2011 NORTHEASTERN NATURALIST 18(4):509–520 The Effects Of Deer Exclosures On Voles and Shrews In Two Forest Habitats David Byman* Abstract - White-tailed Deer (Odocoileus virginianus) have overbrowsed much of the hemlock-mixed northern hardwood forest in northeastern Pennsylvania. I investigated the possible deleterious effect of this overbrowsing on 4 ground-cover-dependent smallmammal species through the use of deer exclosures. From May through September 1996–2005, 4 x 10 Sherman live-trap grids were placed in two 0.65-ha exclosures and their adjacent control sites in forest heavily browsed by deer. One exclosure was located beneath a primarily oak-maple canopy and the other exclosure under Tsuga canadensis (Eastern Hemlock). More Myodes gapperi (Southern Red-backed Vole) were captured in the oakmaple grids than in the hemlock grids and more M. gapperi were taken in the oak-maple exclosure than in the neighboring control. Microtus pinetorum (Woodland Vole) first appeared at the grids six years after the exclosure construction, and a large majority were taken in the oak-maple exclosure. Of Blarina brevicauda (Northern Short-tailed Shrew) captured at the grids, 80% were taken in the oak-maple grids and 58% in the oak-maple exclosure. Beginning in 2000, most of Sorex cinereus (Masked Shrew) captured were taken in the oakmaple habitat. Most of those trapped under the oak-maple canopy were taken in the control. These observations suggest that heavy deer browsing may depress populations of M. gapperi, M. pinetorum, and B. brevicauda, but not Sorex cinereus. Introduction It is well documented that Odocoileus virginianus Zimmermann (Whitetailed Deer) populations in forested regions of the northeastern United States can adversely affect plant regeneration and diversity (Alverson and Waller 1997, Behrend et al. 1970, Brenner 2006, Marquis and Grisez 1978, Russell et al. 2001, Tilghman 1989, Townsend et al. 2002). Studies have revealed the negative effects of deer browsing on young trees (Anderson and Katz 1993, Buckley et al. 1998, Healy 1997), shrubs (Levri et al. 2009, Townsend and Meyer 2002), and herbaceous plants under the forest canopy (Anderson 1994, Augustine and Frelich 1998, Shelton and Inouye 1995). Such heavy browsing by White-tailed Deer has had a particularly deleterious effect on plant regeneration in an uplands hemlock-northern hardwoods forest of northeastern Pennsylvania (Townsend et al. 2002). With the exception of bird populations (McShea and Rappole 2000), the indirect negative effects on wildlife from reduced understory diversity and cover density due to heavy deer browsing are not well studied. Browsing by deer could adversely affect small-mammal populations through the depletion of protective ground cover, modification of ground microclimate conditions, and competition for preferred foods. Presumably, semi-fossorial voles and shrews should be among the small mammals most sensitive to the loss of ground cover. *Penn State Worthington Scranton, 120 Ridge View Drive, Dunmore, PA 18512; dxb14@ psu.edu. 510 Northeastern Naturalist Vol. 18, No. 4 In the second-growth mixed conifer-hardwood upland forests of northeastern Pennsylvania, the most prevalent voles and shrews are Myodes gapperi Vigors (Southern Red-backed Vole), Microtus pinetorum LeConte (Woodland Vole), Blarina brevicauda Say (Northern Short-tailed Shrew), and Sorex cinereus Kerr (Masked Shrew) (Merritt 1987). The Southern Red-backed Vole has been described as a sometimes cyclic, opportunistic feeder whose distribution in the forests of the northeastern United States is determined by moisture availability as well as by litter abundance (Kirkland and Griffin 1974, Merritt 1981, Miller and Getz 1977). The semi-fossorial Woodland Vole normally requires a thick ground cover over well-drained soil (Connor 1953, Hamilton 1938, Miller and Getz 1969, Smolen 1981). The Northern Short-tailed Shrew has broad habitat requirements, but seems to prefer moist, deep litter in hardwood forest (George et al. 1986; Getz 1961; Miller and Getz 1977; Pruitt 1953, 1959; Yahner 1982, 1983). The Masked Shrew is associated with soil moisture maintained by soil