Trends in survival were examined for infants in five birth-weight groups (500 to 999 g, 1,000 to 1,499 g, 1,500 to 1,999 g, 2,000 to 2,499 g, and greater than or equal to 2,500 g). The study population consisted of infants born in Georgia from 1974 through 1981. Survival to the first birthday increased for all birthweight groups. Improvements in postneonatal survival occurred concurrently with improvements in neonatal survival, except for infants with birth weights of 500 to 999 g. Between 1974-1975 and 1980-1981, the proportion of deaths that were attributed to perinatal conditions but occurred during the postneonatal period increased from 1.1% to 4.7%. These observations suggest that advances in perinatal care have led to postponement rather than prevention of a few infant deaths, although overall many more deaths were prevented. For infants with birth weights of less than 1,500 g, lower mortality among those admitted to newborn intensive care units was the major contributor to improved survival. Our findings indicate that both neonatal and postneonatal mortality should be examined in evaluating the effects of perinatal care.