346 



Prof. E. B. Wilson. 



being supplied solely by the egg. Sex-linked characters shown by such 

 females must be derived from the mother. On the other hand, eggs of the 

 no-X class fertilised by the X class of sperm (normally female-producing) 

 should produce males, and these should show only sex-linked characters 

 derived from the father. This hypothesis was first tested, very ingeniously 

 and thoroughly, by combining different sets of sex-linked characters derived 

 respectively from the mother and the father. The results uniformly sustain 

 the hypothesis. Very recently Bridges has tested his assumption cyto- 

 logically. The expectation is that eggs of the XX type fertilised by sperm 

 of the X class should produce females with three X-chromosomes, while if 

 fertilised by sperms of the Y class the females should possess two X's and a 

 Y. The cytological examination has demonstrated that certain females of 

 this race actually possess three of these chromosomes. 



Taken as a whole, the foregoing evidence gives almost crucial proof in 

 favour of the conclusion that both the sex-determining factor and the sex- 

 linked ones are borne by the X-chromosome. Sex-linked heredity of the 

 type seen in birds or in Lepidoptera requires an explanation somewhat 

 different in detail but similar in principle (Spillman, Castle). In the facts 

 of sex-linkage generally the chromosome hypothesis finds, I think, its 

 strongest support, for the linkage of sex-linked factors with one another is of 

 quite the same type as that which appears in other groups that are inde- 

 pendent of sex, and the conclusion can hardly be avoided that in both cases 

 linkage is due to the same cause. 



We now take a final step in order to consider a seeming difficulty which 

 introduces us to the most recent inquiries in this field. If our hypothesis is 

 correct, how does it come to pass that linkage is not complete ? How can 

 we explain the variations in the so-called strength of linkage ? Let me again 

 illustrate by a single example taken from Drosophila, showing the heredity 

 of two pairs of sex-linked characters of the first group. One pair comprises 

 the normal grey body colour (G) and its recessive mutation yellow (Y), the 

 other the normal red eye-colour (E) and its recessive mutation white (W). 

 Let the pure-bred dominant female EG be crossed with the pure-bred 

 recessive male WY and the hybrid offspring be inbred. Were linkage 

 complete we should expect to find in the grandchildren the same combina- 

 tions as those whicli entered the hybrid, and these alone — that is to say, 

 EG or WY. In point of fact, this expectation is nearly always realised, but 

 about one individual in eighty shows one or the other of the new combina- 

 tions EY or WG. Genetically this means that E and G, or W and Y, are 

 strongly linked, yet now and then may dissolve their union and recombine. 

 Cytologically it means that the original X-chromosomes, bearing in one case 



