Plating methods for estimating survival of indicator organisms, such asEscherichia coli, and water-borne pathogens includingVibrio cholerae, have severe limitations when used to estimate viable populations of these organisms in the aquatic environment. By combining the methods of immunofluorescent microscopy, acridine orange direct counting, and direct viable counting, with culture methods such as indirect enumeration by most probable number (MPN) estimation and direct plating, it was shown that bothE. coli andV. cholerae undergo a "nonrecoverable" stage of existence, but remain viable. Following 2-week incubations in saltwater (5-25%o NaCl) microcosms, total counts, measured by direct microscopic examination of fluorescent antibody and acridine orange stained cells, remained unchanged, whereas MPN estimates and plate counts exhibited rapid decline. Results of direct viable counting, a procedure permitting estimate of substrate-responsive viable cells by microscopic examination, revealed that a significant proportion of the nonculturable cells were, indeed, viable. Thus, survival of pathogens in the aquatic environment must be re-assessed. The "die-off" or "decay" concept may not be completely valid. Furthermore, the usefulness of the coliform and fecal coliform indices for evaluating water quality for public health purposes may be seriously compromised, in the light of the finding reported here.