Laws of Modification.

 I have hitherto spoken of the variations—which are so common and varied in the cultivated state of organisms, and to a somewhat lesser degree frequent in the wild—as if they were caused by chance. This is, however, a wholly incorrect way of expression, which is only suitable to document our complete ignorance of the cause of any particular deviation. some writers[p. 158]see it as the task of the reproductive system to produce individual differences or very slight differences in structure rather than to make the child equal to the elders. But the much greater variability, as well as the much more frequent monstrosities, of the organisms subjected to culture lead me to believe that deviations of structure are in some way related to the nature of the external conditions of life to which the elders and their ancestors have been exposed for several generations , hang out. I made the remark in the first chapter—but a long record of facts, which cannot be given here, would be necessary to prove the truth of this remark—that the reproductive system is extremely sensitive to changes in the external conditions of life is sensitive; hence I ascribe chiefly to its functional disturbances in the adults the changeable or plastic nature of their offspring. The male and female elements of the organization appear to have been affected by this prior to their merging to form new offspring of the species. What the play plants (p. 15 ) is concerned, the bud alone is affected, which in its first stage of development is not very essentially different from an oak tree. On the other hand, we are in utter ignorance as to how it is that, through disturbance of the reproductive system, this or that part is affected more or less than another. Notwithstanding this, we succeed in catching a faint ray of light here and there, and we are convinced that there must be some cause, however slight, for every variation.

How much immediate influence differences in climate, food, &c., can exert on any creature, is extremely doubtful. I am convinced that the effect is extremely slight in animals, and perhaps somewhat greater in plants. It may at least be said with certainty that these influences did not produce the many excellent and complex adaptations of the organization of one being to another[p. 159]can have, which we see everywhere in nature. Some small effects may be ascribed to climate, food, etc., such as B. E DUARD F ORBES states with certainty that a species of conchylia may assume brighter colors in warmer climates and shallow waters than in its colder ranges. G OULD believes that birds of the same species are more brilliantly colored in an atmosphere that is always cheerful than on an island or on the coast [18] . W OLLASTON also believes that staying near the sea affects the colors of insects. M OQUIN -T ANDONgives a list of plants which develop more and less fleshy leaves on the sea-coast, though not fleshy inland either. And so many other similar examples could be cited.

The fact that varieties of one species, when they pass into the zone of distribution of another species, assume in a small degree something of the character of that species, agrees with our view that species of all kinds are only more pronounced permanent varieties. Thus the species of conchylia of shallow tropical sea regions are usually brighter in color than those that live in deep and cold waters. Such are the birds-species of the inland countries according to G OULD more vividly colored than those of the islands. Thus the insect species which are confined to the shores are often bronzed and dull in appearance, as any collector knows. Species of plants which grow only along the sea are very often furnished with fleshy leaves. Anyone who believes in the special creation of each individual species will therefore have to say that e.g. For example, these conchylia have been created with brighter colors for a warmer sea, while those others have only acquired the brighter coloring by modification when they migrated to the shallower and warmer waters.

[p. 160]

If a variation is of the slightest use to a creature, we cannot say how much is due to the accumulating action of natural selection, and how much to the influence of external conditions of life. Thus it is well known to fur-dealers that animals of a species have thicker and better pelts the colder the climate they have lived in. But who is to say how much of this difference is due to the warmest clad individuals having been favored and preserved through many generations by Natural Selection, and how much to the direct influence of the severe climate? For it seems well that climate has some direct effect upon the condition of the hair of our domestic animals.

One can cite examples that the same variety has arisen under very different living conditions, while on the other hand different varieties of a species appear under the same conditions [19]. These facts show how indirectly the conditions of life act. Thus every naturalist is aware of innumerable examples of genuinely preserved species without any varieties, although they live in the most opposite climates. Such considerations lead me to lay very little weight upon the direct influence of living conditions. Indirectly, as has been said, they seem to take an important part in the disturbance of the reproductive system, and thereby bring about variability, and Natural Selection then saves up all useful though slight variation until such becomes fully developed and perceptible to us. In a broader sense it may be said that the conditions of life include not only variability, but also natural selection; for it depends on the nature of the conditions of life whether this or that variety can arise. But we see from the process of human breeding that these two elements of change[p. 161]are materially different from each other; the living conditions in the state of domesticity cause variability, and the will of man, acting consciously or unconsciously, accumulates variation in certain directions in certain directions.

Effects of use and non-use. ) The facts adduced in the first chapter leave little doubt about our domestic animals, that use strengthens and expands certain parts, and disuse weakens them, and that such variations are inherited. In the wild there is no standard for comparing the effects of long-continued use or disuse, because we do not know the older forms; but some animals bear formations which can be explained as the result of disuse. Professor R. O WEN has remarked that it is a great anomaly in nature that a bird cannot fly, and yet many are so. The South AmericanBighead duck can only flap above the surface of the water, and has wings of almost the same condition as the Aylesbury domestic duck breed. As the large ground-birds seldom fly for other purposes than to avoid danger, I believe that the almost wingless nature of certain species of birds which some islands of the Great Oceannow inhabit or once inhabited, where they have no fear of being pursued by beasts of prey, from the disuse of their wings. It is true that the ostrich inhabits continents and is threatened by dangers which it cannot escape by flight; but he can defend himself against his enemies by kicking his feet as well as some of the smaller quadrupeds. One can imagine that the forefather of the ostrich had a way of life similar to that of the bustard, and that as a result of natural breeding over a long series of generations, it grew larger and heavier, used its legs more and its wings less, until he finally became completely unable to fly.

KIRBY has remarked (and I have observed the same fact) that the front tarsi of many male dung beetles often[p. 162] have broken off; he examined seventeen specimens from his collection, and found no trace of them in any of them. Onites Apelles has so commonly lost its tarsi that this insect has been described as entirely wanting. In some other genera they are present only in a stunted state. They are wholly absent from the Ateuchus, or sacred beetle, of the Egyptians. It is true that there is insufficient proof that accidental mutilations are hereditary; but the case observed by B ROWN -S EQUARD of the inheritance of that on a guinea-Pig epilepsy in their offspring caused by damage to the spinal cord would have to make us cautious if we wanted to deny it. Hence it seems more advisable to attribute the total want of the front tarsi of the Ateuchus, and their atrophied state, in some other genera to the long-continued effect of their disuse by their progenitors; for as the tarsi of many dung-beetles are mostly wanting, they must be lost early in life, and therefore cannot be of any essential service, nor be of much use, to these insects.

