Such a conception is clear and reasonable in principle, and all work on nervous physiology tends to show that it is a good working hypothesis. We cannot read modern books without feeling that immense advances will be made by its aid. But the complexity of the brain of the higher vertebrate is so incredibly great, and the difficulties of imagining the nature of the necessary physico-chemical reactions in the synapses, and elsewhere, are so immense that experimental verification may be impossible. And all that we have said applies to a single elemental stimulus, yet in any common action the stimulus is a synthesis of almost innumerable simple ones, while the response is also a synthesis. The optical image of almost any object contains a very great number of tints and colours differing almost imperceptibly: there must at least be as many simple stimuli as there are rod or cone elements in the part of the retina covered by the image. The motor responses consist of a multitude of delicately adjusted and co-ordinated muscular contractions and relaxations. If we are to accept a mechanistic hypothesis of action, of this kind, and which includes only such processes as are suggested above, it is not enough that a logical description, consistent in itself, and consistent with physico-chemical knowledge, should be formulated. The mere statement of such an hypothesis does not carry us far. If it is, in essence, mechanistic, it must be capable of experimental verification in detail. Even if it were verified experimentally it would still leave untouched the problem of consciousness. All that we have considered are series of physico-chemical energy-transformations. How, then, does consciousness arise? We cannot even imagine its association in a functional sense with the train of events forming an afferent impulse. In some form or other mechanism must assume a dualism--a parallelism of physical and psychical processes. Physical events in the central nervous system are associated with psychical ones--when the former occur so do the latter--yet the former are not “causes” in any physical sense of the latter. Consciousness follows cerebral energy-transformations as a parallel “epiphenomenon.” At once we leave the province of mechanism, and how can we remain content with such a limitation of our descriptions? And if we conclude, as we seem obliged to do, that consciousness is an affective agency in modifying our responses to external stimuli, does not this in itself show that our concept of behaviour as a purely physico-chemical process is insufficient in its exclusiveness?