Southeastern Naturalist R.W. Perry, D.C. Rudolph, and R.E. Thill 2018 Vol. 17, No. 2 298 2018 SOUTHEASTERN NATURALIST 17(2):298–308 Capture-site Characteristics for Eastern Spotted Skunks in Mature Forests during Summer Roger W. Perry1,*, D. Craig Rudolph2, and Ronald E. Thill2 Abstract - Spilogale putorius (Eastern Spotted Skunk) is an increasingly rare species undergoing population declines throughout many portions of its range. We incidentally captured Eastern Spotted Skunks in snake traps during a study examining effects of woodland restoration on herpetofauna in the Ouachita Mountains of Arkansas. We used extensive habitat data collected at each trap site to determine potential characteristics of sites where Eastern Spotted Skunks were more likely to occur during summer. We recorded 18 Eastern Spotted Skunk captures in 10 of our 36 drift-fence traps. Capture rates of Eastern Spotted Skunks were 6 times greater and occupancy rates were 9 times greater in unmanaged, mature forests with a well-developed midstory than in frequently burned woodlands that lacked a midstory. Higher-occupancy rates were associated with greater total cover, greater cover of woodyunderstory vegetation, and sparse forb cover. Our data support those of previous studies that suggest Eastern Spotted Skunks occur in areas with dense cover, which may include mature forests with well-developed midstories. Introduction Spilogale putorius (L.) (Eastern Spotted Skunk) has undergone widespread declines in abundance over several decades and is a species of concern in many states across its range (Gompper and Hackett 2005). Despite its increasing rarity, studies of Eastern Spotted Skunk habitat use are limited and its habitat associations remain unclear. For example, Eastern Spotted Skunks have been associated with prairies (Crabb 1948), Quercus spp. (oak)–Carya spp. (hickory) forests (forest-age unknown; McCullough and Fritzell 1984), or young, regenerating forest and mature hardwood forests (Lesmeister et al. 2009). In the Appalachian Mountains, they were found in dense thickets of Rhododendron (rhododendron) and Kalmia latifolia L. (Mountain Laurel) (Diggins et al. 2015, Reed and Kennedy 2000, Wilson et al. 2016), and in young to mid-successional (less than 50 years old) forest (Thorne et al. 2017). To achieve better understanding of this species and potential reasons for its decline, more information is needed on the species’ habitat associations and how land management may affect presence. Land managers throughout North America are implementing woodland and savanna restoration programs to recreate the open forest conditions that historically existed in many regions prior to European settlement (e.g., Spetich et al. 2011). In the Ouachita National Forest (ONF) of Arkansas and Oklahoma, ~142,000 ha have been targeted for restoration of Pinus (pine) woodlands. To restore this community, 1US Forest Service, Southern Research Station, Hot Springs, AR 71902. 2US Forest Service, Southern Research Station, Nacogdoches, TX 75965. *Corresponding author - rperry03@fs.fed.us. Manuscript Editor: Marcus Lashley Southeastern Naturalist 299 R.W. Perry, D.C. Rudolph, and R.E. Thill 2018 Vol. 17, No. 2 the overstories of mature (generally >50 years old) forests are thinned, midstories are removed or reduced, and stands are subjected to prescribed burns at 3–5-year intervals. These efforts result in open forest stands with an herbaceous understory and little or no midstory (Fig. 1). Lesmeister et al. (2013) examined habitat use by Eastern Spotted Skunks in restored woodlands of the ONF and found that the species was negatively associated with restored woodlands. In that study, Eastern Spotted Skunks were associated with young, cluttered, regenerating forests and with mature hardwood forests. However, during a study comparing herpetofauna