On the Origin of Species

A PENN STATE ELECTRONIC CLASSICS SERIES PUBLICATION On the Origin of Species by Charles Darwin is a publication of the Pennsylvania State University. This Portable Document file is furnished free and without any charge of any kind. Any person using this document file, for any purpose, and in any way does so at his or her own risk. Neither the Pennsylvania State University nor Jim Manis, Faculty Editor, nor anyone associated with the Pennsylvania State University assumes any re- sponsibility for the material contained within the document or for the file as an electronic transmission, in any way.

On the Origin of Species by Charles Darwin, the Pennsylvania State University, Electronic Classics Series, Jim Manis, Faculty Editor, Hazleton, PA 18202-1291 is a Portable Docu- ment File produced as part of an ongoing student publication project to bring classical works of literature, in English, to free and easy access of those wishing to make use of them.

Cover Design: Jim Manis

The Pennsylvania State University is an equal opportunity university. Charles Darwin too well studied in the book of God’s word, or in the On the Origin book of God’s works; divinity or philosophy; but rather let men endeavour an endless progress or of Species proficience in both.’ Bacon: Advancement of Learning. by Down, Bromley, Kent, Charles Darwin October 1st, 1859.

On the Origin of Species by Means of Natural Selec- ‘But with regard to the material world, we can at tion, or the Preservation of Favoured Races in the least go so far as this—we can perceive that events Struggle for Life. are brought about not by insulated interpositions of Divine power, exerted in each particular case, but By by the establishment of general laws.’ Charles Darwin

W. Whewell: Bridgewater Treatise.

‘To conclude, therefore, let no man out of a weak conceit of sobriety, or an ill-applied moderation, think or maintain, that a man can search too far or be 3 On the Origin of Species Contents Chapter II

Introduction Variation under Nature

Chapter I Variability — Individual Differences — Doubtful species — Wide ranging, much diffused, and com- Variation under Domestication mon species vary most — Species of the larger genera in any country vary more than the species of the Causes of Variability — Effects of Habit — Correla- smaller genera — Many of the species of the larger tion of Growth —Inheritance — Character of Domes- genera resemble varieties in being very closely, but tic Varieties — Difficulty of distinguishing between unequally, related to each other, and in having re- Varieties and Species — Origin of Domestic Variet- stricted ranges. ies from one or more Species — Domestic Pigeons, their Differences and Origin — Principle of Selec- Chapter III tion anciently followed, its Effects —Methodical and Unconscious Selection — Unknown Origin of our Struggle for Existence Domestic Productions — Circumstances favourable to Man’s power of Selection. Bears on natural selection — The term used in a wide sense — Geometrical powers of increase — Rapid increase of naturalised animals and plants —Nature of the checks to increase — Competition universal

4 Charles Darwin — Effects of climate — Protection from the number acter, related to the diversity of inhabitants of any of individuals — Complex relations of all animals small area, and to naturalisation — Action of Natu- and plants throughout nature — Struggle for life ral Selection, through Divergence of Character and most severe between individuals and varieties of the Extinction, on the descendants from a common par- same species; often severe between species of the ent — Explains the Grouping of all organic beings. same genus — The relation of organism to organism the most important of all relations. Chapter V

Chapter IV Laws of Variation

Natural Selection Effects of external conditions — Use and disuse, combined with natural selection; organs of flight and of Natural Selection — its power compared with man’s vision — Acclimatisation — Correlation of growth selection — its power on characters of trifling im- — Compensation and economy of growth — False portance — its power at all ages and on both sexes correlations —Multiple, rudimentary, and lowly — Sexual Selection — On the generality of inter- organised structures variable — Parts developed in crosses between individuals of the same species — an unusual manner are highly variable: specific char- Circumstances favourable and unfavourable to Natu- acters more variable than generic: secondary sexual ral Selection, namely, intercrossing, isolation, num- characters variable —Species of the same genus vary ber of individuals — Slow action — Extinction in an analogous manner — Reversions to long-lost caused by Natural Selection —Divergence of Char- characters — Summary.

5 On the Origin of Species Chapter VI Chapter VII

Difficulties on Theory Instinct

Difficulties on the theory of descent with modifica- Instincts comparable with habits, but different in their tion — Transitions —Absence or rarity of transitional origin —Instincts graduated — Aphides and ants — varieties — Transitions in habits of life — Diversi- Instincts variable — Domestic instincts, their origin fied habits in the same species — Species with hab- — Natural instincts of the cuckoo, ostrich, and para- its widely different from those of their allies — Or- sitic bees — Slave-making ants — Hive-bee, its cell- gans of extreme perfection — Means of transition — making instinct — Difficulties on the theory of the Cases of difficulty — Natura non facit saltum —Or- Natural Selection of instincts —Neuter or sterile in- gans of small importance — Organs not in all cases sects — Summary. absolutely perfect —The law of Unity of Type and of the Conditions of Existence embraced by the Chapter VIII theory of Natural Selection. Hybridism

Distinction between the sterility of first crosses and of hybrids —Sterility various in degree, not universal, affected by close interbreeding, removed by domestication — Laws governing the sterility of hy-

6 Charles Darwin brids — Sterility not a special endowment, but inci- sudden appearance of groups of species — On their dental on other differences — Causes of the sterility sudden appearance in the lowest known fossilifer- of first crosses and of hybrids —Parallelism between ous strata. the effects of changed conditions of life and crossing — Fertility of varieties when crossed and of their Chapter X mongrel offspring not universal — Hybrids and mongrels compared independently of their fertility On the Geological Succession of Organic Beings — Summary. On the slow and successive appearance of new spe- Chapter IX cies — On their different rates of change — Species once lost do not reappear — Groups of species fol- On the Imperfection of the Geological Record low the same general rules in their appearance and disappearance as do single species — On Extinction On the absence of intermediate varieties at the — On simultaneous changes in the forms of life present day — On the nature of extinct intermediate throughout the world — On the affinities of extinct varieties; on their number — On the vast lapse of species to each other and to living species — On the time, as inferred from the rate of deposition and of state of development of ancient forms — On the suc- denudation —On the poorness of our cession of the same types within the same areas — palaeontological collections — On the intermittence Summary of preceding and present chapters. of geological formations — On the absence of intermediate varieties in any one formation — On the

7 On the Origin of Species Chapter XI source with subsequent modification —Summary of the last and present chapters. Geographical Distribution Chapter XIII Present distribution cannot be accounted for by differences in physical conditions — Importance of bar- Mutual Affinities of Organic Beings: riers — Affinity of the productions of the same con- Morphology: Embryology: Rudimentary Organs tinent — Centres of creation — Means of dispersal, by changes of climate and of the level of the land, Classification, groups subordinate to groups — and by occasional means — Dispersal during the Natural system — Rules and difficulties in classifi- Glacial period co-extensive with the world. cation, explained on the theory of descent with modification — Classification of varieties — Descent al- Chapter XII ways used in classification — Analogical or adaptive characters — Affinities, general, complex and Geographical Distribution — continued radiating — Extinction separates and defines groups —Morphology, between members of the same class, Distribution of fresh-water productions — On the between parts of the same individual — Embryol- inhabitants of oceanic islands — Absence of Batra- ogy, laws of, explained by variations not superven- chians and of terrestrial Mammals — On the rela- ing at an early age, and being inherited at a corre- tion of the inhabitants of islands to those of the near- sponding age — Rudimentary Organs; their origin est mainland —On colonisation from the nearest explained — Summary.

8 Charles Darwin Chapter XIV On the Origin of Species.

Recapitulation and Conclusion Introduction. Recapitulation of the difficulties on the theory of Natural Selection —Recapitulation of the general and HEN ON BOARD H.M.S. ‘BEAGLE,’ as natu special circumstances in its favour —Causes of the ralist, I was much struck with certain general belief in the immutability of species — How Wfacts in the distribution of the inhabit- far the theory of natural selection may be extended ants of South America, and in the geological rela- — Effects of its adoption on the study of Natural his- tions of the present to the past inhabitants of that tory — Concluding remarks. continent. These facts seemed to me to throw some light on the origin of species—that mystery of mys- teries, as it has been called by one of our greatest philosophers. On my return home, it occurred to me, in 1837, that something might perhaps be made out on this question by patiently accumulating and reflecting on all sorts of facts which could possibly have any bearing on it. After five years’ work I allowed myself to speculate on the subject, and drew up some short notes; these I enlarged in 1844 into a sketch of the conclusions, which then seemed to me probable: from that period to the present day I have steadily 9 On the Origin of Species pursued the same object. I hope that I may be ex- This Abstract, which I now publish, must neces- cused for entering on these personal details, as I give sarily be imperfect. I cannot here give references and them to show that I have not been hasty in coming to authorities for my several statements; and I must a decision. trust to the reader reposing some confidence in my My work is now nearly finished; but as it will take accuracy. No doubt errors will have crept in, though me two or three more years to complete it, and as I hope I have always been cautious in trusting to my health is far from strong, I have been urged to good authorities alone. I can here give only the gen- publish this Abstract. I have more especially been eral conclusions at which I have arrived, with a few induced to do this, as Mr. Wallace, who is now study- facts in illustration, but which, I hope, in most cases ing the natural history of the Malay archipelago, has will suffice. No one can feel more sensible than I do arrived at almost exactly the same general conclu- of the necessity of hereafter publishing in detail all sions that I have on the origin of species. Last year the facts, with references, on which my conclusions he sent to me a memoir on this subject, with a re- have been grounded; and I hope in a future work to quest that I would forward it to Sir Charles Lyell, do this. For I am well aware that scarcely a single who sent it to the Linnean Society, and it is published point is discussed in this volume on which facts can- in the third volume of the Journal of that Society. Sir not be adduced, often apparently leading to conclu- C. Lyell and Dr. Hooker, who both knew of my sions directly opposite to those at which I have ar- work—the latter having read my sketch of 1844— rived. A fair result can be obtained only by fully stat- honoured me by thinking it advisable to publish, ing and balancing the facts and arguments on both with Mr. Wallace’s excellent memoir, some brief ex- sides of each question; and this cannot possibly be tracts from my manuscripts. here done.

10 Charles Darwin I much regret that want of space prevents my hav- tion which most justly excites our admiration. Natu- ing the satisfaction of acknowledging the generous ralists continually refer to external conditions, such assistance which I have received from very many as climate, food, &c., as the only possible cause of naturalists, some of them personally unknown to me. variation. In one very limited sense, as we shall here- I cannot, however, let this opportunity pass without after see, this may be true; but it is preposterous to expressing my deep obligations to Dr. Hooker, who attribute to mere external conditions, the structure, for the last fifteen years has aided me in every pos- for instance, of the woodpecker, with its feet, tail, sible way by his large stores of knowledge and his beak, and tongue, so admirably adapted to catch in- excellent judgment. sects under the bark of trees. In the case of the In considering the Origin of Species, it is quite con- misseltoe, which draws its nourishment from certain ceivable that a naturalist, reflecting on the mutual trees, which has seeds that must be transported by affinities of organic beings, on their embryological certain birds, and which has flowers with separate relations, their geographical distribution, geologi- sexes absolutely requiring the agency of certain in- cal succession, and other such facts, might come to sects to bring pollen from one flower to the other, it the conclusion that each species had not been inde- is equally preposterous to account for the structure pendently created, but had descended, like variet- of this parasite, with its relations to several distinct ies, from other species. Nevertheless, such a conclu- organic beings, by the effects of external conditions, sion, even if well founded, would be unsatisfactory, or of habit, or of the volition of the plant itself. until it could be shown how the innumerable spe- The author of the ‘Vestiges of Creation’ would, I cies inhabiting this world have been modified, so as presume, say that, after a certain unknown number to acquire that perfection of structure and coadapta- of generations, some bird had given birth to a wood-

11 On the Origin of Species pecker, and some plant to the misseltoe, and that cation. We shall thus see that a large amount of he- these had been produced perfect as we now see them; reditary modification is at least possible, and, what but this assumption seems to me to be no explana- is equally or more important, we shall see how great tion, for it leaves the case of the coadaptations of or- is the power of man in accumulating by his Selec- ganic beings to each other and to their physical con- tion successive slight variations. I will then pass on ditions of life, untouched and unexplained. to the variability of species in a state of nature; but I It is, therefore, of the highest importance to gain a shall, unfortunately, be compelled to treat this sub- clear insight into the means of modification and co- ject far too briefly, as it can be treated properly only adaptation. At the commencement of my observa- by giving long catalogues of facts. We shall, how- tions it seemed to me probable that a careful study ever, be enabled to discuss what circumstances are of domesticated animals and of cultivated plants most favourable to variation. In the next chapter the would offer the best chance of making out this ob- Struggle for Existence amongst all organic beings scure problem. Nor have I been disappointed; in this throughout the world, which inevitably follows from and in all other perplexing cases I have invariably their high geometrical powers of increase, will be found that our knowledge, imperfect though it be, treated of. This is the doctrine of Malthus, applied of variation under domestication, afforded the best to the whole animal and vegetable kingdoms. As and safest clue. I may venture to express my convic- many more individuals of each species are born than tion of the high value of such studies, although they can possibly survive; and as, consequently, there is have been very commonly neglected by naturalists. a frequently recurring struggle for existence, it fol- From these considerations, I shall devote the first lows that any being, if it vary however slightly in chapter of this Abstract to Variation under Domesti- any manner profitable to itself, under the complex

12 Charles Darwin and sometimes varying conditions of life, will have of varieties when intercrossed; and fourthly, the im- a better chance of surviving, and thus be naturally perfection of the Geological Record. In the next chap- selected. From the strong principle of inheritance, ter I shall consider the geological succession of or- any selected variety will tend to propagate its new ganic beings throughout time; in the eleventh and and modified form. twelfth, their geographical distribution throughout This fundamental subject of Natural Selection will space; in the thirteenth, their classification or mu- be treated at some length in the fourth chapter; and tual affinities, both when mature and in an embry- we shall then see how Natural Selection almost in- onic condition. In the last chapter I shall give a brief evitably causes much Extinction of the less improved recapitulation of the whole work, and a few conclud- forms of life and induces what I have called Diver- ing remarks. gence of Character. In the next chapter I shall dis- No one ought to feel surprise at much remaining cuss the complex and little known laws of variation as yet unexplained in regard to the origin of species and of correlation of growth. In the four succeeding and varieties, if he makes due allowance for our pro- chapters, the most apparent and gravest difficulties found ignorance in regard to the mutual relations of on the theory will be given: namely, first, the diffi- all the beings which live around us. Who can explain culties of transitions, or in understanding how a why one species ranges widely and is very numer- simple being or a simple organ can be changed and ous, and why another allied species has a narrow perfected into a highly developed being or elabo- range and is rare? Yet these relations are of the high- rately constructed organ; secondly the subject of In- est importance, for they determine the present wel- stinct, or the mental powers of animals, thirdly, Hy- fare, and, as I believe, the future success and modifi- bridism, or the infertility of species and the fertility cation of every inhabitant of this world. Still less do

