Gynandromorphous Lepidoptera. 327 



present in Amorpha, is very narrow, being another example 

 of an unpaired organ reduced in size because it is developed 

 from half the usual number of cells. From the other side 

 of the saccus runs a small piece of cingula, an example of 

 partial reduplication of what is really a paired structure, 

 which, like the uncus, has become fused. Two valves are 

 present, though not quite normal in shape, and the penis 

 is more slender than usual. The female half is the more 

 interesting. 



In the normal gaea there is a most elaborate ostium 

 bursae, guarded by two folds of chitin, covered with coarse, 

 curved hairs, and in the ventral wall of the bursa itself is 

 a dense mass of chitin with two wings covered with short, 

 thick spines, a structure corresponding to the signum of 

 the geometrid bursa. 



In the gynandromorph the ostium has only one of the 

 chitinous folds present, that lying on the left or female 

 side. 



The whole bursa is small in size, and only the left wing 

 of the signum is present (PL CIV). 



This is the best proof I have met with that the small 

 bursa of the gynandromorph represents a half bursa. The 

 primitive cell mass, half the usual size, could complete the 

 tubular neck and saccular body of the organ, but could not 

 develop more than one half of such an elaborate structure 

 as the chitinous fold guarding the ostium or the signum 

 in gaea. 



Theory of Origin of Halved Gynandromorphs. 



When I published my paper in the Journal of Genetics 

 in 1915 I was unaware of the earlier writings of Morgan on 

 this subject. In his first paper, published in 1905, " An 

 alternative interpretation of Gynandromorphous Insects," 

 he suggests that two spermatozoa may enter the same egg, 

 and that one of these develops without any fusion of ita 

 nucleus with the nucleus of the egg, whereas the other 

 fuses in the usual way. The independently developing 

 spermatozoon derived solely from the male parent pro^ 

 duces the male part of the perfect insect. The other, com- 

 bined with the ovum, produces the female part, which is 

 thus derived from both parents. 



The theory is alternative to that put forward by Boveri, 

 who thought that there was a premature division of the 

 egg nucleus, one half of which combined with the whole 



