288 C. LEVINTHAL 



Other hand, if they are crossed to the wild type, some of the progeny then 

 behave in further crosses as though they contained only a single r mutation. 

 In general, a phage with only a single r mutation will give wild-type re- 

 combinants when crossed to one of the original r stocks but not with the 

 other, whereas a phage containing the double r will not give such recom- 

 binants when crossed with either. 



In a cross ah X a^h^ the number of recombinants of the type ob^ is gener- 

 ally equal to the number of recombinants of the type a+6. This conclusion 

 is not dependent on the nature of the mutations, i.e., the symbols a, 6, etc., 

 can represent any mutations which arise independently. In addition to the 

 equal numbers of the reciprocals in such a cross, one also finds the same total 

 number of recombinants if the two mutations are in one of the parents or if 

 each mutation is introduced into the cross in a different parent. Thus, the 

 last two crosses in Table I would always yield the same number of recom- 



TABLE I 

 Various Crosses and Their Recombinant Types 



binants. These equahties do not, however, reflect any particular property of 

 the elementary mating act. They indicate only that none of the phage types 

 has any growth advantage relative to the others and that the nature of the 

 mutants does not affect the probabihty of the recombinational event occur- 

 ring. Thus, in the cross ah X a^h^ , the number of progeny particles carrying 

 the genetic property a should be equal to the number of those carrying h if 

 there is no selection. This implies that in the progeny the following relation- 

 ship must hold 



ah + a6+ = a6 + a+6 

 or «&■*" = a"^& 