communities in 2 types of mature forest (restored pine woodlands and unmanaged pine-dominated forests >60 years old) in the ONF (Perry et al. 2009), we incidentally captured Eastern Spotted Skunks in many of our traps. These captures presented an opportunity to delineate structural characteristics of mature forests that may affect the presence of Eastern Spotted Skunks. We analyzed capture data for Eastern Spotted Skunks along with several vegetation measures collected at each trap site with the goal of determining attributes of mature forests that may affect presence of this rare species. Materials and Methods We conducted our study on the Poteau–Cold Springs Ranger District of the ONF (Scott County, AR) in the Ouachita Mountains. The Ouachita Mountains extend from central Arkansas into eastern Oklahoma and consist of a series of east–west-oriented mountains, with elevations varying from 100 m to 800 m. The predominant forest type in the area is mixed Pinus echinata Mill. (Shortleaf Pine)–hardwood forests, but hardwood and riparian forests occur throughout the Figure 1. Mature, unmanaged forest (left) and restored Shortleaf Pine woodland (right) where Eastern Spotted Skunks were captured in the Ouachita Mountains of Arkansas, 1999−2001. Photographs © Roger W. Perry (left) and James Guldin (right). Southeastern Naturalist R.W. Perry, D.C. Rudolph, and R.E. Thill 2018 Vol. 17, No. 2 300 Figure 2. A drift-fence trap used for capturing herpetofauna that also captured Eastern Spotted Skunks in the Ouachita Mountains of Arkansas, 1999−2001, based on a modified design presented by Burgdorf et al. (2005). Photograph © Josh Pierce. region. Hardwoods in these forests are diverse and include oaks, hickories, and Acer rubrum L. (Red Maple). We sampled 2 types of forest stands: (1) mature unmanaged (>60 years old), second-growth pine–hardwood forest; and (2) restored Shortleaf Pine woodlands (Fig 1). Restoration of pine woodlands consisted of thinning forest overstories to retain 13.7–16.1 m2/ha of pine basal area (BA) and 1.4–1.6 m2/ha of hardwood BA; all or most midstory trees were removed. In woodlands, cutting treatments were followed by prescribed burns conducted at 2–5-year intervals, and unmanaged forests were not burned. For additional details on the restoration process and our sampling design, see Perry et al. (2009). We surveyed 12 forest stands: 9 restored pine woodlands and 3 unmanaged mature stands. During the 3 years of our study (1999–2001), restored woodland stands were burned on a 3-year rotation. Three of the 9 woodland stands were burned each year in March or April, and all burned stands were part of larger burning units (64.8–1335.5 ha). Thus, most woodland stands were contiguous with large areas of burned forest. Our overall trapping goal was to capture herpetofauna. However, in the course of sampling, we captured numerous small and medium-sized mammals. We used drift-fence traps designed to capture large snakes (Fig. 2). Each trap array consisted Southeastern Naturalist 301 R.W. Perry, D.C. Rudolph, and R.E. Thill 2018 Vol. 17, No. 2 of four 15-m linear fences arranged at 90o angles to one another and constructed of steel hardware cloth (3.2-mm mesh) with a 1.2 m x 1.2 m x 0.46 m (l x w x h) box trap in the center (Burgdorf et al. 2005). Our traps had 10-cm–diameter entrance funnels that allowed larger animals to enter; this entrance size differed from traps presented by Burgdorf et al. (2005), which had 5-cm–diameter entrance holes. We installed 3 traps in each stand. Trap were >150 m apart, >50 m from roads or stand edges, and >75 m from permanent or intermittent streams, ponds, and stream buffers. We checked traps weekly from early April until late September for 3 y (1999–2001); trapping effort was equal among all stands and years (24 weeks each year), and trapping occurred