These granules are small neurones which take up impulses from afferent (“mossy”) fibres, and distribute them to the dendrites of the Purkinje cells—each collecting from a few fibrils only of the sensory channels. (The word “sensory” is used to indicate that sense-organs are their provenance, and not that their messages become sensations.) The numerous spreading branches of a Purkinje cell, disposed in a transverse plane, are obviously arranged to hold up and keep apart these myriads of longitudinal threads. A cerebral pyramid is shaped like a fir-tree. It is placed in a definitely stratified layer. By its branches it collects impulses from the superficial strata, which it transmits through its stem to the white matter beneath the cortex. The various parts of the central nervous system have work of different kinds to do, and we find interposed in the circuits which compose the several parts cells of various types. We speak of the large cells as “motor,” the granules as “sensory,” the small pyramids as “association” cells—such terms indicating the positions which they occupy in the arcs, but not defining their functions. Of specialization of function the physiologist cannot obtain a hint. He cannot classify nerve-cells in groups concerned in reflex action, in feeling, in remembering, in willing, in thought. On the contrary, he can assert with confidence that such distinctions are not to be drawn. In various situations in the central nervous system a certain type of cell is found for which, in the present state of knowledge, it is impossible to account. We mention these cells lest it should be inferred, from what has been said above, that all neurones can be fitted into a simple scheme of conducting arcs. In the spinal ganglia there are neurones whose axons divide to form “baskets” around other ganglion-cells. In the cerebellum there are similar cells, the axons of which divide into branches, which break up to encase Purkinje-cells. Cells of the same kind are found in a few other situations. In some cases the end-branches which enter into the formation of the baskets are few in number, and thick and clumsy. They grasp the body of the cell which they surround, with gouty fingers, as it were. In other cases the basket is a tangle of fine threads. It is difficult to see what rôle cells of this kind can play in conduction. From the olfactory and optic centres nerve-fibres extend outwards to the olfactory bulb and retina. Here again is an arrangement which does not fit in with any scheme. We might multiply examples. But enough has been said, perhaps, to convey the impression which we wish to leave, that, although experiment abundantly proves that the nervous system consists of an association of sensori-motor conducting arcs, and although anatomical investigation demonstrates the existence of chains of neurones which take part in the formation of such arcs, it is impossible to reduce the system to schemata or to prepare diagrams in which all structural