STUDIES ON INSECT SPERMATOGENESIS. 6l 



being apparently indeterminate in position. At the same time it 

 becomes evident that the vacuolated condition of the central mass 

 is likewise clearing up, the substance of the vacuole walls being 

 condensed into concentric shells oi chromophilic material marking 

 the limits between the former concentric rings of vacuoles, while 

 the substance of the vacuoles themselves flows together to form 

 more or less continuous shells of non-staining material alternat- 

 ing with the thinner shells of chromophilic material. This ar- 

 rangement is at first very incomplete and imperfect (Fig. 3), the 

 chromophilic partitions being more or less irregular. Something 

 of this intermediate period in the differentiation of the nebenkern 

 is also figured by other workers, particularly by \Yilke ('07 and 

 '13) in Hydrometra. (See his Fig. 70 in Hydromctra lacitstris 

 ('07) and especially Figs. 77 and 74 in Hydromctra paludiim 



At the period of Fig. 3, the shells of chromophilic substance 

 are still joined at many irregular intervals by cross partitions. 

 Presently, however, the pattern clears up considerably and be- 

 comes much more regular in its main outlines (Fig. 5). The con- 

 centric zones of material are now smoother and more nearly com- 

 plete, and the nebenkern as a whole exhibits a tendency to a 

 division of all the parts into two equal halves (Fig. 5). The gen- 

 eral arrangement recalls the appearance of a hemi-sected " onion " 

 -noted by so many students of insect spermiogenesis. Previous 

 to this time the pattern seems to have been susceptible to the dis- 

 integrative action of acetic acid, but with the stronger develop- 

 ment of the chromophilic plate-work, the pattern is very often 

 preserved after almost any fixative. This stage has thus become 

 the most familiar of the nebenkern "patterns." It was figured 

 by von la Valette St. George ('86a) in Blatta (Fig. 74) and later 

 described in detail by Henking ('91) in Pyrrhocoris. \Yilke in 

 Hydromctra lacttstris ('07) (Figs. 72 and 75), and in H. pa- 

 hidum ('13) (Fig. 78) has figured this stage with what one is 

 inclined to think a trifle too diagrammatic clearness. Attention 

 may also be called to the figures by Stevens ('05) of Staiofcl- 

 matus (Figs. 84 and 85) and Blatella (Fig. 151), by \Yassilieff 

 ('07) of Blatta (Fig. 56), and by Boring ('07) of Pccciloptera 



