2011 NORTHEASTERN NATURALIST 18(1):95–106 Day-Roosts of Myotis leibii in the Appalachian Ridge and Valley of West Virginia Joseph S. Johnson1,*, James D. Kiser2, Kristen S. Watrous3, and Trevor S. Peterson3 Abstract - Currently there is little known about day-roosts used by Myotis leibii (Eastern Small-footed Myotis) in the Central Appalachians. To provide insights on this species’ day-roosting habits, we successfully radiotracked 5 lactating females and 5 non-reproductive males to 57 day-roosts during June and July 2008. Eastern Small-footed Myotis used ground-level rock roosts in talus slopes and rock fields (n = 53), and roosts in vertical cliff faces (n = 4). Ground-level roosts had low canopy cover (males: x̅ = 14.1 ± 2.1 [SE] %, females: x̅ = 19.6 ± 3.1%), but were located close to vegetation (males: x̅ = 3.6 ± 0.4 m, females: x̅ = 4.8 ± 0.8 m). Males switched roosts every 1.1 ± 0.04 days, traveled 41.2 ± 7.8 m between consecutive roosts, and roosted 415 ± 49.0 m from capture locations. Females switched roosts every 1.1 ± 0.06 days, traveled 66.5 ± 14.6 m between consecutive roosts, and roosted 368 ± 24.0 m from capture locations. Ground-level roosts used by females were closer to ephemeral water sources (x̅ = 226 ± 31.2 m, n = 25) than those used by males (x̅ = 458 ± 16.7 m, n = 28; W = 401, P < 0.01). These data illustrate the importance of rock habitat with high solar exposure near protective cover and water in day-roost selection by Eastern Small-footed Myotis. Introduction Myotis leibii Audubon and Bachman (Eastern Small-footed Myotis) is among North America’s less common bat species. At present, it is listed as rare or imperiled throughout its range, and given legal protection in several states (Best and Jennings 1997, Erdle and Hobson 2001, NatureServe 2010). The species is considered rare, in part, because it is uncommonly observed during summer and winter surveys. Whereas localized summer abundance of the species has been reported, reproductive colonies rarely have been observed, and summer populations are thought to be small and scattered across the species’ range (Best and Jennings 1997, Erdle and Hobson 2001). Winter populations of Eastern Small-footed Myotis have been equally difficult to assess due to the tendency of these bats to hibernate individually or in small groups, often under rocks on cave and mine floors, or in crevices in cracked clay cave floors and crevices at cave entrances, where they often remain unobserved by biologists (Barbour and Davis 1974, Whitaker and Hamilton 1998). This lack of information on summer and winter colonies of Eastern Small-footed Myotis has led to uncertainty about population trends and distribution of the species. The Eastern Small-footed Myotis is also among North America’s smallest bat species, with adults typically weighing 3–6 g (Best and Jennings 1997). 1University of Kentucky, Lexington 40546. 2Stantec Consulting, Jeffersonville, IN 47199. 3Stantec Consulting, Topsham, ME 04086. *Corresponding author - joseph.johnson@ uky.edu. 96 Northeastern Naturalist Vol. 18, No. 1 Accordingly, this low mass in relation to size of currently available radiotransmitters has hindered radiotelemetry studies of the species; therefore, data on summer behavior are limited to isolated records. Examples of these include reports of summer maternity colonies in ground-level rock crevices, cliff faces, buildings, and bridge expansion joints (Barbour and Davis 1974; Harvey et al. 1999; J. MacGregor, Kentucky Division of Fish and Wildlife Resources, Frankfort, KY, unpubl. data). Observations of day-roosts during the spring (Johnson and Gates 2008, Tuttle 1964) and fall (Roble 2004) indicate similar roosting behavior during those periods as well. Records indicate that Eastern Small-footed Myotis are influenced by the availability of suitable rock habitat or similar anthropogenic surrogates, such as bridges or buildings (Barbour and Davis 1974; Erdle and Hobson 2001; Harvey et al. 1999; Johnson and Gates 2008; J. MacGregor, unpubl. data; Roble 2004; Tuttle 1964). However, it is unclear what constitutes suitable day-roosting habitat, or whether day-roosting habitat varies among sex and reproductive classes, as has been documented for other North American Myotis species that roost in tree and rock substrates (Baker and Lacki 2006, Chruszcz and Barclay 2002, Lausen 2007, Lausen and Barclay 2002). Johnson and Gates (2008) found no difference in roosting behavior of Eastern Small-footed Myotis among 10 rock