GENETIC INTERACTIONS BETWEEN ANIMAL VIRUSES 279 



for which the limiting infective dose contained only a small number of mor- 

 phological units. Much more difficult problems might be presented if genetic 

 work were attempted with viruses which under optimal conditions showed 

 100 or more visible units per infective dose. 



B. Marker Characteristics 



In any study of genetic interaction, whether in viruses or in sexually 

 reproducing higher organisms, the geneticists' method is to take as parents 

 two varieties which (1) have sufficient common characteristics to allow the 

 production of viable progeny and (2) differ preferably in two or more easily 

 demonstrable qualities. The common characteristics can be taken for granted 

 — it is the redistribution of differences among the progeny which is important. 



An essential requirement for genetic work with viruses is therefore to have 

 available strains differing from one another in easily recognized qualities but 

 still close enough in other qualities to allow ready interaction. The available 

 genetic markers, i.e., those characters for which differences can be readily 

 shown, will differ from one group of viruses to another, and will be discussed 

 in some detail (p. 284) for each of the three groups that have been studied. 



At this point, it is only necessary to say something about three broad groups 

 of qualities that can be used as markers. 



1. Morphological 



In general, recombination is not likely to occur between viruses of different 

 morphology, but in the influenza viruses it can be shown (Burnet and Lind, 

 1957b) that the proportion of filamentous forms is an inheritable character of 

 a strain and that this can be used in genetic studies. 



2. Somatic 



These are qualities depending on the nature of the free surface of the infec- 

 tive virus particles. They are specially important in the hemagglutinating 

 viruses of the myxovirus group, simply because of the ease with which 

 appropriate manipulations can be devised to manifest differences in vitro. 

 A typical example is serological character, as demonstrated by hemagglutinin 

 inhibition. When one of these qualities is demonstrated in vitro, whether by 

 some modification of hemagglutination or by a complement fixation technique, 

 we are concerned with the character of a large population of virus particles ; 

 the result observed will not be influenced if, say, 5 % of the particles are 

 different in type from the majority. Only the nature of the dominant form 

 will be recognized. 