organic matter and abundant understory vegetation (Getz 1961; Pruitt 1953, 1959; Whitaker 2004). Platt and Blakely (1973) suggested that Masked Shrew population levels are inversely correlated with those of the larger Southern Short-tailed Shrew, which does prey on Sorex spp. as well as its primarily invertebrate food species (Eadie 1949, Hamilton 1940). As the Woodland Vole is highly dependent on leaf litter and herbaceous cover for security, the removal of much of this cover by deer should adversely affect this rodent. As the abundance of the Red-backed Vole is dependent on moisture levels as well as cover and as this vole is also highly cyclic in numbers, high deer densities may not obviously affect this species to the degree that would be true for the Woodland Vole. As the body of the Northern Short-tailed Shrew is roughly twice as long and its weight three times as heavy as that of the Masked Shrew in Pennsylvania (Merritt 1987), the bigger animal is likely to require the denser cover for protection against predators. Since White-tailed Deer have depleted the forest ground cover in the Eastern Hemlock-White Pine-hardwoods forest of northeastern Pennsylvania, I predicted that improved habitat within deer exclosures would lead to increased numbers of the cover-dependent voles and shrews. I hypothesized that the populations of voles and shrews in the exclosures would increase through increased reproduction, survivorship, and/or immigration. To test this prediction, I compared vole and shrew populations inside and outside deer exclosures. Prior to the post-winter appearance of the leaves of deciduous trees, plants growing underneath deciduous canopies receive more light in the early growing season than do species underneath evergreen canopies. If as a result, deer exclosures under deciduous and evergreen canopies differ in ground cover density and diversity, vole and shrew densities may then differ between these protected sites as well. I predicted that cover-dependent voles and shrews would achieve higher populations in a deer exclosure placed in deciduous forest than in an exclosure under an evergreen overstory. Field-Site Description Lacawac Sanctuary (41°23'N, 75°17"W) is a 202-ha nature preserve located on the Pocono Plateau in northeastern Pennsylvania (Townsend et al. 2002). 2011 D. Byman 511 Anecdotal observations of high deer abundance at the sanctuary date from the early 1970s (Townsend et al. 2002). In 2000, deer density was estimated at 19 to 29 individuals km-2 (Townsend et al. 2002). At Lacawac, there was an obvious browse line at 1.5 to 1.6 m, with a scarcity of understory plants throughout the sanctuary. Within the sanctuary, Tsuga canadensis (L.) Carrière (Eastern Hemlock) was the dominant canopy tree, with Acer rubrum L. (Red Maple) and Quercus rubra L. (Red Oak) as subdominants. Quercus prinus L. (Chestnut Oak), Fagus grandifolia Ehrhart (American Beech), and Acer saccharum Marshall (Sugar Maple) were also important canopy trees. The understory at Lacawac was almost completely devoid of sizeable seedlings or sprouts or small saplings of these overstory trees (Townsend et al. 2002). Seedlings and sprouts of mature understory trees and the tall shrubs Hamamelis virginiana L. (Witch Hazel), Ilex verticillata (L.) A. Gray (Winterberry), Rhododendron maximum L. (White Laurel), Vitis labrusca L. (Fox Grape), Kalmia latifolia L. (Mountain Laurel), and Vaccinium corymbosum L. (Highbush Blueberry) were nearly absent throughout the sanctuary and were always small (≤20 cm) when present. Flowering herbaceous vegetation was also severely depleted in species composition and abundance. Methods Small mammal trapping I followed guidelines of the American Society of Mammalogists for the capture and handling of mammals (Animal Care and Use Committee 1998). This study was approved by the Penn State University Institutional Animal Care and Use Committee. Two 0.65-ha deer exclosures had been constructed in 1994–1995 at Lacawac Sanctuary. The oak-maple exclosure was in a site dominated by Red Oak (36% of stand density) and Red Maple (30% of stand density), with Chestnut Oak (10%), Eastern Hemlock (8%), Pinus strobus