outcrops used as roosts by adult females and unused rock outcrops where absence was assumed. Unfortunately, their inferences were limited because the research occurred in the staging period immediately post-emergence from hibernation. Given the paucity of roosting data on Eastern Small-footed Myotis and the diversity of findings on North American Myotis species utilizing rock substrates for day-roosts (Chruszcz and Barclay 2002; Johnson and Gates 2008; Lacki and Baker 2007; Lausen 2007; Lausen and Barclay 2002, 2003), the objective of our study was to characterize day-roosting habitat for male and female Eastern Small-footed Myotis and determine whether day-roosting habitat differed between sexes. We also characterized roost-switching behavior and examined differences in these behaviors between sexes. Study Area and Methods Study area Our study took place along New Creek Mountain in Grant and Mineral counties, WV (39.2538°N, 79.1112°W). Average monthly temperatures measured with a Hobo data logger (Model U23-001, Onset Computer Corporation, Pocasset, MA) increased from 17.7 to 20.7 °C from May to July 2008, similar to climate normals for these months (NCDC 2002). The ridgeline of New Creek Mountain is 12 km from north to south, with a maximum elevation of 927 m asl. New Creek Mountain is located at the western edge of the Appalachian Ridge and Valley Physiographic Province within the Ridge and Valley Section of the oak-chestnut (Quercus spp.-Castanea sp.) Forest Region as described by Braun (1950). Historically, this forest was dominated by an overstory of Castanea dentata (Marshall) Borkh (American Chestnut) and various oaks including Quercus prinus L. (Chestnut Oak), Q. alba L. (White Oak), Q. velutina Lambert (Black Oak), and Q. rubra L. (Northern Red Oak) depending on aspect, elevation, and site quality. The loss of American Chestnut from the forest due to Cryphonectria 2011 J.S. Johnson, J.D. Kiser, K.S. Watrous, and T.S. Peterson 97 parasitica (Murrill) Barr (Chestnut Blight) has reshaped the forest composition. Today, the mountain is still predominantly forested, consisting primarily of mixed hardwood stands with some areas of Pinus spp. (pines) along ridetop and upperslope roads and Tsuga canadensis Carrière (Eastern Hemlock) along streams at lower elevations. Forests on xeric and subxeric ridgetops and south- to southwestern-facing slopes are dominated by an overstory of 8–12-m-tall Q. coccinea Münchhausen (Scarlet Oak), Nyssa sylvatica Marshall (Black Gum), Acer rubrum L. (Red Maple), Black Oak, Chestnut Oak, and White Oak, while forests on mesic to submesic locations, such as eastern and southeastern slopes, are dominated by A. saccharum Marshall (Sugar Maple), Fagus grandifolia Ehrhart (American Beech), Liriodendron tulipifera L. (Tuliptree), Betula lenta L. (Sweet Birch), Northern Red Oak, and White Oak. Past infestations of Lymantria dispar L. (Gypsy Moth) have created successive overstory mortality waves, producing a high proportion of snags and a dense shrub layer dominated by Vaccinium angustifolium Aiton (Low Bush Blueberry), V. pallidum Aiton (Blue Ridge Blueberry), Kalmia latifolia L. (Mountain Laurel), and saplings of overstory trees. Non-forested landcover consists of sandstone talus slopes, small house clearings, forest roads, and transmission lines. Talus slopes are characterized by unvegetated, exposed patches of rocks and smaller patches of isolated trees and shrubs. Examination of aerial photographs in ArcMap (ESRI 2008) showed talus slopes ranged in size from a few square meters to 17 ha. Talus slopes are limited to two general areas: the western slope of the mountain, and the base of a large vertical cliff marking the southern terminus of the ridgeline. Rocky groundcover is also present underneath the forest canopy throughout the study area; several forest stands occur on ground cover nearly identical in composition to talus slopes. Forest clearing for two transmission lines created additional exposed rock habitat similar to naturally occurring talus slopes, although smaller in size. Capture and radiotelemetry Capture, handling, and radiotelemetry techniques followed the American Society of Mammalogists’ guidelines for the use of animals in research (Gannon et al. 2007). We captured bats in 38-mm diameter polyester mist-nets (Avinet, Inc., Dryden, NY) placed over ridge-top roads and ponds at twelve distinct locations, each sampled during two non-consecutive evenings. We deployed two nets at each capture