The association between recreational physical activity and the risk of benign proliferative epithelial disorders (BPED) of the breast was examined in a case-control study conducted in Adelaide, Australia, between 1983 and 1985. The study included 383 incident, histologically confirmed cases of BPED identified through the major private pathology laboratory, 383 unbiopsied community controls matched to the cases for age and area of residence, and 192 controls whose biopsies did not show any signs of epithelial proliferation. Subjects reported their level of usual weekly physical activity in a self-administered questionnaire. When cases were compared with community controls, there was some evidence for a negative association between risk of BPED and total recreational physical activity. However, the reverse was observed when the cases were compared with biopsy controls. These patterns were essentially the same when examined by intensity of activity and by menopausal status. Despite the lack of an association between physical activity and risk BPED in this study, further investigation is warranted because of the potential benefits for the primary prevention of breast disease.