288 F. M. BURNET 



the results in Table II were obtained. The great bulk of recombinants was 

 obtained from clones demonstrably heterozygous. 



In accord with this finding, when the primary harvest was titrated in the 

 presence of anti-M serum, only W -f- clones were obtained. This suggests that 

 heterozygotes with M components on their surface were neutralized and could 

 not give rise to W — progeny. 



The result differs from the fluid A mentioned above from Lind and Burnet's 

 (1957a) paper. This gave virtually a full yield of the WS — recombinant when 

 titrated in anti-MEL serum. The significance of these differences between the 

 two sets of results are discussed later. 



E. Use of Nonviable Virus as an Effective "Parent" 



One of the most interesting features of genetic work with influenza viruses 

 is the comparative ease with which recombinants can be obtained from 

 inoculation, either as mixtures or in sequence, of one component killed by 

 heat or ultraviolet irradation and the second in the form of active virus. This 

 was reported for heat-killed virus by Burnet and Lind (1954a) and confirmed 

 for ultraviolet-killed virus by Baron and Jensen (1955) and for both heat- and 

 ultraviolet-killed virus by Hirst and Gotlieb (1956). In recent work (Lind and 

 Burnet, 1957c), we have made a closer analysis of virus, heat-inactivated at 

 various temperatures. The best yield of recombinants was obtained from 

 fluids inactivated by slow thermal degradation at 37°C. for 20 days. 



Nothing has been published to suggest that the type of recombinant 

 obtained from such nonviable/viable crosses differs significantly from those 

 from normal crosses. Burnet and Lind (1954a) found a considerably lower 

 yield, but the types of virus obtained were of normal quality. Hirst and 

 Gotlieb (1956) were more successful in obtaining M + with direct mouse 

 pathogenicity from W + /M — crosses when the M — component was 

 inactivated by ultraviolet light than when active virus was used. Baron and 

 Jensen (1955) do not give data for comparable active/active crosses, but there 

 is nothing to suggest that they would not have obtained similar recombinants. 



