VAEIATION IN VIRULENCE 237 



Pock variants of herpes (Wildy, 1955) and myxoma viruses (Fenner and 

 Marshall, 1957) have also been described. In both cases strains producing 

 small pocks were usually less virulent for other hosts (mouse and rabbit) but 

 the correlation was not complete, as the Californian strains of myxoma virus, 

 though producing very small pocks, were highly virulent for rabbits. 



2. Adaptation of Influenza Virus to Mouse and Hamster Lung 



Many strains of influenza virus have been adapted by passage to produce 

 fatal pneumonia in mice. The ease with which this process occurs varies 

 greatly from strain to strain; some are apparently in the adapted state on 

 first isolation from man (Francis and Magill, 1937; Clampit and Gordon, 

 1937), some are adapted readily, some become adapted only if special 

 procedures are employed to lower the resistance of the mice (Jones, 1950), 

 and some have so far defied all attempts. There is some evidence that 

 adaptation occurs more rapidly if preceded by way of a few passages in 

 ferrets (Andrewes et al., 1935), and less readily if preceded by prolonged 

 passage in the allantois (Smith et al., 1951; Ledinko and Perry, 1955). 

 Because most strains of influenza virus were not, in their natural state, 

 adapted to any of the convenient laboratory hosts available in the 1930's, 

 a large volume of literature has sprung up on the mechanism of adaptation to 

 mouse (or hamster) lung and the nature of unadaptedness. The nature of the 

 tissue being infected is more complex than any so far considered. Perhaps 

 because of this the findings are, in places, singularly difficult to interpret. 



Once again, the process of virus variation is seen to operate by way of an 

 upsurgence of minority types. Selection of the majority type by passage at 

 limit dilution in the allantois, between each passage in lungs, prevents the 

 process of adaptation (Davenport, 1951). Also, as in the case of the O-D 

 change, adaptation to mouse lung is associated with changes in in vitro 

 behavior; thus, there may be a reduction in capacity to agglutinate fowl cells 

 (Hirst, 1947b; Friedewald and Hook, 1948), change in position in the fowl 

 cell receptor gradient (Ledinko, 1956), alteration in enzyme action on and 

 sensitivity of hemagglutinin to ovomucin (Ledinko, 1956), mouse lung 

 inhibitor (Davenport, 1952), and the ^-inhibitor of mouse and ox serum 

 (Chu, 1951; Smith et al, 1951; Brans et al, 1953; Briody et al, 1955) and 

 sheep salivary gland mucoid (Ledinko, 1955). Some of these changes in in 

 vitro behavior have been shown to apply to an increasing proportion of the 

 virus particles as adaptation progresses during passage; indeed, an association 

 can be demonstrated, at any stage in adaptation, between the in vitro 

 behavior of clones isolated from the lungs and their degree of virulence 

 (Ledinko, 1956). Just as an increase in virulence seems to be associated with 

 insensitivity to certain inhibitors, so the converse seems to be true; from a 

 mouse-adapted strain there was isolated, by chance, a variant which was 



