GENETIC INTERACTIONS BETWEEN ANIMAL VIRUSES 391 



in line with the difficulty of "transferring" virulence in influenza virus 

 recombinations. 



As yet these results do little more than establish the fact that typical 

 recombination of characters can be observed in this group of viruses. There 

 is still need of a more controllable means of inducing double infection of cells 

 which will allow analysis of first-cycle yields without the complication of 

 selective factors which must be present in the overlapping area of two 

 dissimilar pocks. Fenner is hopeful that this may be found by the use of 

 HeLa cells in tissue culture. 



TABLE VI 



Twelve Recombinant Clones fkom Mixed Infections with Vaccinia 

 7N 00000 and RP (Rabbitpox) + + + + + 



B. Interactions among Intestinal Viruses 



The position in regard to the polioviruses is still far from clear and so far 

 only preliminary reports have appeared. There seems to be no doubt that 

 when two serotypes are grown together in tissue culture the harvest differs 

 from a simple mixture of the same serotypes grown separately. An unduly 

 large proportion of the plaques do not contain one of the normal serotypes. 

 Some contain both serotypes, others contain doubly neutralized virus 

 (Sprunt et at., 1955; Black and Melnick, 1956). Similar findings 

 have been reported for cultures from mixed infections by ECHOl and 

 polio 1 viruses by Benyesh et al. (1957) from Melnick's laboratory. The 

 significance of these results is far from clear, but there is a distinct resem- 

 blance to the influenza virus findings. A prima facie interpretation would be 

 that the standard virus particle is built up of a small number of subunits (as 

 suggested by Crick and Watson (1957) and Caspar (1957)), each perhaps 

 equivalent to a single genome. If these morphological subunits each carrying 



