The objective of this study was to determine and evaluate some of the factors that affect the formation and decay of disinfection by-products (DBFs), specifically trihalomethanes (THMs) and haloacetic acids (HAAs), in full-scale drinking water distribution systems. Five utilities throughout the United States were sampled 6-8 times over a period of 15 months. This paper addresses four of the five utilities chosen to participate in the study. Utilities were chosen based on source water characteristics, seasonal temperature changes, geographic location, type of secondary disinfectant utilized and several characteristics specific to each distribution system. Minimum levels of DBFs, THMs > 40 μg/L and HAAs > 20 μg/L, were also required for participation. In addition, each utility had a hydraulic model of their distribution system that was used to determine residence times at sampling locations in their distribution system. Although treatment processes were not considered in the selection of utilities, all the utilities chosen to participate in the study employed conventional treatment processes: coagulation, flocculation, sedimentation and filtration. Two utilities utilized free chlorine as their secondary disinfectant and two utilities utilized chloramines as their secondary disinfectant. Sampling locations were chosen on the basis of residence time, pipe material, pipe diameter and other factors specific to each distribution system. Fourteen to sixteen sampling locations were chosen to represent the various conditions that were present in each distribution system. Samples collected were analyzed for pH, temperature, secondary disinfectant residual, assimilable organic carbon (AOC), total organic carbon (TOC), ultraviolet (UV) absorbance, heterotrophic plate counts (HPCs), THMs, HAAs, bromide and ammonia, nitrite and nitrate, where appropriate. For those systems that utilized free chlorine, results show that chlorine residual decreased as residence time increased. In addition, THM and HAA concentrations increased as residence time increased. Evidence of biodegradation of HAAs at high residence times was observed at one utility. At sampling locations with high residence times, loss of chlorine residual and decreases in HAA concentration were coupled with increases in HFCs. Data obtained for those utilities that utilized chloramines demonstrate that disinfectant residual remained constant throughout the distribution system, except at sampling locations where evidence of nitrification and HAA biodegradation were observed. Because of the periodic addition of a second raw water source, the hydraulic model for one of the systems was rerun and new hydraulic scenarios were determined. Consequently, the residence times at each sampling location for this system were altered depending on flow and demand in the distribution system. Thus, for some scenarios, the total chlorine residual remained constant, while for others, a significant decrease in total chlorine residual was observed. THM4 and HAA9 concentrations remained constant at both utilities when the disinfectant residual remained constant. However, at both utilities, a decrease in HAA9 concentrations was seen, particularly at those sampling locations that showed a decrease in total chlorine residual and elevated HPCs. These results are consistent with typical conditions that promote the biodegradation of HAAs. In one of the systems a decrease in THM4 concentration was seen at the same locations where HAA9 biodegradation occurred. Evidence of nitrification was seen at both utilities, specifically at locations with low total chlorine residual.