that the 8^'^C ratio for C.
linariifolia should also reflect influences from
the import of host-derived carbon rather than
simply the influences of increased precipita-
tion. This prediction was supported by C. lin-
ariifolia data, where, despite the large increase
in precipitation, we saw no difference in S^^C
ratios between years.
Photosynthetic rates also responded to sea-
sonal inffuences, although rates were not dif-
ferent between parasites and hosts. Photosyn-
thetic rates declined during the growing sea-
son, which, in part, may be attributed to the
drier conditions late in the season as indicated
338
Great Basin Naturalist
[Volume 56
Table 1. Carbon isotope ratios (S^-^C) for hosts and parasites. Sites and sample sizes are as follows: Uninfected A. tri-
dentata (Strawberry Reservoir 1994; n = 5; Strawberiy Resei-voir 1995: n = 3 for early season; Tintic 1995: »! = 4 for
early season, n — 5 for late season), infected A. tridenfata (Strawbeny Reservoir 1994: n = 5; Stravvberr\' Reservoir
1995: n = 20 for early season, n — 30 for late season; Tintic 1995: n = 10 for early season, n = 11 for late season), C.
linariifolia (Strawbeny Reservoir 1994: n = 9; Strawbeny Resei-voir 1995: n = 7 for early season, n = 8 for late season;
Tintic 1995: 7i = 11 for early season, n — 8 for late season), and O. tobniei (Strawberrx* Resenoir 1995: 0=9 for early
season, n = 19 for late season). Letters denote significant seasonal differences within a site and species (Tukey-Kramer
HSD, a — 0.05). Values shown are means ± 1 .s^. NA denotes data not available.
A. tridentata
Site
Year
Lhiinfected
Infected
C. linariifolia
O. tobniei
1994
-25.56 ± 0.32«
-26.24 ± 0.13«
-28.93 ± 0.34^'
NA
1995
Early
-27.S6±0.ll'
-27.30 ± 0.11''
-28.91 ± 0.15^'
-28.66 ± 0.09^'
Late
NA
-27.80 ± 0.09^
-29.33 ± 0.25«
-29.50 ± 0.10'
1995
Earlv
-27.57 ± 0.19''
-27..32 ± 0.2ll'^'
-29.19 ± 0.22^1
NA
Late
-27.33 ± 0.19''
-27.17 ± 0.16''
-29.23 ± 0.1.5-'
NA
Strawberry Resenoir
Tintic
by predawn water potentials. Perhaps, the de-
cline in C. linariifolia photosynthesis was also
related to the phenology' of the hemiparasite. It
is possible that late in the season when these
hemiparasites set fruit, they rely less on cur-
rent photosynthesis and more on heterotrophic
carbon gain. Most studies of hemiparasite-host
gas-exchange dynamics found that hemipara-
site photosynthesis was much lower than that
of the host'(Hollinger 1983, Press et al. 1987b,
Pate et al. 1990, Marshall, Dawson, and Ehle-
ringer 1994). S. hennonthica and S. asiatica
have photos>'nthetic rates that are half of those
for Sorghum hosts (Press et al. 1987b). In con-
trast, the photosynthetic activities of C. linari-
ifolia in this stud}' were similar to rates of A.
tridentata hosts. This pattern remained stable
from year to year, despite large differences in
precipitation.
Hemiparasite gas-exchange rates have been
used to make inferences about potential het-
erotrophic carbon use. After calculating that
8.8-18.9 h of light-saturated photosyntliesis was
necessaiy for 8 different species of facultative
hemiparasites to reach zero net foliar carbon
gain. Press et al. (1988) hypothesized that they
must have had access to a heterotrophic car-
bon supply. Converseh; in Baii.sia trixago and
Parentucellia viscosa (Scrophulariaceae), where
photosyndietic rates were veiy similar to auto-
trophic plants, it was predicted that these fac-
ultative root hemiparasites were less reliant on
host-derived carbon (Press et al. 1993). Since
C. linariifolia also has photosynthetic rates sim-
ilar to those of its host, it follows that C. linari-
ifolia might not contain significant amounts of
heterotrophically derived carbon.
However, in our study this was not the case.
We found a relatively large difference in 'b^^C
ratios between C. linariifolia and A. tridentata
hosts, which likely indicates hemiparasite het-
erotrophy. Indeed, we calculated that C. linari-
ifolia in this study utilized an average of 40%
host-derived carbon. As with other parasitic
plants, unusually high transpiration rates rela-
tive to the hosts represent the most likeK' dri-
x'ing force for this assimilation of host-deri\ed
carbon. While the estimates of heterotrophy
found in this study are well within the range of
those reported for other parasites, one must
consider the inherent obstacles in using an
instantaneous measure of photosynthesis as a
basis for the predicted 8^^C ratios with an inte-
grated measure of actual leaf 8^'^C ratio. For
instance, differences in gas-exchange charac-
teristics at the time leaf carbon was incorpo-
rated may contribute to differences between
predicted and measured S^'^C ratios. Although,
we found no significant difference between pre-
dicted and measured 8^'^C ratios for infected
and uninfected Aiieniisia, a better contiol would
have been autotrophic C. linariifolia plants if
they had been a\ ailable. As mentioned earlier,
parasites may also access different pools of car-
bon at different times throughout the growing
season; in turn, this ma\' influence the S^'^C
ratios measiued in the leaf carbon. While these
factors may appear troublesome at first, the\'
represent a few of the many areas open to
investigation in parasitic plant ecophysiolog)'.
1996]
Hetekotkoimiv in Castilleja and Ortiiocarpus
339
Table 2. Measvired and predicted S^-^C values for iiniii-
Iccted and infected A. thdentutd {n = 5) and C'. linariifolia
ui = 9) at Strawhem Reserxoir in 1994. Means ± 1 s^ are
presented. Also shown is the difference between the pre-
dicted and measured values. * denotes significant difler-
ence at P < 0.05 (paired t test).
A. tridcntdta
Uninfected
Infected
C. liiiuriijolia
■^predicte
Difference
-25.56 ± 0.32
-0.98+1.19
-26.24 ±0.13
1.13 ±0.94
-28.93 ± 0.34
-1.34 ±0.48*
Though no other study quantifies hetero-
trophic carbon gain by a facultative hemipara-
site, a study by Hansen (1979) impHed potential
heterotroph)' in Castilleja chromosa. Experi-
ments measuring the difference of ^"'C labeled
sugar content in uninfected and infected
Artemisia tridentata indivdduals showed less ^"^C
in the infected host tissues. Hansen (1979)
hypothesized that this difference represented
sugar lost to the C. chromosa parasite. With
this indirect method, C. chromosa utilized, on
average, 10% host-derived carbon. Using a
more precise method, we would suggest from
our study that 10% heterotrophy may be an
underestimate.
Significant heterotrophic carbon gain by the
hemiparasite can be associated with a decrease
in host production. Graves et al. (1989) found
that dry weight of Sorghum infected with S.
hermonthica was 40% less than that of unin-
fected Sorghum, and hypothesized that the
effects of S. hermonthica were due to (1) the
direct reduction in host carbon by parasite het-
erotroph}' and (2) the indirect reduction of host
photosynthetic potential. Press and Stewart
(1987) showed that photosynthetic rates for
Sorghum infected by S. hermonthica were re-
duced by nearly half relative to those for unin-
fected Sorghum; stomatal conductance rates
were also significantly decreased. In contrast,
we saw no decrease in photosynthetic rates nor
stomatal conductance rates for infected A. tri-
dentata. Interestingly, there was an increase in
host photosynthesis relative to uninfected A.
tridentata late in the season at Tintic. Our study
also showed no difference in '^ | between
infected and uninfected A. tridentata, suggest-
ing that hosts in this study were not experienc-
ing detectable water stiess. Taken togedier these
Tablk 3. Calculated heterotropliy of C. liimhifolia in
this study compared to heterotropiu' calculated for other
hemiparasitcs.
.Species
Calculated
heterotrophy
in % (range)
CastiUcja
linariijolia
Striiid Iwnnontlticd,
Striga asiatica
Phoredendron
jiinipcrinum
Mistletoe species
Australian
mistletoe
40(16-69)
28-35
61
60 (49-67)
15 (.5-21)
This study
Press et al. 1987,
Graves et al. 1989
Marshall and
Ehleringer 1990
Schulze et al. 1991
Marshall etal. 1994b
data seem to suggest that C Unariifoha do not
negatively impact A. tridentata hosts. How-
ever, this conclusion may be relevant only dur-
ing unusually wet years; A. tridentata may
respond differently to hemiparasite infection
when drought conditions prevail.
One well-supported aspect of the host-para-
site relationship is the unusually high transpi-
ration rates of tlie parasite, often 10 times greater
than those of the host. It is generally believed
that this high water flux results in a water
potential gradient from the host to the parasite.
Therefore, through this mechanism higher
transpiration rates are thought to represent the
driving force for the transfer of solutes from
die host to parasite. Schulze et al. (1984) sug-
gested that high transpiration rates may be
necessary for mistletoe to acquire adequate
nitrogen for growth. The nitrogen-gathering
hypothesis has been the focus of several stud-
ies (Schulze et al. 1984, Ehleringer et al. 1985,
Marshall, Dawson, and Ehleringer 1994).
However, as Raven (1983) points out, these
plants are also inextricably acquiring signifi-
cant amounts of host carbon. Recent studies
indicated that heterotroph)' may be a wide-
spread phenomenon occurring in a variety of
obligate hemiparasitcs (Press et al. 1987a,
Graves et al. 1989, Marshall and Ehleringer
1990, Marshall et al. 1994, Richter et al. 1995).
Evidence from this study indicates that the
facultative root parasite C. linariifolia obtains a
substantial contribution of host-deri\'ed carbon,
thus extending further emphasis to the impor-
tance of this carbon supply for hemiparasitcs.
340
Great Basin Naturalist
[Volume 56
Acknowledgments
The authors thank Nina Buchmann and
J. Renee Brooks for critical review of the man-
uscript and assistance with statistical analysis;
Sue Phillips, Nathan Richer, and Sylvia Torti
for field assistance; and Craig Cook for assist-
ing with stable isotope measurements. All pre-
cipitation data were generously provided by
Utah Climate Center in Logan, Utah.
Literature Cited
Atsatt, E R., and D. R. Strong. 1970. The population
biology of annual grassland hemiparasites. I. The host
environment. Evolution 24: 278-281.
Ehleringer, J. R., and C. B. Osmond. 1989. Stable iso-
topes. Pages 218-290 in R. W. Pearcy, J. R. Ehle-
ringer, H. A. Mooney, and P W. Rundel, editors.
Plant physiological ecolog>'. Chapman and Hall, New
York.
Ehleringer, J. R., E. D. Schulze, H. Ziegler, O. L.
Lange, G. D. Farquhar, and I. R. Covvar. 1985.
Xylem-tapping misdetoes; water or nutrient parasites?
Science 227: 1479-1481.
Farquhar, G. D., M. H. O'Leary, and J. A. Berry. 1982.
On the relationship between carbon isotope discrimi-
nation and the intercellular carbon dioxide concen-
tration in leaves. Australian Journal of Plant Physiol-
ogy 9: 121-137.
GoviER, R. N., M. D. Nelson, and J. S. Pate. 1967. Hemi-
parasitic nutrition in angiosperms I. The transfer of
organic compounds from host to Odontites verna
(Bell.) Dum. (Scrophulariaceae). New Phvtologist 66:
285-297
Graves, J. D., M. C. Press, and G. R. Stewart 1989. A car-
bon b;Jance model of the Sorghiiin-Striata. Oecologia
60: 396-400.
Hull, R. J., and O. A. Leonard. 1964. Physiological aspects
of parasitism in mistletoes {Arceiithobiuin and Phora-
dendron) I. The carbohydrate nutrition of mistletoe.
Plant Physiology 39: 996-1007.
Marshall, J. D., and J. R. Ehleringer. 1990. Ai-e .xylem-
tapping mistletoes partially heterotrophic? Oecologia
84: 244-248.
Marshall, J. D., T. E. Dawson, and J. R. Ehleringer.
1994. Integrated nitrogen, carbon and water relations
of a xylem-tapping mistletoe following nitrogen fertil-
ization of the host. Oecologia 100: 430^38.
Marshall, J. D., J. R. Ehleringer, E. D. Schulze, and
G. Farquhar. 1994. Carbon isotope composition, gas
exchange and heterotrophy in Australian mistletoes.
Functional Ecology 8: 237-241.
Pate, J. S., N. J. Davidson, J. Kuo, and J. A. Milburn.
1990. Water relations of the root hemiparasite Olax
pliyllanthi (Labill) R.Br (Olacaceae) and its multiple
hosts. Oecologia 84: 186-193.
Press, M. C, and G. R. Stewart. 1987. Growth and pho-
tosynthesis in SorgJiinn bicolor infected with Striga
hennonthica. Annals of Botany 60: 657-662.
Press, M. C., J. D. Graves, and G. R. Stewart. 1988.
Transpiration and carbon acquisition in root hemipar-
asitic angiosperms. Joiunal of Experimental Botanv
39(205): 1009-1014.
Press, M. C, A. N. Parsons, A. W. Macxay, C. A. Vincent,
V Cochr.'VNE, and W. E. Seel. 1993. Gas exchange
characteristics and nitrogen relations of two Mediter-
ranean root hemiparasites: Bartsia trixago and Paren-
tucellia viscosa. Oecologia 95: 145-151.
Press, M. C, N. Shah, J. M. Tuohy, and G. R. Stewart.
1987a. Carbon isotope ratios demonstrate carbon flux
from C_^ host to C3 parasite. Plant Phvsiology 85:
1143_1145.
. 1987b. Gas exchange characteristics of the
Sorghitm-Striga host-parasite association. Plant Phys-
iologx- 84: 814-819.
Raven, J. A. 1983. Phytophages of .xylem and phloem: a
comparison of animal and plant sap-feeders. Advances
in Ecological Research 13: 136-239.
RiCHTER, A., M. Popp, R. Mensen, G. R. Stewart, .vnd
D. J. VON Willert 1995. Heterotrophic carbon gain
of the parasitic angiosperm Tapinanthiis oleifolius.
Australian Journal of Plant Ph\siolog>' 22; 537-.544.
Schulze, E. D., O. L. Lange, H. Zeigler, .vnd G. Gebauer.
1991. Carbon and nitrogen isotope ratios of misdetoes
growing on nitrogen and non-nitrogen fixing hosts
and on CAM plants in the Namib desert confimi par-
tial heterotrophy. Oecologia 88: 457^62.
Schulze, E. D., N. C. Turner, and G. Glatzel. 1984.
Carbon, water and nutrient relations of two mistle-
toes and their hosts: a hypothesis. Plant, Cell and
Environment 7: 293-299.
Shah, N., N. Smirnoff, and G. R. Stewart. 1987. Photo-
synthetic and stomatal characteristics of Sfrigo her-
monthica in relation to its parasitic habit. Physiologi-
cal Plantarum 69: 699-703.
Smith, D., L. Muscatine, D. Lewis. 1969. Carbohydrate
movement from autotrophs to heterotrophs in para-
sitic and mutualistic s\mbiosis. Biological Re\ iew 44;
17-90.
Welsh, S. L., N. D. Atwood, L. C. Hic;gins, .and S. Good-
rich. 1987. A Utali flora. Brighani Young University,
Prove, UT. 577 pp.
Received 22 January 1996
Accepted 10 June 1996
Great Basin Naturalist 56(4), © 1996, pp. 341-347
HABITAT AND SPATIAL RELATIONSHIPS OF NESTING
SWAINSON'S HAWKS {BUTEO SWAINSONI) AND RED-TAILED HAWKS
(B. JAMAICENSIS) IN NORTHERN UTAH
Thomas Bosakowski', R. Douglas Ramsey^, and Dwight C. Smitlv^
Abstract. — A total of 28 Swainson's I lawk {liiiteo swainsDiii) and 30 Red-tailed Hawk {B. jainaiccnsis) nests were
found in Cache Valley, Utah, during the summers of 1992 and 1993. All nests were in trees, hut only Hed-tailed Hawks
nested in dead trees (30^). In the intensive study area, ncstinu; densities were 0.10 nests/km- for Swainson's Hawk and
0.08 nests/km- for Red-tailed Hawk. Nearest-neighhor nest distances were significantly shorter among Swainson's
Hawks (1.74 km) than among Red-tailed Hawks (2.83 km). Congeneric nearest-neighhor distances were significantly
shorter tlian conspecific distances for Red-tailed Hawks (1.59 vs. 2.83 km) hut not for Swainson's Hawks (1.52 vs. 1.74
km). CIS analysis of hahitat types was made for 2-km radii around nest sites. Cropland was the dominant land cover
type at nest sites of both species and no significant difference was found between species. Swainson's Hawk nest sites
contained significantly more pasture, whereas Red-tailed Hawk nest sites contained significantly more juniper, maple,
and sagebrush. Only Red-tailed Hawk nests {n — 8; 27%) were found on the periphery of the valley at the base of
foothills of the Cache Mountains. This preference resulted in a significantly higher elevation for Red-tailed Hawk nest
sites. Swainsons Hawk nests occurred only on the valley floor on level teiTain. Distance to the nearest paved road and
building was veiy similar for both species, implying that little difference exists in tolerance levels for human activities.
Overall, multivariate niche overlap for habitat was high (0.89), indicating a lack of habitat partitioning between these 2
Buteos in Cache Valley.
Key words: Swainsons Hawk, Red-tailed Hawk. Buteo, nest sites, liabitat, CIS.
Relatively few studies have included a com-
parison of nest sites, habitat, or densities of
Swainson s Hawks {Buteo swoinsoni) and Red-
tailed Hawks {B. jamaicensis). Rothfels and
Lein (1983) and Bechard et al. (1990) examined
nearest-neighbor distances of these 2 species
in Alberta and Washington, respectively, and
Bechard et al. (1990) also compared habitats.
Janes (1985) examined habitats associated with
sightings of these 2 Buteos in Oregon. Consid-
ering that these species are sympatric through-
out much of their range in western North
America, further information on their habitat
use and nesting density in overlapping regions
would be useful for understanding patterns of
coexistence.
The present study is also important because
the Swainson's Hawk is considered to be de-
clining in Utah, Nevada, and Oregon, and its
status is listed as a "species of special concern "
in Utah, Nevada, Oregon, and Washington, and
"threatened " in California (Harlow and Bloom
1989). Conversely Red-tailed Hawks are con-
sidered to be increasing (Harlow and Bloom
1989) due to an increase in perching habitat, at
the expense of Swainson's Hawks (Janes 1985,
1987). Therefore, a comparative approach to
the nesting ecology of these 2 species is not
only of ecological importance but has implica-
tions for the ftiture conservation of Swainson's
Hawks.
Study Area and Methods
The study was conducted in the Cache Val-
ley portion of Cache County in northern Utah
(Fig. 1). The valley comprises cropland (alfalfa,
hay, winter wheat, corn), pasture, grassland,
marsh, sagebrush-grassland, barnyards/feed-
lots, residential areas, and commercial com-
plexes. During the summers of 1992 and 1993,
we conducted a vehicle survey of the entire
valley by driving on primary and secondary
(dirt) roads along the valley floor and lower
benches. Searches did not extend into moun-
tainous terrain.
lUtah Divisidii of Wildlife Resources, 146.5 W. 200 N., Logan, UT 84321. Present address: Beak Consultants, Inc., 12931 N.E. 12fitli Place, Kirkland. WA
98034-771.5.
^Department of Geography and Earth Resources, Utah State University; Logan, UT 84322.
-^Biology Department. Southern Connecticut State University, New Haven, CT 06515.
341
342
Great Basin Naturalist
[Volume 56
Fig. 1. GIS shaded relief map of the Cache Valley study area in northern Utah showing the distribution of Swainson's
Hawk (circles) and Red-tailed Hawk (triangles) nests for 1992 and 1993 nesting seasons combined. Light lines indicate
primary and secondaiy roads.
Occupied nests were found by scanning
likely trees, especially if adults were seen near-
by or protested our approach. A nest was con-
sidered occupied only if an incubating or
brooding female and/or young were present.
Nests were not examined for the presence of
eggs. Because of the low density of ti^ees (200 m). Habitats
were classified by the Utah Division of Water
Resources (1991) into 11 major habitat types:
cropland, fallow field, grassland, pasture, sage-
brush, juniper {Jiiniperus spp.), maple {Acer
granclidentatinn), riparian (wetlands, temporaiy
marshes, mud flats), open water, residential, and
commercial (non-residential buildings, indus-
trial stiiictures, junkyards, and parking lots).
Statistical analysis was performed on NCSS
software (Number Cruncher Statistical Soft-
ware, Kaysville, UT). Prior to analysis, habitat
variables were tested for normality (D'Agostino
1990). A number of data transformations were
attempted (Zar 1984), but none were able to
normalize all variables. Therefore, a non-para-
metric rank test (Mann-Whitney C/-test, 2-
tailed) was selected for all habitat comparisons.
To calculate habitat overlap from the GIS data
variables, a full-model (all variables included)
discriminant function analysis (DFA) was run
to detemiine the extent of habitat partitioning
between the 2 species. Multivariate niche
overlap for habitat was calculated with log-
transformed variables with the following for-
mula presented by Maurer (1982):
overlap = exp (-d^ /SI + S2);
where d = the difference between mean dis-
criminant scores for species 1 and 2, and S =
the standard deviation of the discriminant
scores. Maurer (1982) and Klopfer and
Ganzhorn (1985) suggested that stepwise pro-
cedures that eliminate variables always result
in a biased underestimation of niclie overlap.
Therefore, we used a full-model DFA instead
of stepwise DFA because it considers the
whole spectrum of habitat variables available
for partitioning.
Results and Discussion
We located 58 occupied nest sites during
1992 and 1993 field seasons: 28 Swainson's
Hawk nests and 30 Red-tailed Hawk nests. In
a single breeding season (1992), a maximum of
22 occupied nests were found for Swainson's
Hawks and 23 for Red-tailed Hawks. M\ nests
were in trees, although a few cliff sites were
available in the study area but not occupied.
Only Red-tailed Hawks nested in dead trees (9
of 30 trees), which was statistically significant
because Swainson's Hawks nested only in live
trees (Fisher Exact Test, 2-tailed, P = 0.002).
Red-tailed Hawks nested higher aboveground
and in taller trees, but tree diameter was not
significantly larger (Table 1).
