88 



GERT IJONNIER 



ted by the females can not hold good. From this it also follows that if 

 there are genes which are resiK)nsible for the high percentages these 

 genes must l)e strictly recessive. 



TABLE 5. 



Offspring of iinld type daughters from culture 440, table 2, when mated 



by Ci-pl males from stock. 



Percentage of exceptions 2,4:. ± 0,8o. 



The first inference that is to be made with absolute certainty is 

 thus that the X:s are themselves capable of producing the high percen- 

 tage of exceptions. But what is it in the X:s which produces this high 

 percentage? 



In order to have an answer to this question I have used a method 

 consisting in the separation of the X:s and then joining various pieces 

 of them with other X:s and in each case I have calculated the percen- 

 tage of exceptions. Thus I took 7 regular wild type females from 440 

 (table 2) and crossed them to scute broad echinus ruby tan forked 

 males (i. e. 6-pl males) (table 5). Because of the fact that 4 of these 

 7 females produced exceptional sons they showed themselves to be 

 A^A^^F-females. As the wild-type mothers of table 5 had the constitution 



„ „ ^ ,, „ f it is obvious that their scute daughters should have had in- 

 •'<• ^c ^t ^ 9 / 



herited the whole X-chromospme with the exception of the part to the 

 left of echinus and likewise the scute forked daughters must have in- 

 herited the X-chromosome with the exception of this same left part 

 and a part of the right hand end of the chromosome. A number 

 of such females were crossed to males of various kinds also from 



