276 F. M. BURNET 



merely as the potentiality of the infected cell subsequently to give rise to a 

 brood of descendant virus particles. There is an increasing volume of evidence 

 mostly circumstantial, that the essential feature of this stage is the release of 

 free nucleic acid from the virus particle and that the "information" needed 

 for the production of descendant virus particles is carried at this stage wholly 

 in the liberated nucleic acid. The inhibition of influenza virus multiplication 

 by ribonuclease (Le Clerc, 1956; Burnet et ah, 1957) can probably be taken as 

 evidence of such a phase. Recent findings that infective nucleic acid could be 

 extracted from tissues infected with the viruses Mengo (Colter et ah, 1947), 

 eastern equine encephalitis (EEE) (Wecker and Schafer, 1957), and Murray 

 Valley encephalitis (MVE) (Ada and Anderson, 1957) provide strong support. 

 Under such circumstances, it should be expected that if simultaneous 

 infection of a cell by two related but distinguishable viruses can be effected, 

 genetic interactions of one type or another will be likely to occur. This could 

 be recognized by the isolation from the progeny of clones of virus manifesting 

 a mosaic of characters, some corresponding to one parent, others to the second 

 parent. Recent work indicates that such findings are by no means uncommon 

 (Burnet and Lind, 1949). In this chapter, those aspects of the genetics of 

 animal viruses which involve the appearance from double infections of 

 progeny with characters derived from both parents will be considered. It 

 will also be legitimate and convenient to include under genetic interactions 

 those instances where a genetic character is, or appears to be, contributed by 

 inactivated virus or some virus component, and to touch on the possibilities 

 of multiplicity reactivation where the interaction of two or more "killed" 

 virus particles allows the appearance of viable progeny. 



II. The Different Types of Genetic Interaction 



We may define our subject matter, then, as the analysis of those phenomena 

 by which virus particles are produced which have observable characters 

 derived from more than one parent form. The most important type of genetic 

 interaction is that by which progeny can be isolated as a true-breeding clone 

 with inheritable qualities derived from two (or more) parent forms. This will 

 be referred to as true recombination. To date, the best examples have been 

 observed with influenza viruses. 



True recombinants have been obtained in which one of the parent strains 

 has been inactivated as far as conventional tests are concerned by heat or 

 ultraviolet light. It is therefore by no means inconceivable that agents 

 equivalent to the transforming factors of bacterial genetics may be shown to 

 exist. The Berry-Dedrick transformation of fibroma to myxoma is presumably 

 of this character but has not yet been adequately analyzed. 



Evidence has been provided by Gotlieb and Hirst (1954) that (presumably) 

 unstable heterozygotes can be obtained. A heterozygote can be defined as a 