The intensive study area was completely
searched for occupied nests in 1992 and con-
tained 10 Swainson's Hawk nests and 8 Red-
tailed Hawk nests (Fig. 2). Absolute nesting
density in this area was 0.10 nests/km^ for
Swainson s Hawks and 0.08 nests/km- for Red-
tailed Hawks. Gilmer and Stewart (1984)
reported a nesting density of 0.055 nest/km^
for Swainson's Hawk, which was almost half
the density found in Cache Valley. Luttich et
al. (1971) reported a nesting density of 0.145
red-tailed nestsAni^, which is higher than our
study area. Rothfels and Lein (1983) reported
nesting densities of 0.238 nests/km^ for Swain-
son's and 0.508 nests/km^ for Red-tailed Hawks,
which were much greater than the density for
Swainson's and Red-tailed Hawks in Cache Val-
ley. This difference probably reflects the fact
that hawk nests can be dispersed because of
areas of unsuitable and marginal habitat (e.g.,
note that nests were not located within areas of
dense suburban road networks in Fig. 2).
Newton (1979) stated that in continuously
suitable habitat the nests of the same species
are often separated by roughly equal distances.
Mean nearest-neighbor distance (Clark and
Evans 1954) is the measure that can be used to
344
Great Basin Naturalist
[Volume 56
Table 1. Nest tree and topographic variables of Svvainson s and Red-tailed Hawk nest sites in northern Utali. Data
represent means ± s with sample size in parentheses.
Red-tailed Hawk
Swainson's Hawk
p.
Nest tree height (m)
17.3 ±4.1 (25)
13.9 ± 2.9 (23)
0.001
Nest height (m)
14.8 ± 3.4 (23)
11.3 ±3.3 (21)
0.002
Nest tree DBH (cm)
87.2 ± 39.4 (22)
75.5 ± 44.6 (23)
0.226
Distance to paved road (m)
393.6 ± 580.9 (30)
311.6 ±484.2(23)
0.133
Distance to building (m)
246.1 ±174.3 (30)
250.4 ± 174.2(24)
0.649
Elevation (m)
1401 ± 193.6 (29)
1373 ±31.4 (27)
0.001
^Mann-Wliitnev I'-tt-st, 2-tailfcl
quantify these spacing patterns. In Cache Val-
ley we found a significant difference (Student's
t test, t = 2.61, P < 0.02) for this distance,
which was 1.74 km for Swainson's Hawks {n =
10) and 2.83 km for Red-tailed Hawks {n = 8).
In Alberta, Rothfels and Lein (1983) reported
mean nearest-neighbor distances of 1.46 km
for Swainson's Hawks and 0.88 km for Red-
tailed Hawks. These results are similar to Swciin-
son's Hawks in Cache Valley but are much
shorter than our estimate for Red-tailed Hawks.
Rothfels and Lein (1983) noted that their data
on Red-tailed Hawks showed a much denser
population than nomial. The mean for 7 other
Red-tailed Hawk studies was 1.95 km (data
from Rothfels and Lein 1983), which is closer
to the distance found for Cache Valley. The
mean for 5 other Svvainson s Hawk studies is
1.78 km (data from Rothfels and Lein 1983),
veiy close to the mean for Cache Valley. Over-
all, nearest-neighbor distances from our study
area were consistent with the majorit>' of liter-
ature values, demonstrating the regular disper-
sion of nest sites that results from territorial
behavior (Newton 1979).
Congeneric nearest-neighbor distances were
significantly shorter than conspecific distances
for Red-tailed Hawks (1.59 km vs. 2.83 km) but
not for Swainson's Hawks (1.52 km vs. 1.74 km;
Student's t test, ^ = 2.18, P = 0.047 and t =
0.78, P = 0.44, respectively). These results sug-
gest that Red-tailed Hawks are more tolerant
of close nesting by Swainson's Hawks than their
own species, but Swainson's Hawks are equally
intolerant to congeners and conspecifics. In
Alberta, Schmutz (1977) and Rothfels and Lein
(1983) found that congeneric Buteos nested
closer together than conspecifics probably be-
cause competition among congenerics was less
than among conspecifics.
With regard to distribution in the study area,
only Red-tailed Hawks (27%; n = 8) nested
above the vallev floor at the base of foothills of
tlie Cache Mountains (Fig. 1), and this difference
resulted in a statistically significant increase in
elevation (Table 1). Swainson's Hawks did not
nest in this zone at all, possibly because many
of these sites were already occupied by earlier-
nesting Red-tailed Hawks or because of habitat
preferences discussed below. Rothfels and Lein
(1983) noted qualitatively that Swainson's Hawks
usually nested on flatter tenain than Red-tailed
Hawks. In this study, Swainson's and Red-tailed
Hawk nests lacked a significant difference for
the distance to nearest buildings or paved roads
(Table 1). No previous studies of these 2 species
have been conducted in areas with this much
urbanization. Our data suggest that no signifi-
cant differences exist in regard to tolerance of
human activities and structures.
Overall, the CIS indicated that habitat
around nest areas was dominated by cropland
and pasture for both Buteos (Fig. 3). Swainson's
Hawk nest sites had significantly more pasture
(22.4% vs. 12.3%) but not cropland, fallow field,
or grassland. In eastern Washington, Bechard
et al. (1990) noted that Swainson's Hawks uti-
lized wheatland and grassland more than Red-
tailed Hawks. In this study Red-tailed Hawks
nested in areas with significantly more tree
cover (maple and juniper) and sagebiTish, which
predominated uplands along the edge of the
valley floor. The importance of trees to Red-
tailed Hawks was noted by Houston and
Bechard (1983), who documented the increase
in nesting by this species after the expansion of
trees into the prairie regions of Saskatchewan.
Similarly Knight et al. (1982) found that Red-
tailed Hawks nested exclusively in riverine for-
est land along the Columbia River, even though
suitable cliff nesting areas were available. Janes
(1985) noted that Swainsons Hawks depended
more on aerial foraging and occurred in habi-
tats containing few or no perches. In this study
Red-tailed Hawks probably nested more in tree
habitats because of greater perch availability/
1996J
Svvainson's and Red-tailku Hawk Nesting
345
[B Red-tailed Hawk
m Swainson's Hawk
N
Fig. 2. CIS road map of 100-kni- intensive study area
centered at Logan Municipal Airport sliowing nest site
locations for Swainson's and Red-tailed Hawks during the
1992 breeding season.
use, although other factors such as larger prey
species may also be a factor.
Connell (1980) explained that resource par-
titioning (or low overlap) is due to the "ghost of
competition past" (past competition), which has
created evolutionary changes in morphology
and behavior to avoid cuirent competition. Since
raptors are at the top of the food pyramid and
occur at extremely low breeding densities
(Newton 1979, Scheoner 1984), resources are
likely to be limiting, and high overlap in re-
sources between species is likely to result in
current competition (except in rare cases such
as vole plagues). Despite some significant dif-
ferences for 4 of the 12 habitat types, we calcu-
lated a multivariate (DFA) niche overlap of
0.89 for habitat. Niche theory suggests that
overlap values higher than 0.6 are needed to
cause interspecific competition, while lower
values indicate undemtilization of the resource
continuum resulting in intense intraspecific
competition (see reviews by Bosakowski et al.
1992, Bosakowski and Smith 1992).
Prey overlap data were not collected in our
study area, but Smith and Murphy's (1973)
data from northern Utah showed a high prey
overlap value of 0.80 for Red-tailed and Swain-
son's Hawks (as calculated by Jaksic 1983). In
Montana, Restani (1991) found an even higher
food overlap (0.93) for tliese 2 Biiteos. Consid-
ering the findings of high overlap for food (Jak-
sic 1983, Restani 1991) and habitat (this study),
competition between these Butcos sliould be
expected whenever the species occur in close
proximity. As further evidence, Schnuitz et al.
(1980) found that reproductive performance
was significantly reduced in cases where these
Biiteos nested at close range.
Due to man-made alterations, few of the na-
tive plant communities presently exist in Cache
Valley. Not surprisingly, we did not observe
significant habitat partitioning between these 2
Biiteos for the existing habitat t\pes. Elsewhere,
investigators have claimed that significant habi-
tat partitioning (non-overlap) occuned between
these Buteos in Oregon (Janes 1985) and Wash-
ington (Bechard et al. 1990), but the extent of
habitat overlap was not previously quantified.
Our results indicate that statistical tests can
show differences among several habitat vari-
ables, while the overlap value can still remain
critically high.
Competition for habitat has also been
demonstrated by behavioral observations of
Swainson's Hawks frequently usurping por-
tions of Red-tailed Hawk temtories with lower
perch densities (Janes 1994). Alternately, Janes
(1985, 1987) noted that the increase in perch-
ing habitat, caused by the spread of junipers,
homesteads, and utility poles, "favors the Red-
tailed Hawk at the expense of the Swainson's
Hawk. " In addition, Janes (1994) also reported
that territorial Swainson's Hawks are occasion-
ally displaced by Red-tailed Hawks.
Bednarz (1988) noted that availabilit)' of nest
trees could be a limiting factor for Swainson's
Hawks because of their affinity for open grass-
land and desert habitats that are often devoid
of trees. Similarly, Houston and Bechard (1983)
reported the expansion of Red-tailed Hawks in
Saskatchewan following the spread of trees
into prairie regions. In our study area the west-
ern portions of the valley floor were often tree-
less and usually supported little nesting for
either species (note lower density of nests in
Fig. 1). For such situations Schmutz et al. (1984)
recommended installation of artificial nest plat-
forms for Swainson's Hawks, which signifi-
cantly increased nesting densit>' in his experi-
ments. However, if artificial nest platforms are
used, we recommend caution and close moni-
toring so as not to give advantage to the more
346
Great Basin Naturalist
[Volume 56
Hectares
300 400
700
CROPLAND
Fig. 3. Habitat areas around nest sites (2 km radius) of Svvainsons {n — 26) and Red-tailed Hawks [n = 2S) from Cache
Valley, Utah, as determined from CIS analysis. Bars represent the mean and stars indicate that a significant difference
was observed between species (Mann-Whitney t/-test, 2-tailed, P < 0.05).
common Red-tailed Hawks. In our study area
only Red-tailed Hawks nested in snags (30% of
occupied nests) and may be more likely to use
an open-topped artificial platform than Swain-
son's Hawks, which always nested in green
trees. Many of the snags used by Red-tailed
Hawks in Cache Valley were caused by failure
to irrigate croplands during recent drought
conditions, thus changing the suitability of nest
sites in favor of Red-tailed Hawks.
In the future close attention to inigation and
surveillance of land-use changes are likely to
be the most important factors in conserving
Swainson's Hawks in Cache Valley. Economic
conversion of agrarian land use to commercial
and residential real estate is currently in pro-
gress, and impacts to future Swainson s Hawk
populations need to be carefully monitored.
Due to the rapid human population growth in
Cache Valley we recommend annual monitor-
ing for Swainson's Hawk territories and nests,
which may be impacted by future development
or land-use changes. This monitoring manage-
ment will recjuire frequent updating of the CIS
database to track habitat changes in the futuie
so that necessary mitigation steps can be evalu-
ated.
Acknowledgments
This study was entirely a volunteer effort
completely funded by the authors except for
the use of the CIS system at Utah State Uni-
versity. We thank J. C. Bednarz, M. J. Bechard,
and R. C. Whitmore for rexiewang the manu-
script and providing helpful comments and
criticisms.
Literature Cited
Bechard, M. J., R. L. Knight, D. G. Snhth, and R. E.
FiTZNER. 1990. Nest sites and habitats of sympatric
hawks {Bitten spp.) in Washington. Journal of Field
Ornitholog) 61: 159-170.
Bednarz, J. CI 1988. Swainson's Hawk. Pages 87-96 in
Proceedings of the southwest raptor management
SNinposiiun. National Wildlife Federation, Scientific
and Technical Series No. 11. Washington, DC.
BosAKOwsKi, T, and D. G. Snhth. 1992. Comparative
diets of SNinpatric nesting raptors in the eastern
1996]
Swainson's and Rkd-tailkd Hawk Nesting
347
(lec'icluoiis forest l)ioini\ Canadian [oninal of ZooIogN'
70: 9S4-992.
BosAKOWSKi, T, D. G. Smuii, and K. Spkiskh. 1992. Niche
o\erlap of tvvo sympatiic-nt'stinK liavvks Accipiter spp.
in tlic New Jersey-New York llitililaiuls. I-lcoi^rapln
15: 358-372.
CUAHK, P [., AND E C. Evans. 1954. Distance to nearest
neighbors as a measure of spatial relationships in
populations. Ecology' 35: 445-453.
CONNELL, J. H. 1980. Diversity and tlie cocNolntion of
competition, or the ghost of com|ietition past. Oikos
35; 131-138.
Cr.'\ighe.\d, J. J., .\NU E C. Ckak:iieai), Jk. 1956. Hawks,
owls and wildlife. Stackpole Publishing Co., Harris-
burg, PA.
D'Agostino, R. B. 1990. A suggestion for using powerful
and informative tests of normality. American Statisti-
cian 44: 316-322.
FiTZNER, R. E. 1980. Behavioral ecolog> of the Swainson's
Hawk in southeastern Washington. Unpublished doc-
toral dissertation, Washington State University; Pull-
man. 194 pp.
GiLjMER, D. S., and R. E. Stewart. 1984. Swainson's Hawk
nesting ecology in North Dakota. Condor 86: 12-18.
Harlow, D. L., and R H. Bloom. 1989. Biiteos and Golden
Eagle. Pages 102-110 in Proceedings of the western
raptor management symposium and workshop.
National Wildlife Federation, Washington, DC.
Houston, C. S., and M. J. Bechard. 1983. Trees and the
Red-tailed Hawk in southern Saskatchewan. Blue Jay
41: 99-109.
Jaksic, E M. 1983. The trophic structure of sympatric
assemblages of diurnal and nocturnal birds of prey.
American Midland Naturalist 109: 152-162.
Janes, S. W 1985. Habitat selection in raptorial birds.
Pages 159-190 in M. L. Cody, editor. Habitat selec-
tion in birds. Academic Press, Inc., Orlando, EL.
. 1987. Status and decline of of Swainsons Hawks in
Oregon: the role of habitat and interspecific competi-
tion. Oregon Birds 13: 165-179.
. 1994. Partial loss of Red-tailed Hawk territories to
Swainson's Hawks; relations to habitat. Condor 96:
52-57.
Klopfer, P H., and J. H. Ganzhorn. 1985. Habitat selec-
tion; behavioral aspects. Pages 436-454 in M. L. Cody,
editor. Habitat selection in birds. Academic Press,
Inc., Orlando, EL.
Knight, R. L., D. G. S.\irni, and A. Ehickson. 1982. Nest-
ing raptors along the Columbia River in ncjrth-cen-
tral Washington. Murrelct 63: 2-8.
LurnGii, S. N., L. B. Keith, and J. D. Stephenson. 1971.
Population dynamics of the Red-tailed Hawk (Btileo
janiaiccnsis) at Rochester, Albc-rta. Auk 88: 75-87.
Maurek, B. a. 1982. Statistical inference for MacArthur-
Levins niche overlap. Ecology 63: 1712-1719.
Newton, I. 1979. i\)pulation ecology of raptors. Buteo
Books, Vermillion, SD. 399 pp.
Restani, M. 1991. Resource partitioning among three Bitlco
species in the Centennial Vallev, Montana. Condor
93: 1007-1010.
ROTHFELS, M., AND M. R. Lein. 1983. Territoriality in sym-
patiic populations of Red-tailc-d I lawks and Swainson's
Hawks. Canadian Journal of Zoolog\' 61: 60-64.
Schoener, T. W 1984. Size differences among sympatric,
bird-eating hawks: a worldwide siuA'cy. Pages 254—281
in D. R. Strong et al., editors. Ecological communi-
ties: conceptual issues and evidence. Princeton Uni-
versity Press, Princeton, NJ 613 pp.
SCHMUTZ, J. K. 1977. Relationships between three species
of the genus Buteo (Aves) coexisting in the prairie-
parkland ecotone of southeastern Alberta. Unpublished
master's thesis, University of Alberta, Edmonton.
ScHMUTZ, J. K., D. A. Moore, and A. R. Smith. 1984. Arti-
ficial nests for Ferruginous and Swainson's Hawks.
Journal of Wildlife Management 48: 1009-1013.
ScHMUTZ, J. K., S. M. Sghmutz, and D. a. Boag. 1980.
Coexistence of three species of hawks {Buteo spp.) in
the prairie-parkland ecotone. Canadian Journal of
Zoology 58; 1075-1089.
U.S. Department of Interior. 1979. Snake River Birds of
Prey special research report. USDI Bureau of Land
Management, Boise, ID. 142 pp.
Utah Division of Water Resodrces. 1991. Water-related
land use inventoiy report of the Bear River river
basin (Utah portion). Utah Department of Natural
Resources, Salt Lake City, UT.
Zar, J. H. 1984. Biostatistical analysis. Prentice-Hall, Engle-
wood Cliffs, NJ.
Received 12 July 1995
Accepted 30 May 1996
Great Basin Xatxinilist 56(4). © 1996, pp. 348-359
WESTERN BALSAM BARK BEETLE, DRYOCOETES COXFLSUS SWAINE,
FLIGHT PERIODICITY IX NORTHERN UTAH
E. Matthew Hansen^
Abstract. — The flight periodicit>' of western balsam bark beetle {Dryocoetes confusits Swaine) in Big Cottonwood
Can\'on, Utali, was studied during tlie summer months of 1992, 1993, and 1994. Contents of baited funnel traps uere
tallied b\' species up to 3 times weekh'. Two main periods of flight acti\it>- were obsened each >'ear. The first and, gener-
ally; largest occun^ed in eaih' summer soon after flight was initiated for the season. A 2nd period was obsened in late
summer, generalK .\ugust. Timing of die 2 periods was influenced b\' imusualK wann or cool weather in each stud\'
year. The 1st period had more niiiles than females while the 2nd period had a majority of females. E.vcept during periods
of cool or wet weaUier, western balsam bark beedes were found to be active at least at mininiiil levels fioni June through
September
Key words: Dnocoetes confusus, flidit periodicity. Scolyfidae. insect control, insect plienology. .\bies lasiocaipa,
Utah forests.
The western balsam bark beetle. Dri/ococtes
confusits S\\aine (Coleoptera: Scol\ tidae), is a
serious insect pest of tiTie firs. This insects life
cycle is not full\ understood (Johnson 1985),
howe\er, possibK due to the traditionalK low
commercial \ alue of its host. In British Colmn-
bia, for example, timber losses from western
biilsiuii bark beetles ha\e onb' relati\eK' recently
been calculated (Doidge 1981). The need to
understand the life c\ cle and beha\ ior of this
bai^k beede has increased in relation to die in-
creased commercial and aestlietic \iilue of titie
firs.
Drought-subjected subalpine tin Abies hisio-
ctirpa (Hook.) Nutt.. in northern Utah has been
experiencing a western balsam bark beetle out-
break that began in 1989. Most of the iiffected
trees cu-e on the hea\ il> \ isited \\'asatch-Cache
Nationid Forest, including die can\oiis east of
Salt Liike Cit\- where picnic ai-eas, campgiounds,
and ski resorts are common. This caused local
forest managers to seek beede abatement mea-
sures. Baik beede conti'ol sti-ategies, such as de-
plo\ nient of semiochemicals or cultural treat-
ments, rec^uire know ledge of the time frame in
which beetles emerge fi-oni infested host mate-
rial to attack new hosts. This is a report of 3 \r
of western biilsam bark beetle flight periodicitx
data from Big Cottonwood Can\on. Utah. Sex
ratios, weather influences, and associated scoly-
tids are also presented.
Materials and Methods
Fi\ e plots w ere established on .5—6 June 1992
in Big Cottonwood Canyon, Utah, ranging fiom
2000 to 2840 m ele\ation. Plots were selected
from areas of recent beetle actixit)' indicated
b\ lading or red subalpine fir crowns.
The plot at 2000 m has a white fir [Abies
concolor [Cord and Glend.] Lindl. ex Hildebr.)/
Douglas-fir (Pseiidoisuga )nciizicsii [Mirb.]
Franco) mix and also had the least amount of
fading host material of an\ plot. This is gener-
alK the low er ele\ ational band for subalpine fir
in Big Cottonwood Can>on. Small amoimts of
subalpine fir Ciui be found immediateK" uphill
fi"om the plot. The higher plots are dominated
b> subalpine fir sometimes in association with
Douglas-fir tjuiiking aspen (Popuhis trcnwloides
Michx.), or Engelmann spruce [Picca cngcl-
)nannii Pam" ex Engelm.).
Each plot contained three 16-imit Lindgien
fimnel traps- spaced at about 50-m inten ids in
a triangular pattern. Traps were baited with a
semiochemical mixture containing <^vo-bre^ i-
coniin (racemic) released at 1 mg/24 h at 24 °C'^
(Borden et al. 1987). Traps were hung as high
'I SDA Forest Senici-. Iiitt-nuoiintain Rese;irch Station. S60 Nortli 12(K) East. U>,aan. ITS4321.
-Phero Tech Inc.. Delta. B.C.. Canada,
3phero Tech Inc.. Delta. B.C. Ciuiada,
348
1996]
Dryocoetes coxfusus Flight Periodicity
349
as possible on branches, lea\ inu tlie trap eup
about 1.5 ni abcneground. Trap eups were
emptied up to 3 times weekly to reduce losses
to predation. Cups w^ere emptied less fiequentK-
late in each stud\- \ear as captures diminished.
Western balsam bark beetles were tallied aloni:;
with associated scoK tids and important preda-
tors, nameh' checkered beedes {Enoclerius spp.).
Identification of associated scoK tids was pro-
\ided b>" Stephen L. Wbod^. D. coufusus cap-
tures for the entire season were tlien totaled
for each plot. The percentage of the annual
total caught at each obsen ation w as then plot-
ted against date for each location.
The stud\' was repeated starting in mid-Ma\
1993 with 5 plots installed from 1750 to 2840
m ele\'ation. Two sites from 1992 were reused,
2 were mo\ ed a short distance (about 100 m
horizontal), and 1 was new. The lowest ele\a-
tion plot was deliberately established in the
white fir zone. Low^-elevation plots were in-
stalled earlier in the \ear than in 1992 to ensure
placement before beetle flight commenced.
Plots at 1750 and 2350 m were installed 25
Ma\- 1993, and the plot at 2560 m was estab-
lished 27 Max 1993. The plot at 2660 m was
instiilled on 8 June 1993 and the plot at 2840 m
on 21 June 1993.
