The species is therefore a group of organisms all of which exhibit the same morphological characters. This sameness is not absolute, for the individuals composing the species may vary from each other with respect to any one character. But the range of these variations is limited. They fluctuate about an imaginary mean value which remains constant in the case of a species which is not undergoing selection, and is therefore nearly the same throughout a series of generations. The formal characters which we regard as diagnostic of the species are these imaginary mean ones.

It is possible to breed from stock a very great number of animals, all of which are connected by a tie of blood-relationship, that is, all have descended from the same ancestor or ancestors. Such an assemblage of animals would resemble those assemblages living in the wild which we call species, in that a certain morphological similarity would be exhibited by all the individuals. If the breeding were conducted so as to avoid selection, the range of variability would be very much the same as that observed in the wild race. The two groups of animals—that bred artificially, and that observed in natural conditions—would be very much alike, and it is impossible to resist the conclusion that the natural race, like the artificial one, is a family in the human sense, that is, all the individuals composing it are connected together by a tie of common descent.

Let us extend this reasoning to categories of organisms of higher orders than species. We can associate together groups of species in the same way that we associate together the individuals of the same species. There are certain morphological characters which are common to all the species in the category, but there are also differences between specific group and specific group, and these differences may be regarded as variations from the generic morphological type. All the Cats, for instance, have certain characters in common: fully retractile claws, a certain kind of dentition, certain cranial characters, and so on. We postulate a feline type of structure, and we then regard the characters displayed by the cat, lion, tiger, leopard, etc., as deviations from this feline morphological type. Thus we establish the Family FelidÆ. But again we find that the FelidÆ together with the CanidÆ, and many other species of animals, also display common characters, dental and osteological chiefly, and we express this resemblance by assembling all these families in one Order, the Carnivora. The Carnivores, however, are only one large group of Quadrupeds: there are many others, such as the Rodents, Ungulates, Cetacea, etc., and all of these possess common characters. In all of them the integument is provided with hairs, or other similarly developed structures; all breathe by means of a diaphragm; in all, the young are nourished by suckling the mammÆ of the mother; and all develop on a placenta. We therefore group them all in the Class Mammalia. Now the Mammals possess an internal skeleton of which the most fundamental part is an axial rod—the notochord—developing to form a vertebral column; and this notochordal skeleton is also possessed by the Birds, Reptiles, Amphibia, and Fishes. There are also some smaller groups in which the notochord is present but does not develop to form segmented vertebrÆ. Including these, we are able to form a large category of animals—the Chordata—and this phylum is sharply distinguished from all other cognate groups.

All animals and plants may be classified in a similar way. Insects, Spiders, and Crustacea, for instance, are all animals in which the body is jointed, each joint or segment being typically provided with a pair of jointed appendages or limbs. Because of this similarity of fundamental structure we include all these animals, with some others, in one phylum, the Arthropoda. So also with the rest of the animal kingdom, and similar methods may be extended to the classification of the plants. A few small groups in each of the kingdoms are difficult to classify, but it has been possible to arrange most living organisms in a small number of sub-kingdoms or phyla, and even to attempt to trace relationships between these various categories.

The mere systematic description of the organic world would have resulted in such a reasoned classification apart altogether from any notions of an evolutionary process. But the classification, originally a conventional way of making a list of organisms, would at once suggest morphological similarities. It would suggest that all the Cats were Carnivores, that all the Carnivores were Mammals, and that all the Mammals were Chordates. It would suggest that all Wasps were Hymenoptera, that all Hymenoptera were Insects, and that all Insects were Arthropods. It would establish a host of logical relations between animals of all kinds.

It would show us a number of groups of animals separated from each other by morphological dissimilarities. But let us also consider all those animals which lived in the past of the earth, and the remains of which are found in the rocks as fossils. Including all the forms of life known to PalÆontology, we should find that the dissimilarities between the various groups would tend to disappear. The gaps between existing Birds and Reptiles, for instance, would become partially bridged. PalÆontology would also supplement morphology in another way. The study of the structure of animals leads us to describe them as “higher” and “lower”—higher in the sense of a greater complexity of structure. Thus the body of a Carnivore is more complex than that of a Fish, inasmuch as it possesses the homologues of the truly piscine gills, but it also possesses a four-chambered heart instead of a two-chambered one; and it possesses the mammalian lungs, diaphragm, and placenta, structures which are not present in the Fish. Now, so far as its imperfect materials go, palÆontology shows us that the higher forms of life appeared on the earth at a later date than did the lower forms. The remains of Mammals, for instance, are first found in rocks which are younger than (that is, they are superposed upon) those rocks in which Reptiles first appear; and so also Reptiles appear later in the rock series than do Fishes. PalÆontology thus adds to the logical order suggested by morphology a chronological order of this nature: higher, or more complex forms of life appeared at a later date in the history of the earth than did lower or less complex ones.

A parallel chronological sequence would also be suggested by the results of embryology. This branch of biology shows us that all animals pass through a series of stages in their individual development, or ontogeny. The earlier stages represent a simple type of structure, usually a hollow ball of cells, but as development proceeds, the structure of the embryo becomes more and more complex. The process of development is continuous in many animals, but in others (perhaps in most) larval stages appear, that is, development is interrupted, and the animal may lead for a time an independent existence similar to that of the fully developed form. Often these larval stages suggest types of structure lower than that of the fully developed animal into which they transform. Even if larval stages may not appear in the ontogeny, it is very often the case that the developing embryo exhibits traces, or at least reminiscences, of the types of morphology characteristic of the animals which are lower or less complex than itself; thus the piscine gills appear during the development of the tailed Amphibian, and even in that of the Mammal, and then vanish, or are converted into organs of another kind. The individual thus passes through a series of developmental stages of increasing complexity: it repeats, in its ontogeny, the palÆontological sequence in a distorted and abbreviated form.

