Microcosm data provide new kinds of information which go beyond currently used regulations and procedures to evaluate environmental fate and effects of chemicals. Microcosm data provide a basis for considering ecological questions such as the extent of effects at several trophic levels, the breadth of effects on competitive species, the extent of apparent detoxification, and effects on the expected system recovery profile. Several models for applications of microcosm data are discussed: as a model ecosystem in a high-level tier test; as a means of integrating fate and effects processes to validate inferences and estimates; as a low-level tier or screening test when specific concerns require evaluation in an ecosystem context; as a means to anticipate patterns of ecosystem magnification and amelioration; and as a means to rank and compare environmental impacts based on community- or ecosystem-level characteristics. Besides applications in evaluating the impact potential of chemicals, microcosms may be useful in evaluating the potential environmental impact of genetically engineered microorganisms. As these new methods become available and credible, regulatory agencies such as EPA and FDA will need to consider how to incorporate the resulting data into their evaluation schemes.