The 1994 flight periodicit) stud\ utilized
the 4 highest sites from 1992, ranging in ele\a-
tion from 2310 to 2840 m. These areas contin-
ued to contain fading host material throughout
each stud\ \ ear The low-elex ation sites, lacking
a substantial sulialpine fir component, were
dropped from the stud\' due to the small popu-
lations of D. confiisus in those areas. The plot
at 2310 m was established on 10 May 1994,
while the remaining plots were instiilled on 25
Ma\ 1994. This gi\es 3 consecuti\e >ears of
flight period data for 4 locations.
The first 10 D. confiisus from each trap cup
obsenation, totaling 30 per plot, were tiillied for
sex in 1993 and 1994. Females were identified
b\' a prominent setiU biiish on the frons (Borden
et al. 1987). For 1993 and 1994 the sex ratio of
each distinct flight surge was compared. The
di\'ision between surges was determined from
each significant flight activity' lapse not associ-
ated with cool or wet weather.
Weather data from Brighton-SiKer Lake'^
was compared with flight acti\ it> for the plots
at 2600/2660 m and 2840 m (this station is geo-
graphicalK' and elevationalK' between the 2
plots). Daily maximum/minimum temperatures
and dail\- precipitation were plotted from 20
Ma\- dirough 31 October for each stud\' year.
Results
Flight Periodicity
1750 Meters. — This site was used onl\ in
1993 with a total of 42 D. confiisus captures.
ConsequentK; I deleted it from consideration
for the puipose of this study. Nearb\ white fir
mortalit) that was examined contained evi-
dence onh' of fir engraver beetle, Scolytus ven-
tralis LeConte.
2000 Meters. — Because this site had rela-
ti\el> few captures, I used it only in 1992.
Flight acti\ it>' for that year sharpK peaked in
mid- to late June (Fig. 1). A substmitiiilK- smaller
surge occurred in early August. D. confusus
captures totaled 1469. The 1st wa\e of acti\it\^
accounted for 84% of total captures.
2310/2350 Meters.— The substantial num-
ber of captures at the first obsenation of 8
June 1992 indicates diat flight was likeK- initi-
ated before plot establishment (Fig. 2). Cap-
tures peaked in mid- June with activity contin-
uing throughout the month. A 2nd surge began
in mid-July and tapered off in mid-August. D.
confusus captures totaled 19,071. Fort)-one
percent of the total occurred in the 1st surge.
In 1993 traps at this plot began to capture
beedes in mid-June widi a sniiill peak occuning
in late June. A 2nd wave of captures began in
late Jub, peaking in mid- to late August. D.
confusus captures were about 9% of those in
1992, totaling 1800. The 1st wa\e of captures
accounted for 10% of the totiil.
Capture patteiTis of 1994 were very similar
to those of 1992. The first positi\e obsen'ation
was on 3 June 1994. An early sunuuer peak
occuned on 13 June 1994 with acti\it\^ taper-
ing off in late June through earl\- JuK'. A late-
summer surge occuned in mid- to late July with
captmes gradualb' diiuinishing through early
October. D. confusus captures totaled 2574,
with 30% caught in the 1st surge.
2560 Meters.— The first 1992 obsenation
was positi\e, indicating that flight w as possibb'
initiated before plot establishment. Beetles
■'Prottssor i-meritus. Life Science Museum and Department of Zoolog\, Brighani Young L'niversit)', Provo, UT.
-^Salt Lake Cit\ Watershed .Management.
350
Great Basin Naturalist
[Volume 56
50
I 30
S20 +
§10
0)
■ ■I
20 6/1
15
7/1 15
8/1 15
1992
9/1 15
10/1 15
31
Fig. 1. Percentage of total (;; = 1469) seasonal Dnjococtes conjmm captures per obsei-vation at 2000 m, 20 May-31
October Arrow indicates the trough between surges not associated with cool or wet weather
x,20
10
20 6/1
20
Q. iK
s
V)
(U
^ 10
20
15
7/1 15
8/1 15
1992
9/1 15 10/1 15
31
■ ■ ll illl .1 111
20 6/1
15
7/1 15
8/1 15
1993
9/1 15 10/1 15
31
X3
0>
gist
V)
(U
Si
5 -
^^
I I I
20 6/1 15 7/1 15
8/1 15
1994
9/1
15 10/1 15
31
Fig. 2. Percentage of total (nigc), = 19,071, n]yy3 = 1800, /i,c)94 = 2574) seasonal Dnjococtes confii,siis captures per
observation at 2310/2350 n», 20 May-31 October. Arrow indicates die trough between surges not associated with cool or
wet weather
1996J
DmOCOETES CONFUSUS FL1(;11T PERIOIDlCITi'
351
were eau^ht in large numbers throughout June
with a 2ncl surge of aetivity in early to mid-
August (Fig. 3). D. confiisiis trap eaptures totaled
9164, with 669r captured in Hie 1st surge.
In 1993 eaptures began in late June, peak-
ing in early to mid- July. A 2nd, substantially
larger wa\e started in late July and peaked
fi-om mid-August through early September.
Captures were about 41% of those in 1992,
totaling 3763. Twelve percent of that total were
caught in the 1st surge.
The pattern of 1994 captures was similar to
that of 1992 with a sharp peak occurring in
mid- to late June. A 2nd wave began in late
July, peaking in early August. D. confusus cap-
tures totaled 4476, half of which were caught
in the 1st surge.
2600/2660 Meters.— 1992 captures began
in mid- June with a sharp peak occurring in
late June (Fig. 4). A 2nd wave began in late
July with a mid-August peak. D. confusm cap-
tures totaled 7548, with 68% caught in the 1st
surge.
In 1993 activity began in late June with an
early summer peak in mid-July A 2nd, substan-
tially larger wave started in late July and peaked
from late August through early September D.
confusus captures totaled 5882, 16% coming in
the 1st surge.
In 1994 activity began in early June with a
shaip peak in late June. A 2nd wave began in
late July, peaking in early August. Captures were
the fewest for any study year, totaling 1331.
^30
3 25 f
S2O
CO
0)
ii993 = 5882, /iiyg^ = 1331) seasonal Dnjocoetes confiisus captures per
obsei-vation at 2600/2660 in, 20 May-31 October. Airow indicates the trough between surges not associated with cool or
wet weather.
Si.xty percent of these were caught in the 1st
surge.
2840 Meters.— In 1992 the 1st capture
was obsei-ved on 22 June. Captures peaked in
late June, and considerable activity continued
tlirough early July (Fig. 5). A 2nd surge occuned
in mid- to late August. Captures totaled 17,542
with 72% caught in the 1st surge.
In 1993 activity began in early July with a
very large peak occurring in mid- to late July.
A 2nd surge occurred in mid- to late August.
Captures were down fiom 1992 levels but were
still substantial, totaling 10,344. Seventy-six
percent of these were caught in the 1st surge.
In 1994 flight initiated in mid-June with a
distinct spike in late June. A late-summer surge
began in late July and continued through mid-
August. Captures were the greatest of any plot
in an\' year, totaling 20,600. Sixty-seven per-
cent were caught in the 1st surge.
Surge Activity
Considering onb' the 4 plots common to
each study year, there is a trend for the 1st
surge to be larger than the 2nd w ith increasing
elevation (Table 1). The lowest ele\ation plot
consistentK' captured more beetles in the 2nd
wave. The highest plot, however, consistentK
captured more beetles in the 1st surge.
Weather Influences
Periods of cold and/or wet weather coin-
cided with a reduction or pause in beetle
1996]
Dryocoetes coneusvs Flicht PERionicnv
353
■o30
0)
= 25 -
820
CO
0)
0)
£10
i 5 +
0)
Q.
30
= 25 I
820
CO
0)
^15
Z 10 -
£? 5
30
B 25
8 20
^15
y 5
CL „
■
1
r
L
,
1 n. . 1.
1 1 .
20 6/1
15
7/1 15
8/1 15
1992
9/1 15 10/1 15
31
■
•
■
1
f
•
II
.1
20 6/1
15
7/1 15
8/1 15
1993
9/1 15 10/1 15
31
20 6/1
15
7/1 15
8/1 15
1994
9/1 15 10/1 15
31
Fig. 5. Percentage of total (/!i9g2 = 17,542, 711993 = 10,344, nig94 = 20,600) seasonal Dryocoetes confusu.s captnres per
observation at 2840 m, 20 May-31 October. Arrow indicates the trough between surges not associated with cool or wet
weather.
captures (Figs. 6-8). Ver>' little flight occurred
when daily maximum temperatures were less
than 15 °C, confimiing Stock's (1991) findings.
Lapses in flight activity between the main
surges, however, are not necessarily associated
with cool or wet weather. With 4-7 wk
between peaks, warm, diy days were available
during these spans of reduced flight.
One would e.xpect delayed emergence and
flight timing with increasing elevation. Initial
captures at 2840 m were about 2-3 wk later
than at 2310/2350 m each year. Timing of peak
flight activity' was similarly delayed with increas-
ing elevation (Figs. 2-5).
Febmary through May 1992 was the wannest
on record for that period in northern Utah.
June through August 1993 was the coolest on
record while June through August 1994 was
the warmest. The warm spring of 1992 coin-
cided with an earlier than expected flight com-
mencement. D. cunfustis were likely flying be-
fore ti-ap placement, possibly as eaily as late May
at lower elevations. In contrast, the snowy win-
ter and spring of 1993 followed by a record cool
summer resulted in a delayed beetle flight. In
1992 D. confimis were first captured at 2840 ni
on 10 June compared with 6 July in 1993. In
each year, regardless of the overall weather
regime, I obsei-ved that flight did not initiate
until the local snowpack was mostK melted
and that the early summer peak occurred after
all snow patches were gone.
354
Great Basin Naturalist
[Volume 56
Sex Ratio
The early summer surge typically had a
higher portion of male beetles (Table 2). Males
were especially doininant during the first 5—10 d
of emergence, comprising nearly all of those
sampled. The sex ratio tlien became more evenly
mixed for the remainder of the early summer,
including during peak activity. The late-sum-
mer surge was dominated by females in each
year, the ratio being more stable throughout
the period.
Secondaiy Scolytids and Predators
Other scolytids captured include GiiatJiotri-
chiis sulcatus LeConte (ambrosia beetle), Pityok-
teines minutus Swaine, Xylechiniis montanus
Blackman, Hylastes suhopacus Blackman, Scoly-
Table 1. Percentage of total seasonal beetle captures
per plot occuning in the 1st surge.
Year
1992
199.3
1994
Plot
1st surge
(elev. [m])
(%)
2310/2.350
41
2.560
66
2600/2660
68
2840
72
2310/2350
10
2560
12
2600/2660
16
2840
76
2310/2350
30
2560
50
2600/2660
60
2840
67
20 6/1 15 7/1 15 8/1 15 9/1 15 10/1 15 31
^^1 mm prec Lotemp Hi temp
■o30
3 25
Q.
M
0)
e^o
c
<0
2 5
0)
^ 0
20 6/1 15 7/1 15 8/1 15 9/1 15 10/1 15 31
2840m
Fig. 6. 1992 daily weather conditions at Brighton-Silver Lake (2700 ni) with Dnjocoetc.s confiisiis flight acti\it\- at
nearby plots. Arrow indicates the trough bcKveeu surges not associated with cool or wet weather. Note the lag effect,
resulting from 2- to 5-d observation intcnals, which can gi\c the illusion of (hght acti\it\ during acKerse weather
•
f
1
1 ,,. . 1.
1 1 ,
1996]
DRYOCOETES CONEUSUti FL1C;HT Pl^l^lODlCITi'
355
30
5 25 -
820
15
r 10 -
i 5
(U
20 6/1 15 7/1 15 8/1 15 9/1 15 10/1 15 31
2660m
20 6/1 15 7/1 15 8/1 15 9/1 15 10/1 15 31
^^H mm prec Lo temp Hi temp
0)
525
s
20 -
^ 15
i:io
c
Ji 5
0
°- 0
•
I
f
■
II
1
20 6/1 15 7/1 15 8/1 15 9/1 15 10/1 15 31
2840m
Fig. 7. 1993 daih' weather conditions at Brigliton-Silver Lake (2700 m) with Dnjocoetes confusiis fliglit acti\it\- at
nearby plots. Arrow indicates the trough between surges not associated with cool or wet weather Note the lag effect,
resulting from 2- to 5-d obsei-vation intenals, which can give the illusion of flight activit)' during adverse weather.
tus ventralis LeConte (fir engraver beetle), Ips
spp., Cryphahis ritficollis Hopkins, Hyhirgops
porosus (LeConte), Scohjtiis opacus Blaekman,
Dryocoetes affaher (Mannerheim), and Dry-
ocoetes sechelti Swaine. Checkered beetles,
Enoclerus spp., were the most common and
important predaceous insect trapped. Other
predators captured include snakeflies (Raphidi-
idae) and rhizophagids. Gnathotrichus siilcatus,
Xylechinus montanus, Hylastcs siihopaciis , Dryo-
coetes ajfaber, Enoclerus spp., and rhizo-
phagids were caught in sufficient numbers to
suggest that they possibly cue on exo-brevi-
comin.
Clerids, or checkered beetles {Enoclerus
spp.), were captured before western balsam
bark beetle flight commencement in each year.
Captures generally peaked in mid- to late June,
then tapered off through August. Few clerids
were captured later than the end of August in
any year though D. confusus continued to be
active. At the 4 plot locations common to each
study year, a total of 242 clerids were caught in
1992, 357 in 1993, and 307 in 1994.
Discussion
Flight Periodicity
Western balsam bark beetles were caught
throughout the summer months for all study
years. Beede flight typically started in June
and continued well into September with a few
captures as late as early October. Once flight
was initiated, only cool or wet weather could
356
Great Basin Naturalist
[Volume 56
30
B25 -
8 20--
M
^ 15
ifiof
c
S^ 5
I. I ■ I I I I I ll I I I I , I ■ I
20 6/1 15 7/1 15 8/1 15 9/1 15 10/1 15 31
2600m
20 6/1 15 7/1 15 8/1 15 9/1 15 10/1 15 31
^^1 mm prec Lo temp Hi temp
■o30
3 25
S20
(/>
(U
^15
f 10
c
0)
^ 5
0)
°- 0
20 6/1 15 7/1 15 8/1 15 9/1 15 10/1 15 31
2840m
Fig. 8. 1994 daily weather conditions at Brighton-Silver Lake (2700 m) with Dryococtes confusiis flight activity' at
nearby plots. Arrow indicates the trough between surges not associated with cool or wet weather. Note the lag effect,
resulting from 2- to 5-d obsei-vation intei-vals, which can give the illusion of flight activity during adverse weather.
completely curtail it. Some warm, diy periods,
however, had dramatically reduced activity rel-
ative to the peaks. At each plot 2 distinct peaks
of flight activity were seen every year. Some
plots appeared to have a 3rd peak in Septem-
ber of 1992 and 1994, but this can be attrib-
uted to a reduction in trap checking fi-equency
at those times. Generally, the 1st peak was
sharp and occurred within 2-3 wk of initial
emergence. A 2nd peak was obsei'ved 4-7 wk
after the 1st.
This is similar to Stock's (1991) findings in
British Columbia where there "were two major
flight periods per year, the first commencing in
mid- to late June, and the second in mid- to
late August." Stock found this flight period to
correspond well with the life cycle described
by Mathers (1931).
Mathers (1931) studied the western balsam
bark beetle life cycle at Stanley, B.C., using
caged subalpine fir bolts. Young adults were
found to emerge and attack fresh trees in June
and July. Eggs were laid through August before
parents commenced feeding in tunnels before
overwintering. Galleries were advanced the
following June and July with continued egg lay-
ing. Parents then reemerged in July to attack a
fresh host with a 3rd set of brood tunnels exca-
vated, eggs being laid through August. Math-
ers concluded this to be the end of the life
cycle. Eggs from the 1st brood ovenvintered as
lai^vae before pupating the following sunnner.
The 2nd winter was passed as >()ung adults
that emerged to attack fresh trees the following
June and July. Eggs from the 2nd and 3rd
broods overwintered as lanae that pupated the
1996]
Dryocoetes confusus Flight Periodicity
357
Table 2. Percentage of females amonc; sampled beetles
in early and late-smnmer surges.
1993
1994
Plot
1st sur
Se
2nd sui
Ke
1st sm-
Se
2nd singe
(elev. [m])
{%)
{7c)
{%)
(%)
2310/2350
47
74
38
51
2560
46
71
29
(K)
2600/2660
34
75
28
(i6
2840
52
64
44
60
following year and emerged as adults the next
season. Gixen this life cycle, one would expect
to find all life stages represented in any given
year. Stock (1991) suggests that the August
peak obsei"\'ed in his study corresponds to the
July reemergence described b>' Mathers, the
timing difference due to warmer conditions at
Mathers site. Using Mathers life cycle, the 2
flight peaks must be of different generations,
the June peak being of newly emerged young
adults and the August peak being of reemerged
2nd-yr adults. This gives a 3-yr life cycle when
the reemergence year is included. Beetles fi-om
eggs laid in 1991, for example, presumably
might not complete their life cycle until 1994.
Bright (1963) suggests that parent beetles
may die in their 1st or 2nd brood tunnels be-
fore reemerging to attack another host. This
would account for the late-summer surge often
having less activit)' tlian tlie early summer surge.
Bright also believes D. confusus to have a 1-yr
life cycle in the western United States, the life
cycle proposed by Mathers (1931) being a phe-
nomenon restricted to the insect's northern
range. No data or references, however, are cited
for this assertion.
The flight period data presented here pro-
duced noticeably different results each year.
This is almost certainly associated with the
record-setting weather regimes seen each study
year. The double peak pattern was not as evi-
dent during the cool, wet summer of 1993.
Only die highest elevation plot exliibited a large,
early summer peak, July in this case. The lower
elevation plots had a noticeably reduced early
summer peak. The greatest activity at these
plots occurred in late August. The record cool
summer weather likely caused the delayed
emergence seen in 1993, but this does not ex-
plain the diminishment of the early summer
peak. Even with delayed development, I ex-
pected considerable activity once flight was
initiated. Cold weather during emergence was
explored as a possible cause for the reduction
in the early summer peak. Night temperatures
at 2700 m dropped to -4°C on 24 June 1993,
which could have killed some new adults and
lurthei- delayed the early summer flight (Bar-
bara Bentz^ personal communication). Perhaps
development of some new adults was delayed
such that their emergence overlapped with
that of reemerged adults in the late summer
At the highest plot, Mafliers' (1931) hypoth-
esized life cycle corresponds well with the data.
The 2nd and 3rd broods described b\' Vlathers
would assure an early summer surge each year
even though 2 yr is required for sexual matu-
rity. In other words, there is no "off year " such
as with the 2-yr life cycle of the spmce beetle,
Dcndroctomis nifipennis (Kirb\').
Some results from this study did not match
Mathers' (1931) life cycle as well as did Stock's
(1991). For example, assuming that the same
local population was sampled each year, adults
in the early summer flight of 1993 should be
represented again in the late-summer flight of
1994. Allowing for some degree of mortality, I
expected the late-summer surge to have fewer
beetles than the early summer surge of the
previous year for a given location. This study
produced 2 examples where the late- summer
surge contained several times more beetles tiian
the early summer surge of the previous year. If,
in fact, these plots did sample the same popu-
lations each year, then Mathers' (1931) life
cycle may not be accurate for northern Utah.
Though funnel trap captures are not appro-
priate for adequately describing a life cycle,
there are several possible explanations for the
unexpected results at the 3 lower plots. The
record-setting weather regimes in each year
would have certainly affected beetle phenol-
ogy. Perhaps some critical thresholds were not
achieved in 1993, resulting in retarded de\'el-
opment or mortality (Barbara Bentz personal
communication). This may have affected young
adults more so than reemerging adults. Con-
versely, record warm weather in 1992 and 1994
could have advanced development. Perhaps the
2nd and 3rd broods are not important in the
overall life cycle. Amman and Bartos (1991)
found reemerged mountain pine beetle,
Dendroctonus ponderosae Hopkins, females to
^Entomologist, USDA Forest Sen'ice, Intennountain Research Station, Logan, UT.
358
C'UU- AT IvVSlN \ ATI KAl 1ST
[NoUinie 56
produce siiiiiiticaulK tow or oti^^priui:; than now
temiUes with nialos. Forhaps lower elevation
beetles tend to ha\ e a l-\ r lite e\ ele while 2 \ r
is required for beetles at hiiiher oloxations.
Pivlxiblx some ciMubination ot tliese tiictors plus
soiue not explored, sueh as disease and preda-
tion, e».>ntributed to the results.
Sex Ratio
Stock (^IQQl'i found that the late-sunnner
sui"ge is comprised larvioK of females. I sinu:
AuiTust 1 to distiuiTuish 1st and 2nd tlight.
Stock found femiiles to comprise 48*^. 29^,
and 50^ of the 1st tlidit dining 3 consecutive
\-eai-s. The 2ud flight had S0'>. 4S^f . imd GO-^
leniiiles. I found simiUu" tivnds in noitheni I tali.
This predominance b\ feniiiles during 2nd
tlight is t\pic;il tor other scol\ tids tliat exhibit
ivemergence vAnderbrandt et al. 19S5. Flamm
et al. 19S71 This suggests diat tiie late-summer
sui-ge, in fact, comprises reemerged adults.
The dominimce of males dining the initial da\s
of eaiK' summer emergence suggests tliat die\
iU-e likely i-esponsible for host selection and
mate atti-action.
\\Vatlier Influences
Stock 1991^ found tlie majorit>" of western
balsam biuk beetle tlight to occur when tlie
d;ul> maximum temperatme was greater dimi
15 *C. The Siune tivnd is seen hei^e. Periixls ot
cool weather especi;illy when coupled widi
precipitation. essentiiilK stopped beetle cap-
tures. Given w^mn, di"\ da\ s. D. confiisus was
found to be active as late as earh October.
all>eit in greatb reducx^d numbei-s. An\ ctmbol
strategx will need to cxinsider diis extended
flight period.
SinprisingK, tlie cooL wet summer of 1993
feiiled to ha\ e ;m\ obxions etlect on die 1994
beetle population other th;m to dela\ emer-
gence. I h\podiesized diat diis w eadier pattern
\\X)idd ha\ e inciviused lieede morhilit). resulting
in fewer trap captures in 1994. Elsewhere in
the region, mountain pine beede Ian ae were
observ ed widi ivhuxied de\ elopment. possibK
leaving o\ ervvintering life stages more xailnera-
ble to cold weather mortUit\. Assuming a 2-\t
life cycle for western IxUsiun Ixu-k beede. per-
haps the more cxild-susceptible life stages would
not have emerged until June 1995. It is also
possible that winter temperatures in 1993-94
did not reach lethal le\ els.
