Variation in morphology and genotype among individuals of the same species occurs almost everywhere in the natural world. Such variation underlies natural selection and any resulting evolution, and therefore needs to be quantified in order to predict and understand phenotypic and evolutionary dynamics. Where morphology or genotype vary non-randomly in space, structured spatial variation can arise. Such spatial variation can arise from, and cause, further evolutionary processes including local adaptation and speciation. Here, I quantify spatial variation in neutral genetics and morphology, and in selection on morphology, occurring at a very small spatial scale within a single population of European starlings (Sturnus vulgaris) resident to the Island of Fair Isle. I demonstrate weak but statistically significant genetic structure in microsatellite loci by FST comparison but that similar structure is not supported by Bayesian cluster analysis. I also show significant genetic structure between populations of starlings within the UK from England, Scotland, Colonsay, Orkney and Shetland, and show that there are two main genetic clusters supported by Bayesian cluster analysis. I show there are significant differences in weight, tarsus lengths, bill lengths, bill size and, body condition among three, spatially proximate, spatial areas within Fair Isle thereby demonstrating significant small-scale spatial variation in morphology within this island population Using capture mark recapture analysis to estimate survival probabilities I evidence the occurrence of stabilizing selection on weight and disruptive selection on tarsus length in starlings but find no evidence that selection on morphology differed between sexes or among areas within Fair Isle. I conclude that small scale spatial variation in morphology and neutral genetics can occur on small spatial scales even in species with high potential mobility and where there are no distinct differences in environment and movement is not physically inhibited.

The past fifteen years have seen a revolution in human population genetics. We have gone from anecdotal genetic data from a few individuals at a few genetic loci to an avalanche of genome-wide sequencing data, from many individuals in many different human populations. These new data have opened up many new directions of research in human population genetics. In this work, I explore two such directions. Genomic data have uncovered that recent changes in human population size have had dramatic effects of on the genomes of different human populations. These effects have raised the question of whether historic changes in population size have led to differences in the burden of deleterious mutations, or mutation load, between different human populations. In Chapter 1 of this thesis, I show that despite earlier arguments to the contrary only minor differences in load are expected and indeed observed between Africans and Europeans. Over the past decade, genome-wide association studies (GWAS) have begun to systematically identify the genetic variants underlying heritable variation in quantitative traits. The number, frequencies and effect sizes of these variants reflect the selection, and other evolutionary processes, acting on traits. In Chapter 2, I develop a model for traits under pleiotropic, stabilizing selection, relate the model’s predictions to GWAS findings, and show that GWAS findings for height and BMI indeed follow model predictions. In Chapter 3, I develop a method to infer the distribution of selection coefficients acting on genome-wide significant associations made by GWAS.

No description available.

The systematic relationship of two species of the Peromyscus truei group (P. truei and P. difficilis) was analysed through the application of starch gel electrophoresis, numerical taxonomy, and chromosomal techiques. Of 20 loci examined, 11 were monomorphic in all populations, two exhibited variation in only two populations, and seven loci were polymorphic in two or more populations. The mean number of polymorphic loci per population was 0.186, the mean number of polymorphic loci per individual was 0.024, and the proportion of loci heterozygous per individual was 2.4%. Chromosomal forms of P. truei, P. t. gentilis (FN 54) and P. t. truei (FN 62), and P. difficilis, P. d. petricola (FN 56) and P. d. nasutus (FN 58), were consistent for their karyotypes throughout their geographic ranges. No chromosomal hybrids were detected. Numerical analysis of morphological characters and similarity values based on allelic frequencies utilizing Roger's coefficient (S) demonstrated a distinct seperation of karyotypic forms of P. truei (S = 0.902) and P. difficilis (S = 0.924) and were below the mean value of S for conspecifics (S = 0.950). All data indicates that each chromosomal form of the P. truei group examined represents four distinct species. The oldest available name for chromosomal forms of P. true with a fundamental number of 62 is Peromyscus truei Shufeldt (1885) and the oldest available name for chromosomal forms with a fundamental number of 54 is Peromyscus gratus Merriam (1898). The oldest available name for chromosomal forms of P. difficils with a fundamental number of 58 is PeromYSCus nasutus Allen (1891) and the oldest available name for chromosomal forms with a fundamental number of 56 is Peromyscus difficilis Allen (1891).

