30 THE ORIGIN OF GYNANDROMORPHS. 



Since there was no white in the stock, the white eye must have come by 

 mutation and possibly by mutation to a sex-Hnked white-eyed gene. 



(8) In a mating in which both parents were pure bar-eyed flies a 

 male appeared (1917) (text-fig. 12) in which both eyes were round 

 and in addition one eye was three-quarters white, and the other had 

 a fleck of white in it. A germinal mutation in the mother of bar to 

 round eye must have taken place, as shown by the fact that when the 

 fly was bred it produced only normal-eyed offspring. Since this 

 male was normal, it must have come from the union of a Y-bearing 

 sperm and an X egg. Since the bar gene is carried by the X chromo- 

 some, it follows here that mutation must have occurred in one sex 

 chromosome of the mother. It is significant in this connection to call 

 attention to the fact that bar-eye not infrequently mutates (reverts) to 

 normal, as May has clearly proven. 



The other change to white was due 

 to a somatic mutation. 



(9) In stock pure for black and 

 for miniature and impure for white 

 and for red eyes a male appeared 

 that had one white eye (text-fig. 13). 

 It might appear here that simple 

 elimination in a heterozygous female 

 would account for the white eye, but 

 if the fly arose in this way the rest 

 of it should be female. Double elim- 

 ination will, however, give a result 

 of this kind, i. e., a red X is lost 

 from one half and a white X from Text-figure lo. 

 the other side, leaving both parts 



male, one red, the other white. If, on the other hand, the fly started 

 as a red-eyed male and dislocation occurred, so that most of the fly 

 had an X, the other part a Y chromosome, the expectation, based on 

 the evidence from nondisjunction, would be that the male part would 

 die. However, it might be claimed that the evidence appHes to the 

 fly as a whole and not to the survival of a small part of the body, 

 which might very well be capable of living. But we should expect the 

 absence of X to carry other consequences in its train besides loss of 

 eye-color, so that this explanation seems improbable. A third explan- 

 ation is that of somatic mutation. It is not possible to decide between 

 the assumption of double elimination and that of somatic mutation. 



(10) A somewhat similar case is shown in the male figured in plate 1, 

 figure 5. Its ancestry is not now a matter of record, but probably 

 it arose in red-eyed bifid stock that we had at the time. If so, double 

 elimination is excluded and the fly must have arisen by mutation in 

 the sex chromosome. 