\\ hilo i\u"K (.Mnergeneo was associated with
a record warm spring in 1992, die earl\ sum-
mer peak was no earlier than after the more
topical spring of 1994. Timing of the earh" sum-
mer peak for each plot occurred on essentiidh"
the same date in 1992 and 1994. Timing of die
late-summer peak, how o\ er. was about 2 wk
earlier dining the reeord warm summer of
1994 than the more h pieal summer ot 1992.
CONCI.ISION
Once flight begins, actixih t\pie;ill\ builds to
a sliaip peak within 2 wk. This genenilK occurs
fixim mid-June dirough emK" JuK. Acti\it> dien
subsides before building to a 2nd peak 4-7 wk
later. usuiilK in August. Significant acti\it\ CiUi
cxmtinue into earl\ September with some bee-
des fixing as late iis eaiK October
Cultural or semiochemicid iniuiagement of
western balsam bark beetle will need to con-
sider the double peak flight pattern of piu^ent
beetles luid die tact that adult beedes can be
found in some quimtit> diroughout die wiuiiier
months. Remoxid of infested host materiid. for
example, should be done in the tdl w hen flight
is complete. Anti-aggiegation pheromones will
need to be foniiulated to efliuse up to 4 nion or,
possibK. applied twice per seiison.
Further resernxdi is needed to confinii or re-
vise die life c>des described b\ Madiei^s il931)
and suggested by Bright il963i since control
sti-ategies for a 1-\t life cycle can be different
from a 2- or 3-\t life c\ ele. Considering die 1st
versus 2nd surge difterences. diis should be
done for a nmge of eleN"ations. The role of re-
emerged adults in brood production must be
detennined for a more complete understimd-
ing of die oxenill life c> ele. The conditions lead-
ing to iUi outbre;dc also need to be quantified
such diat cidtural guidelines can be estabUshed.
ACKNO\\T_EDGM ENTS
This pivject w:is initiated by Ste\e Munson
• Forest Hciddi Protection. Ogden. VT). Criti-
Cid re\iew of die manuscript w iis proxided b\'
BiU-bara Bentz. -\rt Stock British Colimibia
Forest Service. Nelson). Da\\-n Hansen, and
John .\nhold both Forest Hcidth Protection.
Ogden. UT . Jim Viuidygriff Inteniiountiiin
Research Station. Logiui. UT produced die
graphs. .\lan Dxinersld (Forest HciUth Protec-
tion. Ogden, UT) proxided significant assis-
tance with the annual installation of plots.
1996]
Dryocoetes confusus Flight PpzRioDicrn
359
Thanks to the main who helped collect and
count the trap captures o\er 3 seasons: Alan
Dymersld, Dan Johnson (USDA Forest Ser-
vice, Ogdcn, UT), Jill .\nsted (USDA Forest
Ser\ice, Heber City, UT), Heather Schmidt,
Keith Pflei^er. Rhonda Bishop, Josh Vierej^ge,
Steve Deakins, and Chris Peterson (all Utah
Department of Agriculture, Salt Lake City,
UT). Irene \bit (Intennountain Research Sta-
tion, Ogden, UT) conducted the literature
search. Special thanks to Art Stock and Bar-
bara Bentz for encouragement and insights.
My thanks also to Jesse Logan and L\nn Ras-
mussen O^oth Intennountain Research Station,
Logan, UT) for allowing me to complete this
manuscript. This project was partially funded
by Intennountain Research Station, Mountain
Pine Beetle Project, Logan, UT.
LlTER\TLRE CiTED
A,MMAN, G. D.. AND D. L. Bartos. 1991. Mountain pine
beede offspring characteristics associated with females
producing first and second broods, male presence,
and egg galler\ length. Environmental Entomology
20: 1562-1.567.'
Anderbrwdt, O., E Schlyter, asd G. Birgersson. 1985.
Intraspecific competition affecting parents and off-
spring in the bark beetle /;«• tyjinfiraphus. (Jikos 45:
89-98.
Borden, J. H., A. .VI. Pierce, M. D. Pierce, Jr., L. J. Cik ag,
A. J. STfx.K, AND A. C. Oemi.schi^(;er. 1987. .Semio-
chemicals produced by western balsam bark beetle,
Dryocoetes amfwms Swaine (Coleoptera: Sco^■tidae^
Journal of Chemical Ecology 1.3: 82.3-836.
Bright, D. R., Jr. 1963. Bark beetles of the genus Dry-
ocoetes (Coleoptera: Scolytidae) in .North Amerit-a.
Annals of the Entomological Society of .America 56:
10.3-115.
DoiDGE, D. 1981. Western balsam bark beetle in British
Columbia. Canadian Forest .Service, Pacific Forest Re-
search Centre, Victoria, B.C. Forest Pest Ix-aflet 64.
Fl\mm, R. O., S. R Cook, T. L. Wagner, P E. Plleey, and
R. .\. CoLLSON. 1987. Reemergence and emergence of
Ips acuhwi and Ips calligraphm (Coleoptera: ScoKti-
daej. Environmental Entomologv 16: 869-876.
Johnson, D. W. 198.5. Forest pest management training man-
ual. USD.\ Forest Service, Rock\ .Mountain Region.
Lakevvood, CO. 138 pp.
-Mathers, W. G. 1931. The biology of Canadian baH< beetles.
The seasonal history of Dryocoetes confusus Sw.
Canadian Entomologist 68: 247-248.
Stock, A. J. 1991. The western balsam bark beetle, Dry-
ocoetes confusus Swaine: impact and semiochemical-
based management. Unpublished doctoral disserta-
tion, Simon Eraser University, Bumaby, B.C. 13.3 pp.
Received 11 December 1995
Accepted 12 August 1996
Great Basin Naturalist 56(4), © 1996, pp. 360-368
DISTRIBUTION OF A THERMAL ENDEMIC MINNOW, THE DESERT
DACE {EREMICHTHYS ACROS), AND OBSERVATIONS OF IMPACTS
OF WATER DIVERSION ON ITS POPULATION
Gaiy L. Vinyard^
Abstract. — Population status sui-veys were performed from 1987 to 1996 for desert dace {Eremichthys acros), a
cyprinid endemic to several small thermal springs in Soldier Meadow, Hinnboldt County, Nevada, where the species
occupies 7 spring areas in a single valley. Because spring distributions are patchy and all areas are not linked by surface
flow, each area comprises a more-or-less isolated population, although iiTigation practices or high runoff may occasionally
link several of them. Although limited to thermal springpools and outflows, desert dace were foimd in temperatmes rang-
ing fi'om 37°C near spring sovux'es to 13°C in downstream areas. Between May 1988 and October 1989, most of the dis-
charge fiom a major spring outflow was diverted from its natmal channel into an inigation ditch. Trap catches in the orig-
inal channel were reduced after the diversion, and fish densities were lower in die ditch dian in the channel. Reduced fish
numbers still persist (1996), even though the affected site has been relatively undisturbed since 1989. To improve desert
dace habitat and increase populations, inigation diversion should be discontinued and water returned to the original chan-
nel. Continued protection and increased habitat presewation for desert dace are recommended because of their limited
distiibution, apparently restincted habitat requirements, and the potential for environmental disniption in the area.
Key words: conservation, endemisni, Eremichthys acros. Great Basin, habitat, irrigation, Nevada, springs.
The desert dace {Eremichthys acros, Cyprini-
dae) is a federally listed threatened species
restricted to outflows of thermal springs in
Soldier Meadow, Nevada (Hubbs and Miller
1948, La Rivers 1962). Desert dace have a
unique homy sheath on both jaws and a greatly
elongated intestine; they occupy exceptionally
high temperatures relative to other cyprinids
(Hubbs and Miller 1948, Nyquist 1963). The
distinctive moiphology of desert dace suggests
a long period of isolation extending beyond the
most recent pluvial period. The desert dace
differs significantly from other Great Basin
minnows, and its original description and
assignment to a monoty[Dic genus (Hubbs and
Miller 1948) have been confirmed (La Rivers
1962, Nyquist 1963, Cavender and Coburn
1992). Its taxonomic relationship within the
western cyprinid fauna remains unclear, and
recent workers have judged desert dace most
similar to either relict dace, Relictiis solitarius
(Cavender and Coburn 1992), or to tui chub,
Gila {Siphateles) hicohr (Lugaski 1980).
Few investigations of desert dace have been
undertaken since Nyquist (1963), and little is
known of their behavior, ecology, or physiology'.
I present results of studies of desert dace dis-
tribution and document adverse consequences
to the species fiom water diversion in die valley.
Study Area
Soldier Meadow and Mud Meadow to the
south occupy a gently sloping valley in south-
western Humboldt County, Nevada. Elevation
ranges fi-om about 1400 m MSL at the noitli end
of Soldier Meadow to about 1317 m MSL south
of Mud Meadow Resei-voir (Fig. 1). Although
the lower elevations are near maximum shore-
line level of pluvial Lake Lahontan, the area
was prol^ably not inundated during the Pleis-
tocene (Benson 1978). Soldier Meadow is also
home to an endemic plant, the basalt cinque-
foil {Potcntilla hasaltica), and at least 4 un-
described species of h\'drobiid springsnails
(R. Hershler, Smithsonian Institution, personal
communication).
Ranching operations in Soldier Meadow be-
gan in the late 1860s but ha\ e ne\'er been par-
ticularly successful. After an active period in the
1960s, ranching was largely dormant in the val-
ley through most of the 1970s and early 1980s.
In 1994, \\ ith the help of the Nature Conser-
vancy, much of the property' was transferred to
I Dt'partinciit of Biolo©' /315. University of .Nevada, Reno, NV S95.57.
360
1996]
DisTKiiJU rioN oi' Desert Dace
361
4586
4584
4582 -
4580 -
TO SUMMIT LAKE / I
4578
4576
4574
GENERAL LOCATION
SOLDIER MEADOWS
Enlargement of Area 4
TO
HIGH ROCK
t- LAKE
1 mi
TO GERLACH
1 km
310
315
320
325
Eig. L Map of Soldier Meadow and Mud Meadow showing sampling locations. Solid lines indicate water conrses;
dotted lines are roads. Spring areas discussed in the te.xt are enclosed in ovals. Not all springs in the valley are presented,
and in some areas numerons small springs are represented by single symbols. Springs and streams outside the ovals lack
desert dace {Eremichthtjs acros), but not all sites within ovals contain fish. Values on axes indicate coordinates in relation
to 1000 meter Universal Transverse Mercator Grid, digitized from USGS maps. Insets show location of Soldier Meadow
in relation to Nevada state boundaiy (left) and an enlargement of the spring and ditch system associated with area 4
(right). The area 4 inset depicts the relationships between the spring sources, upstream zone, ditched zone, and old
chaimel as discussed in the te.xt. The ditch was most recently dredged between May 1988 and October 1989. The
approximate upstream limit of desert dace was near the start of die traps.
the U.S. Bureau of Land Management. Most
desert dace habitat now occurs on pubHc land,
and the remainder of the privately held habitat
is protected by a consei"vation easement (Nature
Consei^vancy 1994). Springs and outflow streams
in the valley are all subject to grazing by cattle,
feral horses and burros, and pronghoni. There
is also frequent recreational use of the area by
hunters, campers, bathers, and others.
Fish Distribution
Desert dace distribution is strongly corre-
lated with spring discharge. Fish are absent from
small springs or seeps with little surface water
and fiom larger pools of very hot water lacking
organized discharge. All springs with perennial
surface flow are occupied by desert dace, the
most abundant fish in the valley. Although
desert dace are most often found in habitats
lacking other fish species, they are not confined
to them and have been observed coexisting
with tui chub, speckled dace {R. osciilus), and
Tcilioe suckers {Catostomus tahoensis; La Rivers
1962, Nyquist 1963, Sigler and Sigler 1987).
Desert dace habitat occurs in 7 distinct areas
located within an 8-km (5-mi) radius (Fig. 1).
Because spring distributions are patchy and all
areas are not linked by surface flow, each area
comprises a more-or-less isolated population,
although irrigation practices or exceptionally
high runoff may occasionally allow fish passage
among several of them. Most areas described
below contain many springs varying greatly in
size, and it is often difficult to identif\' the
exact number and location of discharge sources
because of the dense vegetative cover.
362
Great Basin Naturalist
[Volume 56
Area 1
Area 1 (Fig. 1) includes the type locality
(Hubbs and Miller 1948) for desert dace (Table
1, site 19), a spring that issues from the base of
a small cliff It is modified by a valved struc-
ture diverting flow into a pipe for household
use at Soldier Meadow Ranch. Undiverted dis-
charge flows about 100 m east, where it enters
a small impoundment or a series of irrigation
ditches. Desert dace coexist with tui chub in
the spring and impoundment. Desert dace
habitat here is limited by the impoundment
and by the shifting diversion into ditches.
Area 2
In area 2 (Fig. 1) several small springs con-
taining desert dace (Table 1, sites 9, 23-27) are
located around the base of a small hill and flow
generally southward or southwestward into a
large meadow. A large springpool containing
desert dace (Table 1, site 10) is the largest nat-
urally occurring body of open water in the val-
ley (approximately 15 m in diameter and 1.5 m
deep). It has a minor surface outflow south into
a marshy meadow. Nonnally, most of the springs
in this area are unconnected, although proxim-
it)' and common drainage suggest connections
are likely during high lamoff. Natural drainage
from this site is toward area 7, and the outflows
of several of the more southerly springs are
ditched southward through the meadow. The
larger springs in this area receive frequent
recreational use by bathers and campers.
Area 3
Area 3 (Fig. 1) includes several small springs
flowing south approximately 1 km north of Mud
Meadow Resei-voir (Table 1, site 20). Although
at least 3 springs in this area contain desert
dace, and some populations are quite dense, all
springs in the area have been heavily affected
by livestock grazing and irrigation diversion.
The 2 largest springs have long been diverted
into irrigation ditches at a point within 20 m of
the sources. Grazing by cattle, buiTOS, and feral
horses has altered the vegetation and disrupted
soils near the springs. This distin-bance has wid-
ened tlie outflow channel, reduced water depth,
and generally eliminated riparian vegetation.
Area 4
Several large springs issuing from the side
of a small hill are the source for area 4 (Fig. 1).
Widi more than 2600 m of stream, this is the
largest contiguous potential habitat for desert
dace. Water issues from the highest springs at
approximately 50°C (Table 1, sites 1-6, 11, 21)
and cools gradually while flowing downstream
with occasional augmentation by both warm
and cool inflows. These springs produce an
aggregate discharge of approximate!) 60 1/s.
The upper reaches are fishless, presumabh' be-
cause of high water temperatures. Headwater
springs in area 4 probably receive the highest
level of recreational use in the valley, primarily
from bathers and campers. Several small cobble
dams erected across the outflow stream in this
area are mostly upstream from the dace habitat
and pose little impediment to fish passage.
Deposition of soaps and other water pollutants
from bathers may constitute a risk of unknown
magnitude. This area has also been heavily
grazed by cattle and feral horses and buiTos.
Several inigation diveisions ha\'e long existed
in area 4; however, they were poorly maintained
and little used for at least a decade. Between
May 1988 and October 1989, the rancher in
the valley dredged out an old ditch, moving
water away from the original channel in area 4
(Fig. 1, inset). Before the dredging most of
the discharge continued southeastward in the
natural channel and spread into a large, wet
meadow. Speckled dace historically coexisted
with desert dace in the lower sections of this
system near the wet meadow.
By October 1989 most discharge in area 4
was diverted east to the irrigation ditch, and no
water reached the meadow by the original
channel. CurrentK, approximateh' 80% of the
total combined discharge from the source
springs in area 4 is diverted. Speckled dace are
now absent from the system, and the amount of
desert dace habitat was significantK reduced
by this diversion. The loss of discharge into the
lower portions of the wet meado\\' on the down-
stream end of area 4 had additional adverse
impacts on desert dace in area 5 (see below).
Area 5
Area 5 includes a group of \ en- hot springs
that enter a series of old irrigation ditches
approximately 200 m from the source and then
flow southeasterly toward Mud Meadow Reser-
voir (Fig. 1). In 1988 a series of cool springs fed
1)\ discharge from the wet meadow below area
4 entered the outflow stream at area 5 approxi-
mateh' 50 m downstream from the primary
spring sources. Mixing of these waters produced
1996]
DlS THIBUTION OF DESERT DaCE
363
Taiu.K 1. Characteristics of spring Iiahitats in Soldier Meadow. Data were collected at various times between 1987
and 1995. Column desiunations are as follows: AHKA = distribution area (numbers indicate areas indicated on Figure 1;
sites without numbers lack tlesert dace); SITK = site identifier Irom field notes, refers to specific locations within areas;
EAST and NORTH indicate site locations in relation to lOOO meter Universal Transverse Mercator Crid, di^jitized from
uses maps — Mud Meadow, 1972; Soldier Meadow, 1972; FISH = fish six'cies present, E = Eremichlhtis arms, H =
Rhiniclithij.s osnilu.s, C, = Gila bicohn C = Catostonuis taliooisis, a = fish absent; °C = water temperature; DO = dis-
soKed o.xygen concentration (mg O2/I); CX)ND = electrical conductivity (;UMho/cm).
AREA
SITE
EAST
NORTH
FISH
"C
DO
COND
1
19
31SSS5
4584932
E,C
28-29
5.6-6.2
190
1
19A
319038
4584990
E,C
22-26
8.0-8.6
190
2
9
317625
4583030
E
34-38
3.4-4.5
432
2
10
316826
4582540
E
21-34
4.8
410
2
23
317547
4582650
E
21
3
20
318607
4578372
E,R
25
4.5-6.1
270
4
21
315827
4580116
E
17
4
1
314016
4580934
a
36-40
3.6-5.8
370-420
4
2
314068
4580654
a
36-38
5.3-5.8
380-420
4
3
314446
4579892
E
30-35
6.2-6.4
310
4
4
314869
4579580
E
27-28
6.4-7.0
4
5
314005
4580202
a
37-42
1.8-3.7
400-430
4
6
315961
4578982
E,R
19-25
5.8-7.6
280-285
4
11
315331
4579392
E
20-29
6.3-8.0
305
5
12
316691
4578454
E
13-50
1.9-6.3
480
5
13
316888
4578154
E,R
23
6.1
650
5
14
317089
4577836
E,R
23
5.6
650
6
16
314599
4581304
E
23
6.5
295
7
8
316929
4580550
E
34-57
0.9-6.9
470-750
—
1
316682
4580660
R,C
6
8.3
325
—
15
315512
4576460
a
25
5.6
280
—
17
314198
4581512
a
35
3.6
370
—
18
316939
4580684
a
50
3.8
280
—
22
317115
4581286
R
13
—
24
317536
4582660
a
35
—
25
317519
4582660
a
37
—
26
317498
4582640
a
40
—
27
317516
4582620
a
40
—
2.S
316383
4580864
a
10
8.9
—
29
316381
4580940
a
9
5.5
steep temperature gradients, as waters of
±45 °C and <20°C gradually mixed over about
100 m. In May 1988 water in the main spring
outflow was 43 °C at the point where water at
13 °C entered fiom the meadow to the north.
Desert dace were observed actively feeding in
the 13 °C water mass at the point where cold
\\ ater entered the primary channel. Fish also
darted into the turbulent zone between the hot
and cold water masses in pursuit of small drift-
ing food. All observations since October 1989,
after the diversion of water in area 4, have
found the amount of water reaching the lower
sections of the wet meadow above area 5 to
be greatly reduced, and inflow from the cool
springs flowing into area 5 has ceased. Conse-
quently, several hundred meters of the approx-
imately 1 100 m of (ditched) desert dace habitat
have been lost from this area. This area is
grazed and the outflow ditched, but it receives
relatively little recreational or other use.
Ai-ea 6
The desert dace population in area 6 (Table
1, site 16, Fig. 1) occupies a single spring
stocked by U.S. Bureau of Land Management
personnel in the early 1980s at a time of con-
cern over the ftiture of desert dace. This spring
has the smallest discharge of any containing
desert dace (estimated at less than 5 liters per
minute). A gauging box and wooden notch weir
produce a small impoimdment (appro.ximatcK
3 m X 3 m) about 30 m from the source, \\'liicli
contains most of the fish population at this site.
Recently this impoundment was nearly lost by
deterioration of the weir. This system also in-
cludes a somewhat larger eailJien impoundment
(±10 m diameter) approximately 50 m from the
364
Great Basin Naturalist
[Volume 56
source, after which flow disappears into a small
meadow. Desert dace are mostly restricted to
the area above the larger impoundment and
are most abundant near the gauging box and
notch weir. This site is grazed but too small for
recreational use.
Ai-ea7
Area 7 includes several hundred meters of
suitable habitat fed by several springs on the
eastern side of the valley (Fig. 1). Because sev-
eral springs issue at temperatures exceeding
50° C, the extent of suitable habitat varies with
ambient air temperature. This area may be
connected with outflows from area 2 during
periods of high iimoff. Most of the outflow in
this area has been modified to some extent for
inigation, and it is subject to grazing and some
recreational use.
Mud Meadow Resei-voir
Mud Meadow Reservoir contains large-
mouth bass {Micropterus salmoides), goldfish
[Carrasius auratus), and perhaps other species
planted by unknown individuals (Ono et al.
1983). It is unclear whether it is a barrier to
desert dace passage, but it is unlikeK' to provide
any permanently suitable habitat. Although no
nonnative fishes have been observed in any
sites containing desert dace, the potential
threat posed by nonnative fishes spreading into
dace habitat is certainly enhanced by their
establishment in the resei'voir.
Materials and Methods
All desert dace habitats recorded by Nyquist
(1963) and most other springs in the valley
were visited in 1987 to update distribution
information (Vinyard 1988). The dredging of
the irrigation diversion in area 4 significantly
reduced the amount and quality of desert dace
habitat in that area. Investigations thus were
concentrated in the affected locality (area 4)
beginning in 1989.
