This population-based case-control study was conducted in southern Ontario, Canada from 1992 to 1994 to assess the relationship between chlorination by-products in public water supplies and cancers of the colon and rectum. Interviews providing residence and water source histories were completed by 76% of eligible cancer cases and 72% of eligible controls. Supplemental data from municipal water supplies were used to estimate individual exposure to water source, chlorination status, and by-product levels as represented by trihalomethanes (THMs) during the 40-year period before the interview. The analyses included 767 colon cases, 661 rectal cases, and 1545 controls with exposure information for at least 30 of these years (75% of subjects with completed interviews). Among males, colon cancer risk was associated with cumulative exposure to THMs, duration of exposure to chlorinated surface water, and duration of exposure to a THM level > or = 50 microg/liter and 75 microg/liter. Males exposed to chlorinated surface water for 35-40 years had an increased risk of colon cancer compared with those exposed for < 10 years (odds ratio, 1.53; 95% confidence interval, 1.13-2.09). Males exposed to an estimated THM level of 75 microg/liter for > or = 35 years had double the risk of those exposed for < 10 years (odds ratio, 2.10; 95% confidence interval, 1.21-3.66). In contrast, these relationships were not observed among females. No relationship was observed between rectal cancer risk and any of the measures of exposure to chlorination by-products. The results of this study should be interpreted with caution because they are only partially congruent with the limited amount of literature addressing this issue.