2010 NORTHEASTERN NATURALIST 17(3):505–516
Survival and Winter Diet of Sylvilagus obscurus
(Appalachian Cottontail) at Dolly Sods, West Virginia
Alana C. Hartman1,* and Ronald E. Barry2
Abstract - The distribution of Sylvilagus obscurus (Appalachian Cottontail) is disjunct
and restricted to high-elevation refuges in the central and southern Appalachian
Mountains. The purpose of this study was to determine survival and winter diet of this
rabbit at its type locality, the Dolly Sods area of the Monongahela National Forest,
WV. To estimate survival, 44 Appalachian Cottontail individuals were radio-tracked
until death or loss of transmitter signal between October 1997 and June 2000. The
Kaplan-Meier estimate was used to generate finite survival rates. To assess winter
diet, stems browsed within a 1-m radius of winter radiolocations of 15 individuals in
1998–1999 and 1999–2000 were identified and counted. Species and groups of species
browsed were compared to availability, determined by counting the number of
woody stems within a 1-m radius of the same radiolocations. Overall daily survival
rate was 0.9934, finite monthly (28-day) survival rate was 0.8309, and finite yearly
survival rate was 0.0894. No differences in survival were found between sexes or
age groups. The first leaf-off season had lower daily survival rates than those of the
subsequent leaf-on and leaf-off seasons. Gaultheria procumbens (Eastern Teaberry),
Vaccinium spp. (blueberries), Gaylussacia baccata (Black Huckleberry), and Photinia
spp. (chokeberries) were preferred winter browse. Rhododendron spp. and the
abundant Kalmia latifolia (Mountain Laurel) were consumed less than expected.
Introduction
The group of cottontails once considered to comprise the species Sylvilagus
transitionalis Bangs (New England Cottontail) consists of 2 distinct
cytotypes (Ruedas et al. 1989). On the basis of morphometric data, Chapman
et al. (1992) recognized 1 of the cytotypes, the population southwest of the
Hudson River, as S. obscurus Chapman, Cramer, Deppenaar, and Robinson
(Appalachian Cottontail). This taxon has a limited distribution within
the Appalachian Mountain range, from central and eastern Pennsylvania to
northwestern Georgia and north-central Alabama, where it occurs at high
elevations in a mosaic of refugial relicts (Chapman and Stauffer 1981, Chapman
et al. 1982). In West Virginia, it is designated an S3 species (only 21–100
documented occurrences; West Virginia Natural Heritage Program 2007).
Cottontail survival may relate to abundance and distribution of cover,
especially in winter (Trent and Rongstad 1974). Predation is the primary
mortality factor for cottontails (Boland and Litvaitis 2008, Chapman et al.
1982), and previous studies have documented mammalian and avian predators
of Appalachian Cottontails (Barry et al. 1997, Sommer 1997).
1West Virginia Department of Environmental Protection, HC 63 Box 2545, Romney,
WV 26757. 2Biology Department, Bates College, 44 Campus Avenue, Lewiston, ME
04240. *Corresponding author - alana.c.hartman@wv.gov.
506 Northeastern Naturalist Vol. 17, No. 3
Exposure and vulnerability of cottontails to predation is greater during
winter due to the scarcity of high-quality food (Villafuerte et al. 1997). The
winter diet of Appalachian Cottontails is not well known. Spencer (1985)
compared stomach contents of Sylvilagus floridanus J.A. Allen (Eastern
Cottontails) and Appalachian Cottontails in western Maryland and West
Virginia. Most (85%) of the winter (January–March) diet of Appalachian
Cottontails in western Maryland and West Virginia was woody material.
Despite their predominance in the understory, Kalmia latifolia (Mountain
Laurel) and Rhododendron maximum L. (Great Rhododendron) were absent
from stomachs of rabbits in Maryland, and Mountain Laurel and Vaccinium
spp. (blueberry) from Appalachian Cottontails in West Virginia. Spencer
(1985) suggested that plants more palatable to rabbits occurred in these
areas, including Betula lutea Michx. f. (Yellow Birch) and Crataegus spp.
(hawthorn). Mountain Laurel and rhododendrons produce an andromedotoxin
that could render them unpalatable to rabbits. Picea rubens Sarg. (Red
Spruce) was ingested by some Appalachian Cottontails. A relationship was
implied between the use of Tsuga canadensis (L.) Carr (Eastern Hemlock)
in the diet and snow cover or inclement weather (Spencer 1985).
