The association between low physical activity and colon cancer was examined in a Swedish 14-year follow-up study of 16,477 subjects. The relative risk (RR) of colon cancer in subjects with low physical activity was estimated at 3.6 (1.3-9.8, 95% confidence interval). An association was observed for both men and women, and for low physical activity during occupational hours (RR = 1.6, 0.8-2.9) as well as during recreational hours (RR = 1.6, 1.0-2.7). The relative risk for rectal cancer was not elevated. In the study we controlled for age, gender, domicile and, to some extent, for diet. A possible mechanism is that low physical activity could prolong the transit time of the stool in the colon and thereby the duration of contact between the mucosa and fecal carcinogens.