In some cases we may readily ascribe to disuse certain modifications of organization, which, however, result wholly or principally from natural selection. W OLLASTON has discovered the remarkable fact, that of the 550 species of beetle inhabiting Madeira , 200 have wings so imperfect that they cannot fly, and that of the 29 genera exclusive to the island, no fewer than 23 contain all such species . Some facts, such as, among other things, that in many parts of the world flying beetles are constantly being blown into the sea and perish, that the beetles of Madeira are after W OLLASTONSObservation usually lie hidden until the wind calms and the sun shines, that the number of wingless beetles on the exposed bare cliffs is proportionately greater than in Madeira itself, and especially the extraordinary fact on which W OLLASTON so persistently bases, that certain large and very numerous groups of beetles, which are distinguished by their way of life[p. 163]much required to fly, are entirely absent on Madeira —these various reasons lead me to believe that the wingless nature of so many beetles of this island is chiefly due to natural selection, though probably in conjunction with disuse. For during a thousand successive generations, every single beetle that flies least, either because its wings are least developed, or because it is the most indolent, will have the best chance of surviving all the others, because it is not blown into the sea; and, on the other hand, those beetles which prefer to fly will most often be driven out to sea and destroyed.

Those insects on Madeira , on the other hand, which do not stay on the ground and, like the beetles and butterflies that live on flowers, usually have to make use of their wings to earn their living, according to W OLLASTON ' SAssumption by no means stunted, but rather more developed wings. This is quite consistent with the activity of natural selection. For when a new insect first comes to the island, natural selection's tendency to increase or decrease the wings will depend on whether a larger number of individuals are obtained by successfully fighting the winds, or by abstaining from them more or less frequently Try saves yourself. It is the same case as that of the sailors of a ship stranded near shore; for those who swim well, the better they can swim, the better it is to try their luck by swimming on, while for the bad swimmers it would be best if they could not swim at all, and therefore sought rescue on the wreck .

The eyes of moles and some burrowing rodents are stunted in size, and in some cases are entirely covered with skin and fur. This condition of the eyes probably results from continued disuse, the effect of which may be aided by Natural Selection. A South American rodent, Ctenomys, has an even more subterranean[p. 164]way of life than the mole, and a Spaniard, who often caught such things, assured me that such a one was often quite blind; one that I got alive, it was certainly, as the dissection showed, as a result of an inflammation of the nictitating membrane. Since frequent inflammation of the eyes must be detrimental to every animal, and since the eyes are certainly not indispensable for subterranean animals, a reduction in their size, the adhesion of the eyelid to them and the covering of them with fur will be of use to them; and if this be the case, Natural Selection will constantly aid the effect of disuse.

It is well known that several animals of the most diverse classes, inhabiting the caves of Carinthia and Kentucky , are blind. In some crabs the eye-stalk is still present, although the eye is lost: the telescope-frame has remained, although the telescope with its glass is absent. As it is unlikely that eyes, though useless, should become injurious to animals living in darkness, I ascribe their loss entirely to non-use. In one of the blind animal species in particular, in the cave rat (Neotoma), of which Professor S. ILLIMANtwo, half a mile in from the entrance, and thus not quite caught in the background, the eyes were large and bright, and, SILLIMAN informed me , after a month's exposure to gradually increased light, acquired a vague perception of the objects held out to them and began to blink.

It is hard to imagine living conditions more similar than deep limestone caves in nearly similar climates, so that, given the common view that the blind animals were specially created for the American and for the European caves, one too great resemblance in organization and relationship would be expected. However, this is by no means the case between the two faunas on the whole, and SCHIÖDTE remarks with regard to the insects that the whole phenomenon[p. 165]should only be regarded as a purely local one, the resemblance which was found between some inhabitants of Monmouth Cave in Kentucky and the Carinthian Caves being a very simple expression of the analogy existing between the faunas of North America and Europe exist at all. It must be supposed, in my opinion, that American animals of ordinary sight have in successive generations penetrated deeper and deeper into the farthest recesses of the Kentucky cave, as have Europeans into the caves of Carinthiadone. And we have some evidence of this gradual change of residence; for S CHIOEDTE remarks: We accordingly regard these subterranean faunas as small branches, penetrated into the earth, of the geographically limited faunas of the immediate vicinity, which, as they spread further into the darkness, adapted themselves to the conditions around them; Animals, not far removed from ordinary forms, prepare the passage from day to darkness; then follow those formed for the twilight and finally those destined for complete darkness, whose formation is quite peculiar. These remarks S CHIÖDTE ' Stherefore do not refer to the same, but to quite different species. By the time an animal, after innumerable generations, reaches the rearmost parts of the burrow, disuse will thereafter have more or less completely suppressed the eyes, and natural selection will often have wrought other changes, such as elongated antennae and gnawing tips, to some extent replace the face. These modifications notwithstanding, we shall expect to see affinities of the cave-animals of America to the other inhabitants of that continent, and of the cave-dwellers of Europe to the rest of European animals. And this is the case with some American cave animals, as I have heard from Professor DANA listen; and some European cave insects are closely related to some of the species living in the vicinity of the cave. It might be very difficult to give a reasonable explanation of the relationship of the blind cave animals to[p. 166]to give to the other inhabitants of the two continents from the ordinary point of view of an independent creation. That some of the cave-dwellers of the Old and New Worldsto be closely related to one another may be expected from the well-known relationships of most of their other products to one another. As a blind species of Bathyscia is often found on shady rocks outside the caves, the loss of sight in the cave-dwelling species has probably had no relation to the obscurity of that dwelling; and it is quite understandable that an insect already blind should not find its way in the dwelling of a dark cave. Another blind genus, Anophthalmus, presents the curious peculiarity that, as M URRAY remarked, their distinct species are found in various caves in Europe , as well as in that of Kentuckylive, and that the clan only occurs in caves. It is possible, however, that the progenitor or progenitors of these various species were formerly widespread on both continents and (like the elephants of both continents) gradually restricted to their present narrow habitations. Far from wondering that some of the cave animals are of a very anomalous constitution, as A Gassiz remarks of the blind fish Amblyopsis in America , and as of the blind reptile Proteus in Europe If this is the case, I am rather amazed that there are not more wrecks of the old forms of life preserved in it, since such in these dark abysses would probably have been exposed to less severe competition [20] .

acclimatization. ) Habit is hereditary in plants as to the time of flowering, the amount of rain necessary for the process of germination, sleep, etc., and this prompts me to say something more about acclimatization. It is very common for species of the same genera to live in very hot and very cold regions[p. 167]occupancy; and as I believe that all species of a genus are descended from a common ancestor, if this be right, acclimatization must be easily effected during a long breeding period. It is known that each species is adapted to the climate of its own home; Species from an arctic or even a temperate region cannot survive in a tropical climate, etc. Thus some succulent plants cannot survive in a humid climate. However, the degree of adaptation of species to the climate in which they live has often been overestimated. We may infer this from our frequent inability to predict whether an introduced plant will endure our climate or not, and from the great number of plants and animals transplanted to us from warmer climates thrive quite well here. We have reason to believe that in the natural state species are as much or more limited in their distribution by the competition of other organic beings as by their adaptation to particular climates. But whether the adaptation in general be very accurate or not, we have evidence with a few species of plants that they are in some degree accustomed or acclimatized by nature to unequal temperatures. This is how those of Dr. H that they are already in a certain degree accustomed or acclimated by nature to unequal temperatures. This is how those of Dr. H that they are already in a certain degree accustomed or acclimated by nature to unequal temperatures. This is how those of Dr. HOOKER from seeds of Pinus and Rhododendron species raised from different heights of the Himalayas also have different ability to withstand the cold. Mr. T WAITES reports to me that he has observed similar facts in Ceylon , and Mr. HC W ATSON has had similar experiences with plants imported from the Azores to England have been brought. With regard to animals, many well-attested cases could be adduced, that species of the same have extended their distribution far from warmer to colder zones, or vice versa, within historical time; but we do not know with certainty whether these animals were once closely adapted to their native climate, though we assume this to be the case in all ordinary cases—and consequently whether or not they have first needed acclimatization in their new home.

