34O ROBERT H. BOWEN. 



is characteristically developed within the clear vesicle (Figs. 25 

 and 27), but this is not visible as a rule in Golgi preparations. 



The spermatid is now ready to proceed with the general proc- 

 esses of differentiation. The acrosome shrinks somewhat, ap- 

 pearing to express some of its contents into the acroblast, and 

 begins to move around the nucleus toward the anterior end (Fig. 

 30) . As it does so its whole appearance changes ; it becomes 

 more opaque and begins to stain darkly with Fe-hematoxylin. 

 Simultaneously the acroblast begins to draw away toward the tail 

 region, and soon the connection with the remaining portion is 

 severed (Fig. 31). The latter continues its journey around the 

 nucleus, and becomes the acrosome while the former is carried 

 back along the tail. The term acroblast is scarcely applicable to 

 this cast off remnant (Fig. 31), nor is it any longer to be iden- 

 tified with certainty as Golgi apparatus, and I shall therefore 

 refer to it merely as the " Golgi remnant" 1 



For a while the Golgi remnant can be followed as it moves 

 back along the tail, but it soon becomes lost in a group of gran- 

 ules from which it can not be certainly distinguished, and which 

 are doubtless eventually cast out of the sperm with the proto- 

 plasmic ball sloughed off from the tail. It is practically certain 

 that the Golgi remnant is cast off along with the other debris 

 at least I have never been able to find any traces of it in the last 

 stage. Montgomery's ( ! n) statement that no protoplasm is 

 eliminated from the sperm is certainly mistaken, for I have fol- 

 lowed every step in the casting off of the protoplasmic balls and 

 have ascertained that they are eventually ingested by the epi- 

 thelial (nurse) cells which line the cavity of each cyst. 



CRITICAL CONSIDERATIONS. 

 I. Problems of Cytoplasmic Division. 



As yet our knowledge of exact processes in the distribution of 

 the formed elements of the cytoplasm in cell division is exceed- 

 ingly limited and we are, indeed, accustomed to call attention to 



1 This cast off body is homologous with the idiophthartosome of Papani- 

 colaou and Stockard ('18) and the idiozomrest, etc., of many authors. We 

 are really much in need of a convenient and purely descriptive term for this 

 body. 



