Salamanders of the Plethodon ouachitae complex are endemic to the Ouachita Mountains of west-central Arkansas and southeast Oklahoma, where they are largely restricted to high-elevation, mesic forest. Because environmental conditions in the intervening valleys are warmer and drier, the salamanders appear to be isolated on "sky islands" where gene flow among populations on different mountains may be restricted. Previous work using allozymes (Duncan and Highton, 1979) found significant genetic variation among populations on different mountains, and it would not be surprising if the number of distinct taxa within this complex is considerably higher than the three currently recognized species (P. caddoensis, P. fourchensis, and P. ouachitae). Dr. Frank Burbrink (College of Staten Island-CUNY) and Kelly Irwin (AR Game & Fish), and I are sampling the region more intensively and using DNA sequence data to analyze patterns of genetic diversity and assess species boundaries. This work will not only tell us more about species diversity in the region, which is important for conservation, but it will provide information about the processes involved in lineage diversification and ultimately the origin of new species.

As of January 2009, we have collected and sequenced the cyt b and ND4 mitochondrial genes (1816 bp) for 281 P. ouachitae from 55 localities and 130 P. fourchensis from 38 localities throughout the Ouachita Mountains. Results showed that P. ouachitae is composed of seven lineages structured across six major mountains and P. fourchensis is composed of four lineages structured across five montane isolates. Our first paper on P. ouachitae came out in Molecular Ecology in December 2008 and our second paper, on P. fourchensis, is accepted to Molecular Ecology and should be published sometime in the spring of 2009.

During the upcoming Spring 2009 field season, we'll concentrate our fieldwork on P. caddoensis and hopefully be able to complete our sampling of this species. Thus far, we have about 200 individuals from 42 localities and our results indicate this species is also composed of several very divergent, geographically structured lineages. In addition to our phylogeographic work employing mitochondrial genes, we are examing morphological variation and looking at several nuclear loci to better understand diversity and evolutionary history of this group.

Related publications: Shepard and Burbrink. 2008. Molecular Ecology. (pdf); Shepard and Burbrink. In press. Molecular Ecology. (pdf)

Predation is a major selective force in the evolution of morphological characteristics of organisms. Visually oriented predators (e.g. birds) develop search images for prey using features such as pattern and the body contour and outline. Lizards that rely on crypsis as a defense have body shapes that enhance their ability to avoid detection whereas lizards that depend on flight possess body shapes that promote rapid locomotion. In December 2004, I used clay models to test whether predator attack rates varied among lizards of four different body shapes. I conducted this experiment in three Cerrado habitats in Brazil with different vegetative characteristics to examine potential differences in attack rates among habitats. Predator attacks did not differ among the model shapes, but did differ among habitats. Predator attacks were most frequent in the most structurally complex habitat and were higher in more exposed microhabitats (Can you tell which lizards on the left are real and which ones are models?).

Related publication: Shepard. 2007. Herpetologica. (pdf)

Frogs of the genus Leptodactylus deposit their eggs in foam nests; however, nest deposition site and the extent to which larvae develop in the foam nest vary among species. Foam nests may be deposited on the water surface, on land at the margins of water, in burrows below ground adjacent to water, or in burrows below ground away from water. Larvae may enter water shortly after hatching or after an extended period in the foam nest, or they may complete development within the foam nest independent of water. Reproductive modes are hypothesized to have a phylogenetic basis, but data to evaluate this are lacking for most species of Leptodactylus. In 2003, Dr. Jan Caldwell and I studied the reproductive ecology of Leptodactylus labyrinthicus at a locality in the Cerrado region of Brazil. We found that foam nests were deposited on land about 2 m from the water's edge. Less than 20% of the eggs in nests developed and tadpoles remained in the nest after hatching, feeding on the undeveloped eggs. Heavy rains washed tadpoles out of foam nests and into adjacent ponds where they then fed on the eggs of other frogs, primarily small hylids.

Related publication: Shepard and Caldwell. 2005. Copeia. (pdf)

Alternative mating tactics occur in many species. In toads, non-calling males parasitize calling males and attempt to intercept females. These satellite males are usually smaller in size than the calling males they parasitize and this has led to the assumption that because they are smaller, they are also younger. Chris Leary and I analyzed size and age data on caller and satellite Bufo cognatus and Bufo woodhousii and found that although callers were larger in size than satellites, they were not different in age. The size difference between callers and satellites of the same age was the result of different growth trajectories, with callers growing faster.

Related publication: Leary et al. 2005. Animal Behaviour. (pdf)

This project was my Master's work at Illinois State University. I compared aggession, site tenacity and spatial relationships of male Green Frogs (Rana clamitans) during the breeding and non-breeding seasons to test if their behavior was consistent with territoriality. I found that although males are aggressive during the breeding season, their site tenacity and spatial relationships were not entirely consistent with territoriality.

Related publications: Shepard. 2002. American Midland Naturalist. (pdf); Shepard. 2004. Copeia. (pdf)