338 IJ. M. BKIiNAllI). 



simplest way of arranging large masses of chromatin in a 

 long fusiform nucleus.^ That this is the true explanation we 

 learn from fig. 20 a, which represents a rod-nucleus of a 

 kitten (two days old), that is, at an earlier developmental 

 stage than that shown in fig. 20 }> ; Avhile fig. 20 f? shows 

 a nucleus on its way from the middle layer to become a rod- 

 nucleus with the chromatin in it commencing to form clumps. 



Continuing our argument, the large supply of chromatin 

 maintained in these nuclei, in spite of what they are passing 

 on to the rods, must either be manufactured in situ or else 

 come from without. 



The latter is the more probable view, because of the 

 extreme distal position of these rod-nuclei, viz, in the front line 

 of functional activity and furthest from the centres of 

 assimilation. But fortunately we need assume nothing, for 

 figs. 18 a—d show the protomitomic filaments thickly clotted 

 with staining substance, which we can safely assume to be 

 travelling from the nuclei of the middle layer into the rod- 

 nuclei. 



In this connection I believe we can find the explanation of 

 a very common phenomenon in the retina. Fig. 18 a shows 

 two clumps of staining matter, one of which is very large, 

 and fig. 18 6 one clump, outside the nuclei, and clearly them- 

 selves not nuclei. In Part V, I called attention to certain 

 clumps of matter in or on the outer reticular layer which are 

 frequently seen in chambers formed by the ends of what are 

 called the rod-fibres. The resemblance of these clumps within 

 chambers to small nuclei in cells has, indeed, suggested to one 

 observer (Borysiekiewitz) the presence of anew row of "cells" 

 in the retina (cf. fig. 18 6). The very large mass in fig. 18 a 

 is on the horizontal threads of the outer i-eticular layer, but 

 in fig. 18 6 a mass is shown Avithin a chamber of the retina. 

 These, as stated, are certainly not nuclei ; they are contained 

 in no vesicles with sharp contours like the membranes of 



^ Oil this subject compare a discussion in vols, xv and xvi of tlie ' Anat. 

 Anzeigcr.' My own observations arc not quite in accord with Dr. Schaper's, 

 for in very young cats the rod-nuclei have distinct networks (see fig. 20 a). 



