2010 SOUTHEASTERN NATURALIST 9(4):757–772 Identification of Peromyscus gossypinus at Poinsett State Park, South Carolina Pearl R. Fernandes1,*, Justin L. Reynolds1, Nicole Segedin-Garrett1, and Michael J. Dewey2 Abstract - The overlap of external morphometric measurements between Peromyscus leucopus (White-footed Mouse) and P. gossypinus (Cotton Mouse) makes species determination challenging. Peromyscus were live-trapped at Poinsett State Park, Sumter County, SC, and identified using field and laboratory techniques. Our measurements for hind-foot length, total length, and tail length overlapped with published values for both species. Body mass, as measured in the field, was a good criterion that identified our animals as Cotton Mice (n = 29). This identification was confirmed by digital radiography, microsatellite DNA markers, and glucose phosphate isomerase analysis. Our results indicated that a combination of field and laboratory techniques is a valuable approach for positively identifying morphologically similar species. Introduction In many habitats of the eastern United States, mice of the genus Peromyscus (Gloger) are the most abundant small mammals. The genus exhibits a considerable range of morphological, behavioral, and physical variation, and thus serves as a model for studies of population biology, community ecology, adaptive physiology, and evolutionary biology (Carleton 1989). It is also a useful model in genomic approaches to physiological and behavioral adaptations to habitat (Dewey and Dawson 2001). In natural ecosystems, Peromyscus are important not only as prey for mammals, snakes, and birds, but are also predators of insects that cause damage to crop trees (Bellcoq and Smith 1992). Two species, Peromyscus leucopus (Rafinesque) (Whitefooted Mouse) and P. gossypinus (Le Conte) (Cotton Mouse), make up the species group leucopus of Peromyscus, and are believed to have diverged only recently (Carleton 1989, Hall 1981, Hooper 1968, Osgood 1909). This close relationship is supported by a genetic identity of 0.84 (Zimmerman et al. 1978) as well as electrophoretic (Price and Kennedy 1980) and karyologic (Baker et al. 1983) similarities. The species are inter-fertile via captive breeding (Dice 1937) and hybridize in the wild (Barko and Feldhamer 2002, Dice 1940). The White-footed Mouse is generally found at elevations below 900 m in relatively xeric woodlands (Laerm and Boone 1994) or in higher, drier uplands (Howell 1921) of the eastern United States. The Cotton Mouse 1Division of Science, Mathematics and Engineering, University of South Carolina Sumter, 200 Miller Road, Sumter, SC 29150. 2Peromyscus Genetic Stock Center, Department of Biological Sciences, University of South Carolina, Columbia, SC 29208. *Corresponding author - pefernan@mailbox.sc.edu. 758 Southeastern Naturalist Vol. 9, No. 4 is found primarily in mesic lowland hardwood, and swamp forests of the Southeastern United States (Barbour and Davis 1974, Dice 1940, Hoffmeister 1989, Laerm and Boone 1994, Le Conte 1853, McCarley 1954, McCay 2000, Wolfe and Linzey 1977). The White-footed Mouse ranges from Maine southward through the eastern half of the United States to South Carolina, Georgia, and Alabama, and westward through New Mexico to Central Arizona. The Cotton Mouse occurs from Louisiana and southern Florida north in the Mississippi Valley to southern Illinois and southern Kentucky, and through southern South Carolina to eastern North Carolina and Virginia. West of the Mississippi, the Cotton Mouse’s range extends to northeastern Oklahoma and eastern Texas (Whitaker and Hamilton 1998). In areas where both species occur, the White-footed Mouse is generally found in upland woods, and the Cotton Mouse inhabits lowland woods (McCarley 1963, Taylor and McCarley 1963). Both species occur in South Carolina. The White-footed Mouse was reported in Lancaster, Spartanburg, and York counties (Cloninger et al. 1977, Golley 1966), and from all counties in the north-central Piedmont except Union County, where it also likely occurs (Fields 2007). The Cotton Mouse was first reported in the Upper and Lower Coastal Plains in Lancaster County, where it occupied moist mixed woods (Coleman 1948). Cotton Mice have also been reported in Aiken, Barnwell, and Allendale counties, where they use coarse woody debris for primary refuge sites (Hinkelman and Loeb 2007, Loeb 1999, McCay 2000). In the Piedmont Plateau Province of western South Carolina, the Cotton Mouse has been reported in Abbeville, Edgefield, Greenwood, McCormick, and Saluda counties. In areas of overlap, the Cotton Mouse prefers areas with more extensive woody biomass