78 F. B. BANG 



to this. Morgan describes intracytoplasmic viral particles which contain dense 

 nucleus-like bodies, which are separated from the granular ground material 

 by a zone of lesser density. They were enclosed by a single membrane. Near 

 the surface of the host cell they have a double membrane. Gaylord and 

 Melnick (1953) clearly show particles with varying internal density which 

 sometimes look almost like bars. Although a number of hypothetical sequence 

 or life cycles of development for the individual virus units have been proposed 

 it would seem premature to review any of these before electron microscope 

 studies of the changes occurring within one cell cycle of multiplication are 

 available. Furthermore, the increasing recognition of the artifacts of pre- 

 paration, particularly of polymerization of the embedding material (Borysko, 

 1956; Watson, 1957) cautions against the danger of finding in nonpathological 

 appearances too easy confirmation of ideas dependent primarily on studies 

 of the infectious activity of the virus. 



3. Fluorescent Antibody Staining 



The progress of infection with vaccinia virus has been followed in cultured 

 cells by determining the increase of antigen by Coons's fluorescent antibody 

 technique (Noyes and Watson, 1955). This virus destroyed cells of a human 

 epidermoid carcinoma in 48 hours. Early in infection (9 hours) a small amount 

 of antigenic material accumulated near the nucleus. At 16-24 hours increasing 

 amounts in the same area were sometimes sharply delimited. Nuclear staining 

 was observed frequently in cells containing large amounts of antigen. This 

 observation, however, must be interpreted with caution, for tumor cells 

 frequently show nuclei notched with deep indentations of cytoplasm, and 

 viral antigen in this cytoplasm would appear by light microscopy to be within 

 the nucleus. 



Noyes and Watson (1955) suggest, because of the presence of antigen in 

 projections of cells, that virus transmission may take place by such cyto- 

 plasmic connections. This agrees with the several findings of particles at the 

 tips of microfibrils (Robinow, 1950) or microvilli (Bang, 1950). 



It would be profitable if the morphological data here reviewed could be 

 correlated with virological (infectious) measurements. In a recent study, 

 Overman and Tamm (1957) found that vaccinia virus in the chorioallantoic 

 membrane began to increase at 10 hours; that from that time until the forty- 

 eighth hour it increased logarithmically, and that it then continued to be 

 produced until the third day. Nearly all of it was bound to the cells and con- 

 tinued to be released from cells for as long as 7 days. Clearly, then, the time 

 period for infection of individual cells can currently be determined no more 

 precisely than it was morphologically by Bland and Robinow (1939), who 

 found a steady increase in inclusions and elementary bodies up to 24 hours. 



