80 THE ORIGIN OF GYNANDROMORPHS. 



male and ligustica. This is the result which Boveri thinks is shown 

 by the Eugster gynandromorphs. Conversely, if the ligustica chromo- 

 some were lost, all parts containing the descendants of this nucleus 

 would be male and mellijica. This is the result that von Engelhardt 

 claims to have found in his gynandromorphs. Thus both results are 

 expected on the hypothesis of chromosomal elimination ; each is equally 

 possible. Boveri's hypothesis of partial fertilization explains only 

 one case; Morgan's former view of sperm-nuclear development will 

 explain only the other; the hypothesis of elimination will explain both, 

 and for this reason is at present to be preferred. Moreover, since it is 

 demonstrably the way in which gynandromorphs are produced in Droso- 

 phila, this hypothesis is more general than either of the earlier views. 



There is a further implication in these cases of hybrid gynandro- 

 morphs in bees that can now be cleared up. The female parts of the 

 gynandromorph are of hybrid origin. On any view, therefore, these 

 parts are expected to be not necessarily like the mother (unless her 

 character is the dominant one), but hybrid. If the mellijica color is 

 dominant, then on Boveri's views the female side of the gynandro- 

 morph should be mellijica, but according to Newell the Italian yellow 

 color is dominant, hence in half the Eugster gynandromorphs the male 

 and female sides should have the same color. Perhaps this accounts 

 for the astonishing failure on the part of von Siebold to mention the 

 color differences in his gynandromorphs, since the superficial (the 

 racial differences in color) color was often the same on the two sides. 

 If this is the real situation, Boveri must have worked with a deeper 

 color difference, one that is ordinarily not apparent. It is doubtful 

 from his description whether he could determine if the female parts 

 were mellijica or Italian or intermediate. He recognizes the difficulty, 

 for he refrains from making any comparison between the female parts 

 and those of the hybrid workers, but so far as he suggests any com- 

 parison it is with the pure mellijica type. 



In von Engelhardt's case the male parts are described as darker, 

 hence more like mellijica, while the female parts are described as 

 lighter. Since Quinn shows that the yellow (lighter) color is dominant, 

 the two sides should be different, hence the fact strongly supports 

 von Engelhardt's interpretation. In fact, I do not see how we can 

 avoid the conclusion that von Engelhardt's results are supported by 

 much better evidence than are Boveri's own, if any such comparison 

 must be made. Both are probably right, and the theory of chromo- 

 somal elimination not only accounts for both, but on that theory both 

 kinds of results are expected. 



If, as here suggested, both the Eugster and the von Engelhardt 

 gynandromorphs are due to chromosomal elimination, it follows that 

 there must have been also other gynandromorphs present that were 

 not color-hybrids, but show the dominant color both in male and 

 in female regions. In fact, these must have been as conmaon as the 