13 On the Origin of Species we know of the mutual relations of the innumerable Chapter I inhabitants of the world during the many past geological epochs in its history. Although much remains Variation under Domestication obscure, and will long remain obscure, I can entertain no doubt, after the most deliberate study and Causes of Variability — Effects of Habit — Correla- dispassionate judgment of which I am capable, that tion of Growth —Inheritance — Character of Domes- the view which most naturalists entertain, and which tic Varieties — Difficulty of distinguishing between I formerly entertained—namely, that each species has Varieties and Species — Origin of Domestic Variet- been independently created—is erroneous. I am fully ies from one or more Species — Domestic Pigeons, convinced that species are not immutable; but that their Differences and Origin — Principle of Selec- those belonging to what are called the same genera tion anciently followed, its Effects —Methodical and are lineal descendants of some other and generally Unconscious Selection — Unknown Origin of our extinct species, in the same manner as the acknowl- Domestic Productions — Circumstances favourable edged varieties of any one species are the descen- to Man’s power of Selection. dants of that species. Furthermore, I am convinced that Natural Selection has been the main but not ex- WHEN WE LOOK to the individuals of the same variety clusive means of modification. or sub-variety of our older cultivated plants and animals, one of the first points which strikes us, is, that they generally differ much more from each other, than do the individuals of any one species or variety in a state of nature. When we reflect on the vast

14 Charles Darwin diversity of the plants and animals which have been It has been disputed at what period of life the cultivated, and which have varied during all ages causes of variability, whatever they may be, gener- under the most different climates and treatment, I ally act; whether during the early or late period of think we are driven to conclude that this greater vari- development of the embryo, or at the instant of con- ability is simply due to our domestic productions ception. Geoffroy St. Hilaire’s experiments show that having been raised under conditions of life not so unnatural treatment of the embryo causes monstrosi- uniform as, and somewhat different from, those to ties; and monstrosities cannot be separated by any which the parent-species have been exposed under clear line of distinction from mere variations. But I nature. There is, also, I think, some probability in am strongly inclined to suspect that the most fre- the view propounded by Andrew Knight, that this quent cause of variability may be attributed to the variability may be partly connected with excess of male and female reproductive elements having been food. It seems pretty clear that organic beings must affected prior to the act of conception. Several rea- be exposed during several generations to the new sons make me believe in this; but the chief one is the conditions of life to cause any appreciable amount remarkable effect which confinement or cultivation of variation; and that when the organisation has once has on the functions of the reproductive system; this begun to vary, it generally continues to vary for many system appearing to be far more susceptible than generations. No case is on record of a variable being any other part of the organisation, to the action of ceasing to be variable under cultivation. Our oldest any change in the conditions of life. Nothing is more cultivated plants, such as wheat, still often yield new easy than to tame an animal, and few things more varieties: our oldest domesticated animals are still difficult than to get it to breed freely under confine- capable of rapid improvement or modification. ment, even in the many cases when the male and

15 On the Origin of Species female unite. How many animals there are which will mals and plants, though often weak and sickly, yet not breed, though living long under not very close breeding quite freely under confinement; and when, confinement in their native country! This is gener- on the other hand, we see individuals, though taken ally attributed to vitiated instincts; but how many young from a state of nature, perfectly tamed, long- cultivated plants display the utmost vigour, and yet lived, and healthy (of which I could give numerous rarely or never seed! In some few such cases it has instances), yet having their reproductive system so been found out that very trifling changes, such as a seriously affected by unperceived causes as to fail little more or less water at some particular period of in acting, we need not be surprised at this system, growth, will determine whether or not the plant sets when it does act under confinement, acting not quite a seed. I cannot here enter on the copious details regularly, and producing offspring not perfectly like which I have collected on this curious subject; but to their parents or variable. show how singular the laws are which determine the Sterility has been said to be the bane of horticul- reproduction of animals under confinement, I may ture; but on this view we owe variability to the same just mention that carnivorous animals, even from the cause which produces sterility; and variability is the tropics, breed in this country pretty freely under con- source of all the choicest productions of the garden. finement, with the exception of the plantigrades or I may add, that as some organisms will breed most bear family; whereas, carnivorous birds, with the freely under the most unnatural conditions (for in- rarest exceptions, hardly ever lay fertile eggs. Many stance, the rabbit and ferret kept in hutches), show- exotic plants have pollen utterly worthless, in the ing that their reproductive system has not been thus same exact condition as in the most sterile hybrids. affected; so will some animals and plants withstand When, on the one hand, we see domesticated ani- domestication or cultivation, and vary very

16 Charles Darwin slightly—perhaps hardly more than in a state of na- Seedlings from the same fruit, and the young of ture. the same litter, sometimes differ considerably from A long list could easily be given of ‘sporting plants;’ each other, though both the young and the parents, by this term gardeners mean a single bud or offset, as Muller has remarked, have apparently been ex- which suddenly assumes a new and sometimes very posed to exactly the same conditions of life; and this different character from that of the rest of the plant. shows how unimportant the direct effects of the con- Such buds can be propagated by grafting, &c., and ditions of life are in comparison with the laws of re- sometimes by seed. These ‘sports’ are extremely rare production, and of growth, and of inheritance; for under nature, but far from rare under cultivation; had the action of the conditions been direct, if any of and in this case we see that the treatment of the par- the young had varied, all would probably have var- ent has affected a bud or offset, and not the ovules ied in the same manner. To judge how much, in the or pollen. But it is the opinion of most physiologists case of any variation, we should attribute to the di- that there is no essential difference between a bud rect action of heat, moisture, light, food, &c., is most and an ovule in their earliest stages of formation; so difficult: my impression is, that with animals such that, in fact, ‘sports’ support my view, that variabil- agencies have produced very little direct effect, ity may be largely attributed to the ovules or pollen, though apparently more in the case of plants. Under or to both, having been affected by the treatment of this point of view, Mr. Buckman’s recent experiments the parent prior to the act of conception. These cases on plants seem extremely valuable. When all or anyhow show that variation is not necessarily con- nearly all the individuals exposed to certain condi- nected, as some authors have supposed, with the act tions are affected in the same way, the change at first of generation. appears to be directly due to such conditions; but in

17 On the Origin of Species some cases it can be shown that quite opposite condi- a single domestic animal can be named which has not tions produce similar changes of structure. Neverthe- in some country drooping ears; and the view sug- less some slight amount of change may, I think, be gested by some authors, that the drooping is due to attributed to the direct action of the conditions of life— the disuse of the muscles of the ear, from the animals as, in some cases, increased size from amount of food, not being much alarmed by danger, seems probable. colour from particular kinds of food and from light, There are many laws regulating variation, some and perhaps the thickness of fur from climate. few of which can be dimly seen, and will be hereaf- Habit also has a deciding influence, as in the pe- ter briefly mentioned. I will here only allude to what riod of flowering with plants when transported from may be called correlation of growth. Any change in one climate to another. In animals it has a more the embryo or larva will almost certainly entail marked effect; for instance, I find in the domestic changes in the mature animal. In monstrosities, the duck that the bones of the wing weigh less and the correlations between quite distinct parts are very bones of the leg more, in proportion to the whole curious; and many instances are given in Isidore skeleton, than do the same bones in the wild-duck; Geoffroy St. Hilaire’s great work on this subject. and I presume that this change may be safely attrib- Breeders believe that long limbs are almost always uted to the domestic duck flying much less, and accompanied by an elongated head. Some instances walking more, than its wild parent. The great and of correlation are quite whimsical; thus cats with blue inherited development of the udders in cows and eyes are invariably deaf; colour and constitutional goats in countries where they are habitually milked, peculiarities go together, of which many remarkable in comparison with the state of these organs in other cases could be given amongst animals and plants. countries, is another instance of the effect of use. Not From the facts collected by Heusinger, it appears that

18 Charles Darwin white sheep and pigs are differently affected from become plastic, and tends to depart in some small coloured individuals by certain vegetable poisons. degree from that of the parental type. Hairless dogs have imperfect teeth; long-haired and Any variation which is not inherited is unimpor- coarse-haired animals are apt to have, as is asserted, tant for us. But the number and diversity of inherit- long or many horns; pigeons with feathered feet able deviations of structure, both those of slight and have skin between their outer toes; pigeons with those of considerable physiological importance, is short beaks have small feet, and those with long endless. Dr. Prosper Lucas’s treatise, in two large beaks large feet. Hence, if man goes on selecting, volumes, is the fullest and the best on this subject. and thus augmenting, any peculiarity, he will al- No breeder doubts how strong is the tendency to most certainly unconsciously modify other parts of inheritance: like produces like is his fundamental the structure, owing to the mysterious laws of the belief: doubts have been thrown on this principle by correlation of growth. theoretical writers alone. When a deviation appears The result of the various, quite unknown, or dimly not unfrequently, and we see it in the father and child, seen laws of variation is infinitely complex and di- we cannot tell whether it may not be due to the same versified. It is well worth while carefully to study original cause acting on both; but when amongst in- the several treatises published on some of our old dividuals, apparently exposed to the same condi- cultivated plants, as on the hyacinth, potato, even tions, any very rare deviation, due to some extraor- the dahlia, &c.; and it is really surprising to note the dinary combination of circumstances, appears in the endless points in structure and constitution in which parent—say, once amongst several million individu- the varieties and sub-varieties differ slightly from als—and it reappears in the child, the mere doctrine each other. The whole organisation seems to have of chances almost compels us to attribute its reap-

19 On the Origin of Species pearance to inheritance. Every one must have heard sively, or in a much greater degree, to males alone. of cases of albinism, prickly skin, hairy bodies, &c., A much more important rule, which I think may be appearing in several members of the same family. If trusted, is that, at whatever period of life a peculiar- strange and rare deviations of structure are truly in- ity first appears, it tends to appear in the offspring herited, less strange and commoner deviations may at a corresponding age, though sometimes earlier. be freely admitted to be inheritable. Perhaps the cor- In many cases this could not be otherwise: thus the rect way of viewing the whole subject, would be, to inherited peculiarities in the horns of cattle could look at the inheritance of every character whatever appear only in the offspring when nearly mature; as the rule, and non-inheritance as the anomaly. peculiarities in the silkworm are known to appear The laws governing inheritance are quite unknown; at the corresponding caterpillar or cocoon stage. But no one can say why the same peculiarity in different hereditary diseases and some other facts make me individuals of the same species, and in individuals believe that the rule has a wider extension, and that of different species, is sometimes inherited and when there is no apparent reason why a peculiarity sometimes not so; why the child often reverts in cer- should appear at any particular age, yet that it does tain characters to its grandfather or grandmother or tend to appear in the offspring at the same period at other much more remote ancestor; why a peculiar- which it first appeared in the parent. I believe this ity is often transmitted from one sex to both sexes or rule to be of the highest importance in explaining to one sex alone, more commonly but not exclusively the laws of embryology. These remarks are of course to the like sex. It is a fact of some little importance to confined to the first appearance of the peculiarity, us, that peculiarities appearing in the males of our and not to its primary cause, which may have acted domestic breeds are often transmitted either exclu- on the ovules or male element; in nearly the same

20 Charles Darwin manner as in the crossed offspring from a short- variety should be turned loose in its new home. horned cow by a long-horned bull, the greater length Nevertheless, as our varieties certainly do occasion- of horn, though appearing late in life, is clearly due ally revert in some of their characters to ancestral to the male element. forms, it seems to me not improbable, that if we could Having alluded to the subject of reversion, I may succeed in naturalising, or were to cultivate, during here refer to a statement often made by naturalists— many generations, the several races, for instance, of namely, that our domestic varieties, when run wild, the cabbage, in very poor soil (in which case, how- gradually but certainly revert in character to their ever, some effect would have to be attributed to the aboriginal stocks. Hence it has been argued that no direct action of the poor soil), that they would to a deductions can be drawn from domestic races to large extent, or even wholly, revert to the wild ab- species in a state of nature. I have in vain endeav- original stock. Whether or not the experiment would oured to discover on what decisive facts the above succeed, is not of great importance for our line of statement has so often and so boldly been made. argument; for by the experiment itself the conditions There would be great difficulty in proving its truth: of life are changed. If it could be shown that our do- we may safely conclude that very many of the most mestic varieties manifested a strong tendency to re- strongly-marked domestic varieties could not pos- version,—that is, to lose their acquired characters, sibly live in a wild state. In many cases we do not whilst kept under unchanged conditions, and whilst know what the aboriginal stock was, and so could kept in a considerable body, so that free intercross- not tell whether or not nearly perfect reversion had ing might check, by blending together, any slight ensued. It would be quite necessary, in order to pre- deviations of structure, in such case, I grant that we vent the effects of intercrossing, that only a single could deduce nothing from domestic varieties in

21 On the Origin of Species regard to species. But there is not a shadow of evi- gree in some one part, both when compared one with dence in favour of this view: to assert that we could another, and more especially when compared with not breed our cart and race-horses, long and short- all the species in nature to which they are nearest horned cattle, and poultry of various breeds, and allied. With these exceptions (and with that of the esculent vegetables, for an almost infinite number perfect fertility of varieties when crossed,—a sub- of generations, would be opposed to all experience. ject hereafter to be discussed), domestic races of the I may add, that when under nature the conditions of same species differ from each other in the same man- life do change, variations and reversions of charac- ner as, only in most cases in a lesser degree than, do ter probably do occur; but natural selection, as will closely-allied species of the same genus in a state of hereafter be explained, will determine how far the nature. I think this must be admitted, when we find new characters thus arising shall be preserved. that there are hardly any domestic races, either When we look to the hereditary varieties or races amongst animals or plants, which have not been of our domestic animals and plants, and compare ranked by some competent judges as mere variet- them with species closely allied together, we gener- ies, and by other competent judges as the descen- ally perceive in each domestic race, as already re- dants of aboriginally distinct species. If any marked marked, less uniformity of character than in true spe- distinction existed between domestic races and species. Domestic races of the same species, also, often cies, this source of doubt could not so perpetually have a somewhat monstrous character; by which I recur. It has often been stated that domestic races do mean, that, although differing from each other, and not differ from each other in characters of generic from the other species of the same genus, in several value. I think it could be shown that this statement trifling respects, they often differ in an extreme de- is hardly correct; but naturalists differ most widely

