280 P. M. BURNET 



3. Reproductive 



Somatic characteristics may remain unaltered in virus rendered nonviable 

 by appropriate means. By reproductive characters we include the various 

 manifestations of virulence — capacity to multiply in a given tissue and 

 capacity to produce visible lesions or death. With some qualifications we may 

 say that these require only the presence of a minimal infective dose in 

 the inoculum to manifest the character in question, and that the presence 

 of an excess of virus lacking the character will not, in general, prevent its 

 expression. 



Two qualities which may be needed as markers do not fall clearly into the 

 reproductive group but must be considered along with them. These are 

 "toxicity," in which lesions are produced without virus multiplication, and 

 serological tests by neutralization techniques in vivo. 



4. Incidental 



For want of a better term this is used to signify those marker characters 

 which depend on the production during the course of infection of a recogniz- 

 able substance other than infective virus. The significant examples are 

 vaccinia hemagglutinin and the "soluble" complement-fixing antigens 

 produced by many viruses. 



C. Principles of Recombination 'Experiments 



Work on the genetics of animal viruses has necessarily been largely 

 influenced by the methods which have been used with bacterial viruses. In 

 both fields the ideal to be aimed at is to infect a population of uniformly 

 susceptible cells in such a fashion as to ensure that the largest practicable 

 fraction of cells are infected by at least one particle of each parental type. 

 Any unabsorbed virus should be removed so that when the new brood of 

 virus is liberated it should, as far as possible, be derived wholly from cells 

 infected by the two parental viruses. The progeny should be harvested before 

 there is any opportunity for a second cycle of infection to take place. 



With influenza viruses, the allantoic cavity provides highly susceptible 

 cells readily accessible to virus. These cells have been used for all work on 

 recombination of influenza viruses, but other methods will obviously be 

 required for other groups. 



Fenner's finding that a high degree of recombination occurs where two 

 developing pocks on the chorioallantois find newly proliferated cells in common 

 may be a particularly satisfactory method when the main objective is to 

 determine the qualitative range of recombinants that can be produced. With 

 other types of virus, tissue culture methods will undoubtedly be the most 

 satisfactory and there are obvious advantages for quantitative work in 



