nena masthead
SENA Home Staff & Editors For Readers For Authors

Roosts of Rafinesque’s Big-Eared Bats and Southeastern Myotis in East Texas
Leigh A. Stuemke, Christopher E. Comer, Michael L. Morrison, Warren C. Conway, and Ricky W. Maxey

Southeastern Naturalist, Volume 13, Special Issue 5 (2014): 159–171

Full-text pdf (Accessible only to subscribers.To subscribe click here.)

 



Access Journal Content

Open access browsing of table of contents and abstract pages. Full text pdfs available for download for subscribers.

Issue-in-Progress: Vol. 23 (1) ... early view

Current Issue: Vol. 22 (3)
SENA 22(3)

Check out SENA's latest Special Issue:

Special Issue 12
SENA 22(special issue 12)

All Regular Issues

Monographs

Special Issues

 

submit

 

subscribe

 

JSTOR logoClarivate logoWeb of science logoBioOne logo EbscoHOST logoProQuest logo


159 Roosts of Rafinesque’s Big-Eared Bats and Southeastern Myotis in East Texas Leigh A. Stuemke1, Christopher E. Comer1,*, Michael L. Morrison2, Warren C. Conway1, and Ricky W. Maxey3 Abstract - Because diurnal roosts can be important in determining bat occupancy and abundance in forested habitats, we identified characteristics of cavity trees that influence roost selection by Corynorhinus rafinesquii (Rafinesque’s Big-eared Bats) and Myotis austroriparius (Southeastern Myotis) in east Texas. We identified used and non-used cavity trees with a combination of transect searches, radiotelemetry, and historical records at 7 study areas. Both bat species selected similar cavity trees for summer diurnal roosts, showing an affinity for tupelo trees (Nyssa spp.), with 55% of diurnal roosts in Nyssa aquatica (Water Tupelo) and 33% in N. sylvatica (Blackgum). Of 17 tree and habitat variables we measured at used and unused cavity trees, those related to cavity size and availability (cavity height and diameter, tree diameter, density of large trees in the area) were the most important predictors of use. Characteristics of the surrounding stand at both local and landscape scales were less important. Rafinesque’s Big-eared Bats and Southeastern Myotis appeared to use the largest cavity trees present and we speculate that the availability of suitable trees with large cavities may limit abundance in this region. Introduction Suitable diurnal roosts are vital for the success and persistence of all bat species. Diurnal roosts provide protection from predators and ambient environmental conditions, and aid in reducing energetic costs associated with parturition and thermoregulation (Barclay and Kurta 2007). Within forested systems, bats utilize a variety of structures for diurnal roosts, including tree foliage, exfoliating bark, cracks or crevices in the tree, and internal cavities (Brigham 2007). Due to the importance of diurnal roosts and the relative ease of studying bats in their roosts, roost selection has been extensively studied for some species; however, much of the research been focused on certain species (e.g., Myotis sodalis [Indiana Bat]) or certain regions (e.g., the Pacific Northwest) (Brigham 2007). Because roost characteristics are often very species-specific, local information on individual species’ needs is needed to formulate effective plans for conservation. Corynorhinus rafinesquii Lesson (Rafinesque’s Big-eared Bat) and Myotis austroriparius Rhoads (Southeastern Myotis) are closely associated with bottomland hardwood forests and forested swamps throughout their ranges in the southeastern 1Arthur Temple College of Forestry and Agriculture, Stephen F. Austin State University, Box 6109 SFA Station, Nacogdoches, TX 75962. 2Department of Wildlife and Fisheries Sciences, Texas A&M University, 210 Nagle Hall, College Station, TX 77843-2258. 3Texas Parks and Wildlife Department, PO Box 226, Karnack, TX 75661. *Corresponding author - comerce@sfasu.edu. Manuscript Editor: Jerry Cook Proceedings of the 5th Big Thicket Science Conference: Changing Landscapes and Changing Climate 2014 Southeastern Naturalist 13(Special Issue 5):159–171 Southeastern Naturalist L.A. Stuemke, C.E. Comer, M.L. Morrison, W.C. Conway, and R.W. Maxey 2014 160 Vol. 13, Special Issue 5 United States (Clark 2003, Lance et al. 2001, Mirowsky 1998). Although habitat associations are poorly understood, this association is believed to be related primarily to their preference to roost in large, hollow trees of certain bottomland hardwood species. In East Texas, natural roosts occur in mature to over-mature hardwood trees that have formed cavities with basal and/or top openings in a variety of species, including Nyssa sylvatica Marshall (Blackgum), N. aquatica L. (Water Tupelo), Magnolia grandiflora L. (Southern Magnolia), Fagus grandifolia Ehrh. (American Beech), Platanus occidentalis L. (Sycamore), and Taxodium distichum (L.) (Baldcypress) (Clark 1990, Gooding and Langford 2004, Lance et al. 2001, Mirowsky et al. 2004, Trousdale and Beckett 2005). In most cases, Rafinesque’s Big-eared Bats used multiple roosts (i.e., roost switching) in close proximity (e.g., <1 km) within a given season (Stevenson 2008, Trousdale and Beckett 2005,). The