and more complete overstory than does the White-footed Mouse (Mengak and Guynn 2003). Traditionally, pelage coloration, body measurements, and cranial characteristics have been used to distinguish the Cotton Mouse from the Whitefooted Mouse. The dorsum of the Cotton Mouse is dark golden brown with a mid-dorsal dusky area. The White-footed Mouse is grayish brown to dull reddish brown dorsally. Both species are white on the ventral side (Wilson and Ruff 1999). The Cotton Mouse is more massive (17–46 g) than the White-footed Mouse (15–25 g) and has a considerably darker dorsum (Reed et al. 2004, Whitaker and Hamilton 1998). Hind-foot length is generally >22 mm in Cotton Mouse and ≤22 mm in the White-footed Mouse (Dice 1940, McCarley 1954). The two species are best distinguished by skull length, which is generally >28 mm in the Cotton Mouse and less than 25.4 mm in the White-footed Mouse (Lowery 1974). Since skull length and other cranial characteristics require sacrifice of animals, these criteria are not useful in many ecological studies. In certain habitats, morphological differences between the two species can be rather subtle (Hoffmeister 1989, Schwartz and Schwartz 1981) due to 2010 P.R. Fernandes, J.L. Reynolds, N. Segedin-Garrett, and M.J. Dewey 759 their close relationship (Lovecky et al. 1979, McCarley 1963), so misidentifications of species occur. Identification is particularly challenging when small sample sizes preclude use of discriminant function analysis of external measurements (Choate 1973, Feldhamer et al. 1983). Additionally, age and geographic variation can affect accuracy of identification based on external measurements (Sternburg and Feldhamer 1997). Molecular markers allow positive identification of closely related species of Peromyscus. Protein markers have been used to study the genetic structure of natural populations (Lewontin 1991). Genotype for protein loci can be inferred from electrophoretic patterns because of their co-dominant expression, constant number of enzyme subunits in related species, and consistent patterns of tissue-specific expression (Luikart et al. 1998). Salivary amylase variants have been used to accurately identify sympatric species of Peromyscus (Aquadro and Patton 1980) and improve the reliability of field-based identifications (Bruseo et al. 1999, Feldhamer et al. 1983, Lindquist et al. 2003, Rich et al. 1996, Sternburg and Feldhamer 1997). Alleles of glucose phosphate isomerase (GPI) are diagnostic between the Cotton and White-footed Mouse (Barko and Feldhamer 2002, Price and Kennedy 1980, Robbins et al. 1985). However, allozymes overlook mutations that do not produce amino acid changes. These silent mutations could contain important additional information about each species. Microsatellite DNA markers allow detection of genetic variation at an unlimited number of loci with much greater sensitivity than allozyme analysis (Allendorf 1994). These DNA markers permit researchers to study historical patterns of isolation and gene flow in populations which can be obscured in allozyme analysis due to balancing selection at allozyme loci (Karl and Avise 1992). In the present study, we compare effectiveness of microsatellite markers and allozyme electrophoretic mobility of GPI with morphological trait measurements as tools for unequivocal identification of mice trapped at a site in south-central South Carolina. We chose Poinsett State Park as an undisturbed habitat for Peromyscus species and investigated: (1) which species of Peromyscus is/are present, (2) whether individuals can be accurately identified by field techniques, and (3) how laboratory techniques compare with field techniques for identification of Peromyscus populations. Field Site Description Poinsett State Park includes approximately 4 square km of relatively undisturbed habitat in central South Carolina and is 35 km southwest of Sumter in Sumter County, at 33°48'28"N, 80°56'0"W. Elevation ranges from 25–68 m above sea level. The Park is part of Manchester Forest at the edge of the Sandhills, but still lies within the upper portion of the Coastal Plain. A diverse geographic landscape of north- and south-facing slopes, wet swampy 760 Southeastern Naturalist Vol. 9, No. 4 lowlands, and dry, exposed ridges characterize the Park (South Carolina State Parks 2009). A topographical map (Fig. 1) was created using Arc GIS software v8.1 (ESRI, NJ) and GIS data from the South Carolina Department of Natural Resources (2002). Vegetative cover in the Park varies from pine, mixed pine-hardwood, secondary growth hardwood to mature swamp forest. South-facing slopes (upland) contain mainly Pinus