Now I say that all these several component parts of Darwinian doctrine are not matters of theory, but matters of fact. The only element of theory in his doctrine of evolution by natural selection has reference to the degree in which these observable facts, when thus brought together, are adequate to account for the process of evolution.
* * * * *
So much, then, as a statement of the theory of natural selection. But from this statement—i. e. from the theory of natural selection itself—there follow certain matters of general principle which it is important to bear in mind. These, therefore, I shall here proceed to mention.
First of all, it is evident that the theory is applicable as an explanation of organic changes in specific types only in so far as these changes are of use, or so far as such changes endow the species with better chances of success in the general struggle for existence. This is the only sense in which I shall always employ the terms use, utility, service, benefit, and so forth—that is to say, in the sense of life-preserving.
* * * * *
Next, it must be clearly understood that the life which it is the object, so to speak, of natural selection to preserve, is primarily the life of the species; not that of the individual. Natural selection preserves the life of the individual only in so far as this is conducive to that of the species. Wherever the life-interests of the individual clash with those of the species, that individual is sacrificed in favour of others who happen better to subserve the interests of the species. For example, in all organisms a greater or less amount of vigour is wasted, so far as individual interests are concerned, in the formation and the nourishment of progeny. In the great majority of plants and animals an enormous amount of physiological energy is thus expended. Look at the roe or the milt of a herring, for instance, and see what a huge drain has been made upon the individual for the sake of its species. Again, all unselfish instincts have been developed for the sake of the species, and usually against the interests of the individual. An ant which will allow her head to be slowly drawn from her body rather than relinquish her hold upon a pupa, is clearly acting in response to an instinct which has been developed for the benefit of the hive, though fatal to the individual. And, in a lesser degree, the parental instincts, wherever they occur, are more or less detrimental to the interests of the individual, though correspondingly essential to those of the race.
These illustrations will serve to show that natural selection always works primarily for the life-interests of the species—and, indeed, only works for those of the individual at all in so far as the latter happen to coincide with the former. Or, otherwise stated, the object of natural selection is always that of producing and maintaining specific types in the highest degree of efficiency, no matter what may become of the constituent individuals. Which is a striking republication by Science of a general truth previously stated by Poetry:—
So careful of the type she seems, So careless of the single life.
Tennyson thus noted the fact, and a few years later Darwin supplied the explanation.
But of course in many, if not in the majority of cases, anything that adds to the life-sustaining power of the single life thereby ministers also to the life-sustaining power of the type; and thus we can understand why all mechanisms and instincts which minister to the single life have been developed—namely, because the life of the species is made up of the lives of all its constituent individuals. It is only where the interests of the one clash with those of the other that natural selection works against the individual. So long as the interests are coincident, it works in favour of both.
Natural selection, then, is a theory which seeks to explain by natural causes the occurrence of every kind of adaptation which is to be met with in organic nature, on the assumption that adaptations of every kind have primary reference to the preservation of species, and therefore also, as a general rule, to the preservation of their constituent individuals. And from this it follows that where it is for the benefit of a species to change its type, natural selection will effect that change, thus leading to a specific transmutation, or the evolution of a new species. In such cases the old species may or may not become extinct. If the transmutation affects the species as a whole, or throughout its entire range, of course that particular type becomes extinct, although it does so by becoming changed into a still more suitable type in the course of successive generations. If, on the other hand, the transmutation affects only a part of the original species, or not throughout its entire range, then the other parts of that species may survive for any number of ages as they originally were. In the one case there is a ladder-like transmutation of species in time; in the other case a possibly tree-like multiplication of species in space. But whether the evolution of species be thus serial in time or divergent in space, the object of natural selection, so to speak, is in either case the same—namely, that of preserving all types which prove best suited to the conditions of their existence.
* * * * *
Once more, the term "struggle for existence" must be understood to comprehend, not only a competition for life among contemporary individuals of the same species, but likewise a struggle by all such individuals taken collectively for the continuance of their own specific type. Thus, on the one hand, while there is a perpetual civil war being waged between members of the same species, on the other hand there is a foreign war being waged by the species as a whole against its world as a whole. Hence it follows that natural selection does not secure survival of the fittest as regards individuals only, but also survival of the fittest as regards types. This is a most important point to remember, because, as a general rule, these two different causes produce exactly opposite effects. Success in the civil war, where each is fighting against all, is determined by individual fitness and self-reliance. But success in the foreign war is determined by what may be termed tribal fitness and mutual dependence. For example, among social insects the struggle for existence is quite as great between different tribes or communities, as it is between different individuals of the same community; and thus we can understand the extraordinary degree in which not only co-operative instincts, but also largely intelligent social habits, have here been developed. Similarly, in the case of mankind, we can understand the still more extraordinary development of these things—culminating in the moral sense. I have heard a sermon, preached at one of the meetings of the British Association, entirely devoted to arguing that the moral sense could not have been evolved by natural selection, seeing that the altruism which this sense involves is the very opposite of selfishness, which alone ought to have been the product of survival of the fittest in a struggle for life. And, of course, this argument would have been perfectly sound had Darwin limited the struggle for existence to individuals, without extending it to communities. But if the preacher had ever read Darwin's works he would have found that, when thus extended, the principle of natural selection is bound to work in favour of the co-operative instincts in the case of so highly social an animal as man; and that of these instincts conscience is the highest imaginable exhibition.
 For cases, see Animal Intelligence, in the chapters on Ants and Bees; and, for discussion of principles, Mental Evolution in Animals, in the chapters on Instinct.
What I have called tribal fitness—in contradistinction to individual fitness—begins with the family, developes in the community (herd, hive, clan, &c.), and usually ends with the limits of the species. On the one hand, however, it is but seldom that it extends so far as to embrace the entire species; while, on the other hand, it may in some cases, and as it were sporadically, extend beyond the species. In these latter cases members of different species mutually assist one another, whether in the way of what is called symbiosis, or in a variety of other ways which I need not wait to mention. For the only point which I now desire to make clear is, that all cases of mutual aid or co-operation, whether within or beyond the limits of species, are cases which fall under the explanatory sweep of the Darwinian theory.
 Prince Kropotkin in the Nineteenth Century (Feb. 1888, Apr. 1891) has adduced a large and interesting body of facts, showing the great prevalence of the principle of co-operation in organic nature.
* * * * *
Another important point to notice is, that it constitutes no part of the theory of natural selection to suppose that survival of the fittest must invariably lead to improvement of type, in the sense of superior organization. On the contrary, if from change of habits or conditions of life an organic type ceases to have any use for previously useful organs, natural selection will not only allow these organs in successive generations to deteriorate—by no longer placing any selective premium upon their maintenance—but may even proceed to assist the agencies engaged in their destruction. For, being now useless, they may become even deleterious, by absorbing nutriment, causing weight, occupying space, &c., without conferring any compensating benefit. Thus we can understand why it is that parasites, for example, present the phenomena of what is called degeneration, i. e. showing by their whole structure that they have descended from a possibly very much higher type of organization than that which they now exhibit. Having for innumerable generations ceased to require their legs, their eyes, and so forth, all such organs of high elaboration have either disappeared or become vestigial, leaving the parasite as a more or less effete representative of its ancestry.
These facts of degeneration, as we have previously seen, are of very general occurrence, and it is evident that their importance in the field of organic evolution as a whole has been very great. Moreover, it ought to be particularly observed that, as just indicated, the facts may be due either to a passive cessation of selection, or to an active reversal of it. Or, more correctly, these facts are probably always due to the cessation of selection, although in most cases where species in a state of nature are concerned, the process of degeneration has been both hastened and intensified by the super-added influence of the reversal of selection. In the next volume I shall have occasion to recur to this distinction, when it will be seen that it is one of no small importance to the general theory of descent.
