Herrick, Morphology of Brain of Bony Fishes. 57 



ber towards the free margin of the hypoarium. These cells are 

 pyriform in shape ; the thicker end is occupied by the nucleus, 

 in the centre of which in most cases a spot like nucleolus is seen; 

 occasionally, however, this is occupied by a group of granules. 

 Each cell partly occupies a distinct chamber in the neuroglia, 

 into which only the nucleus projects, the pointed end being 

 closely surrounded by neuroglia; occasionally a process, given 

 off from the thicker end, converts them into bipolar cells." "The 

 ventricles are lined by a distinct layer of epithelium, which is a 

 continuation of the epithelium lining the central canal of the 

 spinal cord. They are surrounded by a layer of small circular 

 or pear-shaped cells which extend for some distance into the 

 substance of the hypoaria." "In addition to these small cells 

 others of a much larger size occur, which are situated more par- 

 ticularly under the epithelium of the ventricle of the hypophysis 

 cerebri, and the adjacent anterior edge of the hypoarium ; some 

 are pear shaped, others are e.xpanded at one end and flattened 

 like an enlarged columnar epithelium cell, but they are not of 

 the nature of epithelium, for they lie beneath that layer." It 

 will be seen that more perfect methods greatly modify the above 

 description of the cell structure. 



It must be added to the general description of the hypoaria 

 that they are generally double, either obviously so, as in the 

 cat-fish, or divided by the structure into a caudal and cephalic 

 lobe, each having a special cornu. of the ventricle and a peculiar 

 structure. 



The comparison made by Professor Fritsch with the mam- 

 millaries ceases to have weight when we observe that the entire 

 mass of peduncular fibres in two large distinct bundles enters 

 these bodies and also enters into cellular combination within the 

 substance. The structure of the cells is moreover entirely unlike 

 that of the mammillary bodies. Histologically the caudal lobe 

 of the hypoarium is composed of a cortical and a medullary 

 portion. (In this consideration the corpus rotundus is not reck- 

 oned to the hypoaria but to the mesencephalon). 



That the fibres of the dorsal tract of the hypoarium are not 

 the direct continuation of the ventral peduncle fibres can in 

 some cases be demonstrated. Where the section falls in a plane 

 parallel with these radial fibres the bundles can be traced quite 



