The objective of this study was to examine if individuals living near a uranium processing site have greater mutagenic damage, as measured by three mutagenicity assays, compared with subjects unexposed to any nuclear facilities. The design was a cross-sectional exploratory analysis of 112 subjects; 56 volunteer residents were from within a 5-mile radius of the Fernald Uranium Processing site and 56 'control' subjects were from a geographically separate area unexposed to any known uranium emissions. The groups were constrained to be similar in age and sex composition. The main outcome measures were three human somatic gene mutation assays consisting of the HPRT T-lymphocyte cloning assay to measure 6-thioguanine resistant lymphocytes; the glycophorin A assay to detect the loss of expression of the M or N allele; and the micronucleus assay as a marker of chromosomal damage. The results showed no statistically significant or quantitatively important differences between groups for all three mutagenicity assays; only the unselected cloning efficiency was statistically significantly different between groups (0.42 +/- 0.16 for the Fernald versus 0.35 +/- 0.12 for the comparison groups). In both groups, age was significantly related to HPRT mutant frequency, with a 1.25% rate of increase in mutant frequencies for each 1-year gain of age in the Fernald group and a 1.12% rate of increase in mutant frequencies for each 1-year gain of age in the comparison group. For the micronucleus data, females had a greater mean micronucleus frequency than males. In addition, smokers had an increased mean ln (natural logarithm) HPRT mutant frequency (3.06 +/- 0.14 for current smokers compared with a mean of 2.72 +/- 0.05 for non-current (i.e. never plus former) smokers). Our results are consistent with the previously reported association between sex type and micronucleus frequency, the known relationship between age and T-lymphocyte cloning efficiency and age and HPRT mutant frequency, and verify the wide inter-subject variability for the latter. Finally, we conclude that at a population level, the relationships between current cigarette use and HPRT mutant frequency, and sex type and micronucleus frequency, are stronger than is the association between geographic proximity to a uranium processing site and mutagenic abnormalities.