750 
MR. A. SANDERg ON TEE ANATOMY OF TEE 
two. ItoHON " imagined that he saw several processes cut off close to many of these 
cells, and therefore came to the conclusion that they were multipolar cells, but such is 
not the case. I have examined them very minutely, and in good preparations the 
contour has always appeared perfectly smooth, but in preparations that have not 
succeeded well their outline becomes irregular, and this no doubt misled Rohon. 
The single process, when only one is present, passes towards the outside, go that of 
the cells on each side of the mid-line each sends its process in an opposite direction ; 
they always join the transverse commissure which constitutes the second layer. The 
average length of these cells is 55p, and the width 35/r. Yiault does not appear to 
have seen these cells, as no mention is made of them in his account of the histology of 
the optic lobes.t 
In Scyllium these cells first appear rather suddenly as a large circular group in that 
part of the roof of the aqueduct of Sylvius which corresponds to the valvula 
cerebelli opposite the exit of the fourth nerve; in the next section in advance the 
number of cells grouped together is not so large, and still farther forward they spread 
out into a single row laterally, while still keeping more closely clustered together in 
the mid-line. They extend as far as the anterior end of the optic lobe, where they 
terminate with a few scattered cells. 
In the Rays these cells commence behind also opposite the exit of the trochlearis, 
not, as in Scyllium, by a large group, but by a single cell on each side close to the 
anterior termination of the granular layer of the cerebellum. These two cells are 
rather isolated ; in the next three sections in advance only one is to be found ; in front 
of this they gradually increase in number until they form a layer, one or two, and, in 
some places, three cells in thickness, extending some distance laterally. Anteriorly 
they diminish in number and terminate with two outlying ones some distance behind 
the transition of the optic lobe into the third ventricle. In these fishes occurred the 
only instance of an exception to the rule of each cell having but one process ; in this 
case a process was seen emerging from each end. There was also a single instance 
only of a cell sending its process toward the mid-line instead of toward the outside. 
In Rhina these cells are restricted to the mid-line and do not spread out laterally. 
They commence as a group of four or five cells immediately in front of the exit of the 
trochlearis ; they rather suddenly increase in number, and are restricted to the projec¬ 
tion of the roof into the ventricle ; on the edge of the group one or two outlying cells 
are occasionally found. They extend forward for about half the length of the 
ventricle. In another specimen there was a slight variation. They commence in this 
specimen in the valvula cerebelli by three groups, one on each side and one situated 
superiorly in the mid-line ; this latter disappears, and the two former coalesce centrally 
and extend forward nearer the anterior end of the ventricle than in the last specimen, 
and terminate in a few stragglers. 
* Op. cit., p. 78. 
t Op. cit., p. 501. 