concurrently at all sites. Each box trap contained a water dispenser that maintained a continuous water source in each trap. We recorded all captured vertebrates and immediately released them >50 m from the trap (with the exception of skunks, see below). We did not mark captured mammals because our overall goal was sampling herpetofauna. We released captured Eastern Spotted Skunks at the site of capture by opening trap doors and placing a log in the trap opening to act as a ramp that allowed skunks to exit on their own. We followed the guidelines of the American Society of Mammalogists for the capture, handling, and care of mammals (Animal Care and Use Committee 1998). We measured vegetation in September and early October at 4 plots surrounding each trap (Table 1). Plots were located 7 m beyond the distal end of each drift fence. We measured percent canopy closure (Cover) at plot center with a spherical densiometer held at breast height, and overstory and midstory BA (conifer and hardwood combined) using a prism. We visually estimated (± 10%) downed-wood cover in 3 adjacent 2 m x 2 m subplots. In 3 nested 1 m x 1 m subplots, we visually estimated percent cover of grass, forbs, leaf litter, bare ground, and woody understory vegetation (≤1 m high). We measured litter depth in the center of each 1 m x 1 m subplot. We employed a 0.5 m x 0.5 m density board (Nudds 1977) to Table 1. Vegetation parameters used in models to determine effects of forest condition on Eastern Spotted Skunk presence in the Ouachita Mountains of Arkansas, 1999−2001. Dens1–Dens3: lower values = less distance that can be viewed at that height; thus, lower values indicate the presence of thicker vegetation. Parameter Description Overstory BA (m2/ha) of overstory trees (measured with prism) Midstory BA (m2/ha) of midstory trees (measured with prism) Cover Canopy cover (%) at breast height (measured with densitometer) Woody Percent cover of woody plants in the understory (visually estimated) Forbs Percent cover of forbs in the understory (visually estimated) Grass Percent cover of grasses in the understory (visually estimated) Leaf Percent cover of leaf litter on the forest floor (visually estim ated) Bare Percent cover of bare ground on the forest floor (visually estim ated) LitterD Depth (mm) of leaf litter on the forest floor (measured) Dwood Percent cover of down/dead wood on the forest floor (visually es timated) Den1 Index of vegetation density 0−0.5 m above the forest floor (dens ity board) Den2 Index of vegetation density 0.75−1.25 m above the forest floor ( density board) Den3 Index of vegetation density 1.75−2.25 m above the forest floor ( density board) Southeastern Naturalist R.W. Perry, D.C. Rudolph, and R.E. Thill 2018 Vol. 17, No. 2 302 estimate horizontal vegetation density by measuring the distance at which 50% of the density board was obscured by vegetation at 3 heights: ground level–0.5 m high (Den1), 0.75–1.25 m above the ground (Den2), and 1.75–2.25 m above the ground (Den3). With this measure, denser vegetation resulted in lower numbers and sparser vegetation resulted in higher numbers. We modeled occupancy (Ψ) and detectability (P) of Eastern Spotted Skunks in each trap, each year, using the program PRESENCE (MacKenzie et al. 2006) to determine vegetation parameters (Table 1) at each trap site that affected presence/ absence of skunks. To increase model stability, we removed 5 highly correlated variables (≥0.60) prior to analysis (Overstory, Grass, LitterD, Den3, and Bare). We compared 23 a priori models that included effects of vegetative parameters on occupancy (Table 2). We included effects of detectability in our model set; models contained either the effects of woodlands/unrestored sites on detectability [P(Woodland)] or similar detectability between the 2 forest types [P(.)]