GROWTH AND DIFFERENTIATION OF THE GERM-CELLS 123 



middle-piece, and the point — are likewise of cytoplasmic origin ; and 

 it is certain that the middle-piece is in some cases derived from the 

 attraction-sphere of the spermatid, and contains the centrosome. 



As the spermatid develops into the spermatozoon it assumes an 

 elongate form, the nucleus lying at one end while the cytoplasm is 

 drawn out to form the flagellum at the opposite end. The origin of 

 the axial filament is still in doubt. Many authors (for example, 

 Flemming and Niessing) have described it as growing out from the 

 nucleus; but more recent work by Hermann, Moore, and others, 

 shows that this is probably an error and that the axial filament is 

 derived from the substance of the attraction-sphere. 



The greatest uncertainty relates to the origin of the middle-piece 

 and the apex. By one set of authors the centrosome is believed to 

 pass into the point of the spermatozoon (Platner, Field, Benda, Pre- 

 nant) ; by another set, into the middle-piece (Hermann, Wilcox, Cal- 

 kins). That the latter is a correct view is absolutely demonstrated 

 by the fact that during fertilization the centrosome in every accu- 

 rately known case is derived from the middle-piece (amphibia, echino- 

 derms, tunicates, earthworm, insects, mollusks, etc.). The observations 

 of Platner and others in support of the other view are, however, too 

 detailed to be rejected on this ground alone, and it is not impossible 

 that the position of the centrosome may vary in different forms. The 

 uncertainty is due to the difficulty of tracing out the fate of the cen- 

 trosome and archoplasmic structures of the spermatid. It is certain 

 that each spermatid receives a centrosome or attraction-sphere from 

 the preceding amphiaster. But besides the centrosome (attraction- 

 sphere) the spermatid may also contain a second "achromatic" body 

 known as the paranucleus (Nebenkern) or initosome, which has un- 

 doubtedly been mistaken for the attraction-sphere in some cases ^ and 

 to this circumstance the existing confusion may be in part due. The 

 concurrent results of La Valette St. George, Platner, and several 

 others have shown that the " Nebenkern " is derived from the re- 

 mains of the spindle-fibres ; but the most divergent accounts of its 

 later history have been given by different investigators. According 

 to Platner's studies on the butterfly Pygcera ('89), it consists of a larger 

 posterior and a smaller anterior body, which he calls respectively the 

 large and small niitosonia (Fig. 62, C). The former gives rise to the 

 investment of the axial filament of the tail, the latter to the middle- 

 piece, while the " centrosome " lies at the anterior end of the nucleus 

 at the " apex" (Fig. 62, D). Field ('95) reaches an essentially similar 

 result in the echinoderm spermatozoon, the single " Nebenkern " 

 forming the middle-piece, while the "centrosome" lies at the tip 



^ Compare the confusion between yolk-nucleus and attraction-sphere in the ovum, p. 119. 



