Breast cancer originates in undifferentiated terminal structures of the mammary gland. The terminal duct of the Lob 1 of the human female breast is the site of origin of ductal carcinomas. Cell replication and the concentration of estrogen receptors type at in Lob 1 are at their peak during early adulthood, at a time during which the breast is more susceptible to carcinogenesis, decreasing considerably with aging. More importantly, when treated with carcinogens in vitro they express phenotypes indicative of cell transformation. These studies indicate that in humans there is a target cell of carcinogenesis, which is found in a specific compartment whose characteristics are a determinant factor in the initiation event. These target cells will become the stem cells of the neoplastic event, depending upon: a) topographic location within the mammary gland tree, b) age at exposure to a known or putative genotoxic agent, and c) reproductive history of the host. Epidemiological findings such as the higher incidence of breast cancer in nulliparous women and in women having early menarche support this concept, because it parallels the higher cancer incidence elicited by carcinogens when exposure occurs at a young age. In addition, it has been shown that increase in parity is associated with a pronounced decrease in the risk of breast cancer, each additional live birth conferring a 10% risk reduction. Thus, the protection afforded by early full-term pregnancy in women could be explained by the higher degree of differentiation of the mammary gland at the time in which an etiologic agent or agents act. The relevance of our work lies in the side by side comparison of in vivo and in vitro studies in the human breast that validates experimental data for extrapolation to the human situation. The finding that cell proliferation is of importance for cancer initiation, whereas differentiation is a powerful inhibitor, provides novel tools for developing rational strategies for breast cancer prevention and control.