living things, and there is abundance of opening for discovery with respect to the laws of matter and the laws of life; but matter and life must first be given: this is sufficiently obvious; but it is worth noting, because there is sometimes a tendency to make a confusion between creation and the laws of created things; whereas it is obvious that creation is one thing, and the law governing created things is quite another. But (2) as the original existence of living things is a mystery, so also is the reproduction of them. The continuity of life on the earth's surface, insured in various ways more or less resembling each other, and all agreeing in this, that there is apparently no tendency in vital power to degenerate or wear itself out in the course of ages, is, as it were, a standing mystery of creation. The scientific man has nothing to do with this mystery; to him it is simply a fact or phenomenon; but he who tries to go beyond phenomena and to get at the cause behind them will recognize reproduction as being etiologically equivalent to continuous creation. The great feature, however, of the principle of natural selection is (3) the occurrence of variations. Mr. Wallace lays great stress on the abundance of the variations which occur in nature, and the corresponding importance of this element in the Darwinian theory; and he is obviously wise in doing so. But it is well to observe that it is impossible to regard variations either on the one hand as a necessary feature of reproduction, or on the other as simply fortuitous. With regard to the latter supposition it is, certainly, difficult to conceive of chance as being a principal factor, say, in the production of a horse, to say nothing of a man. But even the former supposition is not quite an easy one: it is difficult to see why variations capable of being made permanent should occur, and why (if there be offspring at all) the offspring should not be exactly like the parent; in not a few cases this seems to be the law of living things. What I wish to point out, however, is this, that from the etiological point of view there ought to be a cause for variations as well as for other phenomena; and that, therefore, when we use the phenomenon of variations as a part of the machinery of natural selection, we do not get rid of the task of inquiring, as philosophers, why those useful variations occurred. In fact, in this as in many other instances, what is done is to shift the process one stage backward, but to leave the question of the primary cause very much where it was. Variations are abundant, says the student of natural history, and advantageous variations are preserved and made permanent by the process of natural selection: let it be granted. But the philosopher may still say: How comes it that advantageous variations should occur? Must not this occurrence be the result of some pre-established principle or law of development? Take the case of the horse, which Mr. Wallace has dwelt upon
