264 bulletin: museum of comparative zoology. 



pletely drawn (the one at the left) is the neurite process, which turns 

 abruptly away from the observer. It thus has such a depth and such 

 a complication of fibrils that they cannot be represented. I have tried 

 to work out the network only in the thinner parts of the cell. Close 

 around the deeply-staining nucleus on one side (the upper in the 

 figure) there remains a bluish mass, which, owing to the fact that the 

 excess of stain has not been washed out, hides the fibrils in that region. 

 Having presented already typical sketches of fibrillae by two methods 

 of treatment, I discuss this figure (Fig. 37) simply for comparison. 

 The network has all the appearance of having been a structural condi- 

 tion of the cell previous to the application of the ammonium molybdate. 

 It is hardly credible that accidental fusions could produce such a per- 

 sistent appearance of net-knots. A casual glance at the figure may 

 suggest the appearance of alveoli, particularly to the left of the nucleus. 

 But with a staining as intense as methylen blue gives, one should be 

 able to see those faces of alveoli which lie parallel to the plane of the 

 section. Such faces, or blue areas, are not present in this preparation, 

 and the trend of the blue lines is for the most part in a longitudinal 

 direction. 



The relation of the parallel fibrils in the distal part of the cell to the 

 network in the region of the nucleus is shown by the fortunate bit of 

 selective staining illustrated in Figure 38. The top of the figure is 

 directed toward the rod-zone. At the bottom there are four knots which 

 are tied to one another by short cross-fibrils. From the internal or cen- 

 tral knot there branch distally two fibrils ; one rises toward the eye of 

 the observer and is cut off, the other extends well out in the direction 

 of the rod. In other directions this knot gives rise to three fibrils which 

 pass to as many knots at different levels laterally. A knot on either 

 side of the first sends fibrils toward the rod. Thus we have clear evi- 

 dence that the axial fibrils in the core of the rod are in communication 

 with a network in the vicinity of the nucleus. The fibrils are so few and 

 distinct in this example that they could not be confused with the edges 

 of alveoli. The fibrils clearly make a few closed meshes before they 

 give off parallel fibrils to the rod, and the knots are not mere branching 

 places of a distally branching system. 



The fibrils in the optic nerve are not so clear. In almost all cases the 

 whole fibre stains so intensely that no hint of a fibrillar condition is 

 obtained. In one of my preparations the staining seems more favor- 

 able ; but even here the evidence is too scanty. Figure 47 shoAvs a few 

 pieces of selected neurites from the same region as that shown in Figure 