use of multiple roosts may be related to site, stand, and/or landscape characteristics that minimize energetic costs of survival (Lewis 1995). In addition to tree roosts, these bats also utilize caves and various anthropogenic structures (e.g., abandoned buildings, bridges, wells, culverts, and artificial roosts) for roosting in the southeastern coastal plain ( Clark 1990, Hoffmeister and Goodpaster 1962, Lance et al. 2001). Many factors related to individual trees, forest stands, and forested landscapes influence selection of diurnal roost sites for various species of forest bats (Barclay and Kurta 2007). For example, tree height, tree diameter at breast height (DBH), percent canopy cover, density of surrounding vegetation, stand age, and proximity to water influence tree roost selection by forest bats (Crampton and Barclay 1998, Jung et al. 2004, Ober and Hayes 2008, Willis and Brigham 2005). Due to the perceived importance of diurnal roosts in the distribution of our target species, recent research activity has defined natural and artificial roost requirements at small spatial extents, including characteristics of the roost itself or the immediately surrounding forest stand. Roosts that were spacious and partially lit (Lance et al. 2001) and within close proximity to water (<1 km) were selected more often by both species (Mirowsky et al. 2004). Rafinesque’s Big-eared Bats generally chose larger trees (82–124 cm DBH) than Southeastern Myotis (76–108 cm DBH) as diurnal roosts (Carver and Ashley 2008, Gooding and Langford 2004, Rice 2009). In Texas, areas with historic records of occurrence consisted of relatively large (>100 ha) patches of contiguous forest. These patches generally consisted of >30% bottomland hardwood forest with at least a portion in cypress-tupelo forest (Mirowsky et al. 2004; Meg Goodman, Texas Parks and Wildlife Department, Austin, TX, unpubl. data). Roosts for both species occurred in landscapes with a minimum of 1/3 bottomland hardwood forest and a prevalence of large Baldcypress or Water Tupelo. Habitat surrounding bottomland forest stands used by Rafinesque’s Big-eared Bats and Southeastern Myotis can be quite variable and may consist of mixed pine-hardwood stands, upland pine forest, pine plantations, and urban interfaces (Carver and Ashley 2008, Gooding and Langford 2004, Trousdale et al. 2008). In Texas, the Rafinesque’s Big-eared Bat is a threatened species with a high conservation priority based on perceived population decline (Bender et al. 2005, Mirowsky et al. 2004). The Southeastern Myotis is considered rare but is not Southeastern Naturalist 161 L.A. Stuemke, C.E. Comer, M.L. Morrison, W.C. Conway, and R.W. Maxey 2014 Vol. 13, Special Issue 5 afforded any state listing; however, the Texas Wildlife Action Plan lists this bat as a high priority for conservation (Bender et al. 2005). Much of the concern related to these species is due to the loss of up to 75% of original bottomland hardwood forests in the region. Many of the remaining bottomland hardwood stands were logged in the middle to late 20th century, and perceived low abundance may be related to limited availability of trees suitable for roosting. Within East Texas, data about roost requirements are limited to a single study performed in the mid-1990s (Mirowsky 1998). Recent data in the region are limited to semiannual monitoring of known roosts for both species. Our objectives in this study were to locate additional roost trees and identify factors to differentiate between used and non-used cavity trees for Rafinesque’s Big-eared Bats and Southeastern Myotis. The results provide additional information on roosting ecology in the region that may assist with prioritizing conservation actions for these species. Study Area Rafinesque’s Big-eared Bat and Southeastern Myotis reach the western extent of their ranges in the Pineywoods eco-region of east Texas. This eco-region occurs in the Gulf Coastal Plain physiographic region (Nixon 2000) and extends from the Red River along the northern border of East Texas southward to the northern suburbs of Houston. The Pineywoods covers 6.3 million ha in East Texas and is an extension of the pine, mixed hardwood and bottomland hardwood forests of the southeastern United States (Diggs et al. 2006). Topography of the area is mostly flat with low rolling hills and elevation ranging from 15–230 m. Average annual precipitation across the region varies from 89–127 cm and, combined with hot summers (mean high temperature 24–25 °C) and mild winters (mean low temperature 11–12 °C), produces a long growing season of 220–270 days, depending on location (Nixon 2000). We selected 7 study areas within the Pineywoods eco-region based on historic occurrence records for our target species, habitat conditions, and accessibility: Caddo Lake National Wildlife Refuge (CLNWR: Harrison County, 3440 ha), Caddo Lake State Wildlife Management Area (CLWMA: Marion County, 3240 ha), Little Sandy National Wildlife Refuge (LSNWR: Wood County, 1538 ha), Big Thicket National Preserve (BTNP: Hardin County, 42,770 ha), Trinity River National