278 



THE SECOND-CHROMOSOME GROUP 



ratios are simplified more than in the ordinary F2. Thus in table 125, 

 the classes of star dachs and wild-type are both simple cross-over 

 classes (x), the dachs are simple non-cross-overs (n), and the star 

 class only is complex, being a double non-cross-over plus a single 

 cross-over {2n-\-x). In the ordinary coupling F2, the wild-type class, 

 which corresponds to the star of this experiment, is more complex 

 {3n+2x). 



Six of the F2 cultures gave the expected results, with about 33 per 

 cent of crossing-over between star and dachs; the other culture (2217, 

 table 126) gave no dachs whatever among 116 flies, one-third of which 

 were expected to be dachs. 



Table 126. — Progeny from star males and females heterozygous 



for dachs lethal ( - — 7- | . 



DACHS-DEFICIENCY? 



The phenomena of ''deficiency" had just been worked out in detail 

 in the case of forked-bar deficiency (Bridges, 1917), and accordingly 

 the non-appearance of the dachs where both parents were expected 

 to be heterozygous was immediately attributed to the occurrence of 

 deficiency for the dachs gene. It was thought that the character dachs 

 was unable to appear because of the physical absence or the complete 

 inactivation of the dachs gene. One of the most striking features of 

 forked-bar deficiency had been the lethal nature of the change. The 

 same process that had removed the genes for forked and bar and the 

 intermediate region had replaced their action with a lethal agency. 

 Immediately the supposed case of dachs-deficiency was examined to 

 see whether in it also the elimination of a gene had resulted in the 

 substitution of a lethal. The ratio of 23 wild-type to 93 stars showed 

 unmistakably that a lethal agency was operative; that the seat of 



