1252 BIOLOGICAL EFFECTS OF RADIATION 



The consideration of the stabiHties of the genes, however, is made 

 possible by the existing experiments, and even leads to certain direct 

 inferences as regards the evolutionary process. As has already been 

 shown, it is possible that the analogy of the Galton polygon holds in the 

 quantitative relations of the mutations at the white locus in Drosophila; 

 that the frequency with which mutation to a given allelomorph occurs, 

 is a direct function of the stability of that gene itself. It is possible that 

 further studies in this direction may establish this relation definitely 

 for many genes and give it a meaning in terms of reaction velocities or 

 of energy relationships. From the studies of reverse mutations it defi- 

 nitely follows that the normal allelomorph, which is prevalent in a 

 population, is for many genes maintained against a mutation pressure 

 which tends to replace it by more stable allelomorphs. This is not 

 surprising, in view of the calculations of Haldane, Wright, and Fisher 

 on the role of mutation pressure in mendelian populations. However, 

 it also suggests that there is a large population of genes whose normal 

 allelomorphs are the most stable of the possible series; and of these we 

 are very unlikely to obtain mutations. 



Such considerations are of the utmost importance in any attempt 

 to calculate the size and number of the genes. The ambitious attempt of 

 Gowen and Gay (52), in particular, which gives a figure seven times higher 

 than that of the preceding computations (92, 108), fails because of the 

 use of lethal genes, not tested for recurrence. These have been shown by 

 Patterson to occur on the average seven times as frequently as the viable 

 mutations at the same locus. Now the calculation of the number of 

 genes depends on the frequency of recurrence of viable mutations (Mul- 

 ler, 108) from which the total number of genes may be obtained. Gowen 

 and Gay by assuming that "lethal genes" are per se different from others, 

 and so multiplying their value for "visible" mutation fall into serious 

 error, which results in their high figure. There is now some reason to 

 believe that a direct cytological check may be possible in the chromosomes 

 of the salivary glands of the Drosophila (128). But until this is possible, 

 the distribution of the stabilities of genes must be taken into account in 

 any such calculation; and the determination of their size (8a, 52) on the 

 assumption of direct electron hits certainly awaits an elucidation of the 

 mutation reaction. 



It is a pleasure to record the stimulus of frequent discussion with Drs. 

 A. H. Sturtevant and Th. Dobzhansky. I am also indebted to Miss E. 

 M. Wallace for help with the preparation of the figures. 



REFERENCES 



1. Agol, I. J. Evidence of the divisibility of the gene. Anat. Rec. 47: 385. 

 1930. (Ahstr.) 



2. AooL, I. J. Step allelomorphism in Drosophila melanogaster. Genetics 16: 

 254-266. 1931. 



