934 
MR. A. SANDERS ON THE ANATOMY OE THE 
an opportunity of testing the method which he followed in obtaining his results, a 
description of which he was kind enough to send me. These results are, however, 
very interesting, showing as they do the mode of origin of the optic nerve primarily 
from a net-work of the finest possible fibrillse, a statement which, however probable, 
had not before been demonstrated. 
The two outer layers occupy only a small part of the tectum, the remainder 
extending to the inner surface corresponds to the second, third, fourth, fifth, and sixth 
layers in M. cephalus; it consists principally of neuroglia in which radial striation is 
indistinctly visible. The fusiform cells which ordinarily form a conspicuous feature in 
this lobe are few and far between; but they certainly are present and occur close 
to the outer edge of this inner layer. They are placed more obliquely than in 
M. cephalus, and as in that species, their outwardly directed process, which is pro¬ 
bably the protoplasmic process, goes into the external layer. Fritsch,* in his large 
work on the structure of the brain in Fishes, denies the existence of these cells, and 
supposes that Stieda mistook a crossing of two capillaries for them; this seems to 
show that the former writer did not make his sections sufficiently thin ; if they are 
properly made, so that only one layer of cells occupies their thickness, such a mistake 
would be impossible. 
The cells which correspond to the sixth layer do not form here a distinctly continuous 
layer, but occur in scattered groups along the inner surface of the tectum ; they are 
slightly larger than the cells in M. cephalus; they measure about 0’005 millim. by 
0*004 millim. The interrupted layer thus formed is interspersed with longitudinally- 
directed fibres, which correspond to the fifth layer of the tecta in Teleostei; these 
fibres are medullated, each of them appearing to be a single fibrilla clothed with a 
very narrow medullary sheath. The reason why longitudinally directed fibres in one 
species correspond to transversely directed fibres in another depends on the position 
of the tori longitudinales, which in Mormyrus are widely separated from each other 
by the development of the wings of the valvula cerebelli. The tori longitudinales are 
only in contact in front, and posteriorly they are placed along the outer edge of the 
tecta and end in a free point. The consequence of this arrangement is that the 
commissure of these lobes only exists at the anterior end, and the fibres destined to 
form it must therefore necessarily run longitudinally. The structure of the tori 
longitudinales in Mormyrus resembles that of the corresponding parts in M. cephalus 
so far as they are made up of the smallest-sized cells ; but the arrangement of these 
cells is different; they do not occur in rows between bundles of radiating fibrillse, but 
quite irregularly, in a net-work of the smallest-sized fibres ; they vary very much in 
form ; some are rounded, some pear-shaped, some oval, and others irregular in outline ; 
they give off numerous processes. 
I termed these bodies “ fornix ” in my paper on the brain of Teleostei, following 
* Loc. cit p. 51. 
