The association between body height and the incidence rate of breast cancer has been examined in 236 cases of breast cancer that occurred among 23,831 Norwegian women during 11-14 years of follow-up. At the time of height measurement they were 35-51 years of age. The age-adjusted incidence rate ratio (IRR) of breast cancer was 2.03 (95% of confidence limits 1.36 and 3.01) for women taller than or equal to 167 cm (mean = 170 cm) compared to women who were less than 159 cm (mean = 155 cm). The positive association with height was stronger among women who were diagnosed before the age of 51 (IRR = 2.63; 95% confidence limits 1.48 and 4.68), than among women diagnosed after this age. Moreover, the association appeared to be confined to women who had lived through their peripubertal growth during a period (1940-45) of nationally increased nutritional variability with reduction in dietary fat and restricted caloric intake. Among women born between 1929 and 1936, the relation with height displayed a strong positive linear trend (chi 2 trend = 13.4, P less than 0.001), which was not present among women born between 1925 and 1928 (chi 2 trend = 0.7, P = 0.40), nor among women born in 1937 or later (chi 2 trend = 1.5, P = 0.20). We hypothesise that a time-dependent diversity in nourishment, which may be of particular importance for women in their peri-menarcheal development, may explain the different association between body height and breast cancer risk that was observed for women in different birth cohorts.