128 ALICK ISAACS 



influenza virus when large doses of RDE were present in the medium. 

 Hemagglutinin appeared at the same time but soluble antigen was not 

 studied in this system. Hoyle first thought that the increase in soluble 

 antigen in the cells occurred at a logarithmic rate but later (Hoyle, 1953) he 

 agreed that a linear increase fitted the experimental observations better. 



2. Hoyle (1950) interpreted his findings on the effect of ether treatment of 

 influenza virus particles as showing that when the particles were disrupted 

 they liberated soluble antigen which had been enclosed within the intact 

 particle. Fulton (1953) interpreted the same findings as indicating that ether 

 treatment of the virus had degraded the viral antigens and blunted their 

 serological specificity in an analogous way to the effect shown by ether 

 treatment of rickettsiae (Fulton and Begg, 1946). These differences in 

 interpretation emphasize the difficulty of identifying soluble antigen derived 

 from infected tissues with the antigen liberated by ether treatment of virus 

 elementary bodies. In favor of Hoyle's interpretation are the findings of Lief 

 and Henle (1956b) that with a standardized technique of ether extraction, a 

 constant amount of soluble antigen was liberated from a given dose of 

 influenza virus; in addition, less soluble antigen was liberated from incomplete 

 virus than from standard virus. This finding is further discussed in Section 

 VI, A, but the implication is that the amount of soluble antigen which can be 

 liberated is closely bound up with the infectivity of the virus, a finding which 

 would not be expected on Fulton's hypothesis. 



3. Evidence has been produced that the soluble antigen is nucleoprotein or 

 is closely bound to nucleoprotein. Hoyle (1952) showed that the serological 

 activity of the soluble antigen was greatly reduced by treatment with trypsin, 

 but only slightly reduced by prolonged incubation with ribonuclease, although 

 the ribonuclease could have been contaminated with small amounts of 

 protease. Hoyle also found that the soluble antigen was precipitated by 

 lanthanum acetate. 



More convincing evidence for the presence of ribonucleic acid in the 

 influenza soluble antigen was provided by Ada and Perry (1954). They 

 prepared extracts by differential centrifugation of infected chick embryo 

 lungs from which the soluble antigen was precipitated by addition of 

 immune mouse antiserum. Such precipitates were fractionated by a modified 

 Schmidt-Thannhauser method and were found to contain roughly 4 % of 

 RNA and 0.5 % DNA. Later, Ada (1957) prepared influenza soluble antigen 

 from infected chorioallantoic membranes and compared it with virus soluble 

 antigens prepared by ether extraction of the virus. The two antigens were 

 found to have comparable complement-fixing activities per milligram of dry 

 weight, but while both showed the presence of ribonucleic acid, the total 

 amount of acid and the proportions of nucleotides differed greatly in the two 

 preparations. On the other hand, Schafer (1957) foxmd that fowl plague virus 



