Dispersal Pathways and Genetic Differentiation among Worldwide Populations of the Invasive Weed Centaurea solstitialis L. (Asteraceae)

The natural history of introduced species is often unclear due to a lack of historical records. Even when historical information is readily available, important factors of the invasions such as genetic bottlenecks, hybridization, historical relationships among populations and adaptive changes are left unknown. In this study, we developed a set of nuclear, simple sequence repeat markers and used these to characterize the genetic diversity and population structure among native (Eurasian) and non-native (North and South American) populations of Centaurea solstitialis L., (yellow starthistle). We used these data to test hypotheses about the invasion pathways of the species that were based on historical and geographical records, and we make inferences about historical relationships among populations and demographic processes following invasion. We confirm that the center of diversity and the native range of the species is likely the eastern Mediterranean region in the vicinity of Turkey. From this region, the species likely proceeded to colonize other parts of Europe and Asia via a slow, stepwise range expansion. Spanish populations were the primary source of seed to invade South America via human-mediated events, as was evident from historical records, but populations from the eastern Mediterranean region were also important. North American populations were largely derived from South America, but had secondary contributors. We suggest that the introduction history of non-native populations from disparate parts of the native range have allowed not just one, but multiple opportunities first in South America then again in North America for the creation of novel genotypes via intraspecific hybridization. We propose that multiple intraspecific hybridization events may have created especially potent conditions for the selection of a noxious invader, and may explain differences in genetic patterns among North and South America populations, inferred differences in demographic processes, as well as morphological differences previously reported from common garden experiments.