Ecological genetics of flower color variation in Southern California bush monkeyflowers

Geographic variation in ecologically important traits suggests a role for adaptation and natural selection in diversification. However, historical isolation and secondary contact can also shape patterns of phenotypic variation, and distinguishing between these scenarios is often difficult. In this dissertation, I explore how geographic variation in flower color in populations of Mimulus aurantiacus (Phrymaceae) has arisen and been maintained in nature. In San Diego County, populations are distributed parapatrically. Coastal (western) populations have red flowers, inland (eastern) populations have larger yellow flowers, and a narrow hybrid zone exists where these forms come into contact. I investigate whether the current pattern of floral variation has been influenced by historical barriers to gene flow or whether (and how) different components of natural selection have shaped it. Finally, I examine the genetic basis of differences in these floral traits to gain insight into the complexity of the genetic systems characterizing floral diversification.; Flower color is markedly structured and has a strong genetic component. However, variation at neutral loci is far less pronounced. This contrasting pattern of floral and molecular variation between populations suggests that divergent natural selection, as opposed to recent secondary contact, is responsible for creating and maintaining floral variation in San Diego County.; Reciprocal transplant experiments demonstrate that red-flowered plants perform better than yellow-flowered plants, suggesting no local adaptation for viability or growth measures. Hummingbird and hawkmoth pollinators show dramatic preferences for alternative floral types, with hummingbirds preferring red flowers and hawkmoths visiting exclusively yellow flowers. Pollinators have likely contributed to the dramatic differentiation in floral traits, but perhaps because of recent changes in relative hummingbird abundance in inland regions, red flowers currently receive more visits than yellow flowers everywhere. This suggests that current and historical patterns of selection may differ, and red-flowered plants may be advancing eastward.; A simple genetic basis appears to underlie this dramatic phenotypic differentiation which is associated with visitation by different pollinators. Statistical analyses demonstrate that between one and two genes control flower color differences in M. aurantiacus, which confirm recent reports that, at least for some traits, a small number of genes can have substantial effects on adaptive population differentiation.