L. (White Pine) (6%), American Beech (4%), and Betula lenta L. (Sweet Birch) (3%) as associates (Townsend and Meyer 2002). The Eastern Hemlock exclosure was on a site with Eastern Hemlock as the primary overstory species (>70%). The two exclosures were 306 m apart, separated by 120 m of hemlock-dominated forest and 186 m of primarily oak-maple forest. The exclosure fences were constructed from livestock fencing (lower 1.2 m), and chicken wire (upper 1.2 m) attached to steel poles at 5-m intervals. Most of the bottom apertures within the livestock fence were 15.5 x 20 cm. However, some apertures on each side of each fence were larger, ranging up to 25 cm in width and 20 cm in height. The bottom wire of the fence generally lay on the ground, although the largest gap between fence and ground was 25 cm for the hemlock fence and 21 cm for the oak-maple fence. This fence design was adopted because it was easy to repair and would be a minor impediment to cross-fence travel by fox-sized and smaller animals. In contrast to the forest outside the exclosures, plant succession proceeded within the exclosures following the erection of the fences (Townsend et al. 2002). New growth of a flowering herbaceous understory and tree seedlings occurred throughout the oak-maple exclosure. In the hemlock exclosure, development of a new understory and ground cover was less extensive, only occurring below the 512 Northeastern Naturalist Vol. 18, No. 4 few breaks in the tree canopy and under scattered deciduous trees. Throughout the 10-year study, exclosure fences were checked for damage every week, and the exclosures were monitored for visible deer, deer pellets, and deer tracks in order to determine whether deer had entered the exclosures. Live-trap grids were placed within the oak-maple and Eastern Hemlock exclosures (OMX and HEX, respectively) in 1996. An additional two live-trap grids, the oak-maple (OMC) and Eastern Hemlock (HEC) controls were placed immediately outside each exclosure fence. The two control grids were within 20 m of the exclosure fence and were constructed so that they would parallel the grids within the exclosure fences. All four trap grids contained 40 large (23 x 9 x 7.5 cm) Sherman live traps (H.B. Sherman Traps models LFA, LFATDG, LFG, and LNG) in a 10 x 4 pattern with 10 m between traps. Traps were set during 3 consecutive evenings each 2-week period from May through September in each year. In 1996–1999, traps were baited with peanut butter and rolled oats. In 2000–2005, traps were baited with a mixture of sunflower seeds, oats, and raisins. When first captured, a mouse was marked by toe clipping from 1996 through 2000 and by ear tags (Scott Roestenberg - Western Tag) from 2001 to 2005. Shrews were marked by toe-clipping throughout the study. After each capture and before release, voles and shrews were weighed with a spring scale (Pesola) and voles were sexed. Data analyses In order to examine the effect of deer exclusion on the number of Southern Red-backed Voles and Woodland Voles captured over the 10-year duration of the study, a series of 2 (Grid) by 10 (Year) repeated-measures analyses of variance (ANOVA, SPSS 13.0 for Windows), with Grid as the between-subjects factor and Year as the within-subjects factor. An alpha level of 0.05 was used to test significance of indicated differences in the repeated-measured ANOVAs when examining grid site effects. The effect of deer exclusion on the number of shrews captured over the 10-year duration of the study was examined with the Wilcoxon signed-rank test (Ott 1988). Results In addition to the Southern Red-backed Vole, Woodland Vole, Northern Short-tailed Shrew, and Masked Shrew, mammals captured in the OMX, OMC, HEX, and HEC grids between 1996 and 2005 were Tamias striatus (L.) (Eastern Chipmunk), Sciurus carolinensis Gmelin (Eastern Gray Squirrel), Tamiasciurus hudsonicus Erxleben (Red Squirrel), Glaucomys volans (L.), Southern Flying Squirrel, Peromyscus leucopus Rafinesque (White-footed Mouse), Napaeozapus insignis Miller (Woodland Jumping Mouse), and 1 immature Didelphis virginiana Kerr (Virginia Opossum). The White-footed Mouse made up more than 90% of