Nitrification is a major problem encountered in many drinking water reservoirs and distribution systems across the country. Nitrification episodes are responsible for a sudden drop in disinfectant residuals, high nitrite levels, high heterotrophic plate counts (HPCs), and occasionally coliform occurrences, including the presence of Escherichia coli. The corrective measures are expensive and laborious, and usually only mitigate the problem for a short period of time, with a return to nitrification within a few weeks or months. To address this problem, a study was undertaken with the goal of identifying the microbial populations and the physicochemical conditions that mediate or interact with nitrification and coliform occurrences in drinking water reservoirs. The goal of this study was to identify the specific parameter(s) that indicate or facilitate theestablishment and subsequent growth of nitrifying and coliform bacteria, and trigger nitrification and coliform episodes. To address this, liquid and biofilm samples were collected from three finished-water reservoirs on a monthly basis. In addition, supplemental experiments were conducted to evaluate: the effect of the most commonly used corrective procedure (superchlorination); the sequence of colonization of nitrifying and coliform bacteria; and, the presence of nitrifying and coliform bacteria at different water depths. To study the biofilm, coupons made of the same material as the reservoirs (concrete or coated steel) were placed at two different depths to evaluate the effects of water level fluctuation. Samples were also monitored for nitrifying bacteria, coliforms and E. coli, as well as for a variety of physical and chemical parameters. Ammonia oxidizing bacteria (AOB) were detected in all reservoirs, at all depths, in both liquid and biofilm samples. They were detected at a very early stage of biofilm formation (3 days following coupon placement). In general, higher HPC and AOB levels were measured on the biofilm coupons placed on the bottom of the reservoirs (area constantly submerged) than on top of the reservoirs (intermittently exposed to air). In the liquid samples, the upstream reservoir exhibited higher HPC and AOB levels than the two downstream reservoirs, which is surprising considering the lower water residence time and higher chlorine residual measured in the upstream reservoir. Results indicate that reservoir cycling does not completely prevent nitrification from occurring. Superchlorination effectively decreased biomass concentrations in both liquid and biofilm samples, although experience has shown that nitrification may re-occur within a period of a few weeks or months. Includes 7 references, figures.