The role of inactivated influenza virus in the von Magnus phenomenon has been studied by exposing standard virus preparations in vitro to 37°C. for periods up to 6 or more days. The rate of inactivation of the infectious property of the line of PR8 virus employed was found to be approximately 1.1 log10 unit per day, denoting a half-life of 6½ hours. The rate of inactivation was similar in the allantoic cavity of chick embryos.
On allantoic passage of such heated seeds without dilution it was seen that with a decrease in the infectivity of the inocula proportionately less infectious virus was found in the harvests. The yields of hemagglutinins were much less affected, and thus ID50/HA ratios of as low as 101.0 were observed in the 24 hour harvests. The ratios of the yields always equalled or were higher than those of the inocula. On 100- or 1000-fold dilution of the heated seeds standard virus was obtained.
Growth curves in intact chick embryos or in deembryonated eggs (differential yields) revealed that non-infectious hemagglutinins appeared in the tissues or were liberated therefrom as soon as any virus activity became detectable. Furthermore, once maximal liberation had been established, infectious virus and non-infectious hemagglutinins were released for extended periods of time at nearly constant rates and in unchanging proportions, the latter depending upon the seed employed.
Heated standard virus and undiluted passage seeds (von Magnus), selected on the basis of similar ID50 and HA concentrations, failed to yield similar results in differential growth curves in deembryonated eggs. Although the hemagglutinin titers in the 2-hourly harvests were nearly identical, the undiluted passage seeds produced as little as 1 per cent of the infectious virus which was derived from the heated inocula. Thus considerable differences exist between the 2 types of seeds.