Northeastern Naturalist 184 K. Jansky, B.W. Schubert, and S.C. Wallace 22001166 NORTHEASTERN NATURALIST 2V3(o1l). :2138,4 N–1o9. 41 Geometric Morphometrics of Dentaries in Myotis: Species Identification and Its Implications for Conservation and the Fossil Record Kyle Jansky1,*, Blaine W. Schubert1,2, and Steven C. Wallace1,2 Abstract - Dentaries of 6 species of Myotis from eastern North America were analyzed, using landmark-based geometric morphometrics, and were distinguished with 83.3% accuracy, although sexes were poorly discriminated using this technique. Fossils of Myotis from Bat Cave, KY, were studied in an attempt to identify these specimens to species level. Southeastern Bats and endangered Indiana Bats dominated the fossil sample, with some Eastern Small-footed Bats and endangered Gray Bats. Such results demonstrate the ability to differentiate Myotis from historic and prehistoric sites and provide a tool for researchers to understand and potentially to conserve these species. Introduction Over 100 extant species of Myotis occur worldwide (Simmons 2005), and identification of fossils of Myotis to the level of species historically has been difficult. In many areas, several morphologically similar species occur, and complete skeletons rarely are preserved intact. Many caves, though, contain remains of Myotis that could help illuminate the natural history of individual species and have broader implications for cave paleoecology because bats differ interspecifically in terms of roost preferences (Schubert and Mead 2012). Identifying fossils of Myotis might also contribute to conservation by highlighting caves that were formerly roosting sites for endangered or at-risk species (Toomey et al. 2002). Six species of Myotis live in eastern North America: M. austroriparius (Rhoads) (Southeastern Bat), M. grisescens Howell (Gray Bat), M. leibii (Audubon and Bachman) (Eastern Small-footed Bat), M. lucifugus (Le Conte) (Little Brown Bat), M. septentrionalis (Trouessart) (Northern Long-eared Bat), and M. sodalis Miller and Allen (Indiana Bat). Indiana and Gray Bats are endangered, whereas Northern Long-eared Bats are threatened in the United States (USFWS 1982, 2007, 2015). All are currently being impacted by white-nose syndrome, an exotic fungal disease that has greatly reduced many populations (Frick et al. 2010). Our study is primarily an attempt to develop a procedure for identifying dentaries of Myotis in eastern North America using landmark-based geometric morphometrics, a technique previously employed to distinguish species of Pipistrellus (Sztencel-Jabłonka et al. 2009) and western Myotis (Gannon and Rácz 2006). Any attempt to identify dentaries of Myotis to species presupposes that they can be accurately identified to genus. In North America, only Corynorhinus 1Don Sundquist Center of Excellence in Paleontology, East Tennessee State University, Johnson City, TN 37614. 2Department of Geosciences, East Tennessee State University, Johnson City, TN 37614. *Corresponding author - kjjansky@gmail.com. Manuscript Editor: Allen Kurta Northeastern Naturalist Vol. 23, No. 1 K. Jansky, B.W. Schubert, and S.C. Wallace 2016 185 and Lasionycteris have a lower alveolar formula similar to Myotis (dental formula 3–1–3–3, with a single-rooted p2 and p3 and a double-rooted p4; Czaplewski et al. 2002). Fortunately these genera can easily be distinguished by dental and dentary morphology (Czaplewski et al. 2002, Fauteaux et al. 2014, Gaudin et al. 2011). For instance, Corynorhinus has a more laterally directed angular process, no lingual cingulum on the molar trigonids, a broad talonid, and mandibular foramen typically exposed. Lasionycteris has a lingual cingulum on molar trigonids, a lateral mental foramen that opens between the roots of p2 and p3, wide p2 and p3, narrow talonid, subequal incisors, and a distinct lingual bulge on the hypoconid of p3. In Myotis, the lingual cingulum of molar trigonids is present, and widths of p2 and p3 are