944 
MR. A. SANDERS ON THE ANATOMY OF THE 
are cut off obliquely as they descend from above downwards ; they extend over the 
dorsal edge of the third layer, even where the first layer is deficient; and here they 
form what may be looked upon as a commissure. 
The third layer (figs. 7 and 8) is the widest of all of them, and consists of & finely 
granular material, showing distinct traces of radial striation, which is more apparent 
in some places than in others. It seems to have the same sort of composition as the 
molecular layer of the cerebellum ; the absence of palisade cells would account for the 
striation being very slightly indicated, and would seem to show how much of these 
markings might be attributed to the presence of the palisade cells, and how much to 
the nature of the material itself and to the direction of the nervous fibrilhe. 
The fourth layer (figs. 7 and 8) consists of cells of two sizes arranged in an arch 
equally curved in both directions., The smaller cells form a continuous line which is 
always conspicuous, in whatever direction the sections are made, on account of the 
intense colour which they take up from the staining fluid. They are generally bipolar 
and are so placed that their processes run in a radial direction, in apparent continuation 
of the striae of the third layer; they thus seem to be a medium of communication 
between the fifth layer and those layers situated more externally. 
These cells have all well-marked nuclei and nucleoli, and are of three sizes; the 
largest, generally spherical, measure 0*0143 millim. long (fig. 18, a.) by 0‘013 millim. 
wide, nucleus 0'008 millim., and nucleolus 0*002 millim. Of the middle-sized cells 
the length of one representative cell (fig. 18, b.) is 0*009 millim. long by 0*006 millim. 
broad, the nucleus is oval, 0*0045 millim. long by 0*003 millim. wide; there is here 
also a disk-like nucleolus measuring 0*0015 millim. in diameter. A process is seen to 
be directed from the small end of this cell towards the third layer; the comparative 
breadth and appearance of this prolongation characterise it as a protoplasmic process. 
A fine fibre coming from the other side of this cell, if it existed, is not visible. This 
kind of cell seems to form a transition between the larger kind resembling motor cells 
of the spinal cord and the smaller cells which some writers consider to be connected 
with sensation, but they have more affinity to the former, as is shown by their large 
circular nucleoli. 
The smallest species of cells (fig. 18, c.) in this layer range from 0*006 millim. to 
0*004 millim. in diameter; they are generally nearly round or oval; they have a 
nucleus and nucleolus; their processes are directed radially; the thicker is turned 
towards the third layer, and the other, which is finer, towards the fifth layer. The 
smaller cells are arranged in groups, irregularly scattered through the granular basis 
of the layer; the larger ones either occur singly or in groups, or else in some places 
occupy a curved row : these are placed at the edge of the fourth layer between it and 
the third. 
The fifth layer consists almost entirely of a complicated interlacement of medullated 
fibres; those at the anterior end belong to the trifacial, and those at the posterior 
end to the vagus. Anteriorly they enclose the sixth layer, while posteriorly they 
