Binary combinations of pure chemicals and complex mixtures were evaluated for their ability to produce additive mutagenicity responses in the spiral and standard Salmonella mutagenicity assays. Single chemicals were selected that were representative of the primary chemical class responsible for much of the mutagenic activity of each complex mixture. The following agents were evaluated in the absence of S9: 1-nitropyrene, diesel exhaust extract, and the chlorinated drinking water mutagen 3-chloro-4-dichloromethyl-5-hydroxy-2-[5 H]-furanone (MX). In the presence of S9, the following agents were evaluated: 4-aminobiphenyl, benzo[ a]pyrene, and an organic extract from the particulate emissions resulting from the combustion of polyethylene in a rotary kiln incinerator. Binary combinations of the agents within each S9 group were tested. The results were analyzed for additivity by determining whether the difference between the expected response of the binary mixture was significantly different from the observed response. In general, the combinations of mutagens produced additive responses at low doses in both assays. However, at high doses, the results generally departed from additivity, especially for combinations of indirect-acting mutagens. A requirement for exogenous metabolic activation imposes conditions, such as different S9 optima for each mutagen, that complicate the application of the additivity assumption to binary combinations of indirect-acting mutagens. Although the two bioassays generally gave similar results for each binary combination of mutagens, the spiral assay permitted a gradient of mutagen and/or S9 concentrations to be evaluated, and it allowed the mutagens to be evaluated in the absence of solvents, eliminating any potential solvent-mutagen interactions. These features of the spiral assay may be advantageous when studying chemical interactions.