ANIMAL VIRUSES 477 



host cells initially adsorbed several virus particles within a short time 

 interval. The adsorbed particles can be either fully infectious or partly non- 

 infectious. Allantoic fluid of very low infectivity to hemagglutination ratio 

 can be obtained by serial passages of midiluted infectious allantoic fluid (von 

 Magnus, 1947). Units that resemble incomplete forms have also been obtained 

 when mice are inoculated intracerebrally with non-neurotropic influenza 

 strains (Schlesinger, 1950), and by the infection of HeLa cells with influenza 

 (Henle et al, 1955). 



The incomplete forms produced under the various conditions are not 

 identical. From studies of Granoff (1955), the influenza incomplete forms, 

 obtained by undiluted passages, differ in both biological and physical pro- 

 perties from those found in the allantoic membrane after infection with small 

 doses of virus. Furthermore, two types of influenza incomplete particles have 

 been observed in the electron microscope. Those obtained from allantoic 

 fluid after serial undiluted passage have nearly the same appearance as the 

 infective particle (see von Magnus, 1954); whereas flat, membrane-like 

 structures, with a rough surface architecture and average diameters 

 larger than those of the infective particles (Werner and Schlesinger, 

 1954), were isolated from extracts of some infected cells. The latter type 

 is very similar to the incomplete forms of fowl plague (Schafer et al., 

 1954). 



In all likelihood the various particles now designated as incomplete forms 

 will have to be reclassified after further investigations. Until such classifica- 

 tion is available, one should always indicate how the 'incomplete forms' in 

 question were obtained. 



Filamentous forms, also found w^ith several representatives of the myxo- 

 virus group (influenza, fowl plague, and Newcastle disease) (Mosley and 

 Wyckoff, 1946; see Schlesinger, 1953), are so large that they can be observed 

 in the light microscope. They can attaui lengths up to several microns and 

 have a width of 60 to 80 m/.t. It has been suggested occasionally that they are 

 subdivided into spherical-shaped particles, but more frequently one observes 

 a single spherical particle at the end of a filament. It has not yet been de- 

 finitely decided whether the filamentous forms are infectious, since they have 

 not been isolated. They are able to hemagglutinate and they possess virus- 

 specific antigen (Chu et al., 1949; Bang and Isaacs, 1957). Since the ultrasonic 

 disruption of these structures tends to increase the hemagglutinating activity 

 (Donald and Isaacs, 1954a), one can assume that they contain a larger 

 number of hemagglutinating subunits. 



There is no clear understanding of the function of the S antigens, the 

 hemagglutmins of the pox viruses, or of the incomplete forms of the myxo- 

 viruses. Three hypotheses have been formulated. The first claims that they 

 are necessary stages in the synthesis of new virus particles. Second, they 