The purposes of our study were to examine survival rates of Appalachian
Cottontails and compare these between adults and subadults, and between
males and females, during leaf-off (October–April) and leaf-on (May–September)
seasons. We also investigated selection of woody plant species in
the winter diet. Male Appalachian Cottontails at Dolly Sods traveled farther
between successive resting locations than females did in the leaf-on season
(Boyce 2001), so we suspected that males would have lower survival rates
than females. We also expected that monthly survival during leaf-off seasons
would be lower than during leaf-on seasons. We examined winter forage use
to determine whether, consistent with previous studies, Appalachian Cottontails
would select Rubus spp. and the bark, buds, and small branches of
Acer spp. (maple) and Betula spp. (birch) trees. In addition, we investigated
whether they would browse Red Spruce and Eastern Hemlock and avoid
Great Rhododendron and Mountain Laurel.
Because stochastic extinction events can occur in small populations,
knowledge of such critical factors as survival and winter diet of Appalachian
Cottontails is important. Survival rates are essential elements of population
trend analyses. Wildlife managers can consider preferred woody plants when
developing habitat plans for Appalachian Cottontails.
Methods
Study area
Dolly Sods is a high-elevation (790–1260 m) area in the Monongahela
National Forest of northeastern West Virginia (Tucker and Grant counties,
39°01'N, 79°19'W) consisting of a 4134-ha wilderness area and a 971-ha
scenic area (Venable 1996). Its climax forest of Red Spruce was logged
and burned in the late 1800s, and the area has since become characterized
by scrubby plains and bogs interspersed with stands of spruce and Eastern
2010 A.C. Hartman and R.E. Barry 507
Hemlock and boulder fields (Chapman and Morgan 1973; Chapman et al.
1977; A.C. Hartman, pers. observ.). Shrubs include Great Rhododendron,
Mountain Laurel, and blueberry. Most of our work was done in the vicinity
of Forest Road 75, which runs north to south through the area, and the Red
Creek Campground. The climate is generally cool, windy, and wet, with
>130 cm of precipitation per year. The area is subject to frosts year-round
and frequent ice storms in winter, and annual snowfall can reach 380 cm
(Venable 1996). Dolly Sods supports 2 other lagomorph species, Lepus
americanus Erxleben (Snowshoe Hare) and Eastern Cottontail.
Procedures
The study was conducted from October 1997 through April 2000. Fortyfour
Appalachian Cottontails were captured in wooden livetraps (18 by 22 by
60 cm) baited with apple slices or alfalfa pellets. Rabbits in traps were transported
to Frostburg State University, Frostburg, MD, where they were weighed,
anesthetized (30–45 mg ketamine HCl/kg body mass), measured (total length,
hind foot, and ear), and gender recorded. Initial species identification was based
on external characteristics and a discriminant function originally developed to
differentiate New England Cottontails and Eastern Cottontails (Litvaitis et al.
1991). Species identification was later confirmed by electrophoretic analysis
of blood serum proteins (Morgan and Chapman 1979, Sommer 1997). Rabbits
were ear-tagged and radiocollared. One collar type was designed for rabbits and
hares (Lotek Engineering, Inc., Newmarket, ON, Canada), and we modified the
other style (Advanced Telemetry Systems, Isanti, MN), designed for Bonasa
umbellus L. (Ruffed Grouse), to fit rabbits. Both styles had an expected battery
life of 14–18 months, and both had mortality/motion sensors.
Monitoring and survival estimates. Rabbits were released at capture
sites, usually within 24 h. We tracked rabbits directly to their daytime resting
locations on average 2 times per week (with considerable variation) and at
various times during daylight hours. In most cases, rabbits were visible in,
or fleeing from, the resting location; if not, the presence of pellets, browse, a
small, oval depression, or melted snow marked the spot from which they had
fled. Each radiolocation was field-marked. We compiled the dates on which
each rabbit entered the study (i.e., capture date) and those on which each
rabbit became no longer part of the study. For rabbits that died (n = 33), we
used the date when the mortality signal was first heard as the end date. For
rabbits whose radio signals were lost (n = 8), we used the last date on which
they were seen alive or their signal was heard as the end date. These 8 animals
were considered “censored” on this end date in all survival estimates.
We saw no justification to use any “adjustment period” (Pollock et al. 1989)
to exclude the few rabbits that died shortly after release. Two survivorship
curves were generated, 1 based on trap mortalities being considered as censoring
events, and the other based on trap mortalities being considered as
death events. We statistically compared the 2 curves using the log-rank test
(Pollock et al. 1989), and the null hypothesis that the curves are equal could
not be rejected (P > 0.05 for the most conservative χ2 test; Cox and Oakes
1984). Therefore, we treated trap mortalities as death events for all survival
508 Northeastern Naturalist Vol. 17, No. 3
estimates. We compared rabbit survival by weight (i.e., age class), season,
and sex. Individuals ≤700 g were considered subadults, and those >700 g
were identified as adults. To determine whether to include the 4 subadults
in survival estimates, we performed the log-rank test on the survival curves
based on all individuals and only adults. The null hypothesis of equality/
similarity could not be rejected (P > 0.05), so the 2 data sets appear to be
equally representative samples of the population. However, we chose not to
include subadults in the survival estimates that were partitioned by sex and
season. The sample size of subadults was too small to reveal a statistical difference
between the survival of subadult and adult cottontails.
