The spermatogenesis of Peripatus (Peripatopsis) balfouri. 307 



the central pole of the nuclear membrane. But at least the majority 

 of the linin fibrils of which we are speaking must have had some 

 différent source of origin, and this could have been only an origin 

 from the linin forming the spirem , or an origin de novo. Thus 

 they may be looked upon as lateral outgrowths of the linin spirem, 

 or as differentiations of the karyolymph. One coincidence may help 

 in determining which of these modes of formation is the true one: 

 when they first appear is the time of the longitudinal splitting of the 

 chromosomes; and where they are seen most clearly, i. e. when they 

 are thickest, is about the late synapsis or telophase when this split 

 is most pronounced. Might this not show that it is by the agency of 

 a contraction on the part of these fibrils , that the splitting of the 

 chromosomes is etfected? This would certainly seem to explain why 

 these fibrils are thickest when the longitudinal split is most pro- 

 nounced, i. e. when the daughter chromatin granules have most fully 

 separated. V\e shall return in the General Part of this paper to the 

 discussion of the agency of the linin structures. 



The nucleoli (« Figs. 64, 65, 67, 71, 72, 80) appear about the 

 commencement of the synapsis stage as oval or flattened bodies close 

 to the inner surface of the nuclear membrane, almost always at that 

 region (the distal pole) nearest to that portion of the cytoplasm con- 

 taining the greatest amount of yolk substance. This would be a cor- 

 roboration of my conclusion (1899) that in many cases the nucleolar 

 substance seems to be genetically connected with yolk substance. Sub- 

 sequently the nucleoli assume a more rounded form. 



During the synapsis the cytoplasm gradually increases in amount, 

 as does the yolk substance (Figs. 64, 65, 69—71, 74, 75, 77, 80, 82). 

 The greatest mass of cytoplasm and yolk lies at the distal pole of 

 the cell, where the persisting Zwischenkörper plate is still found; 

 but around the central pole of the nucleus it forms only a very thin 

 layer containing but little yolk substance. 



3. Telophase ot the Spermatocytes. 



Under "telophase" (Figs. 91—97) is here understood the stage 

 which intervenes between the synapsis and the rest stage. In Peri- 

 patus there is no special 'post-synapsis" stage, such as was described 

 by me for Pentatoma, i. e. no stage of long and slender chromosomes 

 intervening between the synapsis and telophase. 



The synapsis passes gradually into the telophase, and here, as 

 with most other periods of mitosis, it is practically impossible to 



20* 