the captures on the four live-trap grids. Voles Southern Red-backed Voles were captured in both OMX and HEX during 1996, the first year of the study, and then in OMX and OMC in 1997 (Fig. 1). This species 2011 D. Byman 513 was not captured again until 2001, with 1 individual in OMC, and 2002, with 8 individuals in the OMX and 4 in OMC. No Southern Red-backed Voles were taken in 2003, 2004, or 2005. Other than the 3 Southern Red-backed Voles in 1996 that were taken in HEC, all the remaining animals were taken in the oak-maple grids, 12 in OMX and 7 in OMC. As 19 of the 22 Southern Red-backed Voles captured in the four grids were taken in the oak-maple traps, the repeated measures ANOVA was only calculated for the oak-maple exclosure and control grids. Within-subjects analysis. Examination of the fluctuations of the Southern Red-backed Vole populations over the ten-year study period revealed a uniformly significant year effect in all comparisons for both oak-maple grids, which was unsurprising given the large yearly fluctuations in numbers of animals captured (Fig. 1). Between-subjects analysis. Examination of the effect of different trapping grids on numbers of individual Southern Red-backed Voles captured over the 10-year research program failed to reveal a significant year x grid interaction for OMX vs. the OMC, Wilkes lambda = 0.910, F(1,9) = 1.454, P = 0.228. This result was despite the 2-to-1 disparity in the number of Southern Red-backed Voles captured in OMX vs. OMC in 2002. The first Woodland Vole captured at any of the grids was taken in OMC in 2001, 7 years after the construction of the exclosure fences and 6 years after the onset of this study (Fig. 1). In 2002, 1 Woodland Vole was taken in OMX and 1 in HEX. Beginning in 2003, 9 years after the construction of the exclosure fences, 20 individuals were taken in the OMX, 1 in OMC, and none in either hemlock grid. As 23 of the 24 Woodland Voles captured in the 4 grids were taken in the oak-maple traps, the repeated measures ANOVA was only calculated for the oakmaple exclosure and control grids. Within-subjects analysis. Examination of the fluctuations of Woodland Vole populations over the 10-year study period revealed a uniformly significant year effect in all comparisons for both oak-maple grids, the result of large yearly fluctuations in numbers of animals captured (Fig. 1). Between-subjects analysis. Examination of the effect of the exclosure fence on numbers of individual Woodland Voles captured over the 10-year research program on the oak-maple trapping grids found a significant year x grid interaction for OMX vs. OMC, Wilkes lambda = 0.812, F(1,9) = 7.124, P < 001. The 10-year trends in the numbers of the 2 vole species taken in the oak-maple grids were different (Fig. 1). Red-backed Voles were taken in the first 2 years of the study and then only seen again in 2001 and 2002. In contrast, the Woodland Vole first appeared in 2001 and sustained a trappable population for the last 5 years of the study. Shrews Northern Short-tailed Shrews were first captured in the 4 grids in 1999 (Fig. 2). In every year from 1999 to 2005, more Northern Short-tailed Shrews were taken in the two oak-maple grids than in the 2 hemlock grids, with a consistently increasing disparity in the last 4 years of the study. With the exception of 2001, more Northern Short-tailed Shrews were captured in the OMX than in the 514 Northeastern Naturalist Vol. 18, No. 4 OMC. The difference in number of captures in the two oak-maple grids in favor of OMX also steadily increased from 2002 to 2005, with a marked disparity of 19 animals in 2005: 26 in OMX vs. 7 in OMC. Figure 1. Yearly totals of individual Southern Red-backed Voles and Woodland Voles captured in each of four forested live-trap grids from 1996 through 2005. The four grids were placed in oak-maple and hemlock deer exclosures and nearby control sites. 2011 D. Byman 515 The one-sided Wilcoxon signed-rank test indicated that over the 10-year duration of this study significantly more Northern Short-tailed Shrews were captured in OMX than in OMC (n = 7, T = 1, P < 0.025). It was not possible to use this Figure 2. Yearly totals of individual Northern Short-tailed Shrews and Masked Shrews captured in each of four forested live-trap grids from 1996 through 2005. The four grids were placed in oak-maple and hemlock deer exclosures and nearby control sites. 