Observational epidemiology supports the hypothesis that variation in diet and other lifestyle exposures accounts for a large part of the variation in incidence of colorectal cancer (CRC). Physical inactivity is associated strongly with enhanced CRC risk, but no human intervention studies have shown causality. This paper reviews data from all available studies of the effects of exercise interventions on intestinal neoplasia using rat and mouse models. All 5 published studies of effects of increased physical activity (both forced and voluntary) using carcinogen-treated rat models show strong protection against CRC by greater physical activity. In contrast, there is little convincing evidence of reduced intestinal neoplasia after increased physical activity in the 3 published studies using Apc(Min) mice (which develop multiple intestinal polyps spontaneously) although the nature and amounts of physical activity imposed in rats and mice were similar. Major differences in protocol between the 2 groups of studies are that the rat studies were much longer (at least 20 wk and in most cases 38 wk compared with < or =9 wk for the mouse studies) and the primary endpoint was colorectal carcinoma (rats) rather than small bowel adenomas (mice). The epidemiological evidence for protection against adenoma formation by increased physical activity is weaker than that for carcinoma. The limited evidence available suggests that, compared with rats, mice may show a greater compensation for energy expenditure in exercise through reduction in nonexercise physical activity, thus ameliorating effects. The resulting smaller effects on body weight and body fatness may limit changes in intestinal neoplasia in Apc(Min) mice.