Résumé
Marine species with high fecundity and high early mortality may also have high variance in reproductive success among individuals due to stochastic factors, making successful reproduction a "sweepstakes." In some cases, the impact is sufficient to reduce the effective number of breeders in wild populations. We tested two predictions of the sweepstakes reproductive success hypothesis in a French Atlantic population of the European flat oyster, Ostrea edulis, by evaluating (1) whether individuals belonging to temporally discrete recruitment cohorts within a single reproductive season displayed reduced genetic variation relative to the entire adult population, and (2) whether these temporal cohorts of recruits were genetically differentiated from each other. We assayed genetic variation at four nuclear microsatellites and a 12S mitochondrial fragment in four recruitment cohorts. Nuclear markers provided no evidence for differentiation between recruitment cohorts and adults or between temporal cohorts. However, mitochondrial data indicate that the first temporal cohort showed significant differentiation with the last (Fst = 0.052, P < 0.05) and with the adult sample (Fst = 0.058, P < 0.05). These differences are most likely due to the smaller effective size of the mitochondrial genome-and hence its increased sensitivity to drift compared to the nuclear genome. This slight mitochondrial signal indicates a certain limitation in the number of contributing female parents in this species. The "sweepstakes" phenomenon was therefore limited in our case. Hypothetically, this phenomenon may occur or not, with a high variance as a result of the interaction between the oyster reproductive biology and different environmental conditions.