22 Charles Darwin in determining what characters are of generic value; any one wild species; but, in the case of some other all such valuations being at present empirical. More- domestic races, there is presumptive, or even strong, over, on the view of the origin of genera which I shall evidence in favour of this view. presently give, we have no right to expect often to It has often been assumed that man has chosen for meet with generic differences in our domesticated domestication animals and plants having an extraor- productions. dinary inherent tendency to vary, and likewise to When we attempt to estimate the amount of struc- withstand diverse climates. I do not dispute that tural difference between the domestic races of the these capacities have added largely to the value of same species, we are soon involved in doubt, from most of our domesticated productions; but how not knowing whether they have descended from one could a savage possibly know, when he first tamed or several parent-species. This point, if it could be an animal, whether it would vary in succeeding gen- cleared up, would be interesting; if, for instance, it erations, and whether it would endure other cli- could be shown that the greyhound, bloodhound, mates? Has the little variability of the ass or guinea- terrier, spaniel, and bull-dog, which we all know fowl, or the small power of endurance of warmth by propagate their kind so truly, were the offspring of the rein-deer, or of cold by the common camel, pre- any single species, then such facts would have great vented their domestication? I cannot doubt that if weight in making us doubt about the immutability other animals and plants, equal in number to our of the many very closely allied and natural species— domesticated productions, and belonging to equally for instance, of the many foxes—inhabiting different diverse classes and countries, were taken from a state quarters of the world. I do not believe, as we shall of nature, and could be made to breed for an equal presently see, that all our dogs have descended from number of generations under domestication, they

23 On the Origin of Species would vary on an average as largely as the parent pretend to say how long before these ancient peri- species of our existing domesticated productions ods, savages, like those of Tierra del Fuego or Aus- have varied. tralia, who possess a semi-domestic dog, may not In the case of most of our anciently domesticated have existed in Egypt? animals and plants, I do not think it is possible to The whole subject must, I think, remain vague; nev- come to any definite conclusion, whether they have ertheless, I may, without here entering on any de- descended from one or several species. The argu- tails, state that, from geographical and other consid- ment mainly relied on by those who believe in the erations, I think it highly probable that our domestic multiple origin of our domestic animals is, that we dogs have descended from several wild species. In find in the most ancient records, more especially on regard to sheep and goats I can form no opinion. I the monuments of Egypt, much diversity in the should think, from facts communicated to me by Mr. breeds; and that some of the breeds closely resemble, Blyth, on the habits, voice, and constitution, &c., of perhaps are identical with, those still existing. Even the humped Indian cattle, that these had descended if this latter fact were found more strictly and gener- from a different aboriginal stock from our European ally true than seems to me to be the case, what does cattle; and several competent judges believe that it show, but that some of our breeds originated there, these latter have had more than one wild parent. With four or five thousand years ago? But Mr. Horner’s respect to horses, from reasons which I cannot give researches have rendered it in some degree prob- here, I am doubtfully inclined to believe, in opposi- able that man sufficiently civilized to have manu- tion to several authors, that all the races have de- factured pottery existed in the valley of the Nile thir- scended from one wild stock. Mr. Blyth, whose opin- teen or fourteen thousand years ago; and who will ion, from his large and varied stores of knowledge, I

24 Charles Darwin should value more than that of almost any one, thinks but that each of these kingdoms possesses several that all the breeds of poultry have proceeded from peculiar breeds of cattle, sheep, &c., we must admit the common wild Indian fowl (Gallus bankiva). In that many domestic breeds have originated in Eu- regard to ducks and rabbits, the breeds of which dif- rope; for whence could they have been derived, as fer considerably from each other in structure, I do these several countries do not possess a number of not doubt that they all have descended from the com- peculiar species as distinct parent-stocks? So it is in mon wild duck and rabbit. India. Even in the case of the domestic dogs of the The doctrine of the origin of our several domestic whole world, which I fully admit have probably de- races from several aboriginal stocks, has been car- scended from several wild species, I cannot doubt ried to an absurd extreme by some authors. They that there has been an immense amount of inherited believe that every race which breeds true, let the dis- variation. Who can believe that animals closely re- tinctive characters be ever so slight, has had its wild sembling the Italian greyhound, the bloodhound, the prototype. At this rate there must have existed at least bull-dog, or Blenheim spaniel, &c.—so unlike all a score of species of wild cattle, as many sheep, and wild Canidae—ever existed freely in a state of na- several goats in Europe alone, and several even ture? It has often been loosely said that all our races within Great Britain. One author believes that there of dogs have been produced by the crossing of a few formerly existed in Great Britain eleven wild spe- aboriginal species; but by crossing we can get only cies of sheep peculiar to it! When we bear in mind forms in some degree intermediate between their that Britain has now hardly one peculiar mammal, parents; and if we account for our several domestic and France but few distinct from those of Germany races by this process, we must admit the former ex- and conversely, and so with Hungary, Spain, &c., istence of the most extreme forms, as the Italian grey-

25 On the Origin of Species hound, bloodhound, bull-dog, &c., in the wild state. permanent race having been thus formed. Moreover, the possibility of making distinct races On the Breeds of the Domestic Pigeon. — Believ- by crossing has been greatly exaggerated. There can ing that it is always best to study some special group, be no doubt that a race may be modified by occa- I have, after deliberation, taken up domestic pigeons. sional crosses, if aided by the careful selection of I have kept every breed which I could purchase or those individual mongrels, which present any de- obtain, and have been most kindly favoured with sired character; but that a race could be obtained skins from several quarters of the world, more espe- nearly intermediate between two extremely differ- cially by the Hon. W. Elliot from India, and by the ent races or species, I can hardly believe. Sir J. Hon. C. Murray from Persia. Many treatises in dif- Sebright expressly experimentised for this object, ferent languages have been published on pigeons, and failed. The offspring from the first cross between and some of them are very important, as being of two pure breeds is tolerably and sometimes (as I considerably antiquity. I have associated with sev- have found with pigeons) extremely uniform, and eral eminent fanciers, and have been permitted to everything seems simple enough; but when these join two of the London Pigeon Clubs. The diversity mongrels are crossed one with another for several of the breeds is something astonishing. Compare the generations, hardly two of them will be alike, and English carrier and the short-faced tumbler, and see then the extreme difficulty, or rather utter hopeless- the wonderful difference in their beaks, entailing ness, of the task becomes apparent. Certainly, a breed corresponding differences in their skulls. The car- intermediate between two very distinct breeds could rier, more especially the male bird, is also remark- not be got without extreme care and long-continued able from the wonderful development of the selection; nor can I find a single case on record of a carunculated skin about the head, and this is accom-

26 Charles Darwin panied by greatly elongated eyelids, very large ex- hood, and it has, proportionally to its size, much ternal orifices to the nostrils, and a wide gape of elongated wing and tail feathers. The trumpeter and mouth. The short-faced tumbler has a beak in out- laugher, as their names express, utter a very differ- line almost like that of a finch; and the common tum- ent coo from the other breeds. The fantail has thirty bler has the singular and strictly inherited habit of or even forty tail-feathers, instead of twelve or four- flying at a great height in a compact flock, and tum- teen, the normal number in all members of the great bling in the air head over heels. The runt is a bird of pigeon family; and these feathers are kept expanded, great size, with long, massive beak and large feet; and are carried so erect that in good birds the head some of the sub-breeds of runts have very long necks, and tail touch; the oil-gland is quite aborted. Several others very long wings and tails, others singularly other less distinct breeds might have been specified. short tails. The barb is allied to the carrier, but, in- In the skeletons of the several breeds, the develop- stead of a very long beak, has a very short and very ment of the bones of the face in length and breadth broad one. The pouter has a much elongated body, and curvature differs enormously. The shape, as well wings, and legs; and its enormously developed crop, as the breadth and length of the ramus of the lower which it glories in inflating, may well excite aston- jaw, varies in a highly remarkable manner. The num- ishment and even laughter. The turbit has a very ber of the caudal and sacral vertebrae vary; as does short and conical beak, with a line of reversed feath- the number of the ribs, together with their relative ers down the breast; and it has the habit of continu- breadth and the presence of processes. The size and ally expanding slightly the upper part of the oe- shape of the apertures in the sternum are highly vari- sophagus. The Jacobin has the feathers so much re- able; so is the degree of divergence and relative size versed along the back of the neck that they form a of the two arms of the furcula. The proportional

27 On the Origin of Species width of the gape of mouth, the proportional length told that they were wild birds, would certainly, I of the eyelids, of the orifice of the nostrils, of the think, be ranked by him as well-defined species. tongue (not always in strict correlation with the Moreover, I do not believe that any ornithologist length of beak), the size of the crop and of the upper would place the English carrier, the short-faced tum- part of the oesophagus; the development and abor- bler, the runt, the barb, pouter, and fantail in the same tion of the oil-gland; the number of the primary wing genus; more especially as in each of these breeds and caudal feathers; the relative length of wing and several truly-inherited sub-breeds, or species as he tail to each other and to the body; the relative length might have called them, could be shown him. of leg and of the feet; the number of scutellae on the Great as the differences are between the breeds of toes, the development of skin between the toes, are pigeons, I am fully convinced that the common opin- all points of structure which are variable. The pe- ion of naturalists is correct, namely, that all have riod at which the perfect plumage is acquired var- descended from the rock-pigeon (Columba livia), ies, as does the state of the down with which the nest- including under this term several geographical races ling birds are clothed when hatched. The shape and or sub-species, which differ from each other in the size of the eggs vary. The manner of flight differs most trifling respects. As several of the reasons which remarkably; as does in some breeds the voice and have led me to this belief are in some degree appli- disposition. Lastly, in certain breeds, the males and cable in other cases, I will here briefly give them. If females have come to differ to a slight degree from the several breeds are not varieties, and have not each other. proceeded from the rock-pigeon, they must have Altogether at least a score of pigeons might be cho- descended from at least seven or eight aboriginal sen, which if shown to an ornithologist, and he were stocks; for it is impossible to make the present do-

28 Charles Darwin mestic breeds by the crossing of any lesser number: Hence the supposed extermination of so many spe- how, for instance, could a pouter be produced by cies having similar habits with the rock-pigeon seems crossing two breeds unless one of the parent-stocks to me a very rash assumption. Moreover, the sev- possessed the characteristic enormous crop? The eral above-named domesticated breeds have been supposed aboriginal stocks must all have been rock- transported to all parts of the world, and, therefore, pigeons, that is, not breeding or willingly perching some of them must have been carried back again into on trees. But besides C. livia, with its geographical their native country; but not one has ever become sub-species, only two or three other species of rock- wild or feral, though the dovecot-pigeon, which is pigeons are known; and these have not any of the the rock-pigeon in a very slightly altered state, has characters of the domestic breeds. Hence the sup- become feral in several places. Again, all recent exposed aboriginal stocks must either still exist in the perience shows that it is most difficult to get any wild countries where they were originally domesticated, animal to breed freely under domestication; yet on and yet be unknown to ornithologists; and this, con- the hypothesis of the multiple origin of our pigeons, sidering their size, habits, and remarkable charac- it must be assumed that at least seven or eight spe- ters, seems very improbable; or they must have be- cies were so thoroughly domesticated in ancient come extinct in the wild state. But birds breeding on times by half-civilized man, as to be quite prolific precipices, and good fliers, are unlikely to be exter- under confinement. minated; and the common rock-pigeon, which has An argument, as it seems to me, of great weight, the same habits with the domestic breeds, has not and applicable in several other cases, is, that the been exterminated even on several of the smaller above-specified breeds, though agreeing generally British islets, or on the shores of the Mediterranean. in constitution, habits, voice, colouring, and in most

29 On the Origin of Species parts of their structure, with the wild rock-pigeon, the wings have two black bars; some semi-domestic yet are certainly highly abnormal in other parts of breeds and some apparently truly wild breeds have, their structure: we may look in vain throughout the besides the two black bars, the wings chequered with whole great family of Columbidae for a beak like black. These several marks do not occur together in that of the English carrier, or that of the short-faced any other species of the whole family. Now, in ev- tumbler, or barb; for reversed feathers like those of ery one of the domestic breeds, taking thoroughly the jacobin; for a crop like that of the pouter; for tail- well-bred birds, all the above marks, even to the feathers like those of the fantail. Hence it must be white edging of the outer tail-feathers, sometimes assumed not only that half-civilized man succeeded concur perfectly developed. Moreover, when two in thoroughly domesticating several species, but that birds belonging to two distinct breeds are crossed, he intentionally or by chance picked out extraordi- neither of which is blue or has any of the above-speci- narily abnormal species; and further, that these very fied marks, the mongrel offspring are very apt sud- species have since all become extinct or unknown. denly to acquire these characters; for instance, I So many strange contingencies seem to me improb- crossed some uniformly white fantails with some able in the highest degree. uniformly black barbs, and they produced mottled Some facts in regard to the colouring of pigeons brown and black birds; these I again crossed together, well deserve consideration. The rock-pigeon is of a and one grandchild of the pure white fantail and pure slaty-blue, and has a white rump (the Indian sub- black barb was of as beautiful a blue colour, with species, C. intermedia of Strickland, having it blu- the white rump, double black wing-bar, and barred ish); the tail has a terminal dark bar, with the bases and white-edged tail-feathers, as any wild rock-pi- of the outer feathers externally edged with white; geon! We can understand these facts, on the well-

30 Charles Darwin known principle of reversion to ancestral characters, been no cross with a distinct breed, and there is a if all the domestic breeds have descended from the tendency in both parents to revert to a character, rock-pigeon. But if we deny this, we must make one which has been lost during some former generation, of the two following highly improbable supposi- this tendency, for all that we can see to the contrary, tions. Either, firstly, that all the several imagined may be transmitted undiminished for an indefinite aboriginal stocks were coloured and marked like the number of generations. These two distinct cases are rock-pigeon, although no other existing species is often confounded in treatises on inheritance. thus coloured and marked, so that in each separate Lastly, the hybrids or mongrels from between all breed there might be a tendency to revert to the very the domestic breeds of pigeons are perfectly fertile. same colours and markings. Or, secondly, that each I can state this from my own observations, purposely breed, even the purest, has within a dozen or, at most, made on the most distinct breeds. Now, it is diffi- within a score of generations, been crossed by the cult, perhaps impossible, to bring forward one case rock-pigeon: I say within a dozen or twenty genera- of the hybrid offspring of two animals clearly dis- tions, for we know of no fact countenancing the be- tinct being themselves perfectly fertile. Some authors lief that the child ever reverts to some one ancestor, believe that long-continued domestication elimi- removed by a greater number of generations. In a nates this strong tendency to sterility: from the his- breed which has been crossed only once with some tory of the dog I think there is some probability in distinct breed, the tendency to reversion to any char- this hypothesis, if applied to species closely related acter derived from such cross will naturally become together, though it is unsupported by a single ex- less and less, as in each succeeding generation there periment. But to extend the hypothesis so far as to will be less of the foreign blood; but when there has suppose that species, aboriginally as distinct as car-

