In the new USEPA cancer risk assessment guidelines, mode of action (MoA) information, combined with a determination of whether or not a chemical is mutagenic, plays an important role in determining whether a linear or nonlinear approach should be used to estimate cancer risks at low doses. In this article, carbon tetrachloride (CT) is used as an example to illustrate how mixed genotoxicity data can be evaluated and used to identify a likely MoA. CT is essentially negative in inducing gene mutations in Salmonella, but is consistently positive in inducing recombination and aneuploidy in fungi. Negative or equivocal results were seen in most in vitro and in vivo studies in mammals, including mutation studies in transgenic mice. However, DNA adducts, primarily those derived from oxidation- and lipid-peroxidation-derived products as well as DNA double-strand breaks and micronucleated cells, have been seen repeatedly in the liver of CT-treated animals. On the basis of the weight of evidence, CT should not be considered a directly mutagenic agent. Mutagenic as well as other genotoxic effects, as they occur, will most likely be generated through indirect mechanisms resulting from oxidative and lipid peroxidative damage and/or damage occurring during necrosis or apoptosis. As key events in this process are expected to occur in a nonlinear fashion, the expected relationship between CT dose and carcinogenic response in the liver is likely to be nonlinear with a steep dose response. This conclusion is consistent with rodent cancer bioassay results in which steep nonlinear dose responses have been seen.