[p. 168]

Since I believe that our domestic animals were originally tamed by still uncivilized people because they were useful to them and easy to reproduce in captivity, and not because of their suitability for extensive transplantation, which was discovered only later, I believe that the usually extraordinary ability of our domestic animals to endure and reproduce in the most varied climates (a much more weighty testimony) serve to conclude that a comparatively large number of other animals which are now in the state of nature could easily be brought to very different climates endure. We must not, however, push the foregoing conclusion too far, because some of our domestic animals may be of different wild stocks, such as B. in our domestic dog breeds the blood of a tropical and an arctic wolf or wild dog might be mingled. Rats and mice are not to be considered domestic animals; and yet they have been introduced by man into many parts of the world, and now have a wider distribution than any other rodent, roaming free under the cold skies of theFaroe Islands to the north and the Falklands Islandsin the south, as on many islands in the tropics. Hence I am inclined to regard adaptation to a particular climate as a quality easily grafted onto an innate wide flexibility of constitution, which is common to most animals. According to this view there is the ability of man and most of his domestic animals to endure the most varied climates, and such facts as the occurrence of ancient species of elephant and rhinoceros in a glacial climate, while all extant species have a tropical or subtropical home are not to be regarded as unlawful, but merely as instances of a very common flexibility of the constitution, which has only come into its own in special circumstances.

How much of the acclimatization of species to a particular climate is a matter of habit, how much of the natural[p. 169]Breeding of varieties with different body constitutions, or how far both causes work together, is a very difficult question. That habit and practice have some influence, I will well believe, both from analogy and from the constant warnings given in our agricultural works, and even in ancient ChineseEncyclopedias are included, to be very careful in moving animals from one region to another. For it is not probable that so many races and sub-races, with as many constitutions adapted to different regions, have been formed by selection; rather, the result comes from habituation. On the other hand, I see no reason to doubt that natural selection constantly tends to obtain those individuals born with the constitutions of body most suited to their native districts. In writings on different sorts of cultivated plants, certain varieties are said to endure this or that climate better than others. This follows very strikingly from the United Statespublished works on fruit-tree cultivation, in which these varieties are usually recommended for the northern and those for the southern states; and as most of these varieties are of recent origin, the difference in their constitutions in this respect cannot be ascribed to habit. The Jerusalem artichoke, which does not propagate from seed, and has therefore never produced new varieties, has been adduced as evidence that it is not possible to effect acclimatization because it is still as sensitive as it has always been: to the same One has often referred to the make-up bean as a purpose, and indeed with much greater emphasis. But as long as no one sows his beans so early for a few dozen generations,[p. 170]Raise seeds, it will not be possible to say that the attempt has been made for so long. Nor can it be supposed that differences in the constitution of these various bean-seedlings do not sometimes appear; for an account has already appeared of how much harder some of these seedlings are as compared with others.

On the whole, I think it may be concluded that habit, use, and disuse have in some cases exerted a considerable influence on the change in constitution and structure of various organs; that, however, these effects of use and disuse have often been increased to a considerable degree, and sometimes surpassed, by natural selection by means of innate variations.

interrelationships of education.)—I mean by this expression that the whole organization of natural beings is so interlinked that if, during the development and growth of one part, a slight change occurs and is accumulated by natural selection, other parts are also changed Need to become. This is a very important point, but little understood. The most common case is that variations which are accumulated only for the benefit of the larva or young no doubt also affect the organization of the adult; just as a deformity affecting the earliest embryo will seriously affect the whole organization of the old. The several parts of the body which are homologous, and which in the earliest embryonic period still resemble one another, seem to be alliedly inclined to vary; hence the right and left sides of the body tend to vary in the same way, the front limbs in the same way as the hind limbs, and even in the same way as the jaws, since the lower jaw is thought to be a homologue of the limbs. These tendencies can, I have no doubt, be more or less controlled by natural selection; thus there has formerly been a family of deer with an eye sprout on only one antler, and would that trait be of any greater use being more and less dominated by natural selection; thus there has formerly been a family of deer with an eye sprout on only one antler, and would that trait be of any greater use being more and less dominated by natural selection; thus there has formerly been a family of deer with an eye sprout on only one antler, and would that trait be of any greater use[p. 171]if it had been, it would probably have become permanent through natural selection.

Homologous parts, as some writers have remarked, tend to stick together, as is often seen in monstrous plants; and nothing is more common than the union of homologous parts in normal formations, as e.g. B. the union of the crown leaves into a tube [21] . Hard parts seem to affect the shape of adjacent softer ones, and some authors are of the opinion that the difference in shape of the pelvis of birds causes the curious difference in the shape of their kidneys. Others believe that in humans the shape of the mother's pelvis acts by compressing the shape of the child's skull [22] . In snakes, according to SCHLEGEL , the shape of the body and the way it is coiled determine the position of some of the most important viscera.

The nature of the bond of interrelationship is very often quite obscure. I SIDORE G EOFFROY S AINT -H ILAIREhas emphatically pointed out that certain malformations occur very frequently and others very seldom together, without our being able to state the reason. What can be more peculiar than the relation between the blue eyes and the deafness of some white cats, or that of the color of the carapace with the female sex of the tortoise; the relation between the feathered feet and the membrane between the outer toes of the pigeons, or that between the presence of more or less down on the newly hatched birds with the future color of their plumage; or finally between hair and tooth formation of the naked Turkish dog, although here probably homology[p. 172]comes into play. With reference to this last case of correlation, it seems to me scarcely accidental that those two orders of mammals which are most abnormal in dress are also most divergent in tooth formation; namely, the cetaceans (whales) and the edentates (pangolins, armadillos, etc.).

