General health, growth and sexual development were evaluated in 74 teenage boys and 98 girls who had been exposed to medroxyprogesterone acetate (MPA) in utero, and 385 boys and 448 girls not exposed. In this 17-year prospective study, the ascertainment of the end points was 'double blind' in that neither the interviewer nor the subject was aware of our interest in MPA. On average, girls exposed to MPA reported reaching the menarche 4 months earlier than the comparison group. This difference disappeared, however, in a multiple regression analysis taking into account social class, the mother's age at menarche and height of the girl's mother and father. Boys exposed to MPA reported their growth spurt to have occurred an average of 6 months earlier and voices to have broken 5 months earlier than unexposed boys. Again, the differences between them and the comparison group disappeared after controlling for confounding variables. There were no significant differences between the MPA-exposed and comparison groups in a wide variety of indices of health reported by the teenagers' mothers. These findings are consistent with the hypothesis that intrauterine exposure to MPA, in the doses used for pregnancy maintenance or for contraception, poses no threat to the long-term health and development of the progeny.