The traditional approach to estimating the potential impacts of anthropogenic chemicals on organisms has been to conduct single-species for short durations with the underlying premise that relatively low concentrations (compared to LC50 values) will not affect survival. In short, the dose makes the poison. In this talk, we challenge this premise through a series of new experiments on aquatic food webs that demonstrate how low concentrations of several pesticides can have substantial impacts on amphibian survival that are not predictable from traditional single-species tests. Indeed these impacts occur via density- and trait-mediated indirect effects, demonstrating that the full suite of pesticide effects on amphibians (or any organism) in nature can only be known when they are embedded into their more natural community context. These results have widespread application to understanding the potential impact on pesticides on natural amphibian populations.