Analysis of 56 river water samples by the Enterolert defined substrate technique, and standard m-Enterococcus agar isolation followed by confirmation, indicated that after 24 h incubation. Enterolert significantly underestimated the true numbers of enterococci. Extending Enterolert incubatioin to 36 h improved detection but also revealed false positives. These findings prompted the development of a novel confirmation medium we have termed glucosidase agar, which was prepared by dissolving Enterolert substrate in 2% (w/v) bacteriological agar. Analysis of 1,043 colonies arising on m-Enterococcus agar from 280 freshwater, marine and sewage effluent samples, demonstrated that 2-4 h incubation on glucosidase agar was a rapid and accurate means of confirming presumptive enterococci, when compared to standard confirmation procedures that take 48 h. The combination of primary isolation on m-Enterococcus agar followed by confirmation on glucosidase agar permits maximum recovery of Enteroccus whilst effectively eliminating false positives/negatives and provides a reliable alternative use of the Enterolert defined substrate technology.