We describe a new species of hemidactyliine salamander in the genus Eurycea based on a study of variation in morphometric, osteologic, reproductive, and larval characteristics and allozyme-based allele frequencies and fixed differences in samples of two color pattern morphs known to exist in the E. quadridigitata complex in North and South Carolina. The new species differs from its presumptive relative Eurycea quadridigitata (Holbrook) in having a bright yellow venter, a smaller size, relatively longer limbs, fewer paravomerine teeth, a greater number of yolk-laden ovarian eggs, and dorsally spotted hatchling larvae. Osteologic differences include a frontal with a well-developed triangular postero-lateral process, a prevomer that lacks a postero-medial bony shelf, a prevomerine tooth series that is not sharply arched, a tetraradiate os triangulare, and fewer trunk vertebrae. The new species exhibits between four and seven fixed allozyme differences from E. quadridigitata. The two species are largely allopatric in distribution in the Carolinas.

Wildlife and vegetation were sampled on five tree islands in the Arthur R. Marshall Loxahatchee National Wildlife Refuge from June 1999–June 2000 in order to generate a preliminary species list and examine the logistics of using various sampling techniques on bayhead tree islands. Seventy-nine vertebrate species and 19 plant species were recorded. Species richness for plants and vertebrates were similar among islands. Similarity among islands ranged from about 35 to 70% for plants and about 40 to 60% for vertebrates. Richness and similarity values for vegetation were lower than those recorded for tear-dropped shaped islands in wetlands south of the refuge probably due to differences in island topography and geographic location. Number of vertebrate species recorded by sampling method ranged from 2 to 62 with most vertebrate species recorded as observational data rather than during trapping events.

Four years after construction of a 5.54 hectare wetland, we compared its floral composition to that of a nearby, well-established reference wetland and to the original plan for its construction. The created wetland contained 88 % of the plant species originally installed and more total species than the reference wetland. Obligate wetland species comprised 65 % of the plants in the reference wetland and 34 % in the created wetland. The created wetland also contained more early-successional stage species and more facultative upland species than the reference wetland. Weighted averages analysis of the created wetland based on wetland indicator categories indicated appropriate wetland vegetation. The successful performance of the created wetland was due in large part to seed bank transfer from a salvaged marsh surface.

Distributions of two species of fairy shrimps (Crustacea: Branchiopoda: Anostraca) and two species of clam shrimps (Crustacea: Branchiopoda: Laevicaudata and Spinicaudata; formerly Conchostraca) in Carolina bays and other wetland ponds are reported for the Atlantic Coastal Plain province of South Carolina. The distributions are based on extensive sampling of 123 ponds at 11 sites in 10 counties. The fairy shrimp Streptocephalus sealii, which was common (28 ponds), has been previously published as a state occurrence. Here, we present new county occurrences. The other fairy shrimp, Eubranchipus moorei, which was uncommon (3 ponds), has not previously been reported for the state, nor has either of the clam shrimps, Limnadia lenticularis, which was uncommon (3 ponds), and Lynceus gracilicornis, which was common (17 ponds).

We studied nesting success of the Wood Thrush (Hylocichla mustelina) in bottomland and upland hardwood forests in South Carolina. Twenty-one of 26 nests (80.8%) were located in bottomland sites, and 76.2% of these nests were in narrow (<150-m wide) bottomland corridors. No nests were found in upland sites enclosed by fields. The Mayfield success rate for 20 nests was 35.3%. All nest failures were attributed to predation; no nests were parasitized by Brown-headed Cowbirds (Molothrus ater). Nest sites were characterized by a dense overstory and a moderately developed understory. Bottomland hardwoods, especially relatively narrow corridors, appear to provide suitable nesting habitat for Wood Thrush in this region. Brood parasitism by Brown-headed Cowbirds does not appear to be a significant factor in the failure of Wood Thrush nests in these sites.

