A sediment study, involving both chemical and biological analyses, was carried out in the St. Lawrence River near Montreal (Lake St. Louis). About 60 stations were sampled during 1984-85, and the sediments were analyzed for support variables, heavy metals, and organochlorinated compounds. Subsamples were elutriated using a 1 to 4 sediment/water ratio. The resulting elutriates were analyzed for several chemical variables, while toxicity was measured using the Microtox test, algal 4 C assimilation, and lethality/developmental inhibition in cladocerans, rotifers, and nematodes. The results showed a great variety of responses and sensitivity, and correlations between the tests were non-significant. In terms of toxic responses, the algal and Microtox tests were the most sensitive. Toxic responses could not be explained in simple terms of contaminant concentrations. Therefore, the chemistry of elutriates is not predictive of the toxic potential of contaminated sediments. Biotests can give an insight into the hazard assessment of sediments, but no single test will be sufficient; the use of a battery of standardized biotests, representing different levels of organization/food chain, including representative natural species, is highly recommended.