To evaluate the roles of alcohol and genetic factors in hepatic iron overload, we studied prospectively 61 patients selected solely on the basis of increased stainable hepatic iron (grade 3 or 4). Independent comparisons were made between alcoholic (n = 20) and nonalcoholic (n = 41) patients, and between patients wih affected relatives (n = 25) and those without (n = 36). For the entire group, the mean value for mobilizable iron was 19.6 g and the prevalence of HLA-A3 was 69.6%, both findings compatible with genetic hemochromatosis. Subgroups were no different in clinical features (diabetes, pigmentation, cardiomyopathy, hypogonadism, or arthropaty), histologic findings (fat, inflammation, fibrosis), indexes of iron metabolism (serum iron, transferrin saturation, chelatable iron, and mobilizable iron stores), or frequency of HLA-A3 and HLA-B7. The only exception was that mean hepatic iron concentration was lower in alcoholic patients than in nonalcoholic patients (17,344 vs. 28,553 micrograms/g dry wt, p less than 0.001). Similarity between subgroups in almost all parameters examined is consistent with the hypothesis that heavy deposition of hepatic iron, as observed in our patients, is an indication of genetic hemochromatosis, regardless of alcohol consumption or the findings of affected relatives. The lower concentrations of hepatic iron in alcoholic patients, despite equal body stores in both groups, suggest that alcohol may alter the distribution of storage iron in genetic hemochromatosis.