. We compared models and selected the best model based on values of Akaike’s information criterion (AIC; Burnham and Anderson 2002). Vegetation parameters were highly correlated. Therefore, we conducted principal components analysis (PCA), using all vegetative parameters, to Table 2. Model parameters included in models of occupancy (Ψ) and detectability (P) of Eastern Spotted Skunks in restored woodlands and unrestored mature forests in the Ouachita Mountains of Arkansas, 1999–2001, including values of AIC, difference from the best model in each set (ΔAIC), and weight of each model among all models (ωi). An asterisk (*) indicates models that failed to converge or produced highly improbable parameter estimates and standard errors (Dail and Madsen 2011). Model AIC ΔAIC ωi Ψ(Cover, Woody, Forbs) P(.) 91.02 0.00 0.65 Ψ(Cover, Woody, Forbs) P(Woodland) 93.00 1.98 0.24 Ψ(Cover, Woody) P(Woodland) 96.88 5.86 0.03 Ψ(Cover) P(.) 97.06 6.04 0.03 Ψ(Cover) P(Woodland) 98.62 7.60 0.01 Ψ(Woody) P(Woodland) 98.63 7.61 0.01 Ψ(Den2) P(Woodland) 99.36 8.34 0.01 Ψ(.) P(Woodland) 102.13 11.11 0.00 Ψ(Midstory) P(.) 107.18 16.16 0.00 Ψ(.) P(.) 110.81 19.79 0.00 Ψ(Woody) P(.) 112.81 21.79 0.00 *Ψ(Forbs) P(.) *Ψ(Cover, Forbs) P(.) *Ψ(Forbs) P(Woodland) *Ψ(Cover, Forbs) P(Woodland) *Ψ(Cover, Woody, Den2) P(Woodland) *Ψ(Cover, Woody) P(.) *Ψ(Cover, Woody, Den1) P(Woodland) *Ψ(Midstory, Cover) P(Woodland) *Ψ(Woody, Den1, Den2) P(.) *Ψ(Midstory) P(Woodland) *Ψ(Burn, unburn) P(Woodland) *Ψ(Midstory) P(.) Southeastern Naturalist 303 R.W. Perry, D.C. Rudolph, and R.E. Thill 2018 Vol. 17, No. 2 characterize vegetation at trap sites and to differentiate vegetative characteristics between woodlands and unrestored sites. We also used this analysis to better characterize sites where we captured Eastern Spotted Skunks. Results We recorded 18 Eastern Spotted Skunk captures over 3 growing seasons (18,144 trap nights) in 6 of the 12 forest stands and in 10 of the 36 traps. Although there were 3 times as many traps in woodlands (n = 27) than in unmanaged stands (n = 9), we captured Eastern Spotted Skunks twice as often in unmanaged stands (12 captured in unmanaged stands versus 6 captured in woodlands). Capture rate (mean number of captures per week) of Eastern Spotted Skunks was 6 times greater in unmanaged stands (0.019 ± 0.005 SE) than in woodlands (0.003 ± 0.001). The number of Eastern Spotted Skunks captured in woodlands was similar (2 each) among stands sampled the first, second, and third year after burning. Of the 23 a priori models, 11 converged and 12 either did not converge or produced highly improbable parameter estimates and standard errors (e.g., >5000 ± 100,000; Dail and Madsen 2011), likely due to the sparse capture data for Eastern Spotted Skunks (Table 2). The best model included positive effects of total cover (Cover), positive effects of understory woody vegetation (Woody), and a negative effect of forb presence (Forbs) (Tables 2, 3). The best model did not contain effects of woodland restoration on detectability. Mean probability of occupancy at unrestored sites (0.569 ± 0.063 SE) was 9 times greater than at woodland sites (0.063 ± 0.034). Detection probability was similar between restored and unrestored sites (0.108 ± 0.050). The first 6 principal components in our PCA of vegetative parameters at trap sites accounted for 84% of the variance in the data, with components 1 and 2 explaining 52% of the variance (Table 4). Principal component Axis 1 explained 34% of the variance in the data (Table 4, Fig. 3). At higher values of PC1, sites increased primarily in total cover, midstory BA, and horizontal vegetation cover at 1.75−2.25 (Den3), and decreased in amount of horizontal cover at 0−0.5 m (Den1) and forb/grass cover. Sites in unmanaged stands (grouped to the right on PC1) were distinctly separate from woodland sites, which were grouped