28 THOS. H. MONTGOMERY jr., 



In the spermatogonic mitoses the pole and central spindle fibres are 

 probably of cytoplasmic origin, as has been already shown. But since 

 in the following anaphase the mantle fibres appear still attached to the 

 chromosomes within the nuclear cavity, and after the nuclear membrane 

 has been formed, it seems justifiable to conclude, that since they become 

 to great extent enclosed within the nucleus during the anaphase, 

 therefore they must be to great extent produced from the linin of 

 the nucleus in the preceding prophases. Each chromosome in the 

 anaphase represents a thread of linin in which are imbedded, or to 

 the surface of which are apposed , chromatin microsomes ; the linin 

 thread may be observed at places where the chromosome is constricted. 

 In the spermatogonic metaphase it was shown that the connective 

 fibres of the spindle are formed from the stretching out of the axial 

 linin thread of each chromosome, during the division of the latter. 

 Hence those occasional fibres, which, in the anaphase, are found 

 connecting chromosomes with points on the periphery of the nuclear 

 membrane (other than that point where the idiozome is situated), 

 may be considered as remnants of such connective fibres. The daughter 

 cells of the last division of spermatogonia remain so closely apposed, 

 that it is difficult to determine whether connective fibres and central 

 spindle fibres still pass from cell to cell ; but in view of the relations 

 in other forms, this might well be the case. In the synapsis of 

 Pentatoma there are only very few achromatic fibres passing from 

 the chromosomes to the nuclear membrane, whereas in Elasmobranchs 

 (Moore, '95) such fibres are quite numerous. In Pentatoma it is the 

 rule that a few fibres, or a single fibre, pass to that point on the 

 nuclear membrane where the true nucleolus first appears ; I can suggest 

 no explanation for this phenomenon, unless it be that it is the presence 

 of the fibre at this point which causes the nucleolus to be formed 

 there. 



The postsynapsis. 



The postsynapsis, a term introduced here for the first time, is 

 a well marked stage of the anaphase, distinguishable alike from the 

 preceding synapsis as from the following telophase. 



The chromosomes, previously massed together densely, become 

 gradually disentangled from one another, and finally lie in the nuclear 

 cavity as long, bent loops. At the commencement of this stage 

 (Figs. 76, 77) the chromosomes begin to separate, so that meshes 

 appear between tUem , and free ends of some of them may be 



