Herrick, Morphology of Brain of Bony Fishes. 53 



ersed by the ventral commissure, is bounded dorso-mesad by the 

 peduncles and caudo-laterad by the tuber and hypoarium. It 

 sends well-defined fibre tracts to the dorsal peduncle. 



The structure of this nidulus is difficult to clear up. Con- 

 sisting as it does of a meshwork of fibres from various sources 

 and cells of much the character of those of the lateral aspects 

 of the cerebrum, it is hard to follow the connections with cer- 

 tainty. The fact that the commissura ventralis passes through 

 it and divides within it, suggests that it receives fibres from the 

 brachia optica. That a strong contingent derived from the cells 

 of the geniculatus itself pass to the dorsal peduncle, seems to be 

 beyond question. A peculiarity of this body in fishes is the de- 

 velopment of a curiously specialized peripheral cell cluster. 



The isolated position of the geniculate body in Amiurus, 

 especially the young fry, makes this brain a Very useful one. 

 It is easy to trace a strong branch from the dorsal peduncular 

 tract into its substance and to follow some of the fibres into ac- 

 tual connection with the cells, which resemble in size and ap- 

 pearance those of the hypoaria. There is a curious area on the 

 cephalo-dorsal aspect, where a depression of the surface indicates 

 the locus of a group of peculiar elongated cells of larger size 

 than the others which radiate entad from this modified surface. 

 The appearance is so like that of the ependyma in certain ven- 

 tricular kreas that a suppressed diverticle of the ventricle was 

 suspected. This seems excluded by the position and closer in- 

 spection of the cells shows that they resemble those of the super- 

 ficial layer of the hypoaria. Among the cells are several of the 

 peripheral expansions of the ventricular epithelmm so conspicu- 

 ous in the tectum. Each of the cells gives off two fibres which 

 have not been traced further. A similar superficial area occurs 

 in Lucioperca. 



From the examination of horizontal sections of Haploido- 

 notiis it seemed probable that fibres from the ventral commis- 

 sure connect with the peripheral fibre of the superficial cells, 

 while another branch is given off to divide among the protoplas- 

 mic processes of the cells, whose connection with the dorsal 

 peduncles is obvious. The transverse commissure seems to 

 have no relation with the geniculatus. In transverse sections of 

 the drum the larger cells are not restricted to the periphery. In 



