Breast cancer is the second leading cause of cancer deaths among women in the U.S. today, claiming 40,000 lives per year. Several factors, including societal influences on childbearing, diet, and environmental chemicals are believed to be impacting the high rate of breast cancer in this country. Reproductive factors are particularly important in the increased incidence in the U.S., two major factors being the depression in the age of menarche and the extension in the time before first full-term pregnancy. Reasons for the depression in the age of menarche are not fully understood, but appear to be due in part to diet, whereas the delay of childbearing is primarily social. Because the contemporary development and maturation of the breast is during this extended time from menarche to pregnancy, there is further time for toxic influences to have an opportunity to induce tumor formation. Our hypothesis is that diet, in particular fatty acid and phytoestrogen composition and quantity, during the perinatal period and childhood determine the level of adiposity and regulate the hormonal milieu in young children. Through leptin and insulin-like growth factor, adiposity determines the pathway of puberty as being driven primarily by the adrenal gland (adrenarche) or ovaries (thelarche), the latter being associated with early menarche and subsequent risk of breast cancer. In prospective cohorts from in utero to age 3 and age 6 to 12, we shall test the association of diet and adiposity, as well as environmental agents, eg endocrine disruptors and carcinogens, and qualities of the psychosocial environment, with pubertal pathway. Through analysis of hormones, growth factors, and aromatase we shall examine mechanisms through which diet may mediate these effects. In rodent studies we shall test the hypothesis that dietary fatty acid and phytoestrogen composition in utero and early life alter puberty and mammary gland maturation. We further predict that these factors alter the periods of life at which mammary glands are most susceptible to carcinogenic insults. Using the power of gene expression arrays we shall define characteristics of initiated mammary epithelial cells that can be used to examine endogenous and exogenous compounds for their carcinogenic potential and better define initiated mammary cells in animal models and human studies.