to the left on PC1. Eighty-one percent of sites with Eastern Spotted Skunk captures had positive values of PC1. Component 2 explained an additional 18% of the variance. At higher values of PC2, sites increased Table 3. Parameter estimates (Betas) from the best model predicting occupancy of Eastern Spotted Skunks in restored woodlands and unrestored mature forests in the Ouachita Mountains of Arkansas, 1999–2001, including parameter effects on occupancy (Ψ) and detectability (P). Model parameter Estimate SE Intercept -7.99 4.13 Ψ(Cover) 2.41 2.19 Ψ(Woody) 2.14 1.22 Ψ(Forbs) -7.15 5.09 P -2.12 0.52 Southeastern Naturalist R.W. Perry, D.C. Rudolph, and R.E. Thill 2018 Vol. 17, No. 2 304 Figure 3. Principal component analysis based on vegetation parameters (Table 1) of sites where Eastern Spotted Skunks w e r e c a p t u r e d (closed circles) and sites with no captures (open circles) in woodlands and mature unmanaged forests. On PC1, sites to the left represented more-open forest conditions, whereas sites to the right represented denser forest condition with more total cover and midstory. Sites where Eastern Spotted Skunks were captured in unmanaged stands are circled on the right, whereas capture sites in woodlands are circled on the left. mostly in bare ground, but decreased in vegetation density at 0.75−1.75 m (Den2), and understory woody vegetation (Fig. 3). No obvious relationship existed between sites where Eastern Spotted Skunks were captured and PC2. Table 4. Principal component analysis loadings for the first 6 components (Prin1−Prin6) and percent of variance explained by each component using 13 vegetation parameters collected at trap sites (Table 1) in both woodlands and unmanaged forests combined in the Ouachita Mountains of Arkansas, 1999−2001. Parameter Prin1 Prin2 Prin3 Prin4 Prin5 Prin6 Overstory 0.177 0.207 0.487 -0.485 0.084 0.332 Midstory 0.325 0.163 -0.317 0.237 0.059 -0.086 Cover 0.398 0.188 0.181 -0.265 -0.017 0.092 Forbs -0.338 0.128 0.326 0.261 -0.099 -0.146 Grass -0.271 -0.230 -0.372 0.021 -0.081 0.610 Woody -0.192 -0.393 0.045 -0.296 0.025 -0.464 Leaf 0.286 -0.259 0.365 0.310 -0.289 -0.073 Bare -0.262 0.339 -0.211 -0.353 0.286 -0.313 LitterD 0.313 -0.205 -0.265 -0.157 0.293 0.174 Down -0.068 -0.090 0.288 0.380 0.834 0.107 Den1 0.326 0.244 -0.162 0.250 -0.031 -0.118 Den2 -0.063 0.568 -0.084 0.157 -0.020 -0.022 Den3 -0.342 0.240 0.136 0.078 -0.156 0.326 % variance explained 34 18 12 8 7 5 Southeastern Naturalist 305 R.W. Perry, D.C. Rudolph, and R.E. Thill 2018 Vol. 17, No. 2 Discussion Mean probability of occupancy of Eastern Spotted Skunks was 9 times greater and capture rates (mean captures/week) were 6 times greater in mature, unmanaged, second-growth forest than in restored, frequently burned woodlands. Similarly, studies within this same landscape found that Eastern Spotted Skunks used restored woodlands less than their availability (Lesmeister et al. 2013). One of the primary differences between woodlands and unmanaged stands is the presence of a dense midstory in unmanaged stands, which is lacking in woodlands (Fig. 1). Masters et al. (2002) found the density of woody stems (1−15-cm diameter) was 12 times greater in unmanaged mature forests than in woodlands of the ONF. Midstory (BA of midstory trees) was an important predictor in our PCA, but was not included in our best occupancy model. However, our midstory measure only included trees 10−15 cm dbh and did not account for smaller trees, which likely accounted for the lack of inclusion of Midstory in the best occupancy model. Nevertheless, this forest layer provides dense cover, which could potentially deter avian predation, and may provide needed cover for Eastern Spotted Skunks. Lesmeister et al. (2010) found that mortality of Eastern Spotted Skunks in our study area was attributable mostly (63%) to avian