BACTERIOPHAGE GENETICS 293 



Brescli and Starlinger (1958) did a similar analysis, and Hausmann and Bresch 

 (1958) performed crosses with three different parental types with which it is 

 possible to distmguish between the various models. They concluded that the 

 number of recombinants which had received genetic markers from each of the 

 three parents was too large to be accounted for by successive pair-wise 

 matings in the infected cell. On the other hand (Trautner and Bresch, 1958), 

 the ratio of recombinants which had markers from three parents to those which 

 had markers from only two varied with time in a way which was not easily 

 understood m terms of group mating. They interpreted the result as a process 

 in which group mating occurs combined with a progressive mixing of parental 

 types which, at the time of uifection, they assumed had a tendency to be 

 separated from each other. This question is certainly not yet settled, but, at 

 present, the evidence suggests that group mating can occur. 



The mathematical analysis which defines group mating is very general and 

 does not depend on any very specific model. It requires only that several par- 

 ticles can enter a mating event and recombinants can be produced which 

 contain information from all of them. A specific model which is reasonable 

 in view of what is known of other organisms and which is contained in this 

 general definition would be one in which pairing occurred between two 

 structures, but only for a short length of the chromosome. Local pairing 

 could take place between two particles in one region and between one of 

 these and a third elsewhere on the chromosome. Thus repeated local pairing 

 is, formally at least, completely equivalent to group mating as the term is 

 used in these analyses. 



Phage Si 3, the smallest one known, differs from others studied in that it 

 has not been possible with SI 3 to demonstrate any genetic recombination. 

 Mutants for S13 are known, and attempts to show recombmation between them 

 have been made by Zahler (1958) and Tessman and Tessman (1958). Zahler, 

 using both host-range and plaque-type markers, looked for recombinants 

 only m smgle bursts to make certain that only true cases of mixed infection 

 were scored, and found that the recombination frequencies cannot be greater 

 than 10~^. Tessman and Tessman used a variety of host range markers and 

 were not able to demonstrate any recombination down to the level of 10~'. 

 It is not known if this lack of observable recombination is an inherent char- 

 acteristic of the phage or if the appropriate conditions for carrying out the 

 cross have not yet been found. 



V. The Mating Event 



The question which has provoked the most interest in phage genetics over 

 the last decade was originally posed by Hershey and Rotman (1948) and 

 concerns the nature of the elementary mating act. As it was put by them, 