Fishes in area 4 were sampled widi standard
unbaited minnow traps on 5 occasions (14 May
1988, 20-22 October 1989, 3 November 1993,
20-21 October 1995, and 27 April 1996). Dur-
ing May 1988 sampling included the entire
original stream channel (>2.6 km) from spring
sources to disappearance of the stream in a wet
meadow. The section upstream from the diver-
sion and the irrigation ditch were sampled in
October 1989, November 1993, October 1995,
and April 1996. The remnant natural channel
downstream from the diversion was also sam-
pled in October 1995 and April 1996 (Fig. 1,
inset).
Fish traps were 40 cm long by 20 cm diame-
ter, constructed of 0.64-cm-mesh galvanized
hardware cloth, with 2.5-cm entrance holes at
the peak of each concave conical end section.
Traps were placed at 20-m intei-vals along spring
outflows and fished 2 h during dayliglit. Altliougli
the traps were sometimes not completely sub-
merged, they were always placed witli tire open-
ings under water. Captin^ed fishes were identi-
fied, enumerated, and released near the point
of capture. Standard length (SL, in mm) and
weight (gm) were recorded on some sample
dates. Water temperature, dissolved oxygen, and
electrical conductivity were measured using
portable meters at regular intei-vals along the
trap set.
In October 1995 stream velocity' was mea-
sured along cross-section transects with a
Marsh-McBurney model 201D flow meter at 6
or 9 sites each in the upstream, ditch, and old
channel zones of area 4. Measurements were
at 5-cm vertical and either 10- or 20-cm hori-
zontal increments, depending on channel widtli.
In June 1995 electrofishing was performed
using a 3-pass depletion methodology (Van-
Deventer and Platts 1989). Stream sections 10
m long were isolated with blocking nets and
depletion rates on successive passes used to
estimate population size. Three groups of 6
sections were fished: in area 3, and in the old
channel and ditched zones of area 4.
Results
Although resident in thermal springs and
outflows, desert dace have wide temperature
tolerances and were obsened in waters rang-
ing from 13 °C to 38 °C. Occupied waters had
conductivity ranging from 190 to 650 fuS and
dissoKed oxygen concentrations generalK near
saturation, ranging from 4.5 mg/1 to 8.0 mg/1
(Table 1). Although there was considerable
overlap between species, desert dace were
ioimd at higher temperatures and lower dis-
solved oxygen concentrations than speckled
dace, which were not observed at tempera-
tures above 26 °C or in dissolved oxygen con-
centrations below 5.2 mg/1.
1996]
Distribution of Desert Dace
365
In area 4 in 1988 desert daee eateh in min-
now traps was signifieantly negative!) corre-
lated with temperature (linear regression; F =
19.98, /?2 = 0.122, n = 131, P < 0.001), al-
though relati\ely little variance in catch was
explained by temperatiue, and no such cone-
lation was obsened in later years. Desert dace
catch was also not significantly correlated with
speckled dace catch. Catch rates generally
reflect fish abundance but may also be affected
by activity. Temperature, dissolved oxygen con-
centration, and combinations of these and
other factors may affect activity levels.
In 1988, when desert dace and speckled
dace were sympatric in the natural channel
above the meadow in area 4, mean catch per
trap-horn- was significantly greater {t = 2.83, P
= 0.009) for desert dace (4.56 fish per trap-
hour) than for speckled dace (1.04 fish per
trap-hour). Where both species occurred,
desert dace was more abundant, and maximum
densities of both species were observed at
temperatures of about 23 °C.
Cross-section grid transect measurements
of water velocit)' were used to assess mean val-
ues at each transect and to compute discharge.
In 1995 in area 4, velocities were significantly
higher in the ditch (6 transects, z = 24.9 cm/s,
n = 121) than in the upstream zone (9 tran-
sects, z = 17.3 cm/s, n = 318; t test, df = 159,
T = -3.663, P < 0.001), or in the old channel
(6 transects, z = 16.7 cm/s, n = 111; t test, df
= 188, T = 3.733, P < 0.001). However, veloc-
ity measurements did not differ significantly
between the upstream and old channel seg-
ments {t test, df = 226, T = 0.504, P = 0.616).
Volumetric computations indicated that dis-
charge in the ditch was 46.5 1/s while discharge
in the channel was 10.8 1/s, or 18.8% of the
total.
In October 1995 desert dace trapped in the
upstream zone were significantly smaller
(mean SL = 35.9 mm, n = 172) than those
from either the ditch (mean SL = 38.7 mm, n
= 82; t test, df = 105, T = -3.33, P = 0.001) or
the old channel (mean SL = 38.4 mm, n =
135; t test, df = 284, T = -5.50, P < 0.001).
Standard length of the fish in the ditch and in
the old channel did not differ significantly {t
test, df = 113, T = 0.328, P = 0.744).
Electrofishing transects in area 4 June 1995
yielded mean values of 21.8 fish per 10 m {n =
6, ,s' = 27.2) for the old channel and 12.5 fish
per 10 m (n = 6, s = 8.5) for the ditch zones.
Densit>' estimates of 110 fish per 10-m section
{it = 6, s = 51.53) were obtained in area 3 at
that time. These values did not differ signifi-
cantly among the 2 zones of area 4; however,
densities in area 3 were significantly higher
than in either zone of area 4 {t tests, P < 0.01
in both eases). The fish eleetrofished from the
ditch were significantly larger (avg. SL = 37.2
mm, n = 67, s = 7.9) than those from the old
channel (avg. SL = 32.7 mm, n = 122, .s- =
11.3;/test, f = 3.2, P = 0.002).
Discussion
Trap data from area 4 offer an opportimit>'
to assess impacts of habitat alteration by com-
paring catch rates in the zone upstream from
the diversion, in the original channel down-
stream, and in the ditched zone (Fig. 2).
Because traps were set on the same spacing
intei-vals in each sampling period, it is possible
to examine cumulative catch per trap hour to
compare fish densities. These values are com-
puted by summing catch per trap hour for each
trap along the trap set from the upper to the
lower end (Fig. 2).
The total cumulative catch in area 4 was
much larger in May 1988 than at any other
time (Fig. 2). In contrast, the lowest cumulative
catch observed was in October 1989, the first
sample after the ditch dredging. Although
direct comparisons of these 2 samples may be
confounded by seasonal differences, the con-
trast between the largest catch obsen^ed (in
May 1988) and the smallest catch observed (in
October 1989) coincides with the dredging.
Comparison of the autumn sample in October
1995 with the spring sample in April 1996 sug-
gests that populations are larger in die spring
than fall, but that this difference is probably
insufficient to explain the difference between
the 1988 and 1989 data.
By November 1993 the cumulative catch had
recovered somewhat from 1989 (Fig. 2). A not-
able difference in 1993 relative to both earlier
observations was the sharp increase in catch
apparent at about 1000 m, slightly upstream
from the diversion. However, with the excep-
tion of the accumulation of fish at this point,
the general slope of the cumulative catch curve
was little changed from October 1989. In
October 1995 and April 1996 shaip increases
in catch immediately above the diversion were
still apparent. The general slope of the catch
366
Great Basin Naturalist
[Volume 56
cr
D
o
I
cc
LU
Q.
I
O
o
LU
>
250
200
150
100
50
I 200
O
150
100
50
CHANNEL -APR 96
CHANNEL -OCT 95
DITCH - APR 96
UPPER ZONE
APR 96
500 1000 1500
DISTANCE (m)
2000
2500
Fig. 2. Cumulative catch of desert dace in unbaited minnow traps in area 4 from Ma\' 1988 through April 1996: A,
Data collected in May 1988, November 1993, and October 1989. B, Results from 1995 and 1996. In all cases single, un-
bailed minnow traps were fished at 20-m intervals for 2-h sets during daylight. Cumulative catch per hour is computed
by summing catch per hour for each trap beginning at the upstream end of the trap set. Distance on the ordinate is the
distance downstream from the first trap. In May 1988 cumulative catch reached 448 at 2620 m downstream (off scale).
Dredging of the irrigation diversion occurred between May 1988 and October 1989. Samples in October 1988 and 1993
included only the zone upstream above the diversion and the ditched zone downstream. In October 1995 and .\pril
1996, the old channel remaining below the diversion was also sampled.
cui-ve for the ditclied segment changed little
between 1993 and 1996 (Fig. 2). The slope
from the renniant old channel in 1995 and
1996 (Fig. 2) was much steeper than that
observed in the ditch, in spite of the roughly 4
times greater discharge measured in the ditch
in 1995. Catch rates in the ditch or old channel
have never reached levels observed in the
channel in 1988, and even summing the cumu-
lative catch from both the ditch and the old
channel still does not \ ield results comparable
with the catch rates ol^sened in 1988.
Different responses b\- desert dace to the
various habitats in area 4 are also evident in
the percentage of traps with non-zero catch
(Fig. 3). This measure can be used as an indi-
rect indication of the amount of habitat occu-
pied. In 1988 the channel zone had the highest
percentage of traps catching fish, nearly 90%.
In the 4 samples from the ditched segment.
lfJ96]
DisTKiiui ION OF Desp:rt Dace
367
12
■?10
r
.6
. 4
UJ
3
&2
H UPPER ZONE
■ DITCH
□ CHANNEL
J3_Jl
100
80 -
MAY88 OCT89
NOV93
DATE
OCT95 APR96
Fig. 3. Axerage catch per trap hour for all traps, area 4,
from Ma>' 1988 tliroiigh April 1996. Bars indicate different
stream segments. Dredging of the inigation ditch occuiTed
between May 1988 and October 1989. The upper zone
sampled was abo\e the point of the irrigation diversion.
The ditched segment existed prior to dredging in 1988 but
had become overgrown nearly to the point of obstructing
any flow. After dredging, it received most of the discharge
from the system. The channel received nearly all the flow
from the springs in area 4 during the 1988 sample, but
only 20% or less of the total flow in subsequent samples.
never more than 75% of the traps eaught fish.
The old channel zone continues (in both 1995
and 1996) to have a larger percentage of traps
catching fish than either of the other 2 zones
(Fig. 3).
Catch per trap hour may also be used to
estimate relative fish populations. The 3 high-
est average catch rates were observed in the
channel below the present point of diversion,
and the highest value of any was obsei'ved in
1988, prior to the dredging (Fig. 4). Although
catch values were still highest in 1995 and
1996 in the old channel zone, they have not
returned to levels observed in 1988. Catches
fi-om the upstream and ditched zones have var-
ied much less during the sample period.
These data indicate that the natural channel
was the most productive site for desert dace
prior to the ditch dredging, and that it still pro-
vides habitat which is superior to the ditch,
c\ en 8 yr after the dredging and with <20% of
the total discharge.
The obsei-ved aggregation of fish above the
diversion (evident in the cumulative catch data
since 1993) bears examination. If habitat in the
ditch is unsuitable, desert dace may avoid the
ditched zone and accumulate in the upstream
zone. Because no aggregation of fish in this
zone was obsei-ved in 1988 or 1989, it seems
likely to be the result of a behavioral response
to the changed conditions.
i 60
40
20
■ UPPER ZONE
■ DITCH
□ CHANNEL
MAY88 OCT89 NOV93 OCT95 APR96
DATE
Fig. 4. IVrcent traps with non-zero catch, area 4, from
May 1988 through April 1996, Bars indicate dilTerent stream
segments. Dredging of the irrigation ditch occuned between
May 1988 and October 1989. The channel received nearly
all the flow from the springs in area 4 during the 1988
sample, but only 20% or less of the total flow in subse-
quent samples.
The higher mean water velocities observed
in the ditch (24.9 cm/s) relative to the up-
stream zone (17.3 cm/s) suggest that desert
dace may avoid higher velocity flows. It is
likely that smaller fish avoid higher velocity
flows in the ditched section and accumulate in
the region immediately upstream from it.
Although this explanation does not account for
the relatively low abundance of fish in the old
channel, other factors, including reproductive
success and differences in habitat qualit)-', may
be important. The absence of the aggregation
upstream from the ditch in 1988 may reflect a
general population reduction resulting from
the ditching.
Distribution of desert dace reflects poteu-
tially interacting factors including tempera-
ture, dissolved oxygen concentration, and cur-
rent velocity. Distribution may also be affected
by interactions with other species, particularK
speckled dace. Studies are necessaiy to iden-
tify and assess the mechanisms of such interac-
tions. An additional area of interest would be
to assess the relative degree of isolation of the
7 population units identified in this study to
determine whether there are any behavioral,
ecological, or genetic differences among these
groups.
Conclusions
In recent years desert dace have been sub-
jected to relatively minor disturbance com-
pared with many other native fish species in
the Great Basin. Most of the sites historically
368
Great Basin Naturalist
[Volume 56
occupied by desert dace retain suitable habitat,
though it has generally been modified to some
extent. Disturbance levels may have been higher
at times in the 1960s (Nyquist 1963). Desert
dace populations in Soldier Meadow have been
relatively stable since 1989, but most desert
dace habitats have been substantially altered
over the years, and we cannot now directly
assess the magnitude of any persistent popula-
tion reductions that may have occurred before
that time. Desert dace populations persist in
the modified thermal waters that now charac-
terize Soldier Meadow; however, the data from
area 4 demonstrate that adverse effects of habi-
tat modifications linger for many years.
It is appropriate to consider management
options for this unique fish. Their presei^vation
requires continued physical protection of
springs and flowing waters in Soldier Meadow
from excessive grazing and prohibition of the
introduction of nonnative organisms. Restoring
the water to natural stream channels should
also be incoiporated into any management plan
because of the potential positive impacts from
improving habitat quality. Consequences of such
water management for the endemic spring-
snails should either be neutral or positive. They
are generally abundant in the springs where
they occur, and losses fi^om restoration of flows
should be offset by increased habitat stability.
Desert dace seem relatively secure under
current conditions. However, the small num-
ber of occupied sites, restricted geographical
distribution, and generally unknown but possi-
bly specialized habitat requirements of the fish
argue strongly for continued monitoring and
increased investigation into factors regulating
populations. Growing demands on aquatic
resources of the Great Basin make it clear that
increased awareness of and protection for this
unique fish will be necessary for their long-
term sui'vival.
Acknowledgments
I am grateful to A. Berglund, J. Dunham,
T Kennedy, K. Obermeyer, R. McNatt, D. Sada,
M. Sevon, and C. Stock-well for assistance with
various aspects of this work, and to Dave Liver-
more of the Nature Consen'ancy for his efforts
on behalf of desert dace. Members of various
Desert Ecosystems, Aquatic Ecology, and
Ichthyology classes from the University of
Nevada, Reno, assisted in fieldwork.
Literature Cited
Benson, L. V 1978. Fluctuation in the \e\e\ of pluvial Lake
Lahontan during the last 40,000 years. Quaternaiy
Research 9: 300-.318.
Cavender, M. M., and T. M. Coburn. 1992. Interrelation-
ships of North American cyprinid fishes. Pages
328-373 in R. L. Mayden, editor, S\ stematics, histor-
ical ecology and North American freshwater fishes.
Stanford Universit)' Press, Stanford, CA.
HuBBS, C. L., AND R. R. Miller. 1948. Two new, relict
genera of cyprinid fishes froiii Nevada. University of
Michigan Museum of Zoology Occasional Papers
507:1-30.
La Rivers, I. 1962. Fishes and fisheries of Nevada. Nevada
State Fish and Game Commission. 782 pp.
Lugaski, T. p. 1980. Comparative chemota.xonomy of
selected Great Basin native cyprinid fishes. Unpub-
lished doctoral dissertation. UniversitA' of Nevada,
Reno. 2.54 pp.
Nature Conservancy. 1994. Soldier Meadow conserva-
tion project. Great Basin Field Office, Salt Lake City,
UT
Nyquist, D. 1963. The ecolog>' of Eremichthys acws, an
endemic thermal species of cyprinid fish fi'om north-
western Nevada. Unpublished master s thesis. Uni-
versity of Nevada. 247 pp.
Ono, R. D., J. D. VViLLUMS, and a. Wagner. 1983. Vanish-
ing fishes of North America. Stonewall Press, Wash-
ington, DC. 257 pp.
SiGLER, W, AND J. SiGLER. 1987. Fishes of the Great Basin.
University of Nevada Press, Reno. 443 pp.
VanDeventer, J. S., AND W. S. Pu\TTS. 1989. Micro-
computer software system for generating population
statistics from electrofishing data — user's guide for
MICROFISH 3.0. U.S. Forest Service General Tech-
nical Report INT-254.
Vinyard, G. L. 1988. Population status sune\ of tlie Soldier
Meadows desert dace {Ercinichfliys acros). Submit-
ted to the U.S. Fish and Wildlife Sei-vice. Contract
14320-87-00178.
Received 19 Fchruanj 1996
Accepted 26 June 1996
Great Basin Naturalist 56(4). © 1996. pp. 369-374
HELMINTHS OF THE SOUTHWESTERN TOAD, BUFO MICROSCAPHUS,
WOODHOUSE'S TOAD, BUFO WOODHOUSII (BUFONIDAE), AND
THEIR HYBRIDS FROM CENTRAL ARIZONA
Stephen R. CoUlluM-ui. Cliarles H. Bursey-, Keith 13. Malmo.s'^
Brian K. Snllivan"\ and Ha\ Cheain'
AB-STRACT. — The gastrointestinal tracts, lungs, and iirinar\' hladders from 77 Btifo inicroscaijIiiLs, 61 Bufo woodhousii,
and (S of their hyhrids were e.xamined for helminths. One species of trematode {Glyptlwlinins quiela), 1 species of ces-
tode {Distoichotnetra bttfimis), and 5 species of nematodes {Aph'cianu incerta, A. itzocancims, Rhabdias americanus,
Physaloptcra sp., and Physocephalus sp.) were found. The greatest prevalence (41%) and mean intensity (231.7) were
recorded ihv Aplcctiina iiiccrta in Bufo icoodhoii.sii. !t appears h\hrids Iiarhoi- fewei- i:iarasites than either parent species.
Key icords: Iwhniutlis. Bulo microscaphus, Huto woodhousii, hyhrids. Arizniia.
The southwestern toad {Bnjo uiicroscapluts
Cope, 1866) is presently recognized as 3 allo-
patric subspecies: B. )n. californicus Camp, 1915,
which occurs in coastal southern California
and northwest Baja California; B. m. microsca-
phus Cope, 1866, found in southern Nevada
and Utah, Arizona, and New Mexico; and B. m.
mexicanus Brocchi, 1879, which occurs in the
Sierra Madre Occidental of central Me.xico
south to Durango (Price and Sullivan 1988).
Woodhouse's toad {Bufo woodhousii Cirard,
1854) is recognized as 4 subspecies: B. w. wood-
housii Cirard, 1854 occurs in eastern Montana
and North Dakota, south through the plains
states to central Texas and west of the Rocky
Mountains from Idaho south to Colorado and
Arizona with isolated populations in west Texas,
southeastern California, and along the Oregon-
Washington border; Bufo w. austrolis Shannon
and Lowe, 1955 is found from central Colorado
through New Mexico and Arizona to Sonora,
Mexico, and iilong the Rio Crande drainage into
southwest Texas and adjacent Mexico; Bufo w.
velatus Bragg and Sanders, 1951 is restricted
to northeast Texas; and B. w. fowled Hinckley,
1882 is widespread throughout much of the
eastern United States south to the Culf Coast
and west to eastern Te.xas (Beliler and King 1979).
The toads examined during this study, B. in.
microscaphus and B. w. austraUs, are known to
li> bridize in Arizona (Sullivan 1986, Sullivan
and Lamb 1988).
Altliough diere are reports of helminths from
B. microscaphus (Pany and Cnmdmann 1965)
and B. woodhousii (Trowbridge and Heflev 1933,
Brandt 1936, Walton 1938, Reiber et af. 1940,
Kuntz 1941, Kuntz and Self 1944, Rankin 1945,
Fantliam and Porter 1948, Frandsen and Crund-
mann 1960, Pany and Crundmann 1965, Camp-
bell 1968, Brooks 1976, Jilek and Wolff 1978,
Baker 1985, Hardin and Janovy 1988, McAUister
et al. 1989), populations of tliese toads from Ari-
zona have not been examined. Concern over
declining amphibian populations (Heyer et al.
1994) has increased interest in die possible nega-
tive effects of parasites on toads. The puqDOse of
tliis paper is to report on helminths of tliese toads
and dieir hybrids from Aiizona.
This investigation of parasitism in these toads
addresses a hypothesis of hybrid zone theory
and species boundaries. The hypothesis that
populations of hybrid individuals with reduced
fitness act as barriers to gene flow between 2
species separated by a hybrid zone (Biuton 1979,
1980) could have several mechanisms. One
mechanism, increased parasitism of hybrids, is
evaluated in this study. Two previous studies of
parasitism in vertebrates are split. Hybrid mice
{Mus muscuhis X Mus domesticus), specifically
backcrossed hybrids, had greater numbers of
cestode and nematode parasites than either
parental species (Sage et al. 1986). Prevalence
of monogenean parasites for hybrid minnows
{Barhus barhus X Barhus meridiomiUs) was
^Department of Biology, Whittier College, Whittier, CA 90608.
^Department of Biology; Pennsylvania State University, Shenango Valley Campus, 147 Shenango Avenue, Sharon, PA 16146.
■^Department of Life Sciences, Bo.x .37100, Arizona State I'niversity West, Phoeni.x, .AZ 8.5069.
369
370
Great Basin Naturalist
[Volume 56
positi^'el^• associated with the percentage of B.
meridionoUs genes (Le Brun et al. 1992). If we
find that hybrid toads have greater parasitism
than each toad species, then parasitism may be
a mechanism tliat reduces hybrid fitness and
contributes to the hairier between these 2 toad
species.
Materials and Methods
One hundred forty-six toads were collected
in Aiizona during 1991-1995; snout-vent length
(SVL) was measured to the nearest mm after a
minimum of 6 mon in 70% ethanol storage.
Toads were identified using a hybrid index (HI)
and advertisement call structure, if available.
