RECOMBINATION IN BACTERIOPHAGE 53 



The variations shown in fig. 6 are evidently due in part to variations in 

 relative numbers of the two kinds of virus adsorbed, and in part to variations 

 in viral growth. For purposes of comparison, the variation in relative multi- 

 plicity has been computed on the assumption of a random distribution of two 

 types of virus over a population of bacteria receiving on the average five 

 particles of each kind. The distribution for this case is approximately that 

 shown in table 3. 



Table 3 

 Ideal Dislribiition of Multiplicities in Mixed Infection 



The proportions of l^acteria falling into the specified groups classified with respect to relative 

 multiplicity of infection with two kinds of virus have been calculated for random adsorption with 

 average multiplicity of five each of the two kinds. 



NUMBERS OF FIRST KIND 

 AS PERCENT OF TOTAL VIRUS PERCENT OF BACTERIA 



ADSORBED 



or 100 

 1-9 or 91-99 

 10-18 or 82-90 

 19-27 or 73-81 

 28-36 or 64-72 

 37-45, 46-54, or 55-63 

 total 



The distributions of yields actually found (fig. 6) differ from the theoretical 

 distribution of multiplicities in showing a considerably broader spread, and a 

 significant excess of yields containing only one kind of virus. These effects 

 could be due in part to an inhomogeneity of the bacteria with respect to ad- 

 sorbing power for virus; otherwise they suggest that a bacterium infected with 

 two viruses is somewhat less likely to liberate a given one, than a bacterium 

 infected with that one alone. Dulbecco (1949) has shown that the latter is, 

 in fact, the case. 



Another possible contribution to the variations described is connected with 

 the relatively long period (five minutes) allowed for adsorption of virus, which 

 permits some bacteria to be infected with one or more particles of one type of 

 virus considerably in advance of infection by the second. Owing to the slow 

 adsorption of the virus T2H and its mutants, the adsorption time cannot be 

 much reduced without reducing the total multiplicity of infection, or introduc- 

 ing excessive amounts of virus. This contribution to the variation in composi- 

 tion of viral yields has not, therefore, been assessed. 



COMPETITION BETW'EEN VIRAL MUTANTS 



The competition between viral mutants expresses itself in two ways; first, 

 by the complete suppression of one virus or the other in mixedly infected bac- 

 teria, and second, by excessive growth of one of the two types in bacteria liber- 

 ating both. These effects are slight among mutants of T2H so far examined, 



160 