site, with nets opened approximately 30 min prior to sunset and remaining opened until 5 h after sunset. We recorded age, sex, reproductive condition, weight, and right forearm length for all individual bats caught. We aged bats as adult or juvenile by examining ephiphyseal-diaphyseal fusions (calcification) of long bones in the wing (Anthony 1988). Females were determined to be non-reproductive, pregnant, or lactating based on the presence of a fetus or teat condition, and males were determined to be non-reproductive or scrotal based on the swelling of the epididymides (Racey 1988). We fitted adult male and lactating female Eastern Small-footed Myotis weighing at least 5.0 g with 0.35-g radiotransmitters (model LB-2N, Holohil Systems Ltd., Carp, ON, Canada) attached between the shoulder blades using surgical adhesive (Torbot®, Cranston, RI). We tracked radiotagged bats to their day-roosts each day using TRX-1000S telemetry receivers (Wildlife Materials, Inc., Murphysboro, IL) and 3-element yagi 98 Northeastern Naturalist Vol. 18, No. 1 antennas (Advanced Telemetry Systems [ATS], Inc., Isanti, MN). We recorded the location of each roost using a handheld global positioning system (GPS), and chronological accounts of each bat’s day-roosting locations were recorded. We confirmed the number of bats inhabiting each roost through emergence counts or by visual inspection of the roost. Using Wilcoxon rank-sum tests (SAS Institute 2004), we compared length of continuous residency (calculated for each bat by dividing the total number of roost days by the number of observed roost-switches) at day-roosts, distance traveled between consecutive roosts, and distances between roosts and capture sites for males and females. Day-roosting habitat We measured canopy cover, slope, aspect, and distance to vegetation outside each day-roost using a spherical densiometer, clinometer, compass, and meter tape, respectively. We used GIS digital elevation models analyzed in ArcMap to determine the elevation of roosts, and used 2-m resolution aerial photographs to determine the quantity of open rock habitat present within 100-m, 200-m, and 1000-m radius plots surrounding each roost. We chose these distances based on distances radiotagged bats traveled between consecutive roosts (Table 1). For our analysis, we defined rock habitat as cliff-lines or patches of unvegetated rock field or talus slopes with 50% canopy cover or less. We used aerial photographs to determine the distance from each roost to the closest available water feature. Because our capture efforts showed heavy use of small upland ponds and ephemeral road ruts, we examined distance to these water bodies separately from distance to perennial streams located downslope of New Creek Mountain. When bats could be visually confirmed inside rock crevices, we measured the maximum width and depth of each crevice, as well as the length and width of the entrance of each crevice. We characterized the orientation of each crevice as vertical or horizontal when orientation was within ± 20° of the horizontal or vertical plane (Lacki and Baker 2007). All other roost crevices were characterized as diagonal. We compared canopy cover, slope, aspect, distance to vegetation, and amount of rock habitat between male and female roosts using non-parametric Wilcoxon rank-sum tests (SAS Institute 2004). Ground-level roosts in rock crevices represented a different environmental setting than roosts located in clifflines, therefore we excluded cliff roosts from statistical analyses of day-roosting habitat. Because aspect is a circular measurement, these data were transformed (Fisher 1993), and we compared values for male and female roosts using twosample Watson-Williams tests (Kölliker and Richner 2004; SAS Institute 2004). All tests were based on a significance level of 0.05. Results Capture and radiotelemetry We captured 61 adult Eastern Small-footed Myotis (25 females, 34 males, and 2 unknown) over 50 net-nights between 26 May and 5 July 2008. Only Myotis septentrionalis Trouessart (Northern Myotis) were captured more frequently (115 adult males and 24 adult females). Adult weights for Eastern Small-footed Myotis ranged from 3.5 to 6.5 g (x̅ = 4.8 ± 0.1 [SE] g). The first lactating female was captured on 2 June. Eighty percent of Eastern Small-footed Myotis were 2011 J.S. Johnson, J.D. Kiser, K.S. Watrous, and T.S. Peterson 99 captured over road-ruts or ridge-top ponds, where sampling accounted for 44% (n = 22) of the total number of