MUTAGENESIS 159 



of the discussion, supported the fairly strong possibility that, at least 

 in higher organisms, many mutations are due not to a direct effect on 

 DNA but to an effect on either nucleotides or proteins. 



I was greatly interested in Benzer's presentation of the latest data 

 relating to the "topography" of the r II region of phage T4 and should 

 like to comment on it in more detail, because it deals with problems 

 I am much concerned with at present. There are more than 2000 

 mutants of separate origin representing that region and about 300 

 independent sites have been identified. Since calculations indicate the 

 existence of about 400 sites, we can expect that at least 100 remain to 

 be discovered. Mutants of the r // region belong to two complementa- 

 tion groups, A and B, and thus the region includes either one or two 

 gene loci, depending on the definition of a locus. It seems to me that 

 the experimental evidence, rapidly accumulating in work with phages 

 and bacteria gives strong support that a genetically identifiable site 

 corresponds to a nucleotide pair of the DNA molecule of a gene. If 

 we accept this interpretation, it can be estimated that the r // region 

 contains about 400 nucleotide pairs — provided, that is, that every 

 nucleotide pair is a part of the genetic code. At last year's session 

 of this group Levinthal estimated that the gene locus controlling the 

 phosphatase system in Escherichia coli comprises approximately 2000 

 nucleotide pairs. If both these estimates are reasonably accurate, they 

 indicate either that the gene loci of E. coli are longer than those of T4 

 phage or that different genes in different microorganisms vary con- 

 siderably in length, or both. 



The most striking feature of the topographical map of the r // 

 region is the identification of two "hot spots," one in section A and 

 one in B. The hot spot in region B accounted for almost 500, or about 

 25 per cent of all the mutants found, and the other hot spot for almost 

 300, or 15 per cent. Thus about 40 per cent of all mutations occurred 

 at these two locations. How can we explain this observation? Ob- 

 viously, the instability can't be due to the constitution of the 

 nucleotide pair present at a hot spot site in the wild-type gene because, 

 since there are only four kinds of nucleotide pairs, hot spots would 

 then be frequent. So it must be the environment of the site that 

 affects its mutation frequency. The position of a nucleotide pair, there- 

 fore, or even more probably its relationship with surrounding nucleo- 

 tide pairs determines its pattern of reduplication, that is, whether it 

 will be reproduced or replaced by a certain other nucleotide pair. 



There is evidence — and Benzer discussed such data in connection 

 with his material — that reverse mutants may be due to reverse changes 



