A biologically based two-stage model for carcinogenesis is presented that relates events occurring at the cellular level to epidemiologic features of breast cancer in females. This model, which accommodates the physiologic responses of breast tissue to menarche, menopause, and pregnancy, predicts age-specific incidence curves that are in close quantitative agreement with those observed in six test populations: Connecticut, Denmark, Finland, Slovenia, Iceland, and Osaka, Japan. According to the model, hormones influence the epidemiology of breast cancer in females by their action on the kinetics of growth of nonneoplastic breast tissue. As a consequence, it is argued that hormones are likely to be unimportant in determining overall risk in populations. The protective effect of an early first birth predicted by the model is in good quantitative agreement with data from a multinational study. Other epidemiologic features of breast cancer are logically explained within the framework of the model. No feature of the epidemiology of breast cancer requires that premenopausal and postmenopausal breast cancer be considered distinct entities from the point of view of pathogenesis.