In order to obtain Kaplan-Meier survival estimates (Pollock et al. 1989)
and their associated 95% confidence intervals (Cox and Oakes 1984) over the
entire period of interest, we used the dates each rabbit entered the study and the
end dates for each rabbit. The Kaplan-Meier method, or product limit estimator,
has several advantages over the binomial or Mayfield methods. First, it does not
assume a constant hazard function underlying the data. Second, it allows for the
gradual, or staggered, entry of individuals after the study has begun. Finally,
it allows right-censored individuals (i.e., those whose survival time is only
known to be greater than some value, perhaps because of a failed transmitter) to
be included in the analysis until the date of disappearance (Pollock et al. 1989).
The study encompassed 918 days, 25 October 1997 to 30 April 2000
(Sucke 2002). Three leaf-off and 2 leaf-on seasons were identified. Twentyone
individuals generated data for >1 season. The period of interest for each
leaf-on season (1998: n = 15, 1999: n = 13) was 153 days. Periods of interest
for leaf-off seasons were 187 days for 1997–1998 (n = 13 rabbits), 212 days
for 1998–1999 (n = 19), and 213 days for 1999–2000 (n = 9). In the comparison
of survival estimates of adults and subadults, the period of interest
for adults was 918 days (the duration of the study, n = 40) and for subadults
was 58 days (26 August–23 October 1998, n = 4). We divided the data into
2 separate time periods to analyze male versus female survival because the
number “at risk” for males reached 0 after 11 January 1998, which rendered
further values of S(t), the survivorship function at time t, impossible to
obtain. One female rabbit was used in both periods. Although data were
available for females for longer periods than those reported, we truncated
the female study periods to match data available for males. Period 1 was 70
days (31 October 1997–9 January 1998, n = 4 males, 4 females) and Period
2 was 691 days (11 April 1998–2 March 2000, n = 18 males, 14 females).
The estimate of survival rate for the entire period analyzed was used to
estimate the finite survival rate, Ŝ(t), for 1 day, 1 month (28 days) , and 1
year (365 days). Finite yearly survival rate was not calculated for any data
set that had an entire period of <365 days, e.g., all leaf-on/leaf-off data sets,
and Period 1 male/female data sets. The estimated finite survival rate for 1
day (“daily” survival rate) is calculated by [S(t)]1/t, where S(t) is the survival
rate at the end of a study period t days long (Krebs 1999; D.R. Diefenbach,
in litt.). Estimated survival rates for any number of days >1 day are calculated
by [Ŝ(1)]t, where Ŝ(1) is the daily survival rate, and t is the number of
days in the period of interest, e.g., 28 or 365. Ninety-five percent confidence
2010 A.C. Hartman and R.E. Barry 509
intervals for these figures were obtained from the original variance and the
delta method (Powell 2007, Seber 2002; D.R. Diefenbach, in litt.).
Survival estimates were compared by gender, age class, and season by
examining confidence intervals for overlap. The criterion for statistical significance for all analyses was α = 0.05.
Winter diet. Browsed stems were identified and counted within a 1-m radius
of winter (1 November–31 March) radiolocations for each of 15 individuals, 12
in 1998–99 and 5 in 1999–2000. Two individuals generated data for both years,
but observations in different years were considered independent because of the
time between them. Browsed stems were defined as woody stems with evidence
of having been bitten off cleanly by rabbits (Todd 1927) as opposed to having
been torn off by Odocoileus virginianus Zimmerman (White-tailed Deer). We
assumed the stems were not browsed by other lagomorphs because Snowshoe
Hares and Eastern Cottontails were uncommon. Stems such as Vaccinium spp.
or Gaultheria procumbens (Eastern Teaberry) that had >1 browsed portion
were counted as a single stem. During winter 1999–2000, some browse data
were collected immediately after locating the rabbit. For all locations in winter
1998–99 and the remaining in winter 1999–2000, browse locations were immediately
field-flagged, and data were collected by returning to the location at
a later date when more time could be spent in the field. Unidentified browsed or
available vegetation never accounted for >2.9% of browsed stems or >0.6% of
available stems for any individual. Plant species browsed at ≥1 locations were
assigned to 13 diet components (Table 1).
