Introduction

The application of molecular techniques to population biology can be traced back to the origins of allozyme gel electrophoresis (Smithies 1955) and the invention of histochemical stains targeting specific enzymes (Hunter and Markert 1957). Initial surprise at the high level of variation revealed was rationalised by the understanding that much molecular variation must be selectively neutral (Kimura 1968; King and Jukes 1969), and making this assumption, molecular markers were applied to samples of anonymous individuals to quantify population-level processes such as inbreeding, differentiation and phylogeography (Avise 1994).

Alongside these population genetics studies, there was a series of detailed studies in which phenotypic and spatial data about the individuals sampled was also collected, and the enzyme kinetic properties of alleles were quantified, allowing tests of selective neutrality at specific allozyme loci. Examples of allozyme polymorphisms that appear to be under selection include phosphoglucomutase in Colias butterflies (Watt et al. 1983), leucine aminopeptidase in the mussel Mytilus edulis (Koehn and Hillbish 1987) and lactate dehydrogenase-B in the marine killifish Fundulus heteroclitus (Powers et al. 1983). Similarly, several studies demonstrated correlations between average allozyme heterozygosity and phenotypic measures, which might be due to overdominance (heterozygote advantage) at specific loci or inbreeding depression (Allendorf and Leary 1986; Ledig 1986; Mitton 1997).