BACTERIOPHAGE GENETICS 349 



quencies of r2 and rl heterozygotes are equal, one may expect to 

 find the // allele in 50 per cent of the mottled plaques if model I 

 is correct and in all of the mottled plaques if model II is correct. 

 In the case of model II, the frequency of the h allele will be less 

 than 100 per cent because recoinbination in the vegetative pool 

 will decrease the frequency of the parental phage types in the 

 pool from which the heterozygotes must be drawn. As the 

 vegetative pool approaches genetic equilibrium the frequency of 

 the h allele in the heterozygotes will approach 50 per cent. 

 The frequency of the h allele to be expected can be calculated 

 from the known linkage relationships and the observed frequency 

 of r+ recombinants which is a measure of the approach to equi- 

 librium. In the case of model I the frequency of the h allele in 

 the heterozygotes will be the same as the frequency in the total 

 population, will be unaffected by the amount of recombination 

 in the vegetative pool, and will be 50 per cent for equal inputs 

 of the two parental types. From the results of Levinthal's 

 experiment given in Table XVIII, it is apparent that the data 



TABLE XVIII 



The Results of Three-Factor Crosses Made to Determine the Structure of 



Heterozygotes. The Cross is hrl X r2" 



" Modified from C. Levin thai (1954). 



are in good agreement with the predictions for model II, but in 

 complete disagreement with the predictions for model I. The 

 genetic composition of the heterozygous particles changes with 



