The Ecology and Natural History of the Striped Skunk (Mephitis Mephitis) in the Cades Cove Campground, Great Smoky Mountains National Park, Tennessee

University of Tennessee, Knoxville TRACE: Tennessee Research and Creative Exchange

Masters Theses Graduate School

8-1981

The Ecology and Natural History of the Striped Skunk (Mephitis mephitis) in the Cades Cove Campground, Great Smoky Mountains National Park, Tennessee

Deborah M. Goldsmith University of Tennessee - Knoxville

Follow this and additional works at: https://trace.tennessee.edu/utk_gradthes

Part of the Earth Sciences Commons

Recommended Citation Goldsmith, Deborah M., "The Ecology and Natural History of the Striped Skunk (Mephitis mephitis) in the Cades Cove Campground, Great Smoky Mountains National Park, Tennessee. " Master's Thesis, University of Tennessee, 1981. https://trace.tennessee.edu/utk_gradthes/3430

This Thesis is brought to you for free and open access by the Graduate School at TRACE: Tennessee Research and Creative Exchange. It has been accepted for inclusion in Masters Theses by an authorized administrator of TRACE: Tennessee Research and Creative Exchange. For more information, please contact [email protected]. To the Graduate Council:

I am submitting herewith a thesis written by Deborah M. Goldsmith entitled "The Ecology and Natural History of the Striped Skunk (Mephitis mephitis) in the Cades Cove Campground, Great Smoky Mountains National Park, Tennessee." I have examined the final electronic copy of this thesis for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Master of Science, with a major in Ecology and Evolutionary Biology.

Michael R. Pelton, Major Professor

We have read this thesis and recommend its acceptance:

Arthur C. Echternacht, Boyd L. Dearden

Accepted for the Council: Carolyn R. Hodges

Vice Provost and Dean of the Graduate School

(Original signatures are on file with official studentecor r ds.) To the Graduate Counci l:

I am submitting herewi th a thesi s written by Deborah M. Goldsmi th entitled 11The Ecology and Natural History of the Striped Skunk (Mephiti s mephiti s) in the Cades Cove Campground, Great Smoky Mountains Nat1onal Park, Tennessee ... I have exami ned the final copy of thi s thesi s for form and content and recommend that it be accepted in parti al fulfi ll­ ment of the requirements for the degree of Master of Sci ence, wi th a major in Ecology . � ��L-. .�o& �c. ., i�L. Pelton, MaJo�ofessor

We have read thi s thesis recommend its acceptance :

Accepted for the Counci l:

�.J!JI V1ce Chancelfor Graduate Studies and Research THE ECOLOGY AND NATURAL HISTORY OF STRIPED SKUNKS • IN THE CADES COVE CAMPGROUND, GREAT SMOKY

MOUNTAINS NATIONAL PARK, TENNESSEE

A Thesis

Presented for the

Master of Science

Degree

The University of Tennessee, Knoxville

Deborah M. Goldsmith

August 1981

3053783 ' ACKNOWLEDGMENTS

This study could not have been compl eted wi thout the he l p, support, and advice of many individual s. I would like to express my appreciati on to Dr. Mi chael R. Pel ton, Professor , Department of

Forestry, Wildl ife and Fi sheries, for his guidance and support on this project. I would also like to thank the members of my commi ttee,

Dr . Sandy {A. C. ) Echternacht, Associate Professor, Department of

Zool ogy, and Dr. Boyd Dearden , Associate Professor , Department of

Forestry , Wi ldl ife and Fi sheries.

Grati tude is expressed to fel low graduate students and friends for helping in conducti ng fieldwork, as wel l as to the Park personnel.

I would like to extend special thanks to Dr. James T. Tanner for hel ping wi th the population parameters ; to Don Wel ler for support , friendship and time throughout the project; to Carol Hom for long · hours in the field and for recruiting people to help in the field; and to all the others who assisted duri ng the study.

I would al so like to acknowl edge the people wi th whom I shared the field station for their tol erance each time I was sprayed.

i i ABSTRACT

Thi s study was initiated to col l ect information on several aspects

of skunk ecol ogy and natural history in a campground situation. The

areas of skunk ecol ogy and behavior considered were population density,

denning behavior, den site avai lability, acti vity behavior, and

foraging behavior in the Cades Cove campground, GSMNP. Data col - lecti on for the study began in June 1979 and ended in August 1980.

The skunk population within the 156 ha study area , which

. incl uded the campground and picnic area , was estimated by mark­

recapture at 31 ± 4 skunks ; in late July and earl y August the skunk

population temporari ly increased as the juveniles became independent

and were caught more frequently. The estimated res ident population

in the study area was higher than reported population estimates from

agricultural and natural areas. The campground and picnic area appear

to influence the size of the skunk popul ation by increa sing the availabi l ity of food and den sites.

Thirty-three natural dens were located duri ng the course of

this study. These dens fel l into 2 groups corresponding to the

seasons , summer and fall. Summer dens tended to be above ground

be located on wel l -drained shal low soils with steep slopes . Fal l dens

tended to be below ground and be located on moderately to wel l drained

soi ls with gentle slopes . Summer dens al so had wider entrances than

fal l dens. Skunks switched from summer dens to fall dens by October

in response to several factors : temperature , decreased food

iii iv availability, physiological changes and possibly light in�ensity

(Alesiuk and Stewart, 1962) .

The drainage culverts in the campground were used as den sites

by skunks from July through October. Of the 31 culverts in the camp­ ground, 20 were used as dens in 1979 and 25 were used as dens in 1980.

Of the 11 culverts not used as dens in 1979, 9 were filled with water

or debris and 2 were occupied by raccoons, while the 6 culverts not

used as dens in 1980 were filled with debris.

Culvert dens and natural dens were not used exclusively by 1

.skunk, but were used sequentially by several skunks. Five of the

natural dens and all of the culvert dens were used sequentially by

skunks during the study .

The daily activity patterns of striped skunks using the camp­ ground· changed with season. In April and May, few skunks were observed in the campground . The number of skunks in the campground during the night increased in July and August in response to the increased avail­ ability of food and to the increased number of juveniles mature enough

to accompany the females during foraging. During July and August, activity was bimodal with peaks at 2300 hrs. and 0130 hrs. The skunks·

observed foraging before 2400 hrs . tended to tolerate people, while

skunks observed foraging at 0130 hrs. tended to be cautious, avoiding people . By October, the number of skunks observed in the campground

at night had decreased with activity being distributed throughout the

night.

Several factors influencing seasonal activity,were measured:

cloud cover, temperature, precipitation , date , and time. Of these

variables temperature accounted for the most variance in activity v levels of skunks in different months. Other variables that were not measured, such as light intensity and the number of campers in the campground, may explain more of the variance . TABLE OF CONTENTS

CHAPTER PAGE

I. INTRODUCTION ...... 1

II . DESCRIPTION OF THE STUDY AREA 4

Location • ..•• •• ...... 4 Topography and Soils •• . . . 6

Climate •••.•. . . . 6 Vegetation • • • • • • • • . 7 Fauna • • • • • . • 9

III. MATERIALS AND METHODS 10

Capture and Handling Techniques .•.• 10 Population Dynamics ••.•••.•.•••. 13 Seasonal and Daily Activity Patterns 13 Home Range • • • • • • . . 16

Denning • . •.••• 17

IV . RESULTS AND DISCUSSION • 20

Trap Data o • • • o • . . . 20

External Morphology • o • 27

Pelage •••.•o ••o •• 36

Population Dynamics o • o ••••• • • • • 39 Seasonal and Daily Activity Patterns . 42 Human-Skunk·Interactions •••••. 46

Home Range and Territoriality • 47

Denning ••..•.•.•.••• . . . 54

• V 0 RESEARCH-MANAGH1ENT IMPLICATIONS 67

VI. SUMMARY AND CONCLUSION .. 69

LITERATURE CITED ...... 73

VITA • • • . • . • • 77 •

vi LIST OF TABLES

TABLE PAGE

1. Cl imatic Data for Cades Cove , GSMNP, for 1979 and 1980 . • • 8

2. Number of Striped Skunks Captured by Month in the Cades

Cove Campground Area , GSMNP, 1979 • . • • • . • . • • • 21

3. Number of Striped Skunks Captured by Month in the Cades

Cove Campground Area , GSMNP, 1980 • • • • • • • • . 22 4. Trapping Success for Stri ped Skunks Captured in the

Cades Cove Study Area , GSMNP, 1979 • . . . • ...... 23

5. Trapping Success for Stri ped Skunks Captured in the Cades Cove Campground Study Area , GSMNP, 1980. . . . 24

6. Sex Ratios of Striped Skunks by Year Expressed as Ma les pe r Females in the Cades Cove Campground Study Area ,

GSMNP, 1979-1980 • . • • • ...... • . • . • . • . . 25

7. External Measurements (cm) of Adult Ma le and Female Striped Skunks Captured in the Cades Cove Campground Study Area , GSMNP , 1979-1980 .•••••••. 28

8. External Measurements of Striped Skunks from 2 Different

Populations Reported in the Literature • . • • • . • • 29

9. External Measurements(cm) of Juvenile Ma le and Female Striped Skunks Captured in the Cades Cove Campground

Study Area , GSMNP, 1979-1 980 . . • • • • • . • . • . • . • . 32

10. Average Monthly We ights( kg ) of Adult and Juvenile Striped Skunks Captured in the Cades Cove Campground Study Area , 1979-1 980 •.•.•••••.•••.. 34

11. Comparison of Body We ight Losses During the Wi nter in 4 Different Striped Skunk Populations • • • • • . • . 37

12. Numbers of Striped Skunks in Each of the 4 Color Patterns for Cades Cove Campground Study Area, GSMNP ,

1979-1 980 • • • • • • • . • . • • • . • . • • • 38

13. Reported Densities of Stri ped Skunks in Various

Habitats in the U. S. • . • • • . • • • . • • • • • . • . • 41

vii vi ii

TABLE PAGE

14. A Summary of Movement Parameters of 11 Stri ped Skunks in the Cades Cove Campground Study Area, GSMNP,

1979-1 980 . • • • . • • • • • . • . • • • . • . • . • 49

15. Classification of Summer and Fall Dens Used by Skunks in the Cades Cove Campground Study Area, GSMNP ,

1979-1 980 . • • • • • • • • • • • • • . • . • . • . . 55

16. Average Measurements of 33 Natural Skunk Dens in the

Cades Cove Campground Study Area, GSMNP • • . • • • . 57 LIST OF FIGURES

FIGURE PAGE

1. Study Area and Grid System for Cades Cove Campground Areas , GSMNP • • • • . • • • • • . • . • . • . • • • 5

2. A Classification of Color Patterns of Striped Skunks • 12

3. Map of the Roads and Campsites in the Cades Cove

Campground , GSMNP ••.•.•••.•••.•.•• 15

4. The Range of 4 Subspecies of Striped Skunk (Mephitis

mephitis} in the Eastern U.S. ( Hami lton , 1943) ••• 30

5. Atypical col or pattern of 4 stri ped skunks in Cades Cove

Campground Study Area , GS MNP . . . • • • • ...... 40

6. Mean Number of Skunks Observed During the Stri p Counts

in the Cades Cove Campground, GSMNP, 1980 . • . • . . 43

7. Home Ranges of 3 Mal e Striped Skunks in the Cades Cove Campground Study Area , GSMNP, 1979-1 980 • . • . • . 51

B. Home Ranges of 3 Female Striped Skunks in the Cades

Cove Campground Study Area , GSMNP, 1979-1 980 • . • • 52

9. Home Ranges of 3 Juveniles in the Cades Cove Campground

Study Area, GSMNP , 1979-1980 • . • . • . • • • . • . 53

10. A Plot of Factor 2, A General Drainage Factor Vs. Factor 1, A General Elevation Factor wi th Reference to Season ...... 58

11. A Plot of Factor 3, Representing Width of Entrance Vs. Factor 1, A General Elevati on Factor with Reference to Season . • • • • • • • . • • • . • . 59

12. Canonical Correlation Histogram of Fal l and Summer Skunk Dens in the Cades Cove Campground Study Area,

GSMNP 0 ...... 0 . . . . . 0 . . . 0 . 0 . . 61

13. The Number of Road Culverts Used as Dens in 1979 in Cades Cove Campground , GSMNP . 0 . . 0 0 0 . 0 64

14. The Number of Road Culverts Used as Dens in 1980 in Cades Cove Campqround, GSMNP . . 0 . . 0 . . . 65

ix CHAPTER I

INTRODUCTION

Striped skunks {Mephitis mephitis) are distributed throughout

the United States , southern Canada , and northern Me xico (Hall, 1981)

in a variety of habitats ; no single wel l-defined land type has been

clearly establ ished as their preferred habitat {Bennitt and Nagel,

1937) . Schwartz and Schwartz {1 959) bel ieved that skunks preferred

areas that included a mixture of woods , brush, and open woods broken

'by wooded ravines and rock outcrops whi le Mohr {1943) described

preferred skunk habitat as an area of rock outcrops and wooded ravines

close to water. Dean {1 965) found that skunks reached their highest

densities in areas where there was a mixture of pastures, orchards,

woods , and a source of trash or a garbage dump and concluded that such

areas represented the optimum skunk habitat.

