72 ROBERT H. BOWEN. 



equivalent to the outer, chromophilic substance of the original 

 mitochondria; 1 but the history which I have traced shows clearly 

 that exactly the reverse is the case. So much for the dangers of 

 arguing on the basis of a single staining reaction. As a result of 

 this behavior toward crystal violet, the nebenkern appears vacu- 

 olated in various characteristic ways, the vacuoles being at first 

 scattered, and subsequently arranged in longitudinal rows that 

 seem to be fusing to form a non-staining axis fo>r the outer, in- 

 tensely staining substance. (See Bowen ('22) Figs. 59, 60 and 

 66.) 



The most interesting results, however, were obtained with 

 Cajal's formol-uranium nitrate method with which I had been 

 attempting the demonstration of the Golgi elements. This tech- 

 nique gave no certain results in the very early stages of the for- 

 mation of the new, or as we may conveniently refer to it, the 

 central, substance, but when the vacuoles had lined up to form 

 thread-like chains (about the stage of Fig. n), the silver was very 

 heavily reduced by the substance of the vacuoles themselves 

 which were thus brought out with the clearness of a silhouette on 

 a field of bright canary yellow (Fig. 10). The general fixation 

 of the spermatid is usually very poor, the nucleus appearing 

 merely as a light, circular area, often badly distorted. But the 

 nebenkern is well preserved and the beaded threads stand out 

 with remarkable clearness, especially in later stages. At this 

 early period they are not quite so clear on account of the mass of 

 chromophilic material which still occupies the center of the neben- 

 kern and is colored a nondescript brown. Whatever may have 

 been the origin of the substance of these threads, it is abundantly 

 clear that it represents now a material of specific and character- 

 istic chemical composition. I had thought at first (Bowen ('19)) 

 that it represented the original chromophobic material of the 

 chondriosomes, but it is now evident that such can not be the case. 



As the nebenkern draws out, the threads make for a time a 

 rather confused picture since the sections rarely cut them exactly 

 parallel to their length (Fig. 17 is a fair example). The cross- 

 sections of these stages are more interesting, the threads, now cut 



1 It should be remembered that the true chromophilic substance disappears 

 entirely at a relatively early stage, as described above. 