* * * * *
We may now proceed to consider certain misconceptions of the Darwinian theory which are largely, not to say generally, prevalent among supporters of the theory. These misconceptions, therefore, differ from those which fall to be considered in the next chapter, i. e. misconceptions which constitute grounds of objection to the theory.
* * * * *
Of all the errors connected with the theory of natural selection, perhaps the one most frequently met with—especially among supporters of the theory—is that of employing the theory to explain all cases of phyletic modification (or inherited change of type) indiscriminately, without waiting to consider whether in particular cases its application is so much as logically possible. The term "natural selection" thus becomes a magic word, or Sesame, at the utterance of which every closed door is supposed to be immediately opened. Be it observed, I am not here alluding to that merely blind faith in natural selection, which of late years has begun dogmatically to force this principle as the sole cause of organic evolution in every case where it is logically possible that the principle can have come into play. Such a blind faith, indeed, I hold to be highly inimical, not only to the progress of biological science, but even to the true interests of the natural selection theory itself. As to this I shall have a good deal to say in the next volume. Here, however, the point is, that the theory in question is often invoked in cases where it is not even logically possible that it can apply, and therefore in cases where its application betokens, not merely an error of judgment or extravagance of dogmatism, but a fallacy of reasoning in the nature of a logical contradiction. Almost any number of examples might be given; but one will suffice to illustrate what is meant. And I choose it from the writings of one of the authors of the selection theory itself, in order to show how easy it is to be cheated by this mere juggling with a phrase—for of course I do not doubt that a moment's thought would have shown the writer the untenability of his statement.
In his most recent work Mr. Wallace advances an interesting hypothesis to the effect that differences of colour between allied species, which are apparently too slight to serve any other purpose, may act as "recognition marks," whereby the opposite sexes are enabled at once to distinguish between members of their own and of closely resembling species. Of course this hypothesis can only apply to the higher animals; but the point here is that, supposing it to hold for them, Mr. Wallace proceeds to argue thus:—Recognition marks "have in all probability been acquired in the process of differentiation for the purpose of checking the intercrossing of allied forms," because "one of the first needs of a new species would be to keep separate from its nearest allies, and this could be more readily done by some easily seen external mark." Now, it is clearly not so much as logically possible that these recognition-marks (supposing them to be such) can have been acquired by natural selection, "for the purpose of checking intercrossing of allied forms." For the theory of natural selection, from its own essential nature as a theory, is logically exclusive of the supposition that survival of the fittest ever provides changes in anticipation of future uses. Or, otherwise stated, it involves a contradiction of the theory itself to say that the colour-changes in question were originated by natural selection, in order to meet "one of the first needs of a new species," or for the purpose of subsequently preventing intercrossing with allied forms. If it had been said that these colour-differentiations were originated by some cause other than natural selection (or, if by natural selection, still with regard to some previous, instead of prophetic, "purpose"), and, when so "acquired," then began to serve the "purpose" assigned, the argument would not have involved the fallacy which we are now considering. But, as it stands, the argument reverts to the teleology of pre-Darwinian days—or the hypothesis of a "purpose" in the literal sense which sees the end from the beginning, instead of a "purpose" in the metaphorical sense of an adaptation that is evolved by the very modifications which subserve it.
 Darwinism, pp. 218 and 227.
 Since the above was written Prof. Lloyd Morgan has published a closely similar notice of the passage in question. "This language," he says, "seems to savour of teleology (that pitfall of the evolutionist). The cart is put before the horse. The recognition-marks were, I believe, not produced to prevent intercrossing, but intercrossing has been prevented because of preferential mating between individuals possessing special recognition-marks. To miss this point is to miss an important segregation-factor."—(Animal Life and Intelligence, p. 103.) Again, on pp. 184-9, he furnishes an excellent discussion on the whole subject of the fallacy alluded to in the text, and gives illustrative quotations from other prominent Darwinians. I should like to add that Darwin himself has nowhere fallen into this, or any of the other fallacies, which are mentioned in the text.
* * * * *
Another very prevalent, and more deliberate, fallacy connected with the theory of natural selection is, that it follows deductively from the theory itself that the principle of natural selection must be the sole means of modification in all cases where modification is of an adaptive kind,—with the consequence that no other principle can ever have been concerned in the production of structures or instincts which are of any use to their possessors. Whether or not natural selection actually has been the sole means of adaptive modification in the race, as distinguished from the individual, is a question of biological fact; but it involves a grave error of reasoning to suppose that this question can be answered deductively from the theory of natural selection itself, as I shall show at some length in the next volume.
 Of course adaptive modifications produced in the individual lifetime, and not inherited, do not concern the question at all. In this and the following paragraphs, therefore, "adaptations," "adaptive modifications," &c., refer exclusively to such as are hereditary, i. e. phyletic.
* * * * *
A still more extravagant, and a still more unaccountable fallacy is the one which represents it as following deductively from the theory of natural selection itself, that all hereditary characters are "necessarily" due to natural selection. In other words, not only all adaptive, but likewise all non-adaptive hereditary characters, it is said, must be due to natural selection. For non-adaptive characters are taken to be due to "correlation of growth," in connexion with some of the adaptive ones—natural selection being thus the indirect means of producing the former wherever they may occur, on account of its being the direct and the only means of producing the latter. Thus it is deduced from the theory of natural selection itself,—1st, that the principle of natural selection is the only possible cause of adaptive modification: 2nd, that non-adaptive modifications can only occur in the race as correlated appendages to the adaptive: 3rd, that, consequently, natural selection is the only possible cause of modification, whether adaptive or non-adaptive. Here again, therefore, we must observe that none of these sweeping generalizations can possibly be justified by deductive reasoning from the theory of natural selection itself. Any attempt at such deductive reasoning must necessarily end in circular reasoning, as I shall likewise show in the second volume, where this whole "question of utility" will be thoroughly dealt with.
* * * * *
Once more, there is an important oversight very generally committed by the followers of Darwin. For even those who avoid the fallacies above mentioned often fail to perceive, that natural selection can only begin to operate if the degree of adaptation is already given as sufficiently high to count for something in the struggle for existence. Any adaptations which fall below this level of importance cannot possibly have been produced by survival of the fittest. Yet the followers of Darwin habitually speak of adaptative characters, which in their own opinion are subservient merely to comfort or convenience, as having been produced by such means. Clearly this is illogical; for it belongs to the essence of Darwin's theory to suppose, that natural selection can have no jurisdiction beyond the line where structures or instincts already present a sufficient degree of adaptational value to increase, in some measure, the expectation of life on the part of their possessors. We cannot speak of adaptations as due to natural selection, without thereby affirming that they present what I have elsewhere termed a "selection value."
* * * * *
Lastly, as a mere matter of logical definition, it is well-nigh self-evident that the theory of natural selection is a theory of the origin, and cumulative development, of adaptations, whether these be distinctive of species, or of genera, orders, families, classes, and sub-kingdoms. It is only when the adaptations happen to be distinctive of the first (or lowest) of these taxonomic divisions, that the theory which accounts for these adaptations accounts also for the forms which present them,—i. e. becomes also a theory of the origin of species. This, however, is clearly but an accident of particular cases; and, therefore, even in them the theory is primarily a theory of adaptations, while it is but secondarily a theory of the species which present them. Or, otherwise stated, the theory is no more a theory of the origin of species than it is of the origin of genera, families, and the rest; while, on the other hand, it is everywhere a theory of the adaptive modifications whereby each of these taxonomic divisions has been differentiated as such. Yet, sufficiently obvious as the accuracy of this definition must appear to any one who dispassionately considers it, several naturalists of high standing have denounced it in violent terms. I shall therefore have to recur to the subject at somewhat greater length hereafter. At present it is enough merely to mention the matter, as furnishing another and a curious illustration of the not infrequent weakness of logical perception on the part of minds well gifted with the faculty of observation. It may be added, however, that the definition in question is in no way hostile to the one which is virtually given by Darwin in the title of his great work. The Origin of Species by means of Natural Selection is beyond doubt the best title that could have been given, because at the time when the work was published the fact, no less than the method, of organic evolution had to be established; and hence the most important thing to be done at that time was to prove the transmutation of species. But now that this has been done to the satisfaction of naturalists in general, it is as I have said, curious to find some of them denouncing a wider definition of the principle of natural selection, merely because the narrower (or included) definition is invested with the charm of verbal associations.
