The survival of culturable fecal coliforms, fecal streptococci, and Clostridium perfringens spores in freshwater and marine sediments from sites near sewage outfalls was studied. In laboratory studies, the inhibition of protozoan predators with cycloheximide allowed the fecal coliforms to grow in the sediment whereas the presence of predators resulted in a net die-off. C. perfringens spores did not appear either to be affected by predators or to die off throughout the duration of the experiments (28 days). Studies using in situ membrane diffusion chambers showed that, with the exception of C. perfringens, die-off of the test organisms to 10% of their initial numbers occurred in both marine and freshwater sediments within 85 days. The usual exponential decay model could not be applied to the sediment survival data, with the exception of the data for fecal streptococci. It was concluded that application of the usual decay model to the fecal coliform data was confounded by the complex relationship between growth and predation. The survival of seeded Escherichia coli in marine sediment was studied by using an enumeration method which detected viable but nonculturable bacteria. Throughout the duration of the experiment (68 days), the same proportion of E. coli organisms remained culturable, suggesting that sediment provides a favorable, nonstarvation environment for the bacteria.