Is Anguilla anguilla L. (eel) liver ethoxyresorufin O-deethylase (EROD) induction absolutely necessary in order to convert promutagens as benzo[a]pyrene (BaP) into a mutagenic compound? Eels were exposed for 8 h to clean (control) and 0.3 microM beta-naphthoflavone (BNF)-contaminated water. The 8-h exposure to 0.3 microM BNF brought about a very high EROD induction (10 pmol/min/mg protein) compared to control animals (1 pmol/min/mg protein). The Ames test (Maron and Ames, 1983) was carried out with Salmonella typhimurium TA 98 strain (TA98 His-) and eel isolated S9 liver fraction was used as a metabolic BaP activator. The BaP and BNF dose range concentrations tested were 0 (blank), 0.015, 0.08, 0.15, 0.38, 0.75, 1.5, 3.8, and 7.5 microM/plate and 0 (blank), 0.412, 1.235, 3.704, 11.1, 33.0, and 100 nM BNF, respectively. A dose-response relationship between BaP concentration and mutagenic activity was observed in the presence of S9 fractions in control and 0.3 microM BNF-exposed eels. Significant positive results, as TA98 His+ revertants, were observed at 0.38, 0.75, 1.5, 3.8, and 7.5 microM BaP/plate induced by BNF S9 fractions. Significant BaP mutagenic activation by liver control S9 was detected only at 1.5, 3.8, and 7.5 microM/plate. The BaP 1.5, 3.8, and 7.5 microM/plate mutagenic activation by BNF S9 and control S9 were not significantly different. Relative to BNF activation, it was only possible to detect His+ reversion at 11.1 nM BNF concentration with 0.3 microM BNF-induced S9. The above results demonstrate that the eel S9 liver fraction has the capacity to biotransform high BaP concentrations and convert it into a mutagenic compound with or without previous liver BNF biotransformation induction. The same does not apply to low BaP concentrations, where liver S9 induction by BNF is necessary to promote mutagenesis.