History of a transient nervous apparatus in certain Ichthyopsida. 341 



substance of Schwann" it must be exceedingly slight in amount, for 

 they are only "browned" and not blackened with osmic acid. 



Figs. 22 (PI. 22) and 24 (PI. 23) furnish in detail rather different 

 appearances than those above described. Of the two Fig. 22 shows 

 only a part of what is to be seen in the row of sections. In it the 

 spinal cord, with a number of transient ganglion-cells, is figured. 

 On each side is a portion of a nerve whose full course, as far as 

 determined, is given again in Fig. 24, PI. 23. This latter, and the 

 mode of drawing it, may now be described. The section in which 

 the ganglion-cells at the root of the nerve appeared was first drawn, 

 and then from the sections in front and behind (in this instance it 

 happened that all were posterior to that drawn) the details of the 

 two nerves were filled in section by section. In this way long 

 stretches of the two nerves could be completely formed anew. These 

 particular nerves have not quite the actual extension given to them 

 in the figure, or rather, to speak more precisely, they could not be 

 followed so far ventralwards. But in the same row of sections there 

 are other similar nerves, the details of whose extension, when filled 

 into the figure, carried them as far as shown. It is unlikely that 

 they terminated even there. This course, by the way, was adopted, 

 not because it made any difference in the actual composition of each 

 nerve, for each of the original nerves could be followed far beyond 

 the point where the "ganglionic nodes" occur, but to avoid unneces- 

 sary multiplication of figures. The figure (Fig. 24) shows in section 

 spinal cord, notochord, myotomes, somatopleure etc. etc. It depicts 

 in more diagrammatic fashion the central cells and the two ganglion- 

 ated nerve-roots seen in Fig. 22, To the right it shows the gang- 

 lionated root of the nerve of that side, passing into a nerve formed 

 by a plexiform arrangement of the root ganglion-cells with one near 

 the tip of the myotome (see Fig. 22, gl). Here the nerve ends 

 with, hardly in, a ganglion -cell {gl.c^). To this ganglion-cell a similar 

 one (gl.c^) is applied, and the latter gives off the nerve-fibre (s. w), 

 which passes down under the epiblast in the way shown in the figure. 

 And ends where? Probably on the yolk-sac. On the left side the 

 nerve starts in the same way from two ganglion-cells — or perhaps 

 it may be more correct to describe it as starting at the ganglion- 

 cell gl.c^ — and passes upwards to end by being applied to the 

 two "root ganglion-cells". The cell glc^ is itself very closely ap- 

 plied to another cell gl. c *, this could only be made out with the 



23* 



