Physical activity has been proposed as a modifiable risk factor for prostate cancer because of its potential effects on circulating hormones such as testosterone and insulin. We examined the association of various measures of physical activity with prostate cancer risk among men in the American Cancer Society Cancer Prevention Study II Nutrition Cohort, a large prospective study of U.S. adults. Information on recreational physical activity was obtained from a self-administered questionnaire completed at cohort enrollment in 1992/1993, as well as from a questionnaire completed as part of an earlier study in 1982. During the 9-year prospective follow-up, 5,503 incident prostate cancer cases were identified among 72,174 men who were cancer-free at enrollment. Cox proportional hazards modeling was used to compute hazard rate ratios (RR) for measures of recreational physical activity and to adjust for potential confounding factors. We observed no difference in risk of prostate cancer between men who engaged in the highest level of recreational physical activity (>35 metabolic equivalent-hours/wk) and those who reported no recreational physical activity at baseline (RR, 0.90; 95% confidence interval, 0.78-1.04; P for trend = 0.31). We also did not observe an association between prostate cancer and recalled physical activity at age 40 or exercise reported in 1982. However, the incidence of aggressive prostate cancer was inversely associated with >35 metabolic equivalent-hours/wk of recreational physical activity compared with that in men who reported no recreational physical activity (RR, 0.69; 95% confidence interval, 0.52-0.92; P for trend = 0.06). Our findings are consistent with most previous studies that found no association between recreational physical activity and overall prostate cancer risk but suggest physical activity may be associated with reduced risk of aggressive prostate cancer.