MULLENIX: EIGHTH CRANIAL NERVE. 231 
largely occupied by the terminal expansions of axis cylinders, desig- 
nated by Retzius and other investigators as “‘Kelchbildungen.”” Such 
groups of terminal structures are shown in Figures 12, 14, 16 (Plate 3). 
There are other places, however, in which there is no evidence of such 
expansions, and the sense cells appear to rest directly upon the surfaces 
of the fibres which constitute the so-called plexus beneath the sensory 
epithelium. Such a condition is represented in Figure 13 (Plate 3). 
In the cristae there are abundant terminal expansions, which are 
appropriately compared to the calyces of flowers, since they do not 
extend simply in the direction of the plane of the section, but, rather, 
more or less radially from the end of the axis cylinder. Consequently 
they are related to very many more sense cells than are visible in a 
single optical plane. Figure 11 (Plate 3) represents such a structure, 
to which eight sense cells are visibly related. ‘The real number of 
sense cells supplied by this terminal expansion is obviously many times 
as great as the number lying in any one optical plane. I have observed 
no cases of single cells being related to more than one such terminal 
organ. 
The boundary between the sense cells and the nervous material 
of which the terminal expansions is composed is always distinct and 
sharp. ‘There is no evidence, in my preparations, for any gradual 
transition from one to the other, but the dark color of the calyx ends 
abruptly where the yellow of the sensory cells begins. In no case have 
I been able to find structures corresponding in any way to the intra- 
cellular network described by Kolmer (:05*, :07). 
In those preparations in which the cell boundaries are best preserved 
it has not been possible to distinguish the cell wall in places where the 
cell was in contact with nervous material. Figure 24 (Plate 4), how- 
ever, represents two cells in the crista, one of which appears not to 
have been in contact with nervous material at its base, the cell wall 
being traceable with clearness to the basal part of the cell. The ad- 
jacent cell has a nerve fiber running to its base, where it bifurcates, 
one branch passing between the two cells, the other passing between 
the observer and the cell. I believe it is safe to assume that these two 
cells are alike in possessing a cell wall at the base, as well as at the sides, 
although the method employed does not enable us to distinguish that 
portion of the cell wall which is in contact with the nervous material, 
which is stained black. 
Likewise, it seems probable that those sense. cells which rest upon 
the terminal nervous expansions are separated from the expansion 
by cell membranes. 
