BACTERIOPHAGE GENETICS 285 



number of mutant phage are examined for further mutations to other 

 altered properties. Thus, one can obtain an rh phage if one first selects an r 

 mutant and then uses the resistant bacteria to select the h mutation. The 

 phage so obtained will have the plaque morphology of the r and the host 

 range of the h. By plating on mixed indicator all four possible combinations 

 of r and h can readily be detected. These four types are indicated in Fig. 1, 



III, Recombination 



A. Linkage Map 



The characteristics of an individual phage particle which are directly ob- 

 servable are called the phenotype of that particle. Although the number of 

 different phenotypes is small in the phage system, it is possible to demon- 

 strate that independent mutations, even if they produce the same phenotype, 

 are rarely identical. This difference in the genetic properties can be demon- 

 strated by making a cross between two different phages and observing new 

 recombinant types which have properties different from either parent. 



The first paper (Hershey and Rotman, 1948) in which the results of a true 

 genetic analysis were presented concerned the linkage between a number of 

 r mutations, all of which produce the same phenotype. In this and two 

 succeeding papers, Hershey and his collaborators (Hershey and Rotman, 

 1948; Hershey and Chase, 1951) presented most of the basic facts and much 

 of the currently accepted theoretical interpretations about the mechanisms 

 involved in phage genetics. In addition, fundamental questions were posed, 

 which the work of many investigators diu-ing the following decade has only 

 partially answered. 



This first paper demonstrated that linkage structures exist on which muta- 

 tions can be arranged in a linear sequence. The experimental evidence for 

 the one-dimensional nature of the genetic map in phage has since been 

 extended considerably by Doermann and Hill (1953) for the phage T4 and by 

 Benzer (1957), who has studied a small region of the T4 genetic structure 

 very intensively. This linear mapping of mutants, which can be carried out in 

 phage as well as in higher organisms, is probably the single most important 

 conclusion from any genetic analysis; for this reason the principle will be 

 discussed in some detail. 



If one considers four independently arising mutations, which will be indi- 

 cated as a, b, c, and d, there are six possible ways in which these mutants can 

 be used in a two-factor cross. Suppose that of these six possible pairs, the 

 cross with a and b produces more recombinants than any of the others. The 

 two mutations in this pair are taken as the ones with the greatest distance 

 between them. Then the mutation c is placed between a and 6 in proportion 