Following Blair (1955), Sullivan (1986), and Sul-
livan and Lamb (1988), we evaluated die degree
of expression of 4 characters to generate the
HI score for each toad: dark ventral pigmenta-
tion, cranial crest, dorsal stripe, and pale colora-
tion across the eyelids. A numerical score (0, 1,
2, 3) was assigned for each of the following 4
character states: present, weakly present, very
weakly present, or absent. A score of 3 was
assigned for the presence of dark xentral pig-
mentation, cranial crests, a dorsal stripe, and
absence of a pale bar across the eyelids. This
yields scores near 12 (4 X 3) for B. woodhoiisii
and 0 (4 X 0) for B. microscaphus. Numerous
other studies of hybridization between toad
species have used a moiphological hybrid index
such as this (Volpe 1959, Meacham 1962, Hen-
rich 1968, Zweifel 1968). All toads from sites of
sympatiy with scores of 4 through 8 were con-
sidered hybrids, as were all toads with interme-
diate advertisement calls. Inteniiediate calls are
typical of hybrid toads between these species
(Sullivan 1995), and calls have long been used
to delimit hybrid toads of other species pairs
(Blair 1956, Zweifel 1968, Green 1982). Sev-
enty-seven Bufo microscaphus (mean SVL =
61.4 mm ± 8.7 s, range 34-86 nmi, 67 males,
10 females); 61 Bufo wooclhousii (mean SVL =
74.5 mm ± 8.8 s, range 49-91 mm, 53 males, 8
females), and 8 hybrids (mean SVL = 60.5 mm
± 8.4 s, range 45-72 mm, 7 males, 1 female)
were examined. Kruskal-Wallis test statistic
(45.92, 2 df, F < 0.001) indicates significant
difference in SVLs for the samples examined.
After examination all specimens were deposited
in the heipetology collection of Arizona State
University (ASU), Tempe. Collection localities
and ASU accession numbers are given in
Appendix 1.
Toads were anesthetized by immersion in 1
g/1 solution of tricaine methane sulfonate (MS-
222, Sigma, St. Louis, MO). Heart, liver, thigh
muscle, and kidne\' were remo\'ed and fiozen
for future genetic analyses. Toads were then
fixed in neutral-buffered 10% formalin and
moved to ethanol for storage following proce-
dure outlined by Simmons (1987). The body
cavity was opened by a longitudinal incision
from vent to throat, and the gastrointestinal
tract was removed by cutting across the esoph-
agus and rectum. The esophagus, stomach, small
intestine, large intestine, lungs, bladder, and
coelom were examined under a dissecting
microscope. No helminths were found in the
esophagus or urinar>' bladder. All helminths
were removed and identified using a glycerol
wet mount. Specimens were placed in vials of
alcohol and deposited in the U.S. National Par-
asite Collection, Beltsville, Maryland 20705:
(accession numbers, Appendix 1).
Results and Discussion
Prevalence, site, and mean intensity for each
parasite are given in Table 1. Temiinology is in
accordance with Margolis et al. (1982). One
species of trematode {Glijpthehnins quieta
[Stafford, 1900]), 1 species of cestode {Distoi-
chometra bufonis Dickey, 1921), and 5 species
of nematodes {Aplcctana inccrta Caballero,
1949, Aplectana itzocanensis Bravo Hollis, 1943,
Rlxahdicis americanus Baker, 1978, Fhijsaloptera
sp. [lai-vae only], and PhysocepJuiIus sp. [lanae
only]) were found. It would appear fiom Table
1 that both species and their hybrids are sus-
ceptible to infection by the same parasites. The
greatest prevalence (41%) and mean intensit)'
(231.7) in our study were recorded kr Aplectana
incerta in Bufo woodhousii. Thirty-four of 77
(44%) Bufo microscaphus (30/67, 45% males;
4/10, 40% females), 51 of 61 (84%) B. wood-
housii (45/53, 85% males; 6/8, 75% females),
and 4 of 8 (50%) hybrids (3/7 males, 1/1 female)
were infected. Males and females of both Bufo
microscaphus {x^ = 1.17, 1 df P > 0.05) and
B. woodhousii {x^ = 2.79, 1 df P > 0.05) did
not differ significantly in helminth prevalence.
There were too few female hybrid toads for
chi-square analysis. There was statistical differ-
ence in abundance of nematodes betw een B.
microscaphus and B. woodhousii (x^ = 23.72,
1996]
Arizona Toad Helminths
371
Table 1. Prevalence, mean intensih' (range), and location of lielmintlis from Btifo microscaphus, B. tvoodhousii. and
tlitir Iivbrids from Arizona.
Biifo microscaphus
Bitfo woodhousii
llv
l)rids
(.V = 77)
(\ = 6l)
i^'
= 8)
Fr
e\ale
lice
Intensih
Location
Pie\iilencc Intensih' L
ocation
Prevalence
Intensih
IjOcation
r.u, isite species
C^^)
(range)
(%)
(range)
(Vo
(i
range)
1 i;i \l VIODA
('.liiptlichiiiii.s quk'ta
1
1.0
h
2
2.0
1.
13
1.0
h
(JsiODA
Distoichoinctni hufoni.'i
14
2.9 (1-6)
h
38
2.0 (1-8)
b
13
1.0
b
Nl MVrODA
Aplectana incerta
1
156.0
b,c
41
231.7(23-564)
b,c
—
Aplectana itzocanensis
19
75.0 (1-.373)
b,c
26
43.2 (1-204)
b,c
25
1.0
c
Phtjsaloptera sp. (lar\;i)
16
5.5(1-31)
a
5
6.0(2-11)
a
13
1.0
a
Pinjsocephalm sp. (lar\ae)
1
104.0
d
—
—
RJialxlias americanus
5
2.0(1-31
V
38
21.7(1-11])
e
—
large intestine.s, d = cysts on stomach wall, e = lungs.
1 df, P < O.OOlj. When the intermediate
prexalence (50%) of the small hxhrid sample (N
= (S) was included in the chi-square calcula-
tion, statistical significant difference remained
iX- = 23.97, 2 df, P < 0.001).
To test for difference in infection rate, we
used a Kruskal-Wallis rank-order statistic be-
cause of the great variation in mean intensit}'
of parasites harbored by B. microscaphus, B.
woodhousii, and their hybrids (116.3, 19.4, 1.3,
respecti\'ely) and the relative!)' small sample of
In biids {N = 8). This test revealed that hybrid
indixiduals had fewer parasites than do indi-
viduals of either species. E.xamination of more
h\ brids could strengthen this result. Subsequent
work to determine the importance of age,
Uenetic factors, nutrition, and ecology would
also help to establish the significance of hybrid
ancestiy on parasite levels.
Infected frogs appeared healthy; i.e., none
\\ ere emaciated and there were no gross abnor-
malities. Thus, tlie presence of helminths did
not appear to adversely affect the populations
oi B. microscaphus, B. woodhousii, or their hy-
brids. In a stud\' on Couch's spadefoot (Scaphio-
pus couchii) fi-om Arizona, Tinsley (1990) found
no correlation between presence of the trema-
tode Pseudodiplorchis americanus and mating
success, although the presence of P. americanus
reduced fat reserves during hibernation.
Bufo microscaphus is a new host record for
Distoichometra bufonis, Aplectana incerta, A.
itzocanensis, Physocephahis sp., and Wwibdias
americanus. Bufo woodhousii is a new host
record for Aplectana incerta and Physaloptera
sp. Bufo w. woodhousii has been reported by
Baker (19'th
1994). Because these helminths are not species
specific and occur in a variet\- of amphibians,
the distribution of inteiTnediate hosts may play
an important role in determining the distribu-
tion of those parasites with indirect life cycles.
The conditions responsible for determining
distribution of the parasites with direct life
cvcles have vet to be defined.
372
Great Basin Naturalist
[Volume 56
Table 2. Helminth conimuniW of desert toads from Arizona.
Helminth
Host
Reference
Trematoda
Glypthelmins quieta
Pseudodiplorchis americamts
Cestoda
Distoichoinctra bitfonis
Nematotaenia dispar
Nematoda
Aplectana incei'ta
Aplectana itzocanensis
Oswaldocriizia pipiens
Physcdopfera sp. (larva)
Physocephaliis sp. (lai^va)
Wiahdias amcricaniis
Biijo microscaphiis
B. woodhoiisii
Scaphiopu.s couchii
Bufo cognatus
B. microscaphiis
B. punctatiis
B. retifonnis
B. woodhoiisii
Scaphiopits couchii
Bufo alvarius
Bufo uiicroscaphus
B. retifonnis
B. woodhousii
Scapliiopus couchii
Bufo alvarius
B. cognatus
B. microscaphiis
B. punctatiis
B. rctijonnis
B. woodhousii
Bufo alvarius
B. cognatus
B. punctatiis
B. retifonnis
Scaphiopus couchii
Bujo alvarius
B. cognatus
B. microscaphiis
B. retifonnis
B. woodhousii
Bujo alvarius
B. microscaphus
B. retifonnis
Bufo alvarius
B. cognatus
B. microscaphus
B. retifonnis
B. woodhousii
This study
This study
Tinsley 1990
Goldberg and Bursey 1991a
This study
Goldberg and Burse\' 19911)
Goldberg et al. 1996
This study
Goldberg and Bursey 1991a
Goldberg and Bursey 1991a
This study
Goldberg et al
This study
Goldberg and
Goldberg and
Goldberg and
This study
Goldberg and
Goldberg et al
This study
Goldberg and
Goldberg and
Goldberg and
Goldberg et al
Goldberg and
Goldberg and
Goldberg and
This study
Goldberg et al
This study
Goldberg and
This study
Goldberg et al
Goldberg and
Goldberg and
This study
Goldberg et al
This stud\
. 1996
Bursey
Bursey
Bursey
Bursey
. 1996
Burse\'
Bursey
Bursev
. 1996
Burse\
Bursey
Bursey
. 1996
1991a
1991a
1991a
1991b
1991a
1991a
1991b
1991a
1991a
1991a
Bursev 1991a
. 1996
Burse\
Burse\'
. 1996
1991a
1991a
Acknowledgments
Field assistance was provided by Mike
Demlong, Erik Gergus, and Matt Conley. The
Department of Life Sciences at Arizona State
University West provided a vehicle for trans-
portation to collection sites and funds for some
costs of this project. We thank Steven M. Nonis
and Michael E. Douglas for access to the Ari-
zona State University Vertebrate Collection and
the constructive comments of 2 anonymous
reviewers. Rachael Schuessler and Elizabeth
Stikkers assisted with collection of parasites.
Literature Cited
Anderson, R. C. 1992. Nematode parasites of vertebrates.
Their development and transmission. CAB Interna-
tional, Wiillingford, O.xon, U.K. 578 pp.
Baker, M. R. 1985. Redescription of Aplectana itzocanen-
sis and A. inceiia (Nematoda: Cosmocercidae) fi^om
amphibians. Transactions of the American Micro-
scopical Society 104: 272-277.
Barton, N. H. 1979. The dynamics of Inbrid zones.
HereditA 43: 341-359.
. 1980. The h>'brid sink effect. Heredit>- 44: 277-278.
Behler, J. L., AND E VV. KiN(.. 1979. The Audubon Societ>-
field guide to North American reptiles and amphib-
ians. Alfred A. Knopf New York. 743 pp.
1996]
Arizona Toad Hki.mimiis
373
lii.AiR, A. P. 1955. Distribution, variation, and hybridization
in a relict toad (Biifo iiiicroscaphiis) in sontliwi'stern
U tall. American Museum Novitates 1722: 1-38.
BiAiR, W. F. 1956. The mating calls of'lnbrid toads. Texas
Journal of Science 8: 350- 355.
Hii\\[n, B. B. 1936. Parasites of certain North C^arolina
Salientia. Ecolosjical Monographs 6: 491-532.
BiiooKS, D. R. 1976. Parasites ol' amphibians of the C.reat
Plains. Part 2. Plat\ helminths of amphibians in
Nebraska. Bulletin of the University of Nebraska
State Museum 10: 65-92.
(IwiPBELL, R. A. 1968. A comparative study of the para-
sites of certain Salientia from Pocahontas State Park,
Virginia. Virginia Journal of Sciencel9: 13-20.
I'^NTHAM, H. B., .AND A. PoRTKR. 1948. The parasitic fauna
of vertebrates in certain Canadian fresh waters, with
some remarks on their ecolog>', structure and impor-
tance. Prt)ceedings of the Zoological Society of Lon-
don 117: 609-649.
f'RANDSEN, J. C, AND A. W. Grundmann. 1960. The para-
sites of some amphibians of Utah. Journal of Para-
sitology 46: 678.
Goldberg, S. R., and C. R. Bursev. 1991a. Helminths of
tliree toads, Biifo (ili(iriii.s, Btifo cognatus (Bufonidae),
and Scaphiopus coiichii (Pelobatidae), fi'om southern
Arizona. Journal of the Helminthological Society of
Washington 58: 142-146.
. 1991b. Helminths of the red-spotted toad, Biifo
piiiwtafiis (Anura: Bufonidae), from southern Aiizona.
Joimial of the Helminthological Society' of Washing-
ton 58: 267-269.
(ioLDBERG, S. R., C. R. Bursey, B. K. Sullivan, and Q. a.
Truong. 1996. Helminths of the Sonoran green toad,
Bufo retifonnis (Bufonidae), from southern Arizona.
Journal of the Helminthological Societ\' of Washing-
ton 63: 120-122.
(iKEEN, D. M. 1982. Mating call characteristics of hybrid
toads (Bufo cnnericanus X B. fnwieri) at Long Point,
Ontario. Canadian Journal of Zoolog\ 60: 329.3-3297.
Hardin, E. L., and J. Janovt, Jr. 1988. Population dynam-
ics of Disfoicho)netra hufonis (Cestoda: Nematotaeni-
idae) in Bujo woodJwimi. Journal of Parasitology 74:
360- 365.
H ENRICH, T. W 1968. Moiphological evidence of secondaiy
intergradation between Bufo hemiophnjs Cope and
Bufo americanus Holbrook in eastern South Dakota.
Herpetologica 24: 1-13.
Heyer, W R., M. a. Donnelly, R. W McDiarmid, L. C.
Hayek, and M. S. Foster., editors. 1994. Measur-
ing and monitoring biological diversity. Standard
methods for amphibians. Smithsonian Institution
Press, Washington DC. 364 pp.
Jilek, R., and R. Wolfe 1978. Occurrence of Spinitectus
gracilis Ward and Magath 1916 (Nematoda: Spiiin-oi-
dea) in the toad (Bufo woodhousii fowleri) in Illinois.
Journal of Parasitology 64: 619.
KiNTZ, R. W. 1941. The metazoan parasites of some Okla-
homa Anura. Proceedings of the Oklahoma Academy
ofScience 21: 33-34.
KUNTZ, R. E., .\ND J. T. Sele 1944. An ecological study of
the metazoan parasites of the Salientia of Comanche
County, Oklahoma. Proceedings of the Oklahoma
Academy ofScience 24: 35-38.
Le Brun, N., F Renaud, P Berrebi, and A. Lambert
1992. Hybrid zones and host-parasite relationships:
effect on die evolution of parasitic specificits'. Evolu-
tion 46: 56-61.
Margolis, L., G. W. Esch, J. C. Holmes, A. M. Kuris,
and G. a. Schad. 1982. The use of ecological terms
in parasitology' (report of an ad hoc committee of the
American Society of Parasitologists). Journal of I^ara-
sitolog\' 68: 131-133.
McAllister, C. T, S. J. Uiton, and D. B. Cow. 1989. A
comparati\'e study of endoparasites in three species
of sympatric Bufo (Anura: Bufonidae), from Texas.
Proceedings of the Helminthological Society of Wash-
ington 56: 162-167.
Meacham, W R. 1962. Factors affecting secondaiy inter-
gradation between two allopatric populations in the
Bufo woodhou-sei complex. American .Midland Natu-
ralist 67: 282-304.
PARin; J. E., and a. W CiRUNDMANN. 1965. Species compo-
sition and distribution of the parasites of some com-
mon amphibians of Iron and Washington counties,
Utah. Proceedings of the Utah Academy of Science,
Arts and Letters 42: 271-279.
Price, A. H., and B. K. Sullivan. 1988. Bufo inicrosca-
phus Cope, southwestern toad. Catalogue of Ameri-
can Ainphibians and Reptiles 415: 1-3.
Prudhoe, S., AND R. A. BR.4Y. 1982. Platyhelminth para-
sites of the Amphibia. British Museum (Natural His-
toiy), Oxford University Press, London. 217 pp 4- 4
microfiche.
Rankin, J. S., Jr. 1945. An ecological study of the helminth
parasites of amphibians and reptiles of western Mass-
achusetts and vicinity. Journal of Parasitolog)' 31:
142-150.
Reiber, R. J., E. E. Byrd, and M. V. Parker. 1940. Certain
new and alread\' known nematodes from Amphibia
and Reptilia. Lloydia 3: 125-144.
Sage, R. D., D. Heyneman, K. Lim, and A. C. Wilson.
1986. Wormy mice in a hybrid zone. Nature 324:
60-63.
Simmons, J. E. 1987. Heipetologieal collecting and collec-
tions management. Socieb,' for the Stud\' of Amphib-
ians and Reptiles, HerjDetological Circular 16. 70 pp.
Smyth, J. D. 1994. Introduction to animal parasitolo,g\'. 3rd
edition. Cambridge University Press, New York. 549
pp.
Sullivan, B. K. 1986. Hybridization bet\\een tlie toads Bufo
rnicroscaphus and Bufo woodhousei in Arizona: mor-
phological variation. Journal of Heq^etolog)' 20: 1 1-21.
, 1995. Temporal stabilib.' in hybridization between
Bufo rnicroscaphus and Bufi woodhousii (Anura:
Bufonidae): behavior and moiphology. Journal of
Evolutionary Biolog)' 8: 233-247.
Sullivan, B. K., and T. Lamb. 1988. Hybridization between
the toads Bufo microscaphus and Bufo woodhousii in
Arizona: variation in release calls and alloz\nies.
Hei-petologica 44: 325-333.
TiNSLEY, R. C. 1990. The influence of parasite infection on
mating success in spadefoot toads, Scaphiopus couchii.
American Zoologist 30: 313-324.
Trowbridge, A. H., and H. M. Hefley. 1933. Preliminaiy
studies on the parasite fauna of Oklahoma anurans.
Proceedings of the Oklahoma Academy of Science
14: 16-19.
Volpe, E. P 1959. Experimental and natural Inliridization
between Bufo terrestris and Bufo fowh'ri. American
Midland Naturalist 61: 295-312.
Walton, A. C. 1938. The Nematoda as parasites of
Amphibia. IV. Transactions of the American Micro-
scopical Societ)' 57: 38-53.
374
Great Basin Naturalist
[Volume 56 ||
ZvVElFEL, R. G. 1968. Effects of temperature, bocK size,
and h\'bridization on mating calls of toads, Bufo a.
ainericaniis and Bufo woodhoiisii fowleri. Copeia
1968: 269-285.
Received 27 March 1996
Accepted 29 July 1996
Appendix 1
Localities and museum (ASU) numbers for specimens
examined:
Bufo microscaphus. Maricopa County (xV = 6) (34°00'N,
112°45'W, elev 603 m) ASU 30360-61, 30369-72; Yavapai
County (iV = 61); 7 fi-om (34°24'N, 112°13'W, elev. 1323
m) ASU 30328-31, 30347-49; 6 from (34°06'N, 112°09'W,
elev 603 m) (ASU 29166-67, 29170-71, 30351, 30375); 4
from (34°04'N, 112°09'W, elev 488 m) (ASU 30377,
30379-81); 34 from (34°05'N. 112°07"W, elev 616 m) ASU
28845-50; 28852-57, 29172-83, 303.34-40; 30386-88; 10
fi-om (34°24'N, 112°08'W, elev. 1140 m) ASU 30487-96;
Coconino County (N = 10) (34°24'N, 112°08'W, elev
2094 m) ASU 30477-86.
Bufo woodhousii: Maricopa County (N = 53); 14 from
(33°38'N, 112°28'W, elev 410 m) ASU 28821-27, 28829-
31, .30356-59; 19 from (.33°56'N, 112°08'W, elev 628 m)
ASU 28818-19, 28828, 28835, 30.362-64, 30366-68, 29151-
59; 2 fi-om (33°36'N, 112°15'W, elev 365 m) ASU 28834,
28836; 7 from (33°39'N, 112°14'W, elev 389 m) ASU
30497-503; 11 from (33°36'N, 112°11'W, elev. 372 m)
ASU 30504-14; Yavapai County {N = 8); 7 fiom (34°06'N,
112°09'W, elev 488 m) (ASU 29165, 29167-69; 30345,
30350, 30355, 30376); 1 from (.34°04'N, 112°09'W, elev
488 m) (ASU 30385).
Hybrids; 'i'avapai County (N - 8); 7 from (34°06'N,
ll'2°09'\V, elev 603 m) ASU 30346, 30352-54, 30373-74,
30382; 1 from (34°04'N, 112°09'W, elev 488 m) ASU
30378.
Accession numbers for helminths in the U.S. National
Parasite Collection (USNPC):
Bufo microscaphus: Distoichometra bufonis (85910); Ghjp-
thcbnins quieta (85921); Aplectana incerta (85911); Aplec-
tana itzocanensis (85912); Physalopteridae (85915); Phij.so-
cephalus sp. (85914); RJjabdias americamis (85913). Bufo
woodhousii: Distoichometra bufonis (85916); Ghjpthehnins
quieta (85921); Aph'ctana incerta (85917); Aplectana itzo-
canensis (85918); Ph\ salopteridae (85920); Rliabdias ameri-
camis (85919). Hybrids: Distoichometra bufonis (85922);
Ghjpthehnins quieta (85921); Aplectana itzocanensis (85923);
Physalopteridae (85924).
C;reat Basin Naturalist 5fi(4), © 199fi, pp. 375-376
JUVENILE RAZORBACK SUCKER {XYRAUCHEN TEXANUS)
IN A MANAGED WETLAND ADJACENT TO THE GREEN RIVER
Tiiiiotlw Moclck'l
Key words: razorhack sucker, floodplain, wctlaiKl. jiircnilc.
The razorback sucker {Xyrauchen fexanus)
is a kirge, endemic catostomid of the Cokirado
Ri\'er drainage. It was once widespread and
abundant throughout the basin (Minckley et ah
1991). Species abundance and distribution de-
cUned following construction of mainstem dams
and the introduction of many nonnative fishes
(Behnke and Benson 1983, Carlson and Muth
19(S9). The razorback sucker was federally
hsted as endangered in 1991 (USFWS 1991). '
The largest riverine population of razorback
sucker is in the middle Green River (Lanigan
and Tyus 1989). These fish spawn successfully
(Tais and Karp 1990), but Lanigan and Tyus
(1989) reported little or no recruitment. Razor-
back sucker larvae in the Green River drift
downstream from spawning sites (Robert Muth,
Lanal Fish Laboratory', Colorado State Univer-
sity, Fort Collins, CO), but few juvenile have
been found and little is known of their habitat
needs. Taba et al. (1965) captured 8 juveniles
(90-115 mm total length [TL]) from Colorado
Rixer back-water habitat in surveys from 1962
to 1964 between Moab and Dead Horse Point,
Utah. More recendy, Gutermuth et al. (1994)
collected 2 juveniles (37 mm and 39 mm) from
a lower Green River backwater in 1991 and 2
others (59 mm and 29 mm) in a backwater on
tlie Ouray National Wildlife Refuge in 1993
(Robert Muth, Larval Fish Laboratory, Colo-
rado State University, personal communication).
