V. MOLECULAE MECHANISM OF MUTATIONS 253 



sites (Brenner et al., 1958) are more scattered and are nearly all different 

 from the spontaneous ones. 



These observations show that the mutagenic specificity is more 

 complicated than one might have expected from the chemical observa- 

 tions. Clearly the mutability of a given nucleotide pair is influenced by 

 other structural factors. Part of this effect may be due to the different 

 kinds of cytosine bases that are observed in DNA. 5-Methylcytosine is 

 found in many organisms in addition to cytosine and preferentially next 

 to G (Sinsheimer, 1955) ; in T2, T4, T6 phages differently glucosylated 

 5-hydromethylcytosine bases are observed and the extent of glucosyla- 

 tion depends on the neighbors (Lehman and Pratt, 1960; Kornberg et al., 

 1961). Hydroxylamine distinguishes chemically between some of these 

 C bases (Freese et al., 1961b), hence other mutagenic mechanisms, 

 including the spontaneous one, might do the same. 



In addition, the nearest neighbors of a given base undoubtedly can 

 have a direct influence upon its mutability, as is indicated by the sites 

 of \evy high mutability mentioned above. Such effects may be produced 

 for example by the difference in the hydrogen bond strength of the next 

 neighbors, that of A-T pairs being less than that of G-C pairs, as it 

 determines the ease with which the base pairs can open up, or by 

 sterical effects of the next neighbor, purines being bulkier than pyrimi- 

 dines, or by an electronic influence of the Tr-electron clouds of one base 

 upon the next one. 



It is conceivable that certain DNA sites have evolved by such means 

 a special protection against spontaneous mutations and against certain 

 mutagens. 



B. THE SPECIFICITY OF REVERSE MUTATIONS 



The aim of studying the induction of reverse mutations is to divide 

 mutations into classes which do or do not respond to a given mutagen or 

 which respond much and little. When forward mutations, induced by a 

 chemical of known specificity, are used the consistency of the theory of 

 mutations can thus be checked. In addition one can hope to determine 

 for each or at least many mutations by which base pair change they 

 came about; this would also enable one to infer the mutagenic mecha- 

 nism for agents whose chemical specificity is not sufficiently understood, 

 as well as for the spontaneous mutations. 



1. Complications of the Analysis 



The interpretation of experimental results concerned with the induc- 

 tion of reverse mutations is complicated by two factors: the higher 

 mutagenic specificity and suppressor mutations. 



