Ferrous ions were highly lethal and mutagenic to germinated conidia of N. crassa. At comparable survival, treatment with 0.2 mM ferrous ions was 14- and 50-fold more mutagenic than UV irradiation or X-rays, respectively, in the reversion of an inositol auxotroph. Ascorbic acid alone (2 mM) was not reproducibly lethal and inhibited both the lethality and mutagenicity of ferrous ions. Bovine superoxide dismutase (SOD) completely inhibited the residual lethality of ferrous ascorbate. Protection by ascorbic acid and SOD indicates that superoxide radicals, generated by oxidation of Fe(II), are directly or indirectly mutagenic and lethal. Malondialdehyde (MDA) was lethal and appeared to be mutagenic; its action is probably different from that of superoxide. Therefore, superoxide-mediated production of endogenous MDA by way of peroxidation of polyunsaturated fatty acids is probably not an alternate mutagenic pathway, at least in the reversion of the allele of the inl locus examined. These results and the demonstration of superoxide-mediated decrease in the synthetic fidelity of DNA polymerase in vitro (Rana and Munkres, in preparation) warrant additional exploration of the hypothesis that endogenous cellular free radicals, generated by pre- and post-senescent metabolism, may enter into lethal and mutagenic reactions.