and then descends again to the very same motor cells of the spinal cord which intervened in the reflex action. Now what has it gained by this roundabout course, and what did it seek in the so-called sensory cells of the cerebral cortex? I do not understand, I shall never understand, that it draws thence a miraculous power of changing itself into a representation of things; and moreover, I hold this hypothesis to be useless, as will shortly appear. But what I do see clearly is that the cells of the various regions of the cortex which are termed sensory,—cells interposed between the terminal branches of the centripetal fibres and the motor cells of the Rolandic area,—allow the stimulation received to reach at will this or that motor mechanism of the spinal cord, and so to choose its effect. The more these intercalated cells are multiplied and the more they project amoeboid prolongations which are probably capable of approaching each other in various ways, the more numerous and more varied will be the paths capable of opening to one and the same disturbance from the periphery, and, consequently, the more systems of movements will there be among which one and the same stimulation will allow of choice. In our opinion, then, the brain is no more than a kind of central telephonic exchange: its office is to allow communication, or to delay it. It adds nothing to what it receives; but, as all the organs of perception send to it their ultimate prolongations, and as all the motor mechanisms of the spinal