taeda L. (Loblolly Pine), Quercus falcata Michx. (Southern Red Oak), Q. marilandica Münch. (Blackjack Oak), Q. margarettae (Ashe) Small (Dwarf Post Oak), and Q. laevis Walter (Turkey Oak). Dominant vegetation in the swampy lowlands includes Figure 1. Location and topography of Poinsett State Park in Sumter County, SC. 2010 P.R. Fernandes, J.L. Reynolds, N. Segedin-Garrett, and M.J. Dewey 761 Southern Red Oak, Turkey Oak, Q. nigra L. (Water Oak), Liquidambar styraciflua L. (Sweet Gum), Nyssa sylvatica Marsh. (Black Gum), and Oxydendrum arboreum (L.) DC. (Sourwood) in the canopy and an understory of Arundinaria tecta (Walter) Muhl. (Switch Cane), Lyonia lucida (Lam.) K. Koch (Fetterbush), Leucothoe axillaris (Lam.) G. Don (Coastal Doghobble), and Kalmia latifolia L. (Mountain Laurel). Lowland sites had thick overstory, dense vine thickets, and various amounts of coarse woody debris (CWD). Methods Trapping A preliminary feasibility study conducted in the late spring of 2002 identified two appropriate trapping sites for Peromyscus at Poinsett State Park. Within each 1-ha plot, two upland (hill slopes) and three lowland (swamp) sites were selected based on visual inspection of the terrain. A road separated upland from lowland sites. At each site, ten traps were set at approximately 10-m intervals in a 2 x 5 grid pattern (W. Dawson, Department of Biological Sciences, University of South Carolina, Columbia, SC, 2002 personal comm.). Trapping was conducted from May to June in 2002, and May to August in 2003. During each sampling period, 20 traps were set in upland sites and 30 in lowland sites. Total number of trap days was 20 in 2002 and 55 in 2003. Peromyscus were captured with small Sherman live traps (17 cm by 5.4 cm by 6.5 cm; H.B. Sherman Co., Tallahassee, fl) baited with peanut butter. Traps were set on, in, or under tree bases, fallen decaying logs, stumps, root boles, vine thickets, open areas, hill slopes, and low elevations near stream edges. Traps were set on Monday, left in place for 4 days, and brought back to the laboratory on Friday for washing. Traps were checked each morning between 0700 and 0900 h and replaced with clean traps after each capture. In 2003, Sherman traps were placed inside sections of polyvinyl chloride (PVC) pipe with chicken wire ends and a hole for mouse entry. This modification prevented damage from raccoons. Trapped mice were weighed, sexed, and assessed for characters of dorsal and ventral pelage, mass, hind-foot length, total length, and tail length. (Feldhamer et al. 1983, Sternburg and Feldhamer 1997). Measurements were taken to the nearest gram or millimeter. Date and location of capture were also noted. Captured mice were marked by tail clipping for recognition of recaptures. Tail tips were numbed with ethyl chloride spray before a maximum of 4 mm was snipped for molecular analysis. Tips were placed in ice, returned to the laboratory, and stored in a -20 °C freezer. Trapped mice were released at the site of capture. All mouse procedures conformed to IACUC Protocol # 1175 at the University of South Carolina. Digital radiography Two adults trapped in 2003, and identified as Cotton Mice based on pelage color, weight greater than 26 g, and hind-foot length greater than 22 762 Southeastern Naturalist Vol. 9, No. 4 mm (Barko and Felhamer 2002, Hoffmeister 1989), were selected for digital radiography (DEXIS, Redwood, CA). X-rays of left lateral views of skulls and mandibles were conducted. Skull length was calibrated using the premeasured head length of the mouse and compared with published data for regional species of Peromyscus (Laerm and Boone 1994, Reed et al. 2004) to confirm identification. Microsatellite DNA markers DNA was extracted from tail snips of all trapped mice using Qiagen DNeasy Kit (Cat. No.69506). DNA from adults from 2002 and all mice from 2003 were amplified by polymerase chain reaction (PCR), using the method of Prince et al. (2002). Primers were PO-97F (TGGCATTCAAAGTTTTATCTC), PO-97R (CCTGGAGC TTTATCTAGAA), PO-21F (TCTGCAAGTTGGAGGTAGAGA), and PO-21R (GGGAGCTGAGGGTTCAA), which have been used successfully with P. polionotus subgriseus (Wagner) (Oldfield Mouse) from the Peromyscus Genetic Stock Center. DNA from Oldfield Mouse, P. maniculatus bairdii (Wagner) (Deer Mouse), White-footed Mouse, and Cotton Mouse were used as controls. Ten μl of PCR products were run on a 2% agarose gel with a 100 base pair (0.5 μg) DNA ladder (New England Biolabs Inc.) as a size standard. DNA bands were stained using ethidium bromide, and gels were photographed on an ultraviolet trans-illuminator. Allozyme analysis GPI analysis on tail tissue of 12 adult