 The question as to whether natural selection has been the only principle concerned in the origination of species, is quite distinct from that as to the accuracy of the above definition.
* * * * *
So much for fallacies and misconceptions touching Darwin's theory, which are but too frequently met with in the writings of its supporters. We must now pass on to mention some of the still greater fallacies and misconceptions which are prevalent in the writings of its opponents. And, in order to do this thoroughly, I shall begin by devoting the remainder of the present chapter to a consideration of the antecedent standing of the two theories of natural selection and supernatural design. This having been done, in the succeeding chapters I shall deal with the evidences for, and the objections against, the former theory.
* * * * *
Beginning, then, with the antecedent standing of these alternative theories, the first thing to be noticed is, that they are both concerned with the same subject-matter, which it is their common object to explain. Moreover, this subject-matter is clearly and sharply divisible into two great classes of facts in organic nature—namely, those of Adaptation and those of Beauty. Darwin's theory of descent explains the former by his doctrine of natural selection, and the latter by his doctrine of sexual selection. In the first instance, therefore, I shall have to deal only with the facts of adaptation, leaving for subsequent consideration the facts of beauty.
Innumerable cases of the adaptation of organisms to their surroundings being the facts which now stand before us to be explained either by natural selection or by supernatural intention, we may first consider a statement which is frequently met with—namely, that even if all such cases of adaptation were proved to be fully explicable by the theory of descent, this would constitute no disproof of the theory of design: all the cases of adaptation, it is argued, might still be due to design, even though they admit of being hypothetically accounted for by the theory of descent. I have heard an eminent Professor tell his class that the many instances of mechanical adaptation discovered and described by Darwin as occurring in orchids, seemed to him to furnish better proof of supernatural contrivance than of natural causes; and another eminent Professor has informed me that, although he had read the Origin of Species with care, he could see in it no evidence of natural selection which might not equally well have been adduced in favour of intelligent design. But here we meet with a radical misconception of the whole logical attitude of science. For, be it observed, this exception in limine to the evidence which we are about to consider does not question that natural selection may be able to do all that Darwin ascribes to it. The objection is urged against his interpretation of the facts merely on the ground that these facts might equally well be ascribed to intelligent design. And so undoubtedly they might, if we were all simple enough to adopt a supernatural explanation whenever a natural one is found sufficient to account for the facts. Once admit the irrational principle that we may assume the operation of higher causes where the operation of lower ones is sufficient to explain the observed phenomena, and all our science and all our philosophy are scattered to the winds. For the law of logic which Sir William Hamilton called the law of parsimony—or the law which forbids us to assume the operation of higher causes when lower ones are found sufficient to explain the observed effects—this law constitutes the only barrier between science and superstition. It is always possible to give a hypothetical explanation of any phenomenon whatsoever, by referring it immediately to the intelligence of some supernatural agent; so that the only difference between the logic of science and the logic of superstition consists in science recognising a validity in the law of parsimony which superstition disregards. Therefore one can have no hesitation in saying that this way of looking at the evidence in favour of natural selection is not a scientific or a reasonable way of looking at it, but a purely superstitious way. Let us take, as an illustration, a perfectly parallel case. When Kepler was unable to explain by any known causes the paths described by the planets, he resorted to a supernatural explanation, and supposed that every planet was guided in its movements by some presiding angel. But when Newton supplied a beautifully simple physical explanation, all persons with a scientific habit of mind at once abandoned the metaphysical one. Now, to be consistent, the above-mentioned Professors, and all who think with them, ought still to adhere to Kepler's hypothesis in preference to Newton's explanation; for, excepting the law of parsimony, there is certainly no other logical objection to the statement, that the movements of the planets afford as good evidence of the influence of guiding angels as they do of the influence of gravitation.
So much, then, for the illogical position that, granting the evidence in favour of natural descent and supernatural design to be equal and parallel, we should hesitate in our choice between the two theories. But, of course, if the evidence is supposed not to be equal and parallel—i. e. if it is supposed that the theory of natural selection is not so good a theory whereby to explain the facts of adaptation as is that of supernatural design,—then the objection is no longer the one which we are considering. It is quite another objection, and one which is not prima facie absurd. Therefore let us state clearly the distinct question which thus arises.
Innumerable cases of adaptation of organisms to their environments are the observed facts for which an explanation is required. To supply this explanation, two, and only two, hypotheses are in the field. Of these two hypotheses one is intelligent design manifested directly in special creation; the other is natural causation operating through countless ages of the past. Now, the adaptations in question involve an innumerable multitude of special mechanisms, in most cases even within the limits of any one given species; but when we consider the sum of all these mechanisms presented by organic nature as a whole, the mind must indeed be dull which does not feel astounded. For, be it further observed, these mechanical contrivances are, for the most part, no merely simple arrangements, which might reasonably be supposed due, like the phenomena of crystallization, to comparatively simple physical causes. On the contrary, they everywhere and habitually exhibit so deep-laid, so intricate, and often so remote an adaptation of means to ends, that no machinery of human contrivance can properly be said to equal their perfection from a mechanical point of view. Therefore, without question, the hypothesis which first of all they suggest—or suggest most readily—is the hypothesis of design. And this hypothesis becomes virtually the only hypothesis possible, if it be assumed—as it generally was assumed by natural theologians of the past,—that all species of plants and animals were introduced into the world suddenly. For it is quite inconceivable that any known cause, other than intelligent design, could be competent to turn out instantaneously any one of these intricate pieces of machinery, already adapted to the performance of its special function. But, on the other hand, if there is any evidence to show that one species becomes slowly transformed into another—or that one set of adaptations becomes slowly changed into another set as changing circumstances require,—then it becomes quite possible to imagine that a strictly natural causation may have had something to do with the matter. And this suggestion becomes greatly more probable when we discover, from geological evidence and embryological research, that in the history both of races and of individuals the various mechanisms in question have themselves had a history—beginning in the forms of most uniformity and simplicity, gradually advancing to forms more varied and complex, nowhere exhibiting any interruptions in their upward progress, until the world of organic machinery as we now have it is seen to have been but the last phase of a long and gradual growth, the ultimate roots of which are to be found in the soil of undifferentiated protoplasm.
 It is often objected to Darwin's terminology, that it embraces such words as "contrivance," "purpose," &c., which are strictly applicable only to the processes or the products of thought. But when it is understood that they are used in a neutral or metaphorical sense, I cannot see that any harm arises from their use.
Lastly, when there is supplied to us the suggestion of natural selection as a cause presumably adequate to account for this continuous growth in the number, the intricacy, and the perfection of such mechanisms, it is only the most unphilosophical mind that can refuse to pause as between the older hypothesis of design and the newer hypothesis of descent.
Thus it is clear that the a priori standing of the rival hypotheses of naturalism and supernaturalism in the case of all these pieces of organic machinery, is profoundly affected by the question whether they came into existence suddenly, or whether they did so gradually. For, if they all came into existence suddenly, the fact would constitute well-nigh positive proof in favour of supernaturalism, or creation by design; whereas, if they all came into existence gradually, this fact would in itself constitute presumptive evidence in favour of naturalism, or of development by natural causes. And, as shown in the previous chapters, the proof that all species of plants and animals came into existence gradually—or the proof of evolution as a fact—is simply overwhelming.