Wildlife Refuge (TRNWR: Liberty County, 10,117 ha), The Nature Conservancy’s Roy E. Larsen Sandyland Sanctuary (TNC: Hardin County, 2250 ha), and Village Creek State Park (VCSP: Hardin County, 441 ha) (Fig. 1). Among the 7 sites, ecological communities were variable, but the basic habitat structure that supports Rafinesque’s Big-eared Bat and Southeastern Myotis populations was present: cypress and tupelo swamps, bottomland hardwood forests, and mixed deciduous/ pine upland forests. Dominant overstory species for these areas included: Liquidambar styraciflua L. (Sweetgum), Southern Magnolia, Water Tupelo, Blackgum, Pinus taeda L. (Loblolly Pine), Quercus lyrata Walter (Overcup Oak), Q. michauxii Nutt. (Swamp Chestnut Oak), Q. nigra L. (Water Oak), Q. phellos L. (Willow Oak), Baldcypress, Ulmus americana L. (American Elm), and Southeastern Naturalist L.A. Stuemke, C.E. Comer, M.L. Morrison, W.C. Conway, and R.W. Maxey 2014 162 Vol. 13, Special Issue 5 U. crassifolia Nutt. (Cedar Elm). Dominant midstory species included Cephalanthus occidentalis L. (Buttonbush), Forestiera ligustrina (Michx.) Poir. (Swamp Privet), Fraxinus pennsylvanica Marshall (Green Ash), and Planera aquatica J.F. Gmel. (Water Elm). Figure 1. Study area locations for Rafinesque’s Big-eared Bat and Southeastern Myotis roost studies conducted in East Texas, 2008–2009. Southeastern Naturalist 163 L.A. Stuemke, C.E. Comer, M.L. Morrison, W.C. Conway, and R.W. Maxey 2014 Vol. 13, Special Issue 5 Methods Bat surveys We used standardized transect searches and radiotelemetry to locate cavity trees used as diurnal roosts by the target species. In addition, we opportunistically identified a small number of roosts during other research activities and recorded data at used cavity roost-trees identified by historical records. A tree was classified as used when one or more individual(s) of our target species were documented inside the tree. Field surveys were conducted during the season of high bat activity: 12 May–15 August 2008, and 19 May–13 August 2009. We located potential roost trees by visual observation along transects located in 100-ha study blocks randomly selected from our 7 study areas. We overlaid systematic block-sampling grids (100 ha) onto aerial photos of each study area using ArcGIS 9.3 (Environmental Systems Research Institute, Redlands, CA), then used a random number generator to select blocks for transect searches. The block size of 100 ha was selected to match the estimated homerange size of the Rafinesque’s Big-eared Bat (Hurst and Lacki 1999, Menzel et al. 2001). There were 4 study blocks at LSNWR, 4 study blocks at CLNWR, 4 study blocks at CLWMA, 2 study blocks at TRNWR, 3 study blocks at BTNP, 2 study blocks at TNC, and 1 study block at VCSP for a total of 20 study blocks. We randomly located 10 1-km-long by 40-m-wide transects in each study block, and searched all trees within the transect boundary to identify potential cavity openings. If we noted a potential cavity, we investigated for signs of bat presence. Depending on the characteristics of the tree and cavity opening, we used either direct observation with Surefire 9p flashlights (105 lumens) with red filters (SureFire, LLC, Fountain Valley, CA), or hand-held mirrors and flashlights to inspect tree cavities. If the tree cavity possessed a bend or some other hindrance that prevented visual inspection, we used acoustic detectors (Pettersson 240X, Pettersson Elektronik AB, Uppsala, Sweden) to determine if bats were using the tree. Identification of species was made through visual observation, photographic evidence, or acoustic analysis of call characteristics. We used both our own site-specific and freely available call libraries to identify calls to species using Sonobat 2.5.9 (SonobatTM, Arcata, CA). If target bat species were not present in a cavity tree at the time of survey and the tree was ≥30 cm DBH (Clark 2003), the tree was classified as a non-used cavity tree for our analyses. Due to time and logistical constraints, we did not make repeat visits to potential roost trees; therefore, we classified them as used or non-used based on their status the day of the survey. For our radiotelemetry activities, we captured bats by mist net or hand capture at known roost trees. We opportunistically set mist nets in areas of appropriate habitat and hand captured bats from known roosts in artificial structures. Upon capture, we recorded species, sex, forearm length, hind foot length, tragus length, ear length, body weight, reproductive condition, and age class (juvenile or adult, as determined by the epipheyseal-diaphyseal fusion of the finger joints) (Anthony 1998). Captured Rafinesque’s Big-eared Bats and Southeastern Myotis weighing a minimum Southeastern Naturalist L.A. Stuemke, C.E. Comer, M.L. Morrison, W.C. Conway, and R.W. Maxey 2014 164 Vol. 13, Special Issue 5 of 7 g were fitted with 0.52-g transmitters (Model BN-2N, Holohil Systems, Ltd., Carp, ON) or 0.50-g transmitters (Philip Blackburn, SFASU, Nacogdoches, TX). Transmitters were affixed with surgical adhesive (Torbot®, Torbot Group Inc., Cranston, RI) between