It is true that the evidence afforded by palÆontology is very meagre. The preservation of the remains of organisms in the stratified rocks is a very haphazard process, and it depends for its success on a series of conditions that are not always present. As the surface of the earth becomes better known, our knowledge of the life of the past will become fuller, but there can be little doubt that whole series of organisms must have existed in the past, and that no recognisable traces of these are known to us. There is also no doubt that the sequences indicated by palÆontology are very incomplete: they are obscured and shortened by many conditions. The earlier embryologists entertained hopes that the study of embryology would reveal the direction of the evolutionary process in many groups of animals: if the organism repeats in its ontogeny the series of stages through which it passed in its phylogenetic development, then a close study of the embryological process ought to disclose these stages. Although these hopes have not been realised, there is yet sufficient truth in the doctrine of recapitulation to enable us to state that there is a rough parallelism between the palÆontological and embryological sequences.

We therefore state a plausible hypothesis when we assert that different species may be related to each other in the same way that the individuals of the same species are related, that is, by a tie of blood-relationship; and that different genera, families, orders, and so on are also so related. Morphological studies enable us to arrange numbers of species in such a way that series, in each of which there is an increasing specialisation of structure, are formed. Both palÆontology and embryology show, to some extent at least, that these stages of ever-increasing specialisation of structure occurred one after the other. Now, stated briefly and baldly as we have put it, this argument may not appear to the general reader to possess much force, but it is almost impossible to over-state the strength of the appeal which it makes to the student of biology. To such a one a belief in a process of transformism will appear to be inseparable from a reasoned description of the facts of the science.

But it would be no more than a belief, not even a hypothesis, if we did not attempt to verify it experimentally. It is merely logical relationships that we establish, and the chronological succession of forms of life, higher forms succeeding lower ones, does not itself do more than suggest an evolutionary process. All that we have said is compatible with a belief in a process of special creation. But if we cling to such a belief, if we suppose that the organisms inhabiting the earth, now and in the past, are the manifestations of a Creative Thought, we must still accept the notion of logical and chronological relationships between all these forms of life. If we permit ourselves to speculate on the working of the Creative Thought, we seem to recognise that the ideas of the different species must have generated each other, and that the genesis of living things must have occurred in some such order as is indicated by a scientific hypothesis of transformism. An evolutionary process must have occurred somewhere, but the kinships so established between organisms would be logical and not material ones.

Science must not, of course, describe the mode of origin of species in this way. So long as it investigates living things by the same methods which it uses in the investigation of inorganic things, it must hold that the concepts of physical science are also adequate for the description of organic nature. It must assume that matter and energy and natural law are given; and that, even in the conditions of our world, life must have originated from lifeless matter; must have shaped itself, and undergone the transformations that are suggested by the results of biology. It must assume, in spite of the formidable difficulties that the assumption encounters, that cosmic physical processes are reversible and cyclical; and that worlds and solar systems are born, evolve, and decay again. Every stage in such a cosmic process, as well as every stage in the evolution of living things, must have been inevitably determined by the stages preceding it. Such a mechanistic explanation must assume that a superhuman intellect, but still a finite intellect like our own, such a calculator as that imagined by Laplace or Du Bois-Reymond, would be able to deduce any state of the world, or universal system, from any other state, by means of an immense system of differential equations. It would be able, as Huxley says, to calculate the fauna of Great Britain from a knowledge of the properties of the primitive nebulosity with as much certainty as we can say what will be the fate of a man’s breath on a frosty day. Such a fine notion as that of an universal mathematics must ever remain as the ideal towards which science strives to approximate.

Or we may suppose that a plan or design has been superposed on nature, is immanent in matter and energy, and works itself out, so to speak. Such a teleological explanation of inorganic and organic evolution inevitably forces itself upon us if we reject the notion of radical mechanism. We think of an universal system of matter and energies as consisting of elements which, when assembled together, interact in a certain way, and with results which are definite and calculable. The assembling together of the elements of the system would be the result of the previous phases of the system. That is radical mechanism. But let us think of the elements of the system as being differently assembled—thus involving the idea of an agency, external to the system, which rearranges them—then the same energies inherent in this system, as in that previously imagined, will also work out by themselves. But the result will be different, and will depend on the manner in which the elements were originally arranged. That would be radical finalism.

Science must reject this notion as it rejects that of special creation, since it introduces indeterminism into the evolutionary process. It must regard the organism and its environment as a physico-chemical system studied from without. It must avoid all attempts to acquire an intuitive knowledge of the actions of the organism, for the latter, and the things which environ it, are only bodies moving in nature. In the systems studied by it time must be the independent variable, and there must be a strict functionality between the parts of the organism and the parts of the reacting environment, so that any change in the one must necessarily be dependent on a change in the other. Such a system and series of interactions is that which is described in a mechanistic hypothesis of transformism.

All this is indeed suggested to ordinary and aided methods of observation. The plant or animal acts upon, and is acted on by, the environment, though it is usually the modification of the organism to which we attend. A man’s face becomes reddened by wind and sun and rain; manual labour roughens his hands and develops callosities; in the summer he sweats and loses heat; in the winter the blood-vessels of his skin contract and heat is economised. In the winter months the fur of many animals becomes more luxuriant and may change in colour. Fishes which inhabit lightly coloured sand are lightly pigmented, but their skins become dark when they move on to darkly coloured sea-bottoms; prawns which are brown when they live on brown weed, become green when they are placed on green weed. Birds migrate into warmer countries, and vice versa, when the seasons change. Such are instances of the adaptations of the morphology and functioning of organisms consequent on changes of environment.

