Although estrogens have been identified as key endocrine hormones in the control of early mitogenesis and development in the mammary gland, local control of cell proliferation during ductal morphogenesis may be regulated by polypeptides such as TGF-alpha or TGF-beta. Many breast tumors are estrogen dependent, and some breast tumor cell lines are known to produce TGF-alpha, suggesting that the mitogenic pathways controlling early normal mammary growth and the growth of some breast tumors may be similar. While progesterone does not appear to be important in the early program of ductal growth, progesterone and estrogen are necessary for the cyclic proliferation of mammary ductal cells that occurs during the menstrual cycle, and for lobuloalveolar growth during pregnancy. Since increased cell division enhances the chances for the formation of a malignant phenotype in the breast, exogenous hormones containing estrogen alone or estrogen and progesterone may increase breast cancer risk. While DES is no longer prescribed to prevent abortions, it demonstrates that high doses of an estrogen during a period of mammary proliferation can affect breast cancer risk. Whether the addition of progestogens to estrogen replacement therapy enhances breast cancer risk in postmenopausal women remains an unanswered question because of the lack of large, well-controlled prospective studies. There currently is no evidence to indicate that the progestogen-containing subdermal contraceptive Norplant increases breast cancer risk. However, it has not been determined if the elevation of serum estrogens reported in some Norplant users affects breast cancer risk. There is little evidence that combined OCAs enhance breast cancer risk in most women. More research is needed to substantiate the findings that OCA use in young women, especially before a first full-term pregnancy, may enhance breast cancer risk. Animal studies indicate that there are critical periods of susceptibility to chemical carcinogens, since the number and malignancy of tumors are increased when carcinogens are administered to young virgin animals during the proliferative period of ductal morphogenesis. Since the breast appears to be most susceptible to the carcinogenic effects of ionizing radiation during the first decade of life, exposure to other carcinogenic agents during the period of early breast development may be important in determining breast cancer risk. Therefore, more studies are needed to confirm the observation that heavy drinkers and heavy smokers are at higher risk for developing breast cancer when they start smoking or drinking at an early age. The observation that serum and urinary estrogen levels increase with alcohol consumption may provide a basis for the higher risk of developing breast cancer in heavy drinkers. While the restriction of methyxanthine intake may alleviate the symptoms associated with fibrocystic breast disease in some women, there is not enough evidence to suggest that a reduction in caffeine intake will reduce breast cancer risk. Evidence for an association between electromagnetic radiation and breast cancer is limited. Electromagnetic radiation may only pose a risk in certain occupations with exposure to very high levels for extended periods of time. It is not known whether exposure to PCBs transplacentally or though the lipid fraction of human milk can affect breast cancer rates in female offspring. The higher risk of breast cancer in women with elevated DDE levels in their blood underscores the importance of determining the extent to which environmental contaminants affect breast cancer risk.