mice captured in 2003 followed the methods of Eppig et al. (1977). These samples were run with samples from known reference species on cellulose acetate gels. Mobility was scored according to VanZant et al. (1983). Data analysis Mice judged to be adults, based on brown dorsal pelage and weight greater than 18 g (Barko and Feldhamer 2002), were used to compute mean, standard error, and range for each morphometric measurement. Unpaired t tests with α = 0.05 were used to compare mass of adult males between 2002 and 2003, and between all males and all females. We compared our 95% confidence interval (mean ± 1.98 SD) and range for each morphometric measurement with published data (mean and range) for Cotton and White-footed Mouse. Published values outside the 95% confidence interval were considered significant differences between our sample and the relevant species. Due to our small sample size, we used digital radiography, microsatellite markers, and glucose phosphate isomerase analysis as additional techniques to confirm field identification and overcome the problem of low statistical power to detect true differences between the species. 2010 P.R. Fernandes, J.L. Reynolds, N. Segedin-Garrett, and M.J. Dewey 763 Results Trapping A total of 34 mice (29 adults and 5 juveniles) were captured over two years. Twenty six (76%) were caught within 6 m of stream edges, and 8 (24%) were trapped between 8 and 15 m from stream edges. No mice were caught in the upland sites, although 40% percent of traps were set in those locations. Table 1 presents the results of the trapping data. Of 17 adults in 2002, 8 were male (45%) and 9 were female (55%). Trap success in the lowland in 2002 was 0.078 animals per trap night. In 2003 only 12 adult males and 1 juvenile were captured. Trap success for the lowland in 2003 was 0.016. In 2003, heavy raccoon activity destroyed many traps. No mice from 2002 were recaptured in 2003. Morphometric measurements Measurements of mass, hind-foot length, total length, and tail length for trapped mice are summarized in Table 2. There was no statistical difference in mass of males between years (t = 0.59, d.f. = 18, P = 0.563). There was also no significant difference in pooled weights of all males and all females in year 2002 (t = 0.560, d.f. = 27, P = 0.58), so all data (n = 29) were pooled for statistical analysis. Mass ranged from 28.0–46.0 g (mean = 35.5, SE = 0.83), hind foot length from 20.0–26.0 mm (mean = 23.30, SE = 0.3, n = 28), total length from 136.0–237.0 mm (mean = 163.81, SE = 3.91, n = 28), and tail length from 54.0–80.0 mm (mean = 66.78, SE = 1.42, n = 28). Pooled data were compared with published morphological measurements for the Cotton and White-footed Mouse and presented in Table 3. The 95% confidence interval for mass was 26.72–44.34 g, for hind-foot length was 20.07–26.53 mm, for total length was 123.56–204.06 mm, and for tail length was 52.13–81.41 mm. Digital radiography, microsatellite DNA markers, and GPI analysis Skull lengths from radiographs were 28.3 and 28.5 mm, values consistent with Cotton mice. The PO-97 primers produced amplicons of 200 base pairs in Oldfield and Deer Mouse controls and 170 base pairs in Cotton and Table 1. Summary of trapping results at Poinsett State Park. 2002 2003 Trapping information Lowland Upland Lowland Upland Trap dates 20 20 55 55 Trap number 30 20 30 20 Trap nights 600 400 1650 1100 Number of females 9 0 0 0 Number of males 8 0 12 0 Number of juveniles 4 0 1 0 Number of recaptures 26 0 14 0 Trap success rate 0.078 0 0.016 0 764 Southeastern Naturalist Vol. 9, No. 4 Table 3. Comparison of morphological measurements between Peromyscus leucopus and P. gossypinus populations. Mass (g) Hind-foot length (mm) Total length (mm) Tail length (mm) Location n Range Mean Range Mean Range Mean Range Mean Species Source Poinsett 28–29 28–46 35.5 20–26 23.3 136–237 163.8 54–80 66.7 Present study Missouri 24 28.8 20.9 leucopus Barko et al.(2000) Missouri 4 34.8 24.0 gossypinus Barko et al.(2000) North Carolina 4 17–24 20.4 leucopus Boone and Laerm (1993) North Carolina 22 19–37 28.4 gossypinus Boone and Laerm (1993) Illinois 73 21.0 20.6 leucopus Feldhamer et al.