516 Northeastern Naturalist Vol. 18, No. 4 test to evaluate the effect of the exclosure on Northern Short-tailed Shrews in the hemlock habitat. There were only 4 yearly observations (capture totals) with a non-zero difference, and 5 are required by the Wilcoxon signed-rank test. Masked Shrews were first captured in the 4 grids in 2000 (Fig. 2). No more than 2 Masked Shrews were taken each year from 2000 to 2003. As with the Northern Short-tailed Shrew, the number of Masked Shrews captured greatly increased in the last 2 years of the study, with more animals taken in the oak-maple sites than in the hemlock sites. However, within the oak-maple sites, the relative numbers of Masked Shrews taken in the OMX vs. OMC differed markedly from that observed with the Northern Short-tailed Shrew, with more Masked Shrews trapped in the control than in the exclosure (Fig. 2). In 2004, 7 Masked Shrews were captured in OMC and 4 were taken in the neighboring OMX. In 2005, 14 Masked Shrews were trapped in OMC and only 3 captured in OMX. It was not possible to use the Wilcoxon signed-rank test to compare the numbers of Masked Shrews captured over the 10 years of this study between either exclosure trap grid with its corresponding control trap grid. In both the oak-maple and the Eastern Hemlock habitats, there were fewer than 5 pairs of observations (capture totals) with a non-zero difference. The observed increase in the number of both shrew species captured in the last 2 years of the study, 2004 and 2005, occurred as 3 lepidopteran species, Lymantria dispar (L.) (Gypsy Moth), Malacosoma disstria Hubner (Forest Tent Caterpillar), and Malacosoma americanum Fabricius (Eastern Tent Caterpillar) entered into an outbreak phase (T. Marasco, Division of Forest Pest Management, Pennsylvania Bureau of Forestry, Harrisburg, PA, pers. comm.). The “ground crawling” mid-to-late larval stages of the Gypsy Moth were particularly abundant in 2005. None of these 3 Lepidopteran species had been in outbreak phase in Wayne County, PA and in adjacent Pike County since the early 1990s. Discussion Although sample sizes and the lack of site replication limit the conclusions that can be drawn concerning the effects of habitat type and deer exclusion on small-mammal abundance, this study supported the hypothesis that high deer density negatively affects the population density of the Southern Red-backed and Woodland Voles inhabiting a hardwoods forest habitat. So few individuals of either species were captured in the Eastern Hemlock habitat that it was impossible to determine anything about exclosure effects under that canopy (Fig. 1). That the repeated measures analysis of variance did not find a significant difference in the numbers of Southern Red-backed Vole captured in the oak-maple trap grids may be due to the prevalence of years with no captures at all at either site or the small number of individuals captured. This species is reported to exhibit cyclic fluctuations in numbers, with peaks ranging from 6 to 10 years (Grant 1976, Patric 1962). Two population peaks may have been observed during this study, in 1997 and 2002, and a few Red-backed Voles visited the oak-maple grids during those years. The numbers of animals captured were low, but after 7 years of deer-free plant succession, the number of Northern Red-backed Voles taken in 2011 D. Byman 517 OMX more than doubled. The 2-to-1 ratio of Northern Red-backed Voles in the oak-maple exclosure vs. the adjacent control in 2002 suggests a preference for habitat conditions associated with a lack of deer. The repeated measures analysis of variance did find a significant difference in the numbers of Woodland Voles in the oak-maple grids in favor of OMX. This result would be expected from the 20 to 1 disparity in Woodland Voles taken inside and outside the exclosure fence during the 2003–2005 trapping seasons (Fig. 1). The capture data suggest that under the conditions at Lacawac