CENTRAL NERVOUS SYSTEM IN VERTEBRATE ANIMALS. 
759 
the other ; they all cross the roots of the oculomotorii at right angles and plainly 
do not join them. The greater number of the fibres entering into the composition 
of this commissure are derived from the second layer of the optic lobe. In fact it 
may be considered as the commissure of the optic lobe, but in addition to these fibres 
some few appear to be derived from the lateral columns of the cord. 
In the Kays this commissure is not so well developed as in Scyllium and Acanthias. 
In the latter it extends farther behind the oculomotorii than in the others. It seems 
as if, in these latter, the extent of these association fibres compensates for the size of 
the ventricles and the consequent thinness of the walls of the optic lobe, cerebrum, 
and the other parts of the brain. 
Fourth pair.—The trochleares are very difficult to trace; they have a distinct 
decussation in the region of the valvula cerebelli, which is represented by a narrow 
neck of tissue between the cerebellum and the optic lobe, and is not nearly so well 
developed as in the Teleostei. I have not been able to trace the roots of these nerves 
to any ganglion, but in one specimen they were seen turning forward by the side of 
the aqueduct of Sylvius. In the region of the ganglion of the oculomotorius, there 
are no collections of cells to be found from which they might be supposed to have 
been derived. Rohon,'“ in his account of the origin of this nerve, appears to have 
fallen into a slight mistake ; he describes it a3 arising from a swelling which is in 
direct connexion with the cortical part of the anterior tuberosity of the cerebellum. 
A consideration of his figure convinces me that the section has been made through 
the restiform bodies behind and not in front of the crura cerebelli ad medullam; if so, 
the bundle of fibres which he looks upon as the root of the trochlearis must have 
some other signification. Acanthias is a very good subject for this investigation, 
because its cerebellum being small does not extend forward and obscure the relation 
of the parts between it and the optic lobe. Fig. 14 is taken from this animal, 
and shows the narrow valvular portion extending as a tent over the anterior part of 
the fourth ventricle at the point of its transition into the aqueduct of Sylvius. It 
shows also the decussation of the fibres and the commencement of their course along 
the side of the aqueduct. 
Fifth pair (fig. 11).—The trifacial is derived from three roots. One comes forward 
from the posterior part of the medulla oblongata from the lateral columns of the cord. 
This evidently corresponds to the ascending root in Mammalia. Another comes back¬ 
ward from the anterior part of the tuberosity of the trifacial, into which it penetrates 
from the formatio recticularis ; these two bundles pass in opposite directions until they 
arrive at the point of exit of the nerve, when they turn directly outward and cross 
each other to join the emerging root of the nerve. The third origin comes from 
a group of cells situated on the lateral part of the floor of the fourth ventricle ; in 
addition to the fibres from the ganglion, others are derived from the raphe. 
* Op. cit., p. 80, fig. 49. 
