The potential genotoxicity of the rodent liver carcinogen 2,6-dinitrotoluene (2,6-DNT) was evaluated in compliance with the guidelines for genotoxicity studies of drugs (Notification No. 1604, Nov. 1, 1999, Ministry of Health and Welfare, Japan) and the OECD guidelines for the testing of chemicals by performing the bacterial reverse mutation (Ames) assay, the in vitro chromosomal aberration assay, and the in vivo comet assay (alkaline single cell gel electrophoresis) in rat liver. In the Ames assay, 2,6-DNT was moderately positive irrespective of metabolic activation. In the in vitro chromosomal aberration assay, under conditions where the test substance would precipitate out, weak structural aberrations were observed with or without S9 mix at each dose at which the cell growth rate was about 40 to 50%. The in vivo comet assay yielded positive results in rat liver; that is to say, the increases in % tail DNA in liver in the 25 and 50 mg/kg groups were observed statistically significantly and dose-dependent. Our findings are in accordance with previous findings in the in vivo/in vitro unscheduled DNA synthesis (UDS) assay in rat liver and in a young rat liver micronucleus assay, although the rat bone marrow micronucleus assay gave negative results. These results suggest that test systems using liver are a useful method for the in vivo genotoxicity assessment of chemicals that require metabolic activation.