STEPHENS: RESUME OF THE SYMPOSIUM 497 



some process requiring the participation of amino-acids. It is then 

 presumably incorporated as a modified component of RNA, since 

 blocking protein and RNA synthesis prevents its "fixation". Finally, 

 it is coded into the DNA system, which, after replication, requir- 

 ing further protein synthesis, is able to express the new mutant 

 phenotype. Haas has further suggested that some X-ray induced 

 mutants are produced by a similar mechanism; others may be 

 directly incorporated into DNA. 



In the higher organisms where the DNA structure is complex 

 (multistranded bundles, possibly linked tandem-wise by divalent 

 cation bonds, and intimately bound with protein) and where mutant 

 effects can be distinguished only as gross morphological deviants, the 

 probability that similar end-effects may be brought about in vari- 

 ous ways becomes very great. In view of the complexity it is not 

 surprising that different organisms react very differently to the 

 same mutagen, and that, conversely, a particular mutagen will pro- 

 duce its own characteristic reaction pattern, different from that of 

 other mutagens, in any particular organism. While similar "spec- 

 tra of reactions" to mutagens may imply common primary effects 

 in microorganisms, it would seem particularly risky to extend the 

 implication to superficially equivalent cases in higher organisms. 

 For instance, from the theoretical point of view, alkylating agents 

 could react with a variety of chemical groups — all available in the 

 components of an intact chromosome, in their possible precursors, 

 and in a number of cellular constituents not obviously associated 

 with nucleic acid synthesis. One cannot assume that where several 

 types of reaction are possible, the same one is operative; either in 

 different organisms or in different mutagens which possess more 

 than one kind of reactive group in common. At this level compar- 

 isons become idle until it can be shown that all mutagenic path- 

 ways converge at one or other "critical step" — perhaps the incor- 

 poration of a base analog, the breakage of a cation bond, or some 

 other plausible event leading to a change in nucleoprotein structure. 



Smith has reported his success in incorporating labelled base 

 analogs in the nuclei of Vicia faba, but so far it is not known whether 

 this will result in mutagenic effects. If incorporation is followed by 

 miscopying on the Freeseian model, it need not necessarily pro- 

 duce immediate phenotypically recognizable mutants. The multi- 



