Evolution of genotypes during range expansion is driven in part by colonization dynamics. I investigated genetic patterns of colonization and dispersal during initial expansion of an invasive bunchgrass, Brachypodium sylvaticum, into Oregon. Using microsatellite markers, I sampled plants at two different scales: at regular intervals along three parallel roads spanning about 30km, and in populations identified throughout Oregon. I also collected field-generated progeny from a subset of populations and used molecular identification of outcrossing events to estimate selfing rates in both central and peripheral populations. Dispersal patterns were similar at both scales, with non-contiguous dispersal responsible for colonization of new populations. High levels of differentiation were observed at all scales, though newly-colonized populations were more differentiated than older populations. Corvallis populations were responsible for colonization of a majority of populations throughout Oregon, while individuals from Eugene were only occasionally found in new populations. Admixture occurs between Corvallis and Eugene populations, decreasing differentiation, and potentially creating novel phenotypes and increasing evolutionary potential of populations. Selfing rates were high, but two populations in the areas of original introduction had lower rates of selfing, suggesting that selfing rates may decrease as population density and diversity increases with age. The influences of founder effects and bottlenecks on phenotypic evolution during range expansion require further investigation, as inbreeding, lag times, and selection may influence evolutionary trajectories of populations.

Rapid range expansion of newly invasive species provides a unique opportunity for studying patterns of dispersal and gene flow. In this thesis, I examined the effect of landscape features on gene flow in the invasive grass Brachypodium sylvaticum at the edge of its expanding range. I used genome-wide Single Nucleotide Polymorphism (SNP) surveys of individuals from 22 locations in the Clackamas Watershed in the Portland, Oregon metropolitan region to assess genetic diversity and structure, to identify putative source populations, and to conduct landscape genetic analyses. Resistance surfaces were created for each landscape feature, using ResistanceGA to optimize resistance parameters. My STRUCTURE analysis identified three distinct clusters, and diversity analyses support the existence of at least two local introductions. Multiple Regression on distance Matrices (MRM) showed no evidence that development, roads, canopy cover, or agriculture had a significant influence on genetic distance in B. sylvaticum. The effect of geographic distance was marginal and reflected geographic clustering. The model of rivers acting as a conduit explained a large portion of variation in genetic distance. Results indicate that rivers influence patterns of dispersal of B. sylvaticum by human recreational activity centering on use of rivers, and possibly due to movement of deer.