The usefulness of the 32P-post-labeling/t.l.c. method for quantitative DNA adduct dosimetry was evaluated. 2-Acetylaminofluorene (2-AAF)-DNA adducts from three systems were characterized qualitatively and quantitatively by the 3H-radiolabeled technique with subsequent analysis by h.p.l.c. (pre-labeling method) and by the 32P-post-labeling method. Both methods showed N-acetoxyacetylaminofluorene (N-OAc-AAF) reaction products with calf thymus DNA were predominantly N-(deoxyguanosin-8-yl)-2-acetylaminofluorene (dG-C8-AAF) with some N-(deoxyguanosin-8-yl)-2-aminofluorene (dG-C8-AF) and N-(deoxyguanosin-N2-yl)-2-acetylaminofluorene (dG-N2-AAF). In contrast, Chinese hamster ovary (CHO) cells treated with [3H]N-OAc-AAF gave 80 or 90% dG-C8-AF adducts and 20 or 10% dG-C8-AAF adducts with the post- or pre-labeling method, respectively. Likewise in CHO cells treated with 2-AAF in the presence of rat liver homogenate, approximately 90% dG-C8-AF and 10% dG-C8-AAF adducts were detected using the 32P-post-labeling method. In Salmonella typhimurium strain TA1538 treated with 2-AAF or [3H]2-AAF in the presence of a rat liver homogenate, one adduct, dG-C8-AF, was identified. Similar quantitative results were also obtained with the two methods. However, the 32P-post-labeling method was more sensitive and also eliminated the use of radiolabeled-mutagen treatments. Quantitative DNA adduct dosimetry was applied to AAF-induced mutagenesis in the S. typhimurium and CHO/HPRT mutation assays. A linear and reproducible relationship existed between dG-C8-AF levels and AAF-induced mutants in both systems.