SEX-LINKED CHARACTERS IN DROSOPHILA. 



189 



I started six bottles of this cross, and in every case the classes 

 YR and NW were in the majority. But in five of the bottles 

 the food was very wet and the hatch poor. None of them gave 

 any YWs at all. Also, in every case the class NW far out- 

 numbered the class YR. In the sixth bottle (see Table VII., 

 Bottle No. 44) the food was in better condition (i. e., was drier), 

 the hatch larger, the ratios nearer the expected, and class YW 

 represented. This w r ould seem to indicate some relation between 

 the dark body color and the relative dampness of the food, 

 though I realize that this suggestion requires further investiga- 

 tion. 



TABLE VI. 



The members of Fi are taken from classes (i) and (4) of Table V. 



Fi. NWXYRX = NR9 YWXYW = YW<? 



Gametes (i) NRX, (2) YWX, YWX, YW 



(3) YRX, (4) NWX 



F 2 . 



NRX YWX = NR9 



YWX YWX = YW9 



YRX YWX = YRQ 



NWX YWX = AW 9 



NRXYW = NR<? 



YWXYW = YWtf 



YRX YW = YRtf 



NWX YW = NW<? 



TABLE VII. 



GIVING RESULTS OF EXPERIMENT III. 



That the coupling here is in accord with that of the previous 

 experiments is clear. A study of bottle number 44 in which the 

 food was not wet and sticky will present an even fairer view of 

 the case, as well as one in which the results are nearer those 



expected. 



EXPERIMENT IV. 



Here again I selected from virgin members of the Fa generation 

 of Experiment II., this time crossing YR females and NW males. 



