RADIATION AND THE STUDY OF MUTATION IN ANIMALS 1249 



his lead, Jollos (77, 78, 79), using much higher temperatures for a shorter 

 time, obtained curious and specific results which have not been confirmed 

 elsewhere. Rokitzky (145) and Plough and Ives (144), as well as 

 Mackensen (Muller, 113) have found mutational effects which might 

 have been expected on the basis of Muller's earlier data. The most 

 striking thing about the experiments of Goldschmidt and of Jollos is the 

 apparent specificity of the temperature effect. This would imply that 

 at different loci the temperature coefficients of mutation process differ — a 

 not improbable state of affairs. The closest approach to this is contained 

 in the statement of Plough and Iv-es that recurrences are in their data 

 somewhat more frequent for certain loci than for others, although they 

 are by no means sure of the statistical significance of the data. 



It is probable from these results, that temperature is effective in 

 changing mutation rate. Yet the effect of radiation has been shown to 

 be independent of temperature. This in itself does not indicate that the 

 two types of mutation have different origins. It is conceivable that the 

 effect of temperature is on the stability of genes with respect to radiation, 

 in which case at the heavy dosages the effect of a rise of temperature 

 would be inconsequential in comparison with the energy of the incident 

 radiation, and so would be masked. 



Following the discovery of the effect of radiation in producing muta- 

 tions several workers suggested the possibility (not without a certain 

 romantic flavor) that all mutations were due to radiation. On such a 

 basis, the rapidity of the evolutionary process is in part dependent on 

 the intensity of the surrounding short-wave radiation. Accordingly 

 Babcock and Collins (7) and Hanson and Heys (63) felt the necessity 

 of testing the potency of "natural" radiation in inducing mutations. 

 They found, as might be expected, slight increases in the mutation rate 

 corresponding to changes in incident radiation with which they were 

 working. More to the point, however, are certain calculations of Muller 

 and Mott-Smith (119), of Timofeeff-Ressovsky (176), and of Efroimson 

 (43). 



They have compared the total ionization to which Drosophila is 

 subject during its life with that which would be expected were all the 

 "spontaneous" mutations induced by "natural" radiation. This 

 involves an extrapolation from the dosage-effect curve, on the assump- 

 tion that the linear relation holds at low radiation intensities. They 

 have found enormous discrepancies, certainly significant, between the 

 observed and calculated mutation rates. But these discrepancies do 

 not show that "it is accordingly probable that most mutations come 

 about as the result of other causes" than radiation. They show at best 

 that at low dosages the relation between intensity and effect is not linear. 

 That this is quite likely has already been pointed out on other grounds 

 (page 1240). 



