330 



HERMANN J. MULLER 



the genes for black body color, purple eyes, and curved wings. 

 Here, too, some flies were obtained in F2 which showed the charac- 

 ters of both grandparents at once (i.e., were both bent, and 

 black, purple, and curved, and sometimes also barred). This 

 proved that bent did not lie in Chromosome II. The details of 

 the count are shown in table 1. 



TABLE 1 

 Not bent 



The number of barred and non-barred, also of males and females, 

 in each class of moderately large size, were approximately equal. 

 The linkage manifested between black, purple, and curved corre- 

 sponds as closely with expectation based on the previous linkage 

 results of Bridges, as could be demanded for the small numbers 

 involved. The determination of curved was at times uncertain, 

 owing to the tendency of bent to curve too, and the determination 

 of purple in eyes which were barred w^as also sometimes uncertain. 

 But as far as the results go, they show that bent is independent of 

 black, purple, curved, barred, and sex. 



As no counts had been made in the cross with pink, and few were 

 obtained in the cross with black purple curved, it was still con- 

 ceivable that although bent was not absolutely linked, in the male, 

 with the members of one of the three previously known groups, as 

 has always been found to be the case with other genes, still it 

 might perhaps be partially linked, in either or both sexes. New 

 crosses were therefore made with the object of securing accurate 

 counts. Some bent males descended from F2 of the cross with 

 pink were mated to black pink females. These males were found 

 to be heterozygous for pink, as half of the Fi flies were pink (al- 

 though none were black or bent). The red-eyed Fi flies were 