From a still more general point of view I may state the case in another way, by borrowing and somewhat expanding an illustration which, I believe, was first used by Professor Huxley. If, when the tide is out, we see lying upon the shore a long line of detached sea-weed, marking the level which is reached by full tide, we should be free to conclude that the separation of the sea-weed from the sand and the stones was due to the intelligent work of some one who intended to collect the sea-weed for manure, or for any other purpose. But, on the other hand, we might explain the fact by a purely physical cause—namely, the separation by the sea-waves of the sea-weed from the sand and stones, in virtue of its lower specific gravity. Now, thus far the fact would be explained equally well by either hypothesis; and this fact would be the fact of selection. But whether we yielded our assent to the one explanation or to the other would depend upon a due consideration of all collateral circumstances. The sea-weed might not be of a kind that is of any use to man; there might be too great a quantity of it to admit of our supposing that it had been collected by man; the fact that it was all deposited on the high-water-mark would in itself be highly suggestive of the agency of the sea; and so forth. Thus, in such a case any reasonable observer would decide in favour of the physical explanation, or against the teleological one.
Now the question whether organic evolution has been caused by physical agencies or by intelligent design is in precisely the same predicament. There can be no logical doubt that, theoretically at all events, the physical agencies which the present chapter is concerned with, and which are conveniently summed up in the term natural selection, are as competent to produce these so-called mechanical contrivances, and the other cases of adaptation which are to be met with in organic nature, as intelligent design could be. Hence, our choice as between these two hypotheses must be governed by a study of all collateral circumstances; that is to say, by a study of the evidences in favour of the physical explanation. To this study, therefore, we shall now address ourselves, in the course of the following chapters.
EVIDENCES OF THE THEORY OF NATURAL SELECTION.
I will now proceed to state the main arguments in favour of the theory of natural selection, and then, in the following chapter, the main objections which have been urged against it.
In my opinion, the main arguments in favour of the theory are three in number.
First, it is a matter of observation that the struggle for existence in nature does lead to the extermination of forms less fitted for the struggle, and thus makes room for forms more fitted. This general fact may be best observed in cases where an exotic species proves itself better fitted to inhabit a new country than is some endemic species which it exterminates. In Great Britain, for example, the so-called common rat is a comparatively recent importation from Norway, and it has so completely supplanted the original British rat, that it is now extremely difficult to procure a single specimen of the latter: the native black rat has been all but exterminated by the foreign brown rat. The same thing is constantly found in the case of imported species of plants. I have seen the river at Cambridge so choked with the inordinate propagation of a species of water-weed which had been introduced from America, that considerable expense had to be incurred in order to clear the river for traffic. In New Zealand the same thing has happened with the European water-cress, and in Australia with the common rabbit. So it is doubtless true, as one of the natives is said to have philosophically remarked, "the white man's rat has driven away our rat, the European fly drives away our fly, his clover kills our grass, and so will the Maoris disappear before the white man himself." Innumerable other cases to the same effect might be quoted; and they all go to establish the fact that forms less fitted to survive succumb in their competition with forms better fitted.
* * * * *
Secondly, there is a general consideration of the largest possible significance in the present connexion—namely, that among all the millions of structures and instincts which are so invariably, and for the most part so wonderfully, adapted to the needs of the species presenting them, we cannot find a single instance, either in the vegetable or animal kingdom, of a structure or an instinct which is developed for the exclusive benefit of another species. Now this great and general fact is to my mind a fact of the most enormous, not to say overwhelming, significance. The theory of natural selection has now been before the world for more than thirty years, and during that time it had stood a fire of criticism such as was never encountered by any scientific theory before. From the first Darwin invited this criticism to adduce any single instance, either in the vegetable or animal kingdom, of a structure or an instinct which should unquestionably be proved to be of exclusive use to any species other than the one presenting it. He even went so far as to say that if any one such instance could be shown he would surrender his whole theory on the strength of it—so assured had he become, by his own prolonged researches, that natural selection was the true agent in the production of adaptive structures, and, as such, could never have permitted such a structure to occur in one species for the benefit of another. Now, as this invitation has been before the world for so many years, and has not yet been answered by any naturalist, we may by this time be pretty confident that it never will be answered. How tremendous, then, is the significance of this fact in its testimony to Darwin's theory! The number of animal and vegetable species, both living and extinct, is to be reckoned by millions, and every one of these species presents on an average hundreds of adaptive structures,—at least one of which in many, possibly in most, if not actually in all cases, is peculiar to the species that presents it. In other words, there are millions of adaptive structures (not to speak of instincts) which are peculiar to the species presenting them, and also many more which are the common property of allied species: yet, notwithstanding this inconceivable profusion of adaptive structures in organic nature, there is no single instance that has been pointed out of the occurrence of such a structure save for the benefit of the species that presents it. Therefore, I say that this immensely large and general fact speaks with literally immeasurable force in favour of natural selection, as at all events one of the main causes of organic evolution. For the fact is precisely what we should expect if this theory is true, while upon no other theory can its universality and invariability be rendered intelligible. On the beneficent design theory, for instance, it is inexplicable that no species should ever be found to present a structure or an instinct having primary reference to the welfare of another species, when, ex hypothesi, such an endless amount of thought has been displayed in the creation of structures and instincts having primary reference to the species which present them. For how magnificent a display of divine beneficence would organic nature have afforded, if all—or even some—species had been so inter-related as to have ministered to each others wants. Organic species might then have been likened to a countless multitude of voices, all singing in one great harmonious psalm. But, as it is, we see absolutely no vestige of such co-ordination: every species is for itself, and for itself alone—an outcome of the always and everywhere fiercely raging struggle for life.
In order that the force of this argument may not be misapprehended, it is necessary to bear in mind that it is in no way affected by cases where a structure or an instinct is of primary benefit to its possessor, and then becomes of secondary benefit to some other species on account of the latter being able in some way or another to utilise its action. Of course organic nature is full of cases of this kind; but they only go to show the readiness which all species display to utilise for themselves everything that can be turned to good account in their own environments, and so, among other things, the structures and instincts of other animals. For instance, it would be no answer to Darwin's challenge if any one were to point to a hermit-crab inhabiting the cast-off shell of a mollusk; because the shell was primarily of use to the mollusk itself, and, so far as the mollusk is concerned, the fact of its shell being afterwards of a secondary use to the crab is quite immaterial. What Darwin's challenge requires is, that some structure or instinct should be shown which is not merely of such secondary or accidental benefit to another species, but clearly adapted to the needs of that other species in the first instance—such, for example, as would be the case if the tail of a rattle-snake were of no use to its possessor, while serving to warn other animals of the proximity of a dangerous creature; or, in the case of instincts, if it were true that a pilot-fish accompanies a shark for the purpose of helping the shark to discover food. Both these instances have been alleged; but both have been shown untenable. And so it has proved of all the other cases which thus far have been put forward.
Perhaps the most remarkable of all the allegations which ever have been put forward in this connexion are those that were current with regard to instincts before the publication of Darwin's work. These allegations are the most remarkable, because they serve to show, in a degree which I do not believe could be shown anywhere else, the warping power of preconceived ideas. A short time ago I happened to come across the 8th edition of the Encyclopaedia Britannica, and turned up the article on "Instinct" there, in order to see what amount of change had been wrought with regard to our views on this subject by the work of Darwin—the 8th edition of the Encyclopaedia Britannica having been published shortly before The Origin of Species by means of Natural Selection. I cannot wait to give any lengthy quotations from this representative exponent of scientific opinion upon the subject at that time; but its general drift may be appreciated if I transcribe merely the short concluding paragraph, wherein he sums up his general results. Here he says:—
It thus only remains for us to regard instinct as a mental faculty, sui generis, the gift of God to the lower animals, that man in his own person, and by them, might be relieved from the meanest drudgery of nature.
