To assess the potential effect of maternal environments on human embryonic/fetal somatic mutation, we measured the frequencies of hypoxanthine-guanine phosphoribosyltransferase (HPRT, hprt gene), mutant T lymphocytes (Mf), and glycophorin A (GPA) variant erythrocytes (Vf) of both allele-loss (phi/N) and allele-loss-and-duplication (N/N) phenotypes in umbilical cord blood. The mean hprt Mf (1.40 +/- 1.11 x 10(-6), N = 66) and GPA Vf (phi/N 4.0 +/- 2.2 x 10(-6), N = 114; N/N 2.7 +/- 2.0 x 10(-6), N = 91) were significantly lower than those previously reported for adult populations. In addition, the hprt Mf was significantly higher than that of a published study of newborn cord blood samples from a geographically distant population (0.64 +/- 0.41 x 10(-6), N = 45, P < 0.01; t test, P < 0.01, Mann-Whitney U test). An examination of the demographic data from these two populations led to the sampling of 10 additional newborns specifically matched to the published study for maternal socioeconomic status. The hprt Mf (0.70 +/- 0.49 x 10(-6)) of this selected population was consistent with the published report and significantly lower than that of our initial population (P < 0.03, t test; P < 0.01, Mann-Whitney U test). These results indicate that there is an environmental effect related to maternal socioeconomic status on the frequency of embryonic/fetal somatic mutations. Molecular analyses of hprt mutants from this cohort with elevated Mf revealed a significant decrease in the relative contribution of gross structural mutations to the overall Mf (25 of 38, 66% vs. 34 of 41, 83%, P = 0.024, chi 2 test), suggesting that the higher Mf resulted from an elevated level of "point" mutations. No individual maternal demographic or environmental factor was identified as contributing more significantly than other any factor to the observed variability in hprt Mf or GPA Vf.