We analyzed browse data with the rank-preference index (Johnson 1980,
Krebs 1999) using the program PREFER 5.1 (Pankratz 1994). Analysis with
this index usually is not affected by rare items in the diet (Krebs 1999). For
each rabbit analyzed, this method subtracted the rank of each diet component
with respect to availability from its rank with respect to use to find the difference
in ranks (ti) for each diet component. Program PREFER yielded average
(for all rabbits) differences in ranks (T) for each of the 13 diet components.
Availability was derived for all species browsed by any rabbit in either winter.
To determine availability for winter 1998–99, the number of stems of each
available species that occurred within a 1-m radius of winter radiolocations
was tallied from the microhabitat study of Boyce (2001). For winter 1999–
2000, we counted available and browsed stems on the same occasion. Because
Spencer (1985) reported conifer needles in the stomachs of Appalachian Cottontails,
we noted whether Red Spruce needles were available within the 1-m
radius plot at 93 of the 167 food locations studied. The sample included locations
of 10 rabbits in winter 1998–1999 and 5 in winter 1999–2000.
Results
We captured and radio-collared 44 Appalachian Cottontails (24 males, 20
females) and 1 male Eastern Cottontail (not included in subsequent analyses)
from 25 October 1997 to 11 January 2000 over 6751 trap nights. Rabbits
were captured with less effort (12–108 trap nights/individual) during the
months of September through December (except November 1998, with 238
510 Northeastern Naturalist Vol. 17, No. 3
trap nights/individual) than at other times of the year. Mean weight of 18
adult females was 1100 g (range = 850–1350 g). Mean weight of 21 adult
males was 1000 g (range = 750–1300 g); weight for 1 male was not obtained.
Four individuals were subadults.
Survival
The median number of days individuals (n = 44) were in the study was
63 (range = 2 to 869). For adults (n = 40), the median was 80 (range = 2 to
869) days. Mean finite survival rate for all Appalachian Cottontails over the
entire 918-day study was 0.0023 (95% confidence interval = 0–0.0054). This
corresponds to a finite daily survival rate of 0.9934 (0.9922–0.9946), finite
monthly (28-day) survival rate of 0.8309 (0.8023–0.8596), and finite yearly
survival rate of 0.0894 (0.0493–0.1296). The mean finite daily survival rate
was similar for subadults (0.9764) and adults (0.9936). Estimates for the 2
groups were different at the monthly level (0.5121 for subadults and 0.8349
for adults), but variability was much greater for subadults and confidence
intervals overlapped. Survival rate dropped sharply after 60 days (t = 60) and
was <0.1 after 3.5 months (Fig. 1).
During Period 1, daily survival rate for males was 0.9747 (0.9498–
0.9996) and for females was 0.9901 (0.9763–1.0040). During Period 2, daily
survival rate for males was 0.9948 (0.9920–0.9976) and for females was
0.9964 (0.9951–0.9978). Confidence intervals of male and female survival
rates also overlapped at the monthly level in both periods, and at the yearly
level in Period 2.
Daily survival rate during leaf-off 1997–1998 was 0.9852 (0.9801–
0.9902), significantly lower than for all other seasons (means ≥0.9947). The
28-day survival rate was 0.6578 (0.5642–0.7513), also significantly lower
than for other seasons (means ≥0.8628). Confidence intervals overlapped
between all remaining seasons for both daily and 28-day levels. Predation
was the most frequently confirmed cause (n = 15) of mortality among 41
Appalachian Cottontails that died during the study.
Winter diet
Food items browsed by Appalachian Cottontails at winter radiolocations
were placed into 13 categories (Table 1). All diet components were not
preferred equally (F12,5 = 6.10, P < 0.05). Amelanchier spp. (serviceberry),
Lonicera sp. (honeysuckle), and Photinia spp. (chokeberry) were among the
most-preferred items (T < -1.0; Table 1), although honeysuckle was present in
very low numbers (only 2 stems of honeysuckle were browsed of 2 available;
Appendix I). Rubus spp. were used in proportion to availability. The mostavoided
components (T > 1.0 ) were Mountain Laurel, Rhododendron spp.,
and the “other” group. Taxa ranked “preferred” and also representing >20
browsed stems included only chokeberry, Eastern Teaberry, and blueberry/
huckleberry. When we truncated the number of diet items to the only 3 that
comprised >5% of available stems and re-analyzed with the rank-preference
index, Eastern Teaberry and blueberry/huckleberry were preferred, and
Mountain Laurel was avoided. For 7 rabbits, >50% of locations contained Red
Spruce needles. Four rabbits had >50% of locations without spruce needles.
2010 A.C. Hartman and R.E. Barry 511
Figure 1. Kaplan-Meier survival curve for S. obscurus for all individuals (n = 44). S(t) = proportion surviving past time t (number of days).
Error bars indicate 95% confidence intervals. Tracking began at day 6.
