The science of sediment toxicology essentially began in the late 1970s. It was largely a product of dredging concerns and recognition of widespread contamination of sediments. During the past few years, sediment toxicity research activity has increased dramatically. Currently, most tests are of an acute nature with fewer available for determining sublethal endpoints of chronic toxicity. Test systems of single and multiple species have included most levels of biological organization in aquatic ecosystems and have been conducted in the laboratory on whole sediments, interstitial waters, elutriates, or other extractable fractions under a wide variety of conditions. Evaluations of methodological effects and comparisons with in situ toxicity using surrogate test species and indigenous communities have, on occasion, shown significant differences in test responses. These differences may be attributed to laboratory‐controlled parameters (e.g., light, species, life stage, exposure conditions, test phase, spiking method); sampling and laboratory‐induced disruption of sediment integrity; alteration of toxicant partitioning due to manipulations and temporal effects; and failure to recognize other influencing ecosystem variables (e.g., organism niche and life cycle, sediment partitioning and gradient dynamics, physicochemical and biological process integration, biotic and abiotic disturbances, micro‐ and macrobiota patches, food‐web interactions). Optimizing and standardizing test methods will require further studies of these variables to improve inter‐laboratory comparisons and ecosystem validity. Despite the many unknowns that exist, a variety of sediment toxicity tests have been effectively used in assessing toxicant contamination by measuring the bioavailable fraction of the in‐place pollutants. The optimal assays vary with the study and its objectives. Intergrative studies using several chemical, community, and toxicity measures are currently the most effective at defining ecosystem perturbations.