411 
and periphery, finer bundles of fibrils spring from the plexus 
of thicker fibrils, which either immediately after their origin 
spread out into individual fibrils anastomosing with each other, 
or give off fibrils in succession in a manner resembling the 
mode of branching of a weeping willow; these, like others, 
reticulating with each other.! It is at all events certain that 
the subepithelial network, whether its arrangement is that 
of a trellice-work or a network, consists of extremely fine 
fibrils, which are marked by the possession of varicose swell- 
ings. It is maintained by Cohnheim that fibrils of extreme 
tenacity spring from the subepithelian network in a direction 
almost vertical to the surface, finding their way between the 
deepest cells of the epithelium ; and that after giving off con- 
necting fibres in the middle layers, they can be traced to the 
surface where, although some of them terminate among the 
superficial layer of cells, they mostly end in filaments with 
terminal knobshaped swellings (Endknopfschen), which float 
free in the precorneal fluid. As regards this distribution in 
the epithelium my preparations show the following: 
The fibrils which rise among the pallisade-shaped epi- 
thelium, give off very numerous fibrils, often only distinguish- 
able as linear series of granules, which take a horizontal 
course on the ends of the columnar cells next the surface. 
These horizontal filaments wind among the epithelial elements 
in zigzag lines, and are connected together both directly and 
by lateral branchlets, and thus form a network which is not 
much inferior to the subepithelial network in density. After 
giving off the horizontal branches, the vertical fibrils wind 
in a convoluted manner towards the surface. In this part 
of their course filaments spring from them which bifurcate, 
and either anastomose together, or accompany them towards 
the surface. The nerve fibrils which are to be found in the 
superficial layers, exhibit here and there in their course, in 
addition to the minute varicosities above described, swellings 
of relatively much larger dimensions. Separated from the 
surface by one, or at most two, layers of elements, they form 
a very dense network of fibres, the meshes of which are smaller 
than those of the deeper network, and the fibres (which have 
1 While this is going through the press, I have made out in preparations 
from which only the anterior epithelium had been removed, that the network 
which is so well drawn by Cohnheim, aud which is seen in my figure 1, Pl. 
XIX, by no means contains the finest and most numerous fibrils. With an 
immersion 10, I have discovered an immense number of the very finest fibrils 
given off from each of those seen in fig. 1; all of these run nearly parallel 
with one another, join with one another with similar fine fibrils, and are, 
without exception, marked by regularly placed bead-like enlargements. A 
drawing of these fibres will be given in the next part of this essay. 
