Sleep is a condition in which all paths are open. Hence no impulses are effective. Wakefulness, alertness, depends upon the closing of all paths save those which are actually in use. We may go further. The power of concentrating attention is the power of limiting the spread of nerve-impulses in the brain. Alcohol opens extra paths; the concentrated effort which was making progress with a problem becomes more diffuse. The first effect appears in greater brilliance of thought, gained at some sacrifice of cogency. Unexpected analogies are discovered. Imagination takes a wider range. But as the dose is increased, a condition akin to sleep is set up. Nerve-impulses become ineffective because, many paths being open, they do not attain a sufficient intensity in any set of paths. These few illustrations are given for the sake of showing the need of a theory of the opening and closing of paths. It is not suggested that they favour the particular hypothesis here set forth as to the structural arrangement which provides the paths and regulates their accessibility. Recent discoveries in the finer structure of the central nervous system have provided many problems which at present appear insoluble. One of the discoveries most difficult to make use of in constructing theory is the existence of extracellular or pericellular nets, which have the appearance of extraordinarily delicate cases of wire-netting immediately surrounding the nerve-cells. It is somewhat remarkable that the spacing of the nets is often very similar to, if not identical with, the spacing of thorns. While some anatomists look upon the nets as nervous, others regard them as pertaining to the connective tissue of the nervous system. At present it is not known how impulses get across from the finest visible twigs of arborizing axons to the dendrites of the neurones which they influence. The wealth of structural detail which recent research has revealed is an embarrassment to anyone who tries to devise a scheme. Not improbably, pericellular nets are intermediate factors in the exchange; or, if not the nets, the structures whose existence is indicated by the appearance of the nets. In the case of many of the finer markings which staining methods bring into view, it is impossible to say whether they indicate the presence during life of the structure as it appears to be, or whether the markings are due to coagulation of plasma or to strain caused by shrinkage in coagulating agents. In a sense this is not of much consequence. Coagulation in a uniform pattern would mean the existence of an architectural substructure which determines the pattern. We may be looking at the cake or at the tin the cake was baked in.