Dispersal ability and reproductive success are important criteria in determining whether a threatened or endangered species is capable of recovering from bottlenecks and catastrophic population declines. The goal of this dissertation is to understand patterns of population genetic variation and population connectivity (i.e. gene flow), and their implications for the conservation of freshwater mussels (Bivalvia: Unionidae). I generated mitochondrial DNA sequence data and nuclear microsatellite data sets for five mussel species (Leptodea leptodon, Le. fragilis, Lampsilis abrupta, Cyprogenia aberti, C. stegaria) and one fish species (Aplodinotus grunniens) in order to evaluate genetic diversity and population structure within each of these species.

In the first of three chapters, I assessed the genetic variation of two nominal species in the genus Cyprogenia (C. aberti and C. stegaria) in order to delineate the number of evolutionary lineages present within the genus. Based on the molecular data collected from mitochondrial DNA and microsatellite genotyping, I found evidence of mito-nuclear discordance in Cyprogenia. Analyses of mtDNA sequences suggest that two deeply divergent clades co-occur sympatrically in most of our sampling sites; the nuclear microsatellite data support three allopatric clades that correspond to major hydrologic basins. My study also showed that the pigmentation of mussel conglutinates in Cyprogenia is highly correlated with the recovered mtDNA clades. Furthermore, the mtDNA sequences appear to be under selection and therefore not reliable for delineating species, and the recognition of species and evolutionarily significant units was based solely on the microsatellite data.

The second chapter investigates the impact of host fish dispersal on the gene flow of a common and a rare mussel species. I compared the gene flow pattern and population structure of an endangered mussel species Leptodea leptodon and its common and widespread congener Le. fragilis with their shared host fish, the freshwater drum Aplodinotus grunniens. The results showed that population structure of Le. fragilis were more congruent with that of the host fish, while the endangered Le. leptodon populations displayed higher levels of genetic isolation among drainages.

In the third chapter I investigated contributions to mussel gene flow from two sources: gamete-mediated gene flow contributed through sperm dispersal, and zygote-mediated gene flow contributed through larval (glochidia) dispersal. The purpose of this study was to determine whether sperm dispersal or glochidia dispersal (via host fish) contribute more to maintaining connectivity among mussel populations. I developed an approach for estimating sperm gene flow of mussel populations when a paternally inherited genetic marker is not available. This is the first study that attempts to measure the gamete-mediated gene flow (male gene flow) in mollusks using mtDNA and nuclear markers. The results for three mussel species (Le. leptodon, Le. fragilis, and Lampsilis abrupta) showed that sperm gene flow among populations is higher than previously assumed.