654 THE BRAIN. 



fashion into a large number of branches which, spreading out laterally in 

 the transverse plane and stretching as far as the surface, ramify through the 

 molecular layer, and are eventually lost to view as exceedingly fine fibrils. 

 Some observers maintain that some of the fine processes are continuous 

 with processes of the small nerve-cells of the molecular layer, but this is not 

 admitted by all. In any case the fibrillar terminations of these cells of 

 Purkinje contribute to the fine fibrils of the molecular layer. 



The nuclear layer in ordinary stained specimens has 'the appearance of 

 a mass of nuclei closely crowded together in a bed of reticular nature ; and 

 since the nuclei usually stain deeply, the layer stands out in strong contrast 

 to the much less deeply stained molecular layer. Careful examination with 

 special modes of preparation shows, however, that while some of the nuclei 

 are nuclei belonging to neuroglia and bloodvessels, the majority belong to 

 small nerve-cells of a peculiar nature. In these cells the nucleus is sur- 

 rounded by cell substance, which, forming a thin layer immediately around 

 the nucleus, is chiefly disposed as thin spreading branches, some of which end 

 in a peculiar arborescence not unlike a muscle end-plate ; these processes 

 contribute w r ith the neuroglia to form the reticular-looking bed spoken of 

 above. No process can be traced inward to the central white matter ; but 

 one of the processes gives off a branch, which passing vertically outward 

 takes on the appearance of a delicate axis-cylinder process and runs, with- 

 out dividing, into the molecular layer for a variable distance, sometimes 

 reaching close to the surface, but at last divides at right angles into two 

 fibrils, which run in the longitudinal plane in opposite directions for a con- 

 siderable distance, and are ultimately lost to view. Since these cells in the 

 nuclear layer are very numerous and each gives rise in the above manner to 

 longitudinal fibrils, the molecular layer is traversed by a multitude of fibrils, 

 visible as such in longitudinal sections, but appearing as dots in transverse 

 sections, in which the cells of Purkinje are best displayed. 



Besides these longitudinal fibrils proceeding from the cells of the nuclear 

 layer, special modes of preparation similarly disclose numerous transverse 

 as well as more or less oblique fibrils. Many of these appear to result from 

 the branching of the small nerve-cells of the molecular layer, and some of 

 those so arising descend to the layer of the cells of Purkinje and end around 

 the bodies of those cells in remarkable nests of fibrils, without, however, 

 actually making connections with them. 



The medullated fibres of the central white matter of a lamella pass on all 

 sides into the nuclear layer; or, put in another way, medullated fibres pass- 

 ing out of the nuclear layer at all points converge to form the central white 

 matter. Some of these fibres, as we have seen, begin or end in the cells of 

 Purkinje. None of them appear to join the cells of the nuclear layer, and 

 we have no evidence that any of them end or begin in any way in the nu- 

 clear layer. A certain number, however, may be seen to pass through the 

 nuclear layer and between the cells of Purkinje into the molecular layer, 

 where losing their medulla they divide and apparently contribute to the 

 numerous fibrils of the molecular layer. The presumption, therefore, is that 

 all the fibres of the white matter begin or end either in the cells of Pur- 

 kinje or the fibrils of the molecular layer. 



The superficial gray matter of the cerebellum then resembles the gray 

 matter of the spinal cord in so far as it consists of branching nerve-cells, 

 nerve-fibres, and nerve-fibrils imbedded in neuroglia ; but the disposition 

 and features of the several factors are peculiar. We may take, perhaps, as 

 the key of the structure the fibrils of the molecular layer ; this layer is rela- 

 tively very thick, about 400 /x, much thicker than the nuclear which, how- 

 ever, varies in thickness, being generally thickest at the top of the fold ; 