the shoulder blades without clipping fur, and we did not find that the fur hindered attachment duration. To allow for proper attachment of the transmitters, a thin layer of glue was placed on both the transmitter and the bat and allowed to stand for 5 min or until the glue bubbled as indicated by the adhesive instructions. Tags were then affixed to the bats, and the bat was held for another 15–20 minutes to ensure that the glue had set completely prior to release. We attempted to locate all radiomarked bats daily from the day after transmitter attachment until transmitter failure or drop-off. We tracked bats using a handheld R2000 receiver (Advanced Telemetry Systems, Isanti, MN) with a three-element yagi antenna. Whenever radiotracking led to a tree roost, we marked that tree with a handheld GPS unit and visually verified the presence of the bat in the tree. If additional bats were present in the tree, we attempted to determine species and number by visual or photographic examination. Habitat measurements We recorded 17 habitat-related variables at all used and non-used cavity trees at three spatial extents (tree, stand, landscape). Tree characteristics included species, DBH, and tree height. We also measured several characteristics of the cavity, including interior diameter and height, width and height of the main cavity opening, distance from the bottom of the main cavity opening to ground, and total number of entrances to the cavity. We estimated the height of the cavity by locating a point on the exterior of the tree that corresponded with the top of the cavity. We then used a clinometer to measure the height to this point and used this number to determine the height of the cavity. For trees that had multiple cavity openings, the largest opening was used for measuring height and width of the opening and distance to the ground. We collected stand information in 0.01-ha circular plots centered at each focal tree (used or non-used) to characterize the forest immediately surrounding the focal tree. For each plot, we used a spherical densiometer to measure canopy closure on the perimeter of the plot at each cardinal direction. We also recorded stem density in three size classes according to DBH (0–26 cm, 27–52 cm, and ≥53 cm). For permanent water and habitat edge data, we used georeferenced electronic data or 2010 National Agricultural Imagery Program (NAIP) aerial photographs, both available from the Texas Natural Resources Information System (TNRIS). We used handheld GPS units to obtain data points for anthropogenic structures not clearly visible on aerial photographs. To derive distances, we used the Near tool in ArcGIS 9.3 (Environmental Systems Research Institute, Redlands, CA). Because we were concerned that non-used cavity trees were clustered at the landscape scale, we used random points within the study blocks for comparisons using landscape-scale variables. For these measurements, we generated 34 random points in each study block using the Create Random Points tool in ArcGis 9.3. Southeastern Naturalist 165 L.A. Stuemke, C.E. Comer, M.L. Morrison, W.C. Conway, and R.W. Maxey 2014 Vol. 13, Special Issue 5 Data analysis We used univariate analysis of variance (PROC ANOVA; SAS Institute, Inc. 2004) to identify differences between roost trees for the two target species and to identify factors differentiating between used and non-used cavity trees. We set alpha = 0.05. Results We documented 18 trees used as diurnal roosts by our target species, including 11 used by Rafinesque’s Big-eared Bats and 7 used by Southeastern Myotis. This total included 7 trees discovered on 200 roost-search transects, 7 from radio-telemetry methods, and 2 discovered opportunistically during other activities. Two roosts from Texas Parks and Wildlife Department’s occurrence records for the target species were also included in the analysis. We radiomarked and tracked 7 Rafinesque’s Big-eared Bats (4 females, 3 males) to 6 new tree roosts and 3 Southeastern Myotis (all females) to 1 new tree roost for a total of 54 days. We observed very little roost switching during this study: 6 of the 7 Rafinesque’s Big-eared Bats and all of the Southeastern Myotis used only one tree roost during the period they were radiomarked. The remaining Rafinesque’s Bigeared Bat (a male) switched approximately every 2 days among 4 Water Tupelo trees. Two of the male Rafinesque’s Big-eared Bats roosted in an artificial structure for one night each: a concrete bridge and the attic above a garage in an occupied dwelling). One post-lactating female frequently roosted under a concrete bridge. We did not observe radiomarked Southeastern Myotis roosting in any artificial structures. While conducting the 200 roost-search transects, we also documented and recorded data for 244 non-used cavity trees. Based on these data, the study blocks had approximately 12.3 cavity trees (≥30 cm DBH) per 100 ha of habitat. Nyssa spp. were the most common trees used for roosts by both species, with 55% of diurnal roosts in Blackgum and 33% in Water Tupelo. These two species comprised approximately 