The advocates of this theory attempt to meet these objections by saying that the imitative instinct accounts for the particular directions in which the discharges of energy occur. A kitten's plays are like those of the cat tribe because the kitten is accustomed to imitate cats; when it falls to playing it is with cats, and so it sheds its superfluous energies in the customary imitative channels. In this way it grows to learn the games of its own species. There is a good deal in this point; most games are imitative in so far as they are learned at all. But it does not save the theory; for many animal plays are not learned by the individual at all, as we have seen above; on the contrary, they are instinctive. In these cases the animal does not wait to learn the games of his tribe by imitation, but starts-right-in on his own account. Besides this there are many forms of animal play which are not imitative at all. In these the animals co-operate, but do not take the same parts. The young perform actions in the game which the mother does not. All this goes to support another and most serious objection to this theory--in the mind of all those who believe in the doctrine of evolution. The Surplus-Energy Theory considers the play-impulse, which is one of the most widespread characters of animal life, as merely an accidental thing or by-product--a mere using-up of surplus energies. It is not in any way important to the animals. This makes it impossible to say that play has come to be the very complex thing that it really is by the laws of evolution; for survival by natural selection always supposes that the attribute or character which survives is important enough to keep the animal alive in the struggle for existence; otherwise it would not be continued for successive generations, and gradually perfected on account of its utility. On the whole, therefore, we find the Surplus-Energy Theory of play quite inadequate.