The ability of skunks to exploit man-made structures and garbage

reflects the ir opportuni stic feeding and denning habits . Previous

studies have shown that striped skunks use a variety of structures as

den sites {Verts, 1967; Houseknecht, 1969; Houseknecht and Tester,

1978) . All en {1939) found that skunks usually denned in a burrow dug

by themsel ves or by another animal , such as woodchucks {Marmota monax),

badgers {Taxidea taxus), red foxes {Vul pues vulpes) or gray foxes

{Urocyo n cinereoargenteus) ; but sometimes used stump piles and

bui ldings . · Sto�m {1972) and Houseknecht {1 969) showed that skunks

switched den types in response to changes in season using bel ow ground

1 2

dens such as burrows, root cavities, logs, junk piles, and buildings

during the winter and above ground dens such as leaf piles, stumps,

hollow logs, hay fields,·and brush piles during the summer.

The majority of the studies done on foraging and feeding

behavior have been in natural or agricultural areas where the diet of

skunks, in order of descending importance, consisted of insects,

fruits, small mammals, grains, birds, reptiles, and amphibians (Kelker,

1937 ; Hamilton, 1936). Dean (1965) showed that skunks supplement

their diet with garbage and discarded food, particularly in the winter

. when other food items are scarce. Hamilton (1936) found that skunks

moved into rural towns in the fall as the availability of food in the

farm areas decreased and used the garbage in the towns as a food

source.

· Skunks can also be a problem in recreational areas, such as

campground and picnic areas where high densities of both people and

skunks increase the level of contact between the 2 groups and the

probability of� person being bitten or sprayed . The bite of the

skunk is a potential source of local infection, tetanus, rabies, and

leptospirosis (Verts, 1967), while the spray is an inconvenience and

may cause some individuals to become ill (Curtis, 1978). The majority

of skunk-people interactions occur between May and August when levels

of human outdoor activity and skunk activity are high (Hattwick,

et al. 1972).

The skunk population in Cades Cove campground, Great Smoky

Mountains National Park, has increased in recent years, becoming a

nuisance and a potential health hazard (Kent Higgins, Personal Commu­

nications). Unfortunately, little information is available on the 3

behavior and ecology of campground skunks. Earlier studies concen­

trated on agricultural areas (Verts, 1967; Storm, 1972), natural areas

(Allen, 1939; Houseknecht, 1969; Bailey, 1971; Sunquist, 1974) or on

captive populations (Alesiuk and Stewart, 1962). These studies provide

background information on ecology and behavior of striped skunks but

are not adequate for developing an efficient management program for

campground skunks. Since skunks are opportunistic animals, their

behavior probably changes in response to conditions found in campgrounds

and picnic areas. This study was undertaken to collect information on

· several aspects of skunk ecology and natural history in a campground

situation and compare the results with information from other studies

done in agricultural or natural areas. The areas of skunk ecology con­

sidered were: population density, denning behavior, den site selection,

foragfng behavior and activity behavior . CHAPTER II

DESCRIPTION OF THE STUDY AREA

Location

Cades Cove is located in Blount County, Tennessee in the north­

western section of the Great Smoky Mountain National Park (GSMNP). It

is bounded on the east by Bote Mountain, on the north by a series of

mountains which separate it from Tuckaleechee Cove, and on the west by

Hannah Mountains (Figure 1). Due to -the flatness and fertility of the

·land , this area contained a prosperous mountain community during the

1880's. Today Cades Code is an historical site under the jurisdiction

of the National Park Service and receives over a million visitors a

year. To accommodate these visitors a campground and picnic area were

· ouilt at the eastern end of Cades Cove.

The present study was confined to a 156 .4 ha area encompassing

the campground and picnic area. The campground and picnic area

covered 42.2 ha and pasture and second growth mixed pine and cove

hardwood forest covered the remaining 114.2 ha. The campground con­

tains 171 campsites among 3 sections: B section with 85 sites, C section with 76 sites and a group section with 4 sites for 10 people

or more. B and C sections are open to tent and recreational vehicle

camping. Camping in the group section is restricted to organized

youth groups of 7 or more people. The picnic area contains 92 tables

and grill�.

4 5

Figure 1. Study area and grid system for Cad es Cove campgrou nd areas, . GSMNP. 6 Topography and Soi ls

The topography of the study area consi sts of rol ling hills which

grades into mountains. The elevation ranges from 570 m in the camp­

ground to 672 m along the edge of the study area. The study area is

drained by Abrams Creek, Anthony Creek, Cooper Branch, Stillhouse

Branch, and their tri butaries. Duri ng July and August it is not

uncommon for parts of these streams to become dry, particularly

Still house Branch.

Soi ls can be placed in either the Al l en-Hayter Association or

the Ramsey Association. The Al len-Hayter Association, which covers

62% of the area, consists of al l uvium and col luvium chiefly from

quartzi te and slate washed in from the Ramsey Associati on. The Al len­

Hayter soi ls are moderately acidic to strongly acidic. It is in

general well drai ned (Elder, 1956) .

The Ramsey Association covers 38% of the study area . It is

characterized by steep slopes wi th thin soils and numerous slate out­

croppings. The parent material is a mixture of sandstone, quartzite,

and slate. The-soil is strongly acidic and excessively drained

(Elder, 1956) .

Cl imate

Cl imatological data for Cades Cove are limited ; in order to

characterize the genera l cl imate of the Cove the cl imatological data

for the whole Park must be considered . The Great Smoky Mountains has

a humid-temperate climate with moderate winters and hot humid summers

( Elder, 1956) . Precipitation and temperature in the Park v·ary greatly

,with elevati on. Annual precipitati on in Cades Cove averages 140 em , 7

while at the higher elevations precipitation may average over 220 em.

This precipitation is distributed throughout the year, but July is

typically the wettest month (Table 1) .

Data on the daily temperature in the Cove are based on the

temperature in Gatlinburg for 1979 and on the temperature in the Cove

for 1980 (Table 1). The weather during the 2 years of this study

differed from the above general description. In 1979, the summer was

usually wet and cool with 12.3 em of rain falling in July, 6.7 em

above the July average for this area (TVA weather data 1980). In

.1980, the summer and fall were extremely dry and hot. Only 3. 1 em of

rain fell in July and the temperature repeatedly went above 37.7°C.

The rainfall was about 3.0 em below normal for July in 1980.

Vegetation

Vegetation within the study area can be divided into 3 classes:

pasture, vegetation within the campground , and second growth mixed

deciduous and pine stands. Pasture covered approximately 6% of the

study area and consisted of a variety of grasses: fescue (Festuca sp. ),

·orchard-grass (Dactylis sp.), timothy (Phleum sp. ), and red and ladino

clover (Tri folium sp.) (Taylor, 1979). The vegetati on in the campground

covers 27% of the study area . This vegetation consists of a herb

layer of willowgrass (Microstegium sp .), and poison ivy (Rhus radicans),

a shrub layer of huckleberry (Gaylussacia sp. ), dogwood (Cornus florida),

and hemlock (Tsuga canadensis); an open canopy layer dominated by pines

(Pinus sp. ), tulip poplar (Liriodendron tulipifera), red maple (Acer

rubrum) , sweet gum (Liguidambar styraciflua), and a variety of oaks

.(Quercus sp.). Second growth mixed deciduous forest covers approximately TABLE 1. Cl imatic Data for Cades Cove , GSMNP, for 1979 and 1980.

Month Year

June 1979 6.4 7.2 5.0 ------12. 7-25

July 1979 8.5 12.3 ------15.4-27.9

August 1979 4.7 6.6 3.0 18.7-25.8 16. 3-28.7

September 1979 6.7 3.0 3.5 ------13.9-24.6

October 1979 2.2 2. 5 ------6.4-19.2

Apri l 1980 4.9 4.1 5.6 6. 2-20. 1 6.3-20.9

May 1980 2.9 2.4 2.2 11.4-24.2 9.7-24.3

June 1980 2.6 2.8 1.8 15.0-27.2 13.7-28.5

July 1980 4.3 2.6 3.9 18.9-32.2 17.6-31 .4

August 1980 4.7 1.7 1.6 16.9-29.8 17.0-31 .0

*Precipitation data provided courtesy of Tennessee Val ley Authority.

**Temperature data provided courtesy of the Nati onal Park Service. Values are mean monthly minimum followed by mean monthly maximum.

(X) 9 . 67% of the study area. In this area, pines can occur in almost pure

even-aged stands with an understory of hemlock . Rhododendron

(Rhododendron sp.) can be found in almost pure dense thickets along Cooper Branch and Anthony Creek . The canopy of the upland forests is

dominated by oaks {Quercus sp.), maples (Acer sp.), pines (Pinus sp.),

. - and hemlock (Tsuga canadensis), while the understory trees are hickory

(Carya sp.), dogwood, hemlock, rhododendron, pine, and maple . The

shrub understory ranged from absent to dense thickets of rhododendron

or huckleberry.

Fauna

The Great Smoky Mountain National Park is part of the range of

approximately 59 species of mammals, many of which occur in Cades

Cove. Eastern cottontail (Sylvilagus florfdanus), white-tailed deer

(Odocoileus virginianus), woodchucks (Marmota monax), eastern chipmunks

(Tamias striatus), eastern gray squirrels (Sciurus carolinensis), gray

foxes (Urocyon cinereoargenteus), black bears (Ursus americanus), raccoons (Prycyon lotor), and striped skunks (Mephitis mephitis) are a few of the more commonly observed mammals around the campground (Linzey

and Linzey, 1971). In addition to the mammals there are over 200

species of birds, 21 species of snakes, 8 species of lizards, 6 species

of turtles, 12 species of frogs and more than 21 species of salamanders

occurring in the park (Huheey and Stupka, 1967). CHAPTER III

MATERIALS AND ME THODS

Capture and Handl ing Techniques

2 . Data col lection . The study area was divided into 25 , 250m grids to distribute trapping effort evenly over the study area (Figure 1, page 5). In 1979, the trapl ine consisted of 9 traps di stri buted evenl y over 3 randomly chosen grids. The trapl ine was moved to new grids every 8 days. In 1980, the trapline wa s increased to 12 traps evenly distributed among 4 randomly chosen grids and were moved every 5 days. The number of trapnights per trap peri od was changed in 1980 to reduce di sturbance by bears and the recapture of trap- prone skunks. Traps were not set in the campground or the picnic area between June and mid-October due to campers interferring wi th the traps.

