Information on family history and age of onset of breast cancer is examined in an attempt to distinguish between postulated early and late onset forms of breast cancer and to estimate the familial correlation between any subforms that may exist. Data are taken from the Cancer and Steroid Hormone Study, a multicentre, population-based, case-control study conducted by the Centers for Disease Control (CDC). The data set includes 4730 histologically confirmed breast cancer cases aged 20-54 and 4688 controls who were frequency-matched to cases by geographic region and five-year categories of age. Data regarding family history of breast cancer for mothers and sisters of cases and controls are used in this analysis. The methods of Smith (1976) are modified to obtain maximum likelihood estimates of heritability and familial correlation using 35, 40, 45, and 50 years of age as the dividing point between early and late onset. The heritabilities for both the early and late onset forms decrease and approach one another in size with increasing age cutpoint, although the heritability for the early group is always greater than that for the late group. Irrespective of cutpoint, the familial correlation between the two groups does not differ significantly from one, providing evidence for a single genetic form of breast cancer with familial cases increasingly represented among cases with earlier onset.