31 On the Origin of Species riers, tumblers, pouters, and fantails now are, should agrees in habits and in a great number of points of yield offspring perfectly fertile, inter se, seems to me structure with all the domestic breeds. Secondly, al- rash in the extreme. though an English carrier or short-faced tumbler dif- From these several reasons, namely, the improb- fers immensely in certain characters from the rock- ability of man having formerly got seven or eight pigeon, yet by comparing the several sub-breeds of supposed species of pigeons to breed freely under these breeds, more especially those brought from domestication; these supposed species being quite distant countries, we can make an almost perfect se- unknown in a wild state, and their becoming no- ries between the extremes of structure. Thirdly, those where feral; these species having very abnormal char- characters which are mainly distinctive of each breed, acters in certain respects, as compared with all other for instance the wattle and length of beak of the car- Columbidae, though so like in most other respects rier, the shortness of that of the tumbler, and the num- to the rock-pigeon; the blue colour and various ber of tail-feathers in the fantail, are in each breed marks occasionally appearing in all the breeds, both eminently variable; and the explanation of this fact when kept pure and when crossed; the mongrel off- will be obvious when we come to treat of selection. spring being perfectly fertile;—from these several Fourthly, pigeons have been watched, and tended reasons, taken together, I can feel no doubt that all with the utmost care, and loved by many people. our domestic breeds have descended from the They have been domesticated for thousands of years Columba livia with its geographical sub-species. in several quarters of the world; the earliest known In favour of this view, I may add, firstly, that C. record of pigeons is in the fifth Aegyptian dynasty, livia, or the rock-pigeon, has been found capable of about 3000 B.C., as was pointed out to me by Profes- domestication in Europe and in India; and that it sor Lepsius; but Mr. Birch informs me that pigeons

32 Charles Darwin are given in a bill of fare in the previous dynasty. In different breeds can be kept together in the same the time of the Romans, as we hear from Pliny, im- aviary. mense prices were given for pigeons; ‘nay, they are I have discussed the probable origin of domestic come to this pass, that they can reckon up their pedi- pigeons at some, yet quite insufficient, length; be- gree and race.’ Pigeons were much valued by Akber cause when I first kept pigeons and watched the sev- Khan in India, about the year 1600; never less than eral kinds, knowing well how true they bred, I felt 20,000 pigeons were taken with the court. ‘The mon- fully as much difficulty in believing that they could archs of Iran and Turan sent him some very rare ever have descended from a common parent, as any birds;’ and, continues the courtly historian, ‘His naturalist could in coming to a similar conclusion in Majesty by crossing the breeds, which method was regard to the many species of finches, or other large never practised before, has improved them aston- groups of birds, in nature. One circumstance has ishingly.’ About this same period the Dutch were as struck me much; namely, that all the breeders of the eager about pigeons as were the old Romans. The various domestic animals and the cultivators of paramount importance of these considerations in plants, with whom I have ever conversed, or whose explaining the immense amount of variation which treatises I have read, are firmly convinced that the pigeons have undergone, will be obvious when we several breeds to which each has attended, are de- treat of Selection. We shall then, also, see how it is scended from so many aboriginally distinct species. that the breeds so often have a somewhat monstrous Ask, as I have asked, a celebrated raiser of Hereford character. It is also a most favourable circumstance cattle, whether his cattle might not have descended for the production of distinct breeds, that male and from long horns, and he will laugh you to scorn. I female pigeons can be easily mated for life; and thus have never met a pigeon, or poultry, or duck, or rab-

33 On the Origin of Species bit fancier, who was not fully convinced that each they deride the idea of species in a state of nature main breed was descended from a distinct species. being lineal descendants of other species? Van Mons, in his treatise on pears and apples, shows how utterly he disbelieves that the several sorts, for Selection. — Let us now briefly consider the steps instance a Ribston-pippin or Codlin-apple, could by which domestic races have been produced, ei- ever have proceeded from the seeds of the same tree. ther from one or from several allied species. Some Innumerable other examples could be given. The ex- little effect may, perhaps, be attributed to the direct planation, I think, is simple: from long-continued action of the external conditions of life, and some study they are strongly impressed with the differ- little to habit; but he would be a bold man who ences between the several races; and though they would account by such agencies for the differences well know that each race varies slightly, for they win of a dray and race horse, a greyhound and blood- their prizes by selecting such slight differences, yet hound, a carrier and tumbler pigeon. One of the most they ignore all general arguments, and refuse to sum remarkable features in our domesticated races is that up in their minds slight differences accumulated we see in them adaptation, not indeed to the animal’s during many successive generations. May not those or plant’s own good, but to man’s use or fancy. Some naturalists who, knowing far less of the laws of in- variations useful to him have probably arisen sud- heritance than does the breeder, and knowing no denly, or by one step; many botanists, for instance, more than he does of the intermediate links in the believe that the fuller’s teazle, with its hooks, which long lines of descent, yet admit that many of our cannot be rivalled by any mechanical contrivance, is domestic races have descended from the same par- only a variety of the wild Dipsacus; and this amount ents—may they not learn a lesson of caution, when of change may have suddenly arisen in a seedling.

34 Charles Darwin So it has probably been with the turnspit dog; and is man’s power of accumulative selection: nature this is known to have been the case with the ancon gives successive variations; man adds them up in sheep. But when we compare the dray-horse and certain directions useful to him. In this sense he may race-horse, the dromedary and camel, the various be said to make for himself useful breeds. breeds of sheep fitted either for cultivated land or The great power of this principle of selection is not mountain pasture, with the wool of one breed good hypothetical. It is certain that several of our eminent for one purpose, and that of another breed for an- breeders have, even within a single lifetime, modi- other purpose; when we compare the many breeds fied to a large extent some breeds of cattle and sheep. of dogs, each good for man in very different ways; In order fully to realise what they have done, it is when we compare the game-cock, so pertinacious in almost necessary to read several of the many trea- battle, with other breeds so little quarrelsome, with tises devoted to this subject, and to inspect the ani- ‘everlasting layers’ which never desire to sit, and with mals. Breeders habitually speak of an animal’s the bantam so small and elegant; when we compare organisation as something quite plastic, which they the host of agricultural, culinary, orchard, and flower- can model almost as they please. If I had space I could garden races of plants, most useful to man at differ- quote numerous passages to this effect from highly ent seasons and for different purposes, or so beauti- competent authorities. Youatt, who was probably ful in his eyes, we must, I think, look further than to better acquainted with the works of agriculturalists mere variability. We cannot suppose that all the than almost any other individual, and who was him- breeds were suddenly produced as perfect and as self a very good judge of an animal, speaks of the useful as we now see them; indeed, in several cases, principle of selection as ‘that which enables the agri- we know that this has not been their history. The key culturist, not only to modify the character of his flock,

35 On the Origin of Species but to change it altogether. It is the magician’s wand, ported to almost every quarter of the world. The by means of which he may summon into life what- improvement is by no means generally due to cross- ever form and mould he pleases.’ Lord Somerville, ing different breeds; all the best breeders are strongly speaking of what breeders have done for sheep, opposed to this practice, except sometimes amongst says:- ‘It would seem as if they had chalked out upon closely allied sub-breeds. And when a cross has been a wall a form perfect in itself, and then had given it made, the closest selection is far more indispensable existence.’ That most skilful breeder, Sir John even than in ordinary cases. If selection consisted Sebright, used to say, with respect to pigeons, that merely in separating some very distinct variety, and ‘he would produce any given feather in three years, breeding from it, the principle would be so obvious but it would take him six years to obtain head and as hardly to be worth notice; but its importance con- beak.’ In Saxony the importance of the principle of sists in the great effect produced by the accumula- selection in regard to merino sheep is so fully tion in one direction, during successive generations, recognised, that men follow it as a trade: the sheep of differences absolutely inappreciable by an unedu- are placed on a table and are studied, like a picture cated eye—differences which I for one have vainly by a connoisseur; this is done three times at inter- attempted to appreciate. Not one man in a thousand vals of months, and the sheep are each time marked has accuracy of eye and judgment sufficient to be- and classed, so that the very best may ultimately be come an eminent breeder. If gifted with these quali- selected for breeding. ties, and he studies his subject for years, and devotes What English breeders have actually effected is his lifetime to it with indomitable perseverance, he proved by the enormous prices given for animals will succeed, and may make great improvements; if with a good pedigree; and these have now been ex- he wants any of these qualities, he will assuredly

36 Charles Darwin fail. Few would readily believe in the natural capac- In regard to plants, there is another means of ob- ity and years of practice requisite to become even a serving the accumulated effects of selection— skilful pigeon-fancier. namely, by comparing the diversity of flowers in the The same principles are followed by horticultur- different varieties of the same species in the flower- ists; but the variations are here often more abrupt. garden; the diversity of leaves, pods, or tubers, or No one supposes that our choicest productions have whatever part is valued, in the kitchen-garden, in been produced by a single variation from the aborigi- comparison with the flowers of the same varieties; nal stock. We have proofs that this is not so in some and the diversity of fruit of the same species in the cases, in which exact records have been kept; thus, orchard, in comparison with the leaves and flowers to give a very trifling instance, the steadily-increas- of the same set of varieties. See how different the ing size of the common gooseberry may be quoted. leaves of the cabbage are, and how extremely alike We see an astonishing improvement in many florists’ the flowers; how unlike the flowers of the heartsease flowers, when the flowers of the present day are com- are, and how alike the leaves; how much the fruit of pared with drawings made only twenty or thirty the different kinds of gooseberries differ in size, years ago. When a race of plants is once pretty well colour, shape, and hairiness, and yet the flowers established, the seed-raisers do not pick out the best present very slight differences. It is not that the vari- plants, but merely go over their seed-beds, and pull eties which differ largely in some one point do not up the ‘rogues,’ as they call the plants that deviate differ at all in other points; this is hardly ever, per- from the proper standard. With animals this kind of haps never, the case. The laws of correlation of selection is, in fact, also followed; for hardly any one growth, the importance of which should never be is so careless as to allow his worst animals to breed. overlooked, will ensure some differences; but, as a

37 On the Origin of Species general rule, I cannot doubt that the continued se- Explicit rules are laid down by some of the Roman lection of slight variations, either in the leaves, the classical writers. From passages in Genesis, it is clear flowers, or the fruit, will produce races differing from that the colour of domestic animals was at that early each other chiefly in these characters. period attended to. Savages now sometimes cross It may be objected that the principle of selection their dogs with wild canine animals, to improve the has been reduced to methodical practice for scarcely breed, and they formerly did so, as is attested by more than three-quarters of a century; it has certainly passages in Pliny. The savages in South Africa match been more attended to of late years, and many trea- their draught cattle by colour, as do some of the tises have been published on the subject; and the Esquimaux their teams of dogs. Livingstone shows result, I may add, has been, in a corresponding de- how much good domestic breeds are valued by the gree, rapid and important. But it is very far from true negroes of the interior of Africa who have not asso- that the principle is a modern discovery. I could give ciated with Europeans. Some of these facts do not several references to the full acknowledgment of the show actual selection, but they show that the breed- importance of the principle in works of high antiq- ing of domestic animals was carefully attended to in uity. In rude and barbarous periods of English his- ancient times, and is now attended to by the lowest tory choice animals were often imported, and laws savages. It would, indeed, have been a strange fact, were passed to prevent their exportation: the destruc- had attention not been paid to breeding, for the in- tion of horses under a certain size was ordered, and heritance of good and bad qualities is so obvious. this may be compared to the ‘roguing’ of plants by At the present time, eminent breeders try by me- nurserymen. The principle of selection I find dis- thodical selection, with a distinct object in view, to tinctly given in an ancient Chinese encyclopaedia. make a new strain or sub-breed, superior to anything

38 Charles Darwin existing in the country. But, for our purpose, a kind has been less improved. There is reason to believe of Selection, which may be called Unconscious, and that King Charles’s spaniel has been unconsciously which results from every one trying to possess and modified to a large extent since the time of that mon- breed from the best individual animals, is more im- arch. Some highly competent authorities are con- portant. Thus, a man who intends keeping pointers vinced that the setter is directly derived from the naturally tries to get as good dogs as he can, and spaniel, and has probably been slowly altered from afterwards breeds from his own best dogs, but he it. It is known that the English pointer has been has no wish or expectation of permanently altering greatly changed within the last century, and in this the breed. Nevertheless I cannot doubt that this pro- case the change has, it is believed, been chiefly ef- cess, continued during centuries, would improve fected by crosses with the fox-hound; but what con- and modify any breed, in the same way as Bakewell, cerns us is, that the change has been effected uncon- Collins, &c., by this very same process, only carried sciously and gradually, and yet so effectually, that, on more methodically, did greatly modify, even though the old Spanish pointer certainly came from during their own lifetimes, the forms and qualities Spain, Mr. Borrow has not seen, as I am informed by of their cattle. Slow and insensible changes of this him, any native dog in Spain like our pointer. kind could never be recognised unless actual mea- By a similar process of selection, and by careful surements or careful drawings of the breeds in ques- training, the whole body of English racehorses have tion had been made long ago, which might serve for come to surpass in fleetness and size the parent Arab comparison. In some cases, however, unchanged or stock, so that the latter, by the regulations for the but little changed individuals of the same breed may Goodwood Races, are favoured in the weights they be found in less civilised districts, where the breed carry. Lord Spencer and others have shown how the

39 On the Origin of Species cattle of England have increased in weight and in and yet the difference between the sheep possessed early maturity, compared with the stock formerly by these two gentlemen is so great that they have the kept in this country. By comparing the accounts appearance of being quite different varieties.’ given in old pigeon treatises of carriers and tumblers If there exist savages so barbarous as never to think with these breeds as now existing in Britain, India, of the inherited character of the offspring of their and Persia, we can, I think, clearly trace the stages domestic animals, yet any one animal particularly through which they have insensibly passed, and useful to them, for any special purpose, would be come to differ so greatly from the rock-pigeon. carefully preserved during famines and other acci- Youatt gives an excellent illustration of the effects dents, to which savages are so liable, and such choice of a course of selection, which may be considered as animals would thus generally leave more offspring unconsciously followed, in so far that the breeders than the inferior ones; so that in this case there would could never have expected or even have wished to be a kind of unconscious selection going on. We see have produced the result which ensued—namely, the the value set on animals even by the barbarians of production of two distinct strains. The two flocks of Tierra del Fuego, by their killing and devouring their Leicester sheep kept by Mr. Buckley and Mr. Burgess, old women, in times of dearth, as of less value than as Mr. Youatt remarks, ‘have been purely bred from their dogs. the original stock of Mr. Bakewell for upwards of fifty In plants the same gradual process of improve- years. There is not a suspicion existing in the mind of ment, through the occasional preservation of the best any one at all acquainted with the subject that the individuals, whether or not sufficiently distinct to owner of either of them has deviated in any one in- be ranked at their first appearance as distinct variet- stance from the pure blood of Mr. Bakewell’s flock, ies, and whether or not two or more species or races