Skunks were captured in wire live traps (20cm X 20cm X SOcm ,

Tomahawk Li ve Trap Co., Tomahawk , WI ) bai ted wi th sardines. Captured animal s were anesthetized usi ng 4-8 mg/kg of Ketami ne HCl (Personal

Communication , Dr. S. Kennedy ) (Bristol -Myers Lab, Syracuse , NY ) administered intramuscul arly using a 1 cc tubercul in syringe wi th a

5/8 in, 23 ga needle. Skunks were tagged (Nati onal Band and Tag Co. , styl e 1005, size No. 1), sexed, weighed , and morphological parameters measured. Reproductive status of males were determined by the size of the genitals. Reproducti ve status of females were determined by examining the teats and abdominal region for swelling. The tags used in 1979 were backed wi th col or vi nyl of various shapes to facilitate

10 11

identification of an animal at night. The col ors used were red , yellow, white, blue, and green with shapes that incl uded circles, triangles ,

squares , and hearts. The tags used in 1980 were backed wi th yel low or orange reflecti ng tape. Ear tags were attached to the right ear of females and to the left ear of ma les.

Ages of the stri ped skunks were estimated by we ight and denti ­ tion . Weight is most useful in separating adults from juveniles during the spring and summer, but becomes less effective in the fal l

as the young approach their adult size and weight (Al len, 1939) •

. Dentition is useful in aging animals younger than 3 months; after approximately 55 days the last tooth erupts and aging by dentition is no longer possi ble (Verts , 1967) . Thus, by September it is difficult to separate adults from juveniles.

·Trap success for each month of the 1979 and the 1980 field season was calculated by dividing the number of skunks per month by the number of trapnights duri ng that month .

Among skunks the amount of white in the pel age varies depending on the length and width of the dorsal stripes . Each skunk pel age was placed in 1 of 4 grades , created by the fur trade and based on the amount of whi te and coat pattern of the individual: black star, short stripe , long stripe, and broad stripe (Fi gure 2) (Bachrach , 1930) ..

Statisti cal analysis of morphologi cal data. The means and · ranges were calculated on the morphol ogical data for each month of 1979 and 1980. An analysis of variance was performed on these data to determine if there were differences in morphological characteristics within age or sex by month and between age or sex by month . The ANOVA 12

BLACK (STAR)

SHORT STRIPE

LONG STRIPE

BROAO STRIPE

Figure 2. A classification of color patterns of striped skunks. 13 and the simple statistics were done by comput�r using SAS (Helwig and

Council, 1979). Chi-square tests were used to determine if the sex ratio of adult males and adult females and juvenile males and females differed significantly from the expected 1:1 ratio.

Population Dynamics

Popan-2, a data maintenance and analysis system for mark- recapture data was used to estimate the size of the skunk population in the study area. The analysis in Popan-2 is based on the equations developed by Jolly and Seber (Arnason and Baniuk, 1969). This method yields the most information for the fewest assumptions. The basic assumptions are: captures are random with respect to marked and unmarked animals, all previously marked animals can be recognized, and the population is open.

The Popan-2 program allows capture data to be pooled over intervals of time to increase precision of the estimate. It was necessary to pool the capture data from this study because of the low and variable number of captures. When pooling an additional assumption must be met: the population must remain roughly constant within a pooling interval, but may change across intervals (Arnason and Baniuk,

1969). In order to meet this assumption the data were pooled over 4 night intervals.

Seasonal and Daily Activity Patterns

Strip counts. The strip count technique was used in this study

• to determine activity patterns and the size of the skunk population in the campground during the night; it is based on the strip count method 14 descri bed by Tanner (1978). A total of 4 strip counts were performed

duri ng the 1980 fi eld period: 31 May-1 June, 12-1 3 July, 16-17 August,

and 11-12 October. Each stri p count began an hour and a hal f before

sunset and conti nued for 12 hours after sunset. Sunset was used to

·standardize the time. The transects were wal ked every 1.5 hours by an

ob server (Fi gure 3). A minimum of 6 observers were used to walk pre­

determined transects and record al l skunks and their location wi thi n ·

the assigned transect. The streets in the campground were the center

line for the transects. The transects were 35 m wide based on the

. maximum dis tance an observer could sight a skunk wi th a hand-carried ,

battery powered fl ashli ght. Each observer was positi oned at the head

of the assigned transect and began the wal k after receiving a si gnal

by fl ashl ight to begin the wal k; thi s insured that the observers began

the wal k simultaneously.

Due to the design of the campground and the limi ted numbe r of

observers , each stri p count wa lk wa s conducted in 2 parts wi thin the

campground: 8 �ection and the group camping secti on and then C

section. Based on fi eld observations, it was possible to sample the 2 sections sequentially wi thout bias because of the presence of wooded

areas between the 8 and C sections and the fact that the rate at which

an individual skunk moves when foraging is slowe r than the rate of an

observer walki ng. The transects in 8 section and the group camping

section and in C section we re measured and totaled 1.75 km and 1 km,

respecti vely.

Stati stical analysis. Data col l ected during the 4 stri p counts

were used to determine the size of the skunk population in the .campground 15

Figure 3. Ma p of the roads and camps ites in the Cades Cove campground , GSMNP. (Copy of map handed out at campground station. ) 16

during the night and the_dail y activity patterns of the skunks. The

following equation was used to estimate the number of skunks in the

campground for each walk: 2 n � number of type i/skun s ounte � 112L ( k c d )= .N per hectare i =1 °1 i =1

(Tanner, 1978). The symbols L and D represent the total length of the transects and the width of the transects , respectively.

A two-way mixed analysis of variance without replication was run to determine if the daily activity patterns of the campground skunks

·varied significantl y with time or season. A Student-Newmans-Kuels multiple range test was performed to determine which time periods were significantly different. In addition a stepwise multiple regression was performed to determine the relative importance of certain factors

influencing the foraging activity of the skunks in the campground .

Maximum temp�rature, minimum temperature, cloud cover , precipitation , time, and date were the independent variables used to determine the

linear relationship to the dependent variables , the number of skunks

in the campground. The stepwise mul tiple regression was done on the

computer using SAS {Helwig and Council, 1979).

Home Range

Night observations. In order to collect data on home range and

behavior, night observations were conducted at least 3 times a week

during the summer and on the weekends during the spring and fall. The

night observations consisted of observing skunks in the campground for

5 continuous hours during l of 3 time periods : 1700 hrs-2200 hrs , 17

2200 hrs-0300 hrs, and 0300 hrs-0800 hrs. Behavior data were col l ected by scan-sampl ing to obtain an overview of the activity in the camp­ ground (Lehner, 1979). Records were ma intained for each skunk sighted as to where and when it was last observed . This information wa s used to determi ne the degree and intensity of contact between skunks, the factors infl uencing the be havior of the skunks and the home range of the skunks .

Home range estimation. Home range was calculated by the convex polygon method which invol ves drawi ng the smal lest convex pol ygon containing al l the capture and observation points . The home range for each skunk wi th 6 or more points was cal cul ated by the surveyor•s formula

(Olliver and· Glendinning , 1919) where A is the home range area and X and Y are the coordi nates of capture or observati on points on a map of the study area.

Denning

Den characteristics. Dens were located by observing skunks recently released from traps or by observing free-roami ng skunks in the campground. Si tes were confirmed as dens by direct and indirect evidence. Di rect evidence was col l ected by observi ng dens between 1800 hrs . and 2300 hrs. and between 0400 hrs. and 0700 hrs . These intervals correspond to the periods of time when skunks were normal ly leaving or return i ng to dens. Other signs of use , such as fecal material , fresh 18

dirt at entrance(s), paw prints, or smell were considered indirect

evidence that a site was a den. Additional informati on on the location

of dens was obtained from park rangers.

Several measurements were taken at dens : hei ght and width of

entrance(s), number of entrances, and entrance aspect. In addi ti on,

. . dens were classified by their position above or below ground and by den

type. There were 4 different den types : 1) cavities at the base of

dead trees, logs, or stumps, 2) cavities at the base of liv ing trees,

3) burrows, and 4) cavities under rocks . Physical data were recorded

·for each den, whi ch incl uded distance to water, distance to the camp­ ground, elevation, soil type, slope aspect, the degree of slope and microtopography.

In addition to natural dens skunks also used manmade structures,

such a� culverts located in the campground. The dens in the cul verts were found by checking the drainage cul verts every other week in 1979

and once a week in 1980.

Den availability. During the 1980 field period, den parameters

and surrounding physical parameters were measured for each potential

den site to determi ne the availabi lity of dens for striped skunks

in the study area. The potenti al dens were located by searching a 2 40 m plot around 54 randomly chosen points . These plots covered 5.5%

of the total study area . The den parameters and surrounding physical

parameters measured were the same as the data col l ected on the con­

firmed dens. 19

Statistical analysis. A factor analysis using den and physical . parameters was performed to identify those variables which account for

the greatest amount of variance among the dens. A discrimi nant

analysis was performed on the den and physical parameters fo r confirmed

dens and potential dens to look for differences between the 2 ty pes.

A second discriminant analysis and a canonical correlation analysis

were performed using the den parameters and physical parameters of the

confirmed dens to determine if summer and fal l dens consti tute 2 sepa­

rate groups. The factor analysis and the canonical correlation analy-

. sis were done on the computer using SAS (SAS User's Gu i de, 1979). The

discriminant analysis was done by computer using BMDP (Dixon and Brown ,

(1 975 ). CHAPTER IV

RESULTS AND DISCUSSION

Trap Data

Trap success. Out of 1381 trapnights, a total of 180 skunks were trapped during the study: 37 adult males, 65 adult females, 45 juvenile males, 25 juvenile females, and 4 females and 4 males of unknown age (Tables 2 and 3}. Of these 180 skunks, 69 were captured in 1979 only , 25 were recaptured in 1980 from 1979, 16 were captured

only once in 1980, and 45 were recaptured from 1980 only. The low number of 1979 skunks recaptured in 1980 suggests a rapid turnover rate in the skunk population from year to year. Verts (1967} also observed a high turnover rate for the skunk population in Illinois.

Trap success was highest in August of 1979 and July of 1980, corres­ ponding to the time of year when the juveniles were becoming indepen­ dent and available for capture (Tables 4 and 5).

Sex ratio. Adult females were captured more frequently than adult males during the 1979 and 1980 field periods. The sex ratio of adult males to adult females was 1: 1.3 in 1979 and 1: 2.5 in 1980

(Table 6}. The 1979 sex ratio for adult skunks was not significantly different from the expected 1:1 ratio (X2 = .93, df = 1}, but in 1980 there were significantly more adult females captured than adult males

(X2 = 9.0, df = 1} (P<. 005}. Previous studies by Verts (1967), Bailey

(1971}, Hamilton (1937}, and Bennitt and Nagel (1937} have reported

20 TABLE 2. Number of Striped Skunks Captured by Month in the Cades Cove Campground Ar�a. GSMNP, 1979.

Males Femal es Month Adults Juveniles Dnknowns Jt

June 1 - - 2 - - 3

July 3 5 - 9 2 - 19

August 9 15 4 14 10 4 56

September 3 - - - - - 3

October 4 1 - 4 - - 9

November 1 - - 1 - - 2

December 2 - - - - - 2

Total 23 21 4 30 12 4 94

N ..... TABLE 3. Number of Stri ped Skunks Captured by Month in the Cades Cove Campground Area GSMNP , 1980.