(1998) Illinois 5 26.7 22.4 gossypinus Feldhamer et al. (1998) Illinois 18–22 leucopus Hoffmeister (1989) Illinois 22–25 gossypinus Hoffmeister (1989) Southeastern US 108 16–21 19.4 49–83 65.0 leucopus Laerm and Boone (1994) Southeastern US 110 20–24 22.1 58–91 73.0 gossypinus Laerm and Boone (1994) Louisiana 17–23 19.1 17–21 20.0 134–177 157.0 51–87 69.0 leucopus Lowery (1974) Louisiana 25–45 31.1 20–26 22.0 137–210 171.0 55–94 75.0 gossypinus Lowery (1974) Texas 14 19.4–21 20.0 53–76 62.9 leucopus McCarley (1954) Texas 18 21.8–23 22.4 55–80 71.0 gossypinus McCarley (1954) Missouri 11–28 19–25 139–212 63–101 leucopus Schwartz and Schwartz (1981) Missouri 19–25 20–25 161–209 69–101 gossypinus Schwartz and Schwartz (1981) Illinois 168 14.5–32 21.2 18–22 20.3 leucopus Sternburg and Feldhamer (1997) New York 30 16–28 18–23.3 21.0 157–189 170.0 60–92 76.0 leucopus Whitaker and Hamilton (1998) Florida 15 17–46 21–23 22.0 152–189 166.0 63–80 71.0 gossypinus Whitaker and Hamilton (1998) Tennessee 19 15–25 19.5–22 20.0 152–181 165.0 59–83 72.0 leucopus Whitaker and Hamilton (1998) Tennessee 30 25–39 20–26 23.3 160–205 185.0 63–97 80.0 gossypinus Whitaker and Hamilton (1998) Table 2. Summary of morphological data of adult Peromyscus trapped at Poinsett Park. All values in mm except for mass. Males-2002 (n = 8) Females-2002 (n = 9) Males-2003 (n = 12) All mice (n = 29) Character Mean ± SE Range Mean ± SE Range Mean ± SE Range Mean ± SE Range Mass (g) 33.8 ± 1.42 28.0–41.0 36.2 ± 0.81 34.0–41.0 35.1 ± 1.66 29.0–46.0 35.5 ± 0.83 28.0–46.0 Hind-foot length 22.8 ± 1.45 20.0–25.0 23.3 ± 0.50 21.0–25.0 23.6 ± 0.501 21.0–26.0 23.3 ± 0.302 20.0–26.0 Total length 165.6 ± 5.70 147.0–190.0 154.5 ± 3.60 136.0–174.0 169.0 ± 7.601 146.0–237.0 163.8 ± 3.912 136.0–237.0 Tail length 74.0 ± 2.00 67.0–80.0 66.0 ± 1.80 67.0–78.0 63.1 ± 2.001 56.0–75.0 66.8 ± 1.422 56.0–80.0 1n = 11 (1 data point missing). 2n = 28 (1 data point missing). 2010 P.R. Fernandes, J.L. Reynolds, N. Segedin-Garrett, and M.J. Dewey 765 White-footed Mouse controls (Fig. 2). All Poinsett State Park samples produced the 170 base pair amplicon. The PO-21 primers produced amplicons of 170, 150, and 100 base pairs in Oldfield Mouse, Deer Mouse, and White-footed Mouse controls, respectively (Fig. 3). No amplicons were obtained in Cotton Mouse control and 10 samples tested from Poinsett State Park. Some samples from 2002 degraded and could not be analyzed by gel electrophoresis. The 12 samples analyzed in 2003 were homozygous for a slower migrating allele (GPI-1a) than that (GPI-1b) seen in White-footed Mouse, Deer Mouse, and Oldfield Mouse controls (Fig. 4). Control Cotton Mouse tissue was not available. Discussion Identification of the Cotton Mouse is difficult due to the similarity in external appearance and overlap in morphological measurements with the Figure 2. Agarose gel electrophoresis of PCR products with PO-97 primer. Abbreviations are P, M, L, and G for products from P. polionotus, P. maniculatus, P. leucopus, and P. gossypinus, respectively, (-) for a negative control without DNA, Std for a DNA size marker, and numbers for DNA from 10 samples in 2002 (panel a) and 12 samples in 2003 (panel b). This PO-97 marker discriminates P.leucopus and P. gossypinus from other species but fails to identify P. gossypinus. 766 Southeastern Naturalist Vol. 9, No. 4 White-footed Mouse (Hoffmeister 1989, Schwartz and Schwartz 1981). Due to their wide distribution and geographic variation, significant differences in pelage colors and body measurements occur within each species (Blair 1950, Figure 3. Agarose gel electrophoresis of PCR products with PO-21 primer. Abbreviations are P, M, L and G for products from P. polionotus, P. maniculatus, P. leucopus, and P. gossypinus, respectively, (-) for a negative control without DNA, Std for a DNA size marker, and numbers for DNA from 10 samples in 2002 (panel a) and 12 samples in 2003 (panel b). The PO-21 marker discriminates P. gossypinus from all other species. 2010 P.R. Fernandes, J.L. Reynolds, N. Segedin-Garrett, and M.J. Dewey 767 Dice 1940), making reliable identification challenging. The Cotton Mouse has been reported to average a higher body mass, longer hind foot, and larger skull than the White-footed Mouse (Lowery 1974), and these characteristics were of partial value in identifying our individuals. Body mass measurements for Poinsett State Park animals were within the range reported for Cotton Mice and larger than White-footed Mice (Boone and Laerm 1993, Lowery 1974, Schwartz and Schwartz 1981, Whitaker and Hamilton 1998, Wolfe and Linzey 1977). Mean body mass measurements were similar to those reported for Cotton Mice in Missouri (Barko et al. 2000). One “general rule” for identifying a potential Cotton Mouse is hind-foot length greater than 22 mm and/ or body mass greater than 26 g (Hoffmeister 1989). Applying this rule for body mass confirmed our individuals as Cotton Mice. However, 5 (17%) mice had a hind-foot length of <22 mm, and 4 (14%) could have been misidentified. Our range for hind-foot length overlapped with several published ranges for White-footed