Effect of environmental stress on clonal structure of Eucypris virens (Crustacea, Ostracoda)

Environmental stress imposes strong natural selection on clonal populations, promoting evolutionary change in clonal structure. Environmental stress may also lead to reduction in population size, which together with clonal selection may reduce genotypic diversity of the local populations. We examined how clonal structure in wild-collected samples of two parthenogenetic populations of the freshwater ostracod Eucypris virens responded to hypersalinity and starvation, and the combination of the two stressors. We applied the stress treatments in a factorial design for one generation. When 60% of the individuals per experimental unit had died, post-experimental clonal structure was compared to that of the start of the experiment, which reflected the field conditions. We used five polymorphic allozyme loci as genotype markers. All stress treatments reduced survival compared to the control treatment. In the population "Rivalazzetto”, we observed a reduction of clonal richness in the control treatment, with the initially dominant clone maintaining dominance. This may have resulted from interclonal competition and clone-specific survival under the different laboratory conditions. Clonal richness remained high in the salinity treatment while it was reduced in the combined stress and starvation treatments. In the population "Fornovo”, clonal richness reduced in all treatments including control, while the salinity and combined stress treatment reduced clonal evenness. The clone dominating at the start of the experiment increased in frequency in all treatments, but the change in clonal structure during the experiment was more pronounced in this population. These results suggest that in some conditions an intermediate level of environmental stress may lessen the decline in genetic diversity by strong inter-clonal competition. Moreover, the variation in clonal structure among the stress treatments and distinct genetic backgrounds indicates that more general predictions of stress e