Differences in reproductive strategies are a major factor influencing the patterns of genetic variability. Inbreeding and other non-recombining breeding systems can have profound effects on the efficacy of natural selection, which should be manifested in the patterns of genetic diversity within and between species. The impact of an organism’s breeding system can be investigated through a number of approaches. In this thesis, I use mathematical modelling, computer simulations, breeding schemes, quantitative life history measures, and molecular biological techniques to explore many of the consequences of breeding system evolution. Following a general introduction in Chapter 1, I explore the dynamics of transposable elements (TEs)—selfish mobile sequences of DNA that have deleterious effects upon their hosts. Sexual reproduction and recombination are important for constraining TE abundance, and in the absence of sex, an unchecked proliferation of TEs may cause a population to go extinct. In Chapter 2, I use a theoretical framework to analyze TE dynamics under asexual reproduction. Here, I show that while small populations are driven to extinction by element accumulation, large asexual populations can prevent this fate and be cured of vertically transmitted TEs. These results may help explain an "evolutionary scandal": the persistence of ancient asexual lineages, such as the bdelloid rotifers. In Chapter 3, I extend the computer simulations used in the previous chapter to explore the effects of reduced recombination on the distribution and abundance of TEs in sexual populations. I show that TEs become fixed as a result of Hill-Robertson effects in the form of Muller’s ratchet, but only in regions of extremely low recombination when excision is effectively absent and synergism between elements is weak. These results should help explain genomic patterns of TE distributions. In the remainder of the thesis, I turn to testing the genetic effects of androdioecy—the breeding system in which populations are comprised of separate male and hermaphrodite individuals—using the nematode Caenorhabditis elegans and related species. This unusual breeding system promotes high levels of inbreeding, yet males are maintained at appreciable frequencies. In Chapter 4, I measure lifehistory traits in the progeny of inbred versus outcrossed C. elegans and the related outcrossing species, C. remanei, to compare levels of inbreeding depression. I show that highly inbred C. remanei show dramatic reductions in brood size and relative fitness compared to outcrossed individuals, whereas pure strains of C. elegans performed better than crosses between strains, indicating outbreeding depression. The results are discussed in relation to the evolution of androdioecy and the effect of mating system on the level of inbreeding depression. Like C. elegans, C. briggsae reproduces by self-fertile hermaphrodites, and both species have similarly low levels of molecular diversity. But the global sampling of natural populations has been limited and geographically biased. In Chapter 5, I describe the first cultured isolates of C. elegans and C. briggsae from sub-Saharan Africa, characterize these samples for patterns of nucleotide polymorphism and vulva precursor cell lineage variation, and conduct a series of hybrid crosses in C. briggsae to test for genetic incompatibilities. With the new African isolates, I show distinct differences in levels of genetic and phenotypic diversity between the two species. Despite many similarities between C. elegans and C. briggsae, the results indicate that there may be more subtle, and previously unknown, differences in their natural histories. Finally, I return to the question of the impact of reduced recombination on TE dynamics in Chapter 6, by comparing population frequencies of TEs in natural populations of selfing and outcrossing Caenorhabditis species. I show that in the selfing species, C. elegans, transposons are less polymorphic and segregate at higher frequencies compared with the outcrossing species, C. remanei. Estimates of the intensity of selection based on the population frequencies of polymorphic elements suggest that transposons are selectively neutral in C. elegans, but subject to weak purifying selection in C. remanei. These results are consistent with a reduced efficacy of natural selection against transposable elements in selfing populations.

The genus Cochlearia is a taxonomically complex genus with a circumpolar distribution. In common with many other post-glacial colonisers it exhibits complex patterns of morphological and ecological variation. The genus has been the subject of continued taxonomic controversy, especially within the species complex C. officinalis s.l. The focus of this study was to investigate whether the three rare putative endemic Cochlearia officinalis s.l. taxa in Britain: C. micacea, C. officinalis subsp. scotica and C. atlantica were sufficiently distinctive to warrant endemic species or taxon status at any rank. Furthermore, to make conservation recommendations for the species complex based on the outcome of this investigation. The patterns of differentiation in Cochlearia were studied to gain insight into the processes that have driven morphological and ecological diversification in the group. The six putative taxa in Cochlearia officinalis s.l. were considered in this study: C. officinalis s.s., C. officinalis subsp. scotica, C. pyrenaica subsp. pyrenaica, C. pyrenaica subsp. alpina, C. atlantica and C. micacea. Samples of C. danica, a member of the wider genus Cochlearia, were also included for comparison. The samples were screened for variation in AFLP fragments, morphological characters and chloroplast haplotypes. This is the first study focussed on the British Cochlearia to use the amplified fragment length polymorphism (AFLP) technique. Many qualitative morphological characters differences between populations were maintained in cultivation under standard conditions. Variation in some quantitative morphological characters was significantly different between taxon groups. The morphological characters combined did not distinguish between taxonomic groups. Variation was found in samples from the uplands only. Although there were three chloroplast haplotypes all but 6 out of 96 samples had the same haplotype and the chloroplast was not taxonomically informative. The AFLP data did not vary significantly between taxonomic groups, ploidy levels, habitats or geographical regions. There was significant AFLP variation between populations. The morphological and ecological diversity present among populations of Cochlearia officinalis s.l. in Britain is most likely to result from local ecotypic differentiation. The variation in Cochlearia officinalis s.l. could not be divided satisfactorily into taxa of species rank and so specific conservation of taxa within the complex is not recommended. Instead the maintenance of Cochlearia diversity can be achieved by the continued protection of the habitats in which the ecotypes grow.