Xu Jiawu. / "March 2005." / Thesis (Ph.D.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (p. 103-120) / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web. / Abstracts in English and Chinese.

Until relatively recently, synonymous codons have been thought to be under very little evolutionary constraint. That has changed, and now codon bias (the preferential use of a subset of synonymous codons for each degenerate amino acid) is being observed in nearly every organism in which it is adequately searched for. Synonymous codons are evolving under weak selection. As such, it is expected that the selective forces driving the distributions of these codons to be more efficient in species with larger effective population sizes than species with smaller population sizes. With this in mind I investigate patterns of weak selection in the Drosophila subobscura species complex. D. subobscura is a widespread, continental species with a Palearctic distribution. In contrast, its sister species D. guanche and D. madeirensis are island endemic species with low population sizes that each share a relatively recent common ancestor with D. subobscura. In Chapter 2, I investigate patterns of molecular divergence between D. subobscura and the two island species. I show that there is considerable selection acting on synonymous codons in D. subobscura, but not as strong in the island species (although it still exists). The main consequence of this is that purifying selection acting against unpreferred codons is active in D. subobscura, but synonymous codons are much less constrained in the island species. This causes the common measurement of the ratio of nonsynonymous to synonymous mutations to appear larger in D. subobscura while both components are actually increased in the island species. In Chapter 3, I focus on D. subobscura and look at patterns of polymorphism within this species. I show that there is a genome wide skew towards low frequency variants, while only the unpreferred to preferred mutation class segregates at higher than expected frequencies. This is indicative of purifying selection acting upon most of the genome, and positive selection pushing preferred codons to higher frequencies.

Limestone mining in the San Franicso Mountain Range of west central Utah threatens the survival of a rare endemic species of buckwheat (Eriogonum soredium). This species is an edaphic endemic, only found growing on the outcrops of the Ordovician limestone mines in the area. Eriogonum soredium is a candidate for governmental protection under the Endangered Species Act (ESA). However, a common, widespread buckwheat (Eriogonum shockleyi) appears to be closely related to the narrow endemic. The genetic relatedness of the rare and and common species will greatly influence the decision of United States Fish and Wildlife Service (USFW)of whether or not to list the rare species for governmental protection. This study investiaged the amount of genetic divergence between the two species to facilitate the decision. I found levels of population divergence intermediate between a state of no genetic distinction, and complete genetic divergence. However, the two species fall near the genetic divergence end of the continuum. This situation is not uncommon in plants, and suggests that the two species are currently in the process of speciation. Considering their morphological differences, and the ability of the genus Eriogonum to hybridize, these two species show significant amounts of divergence. These results suggest that the continued treatment of E. soredium as distinct from E. shcokelyi is warranted. The USFW will use the results of this study to aid their decision of whether or not to list E. soredium under the ESA. Should the species be listed for protection under the ESA, limitations to the expansion of limestone mining in the San Francisco Mountain Range will be considered.

Bibliography: leaves 180. This study shows the widespread nature of resistance in L. rigidum populations, the tendency to genetic dominance even in early generations of selection and the propensity to display resistance to many herbicides. The fitness of resistant plants and the likely failure of herbicides for weed management suggests that an Integrated Weed Management programme is the best solution.

Thesis (Ph.D.)--York University, 2004. Graduate Programme in Biology. / Typescript. Includes bibliographical references (leaves 109-136). Also available on the Internet. MODE OF ACCESS via web browser by entering the following URL: http://wwwlib.umi.com/cr/yorku/fullcit?pNQ99143

Thesis (Ph. D.)--Ohio State University, 2006. / Title from first page of PDF file. Includes bibliographical references (p. 150-163).