50 of 262 (15%) trees with cavities in the area but 15 of the 18 (83%) used cavity trees. The remaining three used cavity trees were a Sweetgum, a Baldcypress, and a Quercus laurifolia Michx. (Swamp Laurel Oak). All but one of the used cavity-trees contained a hollow trunk cavity with a wide interior diameter and smooth interior walls. Of the 18 used cavity trees measured, 11 (61%) possessed a main cavity that was at ground level, four (23%) had cavities located in the trunk of the tree above ground, and three (18%) were chimney trees with the top broken off allowing entrance to the cavity. Characteristics of cavity trees used by Rafinesque’s Big-eared Bats and Southeastern Myotis were similar (Table 1). In general, Rafinesque’s Big-eared Bats used trees with larger cavities than Southeastern Myotis, but the differences did not reach statistical significance. Roosts used by Southeastern Myotis were in areas with higher stem density at the smallest size class, 0–26 cm; however, other stand and landscape characteristics were similar (Table 1). Used cavity trees for both species combined, had cavities that were nearly twice as large (in both height and diameter) as the cavities in non-used trees (Table 2). Perhaps reflecting the cavity size, used trees were larger in diameter than unused trees, but height did not differ. Cavities Southeastern Naturalist L.A. Stuemke, C.E. Comer, M.L. Morrison, W.C. Conway, and R.W. Maxey 2014 166 Vol. 13, Special Issue 5 of used trees also had more entrances and cavity entrances were higher above the ground. Used trees were located in areas with a higher density of large (≥53 cm) trees, but other stand and landscape characteristics were similar for used and nonused trees (Table 2). Table 1. Means, standard errors (SE), numbers of trees (n), and P-values (P) for tree habitat, and landscape variables for tree roosts used by Rafinesque’s Big-eared Bats and Southeastern Myotis at 7 study areas in East Texas, 2008–2009. Rafinesque’s Big-eared Bat Southeastern Myotis Variable Mean SE n Mean SE n P Tree height (m) 17.6 1.37 11 20.9 2.00 7 0.1903 Tree diameter (cm) 101.3 8.65 11 81.9 4.77 7 0.1140 Cavity, interior diameter (cm) 174.9 74.01 8 80.4 9.43 7 0.1794 Cavity, interior height (cm) 813.6 135.71 10 766.7 358.10 6 0.8707 Cavity, distance from ground (cm) 261.4 117.39 11 53.7 33.71 7 0.1883 Cavity, opening width (cm) 25.6 4.04 8 42.6 10.66 7 0.0804 Cavity, opening height (cm) 96.2 31.11 8 70.0 12.04 7 0.3909 Cavity, number of entrances 1.8 0.30 11 1.1 0.14 7 0.1038 Plot, percent canopy closure 85.8 2.67 11 93.3 1.58 6 0.0508 Plot, stem density 0–26 (cm) 48.6 12.10 11 92.3 10.62 7 0.0236 Plot, stem density 27–52 (cm) 7.0 1.71 11 4.7 1.69 7 0.3804 Plot, stem density ≥53 (cm) 2.5 0.71 11 1.7 0.56 6 0.3132 Plot, canopy height (m) 21.1 2.12 11 21.3 1.58 7 0.9487 Distance to permanent water (m) 452.6 101.85 11 234.0 120.84 7 0.1909 Distance to habitat edge (m) 82.0 12.95 11 81.3 17.56 7 0.9722 Distance to human structures (m) 821.0 129.82 11 632.6 191.87 7 0.4103 Table 2. Means, standard errors (SE), number of trees measured (n) and P-value (P) for univariate ANOVAs for tree, habitat, and landscape variables measured at used roosts and non-used cavity trees for Rafinesque's Big-eared Bats and Southeastern Myotis at 7 stu dy areas in East Texas, 2008–2009. Used Trees Non-usedTrees Variable Mean SE n Mean SE n P Tree height (m) 18.9 1.17 18 21.3 0.45 244 0.1532 Tree diameter (cm) 93.8 5.94 18 77.2 1.90 244 0.0216 Cavity, interior diameter (cm) 130.8 40.47 15 58.3 2.16 205 <0.0001 Cavity, interior height (cm) 796.0 151.56 16 431.7 23.56 164 <0.0001 Cavity, distance from ground (cm) 180.6 75.58 18 29.6 7.86 237 <0.0010 Cavity, opening width (cm) 33.6 5.68 15 25.4 1.21 237 0.1024 Cavity, opening height (cm) 84.0 17.30 15 78.3 3.50 237 0.6958 Cavity, number of entrances 1.6 0.20 18 1.2 0.04 242 0.0440 Plot, percent canopy closure 88.5 1.99 17 88.9 0.89 243 0.8882 Plot, stem density 0–26 (cm) 65.6 9.73 18 61.2 3.27 244 0.7208 Plot, stem density 27–52 (cm) 6.1 1.23 18 5.1 0.25 243 0.2824 Plot, stem density ≥53 (cm) 2.2 0.49 17 1.3 0.11 240 0.0432 Plot, canopy height (m) 21.2 1.40 18 22.6 0.44 244 0.3841 Distance to permanent water (m) 367.6 80.01 18 278.0 9.34 680 0.1280 Distance to habitat edge (m) 81.7 10.13 18 111.6 6.00 680 0.4194 Distance to human structures (m) 747.8 107.74 18 964.1 18.34 680 0.0584 Southeastern Naturalist 167 L.A. Stuemke, C.E. Comer, M.L. Morrison, W.C. Conway, and R.W. Maxey 2014 Vol. 13, Special Issue 5 Discussion Cavity size was the most important factor in determining use of a tree as a diurnal roost by rare forest bats in East Texas, which was similar to the findings of a study in Mississippi (Stevenson 2008). Tree DBH was also important, and may be an appropriate proxy for cavity size. Because both target bat species are known to switch roosts regularly under normal conditions (Lewis 1995), and we visited each potential roost only once, we may have classified some trees as unused that were used at other times. There is no reason to suspect bias based on the day of survey (e.g., we were more