net-nights in the study. The number of Eastern Small-footed Myotis captured decreased with distance from rock habitat; capture rates never exceeded 1 bat per net per night (net-night) at capture sites ≥200 m from rock habitat (Fig. 1). We fitted 5 lactating females and 6 non-reproductive males with radiotransmitters. The maximum increase in wing loading for radiotagged bats was 7.8% (x̅ = 6.8 ± 0.1%). Although this exceeds the 5% increase recommended by Aldridge and Brigham (1988), it was comparable to other bat field studies (Arnett and Hayes 2009, Kurta and Murray 2002), including recent studies of Eastern Small-footed Myotis (Johnson and Gates 2008) and Myotis ciliolabrum Merriam (Western Small-footed Myotis) (Lausen 2007). One radiotagged male was never relocated during the day despite radio signals being detected during the night. We tracked the remaining 10 bats for 4–9 (x̅ = 6.5 ± 0.6) days, resulting in the location of 32 male and 25 female day-roosts. Bats roosted exclusively in rock structures. The majority of roosts (93.0%; n = 53) were in ground-level rock crevices in talus slopes or rock fields within transmission-line clearings. Roost crevices were narrow cracks in sandstone boulders or narrow spaces between adjacent rocks. The remaining 4 roosts were located in a south-facing vertical cliff face at the southern terminus of the mountain. All 4 cliff roosts were used by an individual male that also used ground-level rock crevices. Male Eastern Small-footed Myotis roosted deep within rock crevices; a male bat was visible in only 1 out of 32 (3.1%) day roosts examined visually. Using Figure 1. Decreasing capture rates of Eastern Small-footed Myotis with distance from available rock habitats at New Creek Mountain, WV, 2008. 100 Northeastern Naturalist Vol. 18, No. 1 visual inspection and 4 counts of bats emerging from roosts at sunset, we observed that males were solitary in their roost habits. We frequently observed roosting females near the exterior of roost crevices, where they could be counted within 17 of 25 (68.0%) day-roosts. Over these visual confirmations and 3 emergence counts, we found females roosting solitarily or in groups of up to 8 bats (x̅ = 2.4 ± 0.4). We observed juvenile bats in the roost on several occasions, but not with regularity, and juveniles were not included in estimates of roosting bats. Because no juvenile bats were captured during mist-netting efforts, counts of bats emerging at sunset are assumed to include only adult bats. Both males and females switched roosts frequently, making short-distance movements between consecutive roosts (Table 1). Roosts were rarely observed being used on more than 1 day, with only 2 males and 3 females revisiting 1 roost each on only 1 occasion. The greatest distance between any 2 roosts used by an individual bat throughout that individual’s tracking period was 204 m for a radiotagged male and 140 m for a female. Length of continuous residency at roosts (W = 748, P = 0.46) and distance between consecutive roosts (W = 518, P = 0.30) did not differ between males and females. All roosts were located ≤900 m from capture sites, and distances between roosts and capture locations did not differ (W = 701, P = 0.71) between males and females. Day-roosting habitat Overall, ground-level rock roosts were situated on steep slopes with less than 50% canopy cover, proximal to nearby vegetation (Table 2, Fig. 2). Male and female day-roosts did not differ in mean distance to vegetation (W = 719, P = 0.44), canopy cover (W = 735, P = 0.29), elevation (W = 601, P = 0.19), slope (W = 699, P = 0.68), or aspect (T = 0.24, P = 0.35). Our GIS analysis revealed that 100-m-, 200-m-, and 1000-m-radius plots surrounding roosts were composed of 55%, 37%, and 9% rock habitat (talus slopes, cliff faces, or other open-canopy rock habitat), respectively. Amount of rock habitat within 100-m- (W = 640, P = 0.17), 200-m- (W = 647, P = 0.21), or 1000-m- (W = 659, P = 0.29) radius plots did not differ between male and female roosts (Table 2). Distance to the nearest perennial stream did not differ between male and female (W = 602, P = 0.19) roosts, but female roosts were located closer to upland water sources than male roosts (W = 401, P < 0.01; Table 2). Distances to ephemeral, upland water were less than half the distances to perennial streams (Table 2). No roost was located in abundant rock habitat found beneath forest canopies, with a maximum of 52% canopy cover for any roost. Roosts were located 