Mice (Hoffmeister 1989, Laerm and Boone 1994, Lowery 1974, McCarley 1954, Schwartz and Schwartz 1981, Sternburg and Feldhamer 1997, Whitaker and Hamilton 1998) and did not support the findings of Dice (1940) and McCarley (1954), who found hindfoot length to be the most useful morphological character in distinguishing between the Cotton and White-footed Mouse. As can be observed in Table 3, published ranges for total and tail length overlapped substantially between the two species and varied considerably throughout geographical ranges for each species. Our measurements overlapped considerably with published Figure 4. Allelic mobility of glucose phosphate isomerase (GPI) for several species of Peromyscus. Abbreviations are L, L1, and L2 for P. leucopus, M for P. maniculatus, P for P. polionotus, and 12 samples from Poinsett State Park collected in 2003. 768 Southeastern Naturalist Vol. 9, No. 4 ranges for the White-footed Mouse (Feldhamer et al. 1998, Hoffmeister 1989, Laerm and Boone 1994, Lowery 1974, McCarley 1954, Schwartz and Schwartz 1981, Whitaker and Hamilton 1998), and we could not positively identify our individuals using these criteria. Laerm and Boone (1994) utilized stepwise discriminant analysis to correctly classify all their specimens and overcome the problem of morphological overlap. Our small sample size precluded use of this analysis. From our results, body mass was the only field measurement that correctly identified all our individuals as Cotton Mice. However, age and reproductive condition can affect body mass (Barko et al. 2000, Sternburg and Felhamer 1997). Identification based on body mass was confirmed by digital radiography to compute skull length, microsatellite DNA markers, and glucose phosphate isomerase analysis. Skull length is the best way to identify these species (Lowery 1974), and both specimens examined were Cotton Mice. The microsatellite marker PO-97 did not distinguish between the Cotton Mouse and White-footed Mouse, but was valuable in distinguishing these mice from the two sister species, Oldfield Mouse and Deer Mouse. The PO-21 microsatellite marker gave PCR amplicons in all species except Poinsett samples and Cotton Mouse controls. The apparent null allele for PO-21 in Cotton Mice was a reliable means for distinguishing Cotton Mice from other South Carolina species. Allozyme analysis (Aquadro and Patton 1980, Bruseo et al. 1999, Kilpatrick and Zimmerman 1975, Lindquist et al. 2003, Rich et al. 1996, Robbins et al. 1985, Sternburg and Feldhamer 1997) differentiates between Peromyscus species, and GPI distinguishes between the Cotton and White-footed Mouse (Barko and Feldhamer 2002, Price and Kennedy 1980). The GPI allozyme from our Poinsett samples was different from all other species of Peromyscus in South Carolina, Deer Mouse and Oldfield Mouse. GPI has been used to distinguish Cotton Mice from White-footed Mice in the Midwest (Barko and Feldhamer 2002) and our results suggest a similar distinction for the two species in the South. In conclusion, our results suggest that all mice captured were Cotton Mice. Body mass measured in the field was a good criterion, but hind-foot length, total length, and tail length were poor criteria for identification. Digital radiography for skull length, microsatellite DNA markers, and GPI were laboratory techniques that confirmed the field identification. We recommend using a combination of field and laboratory techniques to ensure accurate identification of morphologically similar species. Acknowledgments We owe a great deal of thanks to Gabor Szalai and the Peromyscus Genetic Stock Center at the University of South Carolina Columbia for assisting in this study. Thank you to Dr. Wayne Van Devender for his detailed review of the manuscript and providing helpful suggestions with the data analysis and presentation. Also, we thank the 2010 P.R. Fernandes, J.L. Reynolds, N. Segedin-Garrett, and M.J. Dewey 769 Department of Parks, Recreation, and Tourism at Poinsett State Park for providing the study site, Professor John Logue for help with plant identifications, Clint Cook for help with Adobe Photoshop, Amber Woodle for assistance in the field, and Dr. Jeff Steinmetz for reading the manuscript. Our thanks extend also to Dr. Michael W. Nimmich for use of the DEXIS digital radiography system. The study was partially funded by the Provost’s Faculty Development Grant from the University of South Carolina Columbia awarded to Pearl Fernandes. Literature Cited Allendorf, F.W. 1994. Comparative utility of genetic markers in the management of Pacific salmon: Proteins, nuclear DNA, and mitochondrial DNA. Pp. 127–133, In L.K. Park., P. Moran, and R.S. Waples (Eds.). Application of DNA Technology to the Management of Pacific Salmon: Proceedings of the Workshop. US Department of Commerce, Seattle, WA. NOAA Technical Memo NMFS-NWFSC-17. 