512 Northeastern Naturalist Vol. 17, No. 3
Fifty percent of locations of 1 rabbit housed spruce needles. Of the remaining
3 rabbits, 2 had spruce needles in the only location examined, and 1 had no
spruce needles in either of the 2 locations examined.
Discussion
Data were insufficient to address the speculation that subadults have
lower survival rates than adults. The expectation that female survival rates
would be higher than those of males was not supported by the data, although
notable, albeit nonsignificant, differences in means (greater for females)
suggest that biological differences between the sexes might be found with a
more powerful experimental design. The assertion that males travel farther
during the leaf-on than leaf-off season (Boyce 2001), thereby increasing
their risk of predation, actually indicates that data should be separated on
the basis of season and sex for analysis in a study with larger sample sizes.
Sample size in the current study was too small to partition in this way.
Our speculation that survival rates would be lower in leaf-off than
leaf-on seasons was supported only by the first of the 3 leaf-off seasons.
The delineation of seasons that was used was based on previous
studies of Appalachian Cottontails in western Maryland (Sommer
1997) and at Dolly Sods (Boyce 2001), which were focused primarily
on patterns of habitat use. Perhaps future survival studies should
define leaf-off, or winter, as a short period during which weather is harsh
(Keith and Bloomer 1993), or delineate seasons on the basis of breeding
onset (Bond et al. 2001), in an effort to reveal other patterns. That
survival rates in leaf-on seasons were not higher might be attributable
to home ranges being larger during the leaf-on than leaf-off season at
Table 1. Winter diet components of Sylvilagus obscurus (Appalachian Cottontail) and associated
T values generated by Johnson’s (1980) rank-preference index. Largest average differences
in browsed and available ranks (T) denote the most preferred (negative) or avoided (positive)
diet components. See Appendix I for names of species and common names of taxa contained
within each multiple-species diet component.
T values
Winter diet component All available taxa 3 most available taxa
Amelanchier spp. -1.647
Lonicera sp. -1.588
Photinia spp. -1.235
Acer rubrum L. (Red Maple) -0.822
Hamamelis virginiana L. (American Witchhazel) -0.412
Gaultheria procumbens L. (Eastern Teaberry) -0.294 -0.412
Vaccinium spp. and Gaylussacia baccata (Wangenh.) -0.059 -0.059
K. Koch (Black Huckleberry)
Rubus spp. 0.000
Viburnum spp. 0.029
Ilex montana Torr. & Gray ex Gray (Mountain Holly) 0.941
and Nemopanthus mucronatus (L.) Loes (Mountain Holly)
Rhododendron spp. 1.059
Kalmia latifolia L. (Mountain Laurel) 1.941 0.471
Other (7 species) 2.147
2010 A.C. Hartman and R.E. Barry 513
Dolly Sods (Boyce and Barry 2007). Although leaves increase concealment
cover, rabbits with large home ranges cover greater distances,
potentially encountering marginal habitat and higher risk of predation.
Converting daily, 60-day, and 70-day estimates reported by others to
monthly estimates allowed us to compare our results, obtained at the type
locality of Appalachian Cottontails where rabbits were widely distributed
and active over a large area (Sucke 2002), with those from other cottontail
populations. The overall monthly estimate (0.8309) for Dolly Sods Appalachian
Cottontails was within the ranges reported for 2 populations of
Eastern Cottontails (Bond et al. 2001, Keith and Bloomer 1993) and was
only slightly lower than survival estimates reported for another population
of Eastern Cottontails (Trent and Rongstad 1974) and a population of New
England Cottontails in large (source) patches (Barbour and Litvaitis 1993).
It was higher than the survival of New England Cottontails in smaller (sink)
patches. Daily survival rates were comparable to those obtained for Eastern
Cottontails at Cape Cod, MA by Boland and Litvaitis (2008).
Serviceberry and chokeberry, preferred food items in the diet of Appalachian
Cottontails, are known foods for cottontails (Van Dersal 1938).
Red Spruce was browsed only at 1 location, and presence/absence of spruce
needles at daytime resting sites (food locations) did not reveal a pattern.
Even if needles were abundant at locations, they may have indicated the
importance of spruce for cover rather than food. Hemlock was not browsed,
rabbits avoided Mountain Laurel, and data were insufficient to analyze Great
Rhododendron separately. Vaccinium spp. was abundant at resting locations
and was browsed frequently in combination with Gaylussacia baccata
(Black Huckleberry). However, this group was used nearly in proportion to
its availability, suggesting that these browse items collectively might constitute
a staple component of the Appalachian Cottontail diet. The expectation
that Rubus spp. would be a preferred item of the winter diet of Appalachian
Cottontails also was not supported. Buds and small shoots of maples were
scarce, and only 1 individual browsed on plants in this taxon (A. rubrum,
specifically). Future investigators should consider radiolocating rabbits
during dusk and dawn hours when they are thought to be most active and
feeding (Dalke and Sime 1941).