predators (e.g., Bubo virginianus [Gmelin] [Great Horned Owl]); the dense canopy and understory vegetation in younger (less than 30 years old) forests is believed to provide protection from aerial predation. Further, 92% of avian-caused mortality reported by Lesmeister et al. (2010) was in restored woodlands. Our occupancy model indicated total cover and understory woody-plant cover had the greatest positive effect, and cover of forbs had a negative effect on captures of Eastern Spotted Skunks. Further, our PCA also indicated Eastern Spotted Skunks were more likely to occur in areas with greater total cover and a lower amount of herbaceous vegetation. Areas with dense cover may be used disproportionally more by Eastern Spotted Skunks than open areas (Lesmeister et al. 2009). Greater total cover was the strongest factor affecting den sites selected by Eastern Spotted Skunks in the Ouachita Mountains (Lesmeister et al. 2008), and Sprayberry and Edelman (2018) found midstory and understory density provided a critical layer of cover for Eastern Spotted Skunk den sites. Forb cover is associated with more-open conditions (sunlight reaching the forest floor), and areas that are frequently burned have lower amounts of woody vegetation in the understory and greater forb coverage (Perry et al. 2009); thus, it seems logical that skunks may avoid areas with dense herbaceous vegetation and forb coverage. Lesmeister et al. (2009) found that Eastern Spotted Skunks used hardwooddominated forests (stream buffers and north slopes) more than or in proportion to their availability across the landscape. Similar to our unmanaged, pine-dominated stands, hardwood stands in the area were typically not subject to woodland restoration. These hardwood stands typically received little or no thinning and limited burning, which produces a structure similar to our unmanaged pine-dominated stands, including a dense midstory. Consequently, we believe that both young stands and mature forests with dense midstories (regardless of forest type) may Southeastern Naturalist R.W. Perry, D.C. Rudolph, and R.E. Thill 2018 Vol. 17, No. 2 306 provide comparable conditions in terms of structural protection from predators. Although previous studies found that Eastern Spotted Skunks avoid woodlands and are more likely to use hardwood stands or young−intermediate seral stages with a dense midstory (Lesmeister et al. 2009, Sprayberry and Edelman 2018, Thorne et al. 2017), our data suggests that mature pine-dominated forests with a well-developed midstory may also be important habitat. The importance of a dense midstory and woody understory for Eastern Spotted Skunks is becoming more apparent as additional studies throughout the range of this species are conducted (e.g., Sprayberry and Edelman 2018). The 10-cm–diameter opening size of our trap-entrance holes was an earlier trap design that allowed large snakes access but prevented larger mammals, such as mature Mephitis mephitis (Schreber) (Striped Skunk) and Didelphis virginiana (Kerr) (Opossum) from entering. However, size of the entry hole allowed juvenile Sylvilagus floridanus (J.A. Allen) (Eastern Cotton-tailed Rabbit), juvenile Opossum, and mature Eastern Spotted Skunks to enter. Later refinements of these traps for other studies changed the opening size to 5 cm, which prevented Eastern Spotted skunks and other medium–large mammals from entering (Burgdorf et al. 2005). During our trapping, we captured numerous herpetofauna and other animals, possibly providing an attractant to Eastern Spotted Skunks, which are known to consume and feed their young snakes, lizards, and small mammals (Sprayberry and Edelman 2016). In addition to the animals captured in the traps, the Eastern Spotted Skunks may also have been attracted to the permanent water source. Food resources for Eastern Spotted