40 Charles Darwin have become blended together by crossing, may has chanced to appear, selecting it, and so onwards. plainly be recognised in the increased size and But the gardeners of the classical period, who culti- beauty which we now see in the varieties of the heart- vated the best pear they could procure, never sease, rose, pelargonium, dahlia, and other plants, thought what splendid fruit we should eat; though when compared with the older varieties or with their we owe our excellent fruit, in some small degree, to parent-stocks. No one would ever expect to get a first- their having naturally chosen and preserved the best rate heartsease or dahlia from the seed of a wild varieties they could anywhere find. plant. No one would expect to raise a first-rate melt- A large amount of change in our cultivated plants, ing pear from the seed of a wild pear, though he thus slowly and unconsciously accumulated, ex- might succeed from a poor seedling growing wild, plains, as I believe, the well-known fact, that in a vast if it had come from a garden-stock. The pear, though number of cases we cannot recognise, and therefore cultivated in classical times, appears, from Pliny’s do not know, the wild parent-stocks of the plants description, to have been a fruit of very inferior qual- which have been longest cultivated in our flower and ity. I have seen great surprise expressed in horticul- kitchen gardens. If it has taken centuries or thousands tural works at the wonderful skill of gardeners, in of years to improve or modify most of our plants up having produced such splendid results from such to their present standard of usefulness to man, we poor materials; but the art, I cannot doubt, has been can understand how it is that neither Australia, the simple, and, as far as the final result is concerned, Cape of Good Hope, nor any other region inhabited has been followed almost unconsciously. It has con- by quite uncivilised man, has afforded us a single sisted in always cultivating the best known variety, plant worth culture. It is not that these countries, so sowing its seeds, and, when a slightly better variety rich in species, do not by a strange chance possess

41 On the Origin of Species the aboriginal stocks of any useful plants, but that once obvious, how it is that our domestic races show the native plants have not been improved by contin- adaptation in their structure or in their habits to ued selection up to a standard of perfection compa- man’s wants or fancies. We can, I think, further un- rable with that given to the plants in countries an- derstand the frequently abnormal character of our ciently civilised. domestic races, and likewise their differences being In regard to the domestic animals kept by so great in external characters and relatively so slight uncivilised man, it should not be overlooked that in internal parts or organs. Man can hardly select, or they almost always have to struggle for their own only with much difficulty, any deviation of struc- food, at least during certain seasons. And in two ture excepting such as is externally visible; and in- countries very differently circumstanced, individu- deed he rarely cares for what is internal. He can never als of the same species, having slightly different con- act by selection, excepting on variations which are stitutions or structure, would often succeed better first given to him in some slight degree by nature. in the one country than in the other, and thus by a No man would ever try to make a fantail, till he saw process of ‘natural selection,’ as will hereafter be a pigeon with a tail developed in some slight de- more fully explained, two sub-breeds might be gree in an unusual manner, or a pouter till he saw a formed. This, perhaps, partly explains what has been pigeon with a crop of somewhat unusual size; and remarked by some authors, namely, that the variet- the more abnormal or unusual any character was ies kept by savages have more of the character of when it first appeared, the more likely it would be species than the varieties kept in civilised countries. to catch his attention. But to use such an expression On the view here given of the all-important part as trying to make a fantail, is, I have no doubt, in which selection by man has played, it becomes at most cases, utterly incorrect. The man who first se-

42 Charles Darwin lected a pigeon with a slightly larger tail, never them, after several breeds have once fairly been es- dreamed what the descendants of that pigeon would tablished. Many slight differences might, and indeed become through long-continued, partly unconscious do now, arise amongst pigeons, which are rejected and partly methodical selection. Perhaps the parent as faults or deviations from the standard of perfec- bird of all fantails had only fourteen tail-feathers tion of each breed. The common goose has not given somewhat expanded, like the present Java fantail, rise to any marked varieties; hence the Thoulouse or like individuals of other and distinct breeds, in and the common breed, which differ only in colour, which as many as seventeen tail-feathers have been that most fleeting of characters, have lately been ex- counted. Perhaps the first pouter-pigeon did not in- hibited as distinct at our poultry-shows. flate its crop much more than the turbit now does I think these views further explain what has some- the upper part of its oesophagus,—a habit which is times been noticed—namely that we know nothing disregarded by all fanciers, as it is not one of the about the origin or history of any of our domestic points of the breed. breeds. But, in fact, a breed, like a dialect of a lan- Nor let it be thought that some great deviation of guage, can hardly be said to have had a definite ori- structure would be necessary to catch the fancier’s gin. A man preserves and breeds from an individual eye: he perceives extremely small differences, and it with some slight deviation of structure, or takes more is in human nature to value any novelty, however care than usual in matching his best animals and thus slight, in one’s own possession. Nor must the value improves them, and the improved individuals which would formerly be set on any slight differ- slowly spread in the immediate neighbourhood. But ences in the individuals of the same species, be as yet they will hardly have a distinct name, and from judged of by the value which would now be set on being only slightly valued, their history will be dis-

43 On the Origin of Species regarded. When further improved by the same slow favourable, as freely giving the materials for selec- and gradual process, they will spread more widely, tion to work on; not that mere individual differences and will get recognised as something distinct and are not amply sufficient, with extreme care, to allow valuable, and will then probably first receive a proof the accumulation of a large amount of modifica- vincial name. In semi-civilised countries, with little tion in almost any desired direction. But as varia- free communication, the spreading and knowledge tions manifestly useful or pleasing to man appear of any new sub-breed will be a slow process. As soon only occasionally, the chance of their appearance will as the points of value of the new sub-breed are once be much increased by a large number of individuals fully acknowledged, the principle, as I have called being kept; and hence this comes to be of the highest it, of unconscious selection will always tend,—per- importance to success. On this principle Marshall haps more at one period than at another, as the breed has remarked, with respect to the sheep of parts of rises or falls in fashion,—perhaps more in one dis- Yorkshire, that ‘as they generally belong to poor trict than in another, according to the state of people, and are mostly in small lots, they never can civilisation of the inhabitants—slowly to add to the be improved.’ On the other hand, nurserymen, from characteristic features of the breed, whatever they raising large stocks of the same plants, are generally may be. But the chance will be infinitely small of far more successful than amateurs in getting new and any record having been preserved of such slow, vary- valuable varieties. The keeping of a large number of ing, and insensible changes. individuals of a species in any country requires that I must now say a few words on the circumstances, the species should be placed under favourable con- favourable, or the reverse, to man’s power of selec- ditions of life, so as to breed freely in that country. tion. A high degree of variability is obviously When the individuals of any species are scanty, all

44 Charles Darwin the individuals, whatever their quality may be, will In the case of animals with separate sexes, facility generally be allowed to breed, and this will effectu- in preventing crosses is an important element of suc- ally prevent selection. But probably the most impor- cess in the formation of new races,—at least, in a tant point of all, is, that the animal or plant should country which is already stocked with other races. be so highly useful to man, or so much valued by In this respect enclosure of the land plays a part. him, that the closest attention should be paid to even Wandering savages or the inhabitants of open plains the slightest deviation in the qualities or structure rarely possess more than one breed of the same spe- of each individual. Unless such attention be paid cies. Pigeons can be mated for life, and this is a great nothing can be effected. I have seen it gravely re- convenience to the fancier, for thus many races may marked, that it was most fortunate that the straw- be kept true, though mingled in the same aviary; and berry began to vary just when gardeners began to this circumstance must have largely favoured the attend closely to this plant. No doubt the strawberry improvement and formation of new breeds. Pigeons, had always varied since it was cultivated, but the I may add, can be propagated in great numbers and slight varieties had been neglected. As soon, how- at a very quick rate, and inferior birds may be freely ever, as gardeners picked out individual plants with rejected, as when killed they serve for food. On the slightly larger, earlier, or better fruit, and raised seed- other hand, cats, from their nocturnal rambling hab- lings from them, and again picked out the best seed- its, cannot be matched, and, although so much val- lings and bred from them, then, there appeared ued by women and children, we hardly ever see a (aided by some crossing with distinct species) those distinct breed kept up; such breeds as we do some- many admirable varieties of the strawberry which times see are almost always imported from some have been raised during the last thirty or forty years. other country, often from islands. Although I do not

45 On the Origin of Species doubt that some domestic animals vary less than ability is governed by many unknown laws, more others, yet the rarity or absence of distinct breeds of especially by that of correlation of growth. Something the cat, the donkey, peacock, goose, &c., may be at- may be attributed to the direct action of the condi- tributed in main part to selection not having been tions of life. Something must be attributed to use and brought into play: in cats, from the difficulty in pair- disuse. The final result is thus rendered infinitely ing them; in donkeys, from only a few being kept by complex. In some cases, I do not doubt that the in- poor people, and little attention paid to their breed- tercrossing of species, aboriginally distinct, has ing; in peacocks, from not being very easily reared played an important part in the origin of our domes- and a large stock not kept; in geese, from being valu- tic productions. When in any country several domes- able only for two purposes, food and feathers, and tic breeds have once been established, their occa- more especially from no pleasure having been felt sional intercrossing, with the aid of selection, has, in the display of distinct breeds. no doubt, largely aided in the formation of new sub- To sum up on the origin of our Domestic Races of breeds; but the importance of the crossing of variet- animals and plants. I believe that the conditions of ies has, I believe, been greatly exaggerated, both in life, from their action on the reproductive system, regard to animals and to those plants which are are so far of the highest importance as causing vari- propagated by seed. In plants which are temporarily ability. I do not believe that variability is an inher- propagated by cuttings, buds, &c., the importance ent and necessary contingency, under all circum- of the crossing both of distinct species and of variet- stances, with all organic beings, as some authors have ies is immense; for the cultivator here quite disre- thought. The effects of variability are modified by gards the extreme variability both of hybrids and various degrees of inheritance and of reversion. Vari- mongrels, and the frequent sterility of hybrids; but

46 Charles Darwin the cases of plants not propagated by seed are of Chapter II little importance to us, for their endurance is only temporary. Over all these causes of Change I am con- Variation Under Nature vinced that the accumulative action of Selection, whether applied methodically and more quickly, or Variability — Individual differences — Doubtful unconsciously and more slowly, but more efficiently, species — Wide ranging, much diffused, and com- is by far the predominant Power. mon species vary most — Species of the larger genera in any country vary more than the species of the smaller genera — Many of the species of the larger genera resemble varieties in being very closely, but unequally, related to each other, and in having restricted ranges.

BEFORE APPLYING the principles arrived at in the last chapter to organic beings in a state of nature, we must briefly discuss whether these latter are subject to any variation. To treat this subject at all properly, a long catalogue of dry facts should be given; but these I shall reserve for my future work. Nor shall I here discuss the various definitions which have been given of the term species. No one definition has as

47 On the Origin of Species yet satisfied all naturalists; yet every naturalist would be called a variety. knows vaguely what he means when he speaks of a Again, we have many slight differences which may species. Generally the term includes the unknown be called individual differences, such as are known element of a distinct act of creation. The term ‘vari- frequently to appear in the offspring from the same ety’ is almost equally difficult to define; but here parents, or which may be presumed to have thus community of descent is almost universally implied, arisen, from being frequently observed in the indi- though it can rarely be proved. We have also what viduals of the same species inhabiting the same con- are called monstrosities; but they graduate into va- fined locality. No one supposes that all the individu- rieties. By a monstrosity I presume is meant some als of the same species are cast in the very same considerable deviation of structure in one part, ei- mould. These individual differences are highly im- ther injurious to or not useful to the species, and not portant for us, as they afford materials for natural generally propagated. Some authors use the term selection to accumulate, in the same manner as man ‘variation’ in a technical sense, as implying a modi- can accumulate in any given direction individual fication directly due to the physical conditions of life; differences in his domesticated productions. These and ‘variations’ in this sense are supposed not to be individual differences generally affect what natural- inherited: but who can say that the dwarfed condi- ists consider unimportant parts; but I could show tion of shells in the brackish waters of the Baltic, or by a long catalogue of facts, that parts which must dwarfed plants on Alpine summits, or the thicker be called important, whether viewed under a physi- fur of an animal from far northwards, would not in ological or classificatory point of view, sometimes some cases be inherited for at least some few gen- vary in the individuals of the same species. I am con- erations? and in this case I presume that the form vinced that the most experienced naturalist would

48 Charles Darwin be surprised at the number of the cases of variabil- portant organs never vary; for these same authors ity, even in important parts of structure, which he practically rank that character as important (as some could collect on good authority, as I have collected, few naturalists have honestly confessed) which does during a course of years. It should be remembered not vary; and, under this point of view, no instance that systematists are far from pleased at finding vari- of any important part varying will ever be found: ability in important characters, and that there are not but under any other point of view many instances many men who will laboriously examine internal assuredly can be given. and important organs, and compare them in many There is one point connected with individual dif- specimens of the same species. I should never have ferences, which seems to me extremely perplexing: I expected that the branching of the main nerves close refer to those genera which have sometimes been to the great central ganglion of an insect would have called ‘protean’ or ‘polymorphic,’ in which the spe- been variable in the same species; I should have ex- cies present an inordinate amount of variation; and pected that changes of this nature could have been hardly two naturalists can agree which forms to rank effected only by slow degrees: yet quite recently Mr. as species and which as varieties. We may instance Lubbock has shown a degree of variability in these Rubus, Rosa, and Hieracium amongst plants, sev- main nerves in Coccus, which may almost be com- eral genera of insects, and several genera of Brachio- pared to the irregular branching of the stem of a tree. pod shells. In most polymorphic genera some of the This philosophical naturalist, I may add, has also species have fixed and definite characters. Genera quite recently shown that the muscles in the larvae which are polymorphic in one country seem to be, of certain insects are very far from uniform. Authors with some few exceptions, polymorphic in other sometimes argue in a circle when they state that im- countries, and likewise, judging from Brachiopod

49 On the Origin of Species shells, at former periods of time. These facts seem to mediate characters, he treats the one as a variety of be very perplexing, for they seem to show that this the other, ranking the most common, but sometimes kind of variability is independent of the conditions the one first described, as the species, and the other of life. I am inclined to suspect that we see in these as the variety. But cases of great difficulty, which I polymorphic genera variations in points of structure will not here enumerate, sometimes occur in decid- which are of no service or disservice to the species, ing whether or not to rank one form as a variety of and which consequently have not been seized on and another, even when they are closely connected by rendered definite by natural selection, as hereafter intermediate links; nor will the commonly-assumed will be explained. hybrid nature of the intermediate links always re- Those forms which possess in some considerable move the difficulty. In very many cases, however, degree the character of species, but which are so one form is ranked as a variety of another, not be- closely similar to some other forms, or are so closely cause the intermediate links have actually been linked to them by intermediate gradations, that natu- found, but because analogy leads the observer to ralists do not like to rank them as distinct species, suppose either that they do now somewhere exist, are in several respects the most important for us. We or may formerly have existed; and here a wide door have every reason to believe that many of these for the entry of doubt and conjecture is opened. doubtful and closely-allied forms have permanently Hence, in determining whether a form should be retained their characters in their own country for a ranked as a species or a variety, the opinion of natu- long time; for as long, as far as we know, as have ralists having sound judgment and wide experience good and true species. Practically, when a naturalist seems the only guide to follow. We must, however, can unite two forms together by others having inter- in many cases, decide by a majority of naturalists,

