312 THOMAS H. M< >N I '< .< >M KRY, JR. 



in the region of the centriole and the rlagellum. \Verc the his- 

 tory of the spermatozoa unknown beyond their conditions in the 

 vasa deferentia, there would be no evidence of the fate of the 

 cytoplasmic lobes and chondriosomes. But fortunately I have 

 numerous preparations of oviducts from female individuals, 

 all crowded with spermatozoa, and in these all the spermatozoa 

 lack entirely the cytoplasmic lobes and chondriosomes (Figs. 23, 

 24); in not a single case was a cytoplasmic lobe observed upon 

 a spermatozoon when within an oviduct. 



Peripatus, ^accordingly, has for us more than a phylogenetic 

 interest, it has a high cytological importance. The sperma- 

 tozoon during its development casts off its cytoplasm, and evi- 

 dently all of it. But this abstriction of the cytoplasm, or a 

 portion of it, is now known to be a quite general phenomenon 

 in animals, and only amphibians and certain insects appear 

 to furnish exceptions to it. Much more important is that all 

 the mitochondrial substance, in the form of a compact chondrio- 

 some, is cast away with it. Further, I had previously described 

 the spermatozoon as possessing a lance or perforatorium, staining 

 differently from the nucleus. Now I can demonstrate that this 

 supposed perforatorium stains differently only on account of its 

 excessive tenuity, that it is only the narrowed proximal end 

 of the nucleus, and that it has no connection with the sphere. 

 We have seen that the sphere arises just behind the nucleus 

 (Figs. 2, 3), and moves forward into the cytoplasmic lobe (Figs. 

 5, 7, 8, 10, 12, 14, 16-19). When the chondriosome is fully 

 developed the sphere lies still in the cytoplasmic lobe, separated 

 from the nucleus (Fig. 20), and no evidence was observed that 

 it moves along the latter to constitute a perforatorium. There- 

 fore it is certain that the sphere as well as the chondriosome 

 becomes thrown off with the cytoplasm. 



The history of the centriole was not followed in detail. In the 

 telophasc of the secondary spermatocytes a minute centriole is 

 present at each pole (Fig. i). At the next stage when it was 

 noticed (Fig. 3) it appeared as a much more voluminous body a I 

 the distal pole of the nucleus, and it retains this position there- 

 after. Later it becomes discoidal with indication of subdivision 

 into two parts (Figs. 4-10), and afterwards lengthened in tin- axis 