This note reports occurrence of juvenile and
adult razorback suckers in a wetland adjacent
to the Green River in Utah.
Old Charley Wash is a 60-ha wetland on
the Ouray National Wildlife Refuge in Uintah
County, northwest Utah, adjacent to river kilo-
meter (RK) 402 on the Green River. The wash
is a historical type IV wetland (Cowardin et al.
1979) with smartweed {Polygonum sp.) and
sago pond weed {Potarnogeton pectinatus) being
the primary aquatic plants. The natural levees
of the wetland have been reinforced with
dikes to retain water through the siunmer and
fall periods. Water in- and outflow is con-
trolled at flows <481 ni'^/s. Water enters the
inlet at river flows of approximately 240 m-^/s.
Typical management is to fill in spring and
then maintain water through the summer and
autumn.
The outlet structure at Old Charley Wash
was modified in April 1995 to facilitate fish
capture by creating a drainable, 12-m concrete-
lined channel in which fish could be concen-
trated and captured with seines.
Spring flow of the Green River peaked at
about 595 mVs in 1995 and inundated Old
Charley Wash between 23 May and 1 July.
Inundation was at flows >481 rn^/s. The wash
was dry prior to inundation. Maximum depth
of the wetland was >2 m. Fish in the wetland
were isolated from the river; when runoff sub-
sided, no additional water was added. Fishes
were sampled by fyke and trammel nets, min-
now and light traps, and seines. Collections
were weekly from 23 May to 1 July and ever>'
2 wk from 2 July to 31 August. The wetland was
drained from 25 September to 12 October, and
fishes were collected from the outlet every
other day during the first 2 wk and daily (except
9 October) during the 3rd week. Twenty-eight
juvenile razorback sucker were collected when
Old Charley Wash was drained in the fall of
1995 [x = 94 mm TL [range = 74-125 mm]
and 9.5 g [range = 3-18 g]; voucher speci-
mens, catalog number LFL 24874, Larval Fish
Laboratory, Colorado State University). Eight
(461-525 mm TL; 1034-1650 g) adults also
were captured, 6 prior to and 2 during the
draining process. A total of 10.1 metric tons of
'Colorado River Fish Project, U.S. Fish and VVildhfe Service, 266 West 100 North, Suite 2, Vernal, UT 84078.
375
376
Great Basin Naturalist
[Volume 56
fish were collected during draining. The iol-
lowing species were represented in order of
contribution by weight: Cyprinus carpio, Pime-
phales promelas, Lepoinis cijaneUus, Ictahinis
pimctatus, Ameiurus melas, Cyprinella hitren-
sis, Pomoxis nigromaculatus, Xijrauchen tex-
anus, Esox liicius, Gila atraria, Catostotmis
latipinnis, Catostomiis commersoni, Ptyclioche-
iliis liicins (7 individuals ranging in TL be-
tween 175 and 207 mm, and weight from 33 to
62 g), Gila rohiista, and Ciilaea inconstans.
Tyus and Karp (1990) reported that razor-
back sucker spawn on the ascending limb of
the hydrograph, allowing drifting larvae to
disperse during peak runoff and thus maximiz-
ing access to wetland habitats. It is unknown
whether the juveniles collected during drain-
ing originated from riverine spawning sites or
were produced in Old Charley Wash. How-
ever, their occurrence in Old Charley Wash in
1995 supports speculation (Tyus and Kaip 1990,
Modde et al. 1966) that floodplains may be
important razorback sucker nursery areas.
Support for this study was provided by the
Recovery Implementation Program for the
Endangered Fishes of the Upper Colorado
River Basin. Thanks to T Hatch, C. Flann, N.
Hoskin, D. Irving, B. Haines, R. Nicoles, K.
Day, and K. Kaczmarek for assisting in fish
collections.
Literature Cited
Behnke, R. J., and D. E. Benson. 19S3. Endangered and
threatened fishes of the Upper Colorado River basin.
Colorado State University Cooperative Extension
Service, Bulletin 503A.
Carlson, C. A., and R. T. Muth. 1989. The Colorado
River; lifeline of the American Southwest. Pages
220-239 ((1 D. R Dodge, editor. Proceedings of the
International Large River Symposium. Canadian Spe-
cial Publication of Fisheries and Aquatic Sciences
106, Ottawa.
CowARDiN, L. M., V. Carter, E C. Golet, and E. T. La
Roe. 1979. Classification of wetlands and deepwater
habitats of the United States. U.S. Department of
Interior, U.S. Fish and Wildlife Service, FWS/OBS-
79/31. 131 pp.
Gutermuth, F B., L. D. Lentsch, and K. R. Bestgen.
1994. Collection of age-0 razorback suckers {Xyrau-
chen texamis) in the lower Green River, Ut;ih. South-
western Naturalist 39: 389-391.
Lanigan, S. H., and Tyus, H. M. 1989. Population size
and status of the razorback sucker in tlie Green River
basin, Utah and Colorado. North American Journal
of Fisheries Management 9: 68-73.
MiNCKLEY, W. L., R C. Marsh, J. E. Brooks, J. E. John-
son, and B. L. Jensen. 1991. Management toward
the recovery of the razorback sucker Pages 303-357
in W. L. Minckley and J. E. Deacon, editors. Battle
against extinction: native fish management in the
American West. University of Arizona Press, Tucson.
MoDDE, T, K. P Burnham, and E. E Wigk. 1996. Popu-
lation status of the endangered razorback sucker in
the Middle Green River. Conservation Biologv 10:
110-119.
Tara, S. S., J. R. Murphy, and H. H. Frost. 1965. Notes
on die fishes of the Colorado River near Moab, Utiih.
Proceedings of the Utah Academy of Sciences, Aj-ts,
and Letters 42: 280-283.
Tyus, H. M., and C. A. Karr 1990. Spawning and move-
ments of razorback sucker, Xtjrauchen texamis, in the
Green River basin of Colorado and Utah. Southwest-
ern Naturalist 35: 427-433.
U.S. Fish and Wildlife Service (USFWS). 1991. Endan-
gered and threatened wildlife and plants: the razor-
back sucker {Xtjrauchen fexaniis). Determined to be
an endangered species. Federal Register 56(205):
54957-54967.
Received 6 November 1995
Accepted 21 June 1996
Creat Basin Naturalist 56(4). © 1996, pj:.. 377-378
CONFIRMATION OF COSEXUALITY IN PACIFIC YEW
{TAXUS BREVIFOLIA NUTT.)
K. E. Hoggi, A. K. Mitchell'-, and M. R. Clayton'
Key words: Pacific yew, Taxiis 1)il'\ iiolia, dioecious, cusexuality, British C'oliiinhia, pollen, seed.
Unlike most evergreen conifers in our forests,
which have both pollen and seed on a single
tree. Pacific yew {Taxiis brevifolia Nutt.) is
dioecious, the 2 sexes being segregated on dif-
ferent trees (Rudolf 1974, Taylor and Taylor
1981, Bolsinger and Jaramillo 1990, Hils 19'93).
In Jul\ 1993 branch samples of T. brevifolia
were taken from an undisturbed stand of
coastal Douglas-fir {Psciidotsuga menziesii) on
southern Vancouver Island (48°26'N. lat.;
123°28'W. long.) near Victoria, British Colum-
bia. One of the samples was obsei^ved to have
both male and female reproductive structures
(bud scales partially removed) on a single twig
(Fig. 1).
Occasionally, male and female structures
can occur on the same tree (Taylor and Taylor
1981). In the instances reported (Owens and
Simpson 1986, DiFazio 1995), female and male
structures occurred together only on branches
of predominantly male trees. We obsen'ed this
phenomenon, termed cosexuality (Lloyd 1980),
on a single yew tiee. On one branch, female and
male reproductive structures were observed
within a few mm of each other (Fig. 1) on an
otherwise male tree. The structures were visu-
ally identical to respective buds from other
dioecious trees. In a study by DiFazio (1995),
cosexuality was found in 17 of 58 male trees
(29.3%). It is not known whether these female
buds found on male trees produce viable seed.
Reproductive buds of the Pacific yew can
be visually differentiated throughout the year
(Taylor and Taylor 1981) and are usually located
on the underside of the shoot on noncurrent
growth. Male buds are small (2-3 mm), round,
and green, and they generally occur in clusters
(Fig. 2). They consist of a number of distinct
segments made up of pillowlike structures
(microsporangia) in which the pollen mature.
In spring microsporangia burst the bud scales
(Fig. 3) and pollen is released. Female buds
generally occur singly (Fig. 4) and are erect,
oval (2-3 mm), and green. The female bud
matures slowly through spring and summer
with the ovule (Fig. 5) growing through the
bud scales and revealing the micropyle (open-
ing for pollen). Beginning in late July or early
August, depending on location, a fleshy red aril
(berry) around the hard-coated seed becomes
visible.
Acknowledgments
The authors thank L. Kaupp (University' of
Victoria) for field sampling and L. Manning
(Pacific Forestiy Centre) for preparation of the
photographic plate.
Literature Cited
Bolsinger, C. L., and K. E. Jaramillo. 1990. Taxtis brevi-
folia Nutt.— Pacific yew. Pages 573-579 in R. M. Bums
and B. H. Honkala, editors, Silvics of North America:
1. Conifers. Agriculture Handbook 654. United States
Department of Agriculture, Forest Service, Washing-
ton, DC.
DiFazio, S. P. 1995. The reproductive ecology of Pacific yew
{Taxtis brevifolia Nutt.) under a range of o\'erstory
conditions in western Oregon. Unpubfished disserta-
tion, Oregon State University, CorvaUis. 178 pp.
HiLS, M. H. 1993. Taxus. Pages 424-426 in Flora of North
America Committee, editors, Flora of North .\merica.
Volume 2. Pteridophytes and Gymnosperms. O.xford
University Press, New York.
Lloyd, D. G. 1980. Sexual strategies in plants III. A (]uan-
titative method for describing the gender of plants.
New Zealand Journal of Botany 18: 103-108.
Owens, J. N., and S. Simpson. 1986. Pollen from conifers
native to British Columbia. Canadian Journal of For-
est Research 16: 955-967.
Rudolf, R O. 1974. Taxus. Pages 799-802 in Seeds of woody
plants in the United States. United States Depart-
ment of Agriculture, Forest Service, Washington, DC.
'Canadian Fcirest Service, Pacific Forestry Centre. 506 West Bumside Rd., Victoria, BC VSZ 1M5, Canada.
-.\uthor to whom all correspondence should be addressed.
377
378
Great Basin Naturalist
[Volume 56
Taylor, R. L., and S. Taylor. 1981. Taxus brevifolia in
British Columbia. Davidsonia 12(4): 89-94.
Received 28 March 1996
Accepted 5 June 1996
Figs 1-5 Scanniiis electron inicrotirapii confirming cose.xualitA in Pacific yew {Taxus brevifolia) fi-om soutliem yancouver
Island British Columbia. Scale bar = 1 mm in each Rgme. 1, Male bud (left) aud fennJe bud (right), both vvidi bud scales
partial'h' removed, on the same twig. 2, Young male bud (March) prior to shedding of pollen; bud scales intact. .3, \oung male
bud (March) showing the emerging microsporangia (M); bud seniles intact. 4, Young female bud (Mmdi), bud scdes intact. 5,
Mature female bud (August) showing die ovule tip (Ov) and micropyle emerging dirough die center of the intact bud scales.
( ;ifut Basin Naturalist 5(i(4), © 199(i, pp. 379-3S()
DIURNAL ABOVEGROUND ACTIVITY BY THE FOSSORIAL
SILVERY LEGLESS LIZARD, ANNIELLA PULCHRA
l^a\icl J. (icMiiiaiu)' and l)a\i(l J. Moralka-
Kcy words: uctivilij, lizardu, Anniella, Calijornki, reptiles, behavior.
Anniella pulchra is a limbless, fossorial lizard.
11iis species occurs from Antioch, California,
to northern Baja California, and is often found
on dune fonnations and in sandy habitats where
it t\picall> can be captured by raking the soil
luider bushes (Miller 1944). It can also be found
in se\eral low, coastal mountain ranges (Steb-
bins 1985), and its range extends into the San
Joaquin Valley and to the edge of the Sonoran
Desert in eastern San Diego County (Klauber
1932, Jennings and Hayes 1994). It seems to
prefer moist soils (Miller 1944) where it is able
to drink (Fusari 1985). Because of its fossorial
habit, A. pulchra is rarely found moving above-
ground, but it sometimes can be found on the
surface at dusk or in the evening (Stebbins
1 985). Here we report tlie previously unrecorded
finding of a single A. pulchra moving above-
ground during the middle of the day.
On 27 April 1995 we were driving on
Crocker Springs Road heading northeast over
the southern end of the Temblor Mountains.
This road is unpaved over the Temblors, and
we found 1 A. pulchra on a hard-packed sec-
tion of the road. The location was at 769 m
(2500 ft) in San Luis Obispo County, approxi-
mately 1.5 km west of the county boundary
\\ ith Keni County. The lizard, an adult male 140
nmi snout-vent length (217 mm total length),
w as found at approximately 1425 h.
The day was partly cloudy and the air tem-
perature when the lizard was found was about
24 °C. Although the road is not steeply inclined
at the location, the surrounding topography
traversed by this section of the road is a steep
hillside of about a 45-degree slope. Dominant
\ egetation on the hillside is alkali goldenbush
{Haplopappus acradenius), and no sandy soil
occurs near the location where we found the
lizard. The lizard was stretched out on the road.
which it probably was crossing when we saw it.
Unfortunately, we ran over the lizard with our
vehicle and were not able to watch its move-
ment after we found it. We salvaged the body
and deposited it in die museum of die Ctilifoniia
Academy of Science (specimen #CAS201173,
taken under California Department of Fish and
Game peniiit #1111). Besides the injuries we
inflicted on the specimen, there were no other
signs of injuiy or obvious infestations by para-
sites. This is the first obsei-vation we know of
showing that A. pulchra sometimes makes above-
ground movements during the day.
Midday aboveground activity of A. pulchra
appears to be a rare behavior. It is possible that
this lizard has narrow physiological tolerances
that often prevent surface activity, particularly
in full sun. A. pulchra has a lower prefeired body
temperature than most other lizards (Bury and
Balgooyen 1976), and its requirement for moist
soil and free water has been known for almost
a century (Coe and Kunkel 1907). We found
this lizard active at an air temperature of about
24 °C, which is consistent with its preferred
thermal range of 24-25 °C (Buiy and Balgooyen
1976). Limited surface acti\4t>', especially away
from plant cover, may also be due to predator
avoidance. Because limbless lizards are adapted
for liurrowing, their ability to move quickly
aboveground is limited (Cans 1975). These phys-
iological and behavioral constraints likely limit
the aboveground activity of A. pulchra to short
durations and distances.
Literature Cited
Bl'RY, R. B., and T. G. BALc;<)()Vii.\. 197tt. Tt'inperatuic
selectivity in the legless lizard, Anniella pitlehra.
Copeia 1976: 152-155.
Coe, W. R., and B. W. Kunkel. 1907. Studies of the Ciili-
fornia limbless lizard, Anniella. Transactions of the
'Departmrnt of Biology; California State University', Bakersfield, CA 93311.
^Department of Biolog>', California State University-, Dominguez Hills, Carson, CA 90747
379
380
Great Basin Naturalist
[Volume 56
Connecticut Academy of Arts and Sciences 12:
349-403 + plates.
FUSARI, M. H. 1985. Drinking of soil water by the Califor-
nia legless lizard, Anniella pulchra. Copeia 1985;
981-986.
Cans, C. 1975. Tetrapod limblessness; evolution and func-
tional con-elates. American Zoologist 15: 455-467.
Jennings, M. R., and M. P Hayes. 1994. Amphibian and
reptile species of special concern in California. Cali-
fornia Department of Fish and Game, Sacramento.
255 pp.
KijVUBER, L. M. 1932. Notes on the silver}' foodess lizard,
A7iniella pulchra. Copeia 1932; 4-6.
Miller, C. M. 1944. Ecologic relations and adaptations of
the limbless lizards of the genus Anniella. Ecological
Monographs 14: 271-289.
Stebbins, R. C. 1985. A field guide to western reptiles and
amphibians. Houghton Mifflin Company, Boston,
MA. 336 pp.
Received 11 March 1996
Accepted 14 June 1996
THE
GREAT BASIN
NATURALIST
N D E X
VOLUME 56 — 1996
BRIGHAM YOUNG UNIVERSITY
Great Basin Naturalist 56(4), © 1996, pp. 382-389
INDEX
Volume 56—1996
Author Index
Austin, Dennis D., 167
Baird, Craig R., 237
Barnard, David, 12
Barneby, R. C, 85
Barnum, Andrew H., 283
Bissonette, John A., 1, 319
Bleich, Vernon C, 276
Bloszyk, Jerzy, 59
Bosakowski, Thomas, 341
Bowker, Robert W, 38
Bradley Peter V, 48
Braker, H. Elizabeth, 172
Brusven, M. A., 22
Bursey, Charles R., 180, 369
Buskirk, Steven W, 247
Cashore, Brian L., 183
Cheam, Ha\', 369
Clancy, Karen M.. 135
Clary, W.E, 119
Clayton, M. R., 377
Conway, John R., 54, 326
Cranney, J. Stephen, 142
Crawford, John A., 177
Davidson, Diane W, 95
Davis, Jeffrey L., 276
Davis, Vicki L., 276
Dodds, Kimberly A., 135
Ducharme, Lori A., 333
Eckel, E M., 197
Ehleringer, James R., 333
Fielding, Dennis J., 22
Fraas, W. Wyatt, 205
Frisina, Michael R., 205
Gergus, Erik W. A., 38
Germano, David J., 379
Goldberg, Stephen R., 180, 369
Grayson, Donald K., 191
Greenberg, David, 135
Crier, Charles C, 211
Gutierrez, R. J., 87
Guyon, John C, 129
Haigh, Sandra L., 186
Haines, G. Bruce, 281
Halford, Anne S., 225
Hansen, E. Matthew, 348
Haiper, Kimball T, 95, 294
Hart, KimberK Hamblin, 188
Heckniann, Richard A., 142
Hoffman, James T, 129
Hogg, K. E., 377
Hubert, Wayne A., 300
Hysell, Molly Thomas, 211
Jenkins, Michael J., 28
Keiter, Robert B., 95
Kish, L. E, 22
Knapp, Eaul A., 162
KiTise, Carter G., 300
Ley\'a, Kathiyn J., 135
Lindquist, John L., 267
Livingston, Stephanie D., 191
Looman, Sandra J., 73
Malmos, Keith B., 38, 369
Manning, Sara J., 183
Mathiasen, Robert L., 129
Mattson, David J., 272
Mattson, Todd A., 247
Maxwell, Bruce D., 267
McAllister, Chris T, 180
McCutchen, Hemy E., 90
Mead, Leroy L., 294
Messmer, Terry A., 254
Mihuc, Janet R., 287
Mihuc, Timothy B., 287
Minshall, G. Wayne, 287
Mitchell, A. K., 377
382
1996]
Index
383
Modde, Timothy, 281, 375
Moiison, Clark S.. 150
Moratka, Da\i(lJ..379
Ncwniark, W'iliam I)., 95
Nowak. Rohcrt S., 225
IV'cn, M. Zachariah, 87
IVlren, Eric C, 177
IVttengill Thomas, 12
rierce! Beck>- M., 276
Popper, Kenneth J., 177
Porter, Eric E., 172
Ports, Mark A., 48
i'rice, Peter W, 135
|{ahel, Frank J., 300
Hamse\', R. Douglas, 341
Ik-dak,' Richard A., 172
Reinhart, Daniel P, 272
Rickart, Eric A., 95, 191
Romin, Laura A., 1
Schroeder, Sue, 254
Schultz, Brad W, 261
Seamans, Mark E., 87
Shaver, Monson W, III, 191
Sherman, Paul W, 237
Shiozawa, Dennis K., 95
Shirley, Dennis L., 73
Sites, Jack W., Jr, 95
Smith, Dvvight (;., 341
Smith, llohart M., 180
Soule, Peter T., 162
Stanton, Nancy L., 247
Steinkamp, Melanie J., 319
Sullivan, Brian K., 38, 369
Szewczak, Joseph M., 183
Szyinkowiak, Pavvcl, 59
Tanner, Wilmer W., 279
Tausch, Robin J., 261
Taylor, Vicki L., 294
Tiedemann, A. R., 119
Truong, Quynh A., 180
Tueller, Paul T, 261
Van Buren, Renee, 188
Vinyard, Gar\' L., 157, 360
Wcxdleigh, Linda L., 28, 129
Wambolt, Carl L., 205
Weaver, T, 267
Wells, Samuel A., 308
Welsh, Stanley L., 85, 93
White, Clayton M., 73
Wurtshaugh, Wayne A., 12
Yensen, Eric, 237
Yuan, Andv C, 157
Key Word Index
Taxa described as new to science in this volume appear in boldface type in this index.
