Seedling Recruitment in Fragmented Populations of Echinacea angustifolia

Human activities have resulted in extensive habitat fragmentation, one consequence of which is the reduction in size of remnant plant populations. Small populations may experience increased inbreeding and genetic drift, and are more vulnerable than large populations to demographic and environmental stochasticity. This dissertation describes three research questions related to the effects of habitat fragmentation on Echinacea angustifolia , a species native to the Great Plains of North America.;First, I ask whether small remnant populations of E. angustifolia are declining relative to larger populations. Population projection matrices, parameterized with data from two transition years, predicted population declines. A matrix for the 5 smallest populations had significantly lower population growth rate than a matrix for the large populations. Mortality was higher for seedlings than for other stages of the life cycle; however, survival of vegetative and reproductive plants made a greater contribution to population growth rate than did either emergence or survival of seedlings. This is one of only a few studies to find a correlation between population size and population growth rate.;Second, I ask whether genetic declines in remnant E. angustifolia populations can be remedied by outcrossing. I performed within-population and between-population crosses using six populations of varying size. Responses to cross type depended on the populations represented and on the fitness trait evaluated. Between-population crosses involving the smallest populations resulted in offspring with higher fitness than the offspring of within-population crosses (i.e., heterosis), while other between-population crosses showed outbreeding depression. Due to the risks of outbreeding depression, managers are cautioned against attempting genetic rescue by performing between-population crosses.;Third, I ask whether E. angustifolia populations exhibit local adaptation. I collected and reciprocally sowed seeds from three widely-separated populations located along a 500-km transect from western South Dakota to western Minnesota. Seed source and sow site had significant effects on seedling emergence and survival, but no clear evidence of local adaptation among the three populations was observed. All three source populations had highest emergence in Minnesota and lowest emergence in western South Dakota. The western South Dakota and Minnesota-sourced seed had higher emergence than the central South Dakota seed at all three sowing sites. The western South Dakota seedlings had the highest survival and the Minnesota seedlings had the lowest survival over the first growing season at all three sites.