Males r Females Month Aau1ts Juvenlies jl\du1ts . Juveniles Total

January 1 - - 1 2

Apri l 3 - 6 - 9

May 3 - 6 - 9

June 2 - 2 - 4

July 1 14 19 8 42

August 4 10 2 4 20

Total 14 24 35 13 86

N N TABLE 4. Trapping Success for Striped Skunks Captured in the Cades Cove Study Area; GSMNP , 1979.

Trap Number of Trap success{%) Number of Number of Number of Trap success for Month nights skunks for skunks opossums raccoons turtles all animals

June 8 3 37 .5 1 - - 50. 0

July 176 19 10.8 3 - - 12.5

August 225 56 24.9 - 3 1 26.7

September 151 3 2.0 1 - - 2.6

October 81 9 11.1 4 4 - 21 .0

November 90 2 2.2 - - - 2.2

December 60 2 3. 3 - - - 3. 3

N w TABLE 5. Trapping Success for Striped Skunks Captured in the Cades Cove Campground Study Area. GSMNP. 1980.

Trap Number of Trap success(S) Nwnber of Nu!OOer of Numb�r of Number of Trap success(%) Month nigh�s skunks for skunks o�ossums raccoons turtles cats for a 11 anima1 s

January 45 1 2.2 - - - - 2.2

Apr11 43 9 20.9 - - - 1 23.3

May 64 9 14.1 - 2 - - 17.2

June 60 4 6.7 4 - - - 13.3

July 198 42 21.2 1 2 1 1 23.2

August 180 20 11.1 3 1 1 - 13.9

N -'=" TABLE 6. Sex Rati os of Striped Skunks by Year Expressed as Males pe r Females in the Cades Cove Campground Study Area , GSMNP, 1979-1 980.

Adults Juveniles

Year Males Females Ratio Miles-----Females-�- -�- -Ratio

1979 23 30 1.0 :1.3 14 35 1:2.5

1980 21 12 1.7:1.0 24 13 1.9 :1.0

N U'1 _, "� 26 adul t sex ratios favoring ma les rather than females.

Among juveniles the ratio of females to ma les was 1:1 .75 in 1979

and 1:1 .85 in 1980. Neither sex ratios�.we re significantly different 2 2 from the expected 1:1 (1979: x =2.4, df=l ; 1980: x =3.24, df=l ). The

above is in agreement wi th reported sex ratios for juvenile skunks in

other areas (Verts , 1967).

The reasons for the differences between the sex ratios calcul ated

in this study and those from other studies and the deviation from the

theoretical 1:1 sex ratio may be due to sampling bias caused by sexual

·differences in behavior, differential mortality between the sexes, and

unknown causes. The low number of adult males captured during this

study was probably the result of not trapping extensively during the

breeding season , March-May . According to .Verts (1967) and Al len {1 939)

males tend to be caught more frequently during the breeding season, but

rarely duri g the spring or summer. �. It was not possible to determi ne the causes of skunk mortal ity

due to the limited number of recorded deaths during the study. Of the

10 recorded deaths , 2 adult males and 1 adult female were kil led by

cars , 2 juvenile males died in traps , 1 skunk of unknown sex was stoned

by campers , 2 skunks of unknown sex died of unknown causes , and 2

skunks of unknown sex were kil led in dog attacks. These limited·

mortality data suggest that deaths due to man's involvement appear to

affect the sexes equally. Other forms of mortality whi ch could not be

measured , such as predation , disease and particularly starvation , may

influence the sex ratio. Verts (1967) felt that starvation during the

winter was more common among females in Illinoi s than among males. 27

External Morphology

Measurements. All body measurements except front foot length

and ear length were larger in male than in female striped skunks

(Table 7). Tail length, hind foot length, and body length were sig­

nificantly different between males and females {F test, P<0.05). Total

length, front foot length, and ear length were not significantly dif­

ferent between males and females (F test, P>0.05). . In general, males

tended to have longer bodies and shorter tails than females. Similar

differences between the sexes were reported by Verts {1967) and Bailey

{1971).

Total, tail, and hind foot measurements from skunks caught in

Cades Cove were comparable to measurements of skunks in Michigan (Allen,

1939) {Table 8). However, ear length could not be compared because Allen

measured the ea r length of his skunks from the crown rather than the

notch. The mean total, body, and hind foot length of Cades Cove and Michigan skunks were significantly smaller than the measurements

recorded by Verts {1967) for Illinois skunks (F test, P<0.05).

Differences in external measurements of these 3 skunk popu­

lations can be explained by examining the characteristics and distri­

bution of subspecies of the striped skunk found in the Eastern United

States (Figure 3) . The 4 subspecies of striped skunk are: the

northern plains skunk (Mephitis mephitis hudsonica), the eastern skunk {Mephitis mephitis nigra), the Florida skunk {Mephitis mephitis

elongata), and the Illinois skunk (Mephitis mephitis avia) (Hamilton,

1943). The northern plains skunk is the largest of the Mephitis

·group, ranging from the.northern peninsula of Michigan, through TABLE 7. External Measurements(cm) of Adult Male and Female Striped Skunks Captured in the Cades Cove Campground Study Area, GSMNP, 1979-1980.

Sample Hind Front Sex size Total .Body Tail foot foot Ear Number of teats Mean 57. 8 35. 4 22. 5 6.0 4. 1 1.6 11

Female 60

Range 42.2-64. 1 27.0-42.5 16-29.4 5.4-7.6 2. 2-5. 0 1.0-3.2 10-14

Mean 58.8 37.3 21.5 6. 5 4. 3 1. 4 2 Male 35

Range 48.1-67. 0 28.7-45.4 14.0-26.6 6. 0-7.7 3.1-5.4 1.0-1. 9 2

N . (X) TABLE 8. External Measurements of Striped Skunks from 2 Different Populations Reported in the Literature .

Verts {1967} Al len {1939}

Aa1e · female Male Female

Total length 63.1 60.2 58.1 56.2 (54.0-76.5) (52. 0-67.0) (56. 0-61 . 0) (54.1-59.3)

Body length 40. 7 37.8 (32 .0-47.8) (30.2-45.0)

Tail length 22 .3 22.5 22.4 22.8 (1 7.5-30.5) ( 1 7 • 3-31 . 0 ) (1 9.5-25.0) (21 .2-24.6)

Hind foot length 7.5 7.1 6. 7 6.3 (6 .6-8 .5) (5.9-7.8) (6 .0-7.0) (5 .8-6.4)

' Ear length 2.9 2.9 (2.5-3 .5) (2 .5-3.3)

N 1.0 30

�mephitis nigra � mephitis elongata Ill mephitis avia hudsonica

;;::::mountains

Figure 4. The range of 4 subspecies of striped skunk (Mephitis mephitis) in the Eastern u.s. (Hamil ton. 1943) . 31

Wisconsin and into Illinois where it intergrades wi th the eastern skunk and the Illinois skunk. The eastern skunk is sl ightly smal ler than the northern plains skunk and is distri buted from Maine westward to the lowe r peninsula of Michigan and southward to central Alabama and Mississippi. The Ill inois skunk is the smallest of the Mephitis group and ranges throughout Illinois. The Fl orida subspecies is smaller than the eastern skunk and ranges from southern Virginia to southern Florida and westward along the gulf of Louisiana (Hal l, 1981 ;

Hami lton, 1943) .

There were no significant differen�es between the external measurements of striped skunks sampl ed in Michigan and Cades Cove.

However, skunks in Illinois were significantl y larger than those in

Mi chigan and Cades Cove. The subspecies studied by Al len (1939) in

. Michigan was the eastern skunk; this same subspecies of inter- grades in East Tennessee wi th the Florida skunk. Based on the simi­ larity to skunks sampl ed in Michigan , the characteristi cs of skunks from Cades Cove.tended more toward the ea stern skunk than the

Fl orida skunk. According to Verts (1967) the characteristics of skunks in Illinois tend more toward the northern plains skunk than the other 2 subspecies in Illinois. Since the northern plains skunk is the largest of the Mephitis group, the differences between sk�nks in Ill inois and those in Michigan and Cades Cove is not surprising.

There were no significant differences between-external measure­ ments of juvenile males and juvenile females (Tabl e 9); data from other studies were not available for comparison. The·juvenil es, by . . late August or early September in Cades Cove , are approxi matel y the same size as adult skunks and may be difficult to disti_nguish from TABLE 9 . . External Measurements (cm) of Juvenile Male and Female Striped Skunks Captured in the Cades Cove Campground Study Area, GSMNP, 1979-1 980.

Sample - Hind Front Month Sex size Total Tail foot foot Ear

mean 42.2 17.7 5.3 3.9 1.4 Female 10 range 37.4-54.0 14.0-23.5 4.8-5.9 2.7-4.6 1.0-1 .6

July mean 43.2 17.8 5.6 3.7 1.6 Male 19 range 35.6-55.3 14.0-25.1 4.8-6.5 3.0-5.0 1.2-2.2

mean 51 .6 21 .5 6.2 4.2 1.3 Female 12 range 45. 8-58.1 19.0-27.0 5.6-7.5 3.3-4.5 1.0-1 .8 August mean 52.6 21 .6 6. 2 4.4 1.2 Male 24 range 39.4-59.8 16.1-25.4 5.1-6. 7 3.9-5.6 1.1-1 .5

w N 33 adults . Al len {1939) found that juveniles in Michigan were still sl i ghtly smaller than adults and could be distingui shed from adults unti l late October. Juveni le skunks in Cades Cove may attain adult size more rapidly because of the greater abundance and qual ity of food.

Weights. The body weights of adul t male and female skunks di ffered signifi cantly (F test, P

Mal es, however, began to gain we ight as early as May. Females in

Ill inois {Verts, 1967 ) and Ohio {Bai ley, 1971 ) exhibi ted a similar tendency to gain weight after the juveniles became independent.

The weights of striped skunks·captured around the Cades Cove

campground were signi ficantly heavier than the we ights of skunks in

Mi chigan ; this difference appears to be related to the greater availa­ bility of food and to the longer foraging season in Cades Cove. The

Cades Cove campground provides a year-long food source; skunks do not expend large amounts of energy to obtain food . The food the skunks were observed obtai ning from the campers--popcorn, hotdogs, cookies,

and peanuts--probabl y has a higher fat content than their natural diet {Hediger, 1964 ) of insects, spiders, fruits, smal l mammal s, and TABLE 10. Average Monthly We ights(kg) of Adult and Juvenile Striped Skunks Captured in the Cades Cove Campground Study Area , 1979-1980.

Adult Juvenile Adult Juvenile Si!llple Sa 111ple Simple Sample Month size Mean Ranse size Mean Ranse size Mean Rani! size Mean Ra nse

Apri 1 3 1.8 1.4-2.5 - - - 6 1.8 0. 9-2.5

May 3 1.5 1.4-1 .5 - - - 6 1.9 1.7-2.3

June 3 2.4 2.3-2 .7 - - - 4 1.9 1.5-2.3

July 4 1. 9 1.4-2.3 19 0.8 0.5-1.4 28 1.7 1.1-2.3 10 0.5 0. 3-1 .0 ' August 7 1.9 1.2-2. 7 25 0.8 0.5-1.5 16 1. 5 0.9-2.7 16 0.8 o. 5-1 .4

September 3 2.5 1.9-3 .2

October 4 3.0 2.3-3.6 1 2.3 2.3 4 2.3 1.8-2.9

November 1 2.3 2.3 -- - 1 1.8 1.8

December 2 2.7 2.7

January 1 2.2 2.2 ------1 1.4 1.4

w � 35

carrion (Hamilton , 1936; Kelker , 1937). This high caloric intake may

account in part for the heavy body weight of the Ca des Cove campground

skunks. Also the mild winters in Tennessee allow skunks in Cades Cove

to remain active for a longer period in the fall than Michigan skunks

·and to store more fat. Hamilton (1936) also recorded higher body

weights for skunks with access to garbage than skunks from farm areas.

