22 THE ORIGIN OF GYNANDROMORPHS. 



to belong to this category are also explained by the assumption that 

 the egg started with two nuclei, and in the description of cases both 

 of these views are given as alternatives. 



THE ORIGIN OF THE GERM-CELLS IN FLIES. 



In several species of flies {Miastor, Chironomus, CalUphora) it is 

 known that the germ-cells of the ovary or testis arise from a single 

 cell at an early stage in the cleavage. In Miastor, for instance, when 

 the four first-formed nuclei divide, one of the eight daughter nuclei 

 moves to one pole of the egg, where it becomes surrounded by the 

 peculiar protoplasm of this pole and subsequently pinches off from the 

 surface of the egg. From this single cell by later division arise all 

 of the germ-cells. A similar process has been described for other 

 species of flies. If this holds also for Drosophila it follows that all of 

 the germ-cells must be either eggs or sperm, regardless of whether 

 the somatic parts are male or female. On the other hand, if the germ- 

 cells in Drosophila and in the bee are formed as in some of the other 

 insects, i. e., in the beetle Calligrapha described by Hegner, where 16 

 cells simultaneously reach the polar field, it would be possible for some 

 of the cells to have descended from one of the first two segmentation 

 nuclei and some from the other. In such a case, if the first-division 

 figure underwent elimination, both ovaries and testes might appear 

 in the same individual. In butterflies and moths, where many gynan- 

 dromorphs have been dissected, several cases in which both testes and 

 ovaries occur are known. This is also the case in bees. A difference 

 in the time of isolation of the germ-cells in these groups and in Dro- 

 sophila may account for the difference in the results. 



COURTSHIP OF GYNANDROMORPHS. 



Sturtevant's paper on sex recognition and sexual selection in Dro- 

 sophila gives a full account of the rather elaborate courtship of this 

 fly, in which the behavior of the two sexes is quite different. The re- 

 actions of an animal, male on one side female on the other, or of one 

 that had a female head and a male abdomen, might be expected to 

 furnish interesting conclusions as to the relative importance of the 

 sense-organs versus the reproductive organs in the behavior during 

 courtship. 



Sturtevant tested 6 gynandromorphs. One was male throughout, ex- 

 cept the genitalia, which were female. It behaved as a male. Sections 

 of the abdomen showed one abnormal egg present. Another had 2 sex- 

 combs, right and left, and the right wing was shorter than the left. The 

 abdomen was female. She produced at least 1 egg. Sections of the 

 abdomen showed 2 large eggs and a degenerate ovary present. She 

 courted and was courted, thus giving both reactions. A third was 



