The present studies assessed the impact of various sources of garlic and their constituents (water- and ethanol-extracts and S-allylcysteine) on the in vivo binding of the carcinogen 7,12-dimethylbenz[a]anthracene (DMBA) to rat mammary cell DNA. The provision of dietary raw garlic powder (2%) or its water-extract (1.5%) reduced DMBA-DNA binding by 33 and 46% respectively. Dietary supplementation with a commercially available deodorized garlic powder (powder A) at 2 or 4% depressed the occurrence of adducts by 50 and 78% respectively, while providing a commercially available high sulfur garlic preparation (powder B) at 2% reduced binding by 56%. A pair-feeding study revealed that the depression in carcinogen binding was independent of food intake or weight gain. Although 1% raw garlic powder did not significantly influence the occurrence of DMBA-DNA adducts, an equivalent as the water-extract (0.75%), the ethanol-extract (0.015%) or commercially available powders (A and B) reduced DMBA adducts in mammary tissue by 44, 25, 71 and 65% respectively. Dietary fortification with S-allylcysteine (SAC), a water-soluble constituent of processed garlic, caused a progressive decrease in the binding of DMBA to DNA. Studies with SAC suggest the primary effect of garlic and its constituents is on the bioactivation and binding of the carcinogen rather than DNA repair. These data reveal that several forms of garlic are effective, although variable, in altering carcinogen bioactivation and presumably chemically induced carcinogenesis.