IMPLICATIONS OF THE RESULTS WITH ULTRAVIOLET 309 



of ultraviolet mutagenesis here without those complications which so 

 often distort its expression at the higher doses. 



That not only ultraviolet but also mustard works through a multihit 

 process is indicated by the synergism between the mutagenic effects of 

 these two agents reported by Swanson et at. (68). This makes it the 

 more likely that spontaneous mutations too, since they probably result 

 from changes of still lower energy level, require a concatenation of rare 

 events. The ionizations induced by high-energy radiation rise above 

 this need, perhaps because their higher energy, in its degradation, has 

 the necessary multiple effects, or because it can accomplish, more di- 

 rectly, what the multiple effects in the other cases converge to do. 



In the work on ultraviolet mutations in Drosophila it has not been 

 feasible to get mutation-frequency data for a number of widely differing 

 doses at levels low enough to make the probable role of differential via- 

 bility unimportant. Hence we have not been able to determine the prob- 

 able number of hits participating in a positive way in ultraviolet muta- 

 genesis in this material. Nevertheless we do have, in the data of the 

 group of Drosophila workers previously mentioned, a very suggestive 

 point of correspondence with the E. coli results, which raises the pre- 

 sumption that essentially the same mechanism may be at work in both. 

 The point in question concerns itself with the amount of absorbed ultra- 

 violet energy necessary to produce a specified type of mutation. 



For this purpose we must take into consideration the data for the 

 lowest dose for which significant results were obtained in Drosophila, for 

 that is the dose at which the complications of differential viability, etc., 

 which reduce the apparent effectiveness of the dose, are at a minimum. 

 It happens that, judging by the factors of the type of lamp, the distance, 

 and the time, this dose (involving an exposure to a G.E. "germicidal" 

 lamp at 200 cm for 3 min) came within the range used by Novick and 

 Szilard with E. coli. Although the intensity of our treatment appeared 

 some 4 times lower than theirs, we have not found such a difference to 

 play a very large role in our own work when the factors of the dose are 

 of the order of magnitude that obtained here. Thus we can arrive at an 

 approximate comparison of the probability of a mutation of a given type 

 (say, to resistance to phage T4) being produced in E. coli at a given 

 dose with the probability of a mutation being produced at an individual 

 locus in Drosophila by the same dose. It is true that in this particular 

 Drosophila work we have not dealt with individual loci, but the relation 

 between the overall lethal frequency (the modulus used in the present 

 work) and the average per-locus frequency of mutation has been approxi- 

 mately determined in other studies. Thus our present results can be 

 translated into these terms. 



