304 GENE MUTATIONS CAUSED BY RADIATION 



to the present author by OHver 20 years ago, in which he reported a 

 simple proportionaHty of the frequency of induced gene mutations to 

 the dose of x-rays apphed, this finding has been ever further substanti- 

 ated and extended. There have been only such discrepancies as are to 

 be expected in consideration of the nature of the difficulties involved, 

 and of human frailty, and these have in general been rather quickly 

 corrected. The same statement applies to the independence of the fre- 

 quency of the induced gene mutations from the time distribution or 

 intensity of the dose of radiation. Thus, despite a number of unfortu- 

 nate publicized statements by non-geneticists, casting doubt on the muta- 

 genic effectiveness of small doses of radiation, the work has gone so far 

 as to make it inconceivable, on physicochemical grounds, for a single 

 ionization track traversing a nucleus to be without its proportionate 

 probabihty of inducing a gene mutation. 



Of course, this does not directly prove that one ionization or activa- 

 tion can cause a mutation, since the ionizations come grouped within 

 tracks, and a good share of them (for the most frequently employed 

 wave lengths about the same share) in tight clusters. This may be one 

 of the bases of Opatowski's (63) mathematically expressed objection to 

 the conclusion that a single ion is effective. It is hardly sufficient to 

 answer that soft x-rays have been found to have the same effectiveness 

 as ordinary x-rays, since the wave lengths tried have seldom been quite 

 long enough to meet possible objections, and the measurement of the 

 amount of penetration of those which are long enough presents great 

 uncertainties. A more telling argument might seem to be the fact that, 

 despite their greater ionization density, neutrons are not more effective, 

 but are in fact somewhat less so, than ordinary x-rays, in inducing sepa- 

 rately recordable gene mutations. This, however, proves somewhat too 

 much, for here the ion density is obviously beyond the optimal for in- 

 dividual effects. 



It might be thought that, if two or more ionizations near together 

 were ordinarily needed for producing a gene mutation, then this would 

 cause the mutation rate to rise more rapidly than the dose. For, al- 

 though the frequency of clusters located at the ends of tracks and of 

 their branches, as in the course of delta rays, would all increase linearly 

 with the dose, there would in addition be some independently produced 

 ionizations, lying in different tracks, which happened to be as near to- 

 gether as those in the natural clusters, and the frequency of such ac- 

 cidental juxtapositions would rise with the square of the dose for two 

 ionizations, with its cube for three, etc. If, however, we calculate the 

 frequency of such groupings in comparison with that of the natural 

 clusters, for any doses of the size actually used for Drosophila, we find 



