We have already seen that, if the claim for the inheritance of acquired characters be, on the evidence, judged unproven, and if instinct cannot be ascribed to transmitted habit, or regarded as a legacy of that which has been ancestrally acquired, the only scientific explanation of instinctive behaviour is one which involves the principle of natural selection. But no one doubts that, in the course of experience, animals acquire modes of procedure which are beneficial to the race. This is well seen in the play of animals as interpreted by Professor Groos. Now, why do animals play? From the psychological point of view, because they like it; from the biological point of view, because they thus gain practice and preparation for the serious business of their after-life. But why do they like it? because, under natural selection, those who did not like it, and therefore did not play, proved unfit for life’s struggle, and were eliminated. Suppose that an animal were born with a rooted hereditary aversion to everything nutritious and an inherited hunger for anything harmful and unfit for food. What chance would it stand of survival? Hereditary likes and dislikes determine the general course of acquired behaviour, just as hereditary nerve-connections determine the course of instinctive behaviour. Wherein, then, lies the difference between the two? In the fact that in the one case the nerve-connections are transmitted ready-made, while in the other they result from association or coalescence in the course of individual life. But in both cases the pursuit and attainment of the beneficial brings satisfaction. Now, the consonance of end has long been regarded as an inevitable deduction from the hypothesis of evolution. “That pains are correlatives of actions injurious to the organism,” wrote Mr. Herbert Spencer in his “Principles of Psychology,”[182] “while pleasures are the correlatives of actions conducive to its welfare, is an induction not based on the vital functions only. It is an inevitable deduction from the hypothesis of evolution, that races of sentient creatures could have come into existence under no other conditions. Those races of beings only can have survived in which, on the average, agreeable or desired feelings went along with activities conducive to the maintenance of life, while disagreeable and habitually avoided feelings went along with activities directly or indirectly destructive of life, and there must ever have been, other things being equal, the most numerous and long-continued survivals among races in which these adjustments of feelings to actions were the best, tending ever to perfect adjustment.” And he safeguards the position by adding: “It is frequently taken for granted that the beneficial actions secured must be actions beneficial to the individual; whereas the only necessity is that they shall be beneficial to the race.”