Previous studies of fetal growth and body proportionality have been based on error-prone gestational age estimates and on inappropriate comparisons of infants with dissimilar birth weights. Based on a cohort of 8719 infants with validated (by early ultrasonography) gestational ages and indexes of body proportionality standardized for birth weight, potential maternal and fetal determinants of fetal growth and proportionality were assessed. Maternal history of previous low birth weight infants, pregnancy-related hypertension (particularly if severe), diabetes, prepregnancy weight, net gestational weight gain, cigarette smoking, height, parity, and fetal sex were all significantly associated with fetal growth in the expected directions. Consistent with previous reports, maternal age, marital status, and onset or total amount of prenatal care had no significant independent effects. Fetal growth ratio (relative weight for gestational age), pregnancy-related hypertension, fetal sex, and maternal height were the only significant determinants of proportionality. Infants who were growth-retarded, those with taller mothers, those whose mothers had severe pregnancy-related hypertension, and males tended to be longer and thinner and had larger heads for their weight, although these variables explained only a small fraction of the variance in the proportionality measures. Among infants with intrauterine growth retardation, gestational age was not independently associated with proportionality (in particular, late term and post-term infants did not tend to be more disproportional), a finding that does not support the hypothesis that earlier onset of growth retardation leads to more proportional growth retardation. The results raise serious questions about previous studies of proportionality, particularly those suggesting a nutritional etiology for proportional intrauterine growth retardation.(ABSTRACT TRUNCATED AT 250 WORDS)