Alcohol is a potent teratogen associated with dysmorphology, growth retardation, and neurological damage in children with the full fetal alcohol syndrome (FAS); alcohol is also associated with growth retardation and behavioral alterations in neonates prenatally exposed to various dosages. Questions remain about the long-term consequences of prenatal alcohol exposure. This study reports on the follow-up of a subsample of 68 children, the majority of whom were low income and black (mean age: 5 years, 10 months) who were first evaluated as neonates. Physical and cognitive outcomes of 25 children of women who drank throughout pregnancy [absolute alcohol (AA)/week: mean = 11.80 oz), even after receiving an educational intervention to stop drinking, were compared with outcomes of children in two contrast groups: a) women (n = 22) who stopped drinking (AA/week: mean = 11.46 oz) in the second trimester after an educational intervention but resumed postpartum; and b) women who did not drink during pregnancy and who drank little postnatally (n = 21). Children were compared for alcohol-related birth defects (ARBDs), growth (height, weight, and head circumference), and cognitive, academic, and adaptive measures. Neonatal and current physical measures were correlated to determine predictability of neonatal status. When the effects of age and gender were controlled, children in the continued-to-drink group showed significantly more ARBDs and had smaller head circumferences than those in the other two groups. When current drinking reported by caretakers was controlled, the children who were exposed throughout pregnancy also showed significant and consistent deficits in several areas of intellectual functioning including sequential processing (short-term memory and encoding) and overall mental processing. Alcohol-exposed children displayed significant deficits in preacademic skills when compared with children of nondrinkers, with both alcohol groups deficient in premath and reading skills. There were no differences in adaptive behavior at follow-up. These data suggest that alcohol exposure throughout pregnancy is correlated with persistent physical differences as well as identifiable deficits in sequential memory processes and specific academic skills. However, even when alcohol use is limited to the first part of pregnancy, significant deficits in academic skills and growth parameters are measurable.