Genetic drift and effective population sizes of hatchery-propagated stocks of the Pacific oyster, Crassostrea gigas

Abstract

Using starch-gel electrophoresis, we scored genetic differences for 14 polymorphic enzymes among individuals from three Pacific oyster populations: natural set from Dabob Bay, Washington, and two cultivated stocks. The stocks were each derived from the Dabob Bay wild population and reproductively isolated from this population and from each other for three generations via hatchery propagation and separate rearing on commercial growout beds in Willapa Bay, Washington, and Humboldt Bay, California. Compared with the wild population, one of the two cultivated stocks had significantly fewer alleles per locus, but proportions of polymorphic loci and average heterozygosities were not statistically different among the three population samples. Hatchery-propagated stocks differed markedly in allelic frequencies from each other and from the wild population at most loci. Allelic frequencies in the Dabob Bay sample are assumed to represent those in the progenitors of the hatchery stocks. Average per locus allele-frequency variances between the progenitor and derived hatchery populations are normally distributed after appropriate transformation, indicating that divergence of hatchery stocks from Dabob Bay population owes to random genetic drift. Based on the inverse relationship between the magnitude of random genetic drift and effective population size ( N e), the per-generation effective sizes of the two commercial oyster stocks are calculated to be only 40.6±13.9 (s.d.) and 8.9±2.2 (s.d.) for the Willapa Bay and Humboldt Bay stocks, respectively. These estimates account well for the loss of alleles in these hatchery-propagated stocks. Analysis of allele-frequency drift is recommended over simple comparisons of genetic diversity for revealing the extent and nature of genetic change in reproductively isolated aquaculture stocks.