476 W, SCHAFEIl 



This chapter will give a critical survey of our present knowledge concerning 

 the chemical composition of the different virus-specific units, and will sum- 

 marize briefly the origin and function of the chemical components. Only 

 those animal viruses that can be regarded with certainty as true viruses will 

 be discussed; therefore, the agents of the psittacosis-lymphogranidoma group 

 and the rickettsiae are not mcluded. 



II. Definition of the Various Virus-Specific Units 



The most detailed chemical analyses were generally made on the infective 

 particles — the virus-specific units that are fully capable of reproducing new 

 copies in the cells. 



Other substances have been fomid which are noninfectious and demon- 

 strable as virus-speciiic products only by seroimmunological reactions. They 

 were first demonstrated with vaccinia (Craigie, 1932), and later in many other 

 virus infections (see Smadel, 1952a). Since they are generally smaller than 

 the corresponding infective particles, or at least sediment slower by ultra- 

 centrifugation, they have been designated as solvble antigen (S antigen), 

 although this designation is not very meaningful. It has been noted in many 

 cases that these S antigens are less specific than the superficial antigens of 

 the infective particles. 



The hemagglutinivis, found with viruses of the pox group (Nagler, 1942; 

 see Burnet, 1955a), are also noninfectious. Like the S antigens, they sediment 

 slower than the infective particles and possess a virus-specific antigen, but 

 they are mainly characterized by their ability to agglutinate red blood cells. 



Of special interest are those hemagglutmating units which accompany 

 the myxoviruses, and are usually designated as incomplete forms (von 

 Magnus, 1947; see 1954). In most cases these can be distmguished physic- 

 ally from the infective particles by their lower sedimentation velocity. In 

 comparison with infective particles, the ratio of infectivity to hemagglutina- 

 tion (ID50/HA) is smaller in preparations of incomplete forms. It has not 

 been determined whether the infectivity, which has always been demon- 

 strated in such preparations, is due merely to the presence of some infective 

 particles or whether the incomplete particles themselves have a low, but 

 constant, capacity to initiate infection. 



One can regularly find incomplete forms in extracts of c hick chorioallantoic 

 membranes infected with influenza, fowl plague, or Newcastle disease virus, 

 regardless of the virus dose used for infection (GranoiT et al., 1950; Granoff, 

 1955; Henle et al, 1956; Schafer and Munk, 1952b; Schafer et al, 1954). 

 Only in the case of mfluenza could appreciable amounts of incomplete 

 particles be detected in the supernatant fluids, e.g., allantoic fluid. According 

 to Henle and co-workers (1956), incomplete forms are found outside when the 



