654 VERMILION-DEFICIENCY 



Relation between Vermilion-Deficiency and Vermilion-Duplication. 



That the deficient region actually does extend to the left of the 

 vermilion locus was indicated by use of vermihon-duplication. The 

 stock of vermilion-duplication consists of flies homozygous for the 

 vermilion gene and at the same time homozygous for the not-ver- 

 milion gene, since these flies carry at the left end (point of spindle- 

 fiber attachment?) a transposed section of chromosome including the 

 loci for vermilion and sable. These flies are not vermilion in ap- 

 pearance since the two wild-type allelomorphs dominate over the two 

 recessive vermilion genes. But if such females are crossed to ver- 

 milion males all the daughters are vermilion and all the sons are wild- 

 type — simulating the Abraxas type of ''criss-cross" inheritance, or 

 dominance of vermilion. These daughters are vermilion because 

 two recessive vermilion genes are present and dominate over the 

 single wild-type allelomorphs of the Fi female. When vermilion 

 females carrying vermilion in one X and deficiency in the other were 

 crossed to males of the duplication stock, all the daughters were ver- 

 milion. In its effects upon the dominance of vermilion, vermilion- 

 deficiency may be substituted for a vermilion gene without distinguish- 

 able difference. This fact also indicates a positive action of vermihon- 

 deficiency. Had the deficiency been neutral in effect, then the flies 

 carrying vermilion, not-vermilion, and vermilion-deficiency should 

 have been wild-type like the normal diploid heterozygote. 



The Fi females were crossed to males from the vermilion-dupli- 

 cation stock (Table IX). The most significant point observed in F2 

 was that the lethal effect of vermilion-deficiency was not annulled 

 by the presence of the duplication. The probable explanation of this 

 is that the two regions do not coincide; i.e., although the duplicating 

 fragment includes the loci from vermilion to the right as far as sable, 

 nevertheless it does not extend so far to the left of vermilion as the 

 deficiency does. This excess of the deficient region to the left of ver- 

 milion would have to be enough to include one or more loci vital to 

 the animal. 



It should be noticed that there was 35.5 per cent of crossing over 

 between zero and vermilion-deficiency, which is about 3.5 units 

 more than the average normal value. 



