236 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
could be traced to the outermost ends of the sense-cells. Such free 
terminations are shown in Figures 28, 33, 34, 36 (Plate 5). 
The most convincing evidence that the nerve fibres do not penetrate 
to the interior of the sense cells, but pass between them, is afforded by 
sections which lie in a plane parallel to the surface of the epithelium, 
or nearly so. Figures 39 (Plate 5) and 44, 46-53 (Plate 6) represent 
such sections. In Figures 44, 48 (Plate 6) the cut ends of nerve fibres 
are shown. Such cut ends are always found to lie between the sense 
cells and not within them. Sometimes they are in close contact with 
the sense cells, and sometimes they are in the matrix of intercellular 
material by which the sense cells are surrounded. In Figures 39 (Plate 
5) and 48, 49, 53 (Plate 6) fibres are represented which make their 
way between the cells of the epithelium. ‘These conditions are typical, 
and are in agreement with those seen in a large number of sections. 
I have been able to find only two sense cells which appeared to be 
exceptions to the statement that the sense cells do not contain nervous 
material. These cells are shown in Figure 53 (Plate 6). In one of 
them two fine filaments were discernible, as represented in the figure, 
and in the other a ring-shaped body was present, which was connected 
to an extracellular fibre. This latter structure at once calls to mind 
the intracellular rings which Bielschowsky und Brihl have figured 
for the sense cells of mammals. Inasmuch as I have seen only one 
example of such a structure, however, I am inclined to regard its 
occurrence as the result of come accident, and to consider it as without 
significance. 
In the series of sections which I have just been describing, which is 
the single series in which the nerve fibres of the sacculus and lagena 
appear to be completely impregnated and the non-nervous matter is 
at the same time in good condition, there are many places where 
delicate fibres may be seen in close proximity to the surfaces of the cells 
(Figs. 39, Plate 5; 46-48, 50-53, Plate 6). Not infrequently a fibre 
divides at the surface of the cell, sending a branch to either side (Figs. 
48, 50, 51, 53, Plate 6). One portion of the series, including a con- 
siderable number of sections, when examined by the aid of an ordinary 
high power objective, gave the impression that the sense cells are 
completely surrounded in a meshwork of anastomosing fibres, which 
mark the cell boundaries. When this material was examined under 
the oil immersion, however, it was found that these fibres, like those of 
the so-called subepithelial plexus of the crista acustica shown in Figure 
13 (Plate 3), do not anastomose, but intertwine (Fig. 52, Plate 6). 
