Before proceeding to give some examples of this fact, and to indicate its bearing on our interpretation of behaviour, it may be well to state distinctly that no attempt is or will be here made to trace in detail the course of the evolution of animal behaviour through the ascending grades of life, nor, indeed, to prove that there has been any such evolution. Evolution by natural genesis is here assumed as the only hypothesis with which science has any concern. If it be false, then have the labours of workers and thinkers, since Darwin and Mr. Herbert Spencer worked and thought, been vain. Special creation is not a scientific hypothesis, but a reference of biological and mental phenomena to an ultimate cause, which lies beyond and altogether apart from the scope of scientific inquiry. The fundamental assumption of the man of science is, that any natural event he may select for detailed study has natural conditions and antecedents. And it is only in such detailed study--taking this or that particular occurrence and endeavouring to ascertain what were its related antecedents--that advance in the evolutionary interpretation of nature can be secured. Such advance has been secured by the labours of those physiologists who have established by careful experiment the quasi-independent action of subsidiary nerve-centres as constituents of the nervous system as a whole. In such animals as the crayfish and the lobster the central nervous system consists of a chain of “ganglia,” or nerve-knots, which are connected together by nerve-strands. If these strands be cut between the thorax, which carries the walking limbs, and the abdomen or hinder portion of the body, the nerve-connection between these parts is severed. If the forepart be irritated through its sense-organs, the limbs of that part will respond; but, whereas an unmutilated crayfish, subjected to such irritation, would give a vigorous flap of the tail, this does not take place in the crippled animal.[188] Still, if the abdomen be irritated, it will respond by a strong and swift contraction. The two portions of the body are each capable of acting independently with well co-ordinated movements, but no longer of working together with unity of purpose. In the hinder portion the abdominal limbs, or swimmerets, all swing backwards and forwards simultaneously with rhythmic strokes; they act in concert. Sever now the connections between their ganglia, and each pair of limbs will continue to swing rhythmically but not with concerted rhythm. We have isolated a number of quasi-independent centres, and rendered them really independent. Each is concerned with its own proper co-ordination, but can no longer combine with others in a wider co-ordination. Mr. Hyde[189] has shown that in the king-crab, _Limulus_, when the nerve-chain is severed just in front of the abdominal region, the rhythmic respiratory movements of the abdominal segments still proceed regularly and co-ordinately. Even when only