Maternal diabetes mellitus is complicated by fetal macrosomia and predisposes the offspring to diabetes, but recent evidence indicates that a low, not high, birthweight is associated with a higher incidence of Type 2 (non-insulin dependent) diabetes in adult life. To clarify the relationships between maternal glucose and insulin levels and birthweight, we measured oral glucose tolerance and neonatal weight in a large group (n = 529) of women during the 26th week of pregnancy. Women with gestational diabetes (n = 17) had more familial diabetes, higher pre-pregnancy body weight, and tended to have large-for-gestational-age babies. In contrast, women with essential hypertension (n = 10) gave birth to significantly (p < 0.01) smaller babies. In the normal group (without gestational diabetes or hypertension, n = 503), maternal body weight before pregnancy and at term, maternal height, week of delivery, gender of the newborn, and parity were all significant, independent predictors of birthweight, together explaining 23% of the variability of neonatal weight. In addition, both fasting (p < 0.006) and 2-h post-glucose (p = 0.03) maternal plasma glucose concentrations were positively associated with birthweight independent of the other physiological determinants, accounting, however, for only 10% of the explained variability. In a subgroup of 134 normal mothers with pre-pregnancy body mass index of less than 25 kg.m-2, in whom plasma insulin measurements were available, the insulin area-under-curve was inversely related to birthweight (p < 0.02) after simultaneously adjusting for physiological factors and glucose area.(ABSTRACT TRUNCATED AT 250 WORDS)