Now, here we have the most extraordinary illustration that is imaginable of the obscuring influence of a preconceived idea. Because he started with the belief that instincts must have been implanted in animals for the benefit of man, this writer, even when writing a purely scientific essay, was completely blinded to the largest, the most obvious, and the most important of the facts which the phenomena of instinct display. For, as a matter of fact, among all the many thousands of instincts which are known to occur in animals, there is no single one that can be pointed to as having any special reference to man; while, on the other hand, it is equally impossible to point to one which does not refer to the welfare of the animal presenting it. Indeed, when the point is suggested, it seems to me surprising how few in number are the instincts of animals which have proved to be so much as of secondary or accidental benefit to man, in the same way as skins, furs, and a whole host of other animal products are thus of secondary use to him. Therefore, this writer not only failed to perceive the most obvious truth that every instinct, without any single exception, has reference to the animal which presents it; but he also conceived a purely fictitious inversion of this truth, and wrote an essay to prove a statement which all the instincts in the animal kingdom unite in contradicting.
This example will serve to show, in a striking manner, not only the distance that we have travelled in our interpretation of organic nature between two successive editions of the Encyclopaedia Britannica, but also the amount of verification which this fact furnishes to the theory of natural selection. For, inasmuch as it belongs to the very essence of this theory that all adaptive characters (whether instinctive or structural) must have reference to their own possessors, we find overpowering verification furnished to the theory by the fact now before us—namely, that immediately prior to the enunciation of this theory, the truth that all adaptive characters have reference only to the species which present them was not perceived. In other words, it was the testing of this theory by the facts of nature that revealed to naturalists the general law which the theory, as it were, predicted—the general law that all adaptive characters have primary reference to the species which present them. And when we remember that this is a kind of verification which is furnished by millions of separate cases, the whole mass of it taken together is, as I have before said, overwhelming.
It is somewhat remarkable that the enormous importance of this argument in favour of natural selection as a prime factor of organic evolution has not received the attention which it deserves. Even Darwin himself, with his characteristic reserve, has not presented its incalculable significance; nor do I know any of his followers who have made any approach to an adequate use of it in their advocacy of his views. In preparing the present chapter, therefore, I have been particularly careful not to pitch too high my own estimate of its evidential value. That is to say, I have considered, both in the domain of structures and of instincts, what instances admit of being possibly adduced per contra, or as standing outside the general law that adaptive structures and instincts are of primary use only to their possessors. In the result I can only think of two such instances. These, therefore, I will now dispose of.
The first was pointed out, and has been fully discussed, by Darwin himself. Certain species of ants are fond of a sweet fluid that is secreted by aphides, and they even keep the aphides as we keep cows for the purpose of profiting by their "milk." Now the point is, that the use of this sweet secretion to the aphis itself has not yet been made out. Of course, if it is of no use to the aphis, it would furnish a case which completely meets Darwin's own challenge. But, even if this supposition did not stand out of analogy with all the other facts of organic nature, most of us would probably deem it prudent to hold that the secretion must primarily be of some use to the aphis itself, although the matter has not been sufficiently investigated to inform us of what this use is. For, in any case, the secretion is not of any vital importance to the ants which feed upon it: and I think but few impartial minds would go so far to save an hypothesis as to maintain, that the Divinity had imposed this drain upon the internal resources of one species of insect for the sole purpose of supplying a luxury to another. On the whole, it seems most probable that the fluid is of the nature of an excretion, serving to carry off waste products. Such, at all events, was the opinion at which Darwin himself arrived, as a result of observing the facts anew, and in relation to his theory.
* * * * *
The other instance to which I have alluded as seeming at first sight likely to answer Darwin's challenge is the formation of vegetable galls. The great number and variety of galls agree in presenting a more or less elaborate structure, which is not only foreign to any of the uses of plant-life, but singularly and specially adapted to those of the insect-life which they shelter. Yet they are produced by a growth of the plant itself, when suitably stimulated by the insects' inoculation—or, according to recent observations, by emanations from the bodies of the larvae which develop from the eggs deposited in the plant by the insect. Now, without question, this is a most remarkable fact; and if there were many more of the like kind to be met with in organic nature, we might seriously consider whether the formation of galls should not be held to make against the ubiquitous agency of natural selection. But inasmuch as the formation of galls stands out as an exception to the otherwise universal rule of every species for itself, and for itself alone, we are justified in regarding this one apparent exception with extreme suspicion. Indeed, I think we are justified in regarding the peculiar pathological effect produced in the plant by the secretions of the insect as having been in the first instance accidentally beneficial to the insects. Thus, if any other effect than that of a growing tumour had been produced in the first instance, or if the needs of the insect progeny had not been such as to have derived profit from being enclosed in such a tumour, then, of course, the inoculating instinct of these animals could not have been developed by natural selection. But, given these two conditions, and it appears to me there is nothing very much more remarkable about an accidental correlation between the effects of a parasitic larva on a plant and the needs of that parasite, than there is between the similarly accidental correlation between a hydated parasite and the nutrition furnished to it by the tissues of a warm-blooded animal. Doubtless the case of galls is somewhat more remarkable, inasmuch as the morbid growth of the plant has more concern in the correlation—being, in many instances, a more specialized structure on the part of a host than occurs anywhere else, either in the animal or vegetable world. But here I may suggest that although natural selection cannot have acted upon the plant directly, so as to have produced galls ever better and better adapted to the needs of the insect, it may have so acted upon the plants indirectly though the insects. For it may very well have been that natural selection would ever tend to preserve those individual insects, the quality of whose emanations tended to produce the form of galls best suited to nourish the insect progeny; and thus the character of these pathological growths may have become ever better and better adapted to the needs of the insects. Lastly, looking to the enormous number of relations and inter-relations between all organic species, it is scarcely to be wondered at that even so extraordinary an instance of correlation as this should have arisen thus by accident, and then have been perfected by such an indirect agency of natural selection as is here suggested.
 Note B.
* * * * *
The third general class of facts which tell so immensely in favour of natural selection as an important cause of organic evolution, are those of domestication. The art of the horticulturist, the fancier, the cattle-breeder, &c., consists in producing greater and greater deviations from a given wild type of plant or animal, in any particular direction that may be desired for purposes either of use or of beauty. Cultivated cereals, fruits, and flowers are known to have been all derived from wild species; and, of course, the same applies to all our domesticated varieties of animals. Yet if we compare a cabbage rose with a wild rose, a golden pippin apple with a crab, a toy terrier with any species of wild dog, not to mention any number of other instances, there can be no question that, if such differences had appeared in nature, the organisms presenting them would have been entitled to rank as distinct species—or even, in many cases, as distinct genera. Yet we know, as a matter of fact, that all these differences have been produced by a process of artificial selection, or pairing, which has been continuously practised by horticulturists and breeders through a number of generations. It is the business of these men to note the individual organisms which show most variation in the directions required, and then to propagate from these individuals, in order that the progeny shall inherit the qualities desired. The results thus become cumulative from generation to generation, until we now have an astonishing manifestation of useful qualities on the one hand, and of beautiful qualities on the other, according as the organisms have been thus bred for purposes of use or for those of beauty.
