Although alcohol is known to be a carcinogen for humans, ethanol-genotoxicity studies are incomplete. Ethanol seems not to be a bacterial mutagen, but the results are conflicting in rodent assays. We investigate the genotoxicity in the bone marrow micronucleus (MN) test and in the dominant lethal mutation (DLM) assay using two long-term ethanol exposure protocols. In the MN test, mice consumed three doses (5, 10 and 15% v/v) for 32 weeks. MN induction was compared to two control groups of 5- and 38-week-old mice (the ages of the treated mice when the treatment was initiated and when they were killed, respectively). For the three groups treated with ethanol there was no significant increase in MN induction as compared to the first control group, but observed MN frequencies were significantly lower than in the 38-week-old control group. This suggests a protective effect against genotoxic damage caused by aging, probably due to ethanol action as a hydroxyl radical scavenger. In the DLM assay, male mice drank ethanol at 15% or 30% (v/v) for 20 weeks. In both groups the number of dead implants was similar to the control, but there was a significant reduction in total implants, indicating a pre-implantation loss.