226 F. FENNER AND J. CAIRNS 



infections produces characteristic lesions on the chorioallantoic membrane or 

 in the rabbit skin, and neither the nature of the lesions nor the "efficiency of 

 plating" for the passage host changes with serial passage in either of these 

 hosts. In other cases minimal signs of infection are observed on first introduc- 

 tion of the virus into a new host, yet after serial passage, sometimes prolonged, 

 a lethal infection is regularly produced, e.g., the adaptation of polio virus and 

 dengue virus to mice (Armstrong, 1939; Sabin and Schlesinger, 1945). 

 During serial passage a process of "adaptation" of the virus to the new host is 

 said to have occurred. 



Once a virus has been adapted to an experimental host, the attention of 

 many workers has turned to a frequent by-product of adaptation, namely, 

 the coincident attenuation of the virus for its original host. Thus, after his 

 experiences with the attenuation and exaltation of virulence of bacteria 

 by passage in different animal hosts Pasteur turned his attention to rabies 

 (Pasteur el al., 1884) and found that by serial passage of this virus in rabbit 

 brain it changes from "street" virus to "fixed" virus, and showed enhanced 

 virulence for the rabbit, and reduced (but still high) virulence for the dog and 

 man. The best-known examples of the production of an attenuated virus 

 vaccine by adaptation to a new host tissue are the yellow fever vaccines. The 

 "French neurotropic" strain was obtained by serial passage in the mouse 

 brain (Peltier et al., 1940) and the 17D vaccine during prolonged passage in 

 tissue culture (Theiler and Smith, 1937). 



In consequence of the practical requirements just outlined, there has grown 

 up an enormous volume of literature concerned with the growth of animal 

 viruses in different laboratory hosts — involving primary adaptations, secon- 

 dary adaptations from one laboratory host to another or one organ or tissue 

 of a particular host to another, and studies of the effect of such adaptations 

 on the pathogenic capacity of the virus in its original host. Because of the 

 complexity of the experimental material, studies on the adaptation of animal 

 viruses to new hosts have so far yielded little basic information, but have 

 given rise to a great volume of superficial observations. These, although 

 invaluable at the practical level, reveal virtually nothing of the mechanism 

 of adaptation. 



For most people, maximum virulence of a virus, as a character, indicates 

 that the strain possesses the maximum destructive power which can be 

 associated with the multiplication of a virus of that sort. With bacterial 

 viruses, where the host is a unicellular microorganism, virulence implies cell 

 lysis. The animal host, however, consists of a vast number of individual cells 

 differing widely in many properties. Some may be suited and others unsuited 

 to support virus multiplication, some are of major and others of minor 

 importance in the economy of the host. In animals, therefore, there is no such 

 simple dichotomy and no general agreement as to how the word virulent can 



