1,3-Butadiene (BTD) is an important commodity chemical and air pollutant that has been shown to be a potent carcinogen in mice, and to a lesser extent, a carcinogen in rats. To better assess butadiene's carcinogenic risk to humans, it is important to understand its mode of action and how this relates to differences in responses among species. In a series of in vitro experiments, lymphocytes from rats, mice, and humans were exposed to 3,4-epoxy-1-butene (EB) or 1,2:3,4-diepoxybutane (DEB) for 1h at the G(0) stage of the cell cycle, stimulated to divide, and cultured to assess the ability of these metabolites to induce sister chromatid exchange (SCE) and chromosome aberrations (CAs). EB induced no increases in SCEs or CAs in the cells from the three species. DEB was a potent SCE- and CA-inducer, with the results being similar in each rodent species. The response for SCEs seen in the human cells was more complex, with genetic polymorphism for glutathione-S-transferases (GST) possibly modulating the response. The single cell gel electrophoresis assay was used on genetically engineered V79 cell lines to investigate a possible influence of GST status. Experiments were also conducted to investigate the reason for EB's failure to induce SCEs or CAs in G(0) cells. The results indicate that EB-induced DNA damage was repaired before DNA synthesis in unstimulated lymphocytes, but EB caused a large increase in SCEs if actively cycling cells were treated. Thus, the results indicate that DEB damage is persistent in G(0) cells, and DEB is a much more potent genotoxicant than EB. The carcinogenic effect of butadiene will most likely depend on the degree to which DEB is produced and reaches target tissues, and to a lesser extent on the ability of EB to reach actively dividing or repair deficient cells.