The disease-producing potential of recreational waters is currently estimated through the use of certain bacterial indicators that are believed to be positively correlated with the presence of fecal contamination. In general, these indicators and their recommended limiting values have been adopted for use from existing standard methods for the analysis of sanitary waste water. However, no indicator currently in use today exists solely in the feces of man and not also elsewhere, e.g., in soils, vegetation, and the feces of animals. Storm-water runoff is often rich in bacteria originating from these nonhuman sources and can contain high densities of indicator bacteria; however, existing epidemiological studies of recreational waters receiving storm-water runoff have reported little correlation between current indicator densities and the incidences of swimming illnesses. In addition, microbial analyses of storm-water runoff have revealed a predominance of nonenteric disease-causing bacteria and viruses that have been linked to respiratory illnesses and skin infections. Fecal-based indicators in use today provide no information on the risks resulting from body contact with these nonenteric pathogens. Consequently, for receiving waters containing discharges that originate primarily from separate storm drainage systems, current bacterial indicators are ill suited to accurately assess the water's total illness-producing capabilities. This paper briefly reviews the development of current bacterial standards and evaluates their adoption in the field of storm-water testing. The unique disinfection requirements of storm-generated runoff are discussed, and advanced disinfection practices are reviewed. The need for additional epidemiological studies that address the disease-causing potential of nonhuman and nonenteric pathogens commonly found in storm-water runoff from urban, agricultural, and rural watershed areas is emphasized in order to determine the actual health risks associated with storm water runoff contact.