Effects of a Predator on Prey Metamorphosis: Plastic Responses by Prey or Selective Mortality?

Models for organisms with complex life cycles predict decreased size at, and altered time to, metamorphosis for organisms experiencing greater mortality rate during the larval stages. We tested this prediction by exposing larvae of the tree hole mosquito, Aedes triseriatus, to predation by the mosquito Toxorhynchites rutilus in a series of five experiments. Populations both sympatric and allopatric to the predator were tested, and meta-analysis was used to synthesize the results of all experiments, and to test for heterogeneity of effects of actual vs. perceived predation (predator unable to kill prey). In the presence of this predator, A. triseriatus from both sympatric and allopatric populations had lower mass at pupation, as predicted. This effect was present in both sexes, but was more pronounced in females. The presence of the predator also increased time to pupation for females (but not for males). Reduced size at pupation appears to result from decreased growth rate in the presence of the predator, and we tested the hypothesis that this predator induced behavioral changes in A. triseriatus that lead to reduced growth (e.g., reduced movement and foraging). Behavioral observations did not support the hypothesis of behavioral change. Meta-analysis revealed significant heterogeneity of the effects of actual predation vs. perceived predation on size at pupation, which decreased significantly only in response to actual predation. There was no significant heterogeneity in the effect of actual predation vs. perceived predation on time to pupation. These results suggest that effects of this predator on metamorphosis of A. triseriatus may be products of either facultative responses to the predator or of selective mortality due to predation in experiments with actual predation. Sympatric and allopatric populations of A. triseriatus used in these experiments did not differ consistently in size at and time to pupation, nor in their developmental responses to T. rutilus. Although these results generally support the predictions of existing models of complex life cycles, they suggest that the mechanisms producing these effects in this system may not be the same as the behavioral mechanisms producing similar effects in amphibian systems.