More than one hundred short-term bioassays are now available for detecting the toxicity, mutagenicity, and potential carcinogenicity of chemicals. These bioassays were developed and validated with individual compounds, and their principal application was perceived to be in evaluating the health hazard of such materials. However, man is rarely exposed to single chemicals; his exposure to hazardous chemicals is more commonly a multifactorial phenomenon. Although chemical analysis can be used to detect known hazardous compounds, it would be a staggering and expensive task to analyze large numbers of samples for all known or suspected hazardous constituents. Furthermore, the biological activity of a complex mixture cannot be reliably predicted from knowledge of its components. On the other hand, bioassays alone cannot tell us which components of complex mixtures are responsible for the biological activity detected. Thus, cost effectiveness and technical feasibility dictate stepwise and perhaps iterative application -of both chemical and biological methods in evaluating the health effects of complex environmental mixtures. Through the coupling of reliable biological detection systems with methods of chemical fractionation and analysis, it is frequently possible to isolate the individual chemical species that show biological activity. Initially, complex mixtures may be separated and bioassayed in carefully defined chemical fractions. The results of such short-term screening bioassays then may be used td guide the course of further fractionation and to determine the need for more stringent and comprehensive biological testing.