subequal; the talonid is narrow, i3 is larger than i1 and i2, and p4 is longer than wide. Attempts to identify the 6 species of Myotis of eastern North America using osteological and dental material have met with partial success. Fauteaux et al. (2014) were able to distinguish among Myotis in eastern Canada, in part because only 3 species occur there. However, identification of fossils based on geographic criteria must be done with caution, because geographic ranges may have changed over time (Bell et al. 2009). Several authors have noted characteristics that distinguish Northern Long-eared Bats from other members of the genus, including strong lingual cingula around the protocones of the upper molars, a rectangular p3 with anteroposterior length greater than labiolingual width, and middle labial cusp of i3 small (Czaplewski et al. 2002, Fauteaux et al. 2014, Gaudin et al. 2011). Gray Bats and Eastern Small-footed Bats can sometimes be distinguished on the basis of size (Fauteaux et al. 2014, Gaudin et al. 2011). Menzel et al. (2005) were able to distinguish among these 6 species using cranial characters, though they did not analyze morphology of the dentary in detail. Our study had several objectives. First, we sought to assess the viability of using landmark-based geometric morphometrics to identify dentaries of the 6 species of Myotis present in eastern North America, and especially the 3 species most difficult to identify (Southeastern Bat, Little Brown Bat, and Indiana Bat). Second, we evaluated whether sexual dimorphism was likely to confound such analyses. Finally, we applied these methods to fossil specimens from Bat Cave, KY, to determine which species were present at the site. Field Site Bat Cave is located in Mammoth Cave National Park, KY, and contains thousands of remains of bats, likely deposited by 1 or more floods (Colburn et al. 2015). Radiocarbon dating of radii of Myotis suggests that the deposits span much of the Holocene epoch, from ~9510 to 2250 years before present (Colburn et al. 2015). Most identifiable material consists of crania, dentaries, and humeri (Colburn et al. 2015, Jegla 1961, Macgregor 1993). Previous studies of fossils from Bat Cave identified Eptesicus fuscus (Palisot de Beauvois) (Big Brown Bat), Perimyotis subflavus (Cuvier) (Tricolored Bat), Corynorhinus sp. (big-eared bat), Eastern Small-footed Bat, Little Brown Bat, Indiana Bat, and a large amount of material, mostly dentaries, classified as Myotis but not assigned to any particular species (Colburn et al. 2015, Jegla 1961). Two crania were identified Northeastern Naturalist 186 K. Jansky, B.W. Schubert, and S.C. Wallace 2016 Vol. 23, No. 1 as probable Little Brown Bats, 4 crania as probable Indiana Bats, and 6 dentaries were assigned to the Eastern Small-footed Bat based on small size (Colburn et al. 2015). A number of species also were collected from or observed to live in Bat Cave, including Southeastern Bats, Gray Bats, Little Brown Bats, Northern Long-eared Bats, Indiana Bats, Tricolored Bats, Big Brown Bats, and Corynorhinus rafinesquii (Lesson) (Rafinesque’s Big-Eared Bat) (Colburn et al. 2015; Smithsonian Institution National Museum of Natural History collections records,Washington, DC). Materials and Methods Adult bats (n = 224) from 6 species in the genus Myotis were included in various statistical analyses. Southeastern Bats (n = 29; 10 male, 18 female, 1 unknown), Gray Bats (n = 40; 23 male, 16 female, 1 unknown), Eastern Small-footed Bats (n = 37; 24 male, 10 female, 3 unknown), Little Brown Bats (n = 38; 21 male, 15 female, 2 unknown), Northern Long-eared Bats (n = 39; 28 male, 8 female, 3 unknown), and Indiana Bats (n = 39; 15 male, 18 female, 6 unknown) were studied because these are the only species of Myotis that occur in eastern North America. These modern specimens were from the Smithsonian Institution National Museum of Natural History, whereas fossils from Bat Cave that were analyzed (n = 48) were curated at the