He attributed the higher weight to the abundance of food (garbage) in

towns.

Juveniles captured in July were approximately 60 days old and

. weighed an average of 0. 55 kg (R=0. 3 to 1.44. kg) (Table 10) . There was

no significant difference between the weights of males and female

juvenile skunks at this time. By the end of August weights of juveniles

had increased by 60% ; this is 30% less than the average weight of an

adult at this time . Weights of juveniles approached adult weight by

October. Data on weights of juveniles or other areas of the United

.. States were not available.

The weig�t of skunks in the fall appears to be important in its

- ability to survive the winter. By fall weights of males increased by

50% and females by 11%. The increase is almost exclusively white

adipose tissue which serves to supply the body with energy during the winter when food is scarce (Alesiuk and Stewart, 1962). little change

in the dry fat-free body weight or weight to body water ratio was

recorded between fall and winter (Mutch and Alesiuk , 1977). If an

insufficient amount of fat is stored, starvation can occur during the

winter. Starvation m�y be one cause of winter mortality among skunks,

particularly juveniles, who tend to enter carnivorean lethargy with

slightly lower body weights. Starvation during the winter may be one 36

reason for the high juvenile turnover rate in the skunk population in

the Cades Cove campground that is evidenced by the low number of

recaptures from one summer to the next.

A considerabl e amount of the body we ight gained during the

summer and fall is lost during the winter (Table 11); this wa s an

- average of 33.5% for males and 10.0% fo r females. These losses are

comparable to losses reported in other areas. Michigan skunks lost

36 .6% and 31 .6% body we ight (Al len, 1939) , Minnesota skunks lost 13.8%

and 37.0% body weight (Houseknecht and Tester, 1978) , and Ohio skunks

·lost 42.4% and 19.3% body we ight (Bai ley, 1971) fo r males and females,

respectively. The vari ation in the percentage we ight lost between

locations is likely due to diffe rences in the environment (cl imate,

food , and den sites). A possible explanation for the variation in

we ight' loss between the sexes is that males tend to remain active

during the winter (Sunquist, 1974) . If fo od is avai lable, it replen­

ishes fat · supplies and less we ight is lost. If food is not available,

activity decreases their fat supply; this results in a higher loss of

. body we ight. Females are not as active during the winter and generally

only emerge when the weather is mi ld (Hami l ton, 1937 ; Davis, 1951).

Since females store less adipose tissue, it is advantageous for them to

be less active and consequently expend less energy on thermoregulation .. elage

Of the 109 skunks available for classification , 9 were bl ack

stars , 20 were short stripes, 29 were long stripes , and 34 were broad

stripes (Table 12). The remaining 17 skunks had atypical coat patterns

' which differed from the typical by having brachial -antebrachial 37

TABLE 11. Comparison of Body We ight Losses During the Winter in 4 Different Stri ped Skunk Populations.

Invest1gator{s} Male Female

Bailey ( 1971 ) 42.2% 19.3%

Houseknecht and Tester (1978) 13.8% 37.0%

Al len (1939) 36.6% 31 .6%

Goldsmith (1981 ) 33.5% 10.0% TABLE 12. Numbers of Striped Skunks in Each of the 4 Co 1or Patterns for Cades Cove .Campground Study ·Area, GSMNP, 1979-1980.

Sex Black(star) Long stripe Narrow stripe Broad stripe

Male 5 12 12 11

Female 44 8 17 23

Total 9 20 29 34

w co 39 stripes, lacking dorsal stripes or a nose stripe, having unequal stripes , _ �tripes, and having broken stripes along the back (Fi gure 5). Brachial­ antebrachial stripes are located on the outside of the front legs I (Stain and Stuckey, 1960).

The coat of stri ped skunks is uni quely conspicuous in nature due to the bl ack and white col or pattern ; thi s likely acts as a wa rning to predators that skunks possess an efficient means of defense.

Predators wi ll in general avoid skunks if the skunks are seen and recognized in time (Cott, 1940).

Population Dynamics

Based on mark-recapture estimates the resident skunk population wa s 31 ± 4 skunks (1 skunk per 5.0 ha ) wi th a temporary increase to

76 ± 31 skunks (1 skunk per 2.5 ha } in late Ju ly and early Augu st of 1980. Pri or to late July and early August, the juveniles were under the care of their mothers and not as available for capture . In late July and early August, family units consisting of the female and . young separated ; the juveniles dispersed into the general population , increasing their probability of being captured . As the juveniles dispersed in late August and September, the number of skunks in the study area decreased.

The estimated popul ation of skunks for thi s study area wa s· higher than that reported from other areas wi thin the species range

(Table 13). In areas of Illinois that conta1 n a mi xture of pastu re, open woods, and brush, broken by wooded ravines and rock outcrops, skunks averaged 1 skunk per 7.2 ha (Verts, l 967 ). The campground and picnic area appear to infl uence the skunk populati on by increasing the availabi lity of both dens and food . 40

LACKS NOSE STRIPE

- ··::-s.. ··--

ISLAND STRIPES

LACKS DORSAL STRIPES

UNEQUAL STRIPES

Figure 5. Atypical color pattern of 4 striped skunks in Cades Cove campground study area, GSMNP . 41

TABLE 13. Reported Densities of Striped Skunks in Various Habitats in the U. S.

Researcher Habitat State Densitl

Dean (1965 ) Forest, pasture, Maine 1/4.5 ha orchard , and a garbage dump

Bailey (1971) Marsh Ohio 1/22.0 ha

Verts (1967) Agricultural Illinois 1/7.2 ha

. Present study Mountains and Campground Tennessee 1/5.0 ha 42

Seasonal and Dai ly Activity Patterns

Seasonal changes in the activity patterns of striped skun ks around the Cade s Cove campground were observed during the course of th is study (Fi gure 6}. Of the 6 vari abl es measured and subjected to a regression analysis, temperature had the most signifi cant infl uence on acti vity and in combination with al l the me asured variabl es accounted for 39.0% of the variance observed in skunk acti vity. Other variabl es such as light intensity and the number of campers in the campground the week before a skunk survey wo ul d probably expl ain more of the variance . Skunks tend to concentrate around areas contai ning den sites and food (Verts , 1967}. In the summer, as the number of campers in the campground increased , food became more available for skunks re sulting in their usi ng the campground mo re extensively.

Al esiuk and Stewart (1962 } reported that tempe rature , although not the only factor , played an important role in control ling level s of activity in skunks ; as the tempe ra ture decreases activity tends to decrease .

During May and Apri l ne i ther sex used the campground extensively as a source of food and few animals we re observed in the campground

(Fi gure 6, page 43) . An average of 5 skunks were observed in the campground during nights in Ap ril and May , bu t the number of skunks observed in the campground increased to 122 in July (Fi gure 6}.

Lactating and pregnant females tend to restrict their movements at I thi s time and rema in close to their den s (Bailey, 1971 }. Males do �ot appear to use the cafl1)ground as a primary food source in the spri ng.

During April and May the campground is only partial ly full during the weekdays and ful l on the wee kends when the weathe r is good , so the 43

C1l Sunset at 2075 c: ::::J � I � I . I ::E ln t='? c=J

30 Sunset at 2090 25 20 >, - ::::J 15 � 10

5 0

30 - Sunset at 2025

25 - ....--- 20 � VI ::::J Cl 15 - ::::J r-- - C( 10 -

. 5 - 0 I------,

15 Sunset at 1900 '- C1l 10 · ..0 0 � u 5 0 0 -1 .5 0 1.5 3.0 4.5 6.0 7. 5 9.0 10.0 12.0 Hours After Sun�et

Figure 6. Mean number of skunks observed during the st�ip counts in the Cades Cove campg round, GSMNP, 1980. 44

available garbage and discarded food may be limited . The area surrounding the campground may contain more food for skunks , since insects should

be emerging after winter dormancy. In the spring, skunks are primarily

nocturnal wi th only an occasional sighting duri ng the day.

By late June or early July, the activity of skunks increased and

. the number observed between 1900 hrs . and 0800 hrs . in the campground

increased signifi cantly since spring (F test P<0.05) (Figure 6) . Thi s

increase can be attributed to several factors . The campground was

ful l duri ng the weekdays and the weekends, so available fo od and

. garbage are not limited . Females are ranging further from the dens

because the young are mature enough to be left alone or they accompany

the female increasing the number of females and family units observed

in the campground. In July and August an average of 30 skunks and 2

·fami ly units were observed during night observations. The fami ly units

were only observed during night observation between 1700 hrs. and 2200

hrs. Of the 16 skunks observed during a night wal k, 5-6 were tagged

females , 3-4 we re tagged males, 3-4 were skunks seen during previous

observation periods , but not tagged , and 3-4 were new skunks .

Skunks were more nocturnal than diurnal throughout the summer.

According to Houseknecht (1971) skunks were usual ly active 40-100% of

each hour during the night from July to October. Skunks began to

forage in the campground approximately 1 hour before sunset (Figure 6) .

At this time campers were preparing supper and there we re plenty of

food scraps and garbage. Skunk activity peaked at 2300 hrs . wi th an

estimated 46 skunks i� the campground (Fi gure 6) . The decrease in

activity after 2300 hrs. may be rel ated to a decrease in the availa­

bility of food scraps and garbage. 45

A second peak in activity occurred at 0150 hrs. and appeared to

consist of skunks that did not utilize the campground as a main food

source. These skunks appeared to be different from the skunks observed

earlier in the evening in 2 ways: they were more cautious of people,

· avoiding campsites with active people, and they were difficult to

approach or observe.

By 0800 hrs : activity of skunks had decreased except for an

occasional animal. Skunks (N=30 per season) were observed in the camp­

ground in the middl e of the da y; however, it is not possibl e to determine

. if this is unusual since data from other studies are not availabl e.

Skunks observed foraging in the middl e of the day in the campground

probably den in the culverts and emerge in response to the smel l of

food.

·The activity of campground skunks was basically in agreement

with Sunquist (1974) and Houseknecht (1971). Cades Cove skunks tended

to emerge ea�lier than skunks in other areas but this appears to be

an adaptation t� the earlier availability of food in the evening. All

other studies showed a pattern of bimodal activity for summer except

for Verts (1967); his data concentrated on juvenil es and juvenil es

tended to have a unimodal pattern of activity.

Fewer skunks were observed in the campground in October due to

a general decrease in their activity in response to decreased temper­

atures, changing light intensity, and a decrease in the availability

of food (Mutch and Alesiuk, 1977; Al esiuk and Stewart, 1962). The

activity was evenly distributed throughout the night, with other ·

investigators (Sunquist, 1974; Houseknecht and Tester, 1978; Verts,

1967) reporting continued low levels of activity during the day 46

(Figure 6). Sunquist (1 974) reported that skunks were active about

45% of each hour of the night in November.

In northern climates the decrease in activity is more noticeable

and by December skunks are completely inactive (Davis, 1951). Skunks

remain active in areas with mild winters and a year long food supply,

except during cold spells , periods of snow cover , and scarcity of food .

Skunks are not adapted to the cold or to locomotion through deep snow

(Verts , 1967). In Cades Cove , where winters are mild and food is

available year round, skunks were observed in the campground and in

·the Cove in low (n=28 per month) numbers throughout the winter. The

campground horse stables, and the garbage of park personnel, provide

continuous food for skunks.

