Colorectal cancers are the second most frequent cause of cancer death among men. To our knowledge, approximately six studies have been able to show an inverse relationship between occupational physical activity and colon cancer mortality. Information drawn from the mortality statistics for the years 1979-1982 was used to study this hypothetical association among Swiss men aged 15-79. At-risk-populations were calculated based on 1980 national census data on occupation of all Swiss men. Estimates of occupational physical activity (OPA) were based on job titles of death certificates and were "blindly" classified into three groups of low, moderate and high OPA by three independent experts. Among the cohort of 1.86 million men, 1995 deaths of colon cancer and 1066 deaths of rectal cancer occurred during the four study years. The standardized mortality ratio showed a significant, graded and inverse relationship between OPA and mortality from colon cancer but not from rectal cancer. The estimated relative risk for colon cancer of the physically inactive, as compared to those active, was 1.3 to 1.4, slightly influenced by minor differences in the way of classification of OPA. For several reasons this estimate of excess risk is probably on the low side. The subgroup of men with jobs with very high OPA showed no further reduction in risk of colon cancer, which suggests that other etiologic factors, such as diet, may play an important role. As sedentary lifestyle and colon cancer are both frequent in central Europe the hypothesized protective effect of habitual physical activity against colon cancer would seem important, especially from the public health point of view.