I know of no case better suited to showing the nature of the laws of interrelationship in the modification of important structures, independently of their utility, and thus of natural selection, than is the difference between the outer and inner flowers in the inflorescence of some Compositiflorae and Umbelliferae. Everyone knows the difference between the middle and the marginal blooms z. B. the daisy (Bellis), and this difference is often associated with the stunting of individual parts of the flower. But in some Compositiflora the fruits of the two flowers also differ in size and sculpture, and even the ovaries, with some subsidiary parts, differ, like C ASSINIproven. These differences have been ascribed by some botanists to pressure, and the fruit-forms in the ray-flowers of the composite flowers support this view; but it is by no means true with the Umbellifera that the species with the densest umbellae exhibit the greatest difference between the inner and outer flowers. It might have been thought that the greater development of the petals at the edge of the inflorescence had caused the other parts of the flower to deteriorate, by depriving them of nourishment; but in some Compositiflora there is a difference in the size of the fruits of the inner and ray flowers, without antecedent difference in the corolla. Possible, that these various differences are connected with some difference in the flow of the juices to the central and marginal flowers; we know, at least, that in irregularly formed flowers those closest to the axis are most often subject to peloria formation and become regular. I want as an example this and[p. 173]to cite at the same time as a striking case of interdependence of development, as I lately observed in some garden Pelargoniums, that the middle flowers of the umbel often lose the darker spots on the two upper petals, and that when this is the case the pendulous nectary is wholly atrophied; if the spot is missing on only one of the two upper petals, the nectary is only greatly shortened.

As to the differences in the corollas of the central and marginal flowers of an umbel or flower head, I hold CC S PRENGEL ' SThe notion that the ray-flowers are destined to attract the insects, whose movements tend to fertilize the plants of those two orders, does not seem so far-fetched as it may seem at first sight; and if it is really of use, natural selection may come into play. On the other hand, it scarcely seems possible that the difference between the structure of the outer and inner fruits, which bears no relation to any different formation of the flowers, can be of any use to the plants. However, in the umbelliferous plants the differences appear of such striking importance (since in several cases, after SWAP , the fruits of the outer flowers are orthospermous and those of the central colospermous) that the elder D E CANDOLLE based his chief divisions in this order of plants on analogous differences. We see, therefore, that variations in structure may be conditioned by wholly unknown laws in the interrelationships of evolution, without offering even the slightest discernible benefit to the species.

We may often erroneously ascribe to the interrelationships of development such formations as are common to whole groups of species, but which in truth depend quite simply on heredity. Because an old progenitor z. eg, some peculiarity of its structure may have been acquired by natural selection, and after a thousand generations some other independent variation, and if then both[p. 174]Modifications transmitted together to a whole group of offspring of different habits will naturally be thought to stand in a necessary interrelationship. So I don't doubt that some of the apparent interrelationships that occur in entire orders of the system are only due to the possible mode of action of breeding. if e.g. B. A LPHONS D E C ANDOLLEremarking that winged seeds never occur in fruit which do not open, I would like to explain this rule by the fact that seeds cannot be gradually winged by natural selection except in fruit which open; so that individual plants with seeds which are somewhat winged, and hence more apt for long-distance propagation, have an advantage over others ill-winged in their prospect of survival, and this process cannot well occur with such fruits which do not burst open.

The elder GEOFFROY and GOTHE formulated their law of the compensation of development almost simultaneously, according to which, like GOTHEexpresses itself, nature is compelled to save on the one hand what she gives more on the other. To a certain extent this applies, it seems to me, quite well to our culture products: for if food flows to one part or organ in abundance, it cannot, or at least not in abundance, also be given to another, hence one cow e.g. B. can not force to give a lot of milk and get fat at the same time. One and the same variety of cabbage cannot produce an abundance of nutritious leaves and at the same time a good yield of oil seeds. If the seeds in our fruit wither, the fruit itself gains in size and goodness. In our hens, a large crest of feathers is usually associated with a smaller crest on the head, and a large feather beard is connected with small rags of the beard.[p. 175]believe truth. However, I do not want to cite any examples here; for I find it difficult to find a way of distinguishing, on the one hand, between the considerable enlargement of a part brought about by natural selection and the reduction caused by the same cause or by disuse of another nearby organ, and on the other hand, the atrophy of an organ through loss of nourishment as a result of excessive use Development of another nearby part.

I also suspect that some of the cases advanced as evidence of compensation may be summed up with some other facts under a more general principle, viz., that natural selection is constantly striving to economize in every part of the organization. If, under changed living conditions, a previously useful device becomes less useful, then natural selection should probably aim at reducing its size, even if only insignificantly, since it is advantageous for the individual if his fluids are not used for the development of useless organs wasted. Only in this way can I understand a fact which surprised me when I was engaged in the investigation of the cirripedes, namely, that, if a Cirripede lives in other organisms as a parasite, and is therefore protected, it more or less loses its own calcareous shell. This is the case with the male of Ibla, and in an exceedingly high degree with Proteolepas; for while the carapace of all other Cirripeds consists of the three all-important anterior segments of the immensely developed head, and is furnished with strong nerves and muscles, on the parasitic and protected Proteolepas the whole anterior part of the head appears as a mere rudiment attached to the bases of the tendrils. Now the sparing of a large and complex structure, when rendered superfluous, as here by the parasitic mode of life of Proteolepas, though only progressively progressive, would be a decided advantage to any later individual of the species[p. 176]be, because in the struggle for existence, which the animal has to fight, each individual proteolepas gains the better chance of asserting itself, the less nutrient is lost for the development of an organ that has become useless.

After that, I believe, natural selection in length will always succeed in reducing and saving every part of the organization as soon as it has become superfluous, without developing another part in a corresponding degree to a greater extent. And in the same way, conversely, it should be perfectly able to develop an organ more strongly, without requiring the diminution of another neighboring part as necessary compensation.

To I SIDORE G EOFFROY S AINT -HILAIRE ' S perception it seems to be the rule in varieties, as in species, that when a part or organ often occurs in the structure of an individual, as the vertebra in snakes, and the stamens in the polyandric flowers, the number of which becomes variable, while the number of the same organ or part remains constant, if it must be repeated less frequently. The same zoologist, as well as some botanists, have also remarked that a great many parts are also subject to great changes in internal structure. Especially since these vegetative repetitions, like R. O WENthey call are an indication of low organization, the preceding remark seems to be connected with the very general view of naturalists that those beings which are low in the scale of nature are more variable than the higher ones. By deep organization I understand in this case a slight differentiation of the organs for various special functions; for so long as one and the same organ has various jobs to do, one reason for its variability may perhaps be found in Natural Selection being less carefully preserved or suppressed of every slight variation of form than if the same organ were dedicated to a particular purpose alone. So knives, which are designed to cut all sorts of things, on the whole pretty much like[p. 177]be of the same form, while a tool intended for only one use must also have a different form for every other use.

