AbstractThe reliability of the L5178Y TK±forward mutation assay as a rapid screen for genotoxicity was evaluated by testing 63 coded chemicals. Replicate treatments were used, and at least two independent experiments were performed for each test condition. The test conditions consisted of no exogenous activation, activation by Aroclor 1254‐induced Fischer 344 rat liver S9 homogenate, and in some cases activation by noninduced Fischer 344 rat liver S9. The results were organized into tables that show the mutant colony counts, mutant frequency, and toxicity for each test chemical treatment, positive control treatment, and solvent negative control cultures. The repeat experiments were highly consistent and yielded contradictory evaluations for only a few of the chemicals studied. Fifty‐one of the chemicals (81%) were evaluated as mutagenic under one or both of the test conditions. A range in minimum effective concentrations of almost 106‐fold (0.008 to 5,000 μg/ml) was observed among the mutagenic chemicals. Nine chemicals (14%) were considered to be nonmutagenic. Three chemicals (progesterone,p‐rosaniline HCl, and 1,1,1‐trichloroethane) gave responses that were not easily evaluated under any test condition: evidence for mutagenesis was obtained in some experiments but not for all repeat studies. Under nonactivation conditions, specifically, the mutagenic activities of 4,4′‐bis(dimethylamino)benzophenone, progesterone, andp‐rosaniline HCl remained uncertain. With S9 activation, uncertain evidence for mutagenesis was obtained for 2‐naphthylamine, progesterone, and 1,1,1‐trichloroethane. In some cases, changes in the treatment conditions could lead to different evaluations of the mutagenic activity, and these possibilities are discussed in the descriptive evaluations of each chemical. Comparisons of the observed responses with published results were possible for 29 of the compounds and yielded highly con