The sparmatogenesîs of Peripatus (Peripatopsis) balfouri. 325 



continuous throughout the nucleus, but only portions of a possible 

 spireni. But it must be remembered that these linin elements are 

 delicate structures which can be clearly followed only when they lie 

 exactly in the plane of the section, whereas the chromosomes are 

 distributed very irregularly through the nuclear cavity. One could 

 see the supposed spirem in its total length only if a view could be 

 had of all the chromosomes lying lengthwise in one plane, but they 

 never all lie in such a position. To test the correctness of my con- 

 clusion of a continuous linin spirem in this stage, I have re-examined 

 my old preparations of Pentatoma : in this form the number of chromo- 

 somes is smaller (only 7) at this stage, and in so far ought to show 

 the linin spirem more clearly than Peripatus does. In some of the 

 preparations of Pentatoma, at about the beginning of the monaster 

 stage, I found in several cases several chromosomes arranged along 

 a continuous thread ; and in one case all 7 chromosomes together 

 with the chromatin nucleolus (the metamorphosed chromosome) were 

 connected together by a single continuous linin thread. In another 

 communication I hope to publish my more recent studies on the 

 spermatogenesis in the Hemiptera, and so have reproduced none 

 of my figures of Pentatoma in this paper. This relation in Pentatoma, 

 then, seems to be fully corroborative of the conclusion of a single 

 linin spirem in the 1st maturation prophases of Peripatus. 



Arguing from these conclusions, we may say that any comparatively 

 thick linin fibre which joins one bivalent chromosomes with another, 

 joins distal ends of bivalent chromosome, and is the distal linin fibre 

 of the earlier stages. It is the point of attachment of such fibres 

 which serves to mark the distal ends of the two univalent chromo- 

 somes which form a bivalent one, and which accordingly, in definitive 

 chromosomes of ring form, show where these distal ends lie, and show 

 further how the axes of such ring chromosomes come to lie in the 

 equator of the spindle. 



Besides these thicker linin elements, which apparently constitute 

 together a linin spirem, may be seen in these prophases much more 

 delicate fibrils (Figs. 135, 137, 138, 144, 147, 148, 155, 160, 161), 

 which connect chromosomes together and with the nuclear membrane, 

 and which prol)ably are the same as the delicate fibrils of the synapsis 

 and telophase: these are for the most part just on the verge of 

 visibility, they are not attached with any regularity to particular 

 points on the surface of the chromosomes, and are mainly of short 

 length. 



