356 BACTERIOPHAGES 



c. The Unit of Recombination 



On the assumption that the Watson and Crick (1953) model 

 of DNA is a reasonable approximation to the structure of the 

 phage genetic material one can make certain inferences. One 

 might expect that in reference to the phenotypic expression of 

 chemical changes in the genetic material, there would be con- 

 siderable variation in different regions within one functional 

 unit. If the functional unit controlled the synthesis of an en- 

 zyme, a change in a single nucleotide pair in a sensitive region 

 might render the derived enzyme inactive, and extensive change 

 in another region might have no effect on the catalytic properties 

 of the enzyme but might alter its antigenic specificity. Also 

 it would seem probable that genetic recombination might occur 

 at the linkage between any two adjacent nucleotides in the 

 genetic structure. Therefore the minimum unit of mutation 

 and the minimum unit of recombination might involve the same 

 chemical structure, the single nucleotide pair of the Watson- 

 Crick model. Benzer (1957) proposed the term recon for the 

 "smallest element in the one-dimensional array that is inter- 

 changeable (but not divisible) by genetic recombination." 

 The size of the recon is of the same magnitude as the size of the 

 muton in Benzer's experiments and it is probable that more 

 refined data will show that both units are referable to the 

 nucleotide pair from which the Watson-Crick model is con- 

 structed. 



By a special technique Benzer (1957) was able to isolate 123 

 mutants that were not resolvable by recombination and pre- 

 sumably resulted from changes at the same point. These 123 

 mutant strains could be classified into three categories on the 

 basis of frequency of reverse mutations. One group of 72 strains 

 had reversion frequencies of the order of 0.2 X 10~^ to 4.0 

 X 10~^, a second group of 50 strains had reversion frequencies 

 of the order of 0.3 X lO"'' to 4.0 X 10"^ and 1 strain had a 

 reversion frequency of about 10~*^. If a point mutation involves 

 a substitution of a single nucleotide pair by another these dif- 

 ferences in reversion frequency may be a result of the substitu- 



