THE MORPHOLOGICAL APPROACH 103 



XII. Discussion and Conclusions 



In summarizing the experimental data on virus infections no attempt will 

 be made to systematize contemporary knowledge of individual points but 

 rather to collate some of the ideas and trends which are reorienting and even 

 reorganizing research in this area. 



1. It is axiomatic that a morphological study of virus infections involves 

 the interaction of virus and cell. In this context the pathology of the cell may 

 be expressed as virus action, cell reaction, and subsequent interaction until 

 the cell overcomes or is destroyed by the virus. Given, then, the pathological 

 state, the first necessity is identification of the virus, and for this there is no 

 set of rules. Criteria for the identification of viruses in the electron microscope 

 have been tabulated and discussed elsewhere (Bang, 1955b). These criteria 

 have been generally fulfilled in the case of the poxviruses and perhaps in 

 several others. Viral antigen may now be identified within the cell by the 

 fluorescent antibody technique (Coons, 1957), and the application of this 

 technique will no doubt be greatly expanded. The determination of that part 

 of cell pathology which is a reaction to the virus is more difiicult. Similar 

 changes which are known to occur under various unfavorable conditions 

 have perhaps been taken too little into account. An excellent example is the 

 hypertrophied paranuclear area in infection with poliomyelitis, which is 

 simulated if not duplicated in degenerating mesenchyme and in a variety of 

 tumor cells. 



2. The pathology of the cell now implies, not only the reaction of the 

 entire cell, but also lesions of parts of the cell independent of apparent 

 simultaneous change elsewhere within it. The adenoviruses are identified as 

 masses of virus particles inside the nuclei of otherwise intact cells. The 

 mitochondi'ia in infections with poliomyelitis or Rous sarcoma virus remain 

 intact until late in the extensive cellular changes which entail the accumula- 

 tion of paranuclear material and, in poliomyelitis, the onset of nuclear de- 

 struction. Great growth of abnormal microvilli, which seem to have virus 

 within them, develop at the surface of a cell infected with influenza or New- 

 castle disease virus, and yet the mitochondria are normal even at high re- 

 solutions in the electron microscope. Thus, we are beginning to get away from 

 the term "inclusion," and we have avoided it here except when used in an 

 established framework. It would be more accurate today to speak of lesions 

 within the cell and to describe them as virus, virus by-products, or reactions 

 on the part of the cell to the virus. 



3. The ways in which a cell can react to a virus are probably limited, and 

 the same reactive processes may be induced under a variety of conditions. 

 The microvilli, which are normal extrusions from epithelial cells, seem to 

 contain and, in some cases, to release virus in vaccinia, influenza, Newcastle 

 disease, and frog adenocarcinoma. Even rickettsia mav be extruded from the 



