NATURE OF THE GENETIC EFFECTS 455 



tions, similar in their effects, of separate but very closely neighboring 

 genes. Although these have usually been interpreted as ''duplicate 

 loci," descended from a common ancestral locus that had undergone a 

 duplication which became established in the evolution of the normal type, 

 it seems probable that some of them are cases in which two essentially 

 different but neighboring genes interact through a position effect. How- 

 ever, in one case at least, that of "scute" and its neighbor gene ''achaete," 

 it was possible to prove by means of radiation-induced structural changes 

 that the two loci are able to exert most of their characteristic effects even 

 when they are widely separated from one another. In this case then the 

 evidence of their origination by duplication is convincing. The present- 

 day difference in their normal function, and in their mutational potential- 

 ities, illustrates well the mutational differentiations which such dupli- 

 cated loci tend to undergo in the course of their evolution subsequent to 

 the duplication. Another very instructive example of this kind is 

 furnished by the loci for sperm motility present in the Y chromosome. 

 It was shown in ingenious work of Neuhaus (1939), utilizing the position 

 effects of a large series of radiation-induced translocations involving 

 breaks at slightly different positions in the Y chromosome, that there are 

 over a dozen different but related genes in that chromosome, the com- 

 bined action of all of which is necessary for sperm motility. These genes 

 must have arisen through repeated duplications of a common ancestral 

 gene, and after duplication have undergone mutually complementary 

 mvitational differentiations. 



It was also proved by fragmentation of the X chromosome of Dro- 

 sophila, brought about by application of X rays, that this chromosome 

 contains a considerable number of loci which act cumulatively in sex 

 determination (Muller and Stone, 1930; Patterson, 1931a; Dobzhansky 

 and Schultz, 1931, 1934; Muller, 1932a; Patterson, Stone, and Bedichek, 

 1937). It can, however, be inferred on evolutionary grounds that there 

 was at first only one such locus. The present multiple condition could 

 hardly have arisen exclusively by duplication of that locus, since this, at 

 the first such step, would have given one X chromosome the potency of 

 two and so would have upset the whole sex-determining mechanism. It 

 is therefore necessary to conclude that there was, to some extent, a dis- 

 persal of the sex-determining function over a number of different genes, 

 by means of gene mutations in them (Muller, 1939a) ; the original sex- 

 determining gene must meanwhile have diminished in its potency, in a 

 number of steps. Thus, as with the sperm motility gene of the Y chromo- 

 some, but by a mechanism to some extent different, the entire collection 

 of these genes finally became necessary in order to fulfill completely the 

 function originally carried out by one gene. 



An example of a very different method whereby the production of 

 chromosome breakage by radiation has made it possible to obtain evi- 



