Experimental studies in animals and epidemiological studies in human populations support an inverse association between exercise and the development of cancer. Physical activity has been shown to be protective against the development of breast and colon cancer and may also be important for other kinds of cancer such as that of the prostate. The proposed biological mechanisms for the physical activity--cancer association include exercise's effect on immune function, transit time of digestion, hormones, and body fat. There has been little research on physical activity and the effect on progression of cancer, although there are studies to suggest that it may slow the clinical course of the disease. Furthermore, exercise may be beneficial in the treatment of cancer through mood elevation, decreased loss of lean tissue, and increased quality of life. Much is still to be learned about the effect of exercise on cancer. The intensity, duration, frequency, and type of exercise that is relevant need to be clarified. As well, the time period during life when exercise is important has not been determined. It seems reasonable to conclude that exercise, a modifiable risk factor, is beneficial in preventing certain forms of cancer. Public health interventions may hold promise for cancer prevention.