The finer Structure of the Nerve Cells of Invertebrates. 39 



into the latter as well as into the main body of the cell (see Figs. 12, 

 13 and 16). 



They are always easily distinguished from the 

 spindles, by the manner in which they retain their 

 stain. In addition to this, they are most frequently spherical in 

 form and homogeneous in character. 



In Fig. 16, which has already been described, the pigment gra- 

 nules are found to be still deeply stained, even after the spindles 

 have been removed by the after-differentiation process. In view of 

 the above mentioned observations, I regard the spindle-shaped col- 

 lections of small chromophilous granules (spindles), as constant struc- 

 tures in these cells, and, in all probability, as homologous to the 

 "Körner" of German writers. 



So far, in the preceding pages, proof, concerning the existence of 

 chromophilous granules in the cell body, has formed the chief topic 

 of inquiry. From now on, we will consider the remaining structures 

 in the cell (nucleus excepted) which include the fibrils and ground 

 substance. 



The fibrillar Structures of the CeU. 



The question as to the existence of fibrils in the nerve cells of 

 Invertebrates, although in the past a disputed one, is at the present 

 writing, at least so far as the more recent investigators are concerned 

 (RoHDE, BiNET and Pflücke), practically settled — and on this subject 

 I have little to add to our present knowledge, except concerning their 

 mode of distribution in the cell. 



I have, however, experienced much more difficulty in clearly 

 bringing out these fibrils in the axis-cylinder processes of the nerve 

 cells of Helix and Arion^ than in those of any other Invertebrates 

 thus far examined by me (as Sycotypus, Cambarus, Astacus, Homarus, 

 Limulus and Lumhricus). 



I account for this difficulty, chiefly on the ground that these 

 fibrils, in Helix and Arion, possess a relatively much finer calibre than 

 those of other forms examined, and for this reason are only seen to 

 advantage in thin sections (3 f.i). 



In thicker sections (5 or 6 f.i), the axis-cylinder processes of 

 Helix and Arion frequently appear more or less homogeneous in struc- 

 ture. This is not the case with the axis-cylinder processes of other 

 forms examined by me, as, for instance, in Sycotypus. 



