EYE COLOR IN DROSOPHILA MELANOGASTER 



339 



cultures (n43, Bridges, '16, p. 147) were somewhat variable in 

 eye color. Certain of the flies were lighter than normal, having 

 markedly less of the pinkish tinge than standard eosin, so that 

 they were of a pale yellow or 'cream' color. Some of the cream 

 males were bred to rather light white-eosin sisters which were 

 suspected of being creams also. This mass culture gave a con- 

 siderable proportion of creams, though whites (males) and eosins 

 were also present. Little effort was made to separate the several 

 classes of females. Among the sons there was no confusion, there 

 being the three classes, white, eosin, and the new light color, 

 cream. 



It was now assumed that the cream was a double recessive, 

 that is, eosin plus a recessive mutant eye color, which by analogy 

 with the known dilution effects produced on eosin by such light 

 eye colors as vermilion and pink, was presupposed to be a color 

 more transparent than the normal red of wild flies. 



When an eosin male is outcrossed to a wild female, all the F2 

 females are wild-type, but half of the males are eosin and the 

 other half wild-type. If a cream male is an eosin male modified 

 by an autosomal recessive mutant as suggested above, then in F2 

 from such a cream male by wild female, the new color should 

 appear in a fourth of each of the classes which occur in the above 

 cross. The F2 females should be in the ratio of 6 wild-type : 2 with 

 the new eye color, while the F2 males should be 3 wild- type : 1 

 new color : 3 eosin : 1 'cream,' the double recessive. • 



TABLE 1 

 Pi, cream cf X wild 9 ; wilde 9 Fi wild-type 9 X Fi wild-type cf 



1 Date on which the culture began to hatch. 



The F2 results of the cross of a cream male to wild female 

 (table 1) showed that while the 'cream' character reappeared in 

 F2 in the expected proportion (1 in 16), none of the other flies 

 differed in eye color from those expected in the F2 from the simple 