Imperfectly developed, rudimentary organs are, according to the remark of some writers, which seems to me to be correct, very inclined to change. In this respect I refer to the discussion of the rudimentary and abortive organs in general, and will only add that their variability appears to be due to their being useless, in which case natural selection is powerless to prevent deviations in their structure. Hence rudimentary parts are subject to the free influence of the various laws of growth, to the effects of long-continued disuse, and to the tendency to return.

A part developed to an extraordinary degree or manner in any species has, in comparison with the same part in other species, a great tendency to change. ) — A few years ago I was surprised by a similar statement published by W ATERHOUSE . I also conclude from a remark made by Professor R. O WENover the length of the orangutan's arms that he came to the same view. There is no hope of convincing anyone of the truth of this assertion, without enumeration of the long series of facts which I have collected, but cannot here relate. I can only express my belief that it is a very general rule. I know of several causes which can give rise to errors in this respect, but I hope to have considered them sufficiently. Above all, it is to be observed that this rule should not apply to any part, however unusually developed in itself, unless it is also unusually developed in relation to the same part in closely allied species. Therefore, however abnormal the wing formation of bats in the class of mammals is,[p. 178]order comes; it would apply only if the wings of one species of bat were enlarged in a remarkable proportion to the wings of other species of the same genus. This rule corresponds very well to the unusually involved "secondary sexual characters" with which expressions H UNDERdesignates those characteristics which belong only to the male or the female alone, but are not directly connected with the act of reproduction. The rule applies to both males and females, but more so to the former, because conspicuous characters of this kind are generally rare in females. The perfect applicability of the rule to these latter cases is probably connected with the great and undoubted variability of these characters in general, whether they are much or little developed. But that our rule does not in fact apply to the secondary characters of this kind alone, is evident from the hermaphrodite cirripeds; and I will add here that in examining this order I have met Mr. W ATERHOUSE ' S remark and am fully convinced of the almost invariable applicability of this rule to the Cirripedes. In my later work I shall give a fuller list of the individual cases; but here I will quote only one, which explains the rule in its widest application. The opercular flaps of the sessile Cirripeds (Balanids) are very important structures in every sense of the word, and vary but little even from one taxon to another. Only in the several species of Pyrgoma alone do these valves present a wondrous degree of differentiation. The homologous valves are sometimes quite dissimilar in form in different species, and the amount of possible variation between the individuals of some species is so great that it can be asserted without exaggeration may,

Since birds within one and the same area are extremely[p. 179]vary little, so I examined them more closely in this regard and found the rule to be very effective in this class as well. I cannot prove that it behaves in the same way in plants, and my confidence in its generality would have been greatly shaken by this if the great variability of plants in general did not make it very difficult to compare the relative degrees of variability of both to compare.

If in any species we see a part or organ developed to a remarkable degree or manner, it would be natural to suppose that it must of that kind be of great importance, and yet the part in this case is exceedingly variable. How come this? From the view that each species, with all its parts, as we now see them, was created independently, we can draw no explanation. On the other hand, the assumption that groups of species share a common ancestry from other species and have been modified only by natural selection seems to shed some light on the question. In our domestic animals, if a single part or the whole animal is neglected and propagated without selection, such a part (as e.g. the comb in the Dorking hens) or the whole breed cease to be uniform in character. It will then be said that she has degenerated. In rudimentary and those organs which have been but little differentiated for a particular purpose, as well as in polymorphic groups, we see an almost concurrent case in nature; for here natural selection cannot, or only to a small extent, come into its own and the organization remains in a state of vacillation. But what concerns us here is that in our domestic animals those characters which are subject to so rapid modification through continued breeding, are just as rapidly inclined to vary. Compare the breeds of pigeons; what a marvelously great measure of change is shown only in the beaks of the somersault-doves,[p. 180]Tail of fantail because the English Lovers pay little attention to these points. Even the sub-breeds, such as the short-faced tumblers, are, as is well known, difficult to find in perfect condition, and often individual animals come to light which deviate far from the model picture. It may, therefore, be said with truth that there is a constant struggle between, on the one hand, a tendency to return to a less differentiated condition and an innate tendency to further change of every kind, and, on the other hand, the power of continual selection for the preservation of the race. In the long run, selection wins the day, and we no longer fear to stray so far from the goal that we only get a mean tumbler from a good, short-faced strain. But as long as breeding is still in rapid progress, great inconsistency in the structure subject to change will always be to be expected. It also deserves to be remarked, that these variable characters produced by artificial breeding, from causes quite unknown to us, are sometimes attached more to one sex than to the other, and usually to the male, like the flesh warts of theEnglish carrier pigeon and the pouter's mighty crop.

But let's return to nature. When a part in any species is exceedingly enlarged in relation to the other species of the same genus, we may suppose that it has undergone an unusual amount of modification since its branching from the common stock. This time of diversion will seldom be exceedingly remote, since species seldom last more than one geologic period. An unusual amount of difference implies an unusually long and extensive measure of variability, the product of which has been continually accumulated by breeding for the good of the species. But since the variability of the extraordinarily developed part or organ has been so great and enduring at a not very distant time that[p. 181]of the organization, which have been established for a much longer time, expect to encounter. And I believe this will happen. But that the struggle between natural selection, on the one hand, and the tendency to return and to further modification, on the other, ceases in time, and even the most abnormally formed organs may become permanent, there is no reason to doubt. Hence if an organ, however regular, be transmitted in much the same condition to many already mutating offspring, as is the case with the wing of the bat, it must, according to my theory, have been in the same condition for an immeasurable length of time been present, and as a result are now no more variable than any other organ. Only in those cases where the modification is comparatively recent and exceedingly great, we shall therefore find what we shall call 'generative variability' still in a high degree persisting. For in this case the variability will only seldom have reached a fixed goal through uninterrupted breeding of the individuals who vary in some intended manner and degree, and through constant repression of those inclined to return.

The principle contained in these remarks is still capable of extension. Namely, it is known that the specific characters tend to change more than the clan characters. I will explain what I mean with a simple example. If, in a large genus of plants, some species have blue flowers and others have red, the color will only be of a species character, and therefore no one will be surprised if a blue-flowered species changes to red, or vice versa. But when all species have blue flowers, the color becomes a characteristic of the clan, and the change in it will be an unusual phenomenon. I have chosen this example precisely because an explanation which most naturalists would otherwise be inclined to adduce is not applicable to it,[p. 182]of less physiological importance than those usually used in the classification of genera. While I believe that this explanation is partly, if only indirectly, correct, I can return to it only in the section on classification. It may be quite superfluous to give examples in support of the above assertion that species-characters are more variable than clan-characters; but I have repeatedly gathered from natural-historical works that when a writer was surprised by the observation that some more important organ, which otherwise tends to be constant in whole large groups of species, changes appreciably in closely related species, then also in varied in the individuals of some of the species. This fact shows that a character which is usually of generic value, often becoming variable when declining to specific value, though its physiological importance remains the same. Something similar applies to monstrosities as well; at least it seems ISIDORE G EOFFROY S AINT -H ILAIRE to be in no doubt that the more an organ differs in the different species of the same group, the more subject it is to individual anomalies.

