80 MUTATIONS 



told me recently tliat there has been hardly any exception to the rule 

 that only two- or three-armed compounds are carcinostatic. 



This was taken over into the theory of mutagenesis, but there it 

 proved to be wrong. A one-armed mustard or other one-armed alkylat- 

 ing agent may be an excellent mutagen, even better than the corre- 

 sponding two-armed compound. But the point is that some very good 

 mutagens were discovered on the basis of the cross-linking theory, and 

 only later were the corresponding one-armed compounds tested and, 

 sometimes, found to be also very effective as mutagens. Ethylene oxide 

 was tested after diepoxybutane had proved to be mutagenic. TEM, tri- 

 ethylene imine, was tested before ethylene imine. At least, this is so for 

 countries west of the Iron Curtain. 



In Russia the story of chemical mutagenesis was different, and this 

 brings me to the third principle on which mutagens were detected. The 

 first was pharmacological, the second was the cross-linkage theory. The 

 third forms the basis of Rapoport's (59) work in Russia, which became 

 known in the west only years after it had been carried out and when 

 Rapoport himself was no longer allowed to work along these lines. 

 Rapoport, like everybody else at that time, believed that the specificity 

 of the genetic material resides in the protein. So he searched for chem- 

 icals that would act on proteins, and one of the many that he tested 

 with success on Drosophila was ethylene imine. 



On the basis of the same theory, Rapoport (58) discovered the 

 mutagenic action of formaldehyde. It is interesting to remember that, 

 likewise on the basis of this theory, tests for mutagenic ability of 

 nitrites were carried out long before the spectacular success with this 

 substance in recent years. In 1939, Thom and Steinberg (75) produced 

 mutations in Aspergillus with nitrites. When formaldehyde had been 

 found to be mutagenic, Haldane suggested trying nitrites because these, 

 like formaldehyde, react with the amino groups of proteins. Robertson 

 and Rendel in our Institute tried nitrites on Drosophila. The first 

 series was very promising, but later series were negative, and they 

 abandoned the work. 



We come now to theories that are closer to what we now think is the 

 truth about the nature of the genetic material. Even before Avery's 

 work so strongly implicated nucleic acid, it was of course known that 

 nucleic acid is an important part of the genetic material. This started 

 off experiments wnth purines or pyrimidines or with substances known 

 to react with nucleic acid, like pyronine and acriflavine. Many of them 

 were found to be chromosome breakers and mutagens. 



There is one claim which has been going on for, I think, twenty 



