56 ROBERT H. BOWEN. 



granules) mitochondria into a compact, spherical chondriosome 

 body. In the large spermatids this body usually appears, imme- 

 diately after formation, of rather homogeneous texture, staining 

 faintly or not at all with crystal violet. In the small spermatids 

 the nebenkern typically stains an intense violet (after Benda). 

 Thus, if there is any structural differentiation of the nebenkern in 

 this initial stage it is lost in one case by failure to stain and in the 

 other by over-staining. Subsequent phenomena indicate, how- 

 ever, that there must be at this time some differentiation of the 

 mitochondria! substance, and it will first be necessary to examine 

 carefully the evidence bearing on this point. 



As other observers have remarked, the mitochondria in the 

 spermatocytes at the time of the maturation divisions appear 

 characteristically in the form either of granules or of rods and 

 threads of various dimensions. As Meves ('oo) was the first to 

 show (in Pygcsra), the granular mitochondria are not homoge- 

 neous spheres, but are actually vesicles the substance of which is 

 differentiated into two distinct parts an intensely staining enve- 

 lope or peripheral layer and a clear, non-staining interior. This 

 structure is very clearly demonstrated in many other Lepidoptera, 

 as recently shown by Gatenby ('17), who has applied the conveni- 

 ent descriptive adjective chromophilic and chromophobe (or 

 better, chroma phobic), to the staining and non-staining portions 

 respectively of the chondriosome vesicles. The same vesicular 

 structure has also been shown to occur in some forms other than 

 insects (e.g., Mollusca), so that it would appear reasonable to ex- 

 pect a similar duplex structure in mitochondria of the rod type. 

 The evidence here is not entirely convincing, due largely to the 

 difficulties of technique. Meves ('07) has, however, noted that in 

 the thread-like mitochondria of the honey-bee, there is a differ- 

 entiation into an outer, chromophilic envelope with a core of non- 

 staining material. Gatenby ('18) has recently stated that in the 

 beetle (Tcnebrio) the mitochondria occur either as "spheres" or 

 as " elongated tubes, containing internally a chromophobic sub- 

 stance." In the Hemiptera, I have never been able to satisfy 

 myself of this duplex structure. The mitochondria are in the 

 form of long, delicate threads (Bowen, '20) which are so fine that 



