142 CELL HEREDITY 



ents than DNA, and more complexity than a supercoiling of a long, 

 double helix of DNA, and whether their mode of division is more com- 

 plex than that of DNA itself are questions to be answered in the future. 



We can ask now, however, whether the bacterial zygote, which is 

 clearly partial, undergoes something akin to meiosis in the production 

 of its haploid progeny among which the recombinants are found. It was 

 already noted that among the progeny of the persistent heterozygote, 

 homozygotes were obtained. This means that the quasi-diploid hetero- 

 zygous products of the zygote do not always resemble it in genotype. 

 A genetic event has intervened which is capable of making some loci 

 homozygous. The same was sometimes true in the transduction heter- 

 ogenote. In Chapter 12 such a mechanism will be described as it 

 operates in the chromosomes of fungi, yeast, and Drosophila. It involves 

 crossing over during mitosis; but it is also possible to account for the 

 bacterial results in terms of a copy-choice mechanism during the replica- 

 tion of DNA. The isolation of the single-cell products of conjugating 

 bacteria and of persistent heterozygotes gives results that are irregular 

 enough to suggest the latter possibility. Sister cells are usually not 

 reciprocal recombinants; in fact, the deviation from equality of recipro- 

 cal recombinants in a single clone may be as much as 15: 1. Apparently 

 the fragment of DNA from the Hfr persists and repeatedly acts as a 

 template in recombination until it is lost, perhaps in a nonviable daughter 

 cell. 



In conformity with the copy-choice hypothesis of recombination is the 

 fact that there seems to be a polarity of incorporation of the newly ar- 

 rived Hfr genetic material into the recombinant chromosome. Although 

 the thr'*^ leu^ markers of a particular Hfr strain may enter nearly 100 per 

 cent of the zygotes, only 20 per cent form thr'^ leu^ recombinants. 

 Furthermore, when some 60 per cent of the zygotes receive the ga/"*" 

 marker, only 5 per cent form, recombinants containing that locus. The 

 disparity between the efficiency ratios of 1:5 and 1:12 suggests some 

 zipper action. On the partial replica hypothesis, in order to produce 

 prototrophs one switch must occur in front of the thr leu loci and another 

 distal to them; the closer a marker is to the distal end of the introduced 

 fragment of DNA, the less likely it is to be incorporated in this partial 

 replica. Again, ultraviolet light increases the chance of recombina- 

 tion and, therefore, reduces the frequency of unselected markers from 

 the Hfr among the progeny. As was the case with irradiated phages, 

 lesions produced randomly in the irradiated genetic material interfered 

 with the process of replication and increased the frequency with which 

 the replicating element shifted from one chromosome to another. 



Similarly, when radioactive Hfr are mated with nonradioactive F~ and 



