INTERMUSCULAR NERVE CELLS OF EARTHWORM 163 



though the neurofibrils might branch frequently, they never 

 anastomose, while in each large cell a definite, dense network is 

 conspicuous. In my preparations (silver nitrate), on the other 

 hand, the difference in the arrangement of the neurofibrils, like 

 the difference in size, does not appear to be at all important. 

 Within all cells a definite network could be distinguished. (In 

 studying the neurofibrillar arrangement within the cells a bin- 

 ocular mon-objective microscope, equipped with a substage 

 condenser and an 1.8 fluorite immersion objective, was used.) 



Moreover, the density of the network in different regions of 

 individual cells was also found to vary. These differences in 

 density seemed to be due to two factors: the presence of large 

 'vacuoles,' and the position of the nucleus. The significance of 

 the 'vacuoles' is still a matter of conjecture (Kowalski, '09; 

 Boule, '13), but in the region of a 'vacuole' (fig. 3) the network 

 was usually sparse. The nuclei are spherical or ellipsoidal and 

 with rare exceptions are eccentric in position, being nearer the 

 proximal pole of the cell (figs. 2, 3, 4). The neurofibrillar net- 

 work in the distal pole of the cell appears coarse and dense, but 

 toward the proximal pole, in the vicinity of the nucleus, the den- 

 sity of the net is decreased and, as the cell narrows to give off 

 its proximal process, the meshes of the net become elongated 

 in the direction of the long axis of the cell (fig. 3). 



The eccentric nuclei of the spindle-shaped bipolar cells give to 

 these nerve elements a well-defined morphological polarity. 

 Just what relation there is between the structure of these cells 

 and the direction of conduction of impulses cannot be decided 

 until some conclusion is reached regarding the nature of these 

 cells; i.e., whether they are connected with the receptor or the 

 effector portion of the nervous system. Kowalski ('09) points out 

 that in the earthworm the neurofibrils which form an efferent 

 ('celluhfuge') process are not united into a net, but converge 

 without branching and fuse, giving rise to the single fiber; while 

 on the other hand, the afferent ('cellulipete') process on entering 

 the body of the cell is broken up into a number of branching 

 and anastomosing neurofibrils. Boule ('13) criticises this view 

 of Kowalski and figures the intermuscular nerve cells with 

 anastomosing neurofibrils in both poles. 



