9700515 Morris Most organisms live in unpredictable environments, and must make decisions such as where and when to reproduce using incomplete information. Despite numerous theoretical studies that have predicted how animals should behave in the face of such uncertainty, few field studies have quantified the ways that animals respond reproductively to mortality risks that are spatially and temporally unpredictable. Even fewer studies have quantified how well animals use information to reduce reproductive risks, when compared to either naive or omniscient strategies. In a seasonal lowland rain forest in Peru, the frog Edalorhina perezi and other amphibians preferentially lay eggs in small tree-fall pools, where the eggs and tadpoles experience mortality risks that differ among pools and change temporally depending on unpredictable rainfall. By observing frog egg-laying behavior and movement among pools, I will assess how well they distribute their offspring. My approach will combine observations of risks and frog behavior in natural pools, an experiment using artificial pools to test the frogs' abilities to track changing risks, and a model to determine how well the frogs perform compared to other strategies that range in risk-assessment capabilities.