likely to find them in larger trees on the survey days) and we feel that our findings are representative of roost tree use by these bats in the region. The importance of large trees for roosting by these species has been welldocumented, and roost trees in this study (mean = 93.8 cm) were similar in diameter to Rafinesque’s Big-eared Bat roost trees previously described in Texas (mean = 99.8 cm, Mirowsky 1998) and Louisiana (mean = 59–103 cm, Lance et al. 2001), and slightly larger than roost trees in Mississippi (mean = 79.4 cm, Trousdale and Beckett 2005). However, several studies have found that Rafinesque’s Big-eared Bats selected roosts that were considerably larger, including in Arkansas (mean = 155.3 cm, Cochran 1999), Louisiana (mean = 120.1 cm; Gooding and Langford 2004), and Tennessee (mean = 124.5 cm; Carver and Ashley 2008). The reason that bats in Texas used smaller trees is not clear but it may reflect a paucity of overmature, very large cavity-trees on the East Texas landscape. We found approximately 12.3 cavity trees per 100 ha on our study sites; however, many of these cavities were small. Using the smallest used cavity size (DBH ≥ 62.9 cm, cavity diameter ≥ 40.6 cm, cavity height ≥ 63.5 cm) as a minimum suitable cavity size, we found only 6 suitable roost trees per 100 ha (1 suitable tree per 16 ha) on these study sites. In approximately 2000 ha surveyed for this study, we measured only 20 trees (12 with large cavities) that were ≥124 cm. Although few studies of roost-selection report cavity-tree density, Gooding and Langford (2004) estimated 65.5 cavity trees/ ha and both these authors and Carver and Ashley (2008) noted that large, hollow trees were common on their sites. In contrast, Trousdale and Beckett (2005) commented that suitable trees were uncommon on their study site in Mississippi. Although the implications of using smaller cavities for roosting by Rafinesque’s Big-eared Bats are unknown, our observations suggest that availability of suitable roosts was low in East Texas. In our radiotracking, we observed very little of the roost-switching behavior often observed in both species (Lewis 1995), perhaps due to the large distances between suitable roost trees (Gooding and Langford 2004). Only one male Rafinesque’s Big-eared Bat switched among multiple tree roosts, using four large Water Tupelos that were all within approximately 200 m of each other. Use of anthropogenic structures (abandoned homes, occupied buildings, concrete bridges, concrete bunkers, abandoned wells) was common in our study, and roost switching that did occur often included moving between a single tree-roost and a structure. No information exists comparing relative selection for natural and anthropogenic roosts by Rafinesque’s Big-eared Bats; however, it is illustrative to note that studies with few large trees for roosting (Lance et al. 2001, Mirowsky Southeastern Naturalist L.A. Stuemke, C.E. Comer, M.L. Morrison, W.C. Conway, and R.W. Maxey 2014 168 Vol. 13, Special Issue 5 1998, Trousdale and Beckett 2005) documented >40% of roosts in manmade structures and studies with more suitable tree roosts found fewer (Carver and Ashley 2008) or none (Gooding and Langford 2004). It is difficult to quantify the relationship between bat abundance and availability of suitable roost sites, but most authors agree that roosts, particularly maternity roosts, are critical components of habitat for forest bats (e.g., Barclay and Kurta 2007). In this study, only one of the 11 cavity trees used by Rafinesque’s Big-eared Bats contained more than two bats, suggesting that use of natural tree roosts as maternity colonies is uncommon. In fact, of 7 maternity roosts documented for this species in East Texas, only one occurs in a natural tree roost (Mirowsky et al. 2004). This roost was located in a large (124 cm DBH, 21 m tall) Black Gum tree at LSNWR in Wood County and has contained at least 30 bats during previous surveys by Texas Parks and Wildlife (Meg Goodman, unpubl. data). The remainder were in anthropogenic structures, primarily abandoned buildings, with several colonies containing >40 individuals. If the availability of suitable natural roosts is an important limiting factor in abundance of Rafinesque’s Big-eared Bats, the lack of trees with cavities in the optimal size range may be limiting abundance in this region. Abundance of forest bats is difficult to estimate (Weller 2007), and we do not have reliable estimates of abundance for either target species. In bat surveys conducted over the same study areas, we found that Rafinesque’s Big-eared Bats were widely distributed but never abundant compared to other species (Stuem ke 2011). In contrast to Rafinesque’s Big-eared Bats, we located four apparent maternity roosts for Southeastern Myotis in natural tree roosts. One maternity roost located in a structure (concrete bridge), was opportunistically located by communication with personnel from the Big Thicket National Preserve (Mona Halverson, pers. comm.). These colonies contained 50–150 bats, and the tree roosts occurred in trees ranging from 71–86 cm