\40 KANSAS UNIVERSITY QUARTERLY. 



together, and have relatively little cytoplasm. ( Fig. D.) The nuclei 

 are large, oval, smooth in contour, and poorer in chromatin than the 

 secondary variety (fig. D); and, ia all stages, the nuclear membrane 

 is represented by a single vesicle instead of by one for each chromo- 

 some, as in the secondary spermatogonia. 



In the testis of the adult insect, the primary spermatogonia are 

 comparatively rare, having for the most part already been transformed 

 iato secondary spermatogonia. In a few follicles, however, I have 

 been able to find these in larger or smaller numbers. Where they are 

 numerous, they lie around the axis of the follicle a short distance 

 from its blind end and sometimes also in a single layer at the ex- 

 tremity (fig. 1 ), indicating that development began, not at the center 

 of the follicle, as we might expect, but at the periphery next to the 

 follicular wall. 



Where the primary spermatogonia are few they lie in the inter- 

 spaces between spermatocysts, and some of them form the "Cysten- 

 kernen"' of St. George. It would hardly be surmised that the nuclei 

 of the cyst walls are formed from the same source as the germ-cells 

 themselves, but nevertheless there is strong evidence to show that 

 such is the case. 



In the first place, the primary sijermatogonia near the cyst walls can- 

 not be distinguished by form or staining qualities from those lying in 

 the interior of masses of their kind. In the second place, even as far up 

 the follicle as the zones of spermatocytes and spermatids, no flat nuclei 

 are to be seen in the cyst walls, as described by a number of authors. 

 Instead, in the angles between cysts, where they meet near the me- 

 dian line of the follicle, as well as where they are contiguous with 

 the follicular wall, may be seen cells with nuclei like that of fig. i^ 

 filling the spaces. (Fig. 41.) As I have said, these are the only 

 nuclei of the cyst walls in this portion of the testis. Now in some 

 cases, where masses of primary spermatogonia are found in the blind 

 end of the follicles, each nucleus will be seen to be exactly similar in 

 shape, staining reaction and arrangement of chromatin to those just 

 mentioned. The latter similarity, of course, is found only when the 

 cells are in the resting condition, for, immediately on the commence- 

 ment of preparations for division, this likeness in arrangement of 

 chromatic sulDstance is lost. 



Where masses of primary spermatogonia in the resting stage or 

 early prophases are found, they appear like a quantity of nuclei, 

 sometimes almost devoid of cytoplasm, enveloped in a matrix of 

 stroma. On the edges of these masses, where they come in contact 

 with cysts of secondary spermatogonia, the cyst walls seem to be 

 continuous with the cell-walls or envelojiing stroma of the nearest 



