To investigate the relation between recreational physical activity and breast cancer, we conducted an analysis of a population-based case-control study of women age 50-64 years in western Washington State. The study included 537 patients with breast cancer diagnosed during 1988-1990 and 492 randomly selected population control women. Detailed information on recreational exercise habits and other risk factors for breast cancer was ascertained by structured in-person interviews. A similar proportion of cases and controls reported exercise in the 2 years before the referent date (about 50%), or during ages 12-21 years (about 40%). Compared with women who reported no exercise, there was a slightly decreased risk of breast cancer in women who exercised more than 1.5 hours per week in the 2 years before diagnosis (cases) or referent date (controls) [odds ratio (OR) = 0.7; 95% confidence interval (CI) = 0.4-1.1], or who engaged in at least some high-intensity exercise (OR = 0.7;95% CI = 0.4-1.1), but there was no clear decline in risk at the highest categories of duration or intensity. There was no association between any intensity exercise at ages 12-21 years and risk of breast cancer. These results indicate a weak negative association between physical activity and risk of breast cancer in middle-aged women.