464 



THE AMERICAN NATURALIST [Vol. LII 



and are used because they can be easily detected in the hetero- 

 zygous condition. Since there is no crossing over in the male in 

 melanog aster, any dichsete fly in the offspring of the back-cross 

 must have obtained one third chromosome from the dichsete 

 stock and one from the semiforked stock. Any fly which was 

 not dichsete traces back both its third chromosomes to the semi- 

 forked stock. 



Examination of the offspring from the back-cross in three 

 cultures showed that no dichaete fly was ever semiforked and, 

 conversely, all not-dichffite females were semiforked, provided 

 they were heterozygous for forked, and this includes all those 

 not homozygous forked. About 500 individuals were obtained 

 from these three cultures and the above statement is based upon 

 them. The result is absolutely clear-cut and shows that the 

 modifying factor is recessive and in the third chromosome. The 

 presence of the star chromosome (II) did not affect the appear- 

 ance of the semiforked character in any way. The location of 

 the gene within the chromosome by its linkage relations to other 

 third chromosome genes has not been carried out. 



Summary 



1. A recessive third chromosome modifying gene converts 

 heterozygous forked females into intermediate semiforked in- 

 dividuals. 



2. Homozygous forked flies are not visibly affected by the 

 modifier. 



3. The semiforked modifier rarely produces any visible effect 

 when homozygous unless the forked gene is present. 



