The antiviral effect of highly purified mouse interferon (IFN) was studied on influenza virus infection in mice, by changing the routes of administration. By repeated intravenous (iv.) administrations of IFN, no antiviral effect was observed. However, IFN was effective when the infected mice received either intraperitoneal (i.p.) or intranasal (in.) administrations. While all of the untreated mice died within 10 days after infection, 20% of the i.p. treated mice and 67% of the i.n. treated mice survived more than 30 days. To explain these different effects after administration by different routes, the distribution of given IFN in mouse organs was examined. Fifteen min after i.v. injection, only a low-titered IFN was detected in the serum, lung, spleen and liver of the mice and soon disappeared. However, a relatively high-titered IFN was detected 30 min after i.p, administration in the lung and serum. After i.n. administration, high-titered IFN was detected in the lung within 15 min after administration and continued for 90 min. Thus the findings demonstrated that the maintenance of a detectable amount of IFN in mice organs is dependent on the route of administration, and the higher the titer of IFN in the lung, the higher the protective effect of IFN against influenza virus infection, mouse interferon; anti-influenza effect; distribution Distribution and metabolism of IFN in animals were well reviewed by Ho and Armstrong (1975). The clearance of IFN given to animals and man was also documented by Cantell et al. (1974), Cantell andPyhala (1976) and other workers (Gresser et al. 1967;Nuwer et al. 1972;Pyhala and Cantell 1974;Edy et al. 1976). Their results may be summarized as follows : (i) In the blood, IFN has an extremely rapid half-life (about 20 min), hence it is difficult to maintain elevated serum concentrations. (ii) IFN in the blood slowly decreases after repeated i.v. injections of IFN. (iii) IFN activity has been detected in the blood for more than 5 hr after i.p. administration. However, the relationship between the effect of IFN on viral infections and the distribution of IFN in the organs of these animals has never been fully investigated.Only Finter (1967) described the fact that, although intramuscular (i.m.) or i.v. injections of IFN were not effective against aerosolized influenza virus infection in mice, levels of IFN sufficient to protect against Semilki forest virus infection were obtained from the blood after those injections. On the other hand, nose-dropped IFN was found to be effective against