In the state of Connecticut, 233 women with congenital heart defects were prospectively followed up through 482 pregnancies that resulted in 372 infants who were examined frequently during their first 3 years of life. Approximately half of the women had undergone cardiac surgery and they were compared with the women without operation. There was no maternal mortality, and no patient had infective endocarditis, brain abscess or a cerebrovascular accident. The proportion of pregnancies resulting in live births did not differ significantly in mothers with and without cardiac surgery; the average live birth rate was 77 percent in all. However, the number and size of live-born infants was much greater in mothers who had become acyanotic as a result of reparative surgery than in the still cyanotic women, whether or not they had had palliative surgery. In cyanotic women, placental size was abnormally large in relation to birth weight, which was abnormally low. When the mothers were classified according to cardiac function, there was a significant difference between the number of infants born alive to mothers in good to excellent status and the number born to mothers in fair to poor condition. The latter had a significant increase in interrupted pregnancies as well as in cardiovascular complications during pregnancy. The total group had a 16.1 percent incidence rate of infants with congenital heart disease. This rate was corrected to 14.2 percent by removal of seven mothers, two with Noonan's syndrome, one with hypertrophic cardiomyopathy and four with a family history of congenital heart defects.