6O ROBERT H. BOWEN. 



up very rapidly, the separate vacuoles fusing together to form 

 large clear spaces separated from each other by septa which pass 

 outward from the central mass to the exterior of the nebenkern, 

 which is marked by a very definite membrane. The whole thing 

 rather recalls the figure given by Meves ('oo) of Pygcera, though 

 no real comparison is, of course, possible. 



It is probably the stages between Figs, i and 2 that have often 

 been noted heretofore and likened to a " blackberry." It is clear 

 that if, as often^happens in the small-cell generations, the neben- 

 kern as a whole took the stain rather heavily, its appearance 

 would be something like that of a spherical mass of soap-bubbles 

 or a blackberry. The many bizarre figures of this condition 

 would seem to indicate that fixation effects may play a role in 

 determining the exact appearance of the final result. That the 

 main features of this stage are real, is indicated, however, by its 

 many reproductions in the literature of insect spermatogenesis. 

 Thus, Henking's ('91) Fig. 63 and Paulmier's ('99) Fig. 43 are 

 excellent reproductions of the " blackberry " stage in Pyrrho- 

 coris and Anasa respectively, very similar to those which I have 

 seen in Murgantia. The early stages in the development of the 

 vesicles as seen in optical cross-section are clearly figured by 

 Voinov ('03) in Cybistcr (Fig. 50), by Zweiger ('07) in Forfic- 

 ula (Fig. 37), and by Shaffer ('17) in Passahts (Fig. 19). 

 Later stages in the condensation of the peripheral layer of vacu- 

 oles are figured by Holmgren ('02) in Silpha (Fig. 9/), by Gross 

 ('07) in Pyrrhocoris (Fig. 98) by Stevens -('05) in Blatella (Fig. 

 150), and (probably) by Doncaster and Camion ('20) in Pedicu- 

 lus (Figs. 20 and 21). However, it is not improbable that some 

 of these cases are artifacts similar to those figured by Meves 

 ('oo) in Pygcera as mentioned above. 



The clearing up of the outer layer of vacuoles now proceeds 

 rapidly, the substance of the vacuole walls being apparently with- 

 drawn into the underlying region of the nebenkern, and eventu- 

 ally the whole peripheral zone appears as a clear, non-staining en- 

 velope enclosing a central, chromophilic core (Fig. 3). Usually 

 a connection or two is still retained between the external mem- 

 brane of the nebenkern and the central portion, these connections 



