O’SHEA, CRYAN & BOGAN: UNITED STATES BAT SPECIES OF CONCERN 
97 
A multi-site roost selection study on the coastal plain of Georgia searched 1,731 hollow trees 
on transects and 22 roosts were found (1.3%), with three the maximum number of bats seen in a 
roost (Clement and Castleberry, 2013a). Tree characteristics suggested that microclimates of 
favored trees were a likely factor in their selection. Tree selection by this species in summer includ¬ 
ed species (water tupelo was nearly always selected), solid wood volume (larger trees favored), and 
lower canopy cover; however, there was a strong site effect, with most sites found in areas with a 
karst substrate and nearby caves also known to be used by this species (Clement and Castleberry, 
2013a). Transect surveys at three study areas with appropriate habitat in South Carolina yielded 
361 trees with cavities, with 12 (3.3%) found to be used as roosts by these bats (Loeb, 2017). Only 
water tupelo trees were observed to be used by this species during summer at Upper Ouachita 
National Wildlife Refuge in northeastern Louisiana, with some individual trees used much more 
frequently than other roost trees (Rice, 2009). Colony sizes varied in the Louisiana study, with 
groups of one to ten bats seen most frequently, but with colony sizes of up to about 300 bats 
observed; only trees with basal hollows and no upper openings were used as roosts, and only the 
ceilings or upper parts of tree cavities were occupied (Rice, 2009). 
In western Tennessee, five southeastern myotis were radio tracked during summer to eight 
roost trees at Pinson Mounds State Archaeological Park; the bats roosted in living hollow water 
tupelo trees which did not differ in size from potential trees that were unused, but were smaller than 
hollow trees used by sympatric Rafinesque’s big-eared bats (Carver and Ashley, 2008). 
Population Ecology.— Litter Size, Natality, and Female Reproduction: This is the only 
species of Myotis in North America that regularly gives birth to twins. Twenty of 28 females (71%) 
at a maternity colony in a Florida attic had twin embryos or young and the remainder had single- 
tons (Sherman, 1930). Over 90% of the litters of 1,489 pregnant females subsequently observed in 
Florida consisted of twins for a mean litter size of 1.9 young per female (Rice, 1957). A rare case 
of triplets has also been documented (Foster et al., 1978). Parturition and lactation take place in 
spring and summer; descriptions of parturition including births of twins were provided by Sherman 
(1930). Sex ratios of newborn are 1:1 based on examination of 2,847 young bats in Florida (Rice, 
1957). 
Sexual maturity in females is reached as yearlings in peninsular Florida, based on subsequent 
recaptures at a maternity roost of 46 female bats banded as juveniles that were all pregnant at age 
one year (Rice, 1957). A sample of 153 females older than one year from the same maternity roost 
were all pregnant (Rice, 1957). Estimated natality based on females captured away from materni¬ 
ty colonies is mostly unavailable but was 100% in a sample of 10 adult females captured over water 
in eastern Texas (Debelica-Lee and Wilkins, 2014), 50% in 16 females netted away from roosts in 
Arkansas (Fokidis et al., 2005), and 93% in 41 females captured in southern Illinois near a mater¬ 
nity roost (Hofmann et al., 1999). Non-reproductive females may roost apart from maternity 
colonies (Hermanson and Wilkins, 1986). 
Survival: No modem data exist on adult survival rates (Gore and Hovis, 1994), although cmde 
estimates on proportion returning annually were provided by Rice (1957). Based on a very simple 
model, Rice (1957) suggested that in order for populations to remain stable in Florida, an annual 
survival of at least 46% was necessary; more recent analyses for other species of temperate 
zone bats suggest this estimate may be too low for stability (for example, species reviewed 
in O’Shea et al., 2011c), but these other species have smaller litter sizes. Modem population 
dynamics simulations have not been published. Survival to weaning at a roost in Florida was 
estimated at 88.2% (Foster et al., 1978). Maximum longevity is at least 6 years (Paradiso and 
Greenhall, 1967). 
Mortality Factors: Young are bom at earlier stages of development than in many other 