The effects of ground lichens on seed germination were studied in rosemary scrub on the Lake Wales Ridge in south-central Florida. Lichen removal and seed addition experiments were established within gaps in scrub patches unburned for more than 30 years. Vegetation in these gaps was dominated by ground lichens and vegetatively propagated Licania michauxii and Selaginella arenicola. All ground lichens were removed from one of each of 21 paired plots in 1995 and 1998. Seeds of four short-lived perennial herbs were added to 9 randomly selected plot pairs in 1998. Plots were monitored for seedling and clonal recruitment and death, as well as feral pig damage from 1995–2000. Occasional feral pig disturbance temporarily decreased lichen cover. Our study confirms that seedling recruitment is naturally low in these long-unburned, lichen-covered sites. Seedling emergence was higher, however, in lichen removal plots throughout the experiment, but only significantly so before seed addition. Low recruitment after seed addition was not limited by the availability of seeds and was probably caused by drought. Lichen cover evidently suppresses recruitment in years favorable for recruitment, but variation in precipitation has an overriding effect. Fire suppression leads to increased lichen cover and reduced recruitment of herbs over long periods of time, but remnant larger gaps retain microsites that support recruitment.

We characterized Indiana bat (Myotis sodalis) roosting habitat at three maternity colony sites in western North Carolina and eastern Tennessee. Using radio telemetry, we tracked six bats a total of 40 bat days (range 4–9 days/bat). In 1999, we located a primary roost in an eastern hemlock (Tsuga canadensis) snag (109 cm DBH) in the Nantahala National Forest, NC. In 2000, we located a primary roost in a pine (Pinus sp.) snag (39 cm DBH) in Great Smoky Mountains National Park (GSMNP), TN. Another primary roost was found in a pitch pine snag (P. rigida; 55 cm DBH) in GSMNP in 2001. Largest exit counts for the three colonies were 28, 23, and 81 bats. Primary roost sites were exposed to direct sunlight during most of the day. We also located six alternate roost trees: three pine snags, two red oak (Quercus rubra) snags, and one live sweet birch (Betula lenta). All three primary roosts located in this study were not used during subsequent summers. The eastern hemlock used in 1999 was still standing as of June 2001, while the two primary roosts in GSMNP had fallen within a year of being located. These records represent one of the first descriptions of Indiana bat maternity habitat in the southern United States.

Until recently, eastern woodrats (Neotoma floridana, Ord) at Pine Hills, Union County, were believed to be the only extant population in Illinois. During 1994, additional populations in adjacent Jackson County were discovered at Fountain Bluff, Horseshoe Bluff, and Little Grand Canyon. Genetic variation among Illinois woodrat populations was estimated based on microsatellite DNA at 6 loci. The four populations were separated by 2–14 km, yet all exhibited significant genetic differentiation. Fountain Bluff was the most geographically isolated population and had the lowest genetic heterozygosity. A sample of individuals from a more contiguous population in Missouri exhibited the highest heterozygosity. Populations were monitored by mark-recapture live- trapping for 39 months at Pine Hills and 19 months at Fountain Bluff. Although the Fountain Bluff population was smaller than the Pine Hills population, seasonal demographics and reproductive output were similar. Natural dispersal from all 4 populations was limited by fragmented ahabitat, natural barriers, and anthropogenic barriers. Reintroductions of woodrats may be necessary for the species to repopulate formerly occupied sites throughout southern Illinois.