Now it is immediately obvious that in these cases the process of artificial selection is precisely analogous to that of natural selection (and of sexual selection which will be considered later on), in all respects save one: the utility or the beauty which it is the aim of artificial selection continually to enhance, is utility or beauty in relation to the requirements or to the tastes of man; whereas the utility or the beauty which is produced by natural selection and sexual selection has reference only to the requirements or the tastes of the organisms themselves. But, with the exception of this one point of difference, the processes and the products are identical in kind. Persevering selection by man is thus proved to be capable of creating what are virtually new specific types, and this in any required direction. Hence, when we remember how severe is the struggle for existence in nature, it becomes impossible to doubt that selection by nature is able to do at least as much as artificial selection in the way of thus creating new types out of old ones. Artificial selection, indeed, notwithstanding the many and marvellous results which it has accomplished, can only be regarded as but a feeble imitation of natural selection, which must act with so much greater vigilance and through such immensely greater periods of time. In a word, the proved capabilities of artificial selection furnish, in its best conceivable form, what is called an argument a fortiori in favour of natural selection.
Or, to put it in another way, it may be said that for thousands of years mankind has been engaged in making a gigantic experiment to test, as it were by anticipation, the theory of natural selection. For, although this prolonged experiment has been carried on without any such intention on the part of the experimenters, it is none the less an experiment in the sense that its results now furnish an overwhelming verification of Mr. Darwin's theory. That is to say, they furnish overwhelming proof of the efficacy of the selective principle in the modification of organic types, when once this principle is brought steadily and continuously to bear upon a sufficiently long series of generations.
In order to furnish ocular evidence of the value of this line of verification, I have had the following series of drawings prepared. Another and equally striking series might be made of the products of artificial selection in the case of plants; but it seems to me that the case of animals is more than sufficient for the purpose just stated. Perhaps it is desirable to add that considerable care has been bestowed upon the execution of these portraits; and that in every case the latter have been taken from the most typical specimens of the artificial variety depicted. Those of them which have not been drawn directly from life are taken from the most authoritative sources; and, before being submitted to the engraver, they were all examined by the best judges in each department. In none of the groups, however, have I aimed at an exhaustive representation of all the varieties: I have merely introduced representatives of as many as the page would in each case accommodate.
The exigencies of space have prevented, in some of the groups, strict adherence to a uniform scale—with the result that contrasts between different breeds in respect of size are not adequately rendered. This remark applies especially to the dogs; for although the artist has endeavoured to draw them in perspective, unless the distance between those in the foreground and those in the background is understood to be more considerable than it appears, an inadequate idea is given of the relative differences of size. The most instructive of the groups, I think, is that of the Canaries; because the many and great changes in different directions must in this case have been produced by artificial selection in so comparatively short a time—the first mention of this bird that I can find being by Gesner, in the sixteenth century.
* * * * *
Now, it is surely unquestionable that in these typical proofs of the efficacy of artificial selection in the modification of specific types, we have the strongest conceivable testimony to the power of natural selection in the same direction. For it thus appears that wherever mankind has had occasion to operate by selection for a sufficiently long time—that is to say, on whatever species of plant or animal he chooses thus to operate for the purpose of modifying the type in any required direction,—the results are always more or less the same: he finds that all specific types lend themselves to continuous deflection in any particulars of structure, colour, &c., that he may desire to modify.
Nevertheless, to this parallel between the known effects of artificial selection, and the inferred effects of natural selection, two objections have been urged. The first is, that in the case of artificial selection the selecting agent is a voluntary intelligence, while in the case of natural selection the selecting agent is Nature herself; and whether or not there is any counterpart of man's voluntary intelligence in nature is a question with which Darwinism has nothing to do. Therefore, it is alleged, the analogy between natural selection and artificial selection fails ab initio, or at the fountain-head of the causes which are taken by the analogy to be respectively involved.
The second objection to the analogy is, that the products of artificial selection, closely as they may resemble natural species in all other respects, nevertheless present one conspicuous and highly important point of difference: they rarely, if ever, present the physiological character of mutual infertility, which is a character of extremely general occurrence in the case of natural species, even when these are most nearly allied.
I will deal with these two objections in the next chapter, where I shall be concerned with the meeting of all the objections which have ever been urged against the theory of natural selection. Meanwhile I am engaged only in presenting the general arguments which support the theory, and therefore mention these objections to one of them merely en passant. And I do so in order to pledge myself effectually to dispose of them later on, so that for the purposes of my present argument both these objections may be provisionally regarded as non-existent; which means, in other words, that we may provisionally regard the analogy between artificial selection and natural selection as everywhere logically intact.
* * * * *
To sum up, then, the results of the foregoing exposition thus far, what I hold to be the three principal, or most general, arguments in favour of the theory of natural selection, are as follows.
First, there is the a priori consideration that, if on independent grounds we believe in the theory of evolution at all, it becomes obvious that natural selection must have had some part in the process. For no one can deny the potent facts of heredity, variability, the struggle for existence, and survival of the fittest. But to admit these facts is to admit natural selection as a principle which must be, at any rate, one of the factors of organic evolution, supposing such evolution to have taken place. Next, when we turn from these a priori considerations, which thus show that natural selection must have been concerned to some extent in the process of evolution, we find in organic nature evidence a posteriori of the extent to which this principle has been thus concerned. For we find that among all the countless millions of adaptive structures which are to be met with in organic nature, it is an invariable rule that they exist in relation to the needs of the particular species which present them: they never have any primary reference to the needs of other species. And as this extraordinarily large and general fact is exactly what the theory of natural selection would expect, the theory is verified by the fact in an extraordinarily cogent manner. In other words, the fact goes to prove that in all cases where adaptive structures or instincts are concerned, natural selection must have been either the sole cause at work, or, at the least, an influence controlling the operation of all other causes.
Lastly, an actually experimental verification of the theory has been furnished on a gigantic scale by the operations of breeders, fanciers, and horticulturists. For these men, by their process of selective accumulation, have empirically proved what immense changes of type may thus be brought about; and so have verified by anticipation, and in a most striking manner, the theory of natural selection—which, as now so fully explained, is nothing more than a theory of cumulative modifications by means of selective breeding.
So much, then, by way of generalities. But perhaps the proof of natural selection as an agency of the first importance in the transmutation of species may be best brought home to us by considering a few of its applications in detail. I will therefore devote the rest of the present chapter to considering a few cases of this kind.
There are so many large fields from which such special illustrations may be supplied, that it is difficult to decide which of them to draw upon. For instance, the innumerable, always interesting, and often astonishing adaptations on the part of flowers to the fertilising agency of insects, has alone given rise to an extensive literature since the time when Darwin himself was led to investigate the subject by the guidance of his own theory. The same may be said of the structures and movements of climbing plants, and in short, of all the other departments of natural history where the theory of natural selection has led to the study of the phenomena of adaptation. For in all these cases the theory of natural selection, which first led to their discovery, still remains the only scientific theory by which they can be explained. But among all the possible fields from which evidences of this kind may be drawn, I think the best is that which may be generically termed defensive colouring. To this field, therefore, I will restrict myself. But, even so, the cases to be mentioned are but mere samples taken from different divisions of this field; and therefore it must be understood at the outset that they could easily be multiplied a hundred-fold.
A vast number of animals are rendered more or less inconspicuous by resembling the colours of the surfaces on which they habitually rest. Such, for example, are grouse, partridges, rabbits, &c. Moreover, there are many cases in which, if the needs of the creature be such that it must habitually frequent surfaces of different colours, it has acquired the power of changing its colour accordingly—e. g. cuttle-fish, flat-fish, frogs, chameleons, &c. The physiological mechanism whereby these adaptive changes of colour are produced differs in different animals; but it is needless for our purposes to go into this part of the subject. Again, there are yet other cases where protective colouring which is admirably suited to conceal an animal through one part of the year, would become highly conspicuous during another part of it—namely, when the ground is covered with snow. Accordingly, in these cases the animals change their colour in the winter months to a snowy white: witness stoats, mountain hares, ptarmigan, &c. (Fig. 108.)