On the common notion that each species was independently created, how would it be explained that that part of the organization which differs more from the same part in other independently created species of the same genus is also more variable than those parts which exist in the different types of a clan almost match. I don't see a way to explain a diess. But if we proceed from the view that species are only well-distinguished varieties that have become constant, then we can certainly expect to see that they still often continue to change in those parts of their organization which only came about in relatively recent times in succession have shrunk from the more ordinary formation because of their variation. Or, to put the case in a different way: the characteristics[p. 183]differing from all species of another genus are called generic, and these characters taken together I derive by inheritance from a common ancestor; for only seldom can it have happened by chance that natural selection has modified in exactly the same way different species adapted to more or less divergent habits; and as these so-called generic characters have been inherited from very early on, from the time they branched off from their common parent species, and have changed little or no longer afterwards, it is not probable that they change this very day. On the other hand, the points by which species differ from other species of the same clan are called specific characters,

In connection with this subject I want to make two other remarks. — Without needing to go into detail, it will be granted to me that secondary sexual characters are very variable; I will also, I suppose, be allowed further that the species belonging to the same group may differ more widely in these characters than in other parts of their organization. If, for example, one compares the magnitude of the difference between the males of Gallinaceous birds, in which these kinds of characters are chiefly developed, with the magnitude of the difference between their females, the truth of that assertion will be admitted. The cause of the original variability of the secondary sexual characters has not been established; but it is understandable how come they have not become so uniform and constant as other parts of the organization; for the secondary sexual characters are by sexual selection[p. 184]been accumulated, which is less severe in its action than the common one, not destroying the less favored males, but merely providing fewer offspring. But whatever may be the cause of the variability of these secondary sexual characters; being very variable, natural selection has found in it a wide scope for its action, and thus can easily impart to the species of a group a greater amount of difference in their sexual characters than in other parts of their organization.

It is a remarkable fact that the secondary sexual differences between the two sexes of a species are usually exhibited in the very same parts of organization in which the different species of a genus differ. To illustrate this I will give just two examples which happen to be first on my list; and as the differences in these cases are of a very unusual nature, the relation can hardly be accidental. Very large groups of beetles have an equal number of tarsal joints in common; only in the Engidae family changes to W ESTWOOD ' SObserve this number very much, even in the two sexes of a species. Also in the Burrowing Hymenoptera the course of the wing veins is a character of the greatest importance, as remaining the same in large groups; in some genera, however, it varies from species to species, and then often in the same way in the two sexes of the same species. LUBBOCK _recently remarked that some small crustaceans offer excellent evidence of this law. In Pontella e.g. B. It is mainly the anterior antennae and the fifth pair of legs which furnish the sexual characters, and the same organs also furnish the most important species differences. This relation has a clear meaning in my way of looking at things: namely I look at with certainty all species of a genus as descendants from the same progenitor, as are the two sexes in each species. Hence: whatever part of the organization of the common[p. 185] become mutable in the progenitor or his first offspring, variations in these parts will most probably have been favored by natural and sexual selection, to adapt the different species to different places in the economy of nature, and also to have the two sexes of the same species apt for one another or to adapt males and females to different ways of life, or finally to enable the males to fight with other males for the females.

All arise chiefly from the species belonging to a group descending from a common ancestor, from whom they have inherited much in common; — that parts which have only recently undergone great changes are more apt to vary than those which have been inherited for a long time without any change; — that sexual breeding is less severe than ordinary; Finally, that variations in the same organs have been accumulated by natural and sexual selection, and adapted for secondary sexual and ordinary specific purposes. tend to vary more easily than those which have been passed on for a long time without any change; — that sexual breeding is less severe than ordinary; Finally, that variations in the same organs have been accumulated by natural and sexual selection, and adapted for secondary sexual and ordinary specific purposes. tend to vary more easily than those which have been passed on for a long time without any change; — that sexual breeding is less severe than ordinary; Finally, that variations in the same organs have been accumulated by natural and sexual selection, and adapted for secondary sexual and ordinary specific purposes.

Different species show analogous variations; and the variety of a species often takes on some of the characters of an allied species,[p. 186]or it reverts to some of the characteristics of the parent species.) These assertions are most easily understood by considering the domestic animal races. The most diverse breeds of pigeons present, in widely separated regions, sub-varieties with inverted feathers on the head and feathers on the feet, characters which the original rock-pigeon does not possess; These, then, are analogous variations in two or more different races. The frequent presence of fourteen to sixteen tail-feathers in the pouter may be regarded as a deviation representing the normal formation of another species, namely the fantail. I suppose no one will doubt that all such analogous variations result from the different breeds of pigeons having inherited the same constitution, and hence, under the same unknown influences, the same tendency to vary.Swedish Turnipses and the Rutabaga, plants which many botanists regard only as cultivated varieties of a species. But if this were not correct, we would have a case of analogous variation in two so-called types, and to these the common turnips can be added as the third. On the common view that each species was created independently, we should have to attribute this resemblance of the three plants in their thickened stalk, not to the true cause of their common descent, and a consequent tendency to vary in a like manner, but to three distinct but closely related related acts of creation. With the pigeons we have another case, viz. the occasional appearance in all breeds of slate-blue birds with two black wing-bars, a white rump, a transverse band on the end of the tail, and a white outer border at the base of the outer tail feathers. As all these characters are characteristic of the parent species, I believe no one will doubt that this is a[p. 187]Return to the original character and not an analogous variation in different races. We shall have the more confidence in this conclusion, as, as we have already seen, these colored characters are very likely to appear in the hybrids of two quite differently colored races; and in this case there is nothing to be found in the external conditions of life either, which could explain the reappearance of the slate-blue color with the other color markings, except the influence of the act of crossing on the laws of heredity.

It is indeed an amazing fact that characters lost for many, and perhaps hundreds, of generations are reappearing. However, when a breed has been crossed only once with another breed, the blind shows a tendency occasionally to revert to the character of the alien breed for some, say 12-20, generations. Now, after 12 generations, according to the usual way of expression, the blood of one foreign ancestor is only 1 in 2048, and yet, according to the general assumption, this extremely small fraction of foreign blood is still sufficient to indicate a tendency to return to that original tribe converse. In a race which has not been crossed, but in which bothOlder having lost some of the characters of their common stock, the greater or lesser tendency to restore the lost character, as before remarked, might persist for a number of generations, notwithstanding what may be seen to the contrary. When a character lost in a race returns after a large number of generations, the most probable hypothesis is not that the descendant is now suddenly reverting to a ancestor several hundred generations older, but that in each of the successive For generations there had still been a striving to restore the character in question, which is now finally breaking through under unknown favorable circumstances. So e.g. B. probably,p. 32 ), which[p. 188]very rarely produces a blue bird with black bars; the striving to adopt this coloration is present. This view is hypothetical, but may be supported by some facts; and I cannot see in and of itself any greater improbability in the supposition of a tendency to resume a character which has been inherited through an endless number of generations than in the inheritance of an organ which is actually quite useless or rudimentary. And yet we can at times perceive such a striving to produce an inherited rudiment, as e.g. For example, in the common snapdragon (Antirrhinum), the rudiment of a fifth stamen is so often seen, that a tendency to produce it must be inherited in this plant.