50 Charles Darwin for few well-marked and well-known varieties can for each birth, and which are highly locomotive, be named which have not been ranked as species by doubtful forms, ranked by one zoologist as a spe- at least some competent judges. cies and by another as a variety, can rarely be found That varieties of this doubtful nature are far from within the same country, but are common in sepa- uncommon cannot be disputed. Compare the sev- rated areas. How many of those birds and insects in eral floras of Great Britain, of France or of the United North America and Europe, which differ very States, drawn up by different botanists, and see what slightly from each other, have been ranked by one a surprising number of forms have been ranked by eminent naturalist as undoubted species, and by one botanist as good species, and by another as mere another as varieties, or, as they are often called, as varieties. Mr. H. C. Watson, to whom I lie under deep geographical races! Many years ago, when compar- obligation for assistance of all kinds, has marked for ing, and seeing others compare, the birds from the me 182 British plants, which are generally consid- separate islands of the Galapagos Archipelago, both ered as varieties, but which have all been ranked by one with another, and with those from the American botanists as species; and in making this list he has mainland, I was much struck how entirely vague and omitted many trifling varieties, but which neverthe- arbitrary is the distinction between species and va- less have been ranked by some botanists as species, rieties. On the islets of the little Madeira group there and he has entirely omitted several highly polymor- are many insects which are characterized as variet- phic genera. Under genera, including the most poly- ies in Mr. Wollaston’s admirable work, but which it morphic forms, Mr. Babington gives 251 species, cannot be doubted would be ranked as distinct spe- whereas Mr. Bentham gives only 112,—a difference cies by many entomologists. Even Ireland has a few of 139 doubtful forms! Amongst animals which unite animals, now generally regarded as varieties, but

51 On the Origin of Species which have been ranked as species by some zoolo- eral interesting lines of argument, from geographi- gists. Several most experienced ornithologists con- cal distribution, analogical variation, hybridism, &c., sider our British red grouse as only a strongly- have been brought to bear on the attempt to deter- marked race of a Norwegian species, whereas the mine their rank. I will here give only a single in- greater number rank it as an undoubted species pe- stance,—the well-known one of the primrose and culiar to Great Britain. A wide distance between the cowslip, or Primula veris and elatior. These plants homes of two doubtful forms leads many natural- differ considerably in appearance; they have a dif- ists to rank both as distinct species; but what dis- ferent flavour and emit a different odour; they flower tance, it has been well asked, will suffice? if that beat slightly different periods; they grow in somewhat tween America and Europe is ample, will that be- different stations; they ascend mountains to differ- tween the Continent and the Azores, or Madeira, or ent heights; they have different geographical ranges; the Canaries, or Ireland, be sufficient? It must be and lastly, according to very numerous experiments admitted that many forms, considered by highly- made during several years by that most careful ob- competent judges as varieties, have so perfectly the server Gartner, they can be crossed only with much character of species that they are ranked by other difficulty. We could hardly wish for better evidence highly-competent judges as good and true species. of the two forms being specifically distinct. On the But to discuss whether they are rightly called spe- other hand, they are united by many intermediate cies or varieties, before any definition of these terms links, and it is very doubtful whether these links are has been generally accepted, is vainly to beat the air. hybrids; and there is, as it seems to me, an over- Many of the cases of strongly-marked varieties or whelming amount of experimental evidence, show- doubtful species well deserve consideration; for sev- ing that they descend from common parents, and

52 Charles Darwin consequently must be ranked as varieties. to consider as specific, and what as varieties; for he Close investigation, in most cases, will bring natu- knows nothing of the amount and kind of variation ralists to an agreement how to rank doubtful forms. to which the group is subject; and this shows, at least, Yet it must be confessed, that it is in the best-known how very generally there is some variation. But if he countries that we find the greatest number of forms confine his attention to one class within one coun- of doubtful value. I have been struck with the fact, try, he will soon make up his mind how to rank most that if any animal or plant in a state of nature be highly of the doubtful forms. His general tendency will be useful to man, or from any cause closely attract his to make many species, for he will become impressed, attention, varieties of it will almost universally be just like the pigeon or poultry-fancier before alluded found recorded. These varieties, moreover, will be to, with the amount of difference in the forms which often ranked by some authors as species. Look at the he is continually studying; and he has little general common oak, how closely it has been studied; yet a knowledge of analogical variation in other groups German author makes more than a dozen species out and in other countries, by which to correct his first of forms, which are very generally considered as va- impressions. As he extends the range of his observa- rieties; and in this country the highest botanical au- tions, he will meet with more cases of difficulty; for thorities and practical men can be quoted to show that he will encounter a greater number of closely-allied the sessile and pedunculated oaks are either good and forms. But if his observations be widely extended, distinct species or mere varieties. he will in the end generally be enabled to make up When a young naturalist commences the study of his own mind which to call varieties and which spe- a group of organisms quite unknown to him, he is at cies; but he will succeed in this at the expense of first much perplexed to determine what differences admitting much variation,—and the truth of this ad-

53 On the Origin of Species mission will often be disputed by other naturalists. which are in any degree more distinct and perma- When, moreover, he comes to study allied forms nent, as steps leading to more strongly marked and brought from countries not now continuous, in which more permanent varieties; and at these latter, as lead- case he can hardly hope to find the intermediate links ing to sub-species, and to species. The passage from between his doubtful forms, he will have to trust alone stage of difference to another and higher stage most entirely to analogy, and his difficulties will rise may be, in some cases, due merely to the long-con- to a climax. tinued action of different physical conditions in two Certainly no clear line of demarcation has as yet been different regions; but I have not much faith in this drawn between species and sub-species—that is, the view; and I attribute the passage of a variety, from a forms which in the opinion of some naturalists come state in which it differs very slightly from its parent very near to, but do not quite arrive at the rank of to one in which it differs more, to the action of natu- species; or, again, between sub-species and well- ral selection in accumulating (as will hereafter be marked varieties, or between lesser varieties and in- more fully explained) differences of structure in cer- dividual differences. These differences blend into each tain definite directions. Hence I believe a well- other in an insensible series; and a series impresses marked variety may be justly called an incipient the mind with the idea of an actual passage. species; but whether this belief be justifiable must Hence I look at individual differences, though of be judged of by the general weight of the several small interest to the systematist, as of high impor- facts and views given throughout this work. tance for us, as being the first step towards such slight It need not be supposed that all varieties or incipi- varieties as are barely thought worth recording in ent species necessarily attain the rank of species. works on natural history. And I look at varieties They may whilst in this incipient state become ex-

54 Charles Darwin tinct, or they may endure as varieties for very long to the nature and relations of the species which vary periods, as has been shown to be the case by Mr. most, by tabulating all the varieties in several well- Wollaston with the varieties of certain fossil land- worked floras. At first this seemed a simple task; but shells in Madeira. If a variety were to flourish so as Mr. H. C. Watson, to whom I am much indebted for to exceed in numbers the parent species, it would valuable advice and assistance on this subject, soon then rank as the species, and the species as the vari- convinced me that there were many difficulties, as ety; or it might come to supplant and exterminate did subsequently Dr. Hooker, even in stronger terms. the parent species; or both might co-exist, and both I shall reserve for my future work the discussion of rank as independent species. But we shall hereafter these difficulties, and the tables themselves of the have to return to this subject. proportional numbers of the varying species. Dr. From these remarks it will be seen that I look at the Hooker permits me to add, that after having care- term species, as one arbitrarily given for the sake of fully read my manuscript, and examined the tables, convenience to a set of individuals closely resem- he thinks that the following statements are fairly well bling each other, and that it does not essentially dif- established. The whole subject, however, treated as fer from the term variety, which is given to less dis- it necessarily here is with much brevity, is rather tinct and more fluctuating forms. The term variety, perplexing, and allusions cannot be avoided to the again, in comparison with mere individual differ- ‘struggle for existence,’ ‘divergence of character,’ and ences, is also applied arbitrarily, and for mere con- other questions, hereafter to be discussed. venience sake. Alph. De Candolle and others have shown that Guided by theoretical considerations, I thought that plants which have very wide ranges generally some interesting results might be obtained in regard present varieties; and this might have been expected,

55 On the Origin of Species as they become exposed to diverse physical condi- the species which are already dominant will be the tions, and as they come into competition (which, as most likely to yield offspring which, though in some we shall hereafter see, is a far more important cir- slight degree modified, will still inherit those ad- cumstance) with different sets of organic beings. But vantages that enabled their parents to become domi- my tables further show that, in any limited country, nant over their compatriots. the species which are most common, that is abound If the plants inhabiting a country and described in most in individuals, and the species which are most any Flora be divided into two equal masses, all those widely diffused within their own country (and this in the larger genera being placed on one side, and is a different consideration from wide range, and to all those in the smaller genera on the other side, a a certain extent from commonness), often give rise somewhat larger number of the very common and to varieties sufficiently well-marked to have been much diffused or dominant species will be found recorded in botanical works. Hence it is the most on the side of the larger genera. This, again, might flourishing, or, as they may be called, the dominant have been anticipated; for the mere fact of many spe- species,—those which range widely over the world, cies of the same genus inhabiting any country, shows are the most diffused in their own country, and are that there is something in the organic or inorganic the most numerous in individuals,—which oftenest conditions of that country favourable to the genus; produce well-marked varieties, or, as I consider and, consequently, we might have expected to have them, incipient species. And this, perhaps, might found in the larger genera, or those including many have been anticipated; for, as varieties, in order to species, a large proportional number of dominant become in any degree permanent, necessarily have species. But so many causes tend to obscure this re- to struggle with the other inhabitants of the country, sult, that I am surprised that my tables show even a

56 Charles Darwin small majority on the side of the larger genera. I will many large trees grow, we expect to find saplings. here allude to only two causes of obscurity. Fresh- Where many species of a genus have been formed water and salt-loving plants have generally very through variation, circumstances have been wide ranges and are much diffused, but this seems favourable for variation; and hence we might expect to be connected with the nature of the stations in- that the circumstances would generally be still habited by them, and has little or no relation to the favourable to variation. On the other hand, if we look size of the genera to which the species belong. Again, at each species as a special act of creation, there is no plants low in the scale of organisation are generally apparent reason why more varieties should occur in much more widely diffused than plants higher in a group having many species, than in one having few. the scale; and here again there is no close relation to To test the truth of this anticipation I have arranged the size of the genera. The cause of lowly-organised the plants of twelve countries, and the coleopterous plants ranging widely will be discussed in our chap- insects of two districts, into two nearly equal masses, ter on geographical distribution. the species of the larger genera on one side, and those From looking at species as only strongly-marked of the smaller genera on the other side, and it has and well-defined varieties, I was led to anticipate invariably proved to be the case that a larger pro- that the species of the larger genera in each country portion of the species on the side of the larger gen- would oftener present varieties, than the species of era present varieties, than on the side of the smaller the smaller genera; for wherever many closely re- genera. Moreover, the species of the large genera lated species (i.e. species of the same genus) have which present any varieties, invariably present a been formed, many varieties or incipient species larger average number of varieties than do the spe- ought, as a general rule, to be now forming. Where cies of the small genera. Both these results follow

57 On the Origin of Species when another division is made, and when all the inasmuch as geology plainly tells us that small gen- smallest genera, with from only one to four species, era have in the lapse of time often increased greatly are absolutely excluded from the tables. These facts in size; and that large genera have often come to their are of plain signification on the view that species are maxima, declined, and disappeared. All that we want only strongly marked and permanent varieties; for to show is, that where many species of a genus have whenever many species of the same genus have been been formed, on an average many are still forming; formed, or where, if we may use the expression, the and this holds good. manufactory of species has been active, we ought There are other relations between the species of generally to find the manufactory still in action, more large genera and their recorded varieties which de- especially as we have every reason to believe the serve notice. We have seen that there is no infallible process of manufacturing new species to be a slow criterion by which to distinguish species and well- one. And this certainly is the case, if varieties be marked varieties; and in those cases in which inter- looked at as incipient species; for my tables clearly mediate links have not been found between doubt- show as a general rule that, wherever many species ful forms, naturalists are compelled to come to a of a genus have been formed, the species of that ge- determination by the amount of difference between nus present a number of varieties, that is of incipi- them, judging by analogy whether or not the amount ent species, beyond the average. It is not that all large suffices to raise one or both to the rank of species. genera are now varying much, and are thus increas- Hence the amount of difference is one very impor- ing in the number of their species, or that no small tant criterion in settling whether two forms should genera are now varying and increasing; for if this be ranked as species or varieties. Now Fries has re- had been so, it would have been fatal to my theory; marked in regard to plants, and Westwood in regard

58 Charles Darwin to insects, that in large genera the amount of differ- be divided into sub-genera, or sections, or lesser ence between the species is often exceedingly small. groups. As Fries has well remarked, little groups of I have endeavoured to test this numerically by aver- species are generally clustered like satellites around ages, and, as far as my imperfect results go, they al- certain other species. And what are varieties but ways confirm the view. I have also consulted some groups of forms, unequally related to each other, and sagacious and most experienced observers, and, af- clustered round certain forms—that is, round their ter deliberation, they concur in this view. In this re- parent-species? Undoubtedly there is one most im- spect, therefore, the species of the larger genera re- portant point of difference between varieties and semble varieties, more than do the species of the species; namely, that the amount of difference be- smaller genera. Or the case may be put in another tween varieties, when compared with each other or way, and it may be said, that in the larger genera, in with their parent-species, is much less than that be- which a number of varieties or incipient species tween the species of the same genus. But when we greater than the average are now manufacturing, come to discuss the principle, as I call it, of Diver- many of the species already manufactured still to a gence of Character, we shall see how this may be certain extent resemble varieties, for they differ from explained, and how the lesser differences between each other by a less than usual amount of difference. varieties will tend to increase into the greater differ- Moreover, the species of the large genera are re- ences between species. lated to each other, in the same manner as the variet- There is one other point which seems to me worth ies of any one species are related to each other. No notice. Varieties generally have much restricted naturalist pretends that all the species of a genus are ranges: this statement is indeed scarcely more than equally distinct from each other; they may generally a truism, for if a variety were found to have a wider

59 On the Origin of Species range than that of its supposed parent-species, their Finally, then, varieties have the same general char- denominations ought to be reversed. But there is also acters as species, for they cannot be distinguished reason to believe, that those species which are very from species,—except, firstly, by the discovery of closely allied to other species, and in so far resemble intermediate linking forms, and the occurrence of varieties, often have much restricted ranges. For in- such links cannot affect the actual characters of the stance, Mr. H. C. Watson has marked for me in the forms which they connect; and except, secondly, by well-sifted London Catalogue of plants (4th edition) a certain amount of difference, for two forms, if dif- 63 plants which are therein ranked as species, but fering very little, are generally ranked as varieties, which he considers as so closely allied to other spe- notwithstanding that intermediate linking forms cies as to be of doubtful value: these 63 reputed spe- have not been discovered; but the amount of differ- cies range on an average over 6.9 of the provinces ence considered necessary to give to two forms the into which Mr. Watson has divided Great Britain. rank of species is quite indefinite. In genera having Now, in this same catalogue, 53 acknowledged vari- more than the average number of species in any eties are recorded, and these range over 7.7 prov- country, the species of these genera have more than inces; whereas, the species to which these varieties the average number of varieties. In large genera the belong range over 14.3 provinces. So that the ac- species are apt to be closely, but unequally, allied knowledged varieties have very nearly the same re- together, forming little clusters round certain spe- stricted average range, as have those very closely cies. Species very closely allied to other species ap- allied forms, marked for me by Mr. Watson as doubt- parently have restricted ranges. In all these several ful species, but which are almost universally ranked respects the species of large genera present a strong by British botanists as good and true species. analogy with varieties. And we can clearly under-