Sanctuary, a minimum number of 6 years of plant succession was required within the exclosure before the habitat was suitable for this vole species. It should be noted, however, that it is possible that a) it simply took 6 years for Woodland Voles emigrating from a population refuge to “find” the exclosure, or b) this pattern of captures is the result of cycles in abundance of Woodland Voles. As this semi-fossorial species requires thick ground cover of either litter or vegetation (Connor 1953, Miller and Getz 1969, Paul 1970) while eating forbs, grasses, roots, and seeds in forested habitats (Benton 1955, Cengel et al. 1978, Hamilton 1938), it is to be expected that this species would prefer OMX over its companion control grid after a sufficient number of years of succession provided adequate cover and food. It seems likely that a high deer density in the Pennsylvania’s mixed hardwood forest negatively affects Woodland Vole populations. This study suggests that the eastern hardwood forest is superior to the Eastern Hemlock forest as habitat for the Northern Short-tailed Shrew (Fig. 2). However, as Northern Short-tailed Shrews were captured in the Eastern Hemlock as well as the oak-maple grids from 1999 on and as in both habitats more Northern Short-tailed Shrews were generally caught in the exclosures than in the controls, it can be assumed that by 1999, all 4 grids had at least minimally adequate habitat for this shrew. In addition, both exclosures may have had better habitat for the Northern Short-tailed Shrew than their corresponding controls. The results of the Wilcoxon signed-rank test for the oak-maple grids suggest that the superiority of the OMX over its control as Northern Short-tailed Shrew habitat after 2002 was more pronounced than was the case of the hemlock exclosure when compared with its corresponding control. It is impossible to distinguish which of 2 possible effects of high deer density negatively affected the Short-tailed Shrew populations in the two controls. The deer removed protective cover, but also may have adversely affected the shrew’s invertebrate prey habitat. The outbreak of the Gypsy Moth in 2005 may give some clarity. If it is assumed that the density of the larvae in the soil was roughly the same on both sides of the exclosure, then the considerably higher number of Northern Short-tailed Shrews inside the fence was made possible by the superior protective ground cover inside the exclosure. The apparent superiority of the ground cover in the deer-free habitat was evident since 2002, but became dramatic when it could protect an unusually high Northern Short-tailed Shrew population, made possible in 2005 by an improved food supply. Although Sherman traps probably are not the most effectual means of assessing shrew abundance, comparison of captures of Masked and Northern Short-tailed Shrews revealed markedly different effects of high deer populations 518 Northeastern Naturalist Vol. 18, No. 4 and the resulting ground-cover depletion. Although the Masked Shrew first appeared in 2000, a year after the first captures of the Northern Short-tailed Shrew, few Masked Shrews were caught until 2004 (Fig. 2). In 2004 and 2005, the years of greater insect food abundance outside as well as inside the exclosure fence, the oak-maple exclosure fence appeared to have a dramatic but unexpected effect on the Masked Shrew population, as the Masked Shrew catch was markedly higher on the control grid. Even though a few more Masked Shrews were captured inside the exclosure in 2004 and 2005 than had been the case in previous years, the Masked Shrew did not appear to greatly benefit from the improved ground cover inside the oak-maple exclosure. One explanation for the greater abundance of the Masked Shrew outside the exclosure fence in the high insect food supply years of 2004 and 2005 would be the combination of the Masked Shrew’s small size and the large numbers of the Northern Short-tailed Shrew inside the fence. Since the Masked Shrew is roughly one quarter the size of the Southern Short-tailed Shrew, it seems possible that the smaller shrew was less likely than the bigger shrew to be spotted by