No abstract

No abstract

When a widespread species is sympatric with a rare, geographically restricted conspecific, recurrent gene flow can pose the risk of extinction for the latter. This can occur via genetic swamping, where the occurrence of hybrids eventually replaces the numerically less abundant species. We took a molecular genetic approach to quantify the occurrence and degree of contemporary hybridization between two species of frog co-occurring in a small geographic area in northwest Florida, USA. The Florida bog frog Lithobates okaloosae is a small ranid limited in distribution to a few acidic seepage and steephead streams that feed the Yellow, Shoal and East Bay river drainages in Walton, Okaloosa and Santa Rosa counties, Florida. The bronze frog Lithobates clamitans is a widespread (Eastern North America) congener. Data from nine microsatellite loci and 350 frogs were analyzed using Bayesian clustering (STRUCTURE) and a Bayesian hybrid classification method implemented in NEWHYBRIDS. Power to detect hybrids with the dataset was assessed through simulations. Both methods detected hybrids in similar proportions (5-10%), including congruent confirmation and rejection of samples previously classified as putative hybrids based on phenotypic characteristics. However, the nine loci lacked sufficient power to differentiate among F 1 and backcrossed individuals, leaving open the question of the extent of genetic introgression. Longitudinal genetic monitoring is recommended to evaluate whether the observed level of hybridization represents a long-term threat to the distinction of one of North America's most geographically restricted frogs.

The Florida bog frog (Rana okaloosae) is restricted to approximately 25 seepage drainages on the Florida Panhandle, southeastern USA. We evaluated finescale (\1-10 km) genetic structure among 80 samples from a long-term study area in one portion of its range. We also included co-distributed samples (N = 48) from bronze frogs (R. clamitans). Individual R. okaloosae were significantly more related to one another than expected under panmixia, though significant patterns of isolation-by-distance were detected reflecting limited dispersal. Bayesian clustering failed to identify discrete genetic clusters within species. Subtle, though important differences in genetic structuring between R. okaloosae and R. clamitans suggests that future efforts to predict the impact of landscape changes on Rana okaloosae, should focus on the species itself rather than R. clamitans as a surrogate species.