Local adaptation in the land snail, Cepaea nemoralis : exploratory genetic analysis with AFLP markers

The polymorphic land snail, Cepaea nemoralis (L.) shows a higher prevalence of yellow, banded shells in open habitats and non-yellow (Pink and brown), unbanded shells in woodland. Shell characters are genetically controlled and this pattern of local adaptation is geographically repeated, implying selection is operating. However, no molecular evidence exists to support this claim. An FSI-outlier approach revealed six AFLP loci with elevated levels of differentiation compared to neutral expectations - three in one open-woodland population pair and another three in another pair - providing evidence of positive selection operating between open and woodland habitats. The non-repeated nature of outliers suggests tlie two sites are adapting to habitats by different strategies. Pairwise F SI and a neighbour-joining tree show the two study sites to be genetically divergent from each other, probably because they are geographically isolated (~60km apart) and there are low levels of differentiation between habitats within each pair. An attempt was made at creating a preliminary AFLP linkage map for C. nemoralis using a pseudo-testcross design. A mapping population segregating for shell colour and mid-banding was established but only three, small AFLP linkage groups were identified in one of the parental maps and none in the other. This was largely due to a small proportion of loci being in testcross configuration. In addition, morph frequency data were compared to historic records to look for changes in morph/allele frequency over time. All open populations showed significant differences in colour or banding, whereas no significant differences were found in any woodland populations. This could be an adaptive response to differences in habitat stability between open and woodland, the latter being more stable. Alternatively, balancing selection could be operating in woodlands and open populations fluctuate randomly. The difference in morph frequency stability between habitats was not reflected in allele frequency changes (assuming Hardy-Weinberg equilibrium). 1.EThOS - Electronic Theses Online ServiceGBUnited Kingdo