Skunks are likely more abundant in woodlands than unrestored mature forest, and studies of primary food types, including herpetofauna (Perry et al. 2009) and small mammals (Masters et al. 1998) have found a greater abundance of those taxa in woodlands. Thus, predator avoidance may override food-resource abundance as a driver of habitat use. Further, greater capture rates and occupancy rates of Eastern Spotted Skunks at unrestored sites and the lack of evidence for differences in detectability between these 2 forest types suggest that presence of prey items in traps did not bias habitat comparisons. Pine woodlands provide important habitat for a number of species and taxa, including endangered Picoides borealis (Vieillot) (Red-cockaded Woodpecker) and other bird species (Wilson et al. 1995), small mammals (Masters et al. 1998), and a number of reptiles (Perry et al. 2009). However, not all species benefit from restoring woodlands. Species such as Seiurus aurocapilla (L.) (Ovenbird; Wilson et al. 1995), Plethodon glutinosus (Green) (Slimy Salamander; Perry et al. 2009), and Eastern Spotted Skunk may be less abundant or absent in these woodlands. Therefore, to provide for diverse faunal assemblages, both woodlands and mature forests with an obvious midstory should be maintained across the landscape. Providing these denser forested areas may provide favored habitats for Eastern Spotted Skunks. On the Ouachita National Forest, the east–west oriented mountain range creates a landscape with sunny, south-facing slopes and shaded north slopes. South-slope areas typically consist of pine-dominated forests and are targeted for pine-woodland restoration, whereas north-slope areas are dominated by hardwood and mixed Southeastern Naturalist 307 R.W. Perry, D.C. Rudolph, and R.E. Thill 2018 Vol. 17, No. 2 hardwood–pine forests that are not typically managed as open woodland. Only around 19% of the total acreage (~142,000 ha) of the ONF has been designated for pine-woodland restoration. Furthermore, maintenance of unharvested riparian buffers or streamside management zones creates additional habitat throughout the woodland areas. Research by Lesmeister et al. (2009) indicated that Eastern Spotted Skunks favored these hardwood-dominated areas (along with regenerating forests) in the Ouachita Mountains. Consequently, a large portion of the forest is currently maintained in the forest conditions that Eastern Spotted Skunks apparently favor. Acknowledgments We thank R.A. Buford, J.H. Williamson, and students from Stephen F. Austin State University, Arkansas Tech University, and the University of Arkansas at Monticello for field assistance. We are grateful to W.M. Ford for reviews of an earlier draft and N.E. Koerth for statistical assistance. We thank L.D. Hedrick, W.G. Montague, and personnel of the Poteau- Cold Springs Ranger District of the Ouachita National Forest for their vision and assistance. Funding was provided by the Southern Research Station and the Ouachita National Forest. Literature Cited Animal Care and Use Committee. 1998. Guidelines for the capture, handling, and care of mammals as approved by the American Society of Mammalogists. Journal of Mammalogy 79:1416–1431. Burgdorf, S.J., D.C. Rudolph, R.N. Conner, D. Saenz, and R.R. Schaefer. 2005. A successful trap design for capturing large terrestrial snakes. Herpetological Review 36:421–424. Burnham K.P., and D.R. Anderson. 2002. Model Selection and Multimodel Inference: A Practical Information Theoretic Approach. Springer, New York, NY. 488 pp. Crabb, W.D. 1948. The ecology and management of the Prairie Spotted Skunk in Iowa. Ecological Monographs 18:201–232. Diggins, C.A., D.S. Jachowski, J. Martin, and W.M. Ford. 2015. Incidental captures of Eastern Spotted Skunks in high-elevation Red Spruce forest in Virginia. Northeastern Naturalist 22:N6–N10. Gompper, M.E., and H.M. Jackett. 