As, on my theory, all the species of a genus are of a common descent, it is to be expected that they will sometimes vary in an analogous manner, so that a variety of one species will resemble in some of its characters another species, which, in my opinion, is itself only a developed one and has become a permanent variety. But the characters thus acquired should only be of an insignificant kind; for the presence of all the essential characters is directed by natural selection in accordance with the different habits of the species, and is not left to the mutual action of the conditions of life and a like inherited constitution. It may further be expected that the species of the same genus will sometimes show a tendency to revert to the characters of their ancient ancestors. However, as we never know the exact character of the common progenitor of a group, we cannot distinguish these two cases. If we e.g. B. not knowing that the rock-pigeon is not provided with feathered feet, or with inverted feathers, we could not have said whether these characters in our domestic pigeon-breeds were appearances of reversion to ancestral form, or merely analogous variations; but we might have assumed that the blue coloring is an example of return, because of the number of other drawings that have the blue coloring whether these characters in our domestic pigeon races are appearances of return to parent form, or merely analogous variations; but we might have assumed that the blue coloring is an example of return, because of the number of other drawings that have the blue coloring whether these characters in our domestic pigeon races are appearances of return to parent form, or merely analogous variations; but we might have assumed that the blue coloring is an example of return, because of the number of other drawings that have the blue coloring[p. 189] reappear at the same time and probably not coincide with it merely as a result of simple modification. And still more should we have inferred it, from the blue color and other markings so often reappearing when different races of differing coloring are crossed together. Therefore, although in the wild it usually remains doubtful which cases are to be regarded as reversions to old parental characters and which as new analogous variations, yet, on my theory, we must sometimes find that the modifying descendants of a species (let it be so by return or by analogous variation) assume characters already present in some other members of the same group. This is undoubtedly the case in nature.

A great deal of the difficulty in recognizing a variable species in our systematic works arises from its varieties imitating, as it were, some of the other species of the same genus. Also, one might give a considerable list of forms which are intermediate between two other forms, which it is doubtful whether they should be regarded as species or varieties; and it follows, unless all these forms are to be regarded as independently created species, that the one has so far assumed, by modification, the characters of the other, as to thereby form an intermediate form. But the best proof is offered by parts or organs of essential and uniform nature sometimes so varying, that they assume somewhat the character of the same organ or part in an allied species. I have compiled a long list of such cases, but unfortunately I cannot report them here, but merely repeat that such cases do occur and seem very strange to me.

I will, however, give a peculiar and compound case, not concerning any important character, but occurring in several species of a genus, partly in the natural and partly in the domesticated state. It is obviously a case of return. The donkey sometimes has very clear transverse bands[p. 190]on his legs, like the zebra. It has been asserted that these are most clearly seen during filling, and my research seems to confirm this. It has also been asserted that the stripe on the shoulder is sometimes double. In any case, the shoulder stripe is very variable in length and outline. A non-albino white donkey has also been described without back and shoulder stripes; and these stripes are sometimes very indistinct, or not seen at all, even in dark-colored animals. The kulan of P ALLAS is said to have been seen with a double shoulder stripe. The hemionus has no shoulder stripe; but according to B LYTH ' S et al., traces of it sometimes occur, and Colonel P OOLEinformed me that the foals of this species are sometimes striped on the legs and faintly on the shoulder. The quagga, though striped on the body as distinctly as the zebra, is without bands on the legs; but has dr. G RAY an individual with very distinct bands on the legs, similar to those of the zebra.

As for the horse, I have collected cases in England of the occurrence of back-stripes in all breeds and colors. Examples of queer bandages on the legs were not uncommon in bays, musky, and once in a chestnut. A faint shoulder stripe also sometimes appears in bays, and I found a trace of it in a berry brown. My son has communicated to me a careful examination and drawing of a bay Belgian Cart-Horse with a double stripe on the shoulder and with stripes on the legs; I myself have seen a brown Devonshire pony, a reliable man has given me the careful description of a small brown Welshmanbangs, both of which were provided with three short parallel stripes on each shoulder.

In the north-western part of the East Indies the Kattywar horse breed is so generally striped that, as I understand from Colonel POOLE , who examined it on behalf of the Government, a horse without stripes is not considered Thoroughbred.[p. 191]The spine is always striped; the stripes on the legs, like the shoulder-stripe, which is sometimes double and even triple, are common; moreover, the sides of the face are sometimes striped. The streaks are most noticeable when the filly is filled and sometimes disappear with age. P OOLE found very young colts both gray and berry brown striped. Also, from information I am indebted to Mr. WW E DWARDS , I have reason to suspect that back-stripes are more common in English racehorses than in old horses. I recently had a foal by a chestnut mare (the daughter of a Turkomannian stallion and a Flemishmare) and a chestnut bay English breed horse. This foal, a week old, was narrowly zebra-striped on the back towards the tail and forehead, and on the legs fainter such stripes, all of which will no doubt soon disappear. Without going into further detail here, I will say that I have collected cases of leg and shoulder stripes in horses of quite different breeds, in different regions from England to East China , and from Norway north to Malaysiaarchipelago in the south. In all parts of the world these stripes are by far the most common on bays and musky browns. Under brown par excellence (“Dun”) I understand horses with a long range of color gradations from black-brown to almost cream-colored [23] .

I know that Colonel H AMILTON SMITH , writing on this subject, supposes our various breeds of horses to have descended from different stocks, one of which, that of the bay, having been striated, and that all the striations above described are of antecedent origin Interbreeding with the brown strain. However, I am not at all satisfied with this theory and do not want it on such different breeds[p. 192]such as the Belgian cart horse, the Valais pony, the Renner, the slender Kattywar breed, and others, which are scattered in various parts of the world.

