Expression of parasite virulence at different host population densities under natural conditions.

It has recently been suggested that the expression of parasite virulence depends on host population density, such that infected hosts have a higher sensitivity to density, and thus reach their carrying capacity earlier than uninfected hosts. In this scenario, parasite-induced reduction in fitness (i.e., virulence) increases with host density. We tested this hypothesis experimentally, using outdoor mesocosm populations of Daphnia magna infected by the microsporidian Octosporea bayeri. Contrary to the prediction, virulence was independent of host density. In a competition experiment with initial prevalence of 50%, O. bayeri reduced the competitive ability of infected Daphnia within the asexual growth phase independent of initial host population density. In an additional experiment we set up populations with 100% and 0% prevalence and followed their population dynamics over the whole season. Consistent with the competition experiment, we found no difference in population dynamics within the asexual growth phase of the host, suggesting that infected hosts are not more sensitive to density than uninfected hosts. The additional experiment, however, included more than the initial growth phase as did the competition experiment. Eventually, after 100 days, 100% infected populations assumed a reduced carrying capacity compared to uninfected populations. We identify and discuss three reasons for the discrepancy between our experiment and the predictions.