30 Journal of Comparative Neurology. 



but fuse with those of the ventral commissure and, going with it, 

 seem to pass into the lateral aspects of the coUiculi. 



The resemblance of this tuber bundle to what Dr. Edinger 

 has described as the " mantle bundle," arising in the cerebrum 

 of Selachii, is so great that repeated attempts have been made 

 to find a similar relation in Teleosts, but thus far without success. 

 Dr. Edinger writes me that he thinks the structures distinct. 

 The close relation of this tract with the dorsal peduncles at one 

 point makes it possible to pass from one to the other if great 

 care is not exercised. 



3. The maimnillai'e, post-cinerea, and conus pt'ccconitnis- 

 suralis. It has been already mentioned that there are definite 

 expansions of the infundibulum upon the caudal aspects of the 

 tuber. Although the protuberances produced thereby are often 

 covered by the saccus and hypoaria we have seen no brain in 

 which they were absent. The cells adjacent are fusiform and 

 more or less concentrically arranged. The peduncles of the 

 saccus pass between them and fibres arise in the neighborhood 

 dorsad and can be traced cephalad to the region of the anterior 

 commissure {fornix tracts, above.) Immediately cephalad arise 

 the fibres which pass to the ventral commissure, while another 

 tract passes toward the habena. When we take into account the 

 position and fibre-connections and constancy of occurrence, it 

 seems impossible to avoid comparing these bodies with tbe mam- 

 millaria. The attempt which has been repeatedly made to 

 homologize the hypoaria with the mamrnillaries is obviously utter- 

 ly futile, for not only are they situated in an entirely different 

 region, but the fibre connections forbid any such reference. .\ 

 tract from the mammillaries to the cerebellum and the reception of 

 the ventral pyramids by those bodies would be unreconcilable. 

 From a level somewhat dorsad the ventricles of the hypoaria 

 diverge latero-cephalad and then suddenly ventrad, dividing into 

 two cornua. Morphologically these ventricles are rather ventro- 

 lateral pouches of the aqueduct than extensions of the third ven 

 tricle, as they seem to be. 



Somewhat dorsad of Fig. 2, Plate IV, there appear two 

 protuberances of the dorso-caudal aspect of the tuber which 

 contain only irregularly arranged spherical granules. These 

 may be followed dorsad until they acquire a peculiar corrugated 



