Bumble bees (Bombus spp.) are adept pollinators of countless cultivated and wild flowering plants, but many species have experienced declines in recent decades. Though urban sprawl has been implicated as a driving force of such losses, urban green spaces hold the potential to serve as habitat islands for bumble bees. As human populations continue to grow and metropolitan areas become larger, the survival of many bumble bee species will hinge on the identification and implementation of appropriate conservation measures at regional and finer scales. North Texas is home to some the fastest-growing urban areas in the country, including Denton County, as well as at least two declining bumble bee species (B. pensylvanicus and B. fraternus). Using a combination of field , molevular DNA and GIS methods I evaluated the persistence of historic bumble bee species in Denton County, and investigated the genetic structure and connectivity of the populations in these spaces. Field sampling resulted in the discovery of both B. pensylvanicus and B. fraternus in Denton County's urban green spaces. While the relative abundance of B. fraternus in these spaces was significantly lower than historic levels gleaned from museum recors, that of B. pensylvanicus was significantly higher. Statistical analyses found …

An examination of the effectiveness of seeding, burning, and mowing in the reestablishment of tallgrass prairie species on overgrazed and abandoned pastureland. The study site is a 20 acre tract on U.S. Corps of Engineers land below Lake Lewisville in Denton County, Texas. The site was partitioned into thirty-nine 40 by 40 meter plots with seeding (carried out in 1996) and management treatment (burning, mowing, and no maintenance carried out in 1998) randomly applied following a two level design. For each plot, nine stratified-random 0.1 m2 subplots were examined and shoot counts for each species recorded. The effects of the treatments on individual species and species richness were analyzed with a two-way ANOVA followed by a SNK multiple range test, both on ranked data. Community level analysis was conducted with both a MANOVA on ranked data and a Canonical Correspondence Analysis on raw data. Results indicate that seeding positively affected species richness, particularly when combined with either burning or mowing in the early spring. Mowing also significantly increased species richness in areas that were not seeded, while burning negatively affected species richness on unseeded plots. Treatments significantly affected community composition with treatments having the most clear effect on spring and …