Evidence for hybridization between three species of Indian paintbrush species: Ecological implications and evolutionary scenarios.

摘要

This dissertation uses computer, field and laboratory based approaches to examine whether hybridization, polyploidy, and natural selection mediated through biotic interactions, influence observed patterns of diversity across a putative hybrid zone between three Indian paintbrush species. These species co-occur across broad regions of North America and in some areas of contact they appear to hybridize whereas in other areas they do not. Previously, hybridization had been suspected based upon floral bract color diversity in some regions of overlap. These species are hemiparasites, use pollinators for reproduction, and experience pest damage; all of which suggested that biotic factors might be important in influencing patterns of phenotypic and genotypic diversity in this system.;First, I examined the statistical underpinnings of phenotypic selection studies and current assessments of the strength of natural selection acting upon phenotypic traits. To detect low levels of selection, I found that several hundred individuals should be observed. Next, I examined evidence for hybridization and explored hypotheses for geographic patterns of phenotypic variability. Using experimental pollinator arrays placed in different areas, I demonstrated that pollinator-mediated hybridization is feasible and that rates of hybrid formation may be context-dependent and spatially variable. I then used a series of hand pollinations along with DNA content and chromosome number surveys to determine that the three species can hybridize in multiple directions and across ploidy levels. This suggested that geographic differences in polyploidy cannot explain the mosaic patterns of hybridization. Next, I examined morphological and genetic evidence for hybridization. I found that the putative hybrids tend to be morphologically intermediate but that they are a genetically distinct coherent group. This suggests that hybridization was an ancient event, and/or that the putative hybrids should be considered separate species. Lastly, I conducted a field census survey and found that the putative hybrids were not more or less resistant or tolerant to pest damage than the three species and that damage may not be an important selective agent for or against different genotypes in this system.;This dissertation includes both my previously published and my co-authored materials.