predators in the limited cover available outside the fence. Since Northern Short-tailed Shrews prey on Sorex spp. (Eadie 1949, Hamilton 1940), the Masked Shrews may have tended to simply avoid contact with the large numbers of Northern Short-tailed Shrews inside the exclosure, “preferring” the predation risk created by the reduced cover outside the exclosure to that created by the unusually high numbers of Northern Short-tailed Shrews within the fence. However, it is also possible that the Masked Shrew preferred the deer-affected vegetation structure and/or the microclimate outside the exclosure. Now that 16 years have passed since the construction of the exclosure fences, it would be valuable to examine the relative abundances and diversity of coverdependent voles and shrews, ground-cover species, and soil invertebrates on the oak-maple grids to improve our understanding of the effects of high deer density on small mammals. It can then be determined whether the trends in vole and shrew abundance on the different sides of the deer exclosure fences described in this study have been sustained. If the trends continue, it will be interesting to examine whether the exclosure-based disparity in voles and shrew numbers is coincident with an increased disparity in vegetative ground cover and soil invertebrates on either side of the exclosure fences. Acknowledgments I wish to thank Dr. Daniel Townsend of the University of Scranton for his leadership in the construction of two deer exclosures in Lacawac Sanctuary. Without his invaluable efforts, this research would have been impossible. I would also like to thank Janice Poppich, the Executive Director of Lacawac Sanctuary and its Board of Directors for their support and interest in my research. Finally, I would like to acknowledge the invaluable financial support from Penn State Worthington Scranton, in both Research Development and Richard Matthews grants. 2011 D. Byman 519 Literature Cited Alverson, W.S., and D.M. Waller. 1997. Deer populations and the widespread failure of hemlock regeneration in northern forests. Pp. 280–297, In W.J. McShea, H.B. Underwood, and J.H. Rappole. (Eds.). The Science of Overabundance: Deer Ecology and Population Management. Smithsonian Institution Press, Washington DC. 402 pp. Anderson, R.C. 1994. Height of White-flowered Trillium (Trillium grandiflorum) as an index of deer browsing intensity. Ecological Applications 4:104–109. Anderson, R.C., and A.J. Katz. 1993. Recovery of browse-sensitive tree species following release from White-tailed Deer, Odocoileus virginianus, browsing pressure. Biological Conservation 63:203–208. 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 Augustine, D.J., and L.E. Frelich. 1998. Effects of White-tailed Deer on populations of an understory forb in fragmented deciduous forests. Conservation Biology 12:995–1004. Behrend, D.F., G.F. Mattfield, W.C. Tierson, and J.E. Wiley III. 1970. Deer density control for comprehensive forest management. Journal of Forestry 68:695–700. Benton, A.H. 1955. Observations on the life history of the Northern Pine Mouse. Journal of Mammalogy 36:52–62. Brenner, J.M. 2006. Browsing impact report for the Pennsylvania State Forests. Pennsylvania Department of Conservation and Natural Resources Bureau of Forestry, Harrisburg, PA Harrisburg, PA. Buckley, D.S., T.L. Sharik, and J.G. Isebrands. 1998. Regeneration of Northern Red Oak: Positive and negative effects of competitor removal. Ecology 79:65–78. Cengel, D.J., J.E. Estep, and R.L. Kirkpatrick. 1978. Pine Vole reproduction in relation to food habits and body fat. Journal of Wildlife Management 42:822–833. Connor, P.F. 1953. Notes on the mammals of a New Jersey pine barrens area. Journal of Mammalogy 34: 227–235. Eadie, W.R. 1949. Predation on Sorex by Blarina. Journal of Mammalogy 30:308–309. George, S.B., J.R. Choate, and H.H. Genoways. 