Nonheme iron is synergistic with 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in producing hepatotoxicity in mice. Fe2+ rather than Fe3+ is the probable toxin and we speculated that TCDD, an inducer of microsomal electron transport, might favour reduction of iron. We have defined a system which will release Fe2+ from ferritin (Fe3+) under anaerobic conditions and in the presence of added flavin mononucleotide (FMN). The rate of reduction of ferritin iron was proportional (a) to microsomal protein from 0.5 to >3 mg/mL, (b) to the activity of NADPH–cytochrome c reductase over 0.1 U/mL, (c) to ferritin at concentrations exceeding iron concentrations >200 μmol/L, and (d) to the concentration of FMN when it was less than 125 μmol/L. The system was approximately twice as active with NADPH as with NADH as electron donor. The linear phase of iron release did not commence immediately, but followed a delay (±0.5 min) after adding FMN to an anaerobic mixture containing microsomes, ferritin, an NADPH-generating system, and an oxygen-scavenging system. When microsomes from untreated, Phenobarbitaltreated (3 days), or TCDD-treated (1 or 3 weeks) rats were compared, iron release correlated most closely with the cytochrome P-450 concentration. However, when the microsomal proteins were solubilized and the NADPH–cytochrome c reductase and cytochrome P-450 activities were separated, reduction of ferritin iron was shown to be a function only of the reductase fraction, except that the delay in initiating release of Fe2+ was increased with purified reductase and decreased when a monooxygenase system was reconstituted with cytochrome (phenobarbital or TCDD induced) and lipid. These studies have defined a potentially important hepatic microsomal system able to release Fe2+ from ferritin iron, but have failed to indicate any feature unique to the dioxin-induced monooxygenase system.