The authors evaluated the effects that maternal bone lead stores have in anthropometry at birth in 223 mother-infant pairs. The participants were recruited between April and November 1994. Anthropometric data were collected within the first 12 hr following delivery. Maternal information was obtained 1 mo after delivery occurred. Bone lead burden was determined with in-vivo K-x-ray fluorescence of the tibia (cortical bone) and the patella (trabecular bone). The authors transformed anthropometric measurements to an ordinal 5-category scale, and the association of measurements with other factors was evaluated with ordinal logistic-regression models. Mean bone lead levels were 9.8 microgram/gm bone mineral and 14.4 microgram/gm bone mineral for the tibia and patella, respectively. Birth length of newborns decreased as tibia lead levels increased. Compared with women in the lower quintiles of the distribution of tibia lead, those in the upper quintile had a 79% increase in risk of having a lower birth length newborn (odds ratio = 1.79; 95% confidence interval = 1.10, 3.22). The authors adjusted by birth weight, and the effect was attenuated--but nonetheless significant. Patella lead was positively and significantly related to the risk of a low head circumference score; this score remained unaffected by inclusion of birth weight. The authors estimated the increased risk to be 1.02 per microgram lead/gm bone mineral (95% confidence interval = 1.01, 1.04 per microgram lead/gm bone mineral). Odds ratios did not vary substantially after the authors adjusted for birth weight and other important determinants of head circumference.