2 Mohr. 



is recorded a striking- case uf this kind, in which each ol two different 

 genes, both sex-linked recessives, produces a very particular alteration — 

 a curling of all the bristles and hairs of the individual. When the 

 first of them, called singed, arose, it was assumed that the character 

 was due to a new allelomoiph of the old gene forked which also pro- 

 duces a general alteration of the bristles not unlike the one here met 

 with. However, the tests proved that the gene which was responsible 

 for the singed character had no connection with forked. 



Later the converse situation arose: another sex-linked recessive 

 mutation was found which looks so strikingly like singed that there 

 seemed to be no doubt that the character was the result of a re- 

 occurence of the singed gene by a new mutation. Nevertheless the 

 tests demonstrated that tliis time the alteration was caused by an alle- 

 lomorph of forked, forked^, and had nothing to do with the singed 

 mutation. 



When the very special nature of the somatic manifestation which 

 the two genes have in common is considered, .it seems remarkable that 

 they are genetically unrelated. That this is the case is show^n not 

 only by their different location in the X-chromosome, but also by the fact 

 that the homoz\'gous singed females are completely sterile, in contrast 

 to the forked^ flies which are of normal fertility in both sexes. The 

 sterility of the singed females is due to a defective condition of their 

 eggs which is accompanied b\' a constant change in their form. The 

 forked* mutation arose in a fly which was homozygous for the ordinary 

 forked gene. It accordingly furnishes an example of a secondary 

 mutation in the forked locus, being removed from normal by two in- 

 dependent umtational steps. 



A few other mutations found during the work are recorded; among 

 these are an additional mutation in the singed and two in the forked 

 locus. Two g\Tiandromorphs are described, one involving the singed 

 character, which is one of the most favorable mutations for the stud}' 

 of gynandromorphs in Drosophila. For the explanation of the other a 

 secondary mutation of ruby- back to normal offers the simplest of the 

 alternative possibilities discussed. 



The singed mutation arose while the author was a guest in the 

 laboratory of Dr. T. H. Morgan at Columbia University; and the author 

 whishes to express his sincere thanks to Dr. Morgan for this hospi- 

 talitA- and for the stocks with which the crosses were carried out. He 



