366 GENES MODIFYING NOTCH. 



When a short Notch female is out-crossed to a wild male, the daughters 

 are atavistic (fig. 94, b) , which proves that short Notch is not due directly 

 to a dominant Notch unless the wild male brings in a dominant gene 

 modifying such a dominant gene. If it does, then the next F2 genera- 

 tion should give 3 short to 1 atavistic. On the other hand, if short 

 Notch is due to a recessive modifier, the F2 ratio should be the reverse, 

 namely, 3 atavistic to 1 short. It may be stated here that the evidence 

 shows that a recessive modifier is present, but present in the sex- 

 chromosome itself, so that the numerical results follow the expectation 

 for sex-linked inheritance. The following tests were made to discover 

 the location of the modifying factor for short Notch : 



FIRST TEST. 



(1) A short Notch female was crossed to a Star Dichsete male. 

 The Star Dichaete sons of this cross get their X chromosome from 

 their mother, as well as one normal autosome carrying the normal 

 allelomorph of Star and another that of Dichsete. The fourth chromo- 

 some pair may be left out of account. \Mien such a son is back-crossed 

 to a short Notch stock female, every Notch daughter will have one X 

 from her mother and one from her father (which in turn came from his 

 mother, hence from the short Notch stock). In other words, all Notch 

 daughters have the same X chromosomes as the short Notch stock 

 females. But some of the Notch daughters will have one Star-bearing 

 second chromosome and one normal second chromosome; others both 

 normal of stock. If a recessive factor for short Notch was in the second 

 chromosome, the latter, containing both such chromosomes, should give 

 a shorter wing than the former. Similarly for Dichsete. Some of the 

 Notch daughters will have a Dichsete and a normal third chromosome, 

 others both normal chromosomes of the short stock. If the modifier 

 (shortener) is in the third chromosome the latter (both chromosomes 

 present) should be shorter than the former, etc. The results are given 

 in table 10. 



This table shows (1) that the short Notch reappears in this second 

 generation (back-cross); (2) that it is not more common in the not- 

 Star than in the Star, which means that the modifier is not present in 

 the second chromosome; (3) that it is not more common in the not- 

 Dichsete than in the Dichaete, which means that the modifier is not in 

 the third chromosome; (4) it follows that it must be present either in 

 the first or the fourth. The second test (below) will show that there 

 is in fact an important modifier in the X chromosome itself. Whether 

 another is present in the fourth chromosome will be examined later 

 when the atavistic Notch flies that also occur in table 10 will be 

 discussed. 