What is an adaptation? The term plays a great part in biological speculation, but it is often used in a loose and inaccurate manner, and not always in the same sense. It suggests that the organism is contained by the environment, and that its form becomes adapted to that of the latter, just as the metal which the ironfounder pours into the mould takes the form of the cavity in the sand. “We see once more how plastic is the organism in the grasp of its environment”—such a quotation from morphological literature is perhaps a typical one. Over and over again this passive change in the organism as the result of the action of something rigid which presses upon it is what is understood by an adaptation. No doubt the organism may be so affected, and often the change which it experiences is of the same order as the environmental change. In the winter many animals become sluggish and may hibernate; their heart-beats slow down; their respiratory movements become less frequent, and generally the rate of metabolism, that is the rapidity with which chemical reactions proceed in their tissues, becomes lessened. All these changes become reversed in sign when the temperature again rises. The time of year at which a fish spawns depends on the nature of the previous season. The rate of development of the egg of a cold-blooded animal varies with the temperature. The quantity of starch formed in a green leaf depends on certain variables—the intensity of light, the temperature, and the quantity of carbonic acid contained in the medium in which it is placed. In all these cases the rate at which certain metabolic processes go on in the body of an organism varies according to the conditions of the environment. In general they are cases of van’t Hoff’s law, that is, the rapidity at which a chemical reaction proceeds varies according to the temperature.

They are changes of functioning passively experienced by the organism as the result of environmental changes, and we must clearly distinguish between them and such changes as are the result of some activity or effort on the part of the organism. A flounder which lives in a river migrates out to sea when the first of the winter snows melt and flood the estuary with ice-cold water. Brown or striped prawns living on brown or striped weeds become green when they are placed on green weed, changing their pigmentation to match that of the alga. A kitten brought up in a cold-storage warehouse develops a sleeker and more luxuriant coat than does its sister reared in a well-warmed house. An animal which recovers from diphtheria forms an antitoxin which enables it to resist, for a time at least, repeated infection. A man who goes exploring in polar seas puts on warmer clothing than he wears in the tropics.

It is not necessary that an environmental change should occur in order that an adaptation should be evoked, for the organism may react actively and purposefully to a change in itself. The athlete acquires by running or rowing a more powerful heart; the blacksmith develops more muscular shoulders and arms; and the professional pianist more supple wrists and fingers. If one kidney is removed by operation, or if one lung becomes diseased, the organ on the other side of the body becomes hypertrophied. Aphasia, which is due to a lesion in the unilateral speech-centre, may pass away if the previously unused centre on the other side of the brain should become functionally active. In general, the continued use of an organ leads to its increase in size and efficiency, and conversely disuse leads to a decrease of size and even to atrophy.

The essence of an adaptation is that it is an active, purposeful change of behaviour, or functioning, or morphology, by which the organism responds to some change in its physical environment, or to some other change in its own behaviour, or functioning, or morphology. It is also a change which remains as a permanent character in the organisation of the animal exhibiting it. It does not matter even if the change of behaviour is one which is willed in response to some change of environment actually experienced, or whether it anticipates some change that is foreseen. A changed mode of behaviour adapted intelligently leaves, at the least, a memory which becomes a permanent part of the consciousness of the animal, and may influence its future actions; or if it is evoked by a process of education it must involve the establishment of a “motor habit.” The education of a singer sets up, in the cortex and lower centres of the brain, a nervous mechanism which controls and co-ordinates the muscles of the chest and larynx, and which did not exist prior to the process of education. Adaptations are therefore acquired changes of some kind or other by means of which the organism is able to exert a greater degree of mastery over its environment, including in the latter both the inert matter of inorganic nature and the other organisms with which the animal competes.

They are acquirements because of which the organism deviates from the morphological structure characteristic of the species to which it belongs. Do they affect the entire organisation of the animal exhibiting them, that is, may an acquired change of structure be so fundamental that it affects not only the body of the animal in which it occurs but also the progeny of this animal? Let us suppose that this is the case; let us suppose that quite a large proportion of all the individuals of a species inhabiting a restricted part of the earth’s surface acquire the same change of character simultaneously and that they transmit this deviation of structure to their progeny. Then we should have an adequate means whereby the specific type becomes modified—a means of transformism.

This is the hypothesis which is associated with the name of Lamarck, and its essential postulate is that characters which are acquired by an organism during its own lifetime are transmitted to its offspring. It seems reasonable to suppose that this transmission of acquired characters should occur—how reasonable we should note when we see that de Vries tacitly assumes that fluctuating variations due to the action of the environment may be inherited by the offspring of organisms which exhibit them. That transmutation of species might occur in this way was a popular and widespread belief in England and Germany throughout the greater part of the nineteenth century; and it was a belief entertained by Darwin himself, and confidently, and even dogmatically affirmed at one time by the majority of biologists in both countries.

How was it, then, that a very general change of opinion with regard to this question occurred both in England and Germany during the last two decades of the last century? Certainly many botanists and zoologists continued to adhere to the older hypothesis, and most physiologists still do not appear to make any clear distinction between morphological characters which are inherited and those which are acquired; but the majority of biologists did not hesitate to conclude that not only was the transmission of acquired characters an unproved conjecture, but that it was even theoretically inconceivable. At the beginning of the nineteenth century this belief had almost become a doctrine dogmatically asserted, and one cannot fail to notice a tone of irritation and impatience on the part of the spokesmen of zoology when the contrary opinions are expressed. “Nature,” says Sir E. Ray Lankester, “(and there’s an end of it) does not use acquired characters in the making and sustaining of species for the very simple reason that she cannot do so.”

There can be little doubt that the interrogation of nature with regard to this question was not a very thorough process. The dogmatic denial of the transmission of acquired characters was not the result of exhaustive experiment and observation, but was due rather to the very general acceptance in England and Germany of Darwin’s hypothesis of the transmutation of species by means of natural selection, and of Weismann’s hypothesis of the continuity of the germ-plasm.

