262 bulletin: museum of compaeatiye zoology. 



of the crowding of the elements. Pigment usually obscures the view in 

 the distal part of the sensory cell as far as the nucleus. Proximal to 

 the nucleus (Fig. 41) there is a network, which passes into long meshes 

 toward the capsule, where it merges into the neurite. Finally, the 

 meshes grow so long and narrow that they cannot be distinguished from 

 separate, sinuous fibrils. Tangential sections of the retina fixed in vom 

 Bath's mixture do not yield good results, because it is difficult to distin- 

 guish the fibrillae from the pigment-granules. 



The nuclei of the sensory cells fixed in vom Rath's fluid have the ap- 

 pearance of large, spheroidal vesicles with a rather scanty network, for 

 the karyoplasm is not stained. The chromatic substance appears in the 

 form of small, dark spherules lodged in the linin substance. There is 

 ahva}-s a very large single nucleolus. 



Fibrillae by Bethe's Method. In the sensory cells stained by Bethe's 

 method the fibrils are more or less conspicuous, depending upon how 

 deeply they are stained and how far they are decolored. By over- 

 staining the entire cell is made blue, so that no network is visible ; 

 by excessive decoloring the blue of the fibrils, as well as that of the 

 general cytoplasm, disappears. As such sections were also stained with 

 orange-G, the fibrils, when evident, appear against a yellow background. 

 If the blue is all washed out the whole cell appears yellow. 



In the rods the fibrils exhibit practically the same arrangement as in 

 vom Bath preparations. There is, however, one difference, — the fibrillae 

 of the mantle are more fused into tufts than when treated by the vom 

 Rath method, because the fixation is inferior. On the other hand, 

 Bethe's method supplements vom Rath's where the latter is almost use- 

 less, namely, in the pigment zone of the retina. Cross sections of the 

 sensory cells in the pigment region distal to the nuclei exhibit conditions 

 resembling those shown in Figure 23 (Plate 3). The outlines of the pig- 

 ment cells are not well defined, but the sensory cells are evident as spots 

 free from pigment, though completely surrounded by it. Under high 

 magnification (Plate 4, Fig. 42) the cells are seen to contain a close net- 

 work. Each cross section contains several knots, which are connected 

 to one another and to the periphery by fibrils of smaller diameter. The 

 meshes of the network are finer than in the larger parts of the cell. 

 The network here more than in any other place which I have seen gives 

 the impression of being like a section through alveoli. The knots are 

 relatively so large that they might represent the end views of two or 

 more fused, longitudinal fibrils which run together in the region of the 

 conjunction of three or more alveoli; but such an interpretation is not 