60 Charles Darwin stand these analogies, if species have once existed Chapter III as varieties, and have thus originated: whereas, these analogies are utterly inexplicable if each species has Struggle for Existence been independently created. We have, also, seen that it is the most flourishing Bears on natural selection — The term used in a wide and dominant species of the larger genera which on sense — Geometrical powers of increase — Rapid an average vary most; and varieties, as we shall here- increase of naturalised animals and plants —Nature after see, tend to become converted into new and of the checks to increase — Competition universal distinct species. The larger genera thus tend to be- — Effects of climate — Protection from the number come larger; and throughout nature the forms of life of individuals — Complex relations of all animals which are now dominant tend to become still more and plants throughout nature — Struggle for life dominant by leaving many modified and dominant most severe between individuals and varieties of the descendants. But by steps hereafter to be explained, same species; often severe between species of the the larger genera also tend to break up into smaller same genus — The relation of organism to organism genera. And thus, the forms of life throughout the the most important of all relations. universe become divided into groups subordinate to groups. BEFORE ENTERING ON THE SUBJECT of this chapter, I must make a few preliminary remarks, to show how the struggle for existence bears on Natural Selection. It has been seen in the last chapter that amongst organic beings in a state of nature there is some indi-

61 On the Origin of Species vidual variability; indeed I am not aware that this see beautiful adaptations everywhere and in every has ever been disputed. It is immaterial for us part of the organic world. whether a multitude of doubtful forms be called Again, it may be asked, how is it that varieties, species or sub-species or varieties; what rank, for which I have called incipient species, become ulti- instance, the two or three hundred doubtful forms mately converted into good and distinct species, of British plants are entitled to hold, if the existence which in most cases obviously differ from each other of any well-marked varieties be admitted. But the far more than do the varieties of the same species? mere existence of individual variability and of some How do those groups of species, which constitute few well-marked varieties, though necessary as the what are called distinct genera, and which differ from foundation for the work, helps us but little in under- each other more than do the species of the same ge- standing how species arise in nature. How have all nus, arise? All these results, as we shall more fully those exquisite adaptations of one part of the see in the next chapter, follow inevitably from the organisation to another part, and to the conditions struggle for life. Owing to this struggle for life, any of life, and of one distinct organic being to another variation, however slight and from whatever cause being, been perfected? We see these beautiful co- proceeding, if it be in any degree profitable to an adaptations most plainly in the woodpecker and individual of any species, in its infinitely complex missletoe; and only a little less plainly in the hum- relations to other organic beings and to external na- blest parasite which clings to the hairs of a quadru- ture, will tend to the preservation of that individual, ped or feathers of a bird; in the structure of the beetle and will generally be inherited by its offspring. The which dives through the water; in the plumed seed offspring, also, will thus have a better chance of sur- which is wafted by the gentlest breeze; in short, we viving, for, of the many individuals of any species

62 Charles Darwin which are periodically born, but a small number can knowledge. Nothing is easier than to admit in words survive. I have called this principle, by which each the truth of the universal struggle for life, or more slight variation, if useful, is preserved, by the term of difficult—at least I have found it so—than constantly Natural Selection, in order to mark its relation to man’s to bear this conclusion in mind. Yet unless it be thor- power of selection. We have seen that man by selec- oughly engrained in the mind, I am convinced that tion can certainly produce great results, and can adapt the whole economy of nature, with every fact on dis- organic beings to his own uses, through the accumu- tribution, rarity, abundance, extinction, and varia- lation of slight but useful variations, given to him by tion, will be dimly seen or quite misunderstood. We the hand of Nature. But Natural Selection, as we shall behold the face of nature bright with gladness, we hereafter see, is a power incessantly ready for action, often see superabundance of food; we do not see, or and is as immeasurably superior to man’s feeble ef- we forget, that the birds which are idly singing round forts, as the works of Nature are to those of Art. us mostly live on insects or seeds, and are thus con- We will now discuss in a little more detail the stantly destroying life; or we forget how largely these struggle for existence. In my future work this sub- songsters, or their eggs, or their nestlings, are de- ject shall be treated, as it well deserves, at much stroyed by birds and beasts of prey; we do not al- greater length. The elder De Candolle and Lyell have ways bear in mind, that though food may be now largely and philosophically shown that all organic superabundant, it is not so at all seasons of each re- beings are exposed to severe competition. In regard curring year. to plants, no one has treated this subject with more I should premise that I use the term Struggle for spirit and ability than W. Herbert, Dean of Manches- Existence in a large and metaphorical sense, includ- ter, evidently the result of his great horticultural ing dependence of one being on another, and includ-

63 On the Origin of Species ing (which is more important) not only the life of the plants, in order to tempt birds to devour and thus individual, but success in leaving progeny. Two ca- disseminate its seeds rather than those of other nine animals in a time of dearth, may be truly said plants. In these several senses, which pass into each to struggle with each other which shall get food and other, I use for convenience sake the general term of live. But a plant on the edge of a desert is said to struggle for existence. struggle for life against the drought, though more A struggle for existence inevitably follows from the properly it should be said to be dependent on the high rate at which all organic beings tend to increase. moisture. A plant which annually produces a thou- Every being, which during its natural lifetime pro- sand seeds, of which on an average only one comes duces several eggs or seeds, must suffer destruction to maturity, may be more truly said to struggle with during some period of its life, and during some sea- the plants of the same and other kinds which already son or occasional year, otherwise, on the principle clothe the ground. The missletoe is dependent on of geometrical increase, its numbers would quickly the apple and a few other trees, but can only in a far- become so inordinately great that no country could fetched sense be said to struggle with these trees, support the product. Hence, as more individuals are for if too many of these parasites grow on the same produced than can possibly survive, there must in tree, it will languish and die. But several seedling every case be a struggle for existence, either one in- missletoes, growing close together on the same dividual with another of the same species, or with branch, may more truly be said to struggle with each the individuals of distinct species, or with the physi- other. As the missletoe is disseminated by birds, its cal conditions of life. It is the doctrine of Malthus existence depends on birds; and it may metaphori- applied with manifold force to the whole animal and cally be said to struggle with other fruit-bearing vegetable kingdoms; for in this case there can be no

64 Charles Darwin artificial increase of food, and no prudential restraint pair of young in this interval; if this be so, at the end from marriage. Although some species may be now of the fifth century there would be alive fifteen mil- increasing, more or less rapidly, in numbers, all can- lion elephants, descended from the first pair. not do so, for the world would not hold them. But we have better evidence on this subject than There is no exception to the rule that every organic mere theoretical calculations, namely, the numerous being naturally increases at so high a rate, that if not recorded cases of the astonishingly rapid increase destroyed, the earth would soon be covered by the of various animals in a state of nature, when circum- progeny of a single pair. Even slow-breeding man stances have been favourable to them during two or has doubled in twenty-five years, and at this rate, in three following seasons. Still more striking is the a few thousand years, there would literally not be evidence from our domestic animals of many kinds standing room for his progeny. Linnaeus has calcu- which have run wild in several parts of the world: if lated that if an annual plant produced only two the statements of the rate of increase of slow-breed- seeds—and there is no plant so unproductive as ing cattle and horses in South America, and latterly this—and their seedlings next year produced two, in Australia, had not been well authenticated, they and so on, then in twenty years there would be a would have been quite incredible. So it is with plants: million plants. The elephant is reckoned to be the cases could be given of introduced plants which have slowest breeder of all known animals, and I have become common throughout whole islands in a pe- taken some pains to estimate its probable minimum riod of less than ten years. Several of the plants now rate of natural increase: it will be under the mark to most numerous over the wide plains of La Plata, assume that it breeds when thirty years old, and goes clothing square leagues of surface almost to the ex- on breeding till ninety years old, bringing forth three clusion of all other plants, have been introduced from

65 On the Origin of Species Europe; and there are plants which now range in In- idly stock every station in which they could any how dia, as I hear from Dr. Falconer, from Cape Comorin exist, and that the geometrical tendency to increase to the Himalaya, which have been imported from must be checked by destruction at some period of America since its discovery. In such cases, and end- life. Our familiarity with the larger domestic animals less instances could be given, no one supposes that tends, I think, to mislead us: we see no great destruc- the fertility of these animals or plants has been sud- tion falling on them, and we forget that thousands denly and temporarily increased in any sensible de- are annually slaughtered for food, and that in a state gree. The obvious explanation is that the conditions of nature an equal number would have somehow to of life have been very favourable, and that there has be disposed of. consequently been less destruction of the old and The only difference between organisms which an- young, and that nearly all the young have been en- nually produce eggs or seeds by the thousand, and abled to breed. In such cases the geometrical ratio of those which produce extremely few, is, that the slow- increase, the result of which never fails to be sur- breeders would require a few more years to people, prising, simply explains the extraordinarily rapid in- under favourable conditions, a whole district, let it crease and wide diffusion of naturalised productions be ever so large. The condor lays a couple of eggs in their new homes. and the ostrich a score, and yet in the same country In a state of nature almost every plant produces the condor may be the more numerous of the two: seed, and amongst animals there are very few which the Fulmar petrel lays but one egg, yet it is believed do not annually pair. Hence we may confidently as- to be the most numerous bird in the world. One fly sert, that all plants and animals are tending to in- deposits hundreds of eggs, and another, like the crease at a geometrical ratio, that all would most rap- hippobosca, a single one; but this difference does

66 Charles Darwin not determine how many individuals of the two spe- In looking at Nature, it is most necessary to keep cies can be supported in a district. A large number the foregoing considerations always in mind—never of eggs is of some importance to those species, which to forget that every single organic being around us depend on a rapidly fluctuating amount of food, for may be said to be striving to the utmost to increase it allows them rapidly to increase in number. But in numbers; that each lives by a struggle at some the real importance of a large number of eggs or period of its life; that heavy destruction inevitably seeds is to make up for much destruction at some falls either on the young or old, during each genera- period of life; and this period in the great majority tion or at recurrent intervals. Lighten any check, miti- of cases is an early one. If an animal can in any way gate the destruction ever so little, and the number of protect its own eggs or young, a small number may the species will almost instantaneously increase to be produced, and yet the average stock be fully kept any amount. The face of Nature may be compared up; but if many eggs or young are destroyed, many to a yielding surface, with ten thousand sharp must be produced, or the species will become ex- wedges packed close together and driven inwards tinct. It would suffice to keep up the full number of by incessant blows, sometimes one wedge being a tree, which lived on an average for a thousand struck, and then another with greater force. years, if a single seed were produced once in a thou- What checks the natural tendency of each species sand years, supposing that this seed were never de- to increase in number is most obscure. Look at the stroyed, and could be ensured to germinate in a fit- most vigorous species; by as much as it swarms in ting place. So that in all cases, the average number numbers, by so much will its tendency to increase of any animal or plant depends only indirectly on be still further increased. We know not exactly what the number of its eggs or seeds. the checks are in even one single instance. Nor will

67 On the Origin of Species this surprise any one who reflects how ignorant we chiefly by slugs and insects. If turf which has long are on this head, even in regard to mankind, so in- been mown, and the case would be the same with comparably better known than any other animal. This turf closely browsed by quadrupeds, be let to grow, subject has been ably treated by several authors, and the more vigorous plants gradually kill the less vig- I shall, in my future work, discuss some of the checks orous, though fully grown, plants: thus out of twenty at considerable length, more especially in regard to species growing on a little plot of turf (three feet by the feral animals of South America. Here I will make four) nine species perished from the other species only a few remarks, just to recall to the reader’s mind being allowed to grow up freely. some of the chief points. Eggs or very young ani- The amount of food for each species of course gives mals seem generally to suffer most, but this is not the extreme limit to which each can increase; but very invariably the case. With plants there is a vast de- frequently it is not the obtaining food, but the serv- struction of seeds, but, from some observations ing as prey to other animals, which determines the which I have made, I believe that it is the seedlings average numbers of a species. Thus, there seems to which suffer most from germinating in ground al- be little doubt that the stock of partridges, grouse, ready thickly stocked with other plants. Seedlings, and hares on any large estate depends chiefly on the also, are destroyed in vast numbers by various en- destruction of vermin. If not one head of game were emies; for instance, on a piece of ground three feet shot during the next twenty years in England, and, long and two wide, dug and cleared, and where there at the same time, if no vermin were destroyed, there could be no choking from other plants, I marked all would, in all probability, be less game than at the seedlings of our native weeds as they came up, present, although hundreds of thousands of game and out of the 357 no less than 295 were destroyed, animals are now annually killed. On the other hand,

68 Charles Darwin in some cases, as with the elephant and rhinoceros, suffer most. When we travel from south to north, or none are destroyed by beasts of prey: even the tiger from a damp region to a dry, we invariably see some in India most rarely dares to attack a young elephant species gradually getting rarer and rarer, and finally protected by its dam. disappearing; and the change of climate being con- Climate plays an important part in determining the spicuous, we are tempted to attribute the whole ef- average numbers of a species, and periodical sea- fect to its direct action. But this is a very false view: sons of extreme cold or drought, I believe to be the we forget that each species, even where it most most effective of all checks. I estimated that the win- abounds, is constantly suffering enormous destruc- ter of 1854-55 destroyed four-fifths of the birds in tion at some period of its life, from enemies or from my own grounds; and this is a tremendous destruc- competitors for the same place and food; and if these tion, when we remember that ten per cent. is an ex- enemies or competitors be in the least degree traordinarily severe mortality from epidemics with favoured by any slight change of climate, they will man. The action of climate seems at first sight to be increase in numbers, and, as each area is already fully quite independent of the struggle for existence; but stocked with inhabitants, the other species will de- in so far as climate chiefly acts in reducing food, it crease. When we travel southward and see a species brings on the most severe struggle between the in- decreasing in numbers, we may feel sure that the dividuals, whether of the same or of distinct spe- cause lies quite as much in other species being cies, which subsist on the same kind of food. Even favoured, as in this one being hurt. So it is when we when climate, for instance extreme cold, acts directly, travel northward, but in a somewhat lesser degree, it will be the least vigorous, or those which have got for the number of species of all kinds, and therefore least food through the advancing winter, which will of competitors, decreases northwards; hence in go-