The newer hypothesis of transmutation was one which seemed adequate to account for the diversity of forms of life, so that it was unnecessary to invoke the older one; though Darwin himself admitted that the individual acquirement of structural modifications might be a factor in the evolutionary process; and for more than twenty years after the publication of the “Origin of Species” Lamarck’s hypothesis was not strenuously denied by naturalists. Early in the ’eighties, however, Weismann published his book on the germ-plasm, and the brilliancy and constructive ability of the speculations contained in this remarkable work, as well as the analogies which they suggested between organic and inorganic phenomena, compelled the attention of biologists. The essential parts of Weismann’s hypothesis, as it was first presented to the world, are as follows: very early in the evolution of living from non-living matter many kinds of life-substance came into existence. These were chemical compounds of great complexity, able to accumulate and expend energy, and capable of indefinite growth and reproduction. They were able to exist in an environment which was hostile to them and which tended always to their dissolution, and which was able to modify their nature and their manner of reacting, though it could not destroy them. These elementary life-substances were very different from those which we know in the world of to-day. They were naked protoplasmic aggregates, undifferentiated into cellular or nuclear plasmata, much less into somatic and germinal tissues. All of their parts were similar, or rather their substance was homogeneous. But even with the evolution of the unicellular organism a profound change was initiated, for henceforth one part of the living entity, the nucleus, became charged with the function of reproduction, although it still continued to exercise general control over the functions of the extra-nuclear part of the cell. When the multi-cellular plant and animal became evolved, the heterogeneity of the parts of the organism became greater still. All the cells of the metazoan animal do indeed contain nuclei, but these structures are only the functional centres of the cells: some of the latter are sensory, others motor, others assimilatory, others excretory, and so on. Only in the nuclei which form the essential parts of the reproductive organs does the reproductive function persist in all its entire potentiality: there only does the protoplasm retain all the properties which were possessed by the primitive life-substance before it became heterogeneous, that is, before nucleus and cytoplasm evolved. When part of the primitive life-substance became secluded in a nuclear envelope, it became, to that extent, shielded from the action of the physical environment, and when the organism became composed of multicellular tissues this seclusion became more complete. Clothed in the garments of the flesh, it was henceforth protected from the shocks of the environment, and it became the immutable germ-plasm. But for a very long time before this evolution of tissues the naked life-substance had been exposed to the action of external physical agencies, and it had been modified by these into very numerous forms of protoplasmic matter. When multicellular plants and animals had been evolved there were, therefore, not one, but many kinds of life-substance in existence, and these have persisted until to-day as the unchanging germ-plasmata of the existing organisms.

The Weismannian hypothesis of to-day, supported and amplified, as it is, by subsidiary hypotheses, does not make the same appeal to the student as did the pristine and altogether attractive speculation of thirty years ago. The analogy which it then presented with the matured chemical theory of matter must have been almost irresistible. Just as the indefinitely numerous compounds of chemistry are only the permutations and combinations of some of eighty-odd different kinds of matter, so all the forms of life are combinations and permutations of some of the many different kinds of life-substance which came into existence before the evolution of the multicellular organism. And just as the chemical elements were regarded (in 1883) as immutable things, preserving their individuality even when they were associated together as compounds, so Weismann and his followers looked upon the different kinds of life-substance contained in the chromatic matter of the nucleus as immutable and immortal living entities. Associated together in indefinitely numerous ways by sexual conjugation, they may build up indefinitely variable living structures, but they remain individualised and lying side by side in the germ-plasmata of organisms, just as the atoms were supposed to lie side by side in the chemical molecule of the inorganic compound.31

If these speculations were true, a change of morphology or functioning, acquired by the body, or somatoplasm, could not possibly be transmitted to the progeny of the organism, for by hypothesis the germ-plasm cannot be affected by external changes, and it is only the germ-plasm contained in the spermatozoon of the male parent, or in the ovum of the female, that shapes and builds the body of the offspring. As if this were not enough, Weismann and his followers argued that the transmissibility of a somatic change to the germ was inconceivable. Why? Because the germ-cells are apparently simple: they are only semi-fluid protoplasmic cell bodies and nuclei, not differing appreciably from the cell bodies and nuclei of the somatoplasm (by hypothesis, it should be noted, the difference is profound). There are no structural connections—no nerves, for instance—which join together the cells of the bodily tissues with the parts of the germ and transmit changes in the former to the latter. How, then, could a somatic change affect the germ so that when the latter developed into an organism this particular change became reproduced? Now this may have seemed a conclusive argument in 1883, but is it so conclusive to-day? We know that the cells and tissues are not isolated particles, but that all are connected together by protoplasmic filaments. We know that specialised nervous tissues are not necessary for the transmission of an impulse from a sensory to a motor surface, but that such an impulse may be transmitted by undifferentiated protoplasm. We know that nerve-cells and nerve-fibres are not structurally continuous with each other but that the impulse leaps across gaps, so to speak. We know that events that occur in one part of the body of the mammal may affect other parts by means of the liberation of a chemical substance, or hormone, into the blood stream. It would be strange indeed if a logical hypothesis capable of accounting for the transmission of a particular change from the soma to the germ could not be elaborated.

But acquired characters were not really transmitted after all. So those who clung to Weismannism argued—an unnecessary task surely if this transmissibility were inconceivable. We cannot discuss the evidence here, and it is unnecessary that we should do so, since it is all considered in the popular books on heredity. There is an apparent consensus of opinion in these books which should not influence the reader unfamiliar with zoological literature, nor obscure the fact that many zoologists and botanists accept the opposite conclusion. The discussion is all very tiresome, but we may glean some results of positive value from it. It is unquestionable that very few conclusive and adequate investigations have been made: one cannot help noticing that the literature contains an amount of controversy out of all proportion to the amount of sound experimental and observational work actually carried out. Most of the experiments deal with the consideration of traumatic lesions or mutilations, and it seems to be proved that such defects are not transmitted, or at least are very rarely transmitted. The tails of kittens have been cut off; the ears of terrier-dogs have been lopped; and the feet and waists of Chinese and European ladies have been compressed, and all throughout very numerous generations, yet these defects are not transmitted from parent to offspring. This kind of evidence forms the bulk of that which orthodox zoological opinion has adduced in favour of the belief in the non-inheritability of acquired characters, but does it all really matter? What might be transmitted is a useful, purposeful modification of morphology, or functioning, or behaviour, induced by the environment throughout a number of generations—an adaptation rather than a harmful lesion. There is little conclusive evidence that such adaptations are inherited, though anyone who carefully studies the evidence in existence will not be likely to say that they are certainly not transmitted. Does, for instance, the blacksmith transmit his muscular shoulders and arms to his sons, or the pianiste her supple wrists and fingers to her daughters? There are no observations and experiments in the literature worthy of the importance attaching to the question at issue.