Management implications
This study provides wildlife managers with initial survival estimates
for a population of Appalachian Cottontails at the type locality in habitat
that provides plentiful concealment and thermal ericaceous cover (Boyce
2001). In addition, it identifies Eastern Teaberry, blueberries, and Black
Huckleberry, and possibly chokeberry, as woody species important to this
rabbit for winter browse. Succession is retarded at Dolly Sods because of
poor soils and a harsh climate (Sucke 2002), so ample habitat that provides
suitable cover and food resources for Appalachian Cottontails is likely to
persist there. Nevertheless, management to preserve critical habitat at the
site by such means as controlled burns has been suggested (Venable 1996),
and management that accommodates factors that contribute to survival, such
514 Northeastern Naturalist Vol. 17, No. 3
as ground cover and preferred winter foods, may be critical to the persistence
of Appalachian Cottontails in smaller, isolated patches.
Acknowledgments
We thank the West Virginia Division of Natural Resources and Frostburg State
University for the funding of this project and the USDA Forest Service for access to the
study site, use of a cabin, and use of GIS files. Mitch Spear donated his snowmobile for
the duration of the study. Drs. J. Edward Gates and Ray Morgan of the Center for Estuarine
and Environmental Studies Appalachian Laboratory (University of Maryland)
loaned us GPS and electrophoretic equipment and provided the use of their labs. Drs.
Gwen Brewer, L. Michelle Bowe, Durland Shumway, Lance Revennaugh, and Chrismarie
Baxter (deceased), all of Frostburg State University, also provided support and
expertise. Kelly Boyce contributed significant help in the field and with the literature
search, home-range, and microhabitat data. The following also contributed to field
work: Nancy Bensley, Randy Mowrer, Bob Cordes, Frank Butera, Kieran O’Malley,
Jeff Peters, Bianca McIntyre, Paul Wenninger, Sandy Davis, Julie York, Sean Kline,
Angela Donroe, Steve Doll, Terry McCullough, Mark Lewis, and Jan Ferrigan.
Literature Cited
Barbour, M.S., and J.A. Litvaitis. 1993. Niche dimensions of New England Cottontails
in relation to habitat patch size. Oecologia 95:321–327.
Barry, R.E., N. Bensley, and K.A. Boyce. 1997. Range, density, habitat breadth, and
daily survival of the Appalachian Cottontail (Sylvilagus obscurus) in western
Maryland. Final Report to the Maryland Heritage and Biodiversity Conservation
Programs, Annapolis, MD.
Boland, K.M., and J.A. Litvaitis. 2008. Role of predation and hunting on Eastern
Cottontail mortality at Cape Cod National Seashore, Massachusetts. Canadian
Journal of Zoology 86:918–927.
Bond, B.T., L. W. Burger, B.D. Leopold, and K.D. Godwin. 2001. Survival of Cottontail
Rabbits (Sylvilagus floridanus) in Mississippi and an examination of
latitudinal variation. American Midland Naturalist 145:127–136.
Boyce, K.A. 2001. Distribution, seasonal home range, movements, and habitat of the
Appalachian Cottontail, Sylvilagus obscurus, at Dolly Sods, West Virginia. M.Sc.
Thesis. Frostburg State University, Frostburg, MD.
Boyce, K.A., and R.E. Barry. 2007. Seasonal home range and diurnal movements
of Sylvilagus obscurus (Appalachian Cottontail) at Dolly Sods, West Virginia.
Northeastern Naturalist 14:99–110.
Chapman, J.A., and R.P. Morgan. 1973. Systematic status of the cottontail complex
in western Maryland and nearby West Virginia. Wildlife Monographs 36:1–54.
Chapman, J.A., and J.R. Stauffer. 1981. The status and distribution of the New England
Cottontail. Pp. 973–983, In K. Meyers and C.D. McInnes (Eds.). Proceedings of
the World Lagomorph Conference. University of Guelph, Guelph, ON, Canada.
Chapman, J.A., A.L. Harman, and D.E. Samuel. 1977. Reproductive and physiological
cycles in the cottontail complex in western Maryland and nearby West
Virginia. Wildlife Monographs 56:1–73.
Chapman, J.A., J.G. Hockman, and W.R. Edwards. 1982. Cottontails. Pp. 83–123,
In J.A. Chapman and G.A. Feldhamer (Eds.). Wild Mammals of North America.
Johns Hopkins University Press, Baltimore, MD.
Chapman, J.A., K.L. Cramer, N.J. Dippenaar, and T.J. Robinson. 1992. Systematics
and biogeography of the New England Cottontail Sylvilagus transitionalis
(Bangs, 1895), with the description of a new species from the Appalachian Mountains.
