NATURE OF THE GENETIC EFFECTS 415 



nia). In such cases, however, it was usually difficult to know, in view of 

 the high variability of the spontaneous mutation frequency, whether 

 differences in genetic constitution and in other conditions than the one 

 under investigation had been rigorously enough ruled out. The same 

 stricture applied to similar work which had been carried out by Stubbe, 

 Baur and their co-workers on visible mutations in the snapdragon, 

 Antirrhinum (Baur, 1924). 



The first outstanding success in the use of a chemical for mutagenesis 

 was obtained in 1941 by Auerbach and Robson in their tests of mustard 

 gas and related substances on Drosophila, although the information could 

 not be published in declassified form until 1946 (Auerbach, and Robson 

 1946 et. seq.). Robson had been led to suggest that tests be made to 

 determine whether mustard resembles ionizing radiation in being muta- 

 genic, from the consideration that its deep, slow-healing burns and other 

 somatic effects showed, as he had noticed, a peculiar resemblance to the 

 somatic effects of ionizing radiation. Moreover (as is now known) 

 mustard, like radiation, also tends to inhibit mitosis. Like radiation 

 again, it was found in Auerbach's tests to produce both gene mutations 

 and structural changes of varied kinds, in considerable abundance. The 

 former can in fact be induced by mustard with practically as high a fre- 

 quency as by radiation. Structural changes, however, are induced 

 somewhat more rarely than by doses of radiation giving the same fre- 

 quency of gene mutations as the mustard. This may be because the 

 chromosome breaks are not so nearly simultaneous in their occurrence 

 as with radiation, being often delayed. In fact, unlike what happens 

 with radiation, the mutation or chromosome change induced by mustard 

 sometimes occurs in a chromosome that is a rather remote descend- 

 ant of the one which had been directly treated, as though the treat- 

 ment had originally induced in the genetic material a metastable 

 state (like that of Baur's spontaneous "premutations" in Antirrhinum), 

 in which the somatic effect was still normal, but which became copied in 

 the process of chromosome reproduction, and finally resulted, in some of 

 the descendant chromosomes, in a stable mutant configuration (Auerbach, 

 1947). 



The mutagenic action of the mustard group is, like that of radiation, 

 very general. It has now been proved for organisms of the most varied 

 kinds, including bacteria, fungi, and (as far as structural changes are con- 

 cerned) higher plants and mammals — in the latter two groups by KoUer, 

 Ansari, and Robson (1943) and Koller (1949). Since the discovery of 

 mutagenesis by mustard, it has become difficult to avoid the conclusion 

 that the similarity in the types of somatic effect of mustard and of ion- 

 izing radiation is, for most of the effects, based on the production of 

 chromosome changes by both of these agents. On this view, it is to be 

 expected that, like radiation, mustard would also have an especially 