178 pp. Aquadro, C.F., and J.C. Patton. 1980. Salivary amylase variation in Peromyscus: Use in species identification. Journal of Mammalogy 61:703–707. Baker, R.J., J.L.W. Robbins., F.B. Stangl, Jr., and E.C. Birney. 1983. Chromosomal evidence for a major subdivision in Peromyscus leucopus. Journal of Mammalogy 64:356–359. Barbour, R.W., and W.H. Davis. 1974. Mammals of Kentucky. University Press of Kentucky, Lexington, KY. 368 pp. Barko, V.A., and G.A. Feldhamer. 2002. Cotton Mice (Peromyscus gossypinus) in southern Illinois: Evidence for hybridization with White-footed Mice (Peromyscus leucopus). American Midland Naturalist 147:109–115. Barko, V.A., B.L. Sloss, and G.A. Feldhamer. 2000. A non-lethal method for identification of the Cotton Mouse, Peromyscus gossypinus (Le Conte, 1853). Transactions of the Illinois State Academy of Science 93:285–291. Bellcoq, M.I., and S.M. Smith. 1992. Management of small mammals for the biological control of White Pine Weevil [Abstract]. Pp. 352–354, In Proceedings, 54th Midwest Fish and Wildlife Conference, Toronto, ON, Canada. Blair, W.F. 1950. Ecological factors in speciation of Peromyscus. Evolution 4(3):253–275. Boone, J.L., and J. Laerm. 1993. Cotton mice, Peromyscus gossypinus LeConte (Rodentia: Cricetidae), in the Great Dismal Swamp and surrounding areas. Brimleyana 18:125–129. Bruseo, J.A., S.H. Vessey., and J.S. Graham. 1999. Discrimination between Peromyscus leucopus noveboracensis and Peromyscus maniculatus nubiterrae in the field. Acta Theriologica 44(2):151–160. Carleton, M.D. 1989. Systematics and evolution. Pp. 7–139, In G.L. Kirklandand J.N. Layne. (Eds.). Advances in the Study of Peromyscus (Rodentia). Texas Tech University Press, Lubbock, TX. 366 pp. Choate, J.R. 1973. Identification and recent distribution of White-footed Mice (Peromyscus) in New England. Journal of Mammalogy 54:41–49. Cloninger, R.A., J.S. Garton., P.M. Cumbie, and S.D. Berg. 1977. The occurrence of nongame wildlife in Piedmont transmission corridor rights-of-way. Duke Power 77-02. Duke Power Company, Charlotte, NC. 770 Southeastern Naturalist Vol. 9, No. 4 Coleman, R.H. 1948. Some mammal notes from South Carolina. Journal of Mammalogy 29:293–294. Dewey, M.J., and W.D. Dawson. 2001. Deer Mice: The Drosophila of North American mammalogy. Genesis 29:105–109. Dice, L.R. 1937. Fertility relations in the Peromyscus leucopus group of mice. Contributions of the Laboratory of Vertebrate Genetics, University of Michigan 4:1–3. Dice, L.R. 1940. Relations between the Wood-mouse and the Cotton Mouse in eastern Virginia. Journal of Mammalogy 21:14–23. Eppig, J.J., L.P. Kozak., E.M. Eicher, and L.C. Stevens. 1977. Ovarian teratomas in mice developed from oocytes that have completed the first meiotic division. Nature 269:517–518. Feldhamer, G.A., J.E. Gates, and J.H. Howard. 1983. Field identification of Peromyscus maniculatus and Peromyscus leucopus in Maryland: Reliability of morphological characteristics. Acta Theriologica 27:417–423. Feldhamer, G.A., J.C. Whitaker, and E.M. Charles.1998. Recent records of the Cotton Mouse (Peromyscus gossypinus) in Illinois. American Midland Naturalist 139:178–180. Fields, S.E. 2007. Mammals of the north-central Piedmont of South Carolina. Southeastern Naturalist 6(4):577–596. Golley, F.B. 1966. South Carolina Mammals. Contributions from the Charleston Museum XV. 181 pp. Hall, E.R. 1981. The Mammals of North America. John Wiley and Sons, New York, NY. 2:601– 1181 pp + 90. Hinkelman, T.M., and S.C. Loeb. 2007. Effect of woody debris abundance on daytime refuge use by Cotton Mice. Southeastern Naturalist 6:393–406. Hoffmeister, D.F. 1989. Mammals of Illinois. University of Illinois Press, Urbana, IL. 348 pp. Hooper, E.T. 1968. Classification. Pp. 27–74, In J.A. King (Ed.). Biology of Peromyscus (Rodentia). American Society of Mammalogists, Stillwater, OK. Special Publication No. 2. 593 pp. Howell, A.H. 1921. A biological survey of Alabama. North American Fauna 45:1–88. Karl, S.A., and J.C. Avise. 1992. Balancing selection at allozyme loci in oysters: Implications from nuclear RFLPs. Science 256:100–102. Kilpatrick, W., and E.G. Zimmerman. 1975. Genetic variation and systematics of four species of mice of the Peromyscus boylii species group. Systematic Zoology 24:143–162. Laerm, J., and J.L. Boone. 