Human-Skunk Interactions

Striped skunks in Cades Cove campground did not panhandle food

directly from the campers, rather they obtained their food in the

campground by foraging in the campsites with apparent 11 dignified

indifference .. to the campers. Preferred foraging sites in the camp­

sites were under picnic tables , by fire places, and in trash bags left

on the ground by campers. Skunks often passed within 30 em of campers

while foraging in a campsite. Skunk-human interactions were observed

on 42 occasions during the course of the study. Skunks were observed

taking food from campers on 5 occasions , all of which were camper

initiated. The only agonistic interactions between skunks and . ca�pers

occurred when campers tried to force a skunk out of a culvert (n=lO)

' or threw obj ects at skunks (n=27). Basically skunks tolerated close ' proximity to people without displaY,ing agonistic behavior even when 47

harrassed by being chased or yel led at. Since harrassment did not

seem to deter skunks from foraging in the campground, other methods

woul d probably be required to decrease human-skunk interaction.

Home Range and Territorial ity

Home range . The estimated home range for ma le, female, and

juvenile skunks in this area were 15.6 ha , 24.3 ha , and 15.8 ha,

respectively. These estimates are significantly lowe r than those

reported by Verts (1967) and Storm (1972). However, estimates from

·the present study were based on location data from mark-recapture and

night observations , both of whi ch tended to underestimate home range ,

whi le the other 2 studies were based on locational data from radio

tel emetry. Mark-recapture data biased the estimate of home range

because skunks moved outside the immediate study area and animals

became trap-prone or tray-shy. Also data col lected by night obser­

vations likely biased estimates of home range since skunks were

monitered only· in the campground.

Food supply does infl uence the size and shape of home ranges

(Verts, 1967); therefore skunks foraging in the campground where food

is concentrated should have smaller home ranges than skunks foraging

elsewhere. Campground skunks probably expend less energy foraging

and have a higher energy gain than skunks outside the campground.

Skunks feeding mostly in the area surrounding the campground probably

have larger home ranges accommodating the seasonal changes in the

distribution of food.

When compared to males and juveniles, females had a signifi­

cantly larger percentage of their home range wi thin the campground, 48 suggesting that femal es forage in the campground to a greater extent

than males (Table 14). Males tend to range over larger areas than

females (Verts, 1967 ; Storm, 1972) , forage over a wider range of

habitats, and swi tch locations in response to changes in density of

food items. Females may be more sensitive to the concentration of

food in their immediate vicinity due to the energy demands of caring

for the young and the need to remain cl ose to the young. Due to the

high density of food , the campground provides females with an excellent

foraging area with few competitors , such as raccoons and opossums .

·In late July and early August, when newly independent juveniles increase the density of skunks foraging in the campground, fewer females are

observed there. It is speculated that the females switched to

foraging more extensively in the surrounding area in response to

crowding and depletion of the easily obtainable food due to the

presence of the juveniles.

Juveni.les tended to have more of their home ranges within the

campground than . males, but less than females since juvenile home

. ranges probably correspond to that fraction of the mother•s home

range over which the fami ly unit foraged before the juveniles became

independent. Prior to the break up of the fami ly uni ts, females and

their young were observed foraging over smal l areas of the campground

between 1900 hrs. and 2100 hrs., while females (n=9) ranged throughout

the campground after 2100 hrs . In early August, 7 juveniles we re

observed foraging alone in the same area their mothers had previously

been observed foraging. TABLE 14. A Summary of Movement Parameters of 17 Striped Skunks in the Cades Cove Campground Study Area, GSMNP, 1979-1980.

. Number of Number of mark- · mark Portion of Pe rcent Skunk recapture recapture home range of home range number Sex Age locations 1 ocations Home range (ha) in campground(ha) in campground

84 Female Adult 2 5 19.4 7.0 36.1 8 Female Adult 5 8 15.5 11.2 72.3 1111 Female Adult 1 11 12.5 12.5 100.0 67 Female Adul t 2 10 16.9 12.5 73.9 68 Female Adult 6 10 43.2 17.3 40.0 75 Female Adult 5 10 34.5 13.5 39. 1 110 Female Adult 4 5 30.5 13.5 44.3 52 Female Adult 7 6 53.3 22.3 31 .8 . 6 ·- Female Adult 4 2 20.0 4.5 22.5 2 Male Adult 11 11 40.3 17.0 42.2 16 Male Adult 4 - 13.6 4.0 29.4 40 Male Adult 5 - 2.8 0.0 0.0 82 Male Adult 2 6 8.9 6.0 67 .4 73 Male Adult 4 7 14.3 8.0 5.6 54 Male Juvenile 3 4 27.5 10.5 38.5 22 Female Juvenile 2 1 3.5 2.0 57 .1 23 Female Juvenile 4 1 16.4 6.4 39 .0 � 1.0 50 Territoriality. Since skunks do not have mutually exclusive

home ranges, they are not territorial (Figures 7- 9) ; rather thay are

generalists and use a variety of habitats for denning and foraging .

The food items eaten by skunks tend to be distributed among several

habitats with the concentration in any given area varying with season.

Under these circumstances a large territory is required so the costs of defending a territory would outweigh the benefits gained. The

absence of territorility may confer the advantage of freedom to travel to areas where food is concentrated instead of being restricted to a

. limited area and possibly limited food supply .

Skunks may exhibit a form of conspecific behavior called mutual

avoidance, such as that observed in cats (Leyhausen, 1965 ), otters

(Lutra Lutra) (Erlinge, 1968), and weasels (Mustela sp.) (Lockie,

1966) •. Mutual avoidance prevents conflicts between conspecifics by having both animals ignore each other or retreat. Under certain con­ ' .ditions, skunks did not exhibit mutual avoidance and encounters led

to agonistic behavior stimulated by the invasion of personal space.

Personal space is defined as the area surrounding an animal, which

when invaded by an intruder, invokes an agonistic response (Alcock, 1975).

The distances at which skunks begin to exhibit mutual avo� dance

or agonistic behaviors varies. Females with young display agonistic

behavior more frequently, since they maintain larger distances between

themselves and other animals than do other skunks. In 6 observed

encounters, females began to display aggression when an intruding

skunk was still 5 m away. The most typical aggressive gestures were 51

0

�- (

... 9 ...... "- - � \ � -�- ..·- --::----. '-....

Figure 7. Home ranges of 3 male striped skunks in the Cades Cove campground study area, GSMNP, 1979-1980. 52

0

�.. c__· �- --. -� ' \ �-- -- .·- ... ---·-. '·

Figure 8. Home ranges of 3 female striped skunks in the Cades cove _ campground study area, GSMNP , 1979-1980 . · 53

0

..

·, ______-...... J •

-···-- . --� · ...... , .

ampgroun,d ove c · in the Cades C of 3 juveniles 9. Home ranges 980 . Figure MNP, 1 979-1 study area, GS 54

foot stamping and tai l raising, but 1 encounter involved foot stamping and a false charge. Although skunks normally allow other skunks to pass very close if both parties ignore each other , the more vigorous response of the female to protect her young is a common response in many speci es.

Skunks that are feeding exhi bi t aggressi on if approached directly by another skunk. Of the 24 confli cts observed during this study , 17 were apparently associ ated wi th food , 6 were between females wi th young and other skunks , and 1 was of unknown cause. Of the conflicts asso-

·ciated wi th food, 13 involved foot-stamping only , 3 involved foot­ stamping and the pushing of dirt under the body wi th the forelegs , and only 1 involved spraying. The confli cts over food were observed in late July when the skunk densi ty in the campground was high or in the fall when food was scarce.

Denni ng

Natural ·dens. Thirty-three natural dens were located duri ng the course of thi s study; 2 were natal dens during the spring. The natural dens were classi fi ed as above-ground summer dens, above-ground fall dens , below-ground summer dens , and below-ground fall dens (Table

15). Of the 13 above-ground summer dens, 6 were in logs , stumps, or dead trees, 4 were in rock caviti es, 2 were at the base of live trees ,

and 1 was a burrow. Of the 4 above-ground fall dens, 3 were in logs ,

stumps or dead trees , and 1 was at the base of a live tree. Of the

10 below-ground summer dens , 1 was in a stump , 1 was in a rock cavity,

5 were at the base of live trees , and 3 were burrows. Of the 6 TABLE 15. Classification of Summer and Fal l Dens Used by Skunks in the Cades Cove Campground Study Area , GSMNP, 1979-1 980 .

Sunmer ·_ Fall Above Below Above Below Den type ground ground ground ground Total

Logs , stumps , base of a dead tree 6 1 3 3 13

Rocks 4 1 0 0 5

Base of a live tree 2 5 1 1 9

Burrows 1 3 0 2 6

Total 13 10 4 6 33

U'l U'l 56

bel ow-ground fa ll dens, 3 were in logs , stumps or dead trees, 1 was at

the base of a live tree , 1 was in a woodchuck bu rrow and 1 wa s in a

hol e in the ground. Dens . in logs , stumps , and dead trees comprised by

fa r the greatest majori ty of the dens; 13 of the 33 dens were in logs ,

stumps , and dead trees. The high percentage of dens in thi s category

does riot necessarily refl ect a preference for this den type but may

simply be due to the relative abundance of this type of denning site .

Because of small sampl e sizes , the measurements from al l the

dens were pool ed in order to calculate the means and ranges of the

.den and phys ical parameters . No significant di fferences we re found.

The number of entrances per den ranged from l-3{X=l .4), but 22 dens

had only 1 entrance. Verts {1 967), Selko (1938), and Al len and Shapton

(1942 ) reported an average of 1.3, 1.5, and 2.2 en trances per den ,

respectively {Tabl e 16).

The results of the factor analys is indicate that the dens

appear to fal.l into 2 cl usters corresponding to the seasons surrmer and

fall (Fi gures 10 and 11), since the factor scores of the dens we re

clumped in seasonal groups described by the first 3 fa ctors .

Sl ope , el evation , and depth to bedrock were the variables most strongly

rel ated to factor 1, wh ich appeared to be a genera l elevation factor.

Factor 2, wh ich wa s a general factor describing position of the dens

in re lation to water, was highly correlated wi th distance from water,

soi l drainage, soil type , and depth to bedrock. On ly the single

variable wi dth of entrance 1 correl ated strongly wi th factor 3. These

3 factors accounted for 16.8%, 19.7%, and 10.7% respectively, of the

total variance among the dens. TABLE 16. Average Measurements of 33 Natural Skunk Dens in the Cades Cove Campground Study Area , GSMNP.