DBH. Although direct comparisons are relatively rare, Southeastern Myotis generally use smaller cavities than Rafinesque’s Big-eared Bats (Carver and Ashley 2008, Rice 2009). We did not find differences between tree size used by the two species, but this may reflect the lack of trees in the preferred size range for Rafinesque’s Big-eared Bats or the presence of more large maternity colonies in the Southeastern Myotis tree roosts. If Southeastern Myotis readily use smaller natural tree-roosts, this may help explain why it was among the most common species detected in bat surveys at these study sites (Stuemke 201 1). Although we did not survey comprehensively across the landscape of East Texas, our study areas were located throughout the region. Furthermore, the areas we surveyed included several of the highest quality remaining bottomland hardwood and forested wetland sites in the region. Therefore, it is likely that the relative scarcity of large, hollow trees is common across the region. The reasons for this are unclear and may be both natural and anthropogenic. The East Texas Pineywoods are at the western extent of the range for both preferred roost tree species (Water Tupelo and Black Gum) and both bat species. The extensive forest swamps of the Mississippi River valley and large river systems in the eastern Gulf Coastal Plain may never have been present in Texas (Conner and Buford 1998). Furthermore, Southeastern Naturalist 169 L.A. Stuemke, C.E. Comer, M.L. Morrison, W.C. Conway, and R.W. Maxey 2014 Vol. 13, Special Issue 5 logging of East Texas cypress and tupelo swamps may have occurred later than locations farther east (Brandt and Ewel 1989, Diggs et al. 2006). More recently, hurricanes Rita (in 2005) and Ike (in 2008) severely impacted the remaining stands of mature bottomland forest in southeastern Texas. In 2006, we documented at least three Southeastern Myotis maternity roosts and one Rafinesque’s Big-eared Bat roost at Village Creek State Park near Beaumont, TX (see Fig. 1) that were either knocked over completely or broken off in Hurricane Rita (C.E. Comer, pers. observ.). None of these trees has been occupied by the target species since. Historic range and abundance for rare forest bats are unknown, so it is difficult to determine the impact of hurricanes on these species. Long-term management goals for Rafinesque’s Big-eared Bat and Southeastern Myotis in Texas should include preservation of known roosts and areas of matureovermature bottomland cypress-tupelo swamp. Additionally, preserving younger stands and allowing them to reach older age classes will potentially improve habitat conditions for these bats. Apparently, anthropogenic roost structures (including artificial towers constructed specifically for Rafinesque’s Big-eared Bats) are playing an important role as maternity roost sites in the region (Mirowsky et al. 2004), but the implications for long-term population health are unknown. Particularly in areas with few large tupelo trees, structures known to support maternity colonies should be preserved. Acknowledgments We thank the Texas Parks and Wildlife Department for providing the primary funding for our study. Additional funding came from the US Department of Agriculture’s McIntire- Stennis Program and from the Arthur Temple College of Forestry and Agriculture, Stephen F. Austin State University. Thanks go to the staff and constituents of Little Sandy National Wildlife Refuge/Little Sandy Fish and Hunt Club, Caddo Lake National Wildlife Refuge, Caddo Lake Wildlife Management Area, Big Thicket National Preserve, Trinity River National Wildlife Refuge, The Nature Conservancy’s Roy E. Larsen Sandylands Unit, and Village Creek State Park for allowing us access to their properties and for providing field lodging. Special thanks go to G. Winans, J. Deatherage, and K. Hammond for their assistance in the field. Literature Cited Anthony, E.L.P. 1998. Age determination in bats. Pp. 47–58, In T. H. Kunz (Ed.). Ecological and Behavioral Methods for the Study of Bats. Smithsonian Institution Press, Washington, DC. 533 pp. Barclay, R.M.R., and A. Kurta. 2007. Ecology and behavior of bats roosting in tree cavities and under bark. Pp. 17–59, In M.J. Lacki, J.P. Hayes, and A. Kurta (Eds.). Bats In Forests: Conservation and Management. The Johns Hopkins University Press, Baltimore, MD. 352 pp. Bender, S., S. Shelton, K.C. Bender, and A. Kalmback. 2005. Texas comprehensive wildlife conservation strategy: 2005–2010. Texas Parks and Wildlife Department, Austin, TX. Brandt, K., and K.C. Ewel. 1989. Ecology and management of cypress swamps: A review. Florida Cooperative Extension Service Bulletin 252. University of Florida, Gainesville, FL. Southeastern Naturalist L.A. Stuemke, C.E. Comer, M.L. Morrison, W.C. Conway, and R.W. Maxey 2014 170 Vol. 13, Special Issue 5 Brigham, R.M. 2007. Bats in forests: What we know and what we need to learn. Pp. 1–15, In M. J. Lacki, J.P. Hayes, and A. Kurta (Eds.). Bats In Forests: Conservation and Management. The Johns Hopkins University Press, Baltimore, MD. 368 pp. Carver, B.D., and N. Ashley. 