, Volume 11, Issue 3, pp 835-854
Date: 07 Apr 2009

Genetic diversity in a narrowly endemic, recently described dusky salamander, Desmognathus folkertsi, from the southern Appalachian Mountains

Rent the article at a discount

Rent now

* Final gross prices may vary according to local VAT.

Get Access

Abstract

To understand patterns of biodiversity and whether populations and species are in decline, the detection and description of cryptic biodiversity are essential. Salamanders are of particular conservation interest because they potentially harbor many undescribed species due to morphological conservatism. The dusky salamanders, genus Desmognathus, are a species-rich group in which morphologically cryptic species are especially common. Using a portion of the mitochondrial genome and amplified fragment length polymorphism (AFLP), we investigated the genetic diversity of the highly endemic, stream-dwelling salamander, Desmognathus folkertsi, across its known range in the Appalachian Mountains. Mitochondrial data revealed three well-supported lineages, one of which consisted of only one specimen; however, AFLP data were not congruent with the mitochondrial data. There was 1.11% uncorrected sequence divergence between the two well-sampled lineages. Desmognathus folkertsi exhibited 4.29% sequence divergence from the closely related D. quadramaculatus. Isolation by distance was found for both the AFLP and mitochondrial data when stream distance, rather than when straight-line (i.e., geographic) distance was used. Although genetic diversity is often partitioned by river drainages in freshwater taxa, we did not observe such a pattern in D. folkertsi. We propose that human-mediated dispersal by bait-bucket release may augment natural gene flow via aquatic dispersal across streams. Because this species was only recently discovered, the full extent of the geographic range is unknown. Therefore, an ecological niche model, using climate variables and the Maxent algorithm, was used to determine whether additional regions may be suitable for the species. The model predicted a small range limited to extreme southwestern North Carolina and extreme northeastern Georgia. We suggest future surveys be focused in these regions.