The aromatic amines 2,4-diaminotoluene (2,4-DAT) and 2,6-diaminotoluene (2,6-DAT) are structural isomers that have been extensively studied for their mutagenic and carcinogenic characteristics. Both compounds are equally mutagenic in the Ames/Salmonella assay in the presence of S9. However, the differences in the results of chronic rodent carcinogen bioassays using these two compounds are significant, in that 2,4-DAT is a potent hepatocarcinogen, whereas 2,6-DAT does not produce an increased incidence of tumors in rats or mice at similar doses. The Big Blue transgenic B6C3F1 mouse carries multiple copies of bacteriophage lambda, each with a lacI mutational target gene, integrated into mouse chromosome 4. Our studies were designed to determine whether the Big Blue system could be used to detect differences in the in vivo mutagenic activity between the carcinogen/non-carcinogen pair 2,4- and 2,6-DAT and to determine whether the in vivo mutagenesis assay results correspond to the rodent carcinogen bioassay results. Male B6C3F1 transgenic mice were exposed to 2,4- or 2,6-DAT at 0 or 1000 p.p.m. in the diet for 30 and 90 days. Mice serving as positive controls were administered five daily i.p. injections of 6 mg/kg dimethylnitrosamine (DMN) in saline and were sacrificed 15 days following the last injection. Mutant frequencies at lacI were determined by recovering the genomically integrated lambda phage using an in vitro packaging reaction followed by infection of an appropriate Escherichia coli host. Complete non-sectored blue mutant plaques were scored against a background of clear non-mutant plaques. Mutant frequencies were nearly identical for all three groups at 30 days, while at 90 days the mutant frequency for the hepatocarcinogen 2,4-DAT (12.1 +/- 1.4 x 10(-5)) was significantly higher (P < 0.01) as compared with both age-matched (spontaneous) controls (5.7 +/- 2.9 x 10(-5)) and the 2,6-DAT-exposed group (5.7 +/- 2.4 x 10(-5)). Mutations at lacI arising ex vivo during replication in E. coli are observed in this system as sectored blue plaques. The sectored plaque frequency in this study was constant across all groups at approximately 9.0 x 10(-5). Results from this study demonstrate that the Big Blue transgenic mutation assay: (i) can distinguish differences in vivo between the mutagenic responses of a carcinogen and a non-carcinogen which elicited comparable mutagenic activity in S.typhimurium; (ii) is sensitive to mutagens through subchronic dietary exposure; and (iii) yields a differential response depending upon the length of time mice are exposed to a mutagen.