passing from these to the ganglionic cells; instead of that there is a ground-substance thickly interspersed with granules and nuclei. _From_ the cells we see processes issue; _to_ the cells none are seen arriving. So with the cerebellum. The large cells send their processes upwards to the surface; but downwards towards the white substance the processes are lost in the granular layer, which most histologists regard as connective tissue. 155. A mere glance at nervous tissue in any part will show that cells are far from forming the principal constituents. In the epidermis or a gland the cell is obviously the chief element, forming the bulk of the tissue, and being the characteristic agent. In nerve-tissue, as in connective tissue, the reverse is the case. We must therefore cease to regard the cell as having the importance now attached to it, and must rather throw the emphasis on the fibres and neuroglia. 156. Before quitting this subject let a word be said on the amazing classification which has attained wide acceptance (although rejected by the most eminent authorities), founded on the size of the cells--the large multipolar cells being specified as motor, the smaller cells as sensory, while those of an intermediate size are sympathetic. I forbear to dwell on the development of this notion which specifies sensational, ideational, and emotional cells, because this does not pretend to have a basis in observation; whereas there are anatomical facts which give a certain superficial plausibility to the original classification. The conception is profoundly unphysiological; yet, if the anatomical evidence were constant, one might give it another interpretation. The evidence is, however, not constant. Large cells are found in regions assigned to sensory nerves, and small cells in motor regions. In the spinal cord of the tortoise Stieda declares that the so-called motor cells are limited to the cervical and lumbar enlargements; all the rest of the motor region being absolutely destitute of them.[182] Again look at the cells of the retina--no one will assign motor functions to them--yet they are the same as those of the cerebellum and the anterior horns of the spinal cord. (It is worth a passing mention that the structure of the nervous parts of the retina more closely resembles that of the cerebellum than of the cerebrum.)