GENES MODIFYING NOTCH. 



359 



X 



X 



ruby 



X 



There is a small chance of contamination from the food when so 

 many cultures as these are carried out, even although all the ordinary 

 precautions are taken. Thus the four normal males that appeared are 

 under suspicion. Two of these were tested to the second generation 

 and found to contain no other than normal genes. Since the male con- 

 tains only one sex-chromosome, it was to be anticipated 

 that such red-eyed normal males would not contain any 

 other sex-linked genes than they show unless something 

 unusual had occurred. It is, however, conceivable that 

 a lethal-bearing male rarely comes through (as happens 

 in the case of a few other lethals) even although no notch 

 is observable in the wing. Were this possible some of his 

 daughters or granddaughters would be expected to show 

 Notch, but as none did so, the presumption is that these 

 red-eyed males were not of this kind. It is also possible 

 to mistake at times an old ruby-eyed fly for a red-eyed fly 

 if only a casual examination is made, but as it was appre- 

 ciated that no red-eyed male was expected, a careful 

 scrutiny of the red males was made. For these and 

 other reasons I have discarded the two untested males of 

 the four from the general calculation in locating the fac- 

 tors, although I have also given the calculations in which 

 these are included. The differences in the two results are 

 too small to be of significance. 



A similar doubt arises about the corresponding double 

 cross-over classes in the females that gave two eosin Notch 

 ruby forked females and two normal females. Both of 

 the latter were tested with eosin ruby forked males and 

 gave normal sex-ratios and no Notch daughters. All the 

 daughters and sons had red eyes. For these three reasons there can be 

 little doubt that both of these females in question were due to con- 

 tamination by wild-type flies. 



The other two daughters can not be so easily dsmissed, because 

 they were obviously not due to contamination, since they showed all 

 of the genes involved in the experiment. Unfortunately I have no 

 records to show whether they were tested, or, if so, whether they lived. 

 It is true that occasionally flies are found that have a nick in their 

 wings due to accident or to some other mutation, and in numbers as 

 large as those here employed, the occurrence of such flies is to be 

 expected. It is to be regetted that I was not aware of the fact that 

 Notch flies (even those phenotypically normal for wing margin) can be 

 identified, under the microscope, by the thicker second and fifth veins. 

 By this means the two normal eosin notch ruby forked females could 

 have been securely identified. While it is highly probable that the 

 same difference holds for the selected notch, this has not been deter- 



farfed 



FIG. 92. 