2005. The long-term, range-wide decline of a once common carnivore: The Eastern Spotted Skunk (Spilogale putorius). Animal Conservation 8:195–201. Lesmeister, D.B., M.E. Gompper, and J.J. Millspaugh. 2008. Summer resting- and den-site selection by Eastern Spotted Skunks (Spilogale putorius) in Arkansas. Journal of Mammalogy 89:1512–1520. Lesmeister, D.B., M.E. Gompper, and J.J. Millspaugh. 2009. Habitat selection and homerange dynamics of Eastern Spotted Skunks in the Ouachita Mountains, Arkansas, USA. Journal of Wildlife Management 73:18–28. Lesmeister, D.B., J.J. Millspaugh, M.E. Gompper, and T.M. Mong. 2010. Eastern Spotted Skunk (Spilogale putorius) survival and cause-specific mortality in the Ouachita Mountains, Arkansas. American Midland Naturalist 164:52–60. Lesmeister, D.B., R.S. Crowhurst, J.J. Millspaugh, and M.E. Gompper. 2013. Landscape ecology of Eastern Spotted Skunks in habitats restored for Red-cockaded Woodpeckers. Restoration Ecology 21:267–275 MacKenzie, D.I., J.D. Nichols, J.A. Royle, K.H. Pollock, L.L. Bailey, and J.E. Hines 2006. Occupancy Estimation and Modeling: Inferring Patterns and Dynamics of Species Occurrence. Academic Press, Boston, MA. 648 pp. Southeastern Naturalist R.W. Perry, D.C. Rudolph, and R.E. Thill 2018 Vol. 17, No. 2 308 Masters, R.E., R.L. Lochmiller, S.T. McMurry, and G.A. Bukenhofer. 1998. Small-mammal response to pine-grassland restoration for Red-cockaded Woodpeckers. Wildlife Society Bulletin 26:148–158. Masters, R.E., C.W. Wilson, D.S. Cram, G.A. Bukenhofer, and R.L. Lochmiller. 2002. Influence of ecosystem restoration for Red-cockaded Woodpeckers on breeding bird and small-mammal communities. Pp. 73−90, In W.M. Ford, K.R. Russell, and C.E. Moorman (Eds). Proceedings: The Role of Fire For Nongame-Wildlife Management and Community Restoration: Traditional Uses and New Directions. USDA Forest Service, General Technical Report NE-288. Newtown Square, PA. 145 pp. McCullough, C.R., and E.K. Fritzell. 1984. Ecological observations of Eastern Spotted Skunks on the Ozark Plateau. Transactions of the Missouri Academy of Science 18:25–32. Nudds, T.D. 1977. Quantifying the vegetative structure of wildlife cover. Wildlife Society Bulletin 5:113–117. Perry, R.W., D.C. Rudolph, and R.E. Thill. 2009. Reptile and amphibian responses to restoration of fire-maintained pine woodlands. Restoration Ecology 17 :917–927. Reed, A.W., and M.L. Kennedy. 2000. Conservation status of the Eastern Spotted Skunk Spilogale putorius in the Appalachian Mountains of Tennessee. American Midland Naturalist 144:133–138. Spetich, M.A., R.W. Perry, C.A. Harper, and S. Clark. 2011. Fire in eastern hardwood forests through 14,000 years. Pp. 41–58, In C.H. Greenberg, B. Collins, and F.R. Thompson III (Eds.). Sustaining Young Forest Communities: Ecology and Management of Earlysuccessional Habitats in the Central Hardwood Region, USA. Springer, New York, NY. 312 pp. Sprayberry, T.R., and A.J. Edelman. 2016. Food provisioning of kits by a female Eastern Spotted Skunk. Southeastern Naturalist 15:N53–N56. Sprayberry, T.R., and A.J. Edelman. 2018. Den-site selection of Eastern Spotted Skunks in the southern Appalachian Mountains. Journal of Mammalogy 99:242–251. Thorne, E.D., C. Waggy, D.S. Jachowski, M.J. Kelly, and W.M. Ford. 2017. Winter habitat associations of Eastern Spotted Skunks in Virginia. Journal of Wildlife Management 81:1042–1050. Wilson, C.W., R.E. Masters, and G.A. Bukenhofer. 1995. Breeding-bird response to pine– grassland community restoration for Red-cockaded Woodpeckers. Journal of Wildlife Management. 59:56–67. Wilson, S.B., R. Colquhoun, A. Klink, T. Lanini, S. Riggs, B. Simpson, A. Williams, and D.S. Jachowski. 2016. Recent detections of Spilogale putorius (Eastern Spotted Skunk) in South Carolina. Southeastern Naturalist 15:269–274.