We examined the effect of different sediment types on the staining effectiveness of the fluorochrome DAPI (4'-6-diamidino-2-phenylindole dihydrochloride) over a wide range of concentrations and on the masking effect of sediment particles on DAPI-stained bacteria. Sediment type greatly affects the staining efficiency of DAPI, and most published studies seem to have underestimated bacterial abundances by using suboptimal concentrations of the fluorochrome. A DAPI concentration of 5 mug ml is required to effectively stain the bacteria in most sediments that can be sampled with a gravity corer. When the sediments are diluted 687 times (a dilution factor similar to those most often used in the literature), sediment particle masking of stained bacteria is highly variable for different sediment types. By using a measure of turbidity (A(750)) to indicate masking and the quartz-corrected water content as a measure of the initial (in situ) dilution of each sediment type, it becomes possible to show a linear relationship between masking and the integrated (initial x experimental) dilution of various sediments. This relationship allows the development of a correction procedure for masking which makes accurate and unbiased counts possible. Data so obtained show a strong relationship between bacteria (cells per milliliter of fresh sediment) and sediment organic matter (grams [dry weight] per milliliter of fresh sediment), one that is not discernable without the correction. The proposed method of staining and correction for sediment masking provides the basis for a standardized interpretation of sediment bacterial counts.