Dose-response relationships for administration of suspensions of different dosages of ferric oxide and a constant quantity of benzo[α]pyrene (BP) given intratracheally to Syrian golden hamsters were examined in this model system for respiratory carcinogenesis. Each group received 10 instillations at 2-week intervals. The amounts of ferric oxide were 3, 6, and 9 mg per dose, all with 3 mg of BP. As previously noted in this animal model, squamous cell carcinomas were the predominant morphologic type of induced tumors, and the bronchi were the most frequently affected. Essentially identical results were observed for the 3 experimental groups. Doses of ferric oxide beyond a 1:1 ratio with BP did not enhance tumor yield. These results demonstrated that beyond a certain critical ratio of BP to ferric oxide, the proportion and quantity of ferric oxide administered were unrelated to the incidence of induced tumors.