The brain of Acipenser. 07 
the body. Although the dendrites of a given cell show a tendency to 
expansion in one plane this plane may be variously placed. Owing to 
the extraordinary irregularity of form of the PuRKINJE cells it almost 
never happens that the neurite can be traced away from the cell. 
In one or two cases I have found the neurite turning outward as 
if to enter the molecular layer. 
The molecular layer. — This layer contains occasional 
granule cells whose neurites pass in the common direction with the 
fine fibres of the layer. There are present also a few cells of the 
II type, which measure 10—12 by 20—28 u. These are situated 
superficially in the molecular layer and are of two forms. One form 
is pyramidal with dendrites directed inward, the other is stellate 
with dendrites and neurite disposed parallel with the surface. In 
a few cases I have found these cells possessed of two short neurites. 
These cells seem to constitute a simple cerebellar cortex. Some 
cells among them are probably of the I type. In the cerebellar 
crest I have found one cell whose short neurite presented the club- 
like thickenings characteristic of the second variety of II type cells 
in the valvula. This cell measured 16—20 u. 
The fibres entering the cerebellum end almost exclusively in 
the granular layer. The ascending fibres of the VIII and lateral 
line nerves mostly spread out in the lateral lobes. Although they 
are not well impregnated they constantly present one or two simple 
varicose branches in the granular layer. In sections taken in 
favorable planes it is readily seen that these fibres do not enter the 
molecular layer, except that their tips may in some cases reach 
that layer. The more mesially situated of these fibres enter the 
body of the cerebellum together with a great part of the ascending 
V tract. Many of the V fibres end in the group of cells which I 
have called the nucleus of the ascending V, but at it is impossible 
to draw a limiting line between this structure and the body of the 
cerebellum, it may be disregarded for purposes of histological de- 
scription. The V fibres penetrate deeply into the lateral or granular 
portion of the body and may reach the valvula. At the caudal end 
of the tectum the tractus tecto-cerebellaris I (Pl. 13) enters the 
cerebellum at the junction of the body and valvula (Fig. E). The 
bundle is not large but is very conspicuous because of its coarse 
varicose fibres which are beautifully impregnated. As they enter 
the cerebellum these fibres thicken and present almost the appearance 
of protoplasmic processes. They divide, with few exceptions, in 
Zool. Jahrb. XV. Abth. f. Morph. 7 