It should be noted also that the germ-plasm is certainly not the immutable substance that the hypothesis originally postulated. Changes in the outer physical environment may certainly affect it; thus the larvÆ bred from animals which live in abnormal physical conditions (temperature, moisture, etc.) may differ morphologically from the larvÆ bred from animals belonging to the same species but living in a normal environment. The latter must therefore react on the germ-plasm, but the environment formed by the bodily tissues which surround the germ-cells may also so react: thus the germ-cells may be affected by such bodily changes as differences in the supply of nutritive matter, for instance. The offspring may deviate from the parental structure as the result of structural modifications acquired by the parent during its own lifetime, and, even if the filial deviation were not of the same nature as the parental modification, its inheritance would be an adequate cause of some degree of transmutation.

It is, however, certainly difficult to prove that organisms transmit to their progeny the same kinds of deviation from the specific structure that they themselves acquire as the result of the action of the environment. Even if they did transmit such acquired deviations, it does not seem clear that this kind of inheritance alone would be a sufficient cause of the diversity of forms of life that we do actually observe in nature. Change of morphology would indeed occur, but we should expect to find insensible gradations of form and not individualised species. Let us suppose that Lamarckian inheritance acts for a considerable time on two or three originally distinct species inhabiting an isolated tract of land, and let us suppose that we investigate the variations occurring among all the organisms which are accessible to our observation with respect to some one variable character.

The diagram A represents what would seem to be the result of this process of transmutation. The numbers along the horizontal line are proportional to their distance from o, the origin, and represent the magnitude of the variation considered; and the height of the vertical lines represents the number of organisms exhibiting each degree of variation. We should expect to find that all the variations were equally frequent in their occurrence, but this is not what a study of variability in such a case as we have supposed—that of the animals inhabiting an isolated part of land—does actually indicate. What we should find would be the conditions represented by the diagram B. There would be two or more modes, that is, values of the variable character which are represented by a greater number of individuals than any other value of the variation. The environmental conditions favour the individuals displaying this variation to a greater extent than they favour the rest.

That is to say, the environment selects some kinds of variations among the many that are exhibited, and this is, of course, the essential feature of the hypothesis of the transmutation of species by means of natural selection of variable characters. Organisms enter the world differently endowed with the power of acting on the medium in which they live, or on the environment consisting of their fellow-organisms. Those that are most favourably endowed live longest and have a more numerous progeny than those that are less favourably endowed, and they transmit this favourable endowment to their offspring. Among the progeny of the progeny there may be some in which the favourable variation is still more favourable than it was when it first appeared. Thus the variations which are selected increase in amount. Elimination of the weakest occurs. The idea is eminently clear and simple, and possesses a great degree of generality: it is self-evident, says Driesch, meaning that it cannot be refuted, for it was certainly not clearly obvious to the naturalists before Darwin and Wallace. But, unless we choose to be dogmatic, we can hardly claim that it is an all-sufficient cause for the evolutionary process, and it is useless to attempt to minimise the difficulties of the hypothesis. It is not easy to make it account for the origin of instincts or tropisms, or for restitutions and regenerations of lost parts, or for the appearance of the first non-functional rudiments of organs which later become functional and useful. It is, indeed, possible to devise plausible hypotheses accounting for all these things in terms of natural selection, but each such subsidiary hypothesis loads the original one and weakens it to that extent.

Natural selection does not, of course, induce or evoke variations; these are given to its activity, and they are the material on which it operates. What, then, is the nature of the deviations from the specific types of morphology that are selected or eliminated? Not those induced by the environment, and transmitted in their nature and direction to the progeny of the organisms first displaying them. It is not unproved that such variations do occur, and it is even probable that they do occur. But we may conclude that the frequency of their occurrence is not great enough to afford sufficient material for natural selection. It is also clear that the ordinarily occurring variations that we observe in any large group of organisms collected at random are not alone the material for selection; for we have seen that experimental breeding from such variations does not lead to the establishment of a stable race or “variety.” Nevertheless some effect is produced, and this may be accounted for by supposing that the observed variations are really of two kinds—fluctuating variations, which are not inherited, and mutations, which are inherited. The small observed effect is due to the selection of the mutations alone: it is a real effect of selection, an undoubted transmutation of the specific form, but experimental and statistical investigations seem to show that selection from the variations that we usually observe is too slow a process to account for the existing forms of life.

Natural selection acts, therefore, on mutations. Now it seems that we are forced to recognise the existence of two categories of mutations, (1) those stable modifications of an “unit-character” which we term “Mendelian characters,” and (2) those groups of stable modifications to which de Vries applied the term mutations. It seems at first difficult to see how permanent modifications of the specific form can be brought about by the transmission of Mendelian characters, for these characters are always transmitted in pairs. Let us take a concrete case—that of a man who has six fingers on his right hand, and let us suppose that this was a real, spontaneously appearing character or mutation which had not previously occurred in the ancestry of the man. Two contrasting characters would then be transmitted, (1) the normal five-fingered hand, and (2) the six-fingered hand. Both of these characters are supposed to be present at the same time in the organisation of the men and women of the family originating in this individual, but one of them is always latent or recessive. There would, however, be individuals in which only one of the characters would be present—either the normal or abnormal number of digits, but intermarriage with individuals belonging to the other pure strain would immediately lead again to the transmission of the contrasting characters, or allelomorphs, although marriage with an individual belonging to the same pure strain would carry on the normal or abnormal unmixed character into another generation. But if the possession of six fingers conveyed an undoubted advantage, and if natural selection did really act in civilised man as regards the transmission of morphological characters, then a stable variety (Homo sapiens hexadactylus, let us say) might be produced by its agency. The mutations which we consider in the investigation of the inheritance of alternating characters are therefore just as much the material for natural selections as the mutations which occur among the ordinary variations displayed by organisms in general: but since only one or two characters appear to be subject to this mode of transmission, the process would be so slow as to be inadmissible as an exclusive cause of evolution.

