Genetic Diversity of Ruth's Golden Aster (Pityopsis ruthii), An Endangered Species

Ruth’s golden aster (Pityopsis ruthii) is federally endangered species endemic to the Hiwassee and Ocoee Rivers in Tennessee. Plants grow in crevices on exposed rocks that are in and between the river channel and the adjacent forested slopes. Populations occur downstream of hydroelectric dams. The available long-term data, though limited in scope, suggests that substantial declines in P. ruthii have occurred since monitoring began in 1986, especially along the Hiwassee River. The reach of the Ocoee River where the plants occur regularly receives water flows, which maintains open habitat for the species. Run-off from precipitation and input from small tributaries comprise the normal flow in the river and represents a small portion of the total flow formerly present in the Hiwassee River. Without normal flow in the river and the associated disturbance, multiple woody and herbaceous plants are colonizing the habitat and displacing P. ruthii. Knowledge of genetic population structure is essential for understanding the magnitude of seed dispersal, genetic drift and selection pressures. The attributes of the genetic structure of P. ruthii, in particular genetic erosion in isolated and fragmented populations, may exacerbate the effects of habitat degradation. To assess genetic diversity, 134 samples from 4 discrete locations were genotyped with 12 microsatellite loci. The number of alleles per locus, observed (H_o) and expected (H_e) heterozygosity, and the proportion of genetic variance (F_ST) were calculated using ARLEQUIN and FSTAT, and genetic structure of the population was evaluated using STRUCTURE. The number of alleles per locus ranged from 4 to 13, H_o was 0.41 and H_e was 0.58, and F_ST was 0.32. STRUCTURE analysis assigned the individuals into three distinct clusters that corresponded well to the a priori grouping of plants from the sampled populations. The AMOVA analysis found that most of the genetic variation was attributed from variance among all individuals (49.10%) rather than among three groups indicated by STRUCTURE (23.47%), among populations within groups (9.24%), and among individuals within populations (18.19%). These results imply that gene flow is limited between the rivers and that nonrandom mating is occurring within populations and warrants further examination of population structure, mating systems, and pollen and seed dispersal.