To understand the impacts of anthropogenic chemicals on natural communities, we often must base our predictions on short-term, single-species tests that are conducted as part of the regulation process. While a valuable first-step, these tests tell us little about the impacts of contaminants under more natural conditions. Thus, many ecotoxicologists have moved to testing contaminants under more natural conditions where natural stressors are abundant. Among the many natural stressors, predator stress is a common one whose interactions with pesticides have only been examined under laboratory conditions. Using two mesoscosm experiments, I examined how predation stress interacted with the effects of Roundup®, the most widely applied herbicide in the world. The first experiment, conducted with three spring-breeding species of anurans, crossed four concentrations of Roundup with the presence of no predators, caged adult newts, or caged dragonfly larvae. The second experiment, conducted with three species of summer-breeding anurans, crossed four concentrations of Roundup with the presence of no predators, caged dragonfly larvae, or lethal dragonfly larvae.

Results/Conclusions

            In the first experiment, Roundup and the caged-predator treatments had interactive effects on tadpole survival, mass, and relative morphology. Increased herbicide concentrations caused increased tadpole mortality, but the amount of mortality decreased in the presence of caged dragonflies. Tadpole mass exhibited little effect of the herbicide when no predators or caged newts were present, but mass declined with higher herbicide concentrations when caged dragonflies were present. Not surprisingly, the cues from caged dragonflies induced adaptive morphological changes in the tadpoles. What was surprising is that the herbicide induced the same morphological changes in the tadpoles as the larval dragonflies. In the second experiment, Roundup and the predator treatments also had interactive effects. Increased herbicide concentrations again caused increased mortality and the amount of mortality again declined with caged dragonflies. With lethal dragonflies, however, there mortality was reduced similarly across all herbicide concentrations. Tadpole mass increased with higher herbicide concentrations when predators were absent, was unaffected by the herbicide when caged dragonflies were present, and decreased with higher herbicide concentrations when lethal dragonflies were present. Once again, the morphology of the tadpoles was induced similarly by caged dragonflies and the herbicide. Collectively, these results suggest that the effects of Roundup on larval amphibians can differ tremendously depending on community context. Moreover, the herbicide is somehow able to induce anti-predator responses.