Sample Distance to Distance to Height of Width of

Season size water(m) __c ampgr_o_u_n_c!(ml_ -- -�ll_t r_ance�IDL____ _E!n tr�nce(cm)

Sull11ler dens 21 Means 157.0 96.0 15.5 18.0

Range 1. 3-500. 0 0.0-625.0 6.3-35.6 7.5-59.0

Fal l dens 12 Means 152.0 115.0 14.3 17.7

Range 2.0-450.0 0.0-625.0 6.3-23.7 7.5-26.3

Al l dens 33 Means 154.4 105.5 14.9 17.8

Range 1.3-500.0 0.0-625.0 6.3-35.6 7.5-59.0

U1 ...... , FACTOR2 I s I z.o • I s I I I loS • I I I F I s loO • I F I s I I o.s • s s I F I FS F I s s I s ! F o.o • s I s s s I s I s I s -o.s • f I s I I I -1.0 • I I s I I -l.S • f I f I I I -z.o • I I f s I I -z .s • I I

--- ·------+------·------· ------·------·------·------+------·------· ------·------· ------· ------·--- -1.0 -o.T -o. � -o. 1 o.z c.s o.R 1.1 1.� 1.1 2.0 z.� ?..� 2.�

FAC Tf"llll

U'1 Figure 10� A plot of factor 2 , a general drai nage factor vs. factor 1, a general elevation factor co with reference to season. r.• FACTOR) I ,.. I 2.5 + I I s I I z.o + s I I I I 1.5 + I I I I F s 1.0 • s s I s I I s I s Oo5 + I s s I s I I s s o.o + I s s I F s .- . I I f F s -o.5 + I f F F F s I F I I -1.0 • I I I I s . -1.5 + F I I s I I -z.o + s I I --· .. ·-----·-----·------.------+-·-----·------·------+------·------·------·------+---

-1 .0 - 0. 1 .-o.. -o 0 1 0 0 2 ( 0 5 0 0 a 1 .1 1 0 (o 1. 7 2 0 0 7. 0 J 2 0 6 2 0 9

FACTCA 1 U'l Figure 11. A plot of factor 3, representing width of entrance 1 vs. factor 1, a general elevation ID factor with reference to season . 60

Summer dens t�nded to have higher scores on factors 1 and 3 but

there were no noticeabl e differences on factor 2. Summer dens tended to

have wider den entrances and be located on higher, steeper sl opes whi le

fal l dens tended to have narrower entrances and lie on gentl er sl opes

with deeper soi l cover.

To determine if these trends in the factor scores refl ected

statistical ly significant differences between summer and fal l dens, a

canonical correlation analysis and a discrimi nant analysis were per­

formed on the 33 dens. The resul ts of the canonical correl ation

. analysis, showed a difference between summer and fal l dens, but thi s

difference was not statistical ly significant (Fi gure 12). The signifi­

cance test from the discriminant analys is also did not show a si gnifi­

cant di fference between the summer and fal l dens. The lack of signifi ­

cance may be due to an insufficient number of dens in the sample.

In the canonical correlation , the variabl es measured accounted

for 42. S% · of the variance in season . Width, soi l drainage, and soi l

type were most' strongly correl ated wi th the canonical variabl e gi ving

the best separation of the summer and fal l dens. These results are in

agreement with the resul ts of the factor analysis.

In summary , summer dens tended to be located on well-drai ned

soil , have wi der entrances and have a shal l ower depth to bedrock. In

addition , summer dens tended to be located on steeper slopes and be

above ground. Fal l dens tended to be located on moderately to well

drained soil, wi th narrower entrance(s), and deeper soi ls. In additi on,

fall dens tende� to be located bel ow ground on gentl er slopes.

Skunks in the Cades Cove campground exhibit denning behavior GROUP MEAN SYMBOL SYMBOL COORDINATE S FOR CASES FOR MEA�

SUMMER 0.57 o.o .s 1 FALL -1.20 o.o F 2

F F S � S S S· F F S F S F F SF S F SF ! S S SSSS S S S S ••••••••••• •••••••••••••• •••2 ...... •••••••••••• •• ••••••••••••••••••••• •••••• ••••• •••• ••••••••• ••••• -2 .1 -2o 1 -1.5 -.90 -. 30 o30 o90 1o5 2o 1 2o7 3o3 -2.4 -1.8 -1.2 -.60 OoO o60 1o2 1o8 2o4 3o0

Figure 12. Canonical correl ation histogram of fal l and summer skunk dens in the Cades Cove campground study area, GSMNP . 0\ ..... 62 simi lar to skunks studied by other investigators (Houseknecht, 1971 ;

Houseknecht and Tester , 1978 ; Ve rts , 1967). During the spring and summer, skunks tend to sleep above ground in logs , stumps , and root systems of dead and live trees. Male skunks begin to use above ground dens earl ier than femal es. Females use bel ow ground dens for parturi­ tion and lactati on (Verts , 1967) and swi tch to above ground dens when the young are old enough to accompany the female. Storm (1972 ) found skunks to be very opportunistic, sleeping in cropl ands and in manmade structures during the summer in Ill inois.

The dens in the present study were not used excl usively by 1 skunk. During the summer , dens (n=5) were used for 2-3 days then deserted for varying lengths of time. Some of these dens were subse­ quently used by other skun ks. The sequential use of dens is a function of several factors : species using the area , abundance of suitable den sites, pattern of den use, and stabi lity and permanence of the dens

(Houseknecht and Tester, 1978). Sequential use of a den occurred in

10 of the 33 dens, but may have occurred more frequently than recorded due to the difficulties in monitoring al l the dens on a regul ar schedule.

Opossums and striped skunks are both generalists and tend to forage and den in similar areas. Sequential use of dens by these 2 species is not unusual. Simultaneou s sharing of dens by 2 species was recorded by Shirer and Fitch (1970). There was no evidence. of shared dens in this study.

Skun ks begin to swi tch from above ground dens to bel ow ground dens in October (Verts, 1967). Temperature appears to be an important 63

factor in initiating preparation of winter dens (Alesiuk and Stewart ,

1962). According to Houseknecht (1969), by November under ground dens

were being used by skunks in Central Mi nnesota , where temperatures were

be l ow freezing. The skunks in Cades Cove continued to use above ground

dens in November, but the temperature in Cades Cove sel dom drops bel ow

freezing for extended pe riods . In another study the pe rcentage of

bel ow ground dens used each month wa s inversely re lated to tempe rature

(Houseknecht and Tester, 1978). Below ground dens are better con­

structed for col d weather than above ground dens . The small entrance (s)

. characteristic of bel ow ground dens reduce the amount of heat lost

through the entrance(s) . In addition, skunks pl ug their entrance(s) in

the wi nter to reta in heat (Verts , 1967) , the smal l entrance(s) may make

constructi on of the pl ug(s) eas ier. Bel ow ground dens create a rel a­

tively warm microcl imate, even when the ground is frozen and the den

is occupied by 1 skunk.

Culvert dens . The drai nage cul ve rts in the campground were a

common denning site for skunks. Of the 31 cul ve rts in the campground ,

20 were classified as above ground , manmade summer dens in 1979 and 25

were cl assi fied as above ground , manmade summer dens in 1980 . Of the

11 cul verts not used in 1979, 9 were fi lled with wa ter or debri s and

2 we re occupied by raccoons. Of the 6 cul verts not used by skunks in

1980, 4 we re fi lled wi th debris and 2 were occupied by raccoons

(Figures 13 and 14). The culverts were 96 em in diameter and extended

under the roads in the campground. The culverts were used as dens

from July through October.

Communal denning was observed in the culverts on 12 occasions. 0 1 Sku� k per Cul vert 9 � 2 Skunks per Cul vert � 3 Skunks per Cul vert Ill 8 s:: � 4 Skunks per Cul vert QJ c � 1 Raccoon per Culvert Ill 7 ItS "0 QJ Ill 6 � en +.l 5 s.. QJ > ...- ::s 4 u � 0 3 s.. QJ ..0 E ::s 2 z:

...... ·,•.; . 0 I I I / 1 •···. ·. 1 I I «1··.·· 1 1 1/ 1 :.·.-.1 I I·.<·: I I I/ I· ..I I I I/ J . -.1 I I r fl. ..I 7/6/79 8/6/79 8/ 8/79 8/ 16/79 8/27/79 8/29/79 9/9/79

Fi gure 13. The number of road culverts used as dens in 1979 in Cades Cove campground . GSMNP.

0'1 .j::lo. 8 1- 0 1 Skunk per Cul vert · 2 Skunks per Cul vert VI r2l . s:: Q) 7 1- 3 Skunks per Cul vert c � VI § 4 Skunks per Cul vert IU 6 � "C Q) VI ::::> 5 VI +.J s.. 4 Q) > .-- :::J u 3 1+- 0 s.. 2 Q) .0 E :::J z: 1

I I I I I l==f I I 10 1 I I I Vl I I I I I l 0 ttl · I I I I 1 1 I 0 6/12/80 7/7/80 7/25/80 8/6/80 8/22/80 10/30/80 6/30/80 7/17/80 7/30/80 8/12/80 8/30/80

Figure 14. The number of road cul verts used as dens in 1980 in Cades Cove campground , GSMNP.

0\ U1 66

The ·average number of animals observed denning communal ly at any 1 time was 3, but up to 4 animal s to a cul vert have been observed. The communal dennings observed in July probably consisted of fami ly units and after July probably consisted of sibling groups. In addition to communal denning, skunks were observed sharing culverts wi th other sk�nk� on 3 occasions (one �kunk at each end of the culvert) .

Den availability. It was estimated that there were 12.0 poten­ tial dens pe r hectare not including potential dens in manmade structures or the cul vert dens. The number of availabl e dens per hectare sur­ passed the density of skunks in the area , thus den sites do not appear to be a limiting factor. The discriminant analysis performed on the potential dens and the confirmed dens did not reveal a significance difference between the 2 groups . CHAPTER V

RESEARCH-MANAGEMENT IMPLICATIONS

The density of skunks in the Cades Cove campground is higher

than the density of skunks in agricultura l or natural areas, but the

campground skunks do not appear to suffer from crowding or ma lnutri­

tion. In addition , park records indicate that there have been no

known cases of rabies in the park (Personal Communication , Stu

Coleman); therefore the present skunk population does not appear to

·be a health hazard to campers. Actually many campers seem to enjoy

observing skunks foraging in the campground ; this may be the only

wildl ife some of them see duri ng thei r stay in the park.

By monitoring the skunk population for changes in population

s ize and for the presence of epizooti cs, health problems between

skunks and �eople may be prevented . If a problem, such as over popu­ lation or an epizootic did arise , several different procedures could

be used to decrease the skunk population. Two obvious procedures

. avail able are : decreasing the number and availability of dens and/or

decreasing the amount of available food in the campground. A reduction

in the avai lability of food or garbage has been used wi th variabl e

success on ••panhandl ing" wild animal s in parks throughout the world

{Craighhead , 1979).

One way to reduce the available food in Cades Cove campground

would be to require al l garbage to be kept off the ground so skunks

could not get into it. This method would require the cooperation of

67 68 the campers. Sine� the road culverts are used extensively as dens,

limiting access to them might decrease the number of skunks denning in

the campground. However, the ada ptability of the skunks for using

alternate den sites is unknown. Limiting access to the cul verts mi ght

reduce skunk-dog interactions which often end with the skunk spraying.

Another possible procedure might invol ve relocating skunks to

areas away from campground and picnic areas , but removal of the skunks

from the campground would be a temporary solution to the problem. The

ecological vacuum created by the removal of the skunks could be fi lled

.by skunks from the surrounding area or by raccoons, whi ch mi ght replace

skunks as a problem animal in the campground. At this time information

is not avai lable to predict the response of the animal community to

the removal of skunks from a campground area . Further studies

should be done before such a removal is attempted.

If this study is continued, several changes should be made in

the methodology in order to collect the data efficiently. The study

area should be · increased to a size that wi ll include the entire home

range of the skunks utilizing the campground. Trapping should be

conducted all year in order to determine seasonal changes in activity

and home range size. Radio-col lars should be used to monitor the

skunks throughout the year. A questionnaire could be developed and

given to the campers to obtain their reaction to skunks. A simi lar

study should � conducted in an area of the park that does not include

a campground.· CHAPTER VI

SUMMARY AND CONCLUSION

1. The sex ratio of adult male skunks to adult female skunks

differed in favor of the females from the expected 1:1 ratio. Previous

" " studies reported sex ratios that favored males rather than females.

The differences between the sex ratios calculated for this study and

other studies and for the deviation from the expected 1:1 ratio may be

due to sampling biases caused by differences in behavior between the

·sexes, differential mortal ity between the sexes , and unknown causes.