If we assume that de Vries’ mutations are the material on which selection works, this difficulty is immediately removed, for we now have to deal with groups of stable deviations: not one or two, but all the characters of the organism appear to share in the mutability. But another difficulty now arises. A species of plant or animal may have got along very well with its ordinary structural endowment, and then a number of individuals begin to mutate. Some of the deviations from the specific type may be of real advantage, but others may not: we can, indeed, imagine an in-co-ordination between the mutating parts or organs which would be fatal to the animal; on the other hand, there might be complete co-ordination, with the result that great advantage might be conferred upon the individual. It is easy to see how co-ordination of mutating parts is absolutely essential. An animal which preserves its existence by successful avoidance of its enemies would not be greatly benefited by a more transparent crystalline lens if the vitreous humour of its eye were slightly opaque; and even if all the parts of the eye were perfectly co-ordinated, increased acuity of vision would not greatly help it if its limbs were not able to respond all the more quickly to the more acute sensation. Un-co-ordinated mutations would therefore tend to become eliminated, while co-ordinated ones would become selected and would become the characters of new species.

We must now ask why some groups of variations are co-ordinated while others are not, and it is here that we encounter the most formidable of the difficulties of any hypothesis of transformism which depends on the concept of natural selection. If we assume that the environment induces the appearance of variations, it seems to follow that these variations are likely to be co-ordinated, but we then invoke the principle of the acquirement of characters and their transmission by heredity. If, on the other hand, we assume that variations appear spontaneously, and quite irresponsibly, so to speak, in the germ-plasm of the organism, the selection, or elimination, by the environment will not occur until the co-ordinated or un-co-ordinated variations appear. It is far more likely that a large number of simultaneously appearing variations will be un-co-ordinated than that they will be co-ordinated. Merely as a matter of probability the progressive modification of a species will take place slowly—too slowly to account for what we see.

Two examples will make it easier to appreciate this difficulty. Evolution has undoubtedly proceeded in definite directions. There are two dominant groups of fishes, the Teleosts and the Elasmobranchs, and both must have originated from a common stock. All the characters in each kind of fish must have been useful (since they were selected), and all must have been modifications of the characters of the common stock. The latter became modified along two main lines, or directions, which are indicated by the characters of the existing Teleosts and Elasmobranchs. The whole skeleton, the gills, the circulatory system, and the brain differ in certain respects in these groups. Therefore a modification of the brain in the primitive Elasmobranchs was associated with a modification of the cranium, and therefore with the jaw-apparatus, and so with the branchial skeleton and the gills, and therefore also with the heart, and so on. Suppose that the evolutionary process included ten useful and co-ordinated variations—not an unlikely hypothesis—and suppose that each of these ten useful variations was associated with nineteen useless ones. The chance that any one of them did occur was therefore one in twenty; and if they all occurred independently, that is, if the occurrence of any one of them was compatible with the occurrence of any other one, or of all the others, then the chance that all the ten variations occurred simultaneously was 20^-10 that is, one in the number 20 followed by 10 cyphers, a rather great improbability.

Most biological students are familiar with the similarity of the so-called eye of the mollusc Pecten and that of the vertebrate. The resemblance is one of general structure: in each of these organs there is a camera obscura, a transparent cornea, and behind that a crystalline lens. On the posterior wall of the camera there is a receptor organ, or retina, and this is composed of several layers of nervous elements. The actual nerve-endings are on the surface of the retina, which is turned away from the light, that is, the optic nerve runs towards the anterior surface of the retina, and then its fibres turn backwards. This “inversion of the retinal layers” occurs in all vertebrate animals, but it is exceptional in the invertebrates. The above general description applies equally well to the eye of the vertebrate and to that of Pecten.

Let us admit that these mantle organs in Pecten are eyes, for there is no conclusive experimental evidence that they really are visual organs, and plausible reasoning suggests that they may subserve other functions. Let us assume that the minute structure of the Pecten eye is similar to that of the vertebrate, and that its development is also similar: as a matter of fact both histology and embryology are different. Then we have to explain, on the principles of natural selection, the parallel evolution of similar structures along independent lines of descent; for mollusc and vertebrate have certainly been evolved from some very remote common ancestor in which the eye could not have been more than a simple pigment spot with a special nerve termination behind it. In each case the organ was formed by a very great number of serially occurring variations, yet these two sets of variations must have been the same at each stage in two independently occurring processes. On any reasonable assumption as to the number of co-ordinated variations required, and their chances of occurrence, the mathematical improbability that these two series of variations did occur is so great as to amount to impossibility so far as our theory of transformism is concerned. Natural selection could not, therefore, have produced these two organs.

This argument of Bergson’s fails, of course, in the particular instance chosen by him, but this is because the case is an unfortunate one. Probably a morphologist could find a very much better case of convergent evolution—the parallelism between the teeth of some Marsupials and some Rodents, for instance. If detailed histological and embryological investigation should show a similarity of structure and development, in such compared organs Bergson’s argument would retain all its force. We should then have to assume that there was a directing agency, or tendency in the organism, co-ordinating, or perhaps actually producing, variations.