Proceedings of the Biological Society of Washington 105:841–866.
2010 A.C. Hartman and R.E. Barry 515
Cox, D.R., and D. Oakes. 1984. Analysis of Survival Data. Chapman and Hall,
London, UK.
Dalke, P.D., and P.R. Sime. 1941. Food habits of Eastern and New England Cottontails.
Journal of Wildlife Management 5:216:228.
Johnson, D.H. 1980. The comparison of usage and availability measurements for
evaluating resource preference. Ecology 61:65–71.
Keith, L.B., and S.E.M. Bloomer. 1993. Differential mortality of sympatric Snowshoe
Hares and Cottontail Rabbits in central Wisconsin. Canadian Journal of
Zoology 71:1694–1697.
Krebs, C. 1999. Ecological Methodology. Second Edition. Addison-Wesley Educational
Publishers, Inc., Menlo Park, CA.
Litvaitis, J.A., D.L. Verbyla, and M.K. Litvaitis. 1991. A field method to differentiate
New England and Eastern Cottontails. Transactions of the Northeast Wildlife
Conference 48:11–14.
Morgan, R.P., and J.A. Chapman. 1979. Serum proteins of the Sylvilagus complex.
1979. Pp. 64–72, In K. Meyers and C.D. McInnes (Eds.). Proceedings of the
World Lagomorph Conference. University of Guelph, Guelph, ON, Canada.
Pankratz, C. 1994. Program PREFER: Preference assessment. V. 5.1. Northern Prairie
Science Center, Jamestown, ND.
Pollock, K.H., S.R. Winterstein, C.M. Bunck, and P.D. Curtis. 1989. Survival analysis
in telemetry studies: The staggered entry design. Journal of Wildlife Management
53:7–15.
Powell, L. A. 2007. Approximating variance of demographic parameters using the
delta method: A reference for avian biologists. Condor 109:950–955.
Ruedas, L.A., R.C. Dowler, and E. Aita. 1989. Chromosomal variation in the New
England Cottontail, Sylvilagus transitionalis. Journal of Mammalogy 70:860–864.
Seber, G.A.F. 2002. The Estimation of Animal Abundance and Related Parameters.
Second Edition. The Blackburn Press, Caldwell, NJ.
Sommer, M.A. 1997. Distribution, habitat, and home range of the New England Cottontail
(Sylvilagus transitionalis) in western Maryland. M.Sc. Thesis. Frostburg
State University, Frostburg, MD.
Spencer, R.K. 1985. Comparison of the feeding habits of two sympatric cottontails
(Sylvilagus transitionalis and S. floridanus) in western Maryland and nearby
West Virginia. M.Sc. Thesis. Frostburg State College, Frostburg, MD.
Sucke, A.C. 2002. Survival, winter diet, density, and macrohabitat of the Appalachian
Cottontail, Sylvilagus obscurus, in West Virginia. M.Sc. Thesis. Frostburg
State University, Frostburg, MD.
Todd, J.B. 1927. Winter food of cottontail rabbits. Journal of Mammalogy 8:222–228.
Trent, T.T., and O.S. Rongstad. 1974. Home range and survival of Cottontail Rabbits
in southwestern Wisconsin. Journal of Wildlife Management 38:459–472.
Van Dersal, W.R. 1938. Native Woody Plants of the US, Their Erosion Control and
Wildlife Values. US Government Printing Office, Washington, DC.
Venable, N.J. 1996. Dolly Sods. West Virginia University Extension Service, Morgantown,
WV.
Villafuerte, R., J.A. Litvaitis, and D.F. Smith. 1997. Physiological responses by lagomorphs
to resource limitations imposed by habitat fragmentation: Implications to
condition-sensitive predation. Canadian Journal of Zoology 75:148–151.
West Virginia Natural Heritage Program. 2007. Rare, threatened, and endangered
species in West Virginia. West Virginia Division of Natural Resources, Elkins,
WV. Available online at http://www.wvdnr.gov/Wildlife/RareSpecList.shtm. Accessed
14 October 2008.
516 Northeastern Naturalist Vol. 17, No. 3
Appendix I. Number of stems of each food component browsed by (and available to) S. obscurus at Dolly Sods (F = female, M = male). * = individuals from winter
1999–2000; other individuals from winter 1998–1999. Amel. = Amelanchier; I. m./N m. = Ilex montana and Nemopanthus mucronatus; Rhod. = Rhododendron. The
group Vaccinium spp. (blueberry and huckleberry) included V. angustifolium Ait., V. myrtilloides Michx., V. pallidum Ait., and Gaylussacia baccata (Wangenh.) K.
