158 THE SECOND-CHROMOSOME GROUP 



chromosomes are independent in heredity. Accordingly, blistered was 

 mated to pink as the tj^pical third-chromosome mutant. The F2 from 

 this cross (table 19) gave a 9 : 3 : 3 : 1 ratio, as expected, which con- 

 firmed the view that bhstered is not in the third chromosome. 



Blistered was also mated to black as a representative of the second 

 chromosome; curved was avoided, since it was planned to continue 

 this line and there still lingered some fear of the masking effect of 

 curved on blistered. 



THE SEMI-DOMINANCE OF BLISTERED— FREE-VEIN. 



"WTien the Fi flies from the cross of blistered to black (table 20) began 

 to hatch, it was noticed (October 23, 1913) that nearly two-thirds of 

 the females and a few of the males showed a small section of vein 

 lying free in the third posterior cell and parallel to the fifth longitudinal 

 vein. The length of this extra vein was oftenest about two-thirds 

 the length of the posterior cross- vein, but it varies to a minute dot. 

 Flies often showed it in only one wing. This character, while it is very 

 irregular in occurrence, is very easy to classify, since the vein is clear 

 and sharp. The work done with this character is therefore exact and 

 accurate, without the approximations and close decisions that are 

 required in working, for example, with variable colors of slight average 

 difference. It was guessed that this free vein occurring in Fi was due 

 to the action of the blistered gene, which thus shows an irregular and 

 partial dominance. If this were true it should be possible to work 

 with blistered as a dominant, though only those flies which actually 

 show the character can be used in calculations; many of the others, 

 while somatically normal, belong genetically with those showing the 

 free vein. Accordingly, the Fi males with the free vein were back- 

 crossed to black females, with the expectation that if the gene for the 

 free vein (bhstered) were in the second chromosome all of the off- 

 spring would be either black or free-veined, there being no crossing- 

 over in the male. The result showed that the gene was in the second 

 chromosome (table 20), for the flies which were extra- veined were 

 nearly all not-black. There were 5 black flies which showed a slight 

 development of an extra vein, but these veins were probably due to 

 other causes, since 2 of the flies when tested gave 81 and 93 offspring, 

 respectively, none of which showed a trace of the free- vein character. 

 While all of the not-black flies must have been of the same consti- 

 tution — i. e., heterozygous for the free- vein gene — only 53 per cent of 

 them showed the extra vein. 



At the same time that the Fi free-veined males were back-crossed, 

 some of the similar females were likewise back-crossed (table 20). 

 Since in the female there is crossing-over, this experiment should show 

 the amount of crossing-over between black, whose position is well 

 known, and the gene for the free- vein ; that is, we should obtain one of 



