Among 288 breast cancer patients (118 with bilateral disease and 165 with diagnosis before 40 years of age), we identified 26 families with a history of breast cancer, including a minimum of three first- or second-degree relatives. Complete pedigrees with verified malignancy data from the Finnish cancer registry were constructed for 22 families. The median age at breast cancer diagnosis of the young probands (< 40 years of age) was 35 years and of bilateral probands was 54 years. The relatives of the young probands were diagnosed with breast cancer at a younger age (median age 54 years) than the relatives of the older (bilateral) probands (median age 60 years). Standard life-table methods were used to compare the risk of breast cancer in the family members with that of the general population. Among the relatives of the young probands, the increased breast cancer risk occurred in the early post-menopausal period, whereas the risk estimate for the relatives of the bilateral probands closely followed that of the general population. In both groups, however, those family members reaching the age of 80 years had a cumulative probability of over 50% of developing breast cancer. The standard life-table method proved useful when assessing the age-specific risk for familial breast cancer, taking into account numerous family members as well as their age at disease onset. This kind of analysis can be performed in populations for which reliable cancer registry data are available. It provides a useful tool for selecting individuals for imaging and mutation screening, counselling and experimental chemoprevention programmes.