122 MUTATION AND PLANT BREEDING 



London. Most of them were found to be mutagenic. The close cor- 

 relation between carcinostatic and mutagenic abilities among the 

 alkylating agents is therefore in large part due to a bias in the selec- 

 tion of mutagens for testing. It is probable that chromosome breakage 

 is the main mechanism by which alkylating agents kill dividing 

 tumour cells. In view of the connection between mutation and chro- 

 mosome breakage it would therefore be expected that alkylating 

 agents with carcinostatic ability will be mutagens. The fact that many 

 of them are also carcinogenic is less readily interpreted. It may be 

 taken as support for the theory that the primary event in carcino- 

 genesis may be a somatic mutation. 



Oncologists have found that all effective carcinostatic alkylating 

 agents carry two or more active groups, e.g., ethylene imine or chloro- 

 ethyl groups, and they put forward the theory that cross-linkage 

 between biologically important molecules is a prerequisite for car- 

 cinostatic as well as mutagenic ability. Yet monofunctional mustards 

 have been found to be very effective in the production of mutations 

 in Neurospora (88), and the monofunctional compounds ethylene 

 oxide and ethylene imine are among the most effective mutagens for 

 barley (31, 32), where they also produce many chromosome breaks. 

 There still remained the possibility that polyfunctional compounds 

 are relatively more effective than monofunctional ones in breaking 

 chromosomes. If this were true, they should produce higher ratios of 

 translocations to lethals in Drosophila. This was tested recently in 

 our laboratory for the two compounds ethylene oxide and diepoxy- 

 butane (65). The results suggested that cross-linkage plays no signifi- 

 cant role in the production of chromosome breakage by alkylating 

 agents. 



Alkylating mutagens have been effective in all tested organisms 

 from bacteria to mammals. Data on mammals are not easy to obtain. 

 Two methods have been used successfully on mice. The first consists 

 of injecting the chemical intraperitoneally and examining the Y x for 

 heritable semisterility caused by induced translocations. By this 

 method, translocations have been detected after treatment with 

 nitrogen mustard (37), but toxicity was so strong that even with the 

 highest tolerated dose the number of translocations was small. Possi- 

 bly, treatment of semen for artificial insemination will be more 

 successful and we intend to try this in our laboratory. Triethylene 