2008. Roost-tree use by sympatric Rafinesque’s Big-eared Bats (Corynorhinus rafinesquii) and Southeastern Myotis (Myotis austroriparius). American Midland Naturalist 160(2):364–373. Clark, M.K. 1990. Roosting ecology of the Eastern Big-eared Bat, Plecotus rafinesquii. M.Sc. Thesis. North Carolina State University, Raleigh, NC. Clark, M.K. 2003. Survey and monitoring of rare bats in bottomland hardwood forests. Pp. 79–90, In T.J. O’Shea and M.A. Bogan (Eds.). Monitoring trends in bat populations of the United States and territories: Problems and prospects. US Geological Survey, Biological Resources Discipline, Information and Technology Report. USGS/BRT/ ITR-2003-03. 287 pp. Cochran, S.M. 1999. Roosting and habitat use by Rafinesque’s Big-eared Bat and other species in a bottomland hardwood forest ecosystem. M.Sc. Thesis. Arkansas State University, Jonesboro, AR. Conner, W.H., and M.A. Buford. 1998. Southern deepwater swamps. Pp. 261–287, In M.G. Messina and W.H. Conner (Eds.). Southern Forested Wetlands: Ecology and Management. CRC Press, Boca Raton, FL. 616 pp. Crampton, L.H., and R.M.R. Barclay. 1998. Selection of roosting and foraging habitat by bats in different-aged aspen mixedwood stands. Conservation Biology 12(6):1347–1358. Diggs, G.M., B.L. Lipscomb, M.D. Reed, R.J. O’Kennon, and R. George. 2006. Illustrated Flora of East Texas. Botanical Research Institute of Texas, Fort Worth, TX. Gooding, G., and J.R. Langford. 2004. Characteristics of tree roosts of Rafinesque's Bigeared Bat and Southeastern Bat in northeastern Louisiana. The Southwestern Naturalist 49(1):61–67. Hoffmeister, D.F., and W.W. Goodpaster. 1962. Observations on a colony of Big-eared Bats, Plecotus rafinesquii. Transactions of the Illinois State Academy of Science 55(2):87–89. Hurst, T.E., and M.J. Lacki. 1999. Roost selection, population size, and habitat use by a colony of Rafinesque's Big-eared Bats (Corynorhinus rafinesquii). The American Midland Naturalist 142(2):363–371. Jung, T.S., I.D. Thompson, and R.D. Titman. 2004. Roost site selection by forest-dwelling male Myotis in central Ontario, Canada. Forest Ecology and Management 202:325–335. Lance, R.F., B.T. Hardcastle, A. Talley, and P.L. Leberg. 2001. Day-roost selection by Rafinesque’s Big-eared Bats (Corynorhinus rafinesquii) in Louisiana forests. Journal of Mammalogy 82:166–172. Lewis, S.E. 1995. Roost fidelity of bats: A review. Journal of Mammalogy 76(2):481–496. Menzel, M.A., J.M. Menzel, W.M. Ford, J.W. Edwards, T.C. Carter, J.B. Churchill, and J.C. Kilgo. 2001. Home range and habitat use of male Rafinesque's Big-eared Bats (Corynorhinus rafinesquii). The American Midland Naturalist 145(2):402–408. Mirowsky, K.M. 1998. Roosting ecology of the Southeastern Myotis (Myotis austropriparius) and Rafinesque's Big-eared Bat (Corynorhinus rafinesquii) in East Texas. M.Sc. Thesis. Texas A&M University, Kingsville, TX. Mirowsky, K.M., P.A. Horner, R.W. Maxey, and S.A. Smith. 2004. Distributional records and roosts of Southeastern Myotis and Rafinesque’s Big-eared Bat in eastern Texas. The Southwestern Naturalist 49(2):294–298. Nixon, E.S. 2000. Trees, Shrubs, and Woody Vines of East Texas. Cunningham Productions, Nacogdoches, TX. 240 pp. Southeastern Naturalist 171 L.A. Stuemke, C.E. Comer, M.L. Morrison, W.C. Conway, and R.W. Maxey 2014 Vol. 13, Special Issue 5 Ober, H.K., and J.P. Hayes. 2008. Influence of vegetation on bat use of riparian areas at multiple spatial scales. Journal of Wildlife Management 72(2):396–404. Rice, C.L. 2009. Roosting ecology of Corynorhinus rafinesquii (Rafinesque’s Big-eared Bat) and Myotis austroriparius (Southeastern Myotis) in tree cavities found in a northeastern Louisiana bottomland hardwood forest streambed. M.Sc. Thesis. University of Louisiana, Monroe, LA. SAS Intitute, Inc. 2004. SAS/STAT 9.1. Cary, NC. Stevenson, C.L. 2008. Availability and seasonal use of diurnal roosts by Rafinesque’s Bigeared Bat and Southeastern Myotis in bottomland hardwoods of Mississippi. M.Sc. Thesis. Mississippi State University, Starkville, MS. Stuemke, L.A. 2011. Quantitative analysis of survey methodology, roost selection, and occupancy of Rafinesque’s Big-eared Bats and Southeastern Myotis in East Texas. M. Sc. Thesis. Stephen F. Austin State University, Nacogdoches, TX. Trousdale, A.W., and D.C. Beckett. 2005. Characteristics of tree roosts of Rafinesque’s Bigeared Bat (Corynorhinus rafinesquii) in southeastern Mississippi. American Midland Naturalist 154(2):442–449. Trousdale, A.W., D.C. Beckett, and S.L. Hammond. 2008. Short-term roost fidelity of Rafinesque's Big-eared Bat (Corynorhinus rafinesquii) varies with habitat. Journal of Mammalogy 89(2):477–484. Weller, T.J. 2007. Assessing population status of bats in forests: Challenges and opportunities. Pp. 263–291, In M.J. Lacki, J.P. Hayes, and A. Kurta (Eds.). Bats In Forests: Conservation and Management. The Johns Hopkins University Press, Baltimore, MD. 352 pp. Willis, C. K., and R. M. Brigham. 2005. Physiological and ecological aspects of roost selection by reproductive female Hoary Bats (Lasiurus cinereus). Journal of Mammalogy 86(1):85–94.