The relationship between prenatal low-level lead exposure and fetal growth was evaluated in a sample of 4354 pregnancies in which the mean umbilical cord blood lead level was 7.0 micrograms/dl (SD = 3.3; 10th percentile, 3.4 micrograms/dl, 90th percentile, 10.9 micrograms/dl). Higher cord blood lead levels were significantly associated with gestations of slightly longer duration. Comparing infants with cord blood lead levels greater than or equal to 15 micrograms/dl to those with levels less than 5 micrograms/dl, adjusted risk ratios of 1.5 to 2.5 were observed for low birth weight (less than 2500 g) and for fetal growth indices that express birth weight as a function of length of gestation (e.g., small-for-gestational age, intrauterine growth retardation). The 95% confidence intervals of these risk ratios included 1, however, precluding rejection of the null hypothesis of no association. We conclude that the risk of adverse fetal growth is not increased at cord blood lead levels less than 15 micrograms/dl but that modest increases in risk may be associated with levels greater than or equal to 15 micrograms/dl.