19.3–236 m (x̅ = 112 ± 9.2) downslope from the ridgeline where bats were captured. Table 1. Summary of roost-switching behavior of Eastern Small-footed Myotis at New Creek Mountain, WV, 2008. Data presented are means ± SE (range). Males Females Number of bats 5 5 Length of continuous residency (days) 1.1 ± 0.04 (1–2) 1.1 ± 0.06 (1–2) Distance from previous roost (m) 41.2 ± 7.8 (1.0–140) 66.5 ± 14.6 (5.1–204) Distance from capture site (m) 415 ± 48.0 (90.4–882) 368 ± 24.0 (98.9–525) 2011 J.S. Johnson, J.D. Kiser, K.S. Watrous, and T.S. Peterson 101 Of the 17 female day-roosts where we could observe roosting bats, 35% (n = 6) were vertically oriented, 35% (n = 6) were diagonal, and 29% were horizontal (n = 5). Of the 7 female day-roosts with more than 2 roosting bats, 4 were vertical crevices, and 3 were diagonal crevices, including the largest maternity roost. Female roost crevices averaged 50.5 ± 0.4 cm wide and 38.3 ± 0.3 cm deep, with an average opening size of 73.5 ± 12.3 cm2. The sole male roost measured was a horizontal crevice 44 cm wide, 39 cm deep, with an opening of 73.5 cm2. The dimensions of other male roosts were not measured because the exact location of the bat could not be determined. Discussion We found no evidence of differences between non-reproductive male and lactating female Eastern Small-footed Myotis in day-roosting habitat or roostswitching behavior. Our sample size of bats (5 males and 5 females) was too small for creation of multivariate models (Cox et al. 2008), but our data represent one of the few datasets describing Eastern Small-footed Myotis ecology. Radiotagged males and females chose day-roosts with remarkably similar habitat characteristics. However, our observations of females close to the exterior of roosting crevices, and inability to visually locate males within roosts, suggests that males and reproductive females may occupy different microhabitats within rock crevices that appear identical at the scale considered in this study. Temperature recordings inside a maternity colony of Eastern Small-footed Myotis in a bridge expansion joint showed maximum daily roost temperatures of 38 °C from June to early September, with fluctuations averaging 10–11 °C over most 24-h time periods (J. MacGregor, unpubl. data). High temperatures inside day-roosts are advantageous for pregnant females by aiding in fetal development, and advantageous during the lactation period by aiding in the growth of young, although lactating females of some species have been observed selecting slightly Table 2. Summary of day-roosting habitat of Eastern Small-footed Myotis at New Creek Mountain, WV, 2008. Data presented are means ± SE. Variable Males Females Number of bats 5 5 Number of roosts 25 28 Canopy cover (%) 14.1 ± 2.1 19.6 ± 3.1 Distance to vegetation (m) 3.6 ± 0.4 4.8 ± 0.8 Elevation (m asl) 718 ± 26.9 691 ± 13.0 Slope (degrees) 27.5 ± 1.1 27.9 ± 1.2 Aspect (degrees)A 295 ± 44.2 281 ± 36.9 Distance to perennial stream (m) 979 ± 104.6 941 ± 84.9 Distance to upland water (m)B 458 ± 16.7 226 ± 31.2 Rock habitat within 100-m-radius plots (ha) 1.9 ± 0.1 1.6 ± 0.2 Rock habitat within 200-m-radius plots (ha) 5.1 ± 0.4 4.2 ± 0.6 Rock habitat within 1000-m-radius plots (ha) 31.4 ± 1.6 26.6 ± 2.6 AMean aspect and two sample Watson-Williams test calculated using circular statistics (Kölliker and Richner 2004). BSum of ranks significantly different between males and females (P < 0.01). 102 Northeastern Naturalist Vol. 18, No. 1 cooler roosts, which facilitate the use of shallow torpor during the day (Lausen and Barclay 2002, 2003; Solick and Barclay 2006; Speakman and Thomas 2003). In contrast, males and non- and post-reproductive females may benefit from selecting roosts with even cooler microclimates that facilitate an increased use of torpor (Lausen and Barclay 2002, 2003; Solick and Barclay 2006; Speakman and Thomas 2003). Although some research has shown that pregnant and lactating females select day-roosts with differing structures and differing micro-climates (Lausen and Figure 2. Aerial photograph of Eastern Small-footed Myotis day-roosts (white stars) situated near the edges of talus slopes (light gray areas) on New Creek Mountain, WV, 2008. 