338 BACTERIOPHAGES 



cluster of Hershey and Rotman. This rll cluster was analyzed 

 in great detail by Benzer (1955, 1957) as will be described later. 

 A peculiar discrepancy between observed and calculated 

 recombination frequencies was noted by Hershey and Rotman 

 (1949) and by Doermann and Hill (1953). With three linked 

 loci in the linear order 1, 2, 3, if the recombination frequency 

 between locus 2 and 3 is symbolized C23, etc., the recombination 

 frequency between locus 1 and 3, C13 should be 



C12 + C23 ~ 2(Ci2 X C23) 

 The negative term is to take account of the fact that if recombina- 

 tion occurs both between 1 and 2 and between 2 and 3 the result 

 will be no recombination between 1 and 3 because the second 

 recombination reverses the effect of the first. When recombina- 

 tion frequencies between terminal loci calculated in this way 

 were compared with observed recombination frequencies, the 

 calculated values were almost invariably higher than those 

 determined experimentally. This suggests that the frequency of 

 double recombinations is greater than that calculated on the 

 basis of the single recombination frequencies, that is, the term 2 

 (Ci3 X C23) is an underestimate of the frequency of double 

 recombinations. In classical genetics it is usually found that the 

 frequency of double recombinations between closely linked 

 markers is less than that calculated from single frequencies, a 

 phenomenon that has been termed "interference" because the 

 occurrence of a crossover appears to interfere with the occurrence 

 of another crossover in adjacent regions. When the opposite 

 phenomenon was observed in bacteriophages it was referred to 

 as "negative interference" without any necessary implications as 

 to mechanism. The term "negative interference" in phage 

 papers refers then to an excess of double recombinations over 

 those calculated on the basis of single recombination frequencies 

 using the classical genetic formula given above. 



e. Multiplication after Recombination 



There are a number of possible time relationships among the 

 processes of multiplication, recombination, and maturation. 



