Maintaining optimal conditions in catchments or distribution systems relies heavily on water authorities having access to rapid and accurate water quality data, including an indication of bacteriological quality. In this study, the BacLight bacterial viability kit and carboxyfluorescein diacetate (CFDA) were coupled with flow cytometry (FCM) for rapid detection of physiologically active bacteria from raw and potable waters taken from various locations around South Australia. Results were compared to the direct viable count (DVC) and quantitative DVC (qDVC), in addition to the culture-based methods of the heterotrophic plate count (HPC) and a commercial SimPlate technique. Raw and potable water analysis revealed that DVC and culture-based techniques reported significantly fewer viable bacteria compared to the number of physiologically active bacteria detected using the rapid FCM assays, where this difference appeared to be nonlinear across different samples. Inconclusive results were obtained using qDVC as a viability assay. In particular, HPC results were 2-4 log orders of magnitude below that reported by the FCM assays for raw waters. Few bacteria in potable waters examined were culturable by HPC, even though FCM assays reported between 5.56 x 10(2) and 3.94 x 10(4) active bacteria ml(-1). These differences may be attributed to the presence of nonheterotrophic bacteria, sublethal injury or the adoption of an active but nonculturable (ABNC) state.