86 T. H. MORGAN 



parallel lines; the upper, heavier one, the maternal, and the lower, 

 lighter, one the paternal chromosome. Lethal and non-white en- 

 tered from the mother and non-lethal and white from the father. 

 The sex-chromosomes of the Fi female are represented with the loci 

 for lethal and white close together. The cross between the lines 

 indicates that crossing-over occurs between the loci in question. 

 (The cross does not represent the postulated twisting, but only 

 that it occurs in this general region.) Although for convenience 

 the cross stands midway between the loci this is only a convention, 

 for it will be understood that crossing over occurs at any point 

 between the loci in question. The gametes of A will be of four 

 classes, the first two, where no break has occurred between the 

 loci L and W; and the second two, where a break has occurred, 

 and crossing over has taken place. The first class of gametes 

 (non-cro9s-overs) will be of two kinds, equally numerous, viz., 

 1 W and L w. The second class of gametes (cross-overs) will 

 also be of two kinds, viz., 1 w and L W, but relatively few in 

 number. 



(lives) 

 (dies) 

 (lives) 

 (dies) •*^^**** • 



The analysis given above shows that rarely a red female 

 should occur which is not heterozygous for lethal. The chance 

 for her occurrence is the same as for the occurrence of the red 

 male that lives, viz., 2:370 (table 3) or 1:185. Since therefore 

 over 99 per cent of the red females are heterozygous for lethal 

 we have kept the stock running by mating in mass cultures the 

 red-eyed females to their white brothers. The presence of a 

 female homozygous for non-lethal would make itself apparent 

 at once by the appearance of red males which otherwise are 

 extremely rare, and the reappearance of the 1 :1 :1 :1 proportion 

 of table 2. In this case the stock could be recovered by breeding 



