592 



folding has brought the molecular layer of the two folds into apposition 

 and they have fused to form the single median molecular layer. The 

 cerebellum of Acipenser seems to differ considerably in the relations 

 of the parts from that of Teleosts as described by Schaper ('94). The 

 interpretation here given will be supported by my final description of 

 the minute structure. Here, I can do no more than give a very brief 

 statement of the main features of the two layers. 



The greater part of the granular layer is made up of very minute 

 cells whose neurites pass into the molecular layer and constitute the 

 fine fibres of that layer. These granule cells are present everywhere 

 in the granular layer of the cerebellum and also in both parts of the 

 acusticum, as above mentioned. In the posterior portion of the body 

 they are almost the only cells present. The neurites of a part of the 

 cells in this region enter the fimbria, pass to the other side, course 

 arond the border of the lateral lobes, and form the greater part of 

 the cerebellar crest. The granule cells are proportionately less 

 numerous in the lateral lobes and in the valvula. I have been unable 

 to satisfy myself of the T-shaped division of the neurites of the 

 granule cells which occurs in the mammalian cerebellum, and has been 

 described by Schapek ('93) for that of Teleosts. 



The other cells of the granular layer are Purkinje cells and cells 

 of GoLGi's II type, of which two or three varieties may be distinguished. 

 The Purkinje cells of the lateral lobes lie near the border line 

 between the granular and molecular layers, their dendrites expanding 

 in the molecular layer, as shown in Figs. 8 and 9. In the body and 

 valvula the Purkinje cells seem to lie wholly within the molecular or 

 middle layer (Figs. 7, 8, 10 and 11). This appearance is probably due 

 to the down-folding and fusion suggested above. The destination of 

 the neurites of the Purkinje cells I have not been able to determine. 

 From such evidence as I have, I think it probable that they run 

 through the acusticum to the base of the medulla. In the valvula are 

 a considerable number of cells in the molecular layer with characteristic 

 Purkinje cell dendrites, and with short neurites which present peculiar 

 club -like thickenings. One of them is shown in Fig. 11. Other, 

 smaller cells of the II type occur in the molecular layer in all parts 

 of the cerebellum, and even far posteriorly in the cerebellar crest. 



The fibres which come to end in the cerebellum come chiefly 

 from the medulla, the tectum, and the 'tween-brain. The ascending 

 fibres of the Vth.-VIIIth.-lateral line nerve complex enter the granular 

 layer of the lateral lobes and body (Fig. 9). Fibres from two separates 

 sources in the tectum enter the cerebellum and are distributed to 



