Herrick, Morphology of Brain of Bony Fishes. 25 



the radix at the point of exit from the pero. These cells are 

 fully twice as large as those of the ganglionic layer of the pero. 

 II. — Tracts from the hemispheres. 



1. Tania thalami. In most brains examined there is a 

 peculiar train of Deiters' granules which first appears on the 

 lateral aspects of the preethalamus near the level of the praecom- 

 missura and not far from the hippocampus It can be traced 

 with accompanying fibres to the mesal aspects of the peduncles, 

 thence with the latter into the mid-thalamus, where the cluster 

 speads out on the surface in front of the habena and the pedun- 

 cular fibres sink into the deeper ventral portions between the 

 optic tracts. This tract contains unmedullated (?) fibres and is 

 probably the homologue of the t^nia. The actual connection 

 of this tract with the fibres from the habena has not been fol- 

 lowed and there are a number of points in respect to this 

 region which for the present must be left open. Other tracts 

 from the hemispheres into the upper and middle parts of the 

 midthalamus have thus far proven impossible to trace definitely to 

 either terminus. 



2. Ventral peduncles. As above stated, we have been so 

 fortunate as to make out the actual cellular connection of the 

 peduncular fibres with apex processes of the cerebrum. It was 

 a gratifying surprise to find that the dorsal and ventral pedun- 

 cles have a distinct origin in the hemispheres as well as a distinct 

 course in the mesencephalon. Although we have carefully 

 avoided attempting specific homologues for the several parts of 

 the axial lobe yet the fact that its cephalic portion gives rise to 

 the ventral peduncle and its caudal portion to the dorsal goes 

 far toward substantiating our suggestion that in the axial lobe 

 there are the representatives of the cortical centres of higher 

 vertebrates and a certain similarity in general arrangement. 

 Moreover, that part of the brain connected by the callosum 

 proves to belong chiefly to the ventral system. After passing 

 m company with the dorsal peduncle'; caudad of the chiasm the 

 ventral portion describes a latero-ventral curvature. In those 

 fishes in which the hypoaria are divaricated, as in the cat fishes, 

 the ventral peduncle turns sharply laterad and is thus easily dis- 

 tinguished from the dorsal portion, even in horizontal section. 

 In other cases the course is almost directly ventrad, and perpen- 



