Vibrio cholerae and Escherichia coli were inoculated into membrane diffusion chambers and placed around two small coral reef islands in Puerto Rico and monitored for 5 days. Several chambers were also buried in the sands of one of the reefs. Both E. coli and V. cholerae densities declined by 2 orders of magnitude, as measured by direct particle counts with a Coulter Counter (Coulter Electronics, Inc., Hialeah, Fla.). However, the density of neither bacteria changed dramatically when the same samples were analyzed by epifluorescent direct counts. Differences in the two direct count methods were accounted for by changes in cell morphology that occurred in both bacteria after exposure to seawater. Morphological changes occurred more rapidly in E. coli compared with those in V. cholerae. Bacteria in chambers exposed to sediment did not show significant changes in morphology and had only a slight decline in density. Physiological activity declined by more than 40% for both bacteria within 24 h. The decline in activity was less severe in the sediments. Tropical coral reef sands and turtle grass beds were shown to be less stressful environments for V. cholerae and E. coli than would have been predicted from temperature and microcosm studies. V. cholerae can survive the in situ conditions of a tropical coral reef and could become a source of bacterial contamination for fish and shellfish in this environment. The simultaneous monitoring of E. coli levels established that this bacteria can not be used as an indicator of V. cholerae or other fecal-borne pathogens in coral reef environments because of the greater stress these environments put on E. coli. Both bacteria could be of greater public health importance in tropical marine areas than previously imagined.