Mechanisms facilitating and evolutionary consequences of gene flow in two crop-wild hybrid complexes: Sunflower and rice

Most of the world's crops hybridize where they co-occur with related wild species, yet the evolutionary consequences of this process remain poorly understood. The dispersal of genes from crops to their wild relatives can have important implications for agriculture, especially if crop genes persist in wild populations. For example, hybridization between crops and wild species has been linked to the evolution of more aggressive weeds. In contrast, extensive hybridization with endangered wild relatives could "swamp out" native genes, and displace valuable germplasm essential for future crop breeding. My dissertation focused on two aspects of crop-wild gene flow: (1) How does gene flow occur? and (2) How can germplasm materials maintained in wild relatives be conserved?; In chapter 2, I compared flowering times and other phenotypic traits of volunteers from both "normal" and "off-type" (multi-headed) crop plants with those of wild sunflowers. I observed sufficient overlap in flowering times to allow gene flow among the three plant types. My results suggest that both types of crop volunteers have the potential to act as conduits for gene exchange between cultivated and wild sunflowers.; In chapter 3, I used a matrix population modeling approach to identify parts of the sunflower lifecycle that are most influential in determining crop-wild gene flow. The model demonstrated that seed banks and large initial recipient population size strongly influence final crop allele frequencies. Most significantly, my results suggest the facility with which introgression of crop alleles occurs will be contingent on local conditions. Confinement strategies designed to limit the dispersal of transgenes may not be universally effective.; In chapter 4, I used microsatellite markers, along with Bayesian methods that consider recent fragmentation events, to develop an in situ conservation strategy for wild rice Oryza rufipogon in the Mekong Delta of Vietnam. O. rufipogon is currently common in the Mekong but rapid development could decrease availability of habitat. Estimates of genetic diversity and genealogical history were obtained by screening 378 individuals from 19 O. rufipogon populations (∼20 individuals per population). My analysis found that O. rufipogon populations were highly differentiated and gene flow among populations was limited.