116 The American Naturalist. [February, 



cease here, but from the deeper parts of the hippocampal lobe 

 an ental tract, corresponding to the hippocampal commissure 

 and descending fornix tract, arises and passes cephalomesad to 

 cross just caudad of the strongly developed callosum. At this 

 point a branch is given off precisely like the fornix which passes 

 to a bilobed cellular mass, projecting ventrad into the ventricle 

 caudad and dorsal of the lamina terminalis. In appearance 

 and structure, as well as fibrous connections, these tubers can 

 only be regarded as identical with the corpus fornicis. Nothing 

 is wanting to make the homology perfect. Fibres from the 

 fornix body can be traced into the thalamus, but nothing 

 indentifiable with the mammillaries could be made out with 

 certainty. 



The radix mesalis of the olfactory is much larger and more 

 distinct, and seems to be derived from a mass of cells occupy- 

 ing the pes, while the radix lateralis springs from the pero. 

 This bundle passes caudo-dorsad and then mesad and crosses 

 in a special ventral portion of the prascommissura quite dis- 

 tinct from the specific homologue of the latter. In the Drum, 

 the olfactories are attached to the cerebrum, or more strictly, 

 to the prsethalamus just at the origin of the lamina terminalis. 

 They are obviously not appendages of the prosencephalon, as we 

 have suggested on morphological grounds it would be impossi- 

 ble for them to be, but the radices have suffered division or 

 latero-fiexion with the outgrowth of the secondary cerebral 

 vesicles. The origin of the hippocampal loop of the olfactory 

 tract becomes obvious by inspection of an embryonic brain or 

 that of a fish, it being the least diverted portion of the walls 

 of the primitive prosencephala ventricle carrying with it the 

 tracts and commissural origins proper to it. It is a curious 

 fact that of the two radices of the olfactory, one belongs to the 

 dorsal and the other to the ventral system and the commissural 

 or decussational fibres likewise belong, one to the ventral the 

 other to the dorsal of the two primary commissural systems of 

 the neural tube. The radix lateralis connects, via hippocam- 

 pus, with the dorsal system associated with the callosum, 'i.e. 

 the fornix and hippocampal commissure. The radix mesalis, 

 on the other hand connect- with the anterior commissure sys- 