2011 J.S. Johnson, J.D. Kiser, K.S. Watrous, and T.S. Peterson 103 Barclay 2002, 2003), this is not consistent among all studies (Solick and Barclay 2006), and additional research is needed to determine if differences exist among various sexes and reproductive classes of Eastern Small-footed Myotis. Although we did not compare day-roosts to random locations similar to other studies of rock-roosting species (Lausen 2007; Lausen and Barclay 2002, 2003), we believe our data indicate that roosts were not chosen randomly by Eastern Small-footed Myotis. Day-roosts were located close to patches of shrubs contained within rock fields/talus slopes, or were close to the edge of surrounding forest. Although areas of unvegetated talus slope were as large as 17 ha, no ground-level rock roosts were located >15 m from vegetation or forest edge. Similarly, no ground-level rock roost was located in an area with >52% canopy cover, despite an abundance of rock habitat within forest stands. We hypothesize that roost selection in Eastern Small-footed Myotis is based on either avoiding detection by predators or minimizing energy expenditures. Importantly, day-roosts were located several meters away from vegetated areas, situating day-roosts in areas with low canopy cover, and therefore high solar exposure. We suggest that rocky habitat associated with patches of vegetation or extensive forest edges is particularly valuable because it provides day-roosting habitat with both high solar exposure and short distance to protective cover. In Maryland, Johnson and Gates (2008) located 10 day-roosts by radiotracking 4 female Eastern Small-footed Myotis during the staging period shortly after emerging from hibernation. All 10 day-roosts were in rock crevices in rock outcrops along south-facing slopes; known roosts did not differ from randomly located rock roosts. Day-roosts of Eastern Small-footed Myotis in Maryland differed from day-roosts in this study by being lower in elevation (x̅ = 182 m), located on steeper slopes (x̅ = 37.7°), and situated under greater canopy cover (x̅ = 85.4%). The lower elevation of day-roosts in Maryland likely result in warmer ambient and roost temperatures, allowing for bats to roost underneath greater canopy cover. Data from day-roosts used by pregnant and lactating female Western Smallfooted Myotis in Alberta, Canada, also serve as a useful comparison with our study given the shortage of data on Eastern Small-footed Myotis (Lausen 2007). Day-roosts from Alberta differed from those in West Virginia (this study) in 4 of 6 measures. Sixty-eight percent of Western Small-footed Myotis in Alberta used roosts of a different substrate (mudstone), had opening sizes less than half the size of those documented in this study (31.7 ± 12.6 cm2), were located on steeper slopes (62 ± 4.0°), and had aspects facing south as opposed to west (Lausen 2007). Many day-roosts located in this study were situated on a west-facing slope, where the availability of potentially suitable rock habitats was limited elsewhere. Differences in roost aspect between studies, therefore, may be a result of availability. Further, roosts used by lactating Western Small-footed Myotis had similar depth (36.2 ± 5.2 cm) and consisted of 40% horizontally oriented roosts. Roosting behaviors among Eastern and Western Small-footed Myotis were more similar than roost habitat. Reproductive females in Alberta switched roosts frequently, revisiting a roost on consecutive days on only 3 of 21 (14.3%) occasions (Lausen 2007). Distances between consecutive roosts were similar, ranging from 6.4–106 m (45.6 ± 6 m), and distances between capture locations from first roosts were comparable (146 ± 23 m, range = 4–580 m) to our measurement of capture 104 Northeastern Naturalist Vol. 18, No. 1 location to all roosts. Colony size of lactating females was also small (1.3 ± 0.2 bats, range = 1–5). Frequent roost-switching and short distances between consecutive roosts have commonly been observed among rock-roosting species (Cryan et al. 2000, Johnson and Gates 2008, Lacki and Baker 2007, Lausen and Barclay 2002, Solick and Barclay 2006, Weller and Zabel 2001). Lausen and Barclay (2002) hypothesized that frequent roost-switching benefits individuals through predator avoidance and reduction in ectoparasite loads. Although we did not collect sufficient data on ectoparasite loads to test this hypothesis, these hypotheses may also explain frequent roost-switching among Eastern Small-footed Myotis. Substantial decline of Eastern Small-footed Myotis capture rates with increasing distance from available rock habitat and short distances between roosts and capture sites suggests these bats have small home ranges, in agreement with observations of four females in Maryland during spring (Johnson et al. 2009). In light of these small home ranges, proximity to upland ponds and ephemeral water sources, such as road ruts, may be important in day-roost selection. This need to be near water may be especially true for lactating females, which need to produce milk for pups. The importance of water is supported by our findings that 80% of Eastern Small-footed Myotis were captured over water bodies despite only 44% of capture effort occurring over water, location of day-roosts no farther than 525 m away from upland water sources, and closer proximity to water of female day-roosts than male roosts. Radiotagged bats did not display any aberrant behavior. Females were observed roosting in social groups, and all bats were documented to exit their roosts and presumably forage each night until the transmitter became detached. Therefore, our data suggest that reproductive females can safely handle a 7–9% increase in wing loading, and can be readily captured in habitats with large amounts of available rock substrates. These data, alongside the conservation status of the species (Best and Jennings 1997, Erdle and Hobson 2001), should encourage increased attention on researching the roosting and foraging behaviors of the species. The current scarcity of data hinders effective management and protection of habitat for the species and potentially masks important geographic variation in day-roost selection (Lacki et al., in press). For example, studies from across the range of Myotis thysanodes Miller (Fringed Myotis), a western North American forest bat species that preferentially roosts in rock substrates, have shown high variability in day-roost selection, including extensive use of dead trees (Cryan et al. 2000, Lacki and Baker 2007, Rabe et al. 1998, Weller and Zabel 2001). Eastern Smallfooted Myotis may exhibit similar flexibility in day-roost selection, and we believe more research across the species’ range to examine day-roost selection of all sex and reproductive classes is necessary to better understand how these bats select day roosts throughout the year. In the absence of more regional data, our data show that Eastern Small-footed Myotis heavily use ground-level rock roosts with high solar exposure yet close to cover, and at a larger scale, in close proximity to water during the summer. These results corroborate reports from Kentucky (J. MacGregor, unpubl. data) and support the long-held anecdotal belief that the species relies heavily on rock roosts or similar man-made structures during the summer months (Erdle and Hobson 2001). Our data also indicate that creation or preservation of upland 2011 J.S. Johnson, J.D. Kiser, K.S. Watrous, and T.S. Peterson 105 water sources may be important for habitat management of Eastern Small-footed Myotis; the location of several day-roosts in rock fields within transmission line clearings suggests that habitat for the species can be created or restored in areas where heavy disturbance has occurred. Acknowledgments This project would not have been possible without the help of H. Peters, who contributed hard work and long hours in the field to collect data for this study. Thanks also go to M. Dionne and N. Gikas for additional help in the field. We also thank M. Lacki, S. Pelletier, T. Carter, and an anonymous reviewer who commented on earlier drafts of this manuscript. Literature Cited Aldridge, H.D.J.N., and R.M. Brigham. 1988. Load carrying and maneuverability in an insectivorous bat: A test of the 5% “rule” of radiotelemetry. Journal of Mammalogy 69:379–382. Anthony, E.L.P. 1988. Age determination in bats. Pp. 47–58, In T.H. Kunz (Ed.). Ecological and Behavioral Methods for the Study of Bats. Smithsonian Institution, Washington, DC. 556 pp. Arnett, E.B., and J.P. Hayes. 2009. Use of conifer snags as roosts by female bats in western Oregon. Journal of Wildlife Management 73:214–225. Baker, M.D., and M.J. Lacki. 2006. Day-roosting habitat of female Long-legged Myotis in Ponderosa Pine forests. Journal of Wildlife Management 70:207–215. Barbour, R.W., and W.H. Davis. 1974. Mammals of Kentucky. University Press of Kentucky, Lexington, KY. 286 pp. Best, T.L., and J.B. Jennings. 1997. Myotis leibii. Mammalian Species 547:1–6. Braun, E.L. 1950. Deciduous Forests of Eastern North America. Blakiston Company, Philadelphia, PA. 596 pp. Chruszcz, B., and R.M.R. Barclay. 2002. Thermoregulatory ecology of a solitary bat, Myotis evotis, roosting in rock crevices. Functional Ecology 16:18–26. Cox, D.R., M.J. Lacki, M.D. Baker, and J.S. Johnson. 2008. Sample size and the characterization of roosting habitat of forest-living bats. Bat Research News 49:1–10. Cryan, P.M., M.A. Bogan, and J.S. Altenbach. 2000. Effect of elevation on distribution of female bats in the Black Hills, South Dakota. Journal of Mammalogy 81:719–725. Erdle, S.Y., and C.S. Hobson. 2001. Current status and conservation strategy for the Eastern Small-footed Myotis (Myotis leibii). Virginia Department of Conservation and Recreation, Division of Natural Heritage, Richmond, VA. Natural Heritage Technical Report # 00–19. 34 pp. ESRI. 2008. ArcMap version 9.3. ESRI, Redlands, CA. Fisher, N.I. 1993. Statistical Analysis of Circular Data. Cambridge University Press, Cambridge, UK. 296 pp. Gannon, W.L., R.S. Sikes, and the Animal Care and Use Committee of the American Society of Mammalogists. 2007. Guidelines of the American Society of Mammalogists for the use of wild mammals in research. Journal of Mammalogy 88:809–823. Harvey, M.J., J.S. Altenbach, and T.L. Best. 1999. Bats of the United States. Arkansas Game and Fish Commission, Little Rock, AR. 64 pp. Johnson, J.B., and J.E. Gates. 2008. Spring migration and roost-selection of female Myotis leibii in Maryland. Northeastern Naturalist 15:453–460. Johnson, J.B., J.E. Gates, and W.M. Ford. 2009. Notes on foraging activity of female Myotis leibii in Maryland. Research paper NRS–8. US Department of Agriculture, Forest Service, Northern Research Station, Newton Square, PA. 8 pp. 106 Northeastern Naturalist Vol. 18, No. 1 Kölliker, M., and H. Richner. 2004. Navigation in a cup: Chick positioning in Great Tit, Parus major, nests. Animal Behaviour 68:941–948. Kurta, A., and S.W. Murray. 2002. Philopatry and migration of banded Indiana Bats (Myotis sodalis) and effects of radiotransmitters. Journal of Mammalogy 83:585–589. Lacki, M.J., and M.D. Baker. 2007. Day roosts of female Fringed Myotis (Myotis thysanodes) in xeric forests of the Pacific Northwest. Journal of Mammalogy 88:967–973. Lacki, M.J., M.D. Baker, and J.S. Johnson. In press. Geographic variation in roost-site selection of Long-legged Myotis in the Pacific Northwest. Journal of Wildlife Management. Lausen, C.L. 2007. Roosting ecology and landscape genetics of prairie bats. Ph.D. Dissertation. University of Calgary, Calgary, AB, Canada. 271 pp. Lausen, C.L., and R.M.R. Barclay. 2002. Roosting behaviour and roost selection of female Big Brown Bats (Eptesicus fuscus) roosting in rock crevices in southern Alberta. Canadian Journal of Zoology 80:1069–1076. Lausen, C.L., and R.M.R. Barclay. 2003. Thermoregulation and roost selection by reproductive female Big Brown Bats (Eptesicus fuscus) roosting in rock crevices. Journal of Zoology (London) 260:235–244. National Climatic Data Center (NCDC). 2002. Monthly station normals of temperature, precipitation, and heating and cooling days 1971–2000: West Virginia. National Climatic Data Center, Asheville, NC. 25 pp. NatureServe. 2010. NatureServe Explorer. Arlington, VA. Available online at http:// www.natureserve.org/explorer/. Accessed 24 May 2010. Rabe, M.J., T.E. Morrell, H. Green, J.C. Devos, Jr., and C.R. Miller. 1998. Characteristics of Ponderosa Pine snag roosts used by reproductive bats in northern Arizona. Journal of Wildlife Management 62:612–621. Racey, P.A. 1988. Reproductive assessments in bats. Pp. 21–45, In T.H. Kunz (Ed.). Ecological and Behavioral Methods for the Study of Bats. Smithsonian Institution, Washington, DC. 533 pp. Roble, S.M. 2004. Notes on an autumn roost of an Eastern Small-footed Bat (Myotis leibii). Banisteria 23:42–44. SAS Institute, Inc. 2004. SAS/STAT 9.1 users guide. Volumes 1–7. SAS Institute, Inc., Cary, NC. Solick, D.I., and R.M.R. Barclay. 2006. Thermoregulation and roosting behavior of reproductive and nonreproductive female Western Long-eared Bats (Myotis evotis) in the Rocky Mountains of Alberta. Canadian Journal of Zoology 84:589–599. Speakman, J.R., and D.W. Thomas. 2003. Physiological ecology and energetics of bats. Pp. 430–492, In T. H. Kunz and M. B. Fenton (Eds.). Bat Ecology. University of Chicago Press, Chicago, IL. 798 pp. Tuttle, M.D. 1964. Myotis subulatus in Tennessee. Journal of Mammalogy 45:148–149. Weller, T.J., and C.J. Zabel. 2001. Characteristics of Fringed Myotis day roosts in northern California. Journal of Wildlife Management 65:489–497. Whitaker, J.O., Jr., and W.J. Hamilton, Jr. 1998. Mammals of the Eastern United States. Third Edition. Cornell University Press, Ithaca, NY. 583 pp.