GENES IN THE MALE OF DROSOPHILA. 



197 



The FI heterozygous females (sisters to the males just tested) 

 were also tested by crossing to black vestigial males. Four 

 classes were produced. 



This result, in contrast to the last one, is explicable on the 

 assumption that crossing over occurs in the female, as the fol- 

 lowing analysis shows: 



Fi Q B V h V - -Bv bv 



11 -f > *-' * g U V g -U v g L/C gj 



Black vestigial cf, bv g bv g 



BV g bi'g, gray long. 

 bV g bi'g, black long. 

 Bv g bi'g, gray vestigial. 



bv 



g, black vestigial. 



The expectation on the basis of free crossing over in the female 

 is 2 gray long, 2 black long, 2 gray vestigial, 2 black vestigial. 

 The two classes gray long and black vestigial are cross over 

 classes, and their smaller numbers result from linkage. It will 

 be noted that the two vestigial classes fall below the corresponding 

 long classes, and this can be accounted for as due to viability. 

 The percentage of crossing over on the basis of the data is 17.9. 



These last two experiments show that the realized classes are 

 explicable on the basis of non-crossing over of factors in the 

 heterozygous male, and of crossing over to the extent of 17.9 

 per cent, in the female. This conclusion can be tested by bring- 

 ing into the cross the same factors in other combinations. 

 One such test is as follows. Gray, long winged females (BVg) 

 were bred to black, vestigial males (bv g ) and produced gray 

 long males and females. The F! heterozygous males produced 

 in this way were mated to black vestigial females, and gave: 