The year-round food habits of white-tailed deer (Odocoileus virginianus seminolus Rafinesque) were studied in the wet prairie-tree island habitat of the southwestern Everglades system during 1988–91. Microhistological analysis of rumen contents revealed that a single plant, swamp lily (Crinum americanum L.), constituted 40.4% of the annual diet by dry weight and occurred in the rumens of 87.7% of 39 adults (≥ 1 yr in age) and 18 fawns (< 6 mo in age). No other food among the 35 plant items identified in the rumens constituted more than 6.6% dry weight. Swamp lily was the top-ranked food by dry weight for adult females (53.3%), adult males (29.6%), fawn females (27.1%), and fawn males (31.6%). Diet, by food category (e.g., aquatic herbs, woody plants) did not differ between the winter dry season and the summer wet season. Adult females consumed more aquatic herbs (69.8% dry weight) than adult males (38.2%), and, conversely, adult males consumed more woody plants (31.9%) than adult females (18.3%). These dietary differences between adult females and males resulted from the differential use of habitats by the sexes, likely to maximize reproductive success. The diet of fawns did not mirror that of adult females. Fawns consumed more woody plants than did adult females, which probably reflected the use of tree islands by fawns for concealment to avoid predation during their first months of life.

Stomach contents were examined from 289 Atlantic sharpnose, Rhizoprionodon terraenovae, 120 blacktip, Carcharhinus limbatus, and 66 finetooth, Carcharhinus isodon, sharks collected from the Mississippi Sound, and the frequency of occurrence (% F) of each prey item was determined for each shark species. Rhizoprionodon terraenovae consumed the highest diversity of prey items, which included teleost fishes (67.0 %), crustaceans (37.0 %), mollusks (8.0 %) and elasmobranchs (2.5 %). Carcharhinus limbatus and C. isodon fed primarily on teleost fishes, which occurred in 94 % and 100 %, respectively, of stomachs that contained prey. All shark species exhibited a medium to high level of dietary overlap, with the highest level of overlap found between C. limbatus and C. isodon. An ontogenetic shift was observed in the dietary habits of R. terraenovae with young sharks feeding primarily on invertebrate prey (65.2 %), and adult sharks feeding primarily on teleosts (69.0 %) and elasmobranchs (5.8 %).

The Eurycea bislineata (Two-lined salamander) species complex is characterized by extensive geographic genetic fragmentation throughout its range. I tested for sexual isolation and divergence in courtship behavior across a contact zone between E. wilderae and E. cirrigera. I found concordance between patterns of geographic genetic fragmentation and sexual isolation that suggests that divergence in the mate-recognition systems of E. cirrigera and E. wilderae prevents free gene exchange across their contact zone. In addition, the courtship behavior of the E. bislineata complex provides insights into alternative hypotheses of evolution of the mental gland and pheromone delivery behavior in plethodontid salamanders.

The breeding chronology of Florida Snail Kites (Rostrhamus sociabilis plumbeus) was studied at 6 wetlands in central and south Florida during 1987–1993. Average hatch date pooled for all nests was 27 April ± 36.1 days and ranged from 10 February to 29 August (n = 630 nests). Significant differences in mean hatch dates existed between wetlands and among years at individual wetlands. While hatch date was not correlated with longitude, there was a significant positive correlation between hatch date and latitude suggesting a general increase in hatch date as nesting progressed north. Snail Kite breeding cycles did not appear to be closely associated with water levels during the January to July period at either Lake Kissimmee or Lake Okeechobee. Kite breeding also did not appear to be closely associated with air temperatures during the same period at either Lake Kissimmee or Lake Okeechobee. However, the passing of cold fronts early in the year did delay the initiation of new nests.

On the central Appalachian Plateau, evergreen trees and shrubs often dominate riparian vegetation. In contrast, deciduous trees dominate upland forests. To evaluate an important aspect of habitat suitability for Neotropical migratory birds, we placed artificial nests baited with a Japanese quail (Corturnix coturnix japonica) egg at regular intervals along successively upland transects paralleling Poplar Lick Run and Little Savage River, Garrett County, Maryland. Nests were alternated between 0–0.5 m and 1–2 m heights and were exposed for 12 days during each of four trials in 1990, and 12 days during each of six trials in 1991. Egg predation was typically higher within 50 m of these streams than farther from them. Although artificial nests may not truly represent natural nests and the losses incurred by nesting birds, our results should alert land managers and conservationists to a potential problem and stimulate further research into the suitability of existing, evergreen riparian zones for nesting Neotropical migrants.