Thesis (Ph. D.)--University of Missouri-Columbia, 2000. / Typescript. Vita. Includes bibliographical references. Also available on the Internet.

Thesis (Ph. D.)--University of Missouri-Columbia, 2000. / Typescript. Vita. Includes bibliographical references. Also available on the Internet.

Differences in patterns of species richness and taxonomic composition across continents are well documented. However, less is known about how these patterns originate from the fundamental processes that contribute to the assembly of continental biotas: speciation, extinction, immigration, and emigration. To truly decipher how these processes operate at a continental scale, it is crucial to understand how Earth-history events and environmental change shaped the biogeographical history of the taxa occurring in a region. The Pleistocene glacial cycles have been hypothesized to be a significant geological event which affected the Earth’s biota over the past three million years. During this period, multiple cycles, in which ice sheets covered a large portion of the Northern Hemisphere, have been hypothesized to facilitate allopatry and ecological differentiation. The central goal of this dissertation is to understand the assembly of continental biotas by integrating the role of Earth’s geological and environmental history and recent (late Quaternary) changes in distributions. Here, I studied the North American wood-warblers, which are passerine birds belonging to the family Parulidae. In the first chapter of this dissertation, I examined the extent to which recent diversification is underestimated by ignoring recently diverged taxa. To do so, I evaluated the effect of taxonomic delimitation on the inferred temporal patterns of diversification of wood-warblers in the genus Setophaga. My results show that species-level taxonomic delimitation in ecological and evolutionary studies is crucially important but is often ignored. Evaluating the effect of taxonomic delimitation in the genus Setophaga is particularly relevant because it has been widely cited as an ideal example of niche saturation, and previous studies suggested that lineages in this particular radiation exhibited an asymptotic accumulation of diversity through time. In this chapter, however, I demonstrate that this pattern was likely a consequence of the ways in which taxonomic diversity was sampled. In the second chapter, I examined how biotic and abiotic factors limit the distribution of species of the genus Oreothlypis at a continental scale. For most of the taxa examined, climatic- and biotic-based areas of suitability were equivalent, and therefore the relative importance of these factors could not be disentangled. However, in some cases, biotic variables limited the distribution in areas climatically suitable, and vice versa. The results of this chapter highlight the importance of considering potential effects of biotic interactions in the study of climate-driven range shifts. This paper is also an important methodological contribution to the general field of ecological niche modeling (ENM) by integrating climatic and palynological data to empirically estimate both abiotically and biotically suitable areas which that has only been done theoretically so far. In the third chapter, I integrated phylogenetic data, biotic and climatic ENMs, and reconstructions of environments during the LGM to test how cycles of fragmentation, differentiation, and expansion during the Pleistocene shaped the biogeographical patterns of the genus Oreothlypis in North America. Based on a time-calibrated phylogeny I identified two groups of taxa that diversified during the last million years and therefore were potentially affected by glacial cycles. My analyses suggest that there were rapid switches in the environmental conditions in which species of the genus occur and that glacial cycles resulted in dramatic range shifts from glacial maxima to interglacials. Distribution patterns during the LGM suggest that divergence in areas isolated by glaciers presumably contributed to the geographical structuring of Oreothlypis, as well as to their taxonomic and ecological diversity in the present. In summary, this work illustrates how wood-warblers are an ideal model system for examining the large-scale history of the North America biota and environment, particularly over the Pleistocene. They are highly diverse, have endemics in virtually all of North America's areas of endemism and ecosystems, and many of these patterns are repeated within and across clades. Furthermore, distributional patterns in warblers show high congruence with those of other organisms; thus inferences made about the history of this group likely have implications for understanding the biotic history of North America in general. In addition, many warblers have narrow ecological preferences in that they occupy forest environments with specific tree compositions. As a consequence, ecological models of their distributions, integrated with the North American pollen record of forest change, provide a new perspective on ecosystem change during glacial cycles, and its impact on the origin and maintenance of biodiversity in the continent.

Presented in this thesis is the mitochondrial DNA (mtDNA) cytochrome b gene analysis of the phylogenetic relationships within the Apalis thoracica complex (600bp) as well as among selected African Bradypterus species (603bp). Within A. thoracica 28 individuals from 20 subspecies were analysed. There was only molecular support for 13 historically isolated clades as opposed to the 21 subspecies recognized based on morphology. Among these molecularly defmed clades were clades lynesi, juscigularis and flavigularis, all three of which are of conservation concern and were previously treated as species by some based on their distinct morphology. Molecular clock dating estimated the clades to be of Pliocene age, whereas genetic differentiation within clades was of Plio-Pleistocene to Pleistocene age. This result is consistent with that of other African montane birds. Under the phylogenetic species concept all 11 clades could be recognised as species although a multifaceted approach to species recognition which will include vocalisation data is suggested. Contrary to expectation, the isolated East African clades did not exhibit more variation than the continuously distributed clades within southern Africa. In fact, significant phylogeographic breaks were identified within South Africa that need further investigation. For the African Bradypterus, 13 individuals from six of the 10 African species were analysed. These species were not monophyletic. Based on sequence divergence B. victorini was as different from the other members of the genus as it was from the out groups. In addition its song is very distinct and both males and females sing. Therefore, based on the molecular results presented in this thesis as well as vocalisation data and, it is proposed that B. victorini is not a Bradypterus. This result is particularly significant in light of the growing evidence that the Cape region acted as a repository for ancient animal taxa and holds implications for the conservation status of the region. The molecular data supported the classification of the taxa based on habitat and song but in addition clarified the placement of B. victorini and B. sylvaticus. The basal position of the latter among the remaining members of the genus indicates that Bradypterus is primarily a forest taxon that radiated into other habitats. It must however be kept in mind that the current dataset is based on six of the 10 African Bradypterus and none of the Asian species. The possibility of hybridisation between B. sylvaticus and B. barratti was raised and needs to be tested through additional sampling. / Dissertation (MSc (Genetics))--University of Pretoria, 2006. / Genetics / unrestricted