69 On the Origin of Species ing northward, or in ascending a mountain, we far which have from some cause, possibly in part oftener meet with stunted forms, due to the directly through facility of diffusion amongst the crowded injurious action of climate, than we do in proceed- animals, been disproportionably favoured: and here ing southwards or in descending a mountain. When comes in a sort of struggle between the parasite and we reach the Arctic regions, or snow-capped sum- its prey. mits, or absolute deserts, the struggle for life is al- On the other hand, in many cases, a large stock of most exclusively with the elements. individuals of the same species, relatively to the That climate acts in main part indirectly by numbers of its enemies, is absolutely necessary for favouring other species, we may clearly see in the its preservation. Thus we can easily raise plenty of prodigious number of plants in our gardens which corn and rape-seed, &c., in our fields, because the can perfectly well endure our climate, but which seeds are in great excess compared with the number never become naturalised, for they cannot compete of birds which feed on them; nor can the birds, with our native plants, nor resist destruction by our though having a superabundance of food at this one native animals. season, increase in number proportionally to the When a species, owing to highly favourable cir- supply of seed, as their numbers are checked dur- cumstances, increases inordinately in numbers in a ing winter: but any one who has tried, knows how small tract, epidemics—at least, this seems gener- troublesome it is to get seed from a few wheat or ally to occur with our game animals—often ensue: other such plants in a garden; I have in this case lost and here we have a limiting check independent of every single seed. This view of the necessity of a large the struggle for life. But even some of these so-called stock of the same species for its preservation, ex- epidemics appear to be due to parasitic worms, plains, I believe, some singular facts in nature, such

70 Charles Darwin as that of very rare plants being sometimes extremely acres of exactly the same nature had been enclosed abundant in the few spots where they do occur; and twenty-five years previously and planted with Scotch that of some social plants being social, that is, abound- fir. The change in the native vegetation of the planted ing in individuals, even on the extreme confines of part of the heath was most remarkable, more than is their range. For in such cases, we may believe, that a generally seen in passing from one quite different plant could exist only where the conditions of its life soil to another: not only the proportional numbers were so favourable that many could exist together, of the heath-plants were wholly changed, but twelve and thus save each other from utter destruction. I species of plants (not counting grasses and carices) should add that the good effects of frequent intercross- flourished in the plantations, which could not be ing, and the ill effects of close interbreeding, prob- found on the heath. The effect on the insects must ably come into play in some of these cases; but on have been still greater, for six insectivorous birds this intricate subject I will not here enlarge. were very common in the plantations, which were Many cases are on record showing how complex not to be seen on the heath; and the heath was fre- and unexpected are the checks and relations between quented by two or three distinct insectivorous birds. organic beings, which have to struggle together in Here we see how potent has been the effect of the the same country. I will give only a single instance, introduction of a single tree, nothing whatever else which, though a simple one, has interested me. In having been done, with the exception that the land Staffordshire, on the estate of a relation where I had had been enclosed, so that cattle could not enter. But ample means of investigation, there was a large and how important an element enclosure is, I plainly saw extremely barren heath, which had never been near Farnham, in Surrey. Here there are extensive touched by the hand of man; but several hundred heaths, with a few clumps of old Scotch firs on the

71 On the Origin of Species distant hill-tops: within the last ten years large spaces barren and so extensive that no one would ever have have been enclosed, and self-sown firs are now imagined that cattle would have so closely and ef- springing up in multitudes, so close together that fectually searched it for food. all cannot live. Here we see that cattle absolutely determine the When I ascertained that these young trees had not existence of the Scotch fir; but in several parts of the been sown or planted, I was so much surprised at world insects determine the existence of cattle. Per- their numbers that I went to several points of view, haps Paraguay offers the most curious instance of whence I could examine hundreds of acres of the this; for here neither cattle nor horses nor dogs have unenclosed heath, and literally I could not see a ever run wild, though they swarm southward and single Scotch fir, except the old planted clumps. But northward in a feral state; and Azara and Rengger on looking closely between the stems of the heath, I have shown that this is caused by the greater num- found a multitude of seedlings and little trees, which ber in Paraguay of a certain fly, which lays its eggs had been perpetually browsed down by the cattle. in the navels of these animals when first born. The In one square yard, at a point some hundreds yards increase of these flies, numerous as they are, must distant from one of the old clumps, I counted thirty- be habitually checked by some means, probably by two little trees; and one of them, judging from the birds. Hence, if certain insectivorous birds (whose rings of growth, had during twenty-six years tried numbers are probably regulated by hawks or beasts to raise its head above the stems of the heath, and of prey) were to increase in Paraguay, the flies would had failed. No wonder that, as soon as the land was decrease—then cattle and horses would become fe- enclosed, it became thickly clothed with vigorously ral, and this would certainly greatly alter (as indeed growing young firs. Yet the heath was so extremely I have observed in parts of South America) the veg-

72 Charles Darwin etation: this again would largely affect the insects; the exotic Lobelia fulgens, in this part of England, is and this, as we just have seen in Staffordshire, the never visited by insects, and consequently, from its insectivorous birds, and so onwards in ever-increas- peculiar structure, never can set a seed. Many of our ing circles of complexity. We began this series by orchidaceous plants absolutely require the visits of insectivorous birds, and we have ended with them. moths to remove their pollen-masses and thus to Not that in nature the relations can ever be as simple fertilise them. I have, also, reason to believe that as this. Battle within battle must ever be recurring humble-bees are indispensable to the fertilisation of with varying success; and yet in the long-run the the heartsease (Viola tricolor), for other bees do not forces are so nicely balanced, that the face of nature visit this flower. From experiments which I have remains uniform for long periods of time, though as- tried, I have found that the visits of bees, if not indis- suredly the merest trifle would often give the victory pensable, are at least highly beneficial to the to one organic being over another. Nevertheless so fertilisation of our clovers; but humble-bees alone profound is our ignorance, and so high our presump- visit the common red clover (Trifolium pratense), as tion, that we marvel when we hear of the extinction of other bees cannot reach the nectar. Hence I have very an organic being; and as we do not see the cause, we little doubt, that if the whole genus of humble-bees invoke cataclysms to desolate the world, or invent became extinct or very rare in England, the heart- laws on the duration of the forms of life! sease and red clover would become very rare, or I am tempted to give one more instance showing wholly disappear. The number of humble-bees in how plants and animals, most remote in the scale of any district depends in a great degree on the num- nature, are bound together by a web of complex re- ber of field-mice, which destroy their combs and lations. I shall hereafter have occasion to show that nests; and Mr. H. Newman, who has long attended

73 On the Origin of Species to the habits of humble-bees, believes that ‘more than tangled bank, we are tempted to attribute their pro- two thirds of them are thus destroyed all over En- portional numbers and kinds to what we call chance. gland.’ Now the number of mice is largely depen- But how false a view is this! Every one has heard dent, as every one knows, on the number of cats; and that when an American forest is cut down, a very Mr. Newman says, ‘Near villages and small towns I different vegetation springs up; but it has been ob- have found the nests of humble-bees more numer- served that the trees now growing on the ancient In- ous than elsewhere, which I attribute to the number dian mounds, in the Southern United States, display of cats that destroy the mice.’ Hence it is quite cred- the same beautiful diversity and proportion of kinds ible that the presence of a feline animal in large num- as in the surrounding virgin forests. What a struggle bers in a district might determine, through the inter- between the several kinds of trees must here have vention first of mice and then of bees, the frequency gone on during long centuries, each annually scat- of certain flowers in that district! tering its seeds by the thousand; what war between In the case of every species, many different checks, insect and insect—between insects, snails, and other acting at different periods of life, and during differ- animals with birds and beasts of prey—all striving ent seasons or years, probably come into play; some to increase, and all feeding on each other or on the one check or some few being generally the most po- trees or their seeds and seedlings, or on the other tent, but all concurring in determining the average plants which first clothed the ground and thus number or even the existence of the species. In some checked the growth of the trees! Throw up a handful cases it can be shown that widely-different checks of feathers, and all must fall to the ground according act on the same species in different districts. When to definite laws; but how simple is this problem com- we look at the plants and bushes clothing an en- pared to the action and reaction of the innumerable

74 Charles Darwin plants and animals which have determined, in the varieties. To keep up a mixed stock of even such course of centuries, the proportional numbers and extremely close varieties as the variously coloured kinds of trees now growing on the old Indian ruins! sweet-peas, they must be each year harvested sepa- The dependency of one organic being on another, rately, and the seed then mixed in due proportion, as of a parasite on its prey, lies generally between otherwise the weaker kinds will steadily decrease beings remote in the scale of nature. This is often the in numbers and disappear. So again with the variet- case with those which may strictly be said to struggle ies of sheep: it has been asserted that certain moun- with each other for existence, as in the case of lo- tain-varieties will starve out other mountain-variet- custs and grass-feeding quadrupeds. But the ies, so that they cannot be kept together. The same struggle almost invariably will be most severe be- result has followed from keeping together different tween the individuals of the same species, for they varieties of the medicinal leech. It may even be frequent the same districts, require the same food, doubted whether the varieties of any one of our do- and are exposed to the same dangers. In the case of mestic plants or animals have so exactly the same varieties of the same species, the struggle will gen- strength, habits, and constitution, that the original erally be almost equally severe, and we sometimes proportions of a mixed stock could be kept up for see the contest soon decided: for instance, if several half a dozen generations, if they were allowed to varieties of wheat be sown together, and the mixed struggle together, like beings in a state of nature, and seed be resown, some of the varieties which best suit if the seed or young were not annually sorted. the soil or climate, or are naturally the most fertile, As species of the same genus have usually, though will beat the others and so yield more seed, and will by no means invariably, some similarity in habits consequently in a few years quite supplant the other and constitution, and always in structure, the

75 On the Origin of Species struggle will generally be more severe between spe- structure of every organic being is related, in the most cies of the same genus, when they come into compe- essential yet often hidden manner, to that of all other tition with each other, than between species of dis- organic beings, with which it comes into competi- tinct genera. We see this in the recent extension over tion for food or residence, or from which it has to parts of the United States of one species of swallow escape, or on which it preys. This is obvious in the having caused the decrease of another species. The structure of the teeth and talons of the tiger; and in recent increase of the missel-thrush in parts of Scot- that of the legs and claws of the parasite which clings land has caused the decrease of the song-thrush. How to the hair on the tiger’s body. But in the beautifully frequently we hear of one species of rat taking the plumed seed of the dandelion, and in the flattened place of another species under the most different cli- and fringed legs of the water-beetle, the relation mates! In Russia the small Asiatic cockroach has ev- seems at first confined to the elements of air and erywhere driven before it its great congener. One water. Yet the advantage of plumed seeds no doubt species of charlock will supplant another, and so in stands in the closest relation to the land being al- other cases. We can dimly see why the competition ready thickly clothed by other plants; so that the should be most severe between allied forms, which seeds may be widely distributed and fall on unoc- fill nearly the same place in the economy of nature; cupied ground. In the water-beetle, the structure of but probably in no one case could we precisely say its legs, so well adapted for diving, allows it to com- why one species has been victorious over another in pete with other aquatic insects, to hunt for its own the great battle of life. prey, and to escape serving as prey to other animals. A corollary of the highest importance may be de- The store of nutriment laid up within the seeds of duced from the foregoing remarks, namely, that the many plants seems at first sight to have no sort of

76 Charles Darwin relation to other plants. But from the strong growth the rigour of the climate alone. Not until we reach of young plants produced from such seeds (as peas the extreme confines of life, in the arctic regions or and beans), when sown in the midst of long grass, I on the borders of an utter desert, will competition suspect that the chief use of the nutriment in the cease. The land may be extremely cold or dry, yet seed is to favour the growth of the young seedling, there will be competition between some few species, whilst struggling with other plants growing vigor- or between the individuals of the same species, for ously all around. the warmest or dampest spots. Look at a plant in the midst of its range, why does Hence, also, we can see that when a plant or ani- it not double or quadruple its numbers? We know mal is placed in a new country amongst new com- that it can perfectly well withstand a little more heat petitors, though the climate may be exactly the same or cold, dampness or dryness, for elsewhere it ranges as in its former home, yet the conditions of its life into slightly hotter or colder, damper or drier dis- will generally be changed in an essential manner. If tricts. In this case we can clearly see that if we wished we wished to increase its average numbers in its new in imagination to give the plant the power of increas- home, we should have to modify it in a different way ing in number, we should have to give it some ad- to what we should have done in its native country; vantage over its competitors, or over the animals for we should have to give it some advantage over a which preyed on it. On the confines of its geographi- different set of competitors or enemies. cal range, a change of constitution with respect to It is good thus to try in our imagination to give climate would clearly be an advantage to our plant; any form some advantage over another. Probably in but we have reason to believe that only a few plants no single instance should we know what to do, so as or animals range so far, that they are destroyed by to succeed. It will convince us of our ignorance on

77 On the Origin of Species the mutual relations of all organic beings; a convic- Chapter IV tion as necessary, as it seems to be difficult to acquire. All that we can do, is to keep steadily in mind Natural Selection that each organic being is striving to increase at a geometrical ratio; that each at some period of its life, Natural Selection — its power compared with man’s during some season of the year, during each genera- selection — its power on characters of trifling im- tion or at intervals, has to struggle for life, and to portance — its power at all ages and on both sexes suffer great destruction. When we reflect on this — Sexual Selection — On the generality of inter- struggle, we may console ourselves with the full crosses between individuals of the same species — belief, that the war of nature is not incessant, that no Circumstances favourable and unfavourable to Natu- fear is felt, that death is generally prompt, and that ral Selection, namely, intercrossing, isolation, num- the vigorous, the healthy, and the happy survive and ber of individuals — Slow action — Extinction multiply. caused by Natural Selection —Divergence of Char- acter, related to the diversity of inhabitants of any small area, and to naturalisation — Action of Natu- ral Selection, through Divergence of Character and Extinction, on the descendants from a common parent — Explains the Grouping of all organic beings.

HOW WILL THE STRUGGLE for existence, discussed too briefly in the last chapter, act in regard to variation?

78 Charles Darwin Can the principle of selection, which we have seen chance of surviving and of procreating their kind? is so potent in the hands of man, apply in nature? I On the other hand, we may feel sure that any varia- think we shall see that it can act most effectually. Let tion in the least degree injurious would be rigidly it be borne in mind in what an endless number of destroyed. This preservation of favourable variations strange peculiarities our domestic productions, and, and the rejection of injurious variations, I call Natu- in a lesser degree, those under nature, vary; and how ral Selection. Variations neither useful nor injurious strong the hereditary tendency is. Under domestica- would not be affected by natural selection, and tion, it may be truly said that the whole organisation would be left a fluctuating element, as perhaps we becomes in some degree plastic. Let it be borne in see in the species called polymorphic. mind how infinitely complex and close-fitting are We shall best understand the probable course of the mutual relations of all organic beings to each natural selection by