Ahies lasiocarpa, 348
acid precipitation, 167
Acrididae, 172
activity, 379
A^ropijron spicatiiin, 267
alfalfa growers, 254
alkalinity, 167
alpine, 225
lakes, 167
amphibian decline, 38
Aiiniclhh 379
ant
mounds, 326
thatching, 54, 326
western hai-vester, 162
Arizona, 38, 180, 369
willow, 294
Astragalus laxmannii, 85
Barn Owl
food, 73
reproduction, 73
bat(s), 48
silver-haired, 247
bees, 95
behavior, 276, 379
big game nutrition, 205
bighorn sheep, 319
biodiversity, 95
biogeography, 191
bitterbrush, 205
black bears, 90
blood parasites, 142
Blue Grouse, 177
British Columbia, 377
Bromus inennis, 267
brook stickleback, 281
bnophvtes, 197
Biifo
microscaphus, 369
rctiformis, 38
bulk density, 211
bunch grass lizard, 180
Bureau of Land Management,
95
Buteo. 341
cactus
prickly pear, 211
California, 276, 379
native grassland, 172
Owens Valley, 183
carbon isotope ratios, 333
cations, 119
cattle, 319
Centaiirea
dijfiisa Lam., 22
maculosa, 267
Cercocarpus, 261
Cestoda, 180
characteristics,
site, 225
check dams, 211
checklist, 197
384
Great Basin Naturalist
[Volume 56
Chiroptera, 48
Choristoneura occidentalism 135
class
maturity, 261
Colorado, 54, 326
competition, 22, 267
consei"vation, 95, 360
biology, 300
cosexuality, 377
creel, 12
crude protein, 205
ciyptobiotic soils, 95
dams
check, 211
Daphnia. 157
decline
amphibian, 38
deer, 1
den, 276
Dendra^apiis ol)sciints. 177
dendrochronologv, 294
density, 172
bulk, 211
detritus, 211
diffuse knapweed, 22
dioecious, 377
dispersal, 87, 150
distribution
historic, 38
present, 38
disturbance, 319
diversity, 172
Douglas-fir, 129
Drijocoetes confitsus, 348
dwarf mistletoes, 129
ecology
seed, 333
shrub, 333
ectoparasites, 237
Elateridae, 308
endemic, 225
species, 95
endemism, 360
Ereiuichthys acros, 360
eruptive species, 135
exotic
plant(s), 183, 267
species, 95
Felis concolon 276
Festitca idahoensis, 267
filter feeders, 287
fire
freciuency, 28
scar, 28
fish
management, 12
parasites, 142
Flaming Gorge Resenoir, 150
flight periodicity, 348
floodplain, 375
flora, 197
foliar qualit}', 135
food habits, 90
forest
subalpine spruce-tir, 28
Formica obscuripes, 54, 326
Fourier series, 272
Gastrophrijne olivacea, 38
genetics, 300
Gila atraria, 142
GIS, 341
grasshopper herbivoiy, 172
Great Basin, 59, 191, 360
ground squirrels, 237
growth, 12
habit, 211
habitat. 1, 48, 341, 360
hai"vest, 12
Hawk
Red-tailed, 341
Swainson's, 341
helminths, 369
hemiparasite, 333
herbivoiy, 22
grasshopper, 172
heterotroph}, 333
highw a\' mortalit); 1
historic distribution, 38
holotype, 308
host, 186
hybrids, 369
Idaho, 237, 319
insect(s), 22
control, 348
phenology, 348
invasive plant, 183
irrigation, 360
island(s)
biogeography, 191
of fertility, 2il
juvenile, 375
Lahontan cutthroat trout, 157
Lahontan redside shiner, 157
Lasionycteris iwctivagans, 247
Lewisia
longipetala, 225
pygtuaca. 225
limber pine, 90
litter, 261
lizard(s), 379
bunch grass, 180
lodgepole pine, 129
mammals, 191
maturity class, 261
Melanoplus sanguinipes, 172
meristic
counts, 300
variation, 300
midden, 272
mineral nitrogen, 211
mistletoe(s), 186
dwarf, 129
mites, 59
Montana, 205
mortality
highway, 1
mountain
lion, 276
mahogany, 261
myrmecophiles, 326
Nassella pulchra, 172
needle age, 135
Negastrius, 308
atrosus, 308
rupicola, 308
solox, 308
stibicki, 308
Neotoma cinerea, 191
neot>'pe, 308
nest
boxes, 73
density, 162
sites, 341
size, 162
Nevada, 48, 360
nitrogen, 119
enrichment, 211
mineral, 211
total, 211
noctin^nal, 177
nomenclature, 85
nonnative, 281
North America, 85
nuptial flight, 54
nutrient cycling, 119
nutrition
big game, 205
OdocoiJetis hciuioniis. 1
Oncorhynchus
clarki henshawi, 157
tnykiss, 12
Oochoristica scclopori. 180
Oregon, 177
organic carbon, 211
Osprey 150
oviposition, 135
Ovis canadensis, 319
Owens Valle\; [California], 183
Owl,
Spotted, 87
1996J
Index
385
I'ac'ific yew, 377
randion hdliai'tus, 150
panisite(s), 1(S6
blood, 142
paiatype, 308
rlioradendron calijoniicwn, 186
phosphorus, 119
photosxnthesis, 333
llinnosoiiiatidae, 180
pine
lodgepole, 129
pinon-juniper woodlands, 21 1
planktix'on; 157
l)lant(s), 95
density, 225
invasive, 183
size, 225
pocket gopher, 183
Pogonoinynncx owyheei, 162
pollen, 377
Porter, 197
precipitation
acid, 167
predation, 157
preference, 135
present distribution, 38
prickly pear cactus, 211
productivity, 12
protein
crude, 205
Pternohylo fodiens, 38
Purshio tridentata, 205
Qiierciis gombclii, 119
rainbow trout, 300
range extension, 281
raptor, 150
razorback sucker, 375
recruitment, 261
Red-tailed Hawk, 341
regulation, 12
relationships
spatial, 261
reptiles, 379
reservoir, 12
Richardsonius egregiiis, 157
roadkill, 1
llocky M()untain(s), 90, 197
streams, 287
roost(s), 177, 247
runofif, 211
SV///.V, 294
saltccdar, 183, 186
Sccloponis scalaris, 180
ScoKtidae, 348
seed, 377
seedling, 261
shrub ecology, 333
sighting, 272
silver-haired bat, 247
site characteristics, 225
size selectivity, 157
snags, 247
snow-bed vegetation, 225
soil(s)
characteristics, 211
ciyptobiotic, 95
nutrients, 119
Sonoran Desert, 38
southern Utah, 294
spatial relationships, 261
species
endemic, 95
-environment relationships,
287
eruptive, 135
exotic, 95
Spennophihis hrunneus, 237
Spotted Owl, 87
springs, 360
squirrels
ground, 237
stem diameter, 294
strains, 12
Strix occidentalism 87
subalpine spruce-fir forest, 28
sucker
razorback, 375
sulfur, 119
sui^veys, 129
Swainson's Hawk, 341
tamarisk, 183
Tainahx rainosi.s.siinu, 183, 186
Taxiis hrevifolia, 377
temperature, 276
thatching ant, 54, 326
Thomomys bottac, 1 83
total nitrogen, 211
Trachytes kaliszewskii, 59
transect, 272
trout, 12
rainbow, 300
Yellowstone cutthroat, 300
Trypanoplasma atraria, 142
turbidity, 157
Tyto (dha, 73
Uropodina, 59
Utah, 59, 73,95, 119, 167
forests, 348
southern, 294
vegetation
clearing, 162
snow-bed, 225
vertebrates, 95
vocalization, 272
water quality, 167
weed control, 267
western harvester ants, 162
western spruce budworm, 135
wetland, 375
wilderness, 95
wildfire, 272
wildlife
damage management, 254
damage perceptions, 254
management, 254
winter range, 205
Wyoming, 197
Yellowstone cutthroat trout, 300
yield, 12
386 Great Basin Naturalist [Volume 56
TABLE OF CONTENTS
Volume 56
No. 1— January 1996
Articles
Temporal and spatial distribution of highway mortality of mule deer on newly constructed
roads at Jordanelle Resei-voir, Utah Laura A. Romin and John A. Bissonette 1
Exceptional fish yield in a mid-elevation Utah trout reservoir: effects of angling regulations
Wayne A. Wurtsbaugh, David Barnard, and Thomas Pettengill 1 2
Consumption of diffuse knapweed by two species of polyphagous grasshoppers (Orthoptera:
Acrididae) in southern Idaho Dennis J. Fielding, M. A. Brusven, and L. R Kish 22
Fire frequency and the vegetative mosaic of a spruce-fir forest in northern Utah
Linda Wiidleigh and Michael J. Jenkins 28
Arizona distribution of three Sonoran Desert anurans: Bitfo retifonuis, Gastrophrijne oliiacea, and
Pteniohyla fodiens Brian K. Sullivan, Robert W. Bowker, Keith B. Malmos,
and Erik W. A. Gergus 38
Habitat affinities of bats from northeastern Nevada Mark A. Forts and Peter V Bradley 48
Nuptial, pre, and postnuptial activity of the thatching ant, Formica obsciiripes Forel, in Colorado
John R. Conway 54
Trachytes kaliszewskii n. sp. (Acari: Uropodina) from the Great Basin (Utah, USA), with remarks on
the habitats and distribution of the members of the genus Trachytes Jerzy Bloszyk
and Pawel Szymkowiak 59
Productivity, food habits, and associated variables of Bam Owls utilizing nest boxes in north central
Utah Sandra J. Looman, Dennis L. Shirley, and Clayton M. White 73
Notes
Astragalus laxmannii Jacquin (Leguminosae) in North America . . . . R. C. Barneby and S. L. Welsh 85
Intermountain movement by Mexican Spotted Owls [Strix occidentalis lucida) R. J. Gutierrez,
Mark E. Seamans, and M. Zachariah Peery 87
Limber pine and bears Heniy E. McCutchen 90
Book Review
Utah wildflowers: a field guide to northern and central mountains and valleys. Richard J. SJiaic . . .
Stanle\' L. Welsh 93
No. 2— April 1996
Articles
Selecting wilderness areas to conserve Utah's biological dixersitv' Diane W Daxidson,
William D. Newniark, Jack W Sites, Jr, Dennis K. Shiozawa, Eric A. Rickart,
Kimbcill T. Haiper, and Robert B. Keiter 95
Nutrient distribution in (^)iierciis gamhehi stands in central Utah A. R. Tiedemann
and W P Claiy 1 1 9
Comparsion of two roadside sui-vey procedures for dwaif mistletoes on the Sawtooth National Forest,
Idaho Robert L. Mathiasen, James T. Hoffman, John C. Gu\on, and Uinda U. Wadleigh 129
1996] Index 387
Effects of Doiitilas-fir foliage age class on western spruce hudwonn oviposition choice and lan'al
performance Kinil)erly A. Dodds, Karen M. Clancy, Kathryn J. Leyva,
David Greenberg, and Peter W. Price 1 35
Tnjpanophmiui atrarid sp. n. (Kinetoplastida: Bodonidae) in fishes from tlu^ Sevier River drainage,
Utah j. Stephen Craimey and Richard A. Heckniann 142
Geographical re\ic\\ of the liistorical and cnrrent stains of Ospreys {Pandion huliaetus) in Utah
Clark S. Monson 150
Effects of turbidity on feeding rates of Lahontan cutthroat trout {Orworhi/ncluis clarki henslmwi)
and Lahontan redside shiner {Richardsonius egrc nopsis of the mosses of Wyoming P M. Eckel 1 97
Variation in bitterbrush {Purshia tridentata Pursh) crude protein in southwestern Montana
Carl L. Wambolt, W. Wyatt Fraas, and Michael R. Frisina 205
Dam-forming cacti and nitrogen enrichment in a piiion-juniper woodland in northwestern
Arizona Molly Thomas Hysell and Charles C. Grier 211
Distribution and ecological characteristics of Leivisia longipetala (Piper) Clay, a high-altitude
endemic plant Anne S. Halford and Robert S. Nowak 225
Larger ectoparasites of the Idaho ground squiirel {Spennophilus brunneus) Eric Yensen,
Craig R. Baird, and Paul W. Sherman 237
388 Great Basin Naturalist [Volume 56
Roost sites of the silver-haired bat {Lasionycteris noctivagans) in the Black Hills, South Dakota
Todd A. Mattson, Steven W. Buskirk, and Nancy L. Stanton 247
Perceptions of Utah alfalfa growers about wildlife damage to their hay crops: implications for
managing wildlife on private land Teny A. Messmer and Sue Schroeder 254
Spatial relationships among young Cercocarpus ledifoliiis (curlleaf mountain mahogany)
Brad W. Schultz, Robin J. Tausch, and Paul T. Tueller 261
Potential for controlling die spread of Ccntaitrca macuhsa with grass competition
John L. Lindquist, Bruce D. Maxwell, and T. Weaver 267
Indicators of red squirrel {Tamiascinnis hiidsonicus) abundance in the whitebark pine zone
David J. Mattson and Daniel P Reinhart 272
Thermal characteristics of mountain lion dens Vernon C. Bleich, Becky M. Pierce,
Jeffrey L. Davis, and Vicki L. Davis 276
James William Bee, 1913-1996 Wilmer W. Tanner 279
Note
Brook stickleback {Ciilaea inconstans [Kirtland 1841]), a new addition to the Upper Colorado River
Basin fish fauna Timothy Modde and G. Bruce Haines 281
Errata 282
Book Review
Snakes of Utah. Doiidas C. Cox and Wihner W. Tanner Andrew H. Bamum 283
No. 4— October 1996
Articles
Species-environment relationships among filter-feeding caddisflies (Trichoptera: Hydropsychidae) in
Rocky Mountain streams Timothy B. Mihuc, G. Wayne Minshall, and Janet R. Mihuc 287
Stem growth and longevity dynamics for Salix arizonica Dorn Vicky L. Ta\'lor
Kimball T. Harper and Leroy L. Mead 294
Sources of variation in counts of meristic features of Yellowstone cutthroat trout {Oncorhijnchiis
clarki boiivicri) Carter G. Kruse, Wayne A. Hubert, and Frank J. Rahel 300
Studies on Nearctic Negastriiis (Coleoptera: Elateridae) Samuel A. Wells 308
Bighorn sheep response to ephemeral habitat fragmentation b\' cattle J. A. Bissonette
and Melanie J. Steinkamp 319
A field study of the nesting ecology of the thatching ant, Fonnica ohsciiripes Forel, at high altitude
in Colorado J"hn R. Conwax 326
Gas exchange, 5^'^C, and heterotrophy for CastiUeja linariifolia and Orflnx^arpus tohnici, faculta-
tive root hemiparasites on Artemisia tridentata Lori A. Ducharme
and James R. Ehleringer 333
Habitat and spatial relationships of nesting Swainson's Hawks [Buteo swainsoni) and Red-tailed
Hawks {B. jamaicensis) in northern Utali Thomas Bosakowski, R. Douglas Ramsey,
and Dwight G. Smith 341
Western balsam bark beetle, Drijocoetes confusns Swaine, flight periodicity in northern Utah
E. Matthew Hansen 348
Distribution ol a thermal endemic iniimow, the desert dace (Ereinichtluis aeros), and obsen'ations of
impacts of water diversion on its population Gary L. Vinyard 360
1996] Index 389
Helminths of the southwestern toad, Bufo iiiicroscaphus, Woodhoiise's toad, Btifo ivoocUwtisii
(Butonidae), and their h> hrids troni eentral Arizona Stephen R. Goldberg,
Charles K. Bursey, Keith B. Malnios, Brian K. Snlli\an, and Hay C^heani 369
Notes
Juvenile razorhaek sneker (Xi/rdiichoi fe.xaiiKs) in a inana^ed wetland adjaeenl to the Green River
Timothy Modde 375
Cont'irmation ol'eosexualit) in Faeiiie yew [Taxiis brevifolia Nutt.) . . . . K. E. Hogg, A. K. Mitchell,
and M. R. Clayton 377
Diurnal abo\ egiound aeti\it>' b\' the fossorial silvery legless lizard, Anniella pulchra
David J. Germano and David J. Morafka 379
p
INFORMATION FOR AUTHORS
The Great Basin Naturalist welcomes previously
unpublished mauuseripts pertaining to the biologi-
cal natural history of western North America.
Preference will be given to concise manuscripts of
up to 12,000 words. Simple species lists are dis-
coin-aged.
SUBMIT MANUSCRIPTS to Richard W. Baumann,
Editor, Great Basin Naturalist, 290 MLBM, PO Box
20200, Brigham Young University, Provo, UT
84602-0200. An accompanying cover letter must
include phone number(s) of the author submitting
the manuscript, and FAX number and E-mail
address when applicable; the letter must also pro-
\ ide information describing the extent to which data,
text, or illustrations have been used in other papers
or books that are published, in press, submitted, or
soon to be submitted elsewhere. Authors should
adhere to the following guidelines; manuscripts not
so prepared ma\' be returned for revision.
MANUSCRIPT PREPARATION. In general, the Great
Basin Naturalist follows recommendations in
Scientific Style and Format: The CBE Manual for
Authors, Editors, and Publishers, 6th edition
(Council of Biology Editors, Inc., 11 South LaSalle
Street, Suite 1400,' Chicago, IL 60603, USA; phone
312-201-0101; FAX 312-201-0214). We do, however,
differ in our treatment of entries in Literature Cited.
\ Authors may consult Vol. 51, No. 2 of this journal
for specific instructions on format; these instruc-
tions. Guidelines for Manuscripts Submitted
TO the Great Basin Naturalist, are printed at the
back of the issue. Also, check the most recent issue
of the Great Basin Naturalist for changes.
TYPE AND DOUBLE SPACE all materials, including
literature cited, table headings, and figure legends.
Avoid hyphenated words at the right-hand margins.
Underline words to be printed in italics. Use stan-
dard bond (22x28 cm), leaving 2.5-cm margins on
all sides.
SUBMIT 3 COPIES of the manuscript (5 copies of
fish manuscripts) and the original on a 5.25- or 3.5-
inch disk utilizing WordPerfect 5.1 or above.
j Number all pages and assemble each copy sepa-
! rately: title page, abstract and key words, text,
i acknowledgments, literature cited, appendices,
tables, figure legends, figures.
TITLE PAGE includes an informative title no longer
than 15 words, names and addresses of authors, a
running head of fewer than 40 letters and spaces,
footnotes to indicate change of address and author
to whom correspondence should be addressed if
other than the first author.
ABSTRACT states the purpose, methods, results,
and conclusions of the research. It is followed by
, 6-12 key words, listed in order of decreasing
importance, to be used for indexing.
TEXT has centered main headings printed in all
capital letters; second-level headings are centered
in upper- and lowercase letters; third-level head-
ings begin paragraphs.
VOUCHER SPECIMENS. Authors are encouraged to
designate, properly prepare, label, and deposit
high-(juality voucher specimens and cultures docu-
menting their research in an established permanent
collection, and to cite the repository in publication.
REFERENCES IN THE TEXT are cited by author and
date: e.g., Martin (1989) or (Martin 1989). Multiple
citations should be separated by commas and listed
in chronological order. Use "et al." after name of
first author tor citations having more than two
authors.
ACKNOWLEDGMENTS, under a centered main
heading, include special publication numbers when
appropriate.
LITERATURE CITED, also under a centered main
heading, lists references alphabetically in the fol-
lowing formats;
Mack, G. D., and L. D. Flake. 1980. Habitat rela-
tionships of waterfowl broods on South Dakota
stock ponds. Journal of Wildlife Management
44: 695-700.
Sousa, W. E 1985. Disturbance and patch dynamics
on rocky intertidal shores. Pages 101-124 in
S. T. A. Pickett and E S. White, eds., The ecolo-
gy of natural disturbance and patch dynamics.
Academic Press, New York.
Coulson, R. N., and J. A. Witter. 1984. Forest ento-
mology: ecology and management. John Wiley
and Sons, Inc., New York. 669 pp.
TABLES are double spaced on separate sheets and
designed to fit the width of either a single column
or a page. Use lowercase letters to indicate foot-
notes.
PHOTOCOPIES OF FIGURES are submitted initially
with the manuscript; editors may suggest changes.
Lettering on figures should be large enough to
withstand reduction to one- or two-column width.
Originals must be no larger than 22x28 cm.
NOTES. If the manuscript would be more appro-
priate as a short communication or note, follow the
above instructions but do not include an abstract.
A CHARGE of $50 per page is made for articles
published; the rate for individual subscribers will
be $35 per page. However, manuscripts with com-
plex tables and/or numerous half-tones will be
assessed an additional charge. Reprints may be pur-
chased at the time of publication (an order form is
sent with the proofs).
FINAL CHECK;
• Cover letter explaining any duplication of
information and providing phone number(s),
FAX number, and E-mail address
• 3 copies of the manuscript and WordPerfect
diskette
• Conformity with instructions
• Photocopies of illustrations
(ISSN 001 7-361 4)
GREAT BASIN NATURALIST Vol 56 no 4 October 1 996
CONTENTS
Articles
Species -environment relationships among filter- feeding caddisflies (Trichoptera:
Hydropsychidae) in Rocky Mountain streams Timothy B. Mihuc,
G. Wayne Minshall, and Janet R. Mihuc 287
Stem growth and longevity dynamics for Salix arizonica Dorn
Vicky L. Taylor, Kimball T Harper, and Leroy L. Mead 294
Sources of variation in counts of meristic features of Yellowstone cutthroat trout
{Oncorhijnchiis clarki bouvieri) Carter G. Kruse, Wayne A. Hubert,
and Frank J. Rahel 300
Studies on Nearctic Negastrius (Coleoptera: Elateridae) Samuel A. Wells 308
Bighorn sheep response to ephemeral habitat fragmentation by cattle
J. A. Bissonette and Melanie J. Steinkamp 319
A field study of the nesting ecology of the thatching ant, Formica obscuripcs
Forel, at high altitude in Colorado John R. Conway 326
Gas exchange, S^-^C, and heterotrophy for Castilleja linariifolia and Orthocarpus
tolmiei, facultative root hemiparasites on Artemisia trideniata
Lori A. Duchamie and James R. Ehleringer 333
Habitat and spatial relationships of nesting Swainson's Hawks {Buteo swainsoni)
and Red-tailed Hawks {B. jamaicensis) in northern Utah
Thomas Bosakowski, R. Douglas Ramsey, and Dwight G. Smith 341
Western balsam bark beetle, Drijocoetes confusus Swaine, flight periodicity in
northern Utah E. Matthew Hansen 348
Distribution of a thermal endemic minnow, the desert dace {Eremichthijs acros),
and observations of impacts of water diversion on its population
Gaiy L. Vinyard 360
Helminths of the southwestern toad, Bufo microscaphus, Woodhouse's toad, Bufo
woodhousii (Bufonidae), and their hybrids from central Arizona
Stephen R. Goldberg, Charles R. Bursey, Keith B. Malmos,
Brian K. Sullivan, and Hay Cheam 369
Notes
Juvenile razorback sucker {Xyrauchen texanus) in a managed wedand adjacent to
the Green River Timothy Modde 375
Confirmation of cosexuality in Pacific yew {Taxus brevifolia Nutt.)
K. E. Hogg, A. K. Mitchell, and M. R. Clayton 377
Diurnal aboveground activity by the fossorial silvery legless lizard, Anniella
pulchra David J. Germano and David J. Morafka 379
Index to Volume 56 381
!ii:isiiii>!ir'!lli!'
3 2044 072 231 145