1994. Mensural discrimination of four species of Peromyscus (Rodentia: Muridae) in southeastern United States. Brimleyana 21:107–123. Le Conte, J.L. 1853. Descriptions of three new species of American Arvicolae, with remarks upon some other American rodents. Proceedings of the National Academy of Sciences 6:404–415. Lewontin, R.C. 1991. Twenty-five years ago in genetics: Electrophoresis in the development of evolutionary genetics: Milestone or millstone? Genetics 128:657–662. Lindquist, E.S., C.F. Aquadro., D. McClearn, and K.J. McGowan. 2003. Field identification of the mice Peromyscus leucopus noveboracensis and P. maniculatus gracilis in central New York. Canadian Field-Naturalist 117(2):184–189. 2010 P.R. Fernandes, J.L. Reynolds, N. Segedin-Garrett, and M.J. Dewey 771 Loeb, S.C. 1999. Responses of small mammals to coarse woody debris in a southeastern pine forest. Journal of Mammalogy 80:460–471. Lovecky, D.V., D.Q. Estep, and D.A. Dewsbury. 1979. Copulatory behaviour of Cotton Mice (Peromyscus gossypinus) and their reciprocal hybrids with Whitefooted Mice (P. leucopus). Animal Behavior 27:371–375. Lowery, G.H., Jr. 1974. The Mammals of Louisiana and its Adjacent Waters. Louisiana State University Press, Baton Rouge, LA. 565 pp. Luikart, G., W.B. Sherwin, B.M. Steele, and F.W. Allendorf. 1998. Usefulness of molecular markers for detecting population bottlenecks via monitoring genetic change. Molecular Ecology 7:963–974. McCarley, W.H. 1954. The ecological distribution of Peromyscus leucopus species group in eastern Texas. Ecology 35(3):375–379. McCarley, W.H. 1963. Distributional relationships of sympatric populations of Peromyscus leucopus and P. gossypinus. Ecology 44:784–788. McCay, T.S. 2000. Use of woody debris by the Cotton Mouse (Peromyscus gossypinus) in southeastern pine forests. Journal of Mammalogy 81:527–535. Mengak, M.T., and D.C. Guynn, Jr. 2003. Small mammal microhabitat use on young Loblolly Pine regeneration areas. Forest Ecology and Management 173:309–317. Osgood, W.H. 1909. Revisions of the mice of the genus Peromyscus. North American Fauna 28:1–285. Price, P.K., and M.L. Kennedy. 1980. Genic relationships in the White-footed Mouse, Peromyscus leucopus, and the Cotton mouse, Peromyscus gossypinus. American Midland Naturalist 103:73–82. Prince, K.L., T.C. Glenn, and M.J. Dewey. 2002. Cross-species amplification among peromyscines of new microsatellite DNA loci from the Oldfield Mouse (Peromyscus polionotus subgriseus). Molecular Ecology 2:133–136. Reed, A.W., P.K. Kennedy., M.L. Beck, and M.L. Kennedy. 2004. Using morphologic characters to identify Peromyscus in sympatry. American Midland Naturalist 152(1):190–195. Rich, S.M., C.W. Kilpatrick., J.L. Shippee, and K.L. Crowell. 1996. Morphological differentiation and identification of Peromyscus leucopus and P. maniculatus in northeastern North America. Journal of Mammalogy 77:985–991. Robbins, L.W., M.H. Smith., M.C. Wooten, and R.K. Selander. 1985. Biochemical polymorphism and its relationship to chromosomal and morphological variation in Peromyscus leucopus and the Cotton Mouse, Peromyscus gossypinus. Journal of Mammalogy 66:498–510. Schwartz, C.W., and E.R. Schwartz. 1981. The Wild Mammals of Missouri, Revised Edition. University of Missouri Press and Missouri Department of Conservation, Columbia, MO. 356 pp. Sternburg, J.E., and G.A. Feldhamer. 1997. Mensural discrimination between sympatric Peromyscus leucopus and P. maniculatus in southern Illinois. Acta Theriologica 42:1–13. South Carolina Department of Natural Resources (SC DNR). 2002. Gap data and dynamic mapping. Available online at http://www.dnr.sc.gov/GIS/gap/mapping. html. Accessed 10 January 2002. South Carolina State Parks. 2009. Poinsett State Park. Available online at http:// www.southcarolinaparks.com/park-finder/state-park/662.aspx. Accessed10 February 2009. 772 Southeastern Naturalist Vol. 9, No. 4 Taylor, R.J., and H. McCarley. 1963. Vertical distribution of Peromyscus leucopus and P. gossypinus under experimental conditions. Southwestern Naturalist 8(2):107–108. VanZant, G., P.W. Eldridge., R.R. Behringer, and M.J. Dewey. 1983. Genetic control of hematopoietic kinetics revealed by analyses of allophenic mice and stem cell suicide. Cell 35:639–649. Whitaker, J., and W. Hamilton. 1998. Mammals of the Eastern United States. Cornell University Press, Ithaca, NY. 583 pp. Wilson, D.E., and S. Ruff. 1999. The Smithsonian Book of North American Mammals. Smithsonian Institution Press in association with the American Society of Mammalogists, Washington, DC. 750 pp. Wolfe, J.L., and A.V. Linzey. 1977. Peromyscus gossypinus. Mammalian Species 70:1–5. Zimmerman, E.G., C.W. Kilpatrick, and B.J. Hart. 1978. The genetics of speciation in the rodent Peromyscus. Evolution 32(3):565–579.