2. Adult male and female skunks differed in external charac-

teristics , males tending to have longer bodies and shorter tai ls than

femal es.

· 3. Skunks in Cades Cove campground we ighed signifi cantly more

than skunks in agricultural and natural areas. The difference is

probably re lated to the availabl e quantity and qual ity of food in . the campground and the longer foraging season in Cades Cove . Skunks

. in the Cades Cove campground area do not have to expend as much energy

to obtain food since garbage is a readily available food source wi th

a higher caloric content than the food obtained by skunks in agricul­

tural and natural areas.

4. The coat patterns of striped skunks in the Cades Cove camp­

ground study area we re placed in 1 of 5 categories, based on the

amount of white and the coat pattern of the individual . The categories

were broad striped, striped, short striped, black star, and atypically

69 70

striped. The coat of the striped skunk is conspicuous in nature due to the bl ack and whi te col or pattern ; this fu nctions to warn

predators that skunks have an efficient means of defense.

5. The dens1�y of skunks in Cades Cove campground is hi gher

than reported for natural or agricul tural areas. The study area contained a res ident population of 1 skunk per 5 hectares. The presence of a campground and picnic area infl uenced the Cades Cove skunk population by increasing the availability of den sites and food .

6. The da ily activi ty patte rns of striped skunks using the camp­ ground changed wi th the seasons. In Apri l and May , few skunks we re observed in the campground. The numbe r of skunks observed in the campground during the night increased in July and August in response to increased availabi lity of food and in the number of juveniles ma ture enough to accompany their mothers on forag ing expediti ons . In July and August skunks were primari ly nocturnal wi th a bimodal distribution of activity through the night. Some skunks began to forage around

1800 hrs. , when campers were preparing supper. This is earl ier than other investigators have reported skunks to be acti ve when food is avai labl e. By October, the number of skunks observed in the campground

has decreased.

7. Stri ped skunks in the Cades Cove campground did not pan­

handle food from the campers , rather they obtai ned their food in the

campground by foraging in the campsites. The only agoni sti c inter­

actions observed between skun ks and campers occurred when campers

tried to force a skunk out of a cul vert or threw objects at a skunk.

Skunks tol erated close proximity to peopl e wi thout displayi ng agonistic 71

behavior even when harrassed by people. Possible methods for reducing

human-skunk interaction are the initiation of a program to remove

garbage and discarded food from the campground and limiting access to

the campground culverts.

8. Since skunks do not have mutually excl usive home ranges ,

they are not territorial . Skunks did exhi bit a form of behavior cal led

mutual avoidance which reduces aggressive confl icts between skunks.

The distance at which skunks begin to exhibit mutual avoidance or

agonistic behavior varies. Females wi th young display agonistic

-behavior more frequently since they try to maintai n a large distance

between their family groups and other skunks. Skunks that are feeding

also exhibit aggression rather than mutual avoidance when approached

directly ..

·g. Five aspects of the denning activities of striped skunks

were brought out in thi s study: a. different den types are used in

the summer and fall, b. drai nage cul verts were used as dens from July

through October , c. drainage culvert dens were used communal ly, d.

natural dens and culvert dens were used sequential ly by skunks , and

e. there are more dens than skunks per hectare .

Summer dens were used from May through October and we re above

ground dens located on well drained shal low soils wi th steep slopes.

Fal l dens were used from September through May and were bel ow ground

dens located on moderate to well drained deep soi ls wi th gentle slopes.

The drai nage cul verts were used as dens from July through

October. Of the 31 culverts in the campground, 20 were used as dens

in 1979 and 25 were used as dens in 1980. Communal denning was 72

observed on 12 occasions in the cul verts.

It was estimated that there were 12.0 potential den sites per

hectare , not incl uding possible dens in manmade structures or in cul ­

verts. The number of dens per hectare is significantly higher than

the number of skunks per hectare , therefore den sites are not a

limiting factor to the skunk population. 10. The density of· skunks in Cades Cove campground is hi gher than

the density in agricul tural or natural areas, but the campground

skunks do not appear to suffer from crowding or mal nutrition . By

.monitoring the skunk population for changes in population size and

for the presence of epizootics , health probl ems between skunks and

peopl e may be prevented . If a problem did arise, several procedures

are available: decrease the number and availability of dens , decrease

the amount of available food in the campground , or relocate the skunks

to areas away from campgrounds and picnic areas . LITERATURE CITED LITERATURE CITED

Al cock , J. 1975� Animal behavior, an evol utionary approach. Sinauer Associates , Inc. , Sunderl and , Mass. 501 pp.

Alesiuk, M. , and A. P. Stewart. 1962. Food inta ke, wei ght changes and activity of confined striped skunks (Mephitis mephitis) in wi nter. Am. Midl . Nat. 98:331 -342 .

Al len, D. L. 1939. Wi nter habits of Mi chigan skunks . J. of Wi ldl . Manage . 3:212-228.

___, and W. W. Shapton . 1942. -An ecol ogical study of wi nter dens wi th sp ecial reference to the eastern skunk. Ecology. 23:59-68 .

Arnason and Baniuk. 1969. Popan-2, a data ma intenance and analysis system for mark-recapture . Charl es Babbage Research Center Press , Ma ni toba , Canada . 239 pp.

Bachrach , M. 1930. Fur, a practical treatise. Prentice-Hal l, Inc. New York. 664 pp.

Bailey, T. N. 1971 . Bi ology of striped skunks on a southwestern Lake Erie marsh. Am. Midl . Nat. 85 :196-207 .

Bennitt, R. , and W. 0. Nagel . 1937. A survey of the resident game and furbearers in Mi ssouri . Uni v. Mi ssouri Studi es . 12(12): 1-25 .

Cott, H. M. 1940. Adaptive col oration in animal s. Oxford Universi ty Press, Oxford . 508 pp.

Craighead, F. C. 1979. Track of the gri zzly. Sierra Cl ub Books , San Francisco . 261 pp.

Curtis, W. 1978. Meet the other polecat. Natl . Wi ldl . 16: 18-22.

Davi s, W. B. 1951 . Texas skunks. Tex . Game and Fish . 9(4):18-21 :31.

Dean, F. G. 1965. Wi nter and spri ng habits and density of Ma ine skunks. J. Mammal . 46:673-675 .

Di xon , W. J., and M. Brown (eds.). 1975 . BMDP , biomedical computer programs . Uni versity of Cal ifornia Press , Berkel ey". 792 pp.

El der , J. A. 1956. Soi l survey , Blount County. U. S. Dept. of Agr ic. , So il Conservation Service. 119 pp .

Erl inge, s. 1968 . Territorial ity of the otter (Lutra lutra}. Oikos. 19:81-91 .

74 75

Hal l� E. R. 1981 . The mammal s of North America . John Wiley and Sons , N.Y. 1181 pp.

Hamilton , W. J. 1936 . Seasonal food of skunks in New York. J. Mammal . 17:240-246.

___• 1937 . Wi nter acti vity of the skunks . Ecology . 18:326-327 .

1943 . The mammals of ea stern U. S. Vol . II. Comstoc k --� . Pu bl ishi ng Co ., Inc., Ithaca, N. Y. 432 pp.

Huheey , J. E. , and A. Stupka . 1967 . Amphi bians and reptiles of Great Smoky Mo untains Nati onal Pa rk . Uni vers ity of Tennessee Press , Knoxville. 112 pp.

Hattwick, M. A. , F. H. Hockberg , P. J. Landrigan , and M. B. Gregg� 1972. Skunk-associated human rabies. JAMA. 222: 44-47 .

. Hediger, H. 1964. Wi ld animal s in captivity. Dover Publ ications , Inc. , New York. 207 pp .

Helwi g, J. T. , and K. A. Council. 1979. SAS User•s Guide, statistical analysis system. 491 pp .

Houseknecht, C. R. 1969. Denni ng habi ts of the striped skunk and the exposure to potential disease. Proc . Ann. Conf. Bull . Wi 1dl . · Di sease Assoc. 5:302-306.

Houseknecht, C. R. 1971 . Movements , acti vity patterns and denn ing habits of striped skunks (Mephitis mephitis) and the exposure to potential fo r disease . Unpubl . Ph. D. Di ssertation. Un iversity of Mi nnesota , Minneapol is. 45 pp.

Hou se knecht, C. ·R. , and J. R. Tester . 1978. Denning habits of the striped skun k (Mephitis mephi tis ). Am. Midl . Nat. 100:424-438.

Kel ke r, G. H. 1937. Insect food of skunks . J. Mammal . 18:164-1 70 .

Lehner, P. N. 1979. Handbook of ethol ogical methods . Ga rl and Press , New York . 401 pp .

Leyhausen , P. 1965. The communal organization of sol i tary mammals. Symp. Zool . Soc . London . 14:249-263.

Linzey, A. V. , and D. W. Linzey. 1971 . Mammals of Great Smo ky Mounta ins Nati onal Park. The University of Tennessee , Knoxvi lle. 111 pp.

Lockie, J. D. 1966. Terri tory in smal l carnivores. Symp . Zool . Soc . London . 18: 143-163 .

. Mohr, C. 0. 1943. Illinois furbearers di stribution and income . Il linois Nat. Hist. Survey Bul l. 27 (7) :505-537. 76

Mutch, G. R. , and M. Al esiuk. 1977. Ecological aspects of winter

dormacy in the stri ped skunk {Mephitis mephitis). Can. J. Zool .· 55 :607-615.

Ol liver, J. G. , and J. Cl endinning . 1979. Principles of surveyi ng, Vol . I. Va n Nostrand Reinhold Co ., New York. 195 pp.

Pianka , E. R. 1978. Evolutionary ecology. Harper and Row Publ i shers , New York. 328 pp.

Sc hwartz , C. W. , and E. R. Schwartz . 1959. The Mammals of Missouri . Univ. of Mi ssouri Press and Missouri Conserv . Comm. 80 pp.

Sel ko , l. F. 1938. Notes on the den ecology of striped skunks in Iowa Am. Midl . Nat. 20:455-563.

Shirer, H. W. , and H. S. Fitch . 1970. Comparison from ra dio-tracking of movements and denning habi ts of raccoon , striped skunks , and opossums 1n northeastern Kansas. J. Mammal . 51 :463-491.

Stains , H. J .• and D. Stuckey. 1960 . Brachial-antebrachial stri pes on a striped skunk. J. Mammal . 41 :139.

Storm, G. l. 1972. Daytime retreats and movements of skunks on farm­ land in Illinois. J. Wi ldl . Manage. 36: 31 -45.

Sunquist, L. 1974. Wi nter activity of striped skunks (Mephitis mephitis) in East-Central Minnesota . Am. Midl . Nat. 83:434-445.

Tanner, J. 1978. Guide to the study of animal populations . The University of Tennessee Press, Knoxville. 172 pp.

Taylor, C. 1979. The Distribution and Density of the Woodchuck {Marmota .monax ) in Cades Cove , GSMNP. Unpubl . M.S. Thesis, The Uni vers1ty of Tennessee , Knoxvi lle. 218 pp.

Verts, B. J. 1967. The biol ogy of the striped skunk. Universi ty of Illinoi s Press, Urbana. 204 pp. VITA

Deborah M. Goldsmi th was born on December 30 , 1956, in Ea st

Lansing, Michigan. She attended Robert E. Peary Hi gh School in

Rockville, Maryl and , graduating in 1974. In 1974 she entered Knox

College in Galesburg , Il l inois and received a Bachelor of Arts degree in 1978. She entered The University of Tennessee , Knoxvi lle, in

1978 and accepted a teaching assistantship from the Graduate Program of Ecol ogy. She received her Master's degree in ecology in August

1981 .

77