The relation between various measures of physical activity and colorectal cancer risk was considered in a case-control study conducted between 1992 and 1997 in the Swiss canton of Vaud. Cases were 223 patients (142 men, 81 women) below age 75, with colon (n = 119) or rectal (n = 104) cancer; controls were 491 patients (211 men, 280 women) admitted to hospital for acute, non-neoplastic conditions. Compared with the lowest level of physical activity at age 30-39 years, the odds ratios (OR) of colorectal cancer for the highest level were 0.44 (95% confidence interval, CI, 0.26-0.73) for occupational and 0.53 (95% CI 0.33-0.86) for leisure-time activity. An inverse association was also observed for physical activity at age 15-19 and 50-59 years. The inverse relation between physical activity and colorectal cancer was observed across strata of sex, age, education, body mass index and alcohol drinking; was somewhat stronger in subjects reporting high total energy, and low vegetable and fibre intakes; and was observed across various colon subsites and rectum. In terms of population attributable risk, increasing physical activity would avoid one-fifth to one-third of incident colorectal cancer cases.