OF MUTANT CHARACTERS. 



219 



The back-cross tests of the female (table 62), somewhat delayed, 

 gave a total of 933 flies, of which 103 were cross-overs. The percentaj^e 

 of crossing-over calculated from these back-cross results was 17.7 — 

 a much more dependable figure than that derived from the F2. 



Table 63. — Pi, dachs black vestigial cf c?' X wild 9 9 . 



dachs black vestigial cf cf from stock. 



Fi wild-type 9 X 



One more experiment was needed to finish the determination of the 

 locus of dachs. The locus of dachs was known to be about 18 units 

 from that of black, and while it had been assumed that dachs lies to 

 the right of black, this could only be established by finding the cross- 

 over value for dachs and another of the mutants whose locus was known. 

 Vestigial was chosen for this test and preparations were made to carry 

 out a balanced triple back-cross. The triple-recessive dachs black ves- 

 tigia,l was obtained in F3 from the cross of dachs black to black vestigial. 



Table 64. — Pi, dachs cf cf X black vestigial 9 9. Fi wild-type 



dachs black vestigial cf cf' from stock. 



9 X 



The triple back-cross was made to the extent of about 1,000 flies in 

 three of the four possible ways (tables 63, 64, and 65) and gave a grand 

 total of 4,892 flies (table 66) of which 3,329 were non-cross-overs, 757 

 were cross-overs between dachs and black, 689 were cross-overs 

 between black and vestigial, and 117 were double cross-overs. 



