The aim of this study was to screen for potential agents affecting the mutagenicity of tobacco products. The influence of a number of compounds which have been suggested to be antimutagenic, some of which are present in tobacco products, was investigated on the mutagenicity of a cigarette smoke condensate (CSC) and, in some cases, an extract of oral Swedish moist snuff (SNUS), using a screening procedure of the Ames Salmonella/ microsome assay (STY). For some of the compounds the V79/hprt mutagenicity assay with benzo[a]pyrene metabolites as mutagens was used to obtain complementary and confirmatory information on mammalian cells. The antimutagens used included two selenium compounds, sodium selenite and ebselen; the flavonoids and polyphenols, ellagic acid, (+)-catechin hydrate, scopoletin, chlorogenic acid and rutin trihydrate; the porphyrin derivatives, bovine hemin, biliverdine dihydrochloride, chlorophyllin and a plant extract containing chlorophyll; the terpenoids, beta-carotene, retinol and a mixture of the two epimers (4R) and (4S) of (1S,2E,6R,7E,11E)-cembra-2,7,11-triene-4,6-diols (CBD); and cyclohexanol and ubiquinone. Screening of antimutagenic activities using the STY involves problems with toxicity. In several cases in this study mutagenicity was decreased below the control level without signs of toxicity in the background growth of bacteria. Since the survival of mutants and slight bacteriostatic effects on the background growth cannot be determined accurately in the STY, a reduction in mutagenicity may simply be due to toxicity. Only in cases where a dose-response curve declines to a level at or above the background and then levels off, can toxicity be excluded. An antimutagenic effect determined using this test system is therefore often not sufficient for classifying a compound as antimutagenic until these findings are confirmed in other test systems and, preferably, the mechanism behind this effect is clarified. The results obtained with the selenium compounds were considered to be inconclusive since the reduction in the mutation rate declined below the background level and might only reflect the toxic effects of these compounds. For ellagic acid an almost complete inhibition of the mutagenicity of CSC and SNUS in STY was indicated. This indication of antimutagenicity was confirmed in V79 cells using two metabolites of the CSC constituent benzo[a]pyrene, i.e., trans-7,8-dihydroxy-7,8-dihydrobenzo[ a]pyrene and (+)-7 beta,8 alpha-dihydroxy-9 alpha,10 alpha-oxy-7,8,9,10-tetrahydrobenzo[a]pyrene (BPDE). Chlorogenic acid and (+)-catechin reduced the mutagenicity of CSC and chlorogenic acid also strongly inhibited SNUS mutagenicity. Scopoletin and rutin trihydrate inhibited the mutagenicity of CSC, but showed confounding effects with SNUS. Toxicity could not be excluded for these compounds. Among the porphyrin derivatives, bovine hemin, chlorophyllin and also the extract containing chlorophyll reduced the mutagenicity of CSC in STY down to control level. For bovine hemin and chlorophyllin toxicity could be excluded due to the shape of the dose-response curves. Biliverdine dihydrochloride was found to be less effective in reducing the mutagenicity of CSC. beta-Carotene increased the mutagenicity of CSC in strain TA100 and decreased the mutagenicity with strain TA98. Retinol decreased the mutagenicity of CSC in both these strains. CBD decreased the mutagenicity of CSC to below the control level in strain TA100, indicating toxicity. The frequency of BPDE-induced gene mutations in V79 cells was not affected by CBD. Cyclohexanol was highly toxic towards Salmonella, while ubiquinone did not influence the mutagenicity of CSC in STY. Thus, the screening in this study showed three of the compounds, ellagic acid, chlorophyllin and bovine hemin, to exhibit clear antimutagenic properties towards tobacco-derived mutagens, while for the other compounds for which antimutagenicity was indicated the data have to be confirmed in another test system.