The developmental impact of prenatal and early postnatal low-level lead exposure was assessed in a prospective study of 249 middle and upper-middle class infants with umbilical cord blood lead levels in the range currently considered "normal." Infants were classified into three exposure groups: "low" (less than 3 micrograms/dl), "mid" (6 to 7 micrograms/dl), and "high" (greater than or equal to 10 micrograms/dl). At 6 and 12 months, the lead concentration of capillary blood was measured, and the Bayley Scales of Infant Development administered. At both ages, Mental Development Index scores, adjusted for confounding, were inversely related to infants' umbilical cord blood lead levels. The difference between the mean adjusted scores of the infants in the low and high cord blood lead groups was 5.8 points at 6 months and 7.3 points at 12 months. At neither age were scores significantly related to postnatal blood lead levels. Prenatal exposure to lead levels relatively common among urban populations appear to be associated with less favorable development through the first year of life.