106 A. B. DAWSON 



of the earthworm. It is difficult to decide whether this is the 

 normal distribution or whether only the cells in this position were 

 stained, since the preparations were not obtained by intracoelomic 

 injections, but by partially immersing the live animal in the stain- 

 ing solution. The dorsal portion of the animal accordingly was 

 not in direct contact with the stain. 



The bases of the cells were found to lie at the level of the nerve 

 ring, while the long slender cell bodies extended outward through 

 the muscle layer and pierced the epidermis (fig, 6). The intra- 

 epidermal portion, lying between the columnar cells, was repre- 

 sented only by a very fine process which in no case could be 

 positively traced further than one-half the way to the cuticular 

 surface. One main fiber and one or two accessory fibers usually 

 arose from the bases of the cells. The main fiber in one case was 

 traced through five successive sections, each 20 m thick. It was 

 found to enter the lateral nerve trunk, to pass into the ventral 

 nerve cord, and, branching slightly, to terminate on its own side 

 in a dorsal position. The accessory processes could rarely be 

 traced among the fibers of the nerve ring, but in one cell a short 

 process was seen to branch slightly and end at the surface of the 

 longitudinal muscle layer (fig. 6). 



In general appearance and structure the slender nerve cells dif- 

 fer in several respects from the other intermuscular cells as seen 

 in methylene-blue preparations, but exhibit several features 

 which are strikingly like those observed in epidermal sense cells 

 (figs. 6, 7). Their nuclei are oval or ellipsoidal, contain no well- 

 defined nucleolus, and stain densely. Their bodies stain less 

 heavily, but take the color uniformly. The other intermuscular 

 nerve cells and the cells of the ventral cord, on the other hand, 

 have spherical nuclei, delicate chromatin networks, definite nucle- 

 oli, and well-defined neurofibrillar networks. 



SIGNIFICANCE OF THE INTERMUSCULAR NERVE CELLS 



The question of the significance of the intermuscular nerve 

 cells is a difficult one to discuss. Schneider ('02, p. 392) sug- 

 gested that these scattered cells were possibly sensory in nature. 



