THE MORPHOLOGICAL APPEOACH 93 



few hours, therefore, changes in the nucleus and cytoplasm appeared almost 

 simultaneously. In comparing these findings with the data on fixed and 

 stained cells, it is important to remember that only extremely careful cyto- 

 logical preparation would preserve intact the irregular peripheral spread of 

 endoplasm so that it could be identified in fixed and stained material. 



Barski et al. (1955) recorded the changes produced in large fibroblasts 

 derived from tonsils in motion pictures taken by phase microscopy and 

 described the development of large eosinophilic paranuclear masses. These 

 occurred with both types 1 and 2 in cultures kept at 30 to 31°C. for 20 hours 

 after the addition of a virus. The cell remained motile, and filamentous mito- 

 chondria persisted. Later, a series of bubbling extrusions developed, and the 

 authors suggest that this is the mechanism of virus release. Very similar, if 

 not identical, paranuclear lesions have been described in anterior horn cells 

 by Bodian (1948); it is difficult to differentiate this lesion from the giant 

 centrosphere described by Lewis (1920) in degenerating mesenchyme and in 

 tumor cells. It may represent an abnormal accumulation of normal cellular 

 material, rather than a specific virus "inclusion." The material has been 

 shown to be Feulgen-negative (Harding et al., 1956). 



2. Relation to Virus Release 



The exact relation of these cell changes to the time of virus release has not 

 been settled. In studies on isolated individual cells, Lwoff et al. (1955) showed 

 that about one hour before the beginning of virus release into the medium 

 the cell started to contract, and a hyaline zone developed at the periphery 

 of the cell. At the time of virus release this zone became vacuolized and then 

 disintegrated, by which time virus release ceased. 



Riessig et al. (1956) have followed the changes in cultures of monkey 

 kidney cells during one phase of virus multiplication. Most of the preparations 

 utilized hematoxylin, eosin, and Feulgen stains. Osmium fixation was not 

 used. Some of the living cells were studied by phase microscopy. Although 

 gross changes were not apparent in the infected test tubes until 8 hours after 

 infection, the stained preparations showed a patchy disappearance of the 

 chromatin network of the nucleus and development of an eosinophilic cyto- 

 plasmic mass by 5 hours, with the persistence of the nucleolus. The formation 

 of intracellular virus began at about 4 hours, and extracellular virus increased 

 from 5 hours on. The authors suggest that the changes in the appearance of 

 the cell seen with phase microscopy do not begin until 1 or 2 hours after 

 intracellular virus increases. 



In studies on monkey kidney, HeLa, and human fetal cells, Dunnebacke 

 (1956a,b) found that virus release occurred several hours after nuclear 

 pyknosis, contraction of cytoplasm, and rounding of the cell. In contrast, 

 human amniotic cells released virus much later, and the lesion in these cells 