Now, it is sufficiently obvious that in all these classes of cases the concealment from enemies or prey which is thus secured is of advantage to the animals concerned; and, therefore, that in the theory of natural selection we have a satisfactory theory whereby to explain it. And this cannot be said of any other theory of adaptive mechanisms in nature that has ever been propounded. The so-called Lamarckian theory, for instance, cannot be brought to bear upon the facts at all; and on the theory of special creation it is unintelligible why the phenomena of protective colouring should be of such general occurrence. For, in as far as protective colouring is of advantage to the species which present it, it is of corresponding disadvantage to those other species against the predatory nature of which it acts as a defence. And, of course, the same applies to yet other species, if they serve as prey. Moreover, the more minutely this subject is investigated in all its details, the more exactly is it found to harmonise with the naturalistic interpretation.
 Were it not that some of Darwin's critics have overlooked the very point wherein the great value of protective colouring as evidence of natural selection consists, it would be needless to observe that it does so in the minuteness of the protective resemblance which in so many cases is presented. Of course where the resemblance is only very general, the phenomena might be ascribed to mere coincidence, of which the instincts of the animal have taken advantage. But in the measure that the resemblance becomes minutely detailed, the supposition of mere coincidence is excluded, and the agency of some specially adaptive cause demonstrated. Again, it is almost needless to say, no real difficulty is presented (as has been alleged) by the cases above quoted of seasonal imitations, on the ground that natural selection could not act alternately on the same individual. Natural selection is not supposed to act alternately on the same individual. It is supposed to act always in the same manner, and if, as in the case of a regularly recurring change in the colours of the environment, correspondingly recurrent changes are required to appear in the colours of the animals, natural selection sets its premium upon those individuals the constitutions of which best lend themselves to seasonal changes of the needful kind—probably under the influence of stimuli supplied by the changes of external conditions (temperature, moisture, &c.).
In the first place, we always find a complete correspondence between imitative colouring and instinctive endowment. If a caterpillar exactly resembles the colour of a twig, it also presents the instinct of habitually reposing in the attitude which makes it most resemble a twig—standing out from the branch on which it rests at the same angle as is presented by the real twigs of the tree on which it lives.
Here, again, is a bird protectively coloured so as to resemble stones upon the rough ground where it habitually lives; and the drawing shows the attitude in which the bird instinctively reposes, so as still further to increase its resemblance to a stone. (Fig. 109.)
[Illustration: FIG. 109.—Oedicnemus crepitans, showing the instinctive attitude of concealment. Drawn from a stuffed specimen in the British Museum, 1/6 nat. size, with appropriate surroundings supplied.
To take only one other instance, hares and rabbits, like grouse and partridges—or like the plover just alluded to,—instinctively crouch upon those surfaces the colours of which they resemble; and I have often remarked that if, on account of any individual peculiarity of coloration, the animal is not able thus to secure concealment, it nevertheless exhibits the instinct of crouching which is of benefit to all its kind, although, from the accident of its own abnormal colouring, this instinct is then actually detrimental to the animal itself. For example, every sportsman must have noticed that the somewhat rare melanic variety of the common rabbit will crouch as steadily as the normal brownish-gray type, notwithstanding that, owing to its abnormal colour, a "nigger-rabbit" thus renders itself the most conspicuous object in the landscape. In all such cases, of course, there has been a deviation from the normal type in respect of colour, with the result that the inherited instinct is no longer in tune with the other endowments of the animal. Such a variation of colour, therefore, will tend to be suppressed by natural selection; while any variations which may bring the animal still more closely to resemble its habitual surroundings will be preserved. Thus we can understand the truly wonderful extent to which this principle of protective colouring has been carried in many cases where the need of it has been most urgent.
Not only colour, but structure, may be profoundly modified for the purposes of protective concealment. Thus, caterpillars which resemble twigs do so not only in respect of colour, but also of shape; and this even down to the most minute details in cases where the adaptation is most complete: certain butterflies and leaf-insects so precisely resemble the leaves upon which, or among which, they live, that it is almost impossible to detect them in the foliage—not only the colour, the shape, and the venation being all exactly imitated, but in some cases even the defects to which the leaves are liable, in the way of fungoid growths, &c. There are other insects which with similar exactness resemble moss, lichens, and so forth. A species of fish secures a complete resemblance to bunches of sea-weed by a frond-like modification of all its appendages, and so on through many other instances. Now, in all such cases where there is so precise an imitation, both in colour and structure, it seems impossible to suggest any other explanation of the facts than the one which is supplied by Mr. Darwin's theory—namely, that the more perfect the resemblance is caused to become through the continuous influence of natural selection always picking out the best imitations, the more highly discriminative becomes the perception of those enemies against the depredations of which this peculiar kind of protection is developed; so that, in virtue of this action and re-action, eventually we have a degree of imitation which renders it almost impossible for a naturalist to detect the animal when living in its natural environment.
In strange and glaring contrast to all these cases of protective colouring, stand other cases of conspicuous colouring. Thus, for example, although there are numberless species of caterpillars which present in an astonishing degree the phenomena of protective colouring, there are numberless other species which not only fail to present these phenomena in any degree, but actually go to the opposite extreme of presenting colours which appear to have been developed for the sake of their conspicuousness. At all events, these caterpillars are usually the most conspicuous objects in their surroundings, and therefore in the early days of Darwinism they were regarded by Darwin himself as presenting a formidable difficulty in the way of his theory. To Mr. Wallace belongs the merit of having cleared up this difficulty in an extraordinarily successful manner. He virtually reasoned thus. If the raison d'etre of protective colouring be that of concealing agreeably flavoured caterpillars from the eye-sight of birds, may not the raison d'etre of conspicuous colouring be that of protecting disagreeably flavoured caterpillars from any possibility of being mistaken by birds? Should this be the case, of course the more conspicuous the colouring the better would it be for the caterpillars presenting it. Now as soon as this suggestion was acted upon experimentally, it was found to be borne out by facts. Birds could not be induced to eat caterpillars of the kinds in question; and there is now no longer any doubt that their conspicuous colouring is correlated with their distastefulness to birds, in the same way as the inconspicuous or imitative colouring of other caterpillars is correlated with their tastefulness to birds. Here then is yet another instance, added to those already given, of the verification yielded to the theory of natural selection by its proved competency as a guide to facts in nature; for assuredly this particular class of facts would never have been suspected but for its suggestive agency.
As in the case of protective imitation, so in this case of warning conspicuousness, not only colour, but structure may be greatly modified for the purpose of securing immunity from attack. Here, of course, the object is to assume, as far as possible, a touch-me-not appearance; so that, although destitute of any real means of offence, the creatures in question present a fictitiously dangerous aspect. As the Devil's-coach-horse turns up his stingless tail when threatened by an enemy, so in numberless ways do many harmless animals of all classes pretend to be formidable. But the point now is that these instincts of self-defence are often helped out by structural modifications, expressly and exclusively adapted to this end. For example, what a remarkable series of protective adjustments occurs in the life-history of the Puss Moth—culminating with so comical an instance of the particular device now under consideration as the following. I quote the facts from Mr. E. B. Poulton's admirable book on The Colours of Animals (pp. 269-271).
