Antepartum dexamethasone administration was associated with a significant lowering of the incidence of respiratory distress syndrome (RDS) and with marked reduction of early neonatal mortality in premature infants. Dexamethasone had its greatest effect in infants delivered between 28 and 32 weeks' gestation (P < .001), and the difference was still significant up to 34 weeks' gestation (P < .05). Although the effect of dexamethasone was more marked in patients with intact membranes, it was also effective in those with prolonged rupture of the membranes. Membrane condition before delivery was unrelated to the incidence of RDS. The effectiveness of dexamethasone was analyzed with respect to the condition of the infant at delivery, the mode of delivery, and multiple births. As compared to the controls, dexamethasone was most effective in singletons, infants delivered vaginally, and those with high Apgar scores. The effect of dexamethasone was not significant in twins, breech deliveries, cesarean section deliveries, or infants with low Apgar scores. The effects of dexamethasone appear to be modified by intrauterine asphyxia. Dexamethasone therapy cannot be substituted for optimal delivery conditions of the premature.