Mechanistic biology can suggest no means whereby simultaneously occurring variations are co-ordinated: let us therefore think of these variations as occurring independently of each other, and let us ignore the difficulty of the infrequency of occurrence of suitably co-ordinated variations. Variations are exhibited by the evolving organism, and the selection of co-ordinated series is the work of the environment. But the environment is merely a passive agency, and it has to confer direction on the innumerable variations presented to it by the organism, rejecting most but selecting some. Let us think of the environment, says a critic of Bergson, as a blank wall against which numerous jets of sand are being projected. The jets scatter as they approach the wall: each of them represents the variations displayed by some organ or organ-system of an animal. Let us think of a pattern drawn on the wall in some kind of adhesive substance: where the wall is blank the sand would strike, but would fall off again, but it would adhere to the parts covered by the adhesive paint. The sand grains strike the wall from all sides, that is, their directions are un-co-ordinated. The wall is passive, yet a pattern is imprinted upon it. From passivity and un-co-ordination come symmetry and order.

This argument withstands superficial examination, but to accept it is truly to be “fooled by a metaphor.” For what is the pattern on the wall? It is the environment, says the critic. But what is the environment? Inevitably we think of it as something that makes or moulds the organism, a way of regarding it that drags after it all the confusion of thought implied in the above analogy. Clearly the environment is made by the organism. Its form, that is, space, is only the mode of motion possible to the organism; it is clear that whether the space perceived by an organism is one-, two-, or three-dimensional, space depends upon its mode of motion. Its universe is whatever it can act upon, actually or in contemplation. Atoms and molecules, planets and suns are its environment because it can in some measure act upon these bodies, or at least they can be made useful to it. Chloroform or saccharine, or methyl-blue and all the dye-stuffs prepared from coal-tar by the chemists, are part of our environment because we have made them. They existed only in potentiality prior to the development of organic chemistry. They were possible, but man had to assemble their elements before they became actual. In making them, he conferred direction on inorganic reactions.

Surely the organism itself selects the variations of structure and functioning that are exhibited by itself. If we hesitate to say that these modifications are creations, let us say that they are permutations of elements of structure, and that they were potential in the organisation of the creature exhibiting them. They occur in the latter if we must not say that they are produced. If they are detrimental, the organism is the less able to live and reproduce, and if it does reproduce, its progeny are subject to the same disability. If, as is usual, they simply do not matter, they may or may not affect the direction of evolution. If they are of advantage, that is, if they confer increased mastery over the environment, over the inert things with which the organism comes into contact, the latter enlarges its universe or environment, lives longer, and transmits to its progeny its increased powers of action. Indefinite increase of power over inert matter is potential in living things, and variation converts this potentiality into actuality.

This discussion is all very formal, but two conclusions emerge from it: (1) the insufficiency of the mechanistic hypotheses of transformism to account for all the diversity of life that has appeared on the earth during the limited period of time which physics allows for the evolutionary process. There does not appear to be any possibility of meeting this objection if we continue to adhere to the hypothesis of transformism already discussed: it faces us at every turn in our discussion. How great a part is played, for instance, by “pure chance” in the elimination of individual organisms during the struggle for existence! Let us think of a shoal of sprats on which sea-birds are feeding: it is chance which determines whether the birds prey on one part of the shoal rather than another. Or let us think of the millions of young fishes that are left stranded on the sea-shore by the receding tide: it is chance that determines whether an individual fish will be left stranded in a shallow sandpool which dries up under the sun’s rays, rather than in a deeper one that retains its water until the tide next flows over it. It is no use to urge that there is no such thing as “pure chance,” and that what we so speak of is only the summation of a multitude of small independent causes. Let us grant this, and it still follows that the alternative of life or death to multitudes of organisms depends not upon their adaptability but upon minute un-co-ordinated causes which have nothing to do with their morphology or behaviour. These are instances among many others which will occur to the field naturalist: they shorten still further the time available for natural selection in the shaping of species, for they reduce the material on which this factor operates.

The other result of our discussion is to indicate that the problem of transformism of species is in reality the problem of organic variability. Let us assume that all the hypotheses of evolution are true: that the environment may induce changes of morphology and functioning in animals and plants, and that these changes themselves—the actual acquirements themselves, that is—are transmissible by heredity. Let us assume that the germ-cells may be affected by the environment, either the outer physical environment, or the inner somatic environment, and that mutations may thus arise. Let us assume that mutations may be selected in some way, so that specific discontinuities of structure—“individualised” categories of organisms, or species—may thus come into existence. Even then transformism is still as great a problem as ever, for the question of the mode of origin of these variations or modifications still presses for solution.

The simplest possible cases that we can think of present the most formidable difficulties. The muscles of the shoulders and arms of the blacksmith become bigger and stronger as the result of his activity. Why? We say that the increased katabolism of the tissues causes a greater output of carbonic acid and other excretory substances, and that these stimulate certain cerebral centres, which in turn accelerate the rate of action of the heart and respiratory organs. An increased flow of nutritive matter and oxygen then traverses the blood-vessels in the muscles of the shoulders and arms, and the latter grow. Probably processes of this kind do occur, but to say that they do is not to give any real explanation of the hypertrophy of the musculature of the man’s body, for what essentially occurs is the division of the nuclei and the formation of new muscle fibres. How precisely does an increased supply of nutritive matter cause these nuclei to divide and grow? This is a relatively simple example of the adaptability of a single tissue-system to a change in the general bodily activity, that is to say it is a variation of structure induced by an environmental change.