The larva of the Puss Moth (Cerura vinula) is very common upon poplar and willow. The circular dome-like eggs are laid, either singly or in little groups of two or three, upon the upper side of the leaf, and being of a reddish colour strongly suggest the appearance of little galls, or the results of some other injury to the leaf. The youngest larvae are black, and also rest upon the upper surface of the leaf, resembling the dark patches which are commonly seen in this position. As the larva grows, the apparent black patch would cover too large a space, and would lead to detection if it still occupied the whole surface of the body. The latter gains a green ground-colour which harmonises with the leaf, while the dark marking is chiefly confined to the back. As growth proceeds the relative amount of green increases, and the dark mark is thus prevented from attaining a size which would render it too conspicuous. In the last stage of growth the green larva becomes very large, and usually rests on the twigs of its food-plant (Fig. 111). The dark colour is still present on the back but is softened to a purplish tint, which tends to be replaced by a combination of white and green in many of the largest larvae. Such a larva is well concealed by General Protective Resemblance, and one may search a long time before finding it, although assured of its presence from the stripped branches of the food-plant and the faeces on the ground beneath.
As soon as a large larva is discovered and disturbed it withdraws its head into the first body-ring, inflating the margin, which is of a bright red colour. There are two intensely black spots on this margin in the appropriate position for eyes, and the whole appearance is that of a large flat face extending to the outer edge of the red margin (see Fig. 112). The effect is an intensely exaggerated caricature of a vertebrate face, which is probably alarming to the vertebrate enemies of the caterpillar. The terrifying effect is therefore mimetic. The movements entirely depend on tactile impressions: when touched ever so lightly a healthy larva immediately assumes the terrifying attitude, and turns so as to present its full face towards the enemy; if touched on the other side or on the back it instantly turns its face in the appropriate direction. The effect is also greatly strengthened by two pink whips which are swiftly protruded from the prongs of the fork in which the body terminates. The prongs represent the last pair of larval legs which have been greatly modified from their ordinary shape and use. The end of the body is at the same time curved forward over the back (generally much further than in Fig. 112), so that the pink filaments are brandished above the head.
Lastly, these facts as to imitative and conspicuous colouring lead on to the yet more remarkable facts of what is called mimicry. By mimicry is meant the imitation in form and colour of one species by another, in order that the imitating species may be mistaken for the imitated, and thus participate in some advantage which the latter enjoys. For instance, if, as in the case of the conspicuously-coloured caterpillars, it is of advantage to an ill-savoured species that it should hold out a warning to enemies, clearly it may be of no less advantage to a well-savoured species that it should borrow this flag, and thus be mistaken for its ill-savoured neighbour. Now, the extent to which this device of mimicry is carried is highly remarkable, not only in respect of the number of its cases, but also in respect of the astonishing accuracy which in most of these cases is exhibited by the imitation. There need be little or virtually no zoological affinity between the imitating and the imitated forms; that is to say, in some cases the zoological affinity is not closer than ordinal, and therefore cannot possibly be ascribed to kinship. Like all the other branches of the general subject of protective resemblance in form or colouring, this branch has already been so largely illustrated by previous writers, that, as in the previous cases, I need only give one or two examples. Those which I choose are chosen on account of the colours concerned not being highly varied or brilliant, and therefore lending themselves to less ineffectual treatment by wood-engraving than is the case where attempts are made to render by this means even more remarkable instances. (Figs. 113, 114, 115.)
It is surely apparent, without further comment, that it is impossible to imagine stronger evidence in favour of natural selection as a true cause in nature, than is furnished by this culminating fact in the matter of protective resemblance, whereby it is shown that a species of one genus, family, or even order, will accurately mimic the appearance of a species belonging to another genus, family, or order, so as to deceive its natural enemies into mistaking it for a creature of so totally different a kind. And it must be added that while this fact of mimicry is of extraordinarily frequent occurrence, there can be no possibility of our mistaking its purpose. For the fact is never observable except in the case of species which occupy the same area or district.
Such being what appears to me the only reasonable view of the matter, I will now conclude this chapter on the evidences of natural selection as at all events the main factor of organic evolution, by simply adding illustrations of two further cases of mimicry, which are perhaps even more remarkable than any of the foregoing examples. The first of the two (Fig. 115) speaks for itself. The second will be rendered intelligible by the following few words of explanation.
There are certain ants of the Amazons which present the curious instinct of cutting off leaves from trees, and carrying them like banners over their heads to the hive, as represented in Fig. 116, B, where one ant is shown without a leaf, and the others each with a leaf. Their object in thus collecting leaves is probably that of growing a fungus upon the "soil" which is furnished by the leaves when decomposing. But, be this as it may, the only point we are now concerned with is the appearance which these ants present when engaged in their habitual operation of carrying leaves. For it has been recently observed by Mr. W. L. Sclater, that in the localities where these hymenopterous insects occur, there occurs also a homopterous insect which mimics the ant, leaf and all, in a wonderfully deceptive manner. The leaf is imitated by the thin flattened body of the insect, "which in its dorsal aspect is so compressed laterally that it is no thicker than a leaf, and terminates in a sharp jagged edge." The colour is exactly the same as that of a leaf, and the brown legs show themselves beneath the green body in just the same way as those of the ant show themselves beneath the leaf. So that both the form and the colouring of the homopterous insect has been brought to resemble, with singular exactness, those belonging to a different order of insect, when the latter is engaged in its peculiar avocation. A glance at the figure is enough to show the means employed and the result attained. In A, an ant and its mimic are represented as about 2-1/2 times their natural size, and both proceeding in the same direction. It ought to be mentioned, however, that in reality the margin of the leaf is seldom allowed to retain its natural serrations as here depicted: the ants usually gnaw the edge of the real leaf, so that the margin of the false one bears an even closer resemblance to it than the illustration represents. B is a drawing from life of a group of five ants carrying leaves, and their mimic walking beside them.
 For a full account of this instinct and its probable purpose, see Animal Intelligence, pp. 93-6.
 Both drawings are reproduced from Mr. Poulton's paper upon the subject (Proc. Zool. Soc., June 16, 1891).
CRITICISMS OF THE THEORY OF NATURAL SELECTION.
I will now proceed to consider the various objections and difficulties which have hitherto been advanced against the theory of natural selection.
Very early in the day Owen hurled the weight of his authority against the new theory, and this with a strength of onslaught which was only equalled by its want of judgment. Indeed, it is painfully apparent that he failed to apprehend the fundamental principles of the Darwinian theory. For he says:—
Natural Selection is an explanation of the process [of transmutation] of the same kind and value as that which has been proffered of the mystery of "secretion." For example, a particular mass of matter in a living animal takes certain elements out of the blood, and rejects them as "bile." Attributes were given to the liver which can only be predicated of the whole animal; the "appetency" of the liver, it was said, was for the elements of bile, and "biliosity," or the "hepatic sensation," guided the gland to their secretion. Such figurative language, I need not say, explains absolutely nothing of the nature of bilification.
 Anatomy of Vertebrates, vol. iii. p. 794.
Assuredly, it was needless for Owen to say that figurative language of this kind explains nothing; but it was little less than puerile in him to see no more in the theory of natural selection than such a mere figure of speech. To say that the liver selects the elements of bile, or that nature selects specific types, may both be equally unmeaning re-statements of facts; but when it is explained that the term natural selection, unlike that of "hepatic sensation," is used as a shorthand expression for a whole group of well-known natural causes—struggle, variation, survival, heredity,—then it becomes evidence of an almost childish want of thought to affirm that the expression is figurative and nothing more. The doctrine of natural selection may be a huge mistake; but, if so, this is not because it consists of any unmeaning metaphor: it can only be because the combination of natural causes which it suggests is not of the same adequacy in fact as it is taken to be in theory.
Owen further objected that the struggle for existence could only act as a cause of the extinction of species, not of their origination—a view of the case which again shows on his part a complete failure to grasp the conception of Darwinism. Acting alone, the struggle for existence could only cause extermination; but acting together with variation, survival, and heredity, it may very well—for anything that Owen, or others who followed in this line of criticism, show to the contrary—have produced every species of plant and animal that has ever appeared upon the face of the earth.