Several variations on the standard primary rat hepatocyte DNA/repair assay were evaluated for their ability to enhance the sensitivity of this genotoxicity test system. The use of hamster hepatocytes proved to be a much more sensitive system than rat hepatocytes for detecting the DNA repair inducing ability of the nitrosamines, dimethylnitrosamine and diethylnitrosamine, and the aromatic amines, 2-acetylaminofluorene, 9-aminoacridine, 1naphthylamine and benzidine. In addition, hamster hepatocytes were a more sensitive indicator of the genotoxicity of the azo dyes, o-aminoazotoluene, Congo Red and Evans Blue. However, the azo reduction product of the azo dyes Congo Red, Trypan Blue and Evans Blue, benzidine and o-tolidine, respectively, were active in both rat and hamster hepatocytes at concentrations that were 10-lOO foM lower than the parent dyes. This suggests that little or no azo reduction of the dyes occurred in the in vitro assay systems. The in vivo-in vitro variation of the rat hepatocytes DNA/repair assay exhibited a positive DNA repair response with the azo dye solvent Yellow 5, which was negative in the standard in vitro assay. The in vivo-in vitro hepatocyte DNA repair assay was also more sensitive for detecting the genotoxic activity of Evans Blue, which was positive in the in vivo-in vitro assay and equivocal in the standard in vitro assay. Also, Solvent Yellow 14 was negative in the in vitro assay, but induced an equivocal DNA repair response in the in vivo-in vitro assay system. A treatment/ 3 H-thymidine labeling period of approximately 18 hours, compared to 4 hours, was demonstrated to be superior for detecting the DNA repair elicited by the mutagens 4-nitroquinoline-l-oxide, mitomycin C, dimethylnitrosamine and methyl methanesulfonate in the in vitro rat hepatocyte assay. There was little or no difference observed between the 4 hour and 18 hour treatment/labeling incubation periods for the detection of DNA repair induced by 2-acetylaminofluorene, ariatoxin B~, and benzidine. The data suggest that these several variations on the standard rat hepatocyte DNA/repair assay should be considered when evaluating the genotoxicity of chemicals for safety purposes.