134 Journal of Comparative Neurology. 



common type in which at first cells are everywhere collected 

 about the ventricles. 



Respecting the cerebellum, Osborn, in 1883, described in 

 the Amphiuma, (i) a continuous band of fibres arching fi-om 

 side to side of the medulla, (2) a fine layer of fibres which have 

 an antero-posterior direction, (3) an investing layer of cells one 

 or two rows deep. The transverse band was homologized with 

 the inferior peduncles and restiformia. The fine layer termi- 

 nates in the lateral portions of the cerebellum and optic lobe 

 and is morphologically ventrad of the preceeding. The cells 

 are thought to be of a nervous character but no processes were 

 discovered. In a later paper^ he states that in Cryptobranchus 

 the "cerebellum is chiefly composed of decussating tracts, pass- 

 ing on the one side into the lateral regions of the medulla and 

 into the mesencephalon. It may even be questioned whether 

 we have here the essential elements of the cerebellum, the 

 structure is so extremely simple." He withdraws from the 

 homology of the tract to the mesencephalon with the prepedun- 

 cles of the cerebellum and states that it is really the mesenceph- 

 alic or descending trigeminal tract. The cerebellum of Am- 

 phiuma contains no cells except the ependyma, while in 

 Cryptobranchus there is a small mass of round cell similar to 

 those of the central substance of the optic lobes. The fine 

 fibres of the first sort above described are derived chiefly from 

 the eighth nerve though some come from the seventh. 



Alborn considers that the fibres entering the cerebellum 

 are commissural between the auditory niduli of opposite sides 

 while Koppen and Osborn think they are decussating tracts 

 from the nerve. 



The middle peduncle of the cerebellum is fairly well devel- 

 oped in Anura but instead of tracing it unbrokenly into the in- 

 fundibulum with Dr. Edinger it seems to us that most of its 

 fibres either mingle with those of the ventral peduncles or come 

 into relations with them through the scattered "niger" cells of 

 the pes region. The decussation of these bundles ("Bindearme") 



^Journ, Morph. II., i. i{ 