Illinois State Museum. We selected the dentary for this project because it is among the most common elements in the fossil record of North American Myotis (e.g., Czaplewski and Peachey 2003), and previous studies using geometric morphometrics to distinguish among dentaries of other taxa of bats have met with success (Gannon and Rácz 2006, Sztencel-Jabłłonka et al. 2009). We used 17 landmarks, most of which were tips of processes and the posterior margins of alveoli (Fig. 1, Table 1). We omitted dental characteristics because teeth often disarticulate from dentaries post-mortem. We photographed dentaries in labial view, with the orientation standardized by ensuring that the tip of the coronoid process, tip of the angular process, and distal tip of the mandible were on a single plane, perpendicular to the photographic angle. We then used TPSUtil to convert the photographs into thin-plate-spline (TPS) files Figure 1. Dentary of Myotis, showing the landmarks used in this project. Northeastern Naturalist Vol. 23, No. 1 K. Jansky, B.W. Schubert, and S.C. Wallace 2016 187 Table 1. Definitions of landmarks used. Landmark Definition 1 Tip of coronoid process: point of maximum curvature 2 Angle of mandibular condyle and coronoid process: point of maximum curvature 3 Tip of mandibular condyle: point of maximum curvature 4 Angle of mandibular condyle and angular process: point of maximum curvature 5 Tip of angular process: point of maximum curvature 6 Most ventral point of mandibular symphysis: point of maximum curvature 7 Ventral angle of masseteric fossa: point of maximum curvature 8 Posterior of the base of m3 (at the alveolus) 9 Posterior of the base of m2 (at the alveolus) 10 Posterior of the base of m1 (at the alveolus) 11 Posterior of the base of p3 (at the alveolus) 12 Posterior of the base of p2 (at the alveolus) 13 Posterior of the base of p1 (at the alveolus) 14 Posterior of the base of c (at the alveolus) 15 Posterior of the base of i3 (at the alveolus) 16 Posterior of the base of i2 (at the alveolus) 17 Posterior of the base of i1 (at the alveolus) (Rohlf 2010a). Landmarks were digitized with tpsDIG2 (Rohlf 2010b) and transformed into shape variables by a Procrustes superimposition, using TPSSUPER (Rohlf 2004). We used SPSS (IBM Corporation 2012) to carry out principal component analyses (PCA) and discriminant function analyses (DFA). Discriminant analyses analyzed the shape variables, not principle components, and results were cross-validated We analyzed dentaries from the 6 species of Myotis from eastern North America to determine whether landmark-based geometric morphometric analysis is a viable method for identifying these species. We also analyzed these dentaries for sexual dimorphism using DFA. Specimens of unknown sex were included in analyses as unknowns, so superimposition would be based on a more robust sample size. Restricting analyses to certain species of Myotis introduced several assumptions. We assumed the fossil dentaries belonged to 1 of the 6 species of Myotis that currently occur in eastern North America (i.e., the fossils did not belong to an extinct species or a species today restricted to another geographic region) and further assumed that mandibular morphology had not substantially changed since deposition of the fossils and that modern specimens have been correctly identified to species. We included specimens from Bat Cave as unknowns to identify species present in the fossil record. To allow for the possibility that fossils might have violated the above assumptions, we classified unknowns as none of the species if their typicality probabilities fell outside the 95% confidence interval around the mean for the most likely species, a procedure suggested by Kovarovic et al. (2011). Variants of this analysis were also performed excluding fossils, but these results did not differ substantially from those reported here. Early results indicated neither Little Brown Bats nor Northern Long-eared Bats were major