( 903 ) 
This diagram, at the same time, demonstrates clearly that the space 
in which the frontal root decussates and runs back in a caudal 
direction, does not lie 7 the cerebellum but between the valvula 
and the fold connecting it with the teetum. 
Summing up my results, | can state the following : 
Oculomotor nucleus. 
The IH nucleus in the lower vertebrates occupies a more constant 
place in the longitudinal-axis of the brain than LV nucleus. Never- 
theless it certainly undergoes a distinct frontal shifting. In a dorso- 
ventral direction the oculomotor nucleus of the Teleosts undergoes a 
considerable displacement, which agrees with the ventral displacement 
of the abducens nucleus in these animals. 
Trochlear nucleus and root. 
The trochlear nucleus in the lower vertebrates exhibits suill greater 
differences in its position with regard to the longitudinal-axis of the 
brain. In Petromyzon it lies at a great distance behind the III nucleus 
even behind its own root-entrance, on the level of the trigeminus 
root-entrance, as has also been proved by Tretsakorr for Ammocoetes. 
Moreover it lies more dorsally, above the aquaeduct in the velum. 
In the Selachiu the nucleus lies at a great distance from the 
trigeminus root and close to the III nucleus, passing into the latter. 
In Selache it extends partly behind, though for the greater part in 
front of the IV root-entrance. Moreover, it has come to lie under 
the aquaeduct. In Teleosts the frontal shifting has reached its maximum. 
Tbe nucleus lies entirely in front of the IV root-entrance. 
In some Teleosts the trochlear root undergoes a peculiar spreading 
in bundles, e.g. Gadidae and Pleuronectidae, owing firstly to the 
root being split into two parts before decussation by the passing 
through of the tr. cerebello-mesencephalicus; secondly, the part which 
runs round outside that bundle is drawn forward by the frontal 
growth of the velum that grows out to valvula cerebelli, in conse- 
quence of which it decussates more frontally, and then again runs 
caudally between valvula and its connection with the tectum. 
This severing of the decussation levels also explains the splitting 
of the trochlear root into two roots at the exit, which until 
now has not been found in fishes without valvula (sharks) nor in 
those bony fishes where the valvula is very small (Lophius). 
59 
Proceedings Royal Acad. Amsterdam. Vol. XIII. 