Koch. Rubus spp. (blackberry and raspberry) was represented mainly by R. hispidus L. (Bristly Dewberry), which usually occurred at Dolly Sods as small (< 3-mm
dia.), somewhat woody shoots close to the ground. Other species in this category in small numbers were R. allegheniensis Porter and R. occidentalis L. Photinia spp.
(chokeberry) included mostly P. pyrifolia (Lam.) Robertson & Phipps, but also some P. melanocarpa (Michx.) Robertson & Phipps. Amelanchier spp. (serviceberry)
included A. arborea (Michx. f.) Fern. and A. laevis Weig. Ilex montana Torr. & Gray ex Gray and Nemopanthus mucronatus (L.) Loes (both Mountain Holly) comprised
a category. Most of the Viburnum spp. counted were V. nudum L.var. cassinoides (L.) Torr. & Gray, although some may have been V. lentago L. Rhododendron
spp. (azaleas) included R. viscosum (L.) Torr. and R. prinophyllum (Small) Millais. The “other” diet component consisted of 7 plant species—Quercus ilicifolia Wangenh.
(Bear Oak), Quercus alba L. (White Oak), R. maximum, Pinus resinosa Soland.(Red Pine), Picea rubens Sarg. (Red Spruce), Alnus incana ssp. rugosa (Du Roi)
Clausen (Speckled Alder), and Hydrangea arborescens L. (Wild Hydrangea)— that each were browsed at only 1 location.
Browse item
Vaccinium Gaultheria Rubus Photinia Kalmia Amel. I. m./ Hamamelis Viburnum Rhod. Acer Lonicera Other
Rabbit ID spp. procumbens spp. spp. latifolia spp. N. m. virginiana spp. spp. rubrum sp. (7 species)
F4 161(622) 11(90) 35(95) 1(38) 7(230) 6(6) 0(19) 0(0) 2(2) 1(18) 2(2) 0(0) 2(8)
F10 139(989) 21(250) 5(28) 9(44) 1(218) 2(2) 1(16) 0(0) 0(7) 0(0) 0(0) 1(1) 0(26)
F12 137(1041) 14(70) 0(0) 0(0) 14(217) 0(0) 0(13) 0(1) 1(12) 0(0) 0(0) 0(0) 2(177)
F15 169(1404) 21(491) 3(128) 20(27) 3(111) 0(0) 0(0) 0(1) 0(0) 0(0) 0(0) 0(0) 0(8)
F14 61(421) 1(41) 0(0) 5(17) 0(98) 0(0) 0(0) 0(0) 0(0) 0(10) 0(0) 0(0) 0(10)
M7 65(1158) 8(22) 6(6) 10(17) 8(140) 0(0) 0(6) 2(2) 0(8) 0(16) 0(1) 1(1) 0(9)
M10 197(613) 14(49) 0(12) 1(16) 4(158) 0(0) 0(40) 0(0) 0(0) 0(11) 0(0) 0(0) 0(2)
M13 180(711) 12(70) 5(53) 5(10) 15(168) 0(0) 4(14) 0(0) 0(6) 1(31) 0(0) 0(0) 2(11)
M15 129(671) 4(33) 16(47) 5(31) 3(282) 0(0) 1(38) 0(0) 3(3) 0(0) 0(1) 0(0) 12(128)
M16 95(425) 11(55) 0(57) 2(4) 3(206) 0(0) 0(0) 0(2) 0(0) 0(0) 0(0) 0(0) 0(17)
M17 151(878) 14(94) 2(2) 25(36) 12(238) 1(1) 0(1) 0(2) 0(2) 0(13) 0(1) 0(0) 0(4)
M18 16(914) 1(215) 0(1) 7(15) 0(32) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(5)
F4* 182(1019) 16(275) 27(80) 17(39) 14(242) 0(0) 3(9) 0(0) 0(13) 0(29) 0(3) 0(0) 0(6)
F17* 294(1648) 16(274) 2(10) 8(50) 11(233) 0(0) 0(0) 2(15) 0(0) 0(0) 1(3) 0(0) 0(14)
F20* 29(81) 1(3) 0(5) 0(0) 12(107) 2(7) 0(5) 3(6) 0(0) 0(3) 0(0) 0(0) 3(4)
M15* 167(1395) 7(216) 3(27) 4(33) 5(101) 0(0) 0(1) 0(0) 1(5) 0(0) 0(2) 0(0) 0(10)
M24* 89(403) 17(109) 17(56) 1(4) 7(90) 0(0) 0(0) 0(0) 0(4) 2(15) 0(0) 0(0) 0(2)
Totals 2261(14,393) 334(2357) 121(707) 120(381) 119(2871) 11(16) 9(162) 7(29) 7(62) 4(146) 3(13) 2(2) 21(441).