In most cases, however, the variations of structure that form the starting-points of transmutation processes cannot clearly be related to environmental changes. Some fishes produce very great numbers of ova in single broods—a female ling, for instance, is said to spawn annually some eighteen millions of eggs. If we examine these ova we shall find that there is considerable variation in the diameter and in other measureable characters. We may attempt to correlate these deviations from the mean characters with environmental differences. All the eggs “mature,” that is, they absorb water and swell, while various parts, such as the yolk, undergo chemical changes, during the month or so before the fish spawns. This process of maturation takes place in the closed ovarian sac; and the eggs lie practically free in this sac, and are bathed in a fluid which exudes from the blood-vessels in its walls. It may indeed be the case that there are variations in the composition of this fluid in the different parts of the sac; but these variations cannot be great; the fluid is not really a nutritive one; and the process of maturation is not hurried. We can hardly believe that the differences in morphology are due to these minute environmental differences. We may indeed say that we do not really study the germ cells when we measure the diameter of the egg or investigate any other measurable character, for the real germ-plasm is the chromatic matter of the nucleus. But this obviously begs the whole question: all the parts of the egg that are accessible to observation do vary, and ought we to conclude that the parts which are not accessible do not vary? They must vary: the germ-plasm of each egg must be different from that of all the others, for the organisms which develop from these germs show inheritable differences. Further, can we contend that such minute environmental differences as we have indicated affect the germ-plasm? Is it so susceptible to external changes? A high degree of stability of the germ-plasm is postulated in the mechanistic hypothesis that we have considered, and indeed everything indicates that the specific organisation is very stable. Can it then be upset by such minute differences in the somatic environment?

But the germ-plasm is not really simple, says Weismann; it is a complex mixture of ancestral germ-plasms. The individual fish that we were considering arose from an aggregate of determinants, and half of these determinants were received from the male parent and half from the female one. But each of these parents also arose from a similar aggregate of determinants, which again were received from both parents, and so on throughout the ancestry of the fish. It is true that the germ-plasms contributed by the ancestors were not quite different, but they differed to some extent. Then there must have been as many permutations of determinants in the ovum from which the fish developed as there were permutations of characters in the eighteen millions of ova produced by it. Does not the hypothesis collapse by its own weight?

It could only have been such difficulties as are here suggested that led Weismann to formulate his hypothesis of germinal selection. All those eighteen millions of eggs arose from the division of relatively few germ cells. Each of these original cells contained the specific assemblage of determinants, and the elements of the latter are of course the biophors. The biophors, it will be remembered, are either very complex chemical molecules, or aggregates of such. When the germ cells of the germinal epithelium divide to form those cells which are going to become the ova, the biophors must divide and grow to their former size, and again divide—it is really a chemical hypothesis that we are stating, though we have to employ language which seems to do violence to all sound chemical notions! Now while the biophors were dividing and growing they were “competing” for the food matter which was in the liquid bathing them, and some got less, while others got more than the average quantity. In this way their characters became different, so that the eggs, on the attainment of maturity, became different from each other. Now, apart altogether from the impossibility of applying any test as to the objective reality of this hypothesis, it must be rejected, for it confers on bodies which belong to the order of molecules properties which are really those of aggregates of molecules. The typical properties of a gas, for instance, are not the properties of the molecules of which the gas is composed, but are statistical properties exhibited by aggregates of molecules. On the hypothesis of germinal selection the properties of the animals which develop from the biophors are extended to the biophors themselves. It was surely a desperate plight which evoked this notion! It is, as William James said about Mr Bradley’s intellectualism, mechanism in extremis!

We seem forced to the conclusion—and this is the result to which all this discussion is intended to approximate—that variations, heritable variations at least, arise spontaneously. That is, there are organic differences which have no causes, a conclusion against which all our habits of reasoning rebel. Yet it may be possible to argue that the problem of the causes of variations is really a pseudo-problem after all, and that there is no logical reason why we should be compelled to postulate such causes. When we think of organic variability, do we not think, surreptitiously it may be, of something that varies, that is, something that ought to be immutable but which is compelled to deviate? But what is given to our observation is simply the variations among organisms.

Let us think of the crude minting machines of Tudor times which produced coins which were not very similar in weight and design. From that time onward minting machines have continually been improved, each successive engine turning out coins more and more alike in every respect, so that we now possess machines which stamp out sovereigns as nearly as possible identical with each other. Yet they are not quite alike, and this is because the action of the engine, in all its operations, is not invariably the same. In imagination, however, we make a minting machine which does work perfectly, and turns out coins absolutely alike, but this ideal engine is only the conceptual limit to a series of machines each of which is more nearly perfect than was the last one. It is unlikely that matter possesses the rigidity and homogeneity which would enable us to obtain this perfect identity of result; nevertheless this identity has a very obvious utility, and we strive after it, so that the result of our activity is the conception of a perfect mechanism, and of products which are identical. We assume that the reasons why our early and cruder machines were imperfect are also the reasons why our later and more perfect ones do not produce the results that we desire.

We are artisans first of all, and then philosophers, and so we extend this ingrained mechanism of the intellect into our speculations. To the biologist the organism is a mechanism which, in reproduction, ought to turn out perfect replicas of itself. It does not do so. Now, if biology shows us anything, it shows us that living matter is essentially “labile,” that is, something fluent, while lifeless matter is essentially rigid, or nearly so. Yet, ignoring this difference, we expect from the organism that identity of result and operation that we conceptualise, but do not actually obtain from the artificial machine. We regard the organism, not only as a mechanism like the minting machine, but as the conceptual limit to a series of mechanisms. The reproductive apparatus of our fish does not turn out ova which are identical, but which differ from each other. Some of this variation, we say, is due to the action of the environment; and some of it is due to the condition that each ovum receives a slightly different legacy of characters from the multitude of ancestors. The rest we conceive as due to the imperfect working of the reproductive machinery.

It is useful that science should so regard the working of the organism, for in the search for the causes of variation our analysis of the phenomena of life becomes more penetrating. But does any result of investigation or reasoning justify us in assuming, as a matter of pure speculation, that deviations from the specific type of structure are physically determined in all their extent? Have we not just as much justification for the belief that these deviations are truly spontaneous, and that they arise de novo? So we approach, from the point of view of experimental biology, Bergson’s idea of Creative Evolution.