1986. Blarina brevicauda. Mammalian Species 261:1–9. Getz, L.L. 1961. Factors influencing the local distribution of shrews. American Midland Naturalist 65:67–88. Grant, P.R. 1976. An 11-year study of small-mammal populations at Mont. St. Hilaire, Quebec, Canadian Journal of Zoology 54:2156–2173. Hamilton, W.J. 1938. Life-history notes on the Northern Pine Mouse. Journal of Mammalogy 19:163–170. Hamilton, W.J. 1940. The food of the Soricidae. Journal of Mammalogy 11:26–39. Healy, W.M. 1997. Influence of deer on the structure and composition of oak forests in central Massachusetts. Pp. 249–266, In W.J. McShea, H.B. Underwood, and J.H. Rappole (Eds.). The Science of Overabundance: Deer Ecology and Population Management. Smithsonian Institution Press, Washington DC. 402 pp. Kirkland, G.L., Jr., and R.J. Griffin. 1974. Microdistribution of small mammals at the coniferous-deciduous forest ecotone in northern New York. Journal of Mammalogy 55:417–427. Levri, E.P., M.A. Helsel, and M.A. Levri. 2009. Deer herbivory reduces juvenile recruitment in Mountain Laurel (Kalmia latifolia). Journal of the Pennsylvania Academy of Science 83:63–66. Marquis, D.A., and T.J. Grisez. 1978. The effect of deer exclosures on the recovery of vegetation in failed clearcuts on the Allegheny Plateau. United States Forest Service Research Note NE-475, Broomall, PA. 520 Northeastern Naturalist Vol. 18, No. 4 McShea, W.J., and J.H. Rappole. 2000. Managing the abundance and diversity of breeding bird populations through manipulation of deer populations. Conservation Biology 14:1161–1170. Merritt, J.F. 1981. Clethrionomys gapperi. Mammalian Species 146:1–9. Merritt, J.F. 1987. Guide to the Mammals of Pennsylvania. University of Pittsburgh Press, Pittsburgh, PA. 408 pp. Miller, D.H., and L.L. Getz. 1969. Life-history notes on Microtus pinetorum in central Connecticut. Journal of Mammalogy 50:777–784. Miller, D.H., and L.L. Getz. 1977. Factors influencing the local distribution and species diversity of forest small mammals in New England. Canadian Journal of Zoology 55:806–814. Ott, L. 1988. An Introduction to Statistical Methods and Data Analysis. PWS-Kent Publishing Company, Boston, MA 835 pp. Patric, E.F. 1962. Reproductive characteristics of Red-backed Mouse during years of differing population densities. Journal of Mammalogy 43:200–205. Paul, J.R. 1970. Observations on the ecology, populations, and reproductive biology of the Pine Vole, Microtus pinetorum, North Carolina. Illinois State Museum Reports of Investigations 20:1–28. Platt, W.J., and N.R. Blakely. 1973. Short-term effects of shrew predation upon invertebrate prey sets in prairie ecosystems. Proceedings of the Iowa Academy of Sciences 80:60–66. Pruitt, W.O., Jr. 1953. An analysis of some physical factors affecting the local distribution of the Shorttail Shrew (Blarina brevicauda) in the northern part of the Lower Peninsula of Michigan. Miscellaneous Publications of the Museum of Zoology, University of Michigan 79:1–39. Pruitt, W.O., Jr. 1959. Microclimate and local distribution of small mammals on the George Reserve, Michigan. Miscellaneous Publications of the Museum of Zoology, University of Michigan 109:1–27. Russell, F.L., D.B. Zippin, and N.L. Fowler. 2001. Effects of White-tailed Deer (Odocoileus virginianus) on plants, plant populations, and communities: A review. American Midland Naturalist 146:1–26. Shelton, A.L., and R.S. Inouye. 1995. Effect of browsing by deer on the growth and reproductive success of Lactuca canadensis (Asteraceae). American Midland Naturalist 134:332–339. Smolen, M.J. 1981. Microtus pinetorum. Mammalian Species 147:1–7. Tilghman, N.G. 1989. Impacts of White-tailed Deer on forest regeneration in northwestern Pennslvania. Journal of Wildlife Management 53:524–532. Townsend, D.S., and A.D. Meyer. 2002. Rapid recovery of Witch Hazel (Hamamelis virginiana L.) by sprouting, following release from White-tailed Deer (Odocoileus virginianus Zimm.) browsing. Natural Areas Journal 22:290–295. Townsend, D.S., J.S. Seva., C. Hee-Seagle, and G. Mayers. 2002. Structure and composition of a northern hardwood forest exhibiting regeneration failure. Bartonia 61:1–13. Whitaker, J.O., Jr. 2004. Sorex cinereus. Mammalian Species 743:1–9. Yahner, R.H. 1982. Microhabitat use by small mammals in farmstead shelterbelts. Journal of Mammalogy 64:440–445. Yahner, R.H. 1983. Population dynamics of small mammals in farmstead shelterbelts. Journal of Mammalogy 64:380–386.