RECOMBINATION IN VIRUSES AND BACTERIA 143 



P'^" decay is allowed to occur in the zygotes just after mating, the types 

 of recombinants can be measured as a function of that decay, and its 

 effect upon integration will be determined independently of any effect on 

 transfer. The linkage of genes in the Hfr parent is decreased with time 

 and in proportion to their distance from the thr leu markers at the head 

 end of the chromosome, suggesting that the P^^ disintegration destroys 

 the integrity of genetic material and that the sensitivity of any segment is 

 proportional to its length. This combination of physical with genetic 

 studies has not yet clarified the question of how Hfr genes are incorporated 

 into the F~ genetic material. The polarity so evident in the experiments 

 just reported could be no more than a pairing phenomenon. Although the 

 results do not demonstrate it, they are compatible with a copy-choice 

 mechanism of the sort shown in Figure 5.9. The critical experiments 

 are yet to be done. 



Analysis of the mating system of E. coli has led to a much clearer 

 understanding of the organization of the bacterial chromosome than has 

 recombination analysis by itself. F^ bacteria possess a substance (called 

 F) which is contagious, transforming F~ into F^ upon contact. All the 

 progeny of an F'^ x F~ cross are F^ by infection. The offspring of an 

 Hfr X F~ cross are almost invariably F"; there is no contagious sub- 

 stance in Hfr cells. F"^ cultures consist of a mixture of cells that are 

 phenotypicallv F'^ and some that temporarily behave as F~ , similarly, 

 Hfr cultures are mixtures of cells that behave as Hfr and as F~. Thus 

 intratype crosses produce recombinants. Beyond that, F^ cells can lose 

 their F substance irreversibly and become F~ by growth in the presence 

 of cobalt or acridine dyes. F^ cells may also mutate to Hfr; in fact, it 

 seems that the only cells which are actually able to mate in an F popu- 

 lation are the Hfr mutants. Wollman and Jacob showed this with the 

 replica plating technique (p. 52). They imprinted the surface of a plate 

 spread with F^ thr~ leu" bacteria onto velvet and replicated this onto 

 a plate of minimal agar spread with F~ met~ bio" bacteria. Only the 

 recombinants resulting from a mating formed colonies on the second 

 plate. When isolations were made from homologous positions on the 

 first plate, pure Hfr clones were obtained. Each of these had a muta- 

 tional origin and, interestingly, each independent Hfr had a different 

 genetic structure, as shown by interrupted mating experiments. The 

 same order of genes was found, but each independent Hfr mutant had a 

 different gene at the anterior end of its linkage group, as shown in Table 

 5.8. From data such as these, it can be seen that every gene is linked 

 to every other, as would be the case if the chromosome were not a rod 

 but a continuous structure with no end. It can be represented as a 

 circle, but such a structure could not transfer genes from the Hfr to the 



