The families of 30 workers employed in a plant which, until 1977, had produced 1,2-dibromo-3-chloropropane (DBCP), have been monitored for 5 yr. Thirteen men who were azoospermic in 1977 had more than 100 hr of estimated exposure in production; of these, 4 have recovered spermatogenesis and 2 have fathered 4 female children. Eight men who were oligospermic in 1977 had been exposed for 34-95 hr; of 8 children subsequently born to their wives, 6 were female. Five men with exposure times estimated to be between 10 and 60 hr were normospermic in 1977, and of 5 children subsequently born to them, 4 were male and 1 was female. The outcome of the pregnancies among families of 4 non-exposed men and of exposed men prior to exposure resulted in 27 male and 24 female children. No increase in spontaneous abortion occurred in pregnancies fathered by exposed men. Four infant deaths and 5 congenital anomalies occurred among the babies born of unexposed pregnancies. No infant or fetal deaths and no congenital abnormalities occurred among children born to wives of men after the men were exposed. The birth of at least 10 females of 12 babies born to wives of exposed men with evidence of testicular toxicity would be expected according to binomial probabilities with a P = .015. The children born to families of exposed workers are in good health. A possible explanation for the observed female sex predominance might be Y chromosome non-disjunction, which was reported to be more frequent in sperm of DBCP workers by Kapp et al.