BIOLOGICAL ASPECTS OF INTRACELLULAR STAGES OF VIRUS GROWTH 129 



soluble antigens, prepared in the same way from infected tissues and from 

 ether extracts of the virus, were very similar to one another in their behavior 

 with antisera, nucleic acid content, and ultraviolet absorption spectrum. 



4. Wiener and associates (1946) showed that soluble antigen could be 

 liberated from influenza virus by ultrasonic vibration and this was confirmed 

 by Lief and Henle (1956a). However, Lief and Henle found that after the 

 ultrasonic treatment there was always more antigen which could be extracted 

 subsequently with ether, and they interpreted this to mean that some 

 soluble antigen was loosely adsorbed to the virus surface and some was 

 present in the virus particles themselves. 



In summary, while the evidence is incomplete, it seems likely that the 

 soluble antigen is a nucleoprotein which is produced in infected cells before 

 mature virus appears, and that some of it becomes incorporated in the virus 

 particles. While this statement may be an oversimplification of the events 

 which occur and leaves many points unexplained, it suggests further useful 

 lines of investigation. 



B. The Hemagglutinin of Myxoviruses 



In parallel with studies of the appearance of influenza soluble antigen in 

 infected cells, the development of viral hemagglutinin (or V antigen) has been 

 intensively investigated. However, hemagglutination may be caused by the 

 virus elementary bodies, by particles which are either smaller or less dense 

 than the elementary bodies (see below), or by small fragments produced, e.g. 

 by ether treatment of elementary bodies — all these particles showing similar 

 serological specificity. Unfortunately, in most studies the particle size of the 

 hemagglutinin is not mentioned. This means that a great deal of useful 

 information on the development of viral hemagglutinin in infected cells 

 cannot yet be applied to the problem of which antigens might qualify as 

 possible viral precursors. 



At this point it is necessary to anticipate some of the discussion in Section 

 VI by considering the concept of incomplete viruses. Von Magnus (1946) 

 showed that on serial passage of large inocula of influenza virus in the allantoic 

 cavity incomplete virus was liberated into the allantoic fluid. The incomplete 

 virus showed normal hemagglutinating and serological activity, but had a 

 very low infectivity relative to that of virus similarly passaged at high dilution. 

 There is still controversy as to whether incomplete virus represents virus 

 which was prematurely liberated from the cells before it had acquired full 

 infectivity, i.e., its normal development was interrupted; or whether it 

 should be considered as malformed virus produced by overloaded cells. In 

 the present discussion the term incomplete virus will be used strictly for the 

 virus spontaneously liberated from cells when the experimental technique of 

 von Magnus is applied. This limitation is necessary because there are two 



vol. in — 9 