Turning now to the consequences of cross-breeding between the various species of the horse-kin : R OLLIN affirms that the common mule, of donkey and horse, very often has transverse stripes on its legs, and after GOSSE comes Diess in the United States nine times in ten cases. I once saw a mule with legs so heavily striped that anyone would have been inclined to derive it from the zebra; and Mr. WC M ARTIN has given in his excellent work on the horse the figure of a similar mule. In four colored pictures of donkey-zebra hybrids I found the legs striped much more conspicuously than the rest of the body, and in one of these there was a double shoulder stripe. On Lord M ORTON 'S famous crossbreed by a quagga stallion and a chestnut mare, and on a subsequently obtained pure foal by the same mare by a black Arabian, the legs were much more queer-striped than even the pure quagga. Recently, and this is another very curious case, Dr. GRAY _(who is aware of a second example of this kind) depicts a hybrid of Donkey and Hemionus, on which hybrid, although the Donkey only occasionally and the Hemionus never has stripes on the legs and the latter not even a shoulder-stripe, all four legs are queer striped and also the shoulder has three stripes like the Valais and Devonshire pony browns ( p. 190 ), and even some stripes like the zebra are present on the sides of the face. This last fact has convinced me that not even a streak of color arises by so-called accident, hence I was prompted by this appearance alone in a hybrid of Donkey and Hemionus, Colonel POOLEto ask whether such facial stripes ever occur in the heavily striped Kattywar horse breed, which, as we have seen above, he affirmed.

What remains for us now to these various facts[p. 193]still to say? We see several substantially different species of the equine family, by simple modification, acquiring stripes on the legs as in the zebra, or on the shoulder as in the donkey. In the horse we see this tendency strongly accentuated whenever one of the most natural colors of a horse comes to the fore. The appearance of the stripes is not accompanied by any change in shape or new character. We see this tendency to streak develop most frequently in hybrids between several distinct species. Let us compare with this the preceding case of the pigeons: they descend from a parent species (having 2-3 geographical varieties or subspecies) bluish in color and provided with some definite band-markings, and take, when one of their races, through simple modification, again produces a blue brood, those bands of parent-form inevitably reappear, but without any other change in race-character. When the oldest and truest breeds of different coloring are crossed together, there is a strong tendency in the mongrels to resume the original slate-blue color with the black and white bands and stripes. I have argued that the most probable hypothesis for explaining the reappearance of very old characters is the assumption of a "tendency" in the young of each new generation to recover the long-lost character, a tendency which sometimes breaks through from unknown causes. Then we saw that in different species of the horse-sex the stripes are more conspicuous or common in the young than in the old. If one wanted to designate the pigeon races, some of which have been propagated for centuries by pure inbreeding, as a species, the appearance would be exactly the same as with the horse race. Looking back thousands and thousands of generations, I recognize with confidence a zebra-striped but otherwise perhaps very differently built animal as the common ancestor of the domestic horse (let's touch it on one like the horse tribe. Looking back thousands and thousands of generations, I recognize with confidence a zebra-striped but otherwise perhaps very differently built animal as the common ancestor of the domestic horse (let's touch it on one like the horse tribe. Looking back thousands and thousands of generations, I recognize with confidence a zebra-striped but otherwise perhaps very differently built animal as the common ancestor of the domestic horse (let's touch it on one[p. 194]or from several wild tribes), of the donkey, the hemionus, the quagga, and the zebra.

Those who believe in the independent creation of each species of horse will probably say that each species has been so peculiarly created in the tendency to vary in the free as in the domesticated state, that it often appears streaked like other species of the same genus; and that each of them, when crossed with species from the farthest corners of the world, has a strong tendency to produce hybrids resembling not their own elders in their stripes, but other species of the same genus [24]. To profess this view is, in my opinion, to give up an actual cause for an immaterial, or at least unknown, cause. It makes of the works of God only deceit and aping; — and I was almost as ready to suppose, with the ancient and ignorant cosmognists, that the fossil shells never belonged to any living animal, but were created in the rock to imitate the shellfish now living on the sea-coast.

Summary. ) We are in deep ignorance of the laws according to which changes are made. Not in one case in a hundred can we pretend to know the reason why this or that part of an organism differs more or less from the same part in its parents. But wherever we have the means of making a comparison, the same laws seem to have acted in producing lesser differences between varieties of the same species, and in producing greater differences between species of the same genus. The external conditions of life, such as climate, food, and the like, have probably only a few minor changes[p. 195]conditional. Getting used to the constitution of the body, use of the organs to their strengthening, disuse to their weakening and diminution may have had more important consequences. Homologous parts are apt to vary in the same way, and tend to be connected among themselves. Variations in the hard and external parts sometimes affect softer and internal organs. When one part develops greatly, it may tend to deprive other parts of its nourishment; — and every part of the organic structure which can survive without harm to the individual is preserved. A change in organization at an early age also affects the later developing parts; but then there are still many interrelationships of development, the nature of which we are by no means able to comprehend. Numerous parts are more variable in number and structure, perhaps because they have not yet been sufficiently adapted and differentiated by natural selection for particular functions. For the same reason, organisms that are on a lower level of organization will probably be more variable than those that are more highly developed and more differentiated in all respects. Vestigial organs are ignored by Natural Selection because of their uselessness, and are therefore probably variable. Specific characters, namely those which have only diverged since the various species of a clan have diverged from a common progenitor, are more variable than generic characters which have long since been inherited as such without undergoing any modification during that time. We have here referred only to the individual parts and organs which are still variable, because they have only recently varied and become dissimilar to one another; but we have already seen in the second chapter that the same principle applies to the whole animal; for in a district where many species of one genus are found, i. H. where formerly much variation and differentiation took place, and the fabrication of new species-forms was vigorously pursued, in this district and[p. 196]among these species we now find on average the most varieties. — Secondary sex characters are very variable, and such characters vary most in the species of the same group. Variability in the same parts of organization has usually furnished the secondary sexual differences for the two sexes of a species, as well as the species differences for the various species of the same genus. A limb or organ developed in an extraordinary size or way - namely, taken in comparison with the development of the same limb or organ in the most closely related species - must have undergone an extraordinary degree of change since the appearance of the genea, from which we then find it understandable that why it is still subject to change in a higher degree than other parts; for modification is a slow and long-lasting process, and natural selection in such cases will not have had time to overcome the tendency to further change and return to a less modified state. But if a species with some extraordinarily developed organ has become the ancestor of many modified offspring - which in my opinion is a very slow process and therefore takes a lot of time - then natural selection may have been able to produce the organ, however extraordinarily developed it may be may be, already to impress a firm character. When species have inherited nearly the same constitution from a progenitor and have been exposed to similar influences, so of course they will be inclined to make analogous variations, and will at times revert to some of the characters of their earliest ancestors. Although new and important modifications will not result from this reversal and those analogous variations, such modifications add to the beauty and harmonious diversity of nature.

But whatever may be the cause of the first small difference between parents and offspring, and there must be a cause for it, it is only the constant accumulation of such[p. 197]differences useful to the individual through the natural selection which produces all the more important modifications of structure, through which the innumerable beings of our surface of the earth are enabled to struggle with one another for existence, and through which that which is best adapted for this purpose survives the others .