components of the fauna of Bat Cave (i.e., no fossils were classified as either species), so we repeated the analyses excluding these 2 species to determine more accurately which remaining species Northeastern Naturalist 188 K. Jansky, B.W. Schubert, and S.C. Wallace 2016 Vol. 23, No. 1 Table 2. Discriminant analysis of the 6 Myotis of eastern North America, with fossils from Bat Cave treated as unknowns. For each species, the number of dentaries assigned is followed by the percent in parentheses. Values reported for specimens of known species are cross-validated. Fossils were classified as “none” if the typicality value for the most likely species was below 0.05. Specimens of known species were correctly classified in 83.3% of cases (93.2% when not cross-validated). Predicted Species Eastern Northern Small- Little Long- Bat species Southeastern Gray footed Brown eared Indiana None Total Southeastern 21 (72.4) 4 (13.8) 0 (0) 4 (13.8) 0 (0) 0 (0) 29 Gray 5 (12.5) 32 (80) 0 (0) 2 (5) 0 (0) 1 (2.5) 40 Eastern Small-footed 0 (0) 0 (0) 34 (91.9) 1 (2.7) 0 (0) 2 (5.4) 37 Little Brown 1 (2.6) 3 (7.9) 1 (2.6) 29 (76.3) 0 (0) 4 (10.5) 38 Northern Long-eared 0 (0) 0 (0) 0 (0) 0 (0) 38 (97.4) 1 (2.6) 39 Indiana 1 (2.6) 1 (2.6) 2 (5.1) 3 (7.7) 1 (2.6) 31 (79.5) 39 Fossils (species 16 (33.3) 2 (4.2) 1 (2.1) 0 (0) 0 (0) 12 (25) 17 (35.4) 48 unknown) were present. Analysis excluding these 2 species was intended to supplement, not supersede, the primary analysis of the fossils. Southeastern Bats, Little Brown Bats, and Indiana Bats are difficult to distinguish based on mandibular characteristics (Gaudin et al. 2011) and are often grouped together as simply “medium-sized Myotis” (Colburn et al. 2015, Toomey et al. 2002). For this reason, and because DFA is more successful if fewer groups are analyzed, we also performed a DFA of these 3 species alone to assess the viability of distinguishing them with landmark-based geometric morphometrics. Results A PCA of the 6 species of Myotis of eastern North America and the fossils exhibited a gradient along the first principal component, with Northern Long-eared Bat on 1 side, Eastern Small-footed Bat, Little Brown Bat, and Indiana Bat in the middle, and Southeastern Bat and Gray Bat on the other side (Fig. 2). Most fossils clustered together and scored along the middle-to-high end of the first principal component and positively along the second component. Some fossils extended beyond the morphospace occupied by known species. Overall, 83.3% of the 224 modern specimens were correctly identified by DFA (Table 2). Correct identification for individual species ranged from 72.4% (Southeastern Bat) to 97.4% (Northern Long-eared Bat), and the number of false positives ranged from 1 (2.5%) for Northern Long-eared Bat to 10 (25.6%) for the Little Brown Bat. Of the 48 fossils, 16 (33.3%) were classified as Southeastern Bats, 12 (25%) as Indiana Bats, 2 (4.2%) as Gray Bats, and 1 (2.1%) as an Eastern Small-footed Bat. No fossils were assigned to the Little Brown Bat or Northern Long-eared Bat. Seventeen fossils (35.4%) were classified as none of the 6 species because the typicality probabilities for their most likely species were below 0.05. When we reran the analysis excluding Little Brown Bat and Northern Longeared Bat, the remaining 4 species were correctly classified 86.2% of the time Northeastern Naturalist Vol. 23, No. 1 K. Jansky, B.W. Schubert, and S.C. Wallace 2016 189 Figure 2. First 2 axes of a principal component analysis of the dentaries of the 6 species of Myotis from eastern North America and fossils from Bat Cave, KY. PC 1 explained 22.8% of the variance, and PC 2 explained 14.4%. Figure 3. First 2 axes of a discriminant function analysis of the dentaries of the Southeastern Bat, Gray Bat, Eastern Small-footed Bat, and Indiana Bat, treating fossils from Bat Cave, KY, as unknowns. DF 1 explained 64.8% of the variance, and DF 2 explained 23.4%. Northeastern Naturalist 190 K. Jansky, B.W. Schubert, and S.C. Wallace 2016 Vol. 23, No. 1 Table 3. Results of discriminant analysis of the Southeastern Bat, Little Brown Bat, and Indiana Bat. For each species the number of dentaries assigned is followed by the percent in parentheses. Values reported are cross-validated. Specimens were correctly classified in 82.1% of cases (97.2% when not cross-validated). Predicted Species Species Southeastern Bat Little Brown Bat Indiana Bat Total Southeastern Bat 24 (82.8) 4 (13.8) 1 (3.4) 29 Little Brown Bat 4 (10.5) 29 (76.3) 5 (13.2) 38 Indiana Bat 1 (2.6) 4 (10.3) 34 (87.2) 39 (Fig. 3). Nineteen fossils (39.6%) were assigned as Southeastern Bat, 11 (22.9%) as Indiana Bat, 4 (8.3%) as Gray Bat, 2 (4.2%) as Eastern Small-footed Bat, and 12 (25%) as none of these 6 species. The first PCA axis separated Eastern Small-footed Bats from Southeastern Bats and Gray Bats, whereas Indiana Bats were intermediate. Indiana Bats largely separated from the other 3 species along the second axis. Fossils mostly clustered between Indiana Bats and the Southeastern Bat/Gray Bat group, with only a few specimens plotting close to Eastern Small-footed Bats. A DFA of 3 medium-sized species of Myotis, a group previously difficult to differentiate in the fossil record, correctly classified 82.1% of individuals (Table 3). Five Southeastern Bats were misidentified, and there were 5 false positives. The number of misidentifications and false positive were 9 and 8, respectively, for Little Brown Bats, and 5 and 6, respectively, for Indiana Bats. Analyses of sexual dimorphism typically did not distinguish sexes substantially better than chance. The only exception was the Northern Long-eared Bat, which had the least-balanced sex ratio, with 28 males and 8 females. The sexes in this species were discriminated with 69.4% accuracy. Discussion Although Northern Long-eared Bats exhibited some sexual dimorphism, the analysis was based on an unbalanced ratio of males to females; thus, these results must be interpreted with caution. Because the other 5 species were not sexually dimorphic, landmark-based geometric morphometrics is unlikely to be useful for determining sex ratios of fossil populations of Myotis. Nevertheless, our results suggest that sex was unlikely to be a major confounding factor in other analyses. The main goal of our research was to devise a method of identifying dentaries of Myotis to species. DFA correctly identified 83.3% of specimens, a success rate high enough that general conclusions may be drawn regarding species composition of a sample of fossils. That is, although individual identifications must be treated with caution, if many fossils are attributed to the same species, then there is a high probability that the species in question was present in the fauna. Menzel et al. (2005) discriminated among these 6 species with a 99.4% success rate by incorporating both cranial measurements and external body measurements recorded on specimen tags. Using only cranial dimensions, they had a success rate of 96.9% (Menzel et al. 2005). Our analysis relied on the dentary alone and Northeastern Naturalist Vol. 23, No. 1 K. Jansky, B.W. Schubert, and S.C. Wallace 2016 191 provides a viable alternative for situations in which an adequate sample of cranial material is not available. Furthermore, our method identified certain species (i.e., Eastern Small-footed Bat and Northern Long-eared Bat) in over 90% of cases, and if some species can be ruled out in advance, analysis of remaining species is more effective. In particular, Gray Bats often can be distinguished based on size alone (Gaudin et al. 2011), and if size were combined with our methods, Gray Bats probably could be recognized reliably. We initially planned to incorporate size into our study, but were unable to do so because the program used to append scale bars to our digital photographs provided erroneous measurements, a fact not noticed until after data had been collected. Once other species have been removed (either a priori or with a preliminary DFA), Southeastern Bats, Little Brown Bats, and Indiana Bats—species previously difficult to distinguish—were recognized with 82.1% accuracy. The endangered Indiana Bat was identified with 87.2% accuracy, with only 5 false positives, and this success rate could have important applications for conservation. If a fossil population is analyzed and a significant proportion of specimens are assigned to this species, Indiana Bats likely were indeed present. If the bats have disappeared due to recent alteration or disturbance to their roost, steps might be taken to restore the site to earlier conditions, in hopes that a colony would be reestablished (Toomey et al. 2002). Such restoration could prove important, because only 23 hibernacula serve approximately 82% of all Indiana Bats (USFWS 2007). Restoring altered caves has proven a viable conservation method for Gray Bats (Decher and Choate 1995). Discriminant analysis of material from Bat Cave strongly indicated presence of Southeastern Bats and Indiana Bats, with some support for presence of Gray Bats and Eastern Small-footed Bats. Many fossils had extremely low typicality probabilities for their most likely species, suggesting that they did not belong to any species of Myotis found in the region today. No evidence of Little Brown Bats or Northern Long-eared Bats was found. Though this did not prove absence of these species, our results implied that neither species contributed substantially to the assemblage. It is notable that our analysis did not identify any fossils as Little Brown Bats since previous studies attributed some cranial material to this taxon (Colburn et al. 2015), and it is the most common species at the cave today; typically 200–300 Little Brown Bats hibernated there each year before the appearance of white-nose syndrome, though only 97–161 animals have overwintered there annually since 2009 (R. Toomey, Director, Mammoth Cave International Center for Science and Learning, Mammoth Cave National Park, Mammoth Cave, KY, 2015 pers. comm.). Our results suggested that the predominance of Little Brown Bats at Bat Cave today was a recent development. Absence of Northern Long-eared Bats was less surprising, because few have been observed at the cave in modern times (R. Toomey, 2015 pers. comm.). Once the Little Brown Bat and Northern Long-eared Bat were removed from the analysis, relative proportions of the other 4 species (Southeastern Bat, Gray Bat, Eastern Small-footed Bat, and Indiana Bat) remained largely unchanged. Most fossils appeared to be Southeastern Bats and Indiana Bats. The large proportion of Northeastern Naturalist 192 K. Jansky, B.W. Schubert, and S.C. Wallace 2016 Vol. 23, No. 1 Southeastern Bats (39.6%) was intriguing. This species had not previously been identified among the fossils and had rarely been encountered in the hibernaculum (Colburn et al. 2015, Hall 1961). Nevertheless, this result was consistent with previous attribution of many fossils from Bat Cave to “medium-sized Myotis” (Colburn et al. 2015:94). Presence of Indiana Bats was expected. Indiana Bats have consistently been present in the cave, with a population comparable to that of Little Brown Bats, though their numbers have decreased to 30–40 animals in recent years (R. Toomey, 2015 pers. comm.). Additionally, several crania from Bat Cave have been identified as Indiana Bats (Colburn et al. 2015). Consequently, our results provide additional support for long-term use of Bat Cave as a hibernaculum by this endangered species. Apparent presence of Gray Bats and Eastern Small-footed Bats should be interpreted cautiously because our analyses were less than 100% successful at distinguishing among species. However, a bachelor colony of Gray Bats roosts in the cave during summer (Colburn et al. 2015), and several dentaries from the deposit have previously been identified as Eastern Small-footed Bats due to their small size (Colburn et al. 2015). Thus, our detection of these species among the fossils at Bat Cave is consistent with previous research. When all 6 Myotis from eastern North America were analyzed, 35.4% of specimens fell below the 0.05 cutoff for typicality probabilities; this value dropped to 25% when Little Brown Bats and Northern Long-eared Bats were excluded. That so many specimens fell outside the 95% confidence interval for their assigned species strongly suggested that 1 or more groups not represented among the modern specimens that we included in our analyses were present at Bat Cave in the past. Possible explanations included presence of an extinct species, presence of an extant species no longer found in eastern North America, and change in the shape of the dentary since the fossils were deposited, though we could not assess the relative probability of these hypotheses. Most specimens with typicality probabilities below 0.05 were assigned to either Southeastern Bat or Indiana Bat. Thus, future research may revise relative proportions of these species within the fossil deposit because other specimens may fall within the 95% confidence interval for these species but be closer to the group centroid of a species not represented in our study. Similarity to these 2 species could also be used as a starting point for design of future studies to assess fossils from Bat Cave (i.e., by suggesting inclusion of species likely to be morphologically similar to these species but not currently found in the region). From our results, it seems that Bat Cave has long been home to several species of bats, though at proportions differing from today. Specifically, the Southeastern Bat and Indiana Bat appear to have had substantial colonies at Bat Cave in the past. One or more species of Myotis no longer present in eastern North America (whether due to extinction or extirpation) may have made substantial contributions to the fossil deposit. A small number of Gray Bats and Eastern Small-footed Bats may also have been present, along with a few Big Brown Bats, Tricolored Bats, and Corynorhinus sp. (Colburn et al. 2015, Macgregor 1993). Although Indiana Bats continue to use the cave today, Southeastern Bats have largely abandoned this site. At some point, Northeastern Naturalist Vol. 23, No. 1 K. Jansky, B.W. Schubert, and S.C. Wallace 2016 193 possibly after deposition of the fossils, a small colony of Little Brown Bats became established. Explaining these inferred changes would be an interesting subject for future research. Acknowledgments We thank the Smithsonian Institution National Museum of Natural History, especially S. Peurach, for access to specimens and general assistance. Thanks are also due to the Illinois State Museum for permitting study of fossils from Bat Cave, and particularly to C. Widga for photographing those specimens. The research of R. Toomey and M. Colburn provided much background information. J. Mead, C. Jansky, and A. Cardini provided helpful feedback on an earlier version of this manuscript, and D. Jansky helped prepare the figures. This research was supported in part by the National Science Foundation (EAR-0958985) and East Tennessee State University through the Office of Research and Sponsored Programs and the Don Sundquist Center of Excellence in Paleontology. Literature Cited Bell, C.J., J.A. Gauthier, and G.S. Bever. 2009. Covert biases, circularity, and apomorphies: A critical look at the North American Quaternary Herpetofaunal Stability Hypothesis. Quaternary International 217:30–36. Colburn, M., R. Toomey, C. Widga, and R. Olson. 2015. Holocene paleontology of Bat Cave, Edmonson County, Kentucky. Journal of Cave and Karst Studies 77:91–98. Czaplewski, N.J., and W.D. Peachey. 2003. Late Pleistocene bats from Arkenstone Cave, Arizona. Southwestern Naturalist 48:597–609. Czaplewski, N.J., W.L. Puckette, and C. Russel. 2002. A Pleistocene tapir and associated mammals from the southwestern Ozark Highland. Journal of Cave and Karst Studies 64:97–107. Decher, J., and J.R. Choate. 1995. Myotis grisescens. Mammalian Species 510:1–7. Fauteaux, D., G. Lupien, F. Fabianek, J. Gagnon, M. Séguy, and L. Imbeau. 2014. An illustrated key to the mandibles of small mammals of eastern Canada. Canadian Field- Naturalist 128:25–37. Frick, W.F., J.F. Pollock, A.C. Hicks, K.E. Langwig, D.S. Reynolds, G.G. Turner, C.M. Butchkoski, and T.H. Kunz. 2010. An emerging fungal disease causes regional population collapse of a common North American bat species. Science 329:679–682. Gannon, W.L., and G.R. Rácz. 2006. Character displacement and ecomorphological analysis of two long-eared Myotis (M. auriculus and M. evotis). Journal of Mammalogy 87:171–179. Gaudin, T.J., A.N. Miller, J.L. Bramblett, and T.P. Wilson. 2011. Holocene and Late Pleistocene bat fossils (Mammalia: Chiroptera) from Hamilton County, TN, and their ecological implications. Southeastern Naturalist 10:609–628. Hall, J.S. 1961. Myotis austroriparius in central Kentucky. Journal of Mammalogy 42:399–400. IBM Corporation. 2012. IBM SPSS Statistics 21 Core System User’s Guide. IBM Corporation, Armonk, NY. 426 pp. Jegla, T.C. 1961. A study on deposits of bat bones in Mammoth Cave National Park. Upubl. Report. Mammoth Cave National Park, Mammoth Cave, KY. 26 pp. Kovarovic, K., L.C. Aiello, A. Cardini, and C.A. Lockwood. 2011. Discriminant function analyses in archaeology: Are classification rates too good to be true? Journal of Archaeological Science 38:3006–3018. Northeastern Naturalist 194 K. Jansky, B.W. Schubert, and S.C. Wallace 2016 Vol. 23, No. 1 MacGregor, J. 1993. Responses of winter populations of the federal [sic] endangered Indiana Bat (Myotis sodalis) to cave gating in Kentucky. Pp. 364–370, In Foster, D.L., D.G. Foster, M.M. Snow, and R.K. Snow (Eds.). 1991 National Cave Management Symposium Proceedings. American Cave Conservation Association, Horse Cave, KY. 405 pp. Menzel, M., S. Castleberry, J. Boone, M. Haag, and J. Menzel 2005. Mensural discrimination of Myotis (Chiroptera, Vespertilionidae) in the southeastern United States. Mammalia 69:359–373. Rohlf, F.J. 2004. TPSSuper (version 1.14). Department of Ecology and Evolution, State University of New York at Stony Brook, Stony Brook, NY. Rohlf, F.J. 2010a. TPSUtil (version 1.46). Department of Ecology and Evolution, State University of New York at Stony Brook, Stony Brook, NY. Rohlf, F.J. 2010b. tpsDIG2 (version 2.16). Department of Ecology and Evolution, State University of New York at Stony Brook, Stony Brook, NY. Schubert, B.W., and J.I. Mead. 2012. Paleontology of caves. Pp. 590–598, In W.B. White and D.C. Culver (Eds.). Encyclopedia of Caves. Elsevier, Burlington, MA. 945 pp. Simmons, N.B. 2005. Order Chiroptera. Pp. 312–529, In D.E. Wilson and D.M. Reeder (Eds.). Mammal Species of the World: A Taxonomic and Geographic Reference. Vol. 1. 3rd Edition. Johns Hopkins University Press, Baltimore, MD. 744 pp. Sztencel-Jabłonka, A., G. Jones, and W. Bogdanowicz. 2009. Skull morphology of two cryptic bat species: Pipistrellus pipistrellus and P. pygmaeus—a 3D geometric morphometrics approach with landmark reconstruction. Acta Chiropterologica 11:113–126. Toomey, R.S., M.L. Colburn, and R.A. Olson. 2002. Paleontological evaluation of use of caves: A tool for restoration of roosts. Pp. 79–85, In A. Kurta and J. Kennedy (Eds.). The Indiana Bat: Biology and Management of an Endangered Species. Bat Conservation International, Austin, TX. 253 pp. US Fish and Wildlife Service (USFWS). 1982. Gray Bat Recovery Plan. Available online at http://pbadupws.nrc.gov/docs/ML1214/ML12146A326.pdf. Accessed 3 October 2015. USFWS. 2007. Indiana Bat (Myotis sodalis) Draft Recovery Plan: First Revision. Available online at http://ecos.fws.gov/docs/recovery_plan/070416.pdf. Accessed 3 October 2015. USFWS. 2015. Threatened species status for the Northern Long-eared Bat with 4(d) rule. Available online at http://ecos.fws.gov/speciesProfile/profile/speciesProfile. action?spcode=A0JE. Accessed 7 June 2015.