During pilot year studies for the Estuarine Monitoring and Assessment Program (EMAP) in the Carolinian Province, it was observed that estuarine sediments characterized by high silt-clay content (contaminated as well as uncontaminated) consistently had low Microtox EC50s (as measured by decreased light production). It was hypothesized that the low EC50 values associated with silty sediments were due to adsorption of the bacteria to the silt-clay particles. During the assay, the bioluminescent bacteria are exposed to a sediment suspension, and the effects on light production are evaluated in the liquid phase that remains after removal of the sediments by filtration. If the bacteria adsorb to the sediments, then the decreased light production observed with silty sediments may be the result of significant bacterial removal, rather than toxicity. This sediment bias could affect the interpretation of results from field studies by overestimating the potential for contaminant effects in muddy areas. We report on studies in which artificial sediments (mixtures of pure sand and clay) were used to evaluate the effects of sediment type (in the absence of contaminants) on Microtox EC50 values and bacterial counts of the liquid phase. The results supported the hypothesis. Microtox EC50s and bacterial concentrations in the liquid phase decreased in a predictable manner as clay composition increased. Methods of normalizing for sediment composition and developing toxicity criteria are discussed.