BIOLOGICAL ASPECTS OF INTRACELLULAR STAGES OF VIRUS GROWTH 131 



hemagglutinin remains intracellular, as in the case of the "S" hemagglutinin. 

 Also, the morphological appearances described by Werner and Schlesinger 

 (1954) for influenza hemagglutinin from mouse brain are very reminiscent of 

 the tissue forms described by Schafer for fowl plague; in unpublished observa- 

 tions made in collaboration with R. C. Valentine, it was noted that viral 

 hemagglutinin from infected HeLa cells showed the same morphological 

 appearance as the forms described by Schafer (1957). On the other hand, 

 incomplete virus obtained by the method of von Magnus and present in 

 allantoic fluid appears on electron microscopy (Donald and Isaacs, 1954c; 

 Pye et al., 1956) to be much less pleomorphic than the tissue form of the 

 hemagglutinin, and von Magnus (1954) states that incomplete and standard 

 virus do not show any pronounced differences in size and shape on electron 

 microscopy. It is tempting therefore, to suggest that in the cells of these three 

 sites there is a lack of some factor required for completing the virus multiplica- 

 tion cycle and that viral building blocks accumulate in the cells. There is, 

 however, no firm experimental evidence which would let us identify any viral- 

 associated antigen as a virus precursor. 



Henle et al. (1956) studied the ratio of infectivity : hemagglutinin titer 

 (I /HA ratio) in the membranes and fluid media of de-embryonated eggs 

 infected with influenza virus. With seeds of different varieties, i.e., diluted or 

 undiluted standard seeds and seeds prepared by serial passage of large 

 inocula to produce incomplete virus (see Section VI), it was regularly found 

 that the I /HA ratio hi the membranes was about 1.5 log less than that of the 

 liberated progeny. In any 2-hour liberation period about ten times the amount 

 of infective virus was shed into the medium as was present in the membrane, 

 whereas only about one-quarter of the hemagglutinin was released in the 

 same time. In addition, V or viral antigen, measured by complement fixation 

 and distinct from the viral hemagglutinin, could be shown to be present in the 

 tissues but not in the fluid in any 2-hour period. Henle et al. also found that on 

 adding potassium cyanide the infectivity and hemagglutinin titers in the 

 membranes decreased sharply and rose again, together, as the cyanide was 

 removed. They concluded that the hemagglutinin in the membrane was 

 noninfective at first and acquired infectivity later in its development, the 

 production of noninfective hemagglutinin and conversion to mature virus 

 forming a dynamic process. The noninfective hemagglutinin in these experi- 

 ments may be the same as the "S" hemagglutinin of Granoff et al. and it 

 seems important to investigate this possibility. 



In many studies, the times of first appearance of hemagglutinin, soluble 

 antigen, and infective virus have been compared. As discussed earlier, it is 

 difficult to compare strictly the time of appearance of mature virus with that 

 of the other two antigens since they are not normally measured under 

 identical conditions, although it appears that infective virus develops after 



