Abstract2‐Chlorethanol, 8‐hydroxyquinoline, 2,6‐toluenediamine, and eugenol, previously found to behave as genotoxins in in vitro systems and as noncarcinogens in rodents, were evaluated for their ability to induce genotoxic effects in vivo. Rats were given by gavage a single or two successive doses equal to one‐half the corresponding LD50, killed at different times after treatment, and examined for the following end points: the frequency of both micronucleated polychromatic erythrocytes in the bone marrow and micronucleated hepatocytes (after partial hepatectomy); the in vivo—in vitro induction of DNA fragmentation, as measured by the alkaline elution technique, and of unscheduled DNA synthesis, as measured by autoradiography, in hepatocyte primary cultures. The two latter end points were also evaluated after in vitro exposure of hepatocytes to log‐spaced subtoxic concentrations. 2‐Chloroethanol, 8‐hydroxyquinoline, and eugenol never produced effects indicative of genotoxic activity. The same happened with 2,6‐toluenediamine, with the exception of a significant increase over controls in the amounts of DNA damage and repair displayed by hepatocyte cultures obtained from rats given two 1/2 LD50 separated by a 24 h interval. Our results, which, apart the above mentioned exception, are in concordance with the rodent carcinogenicity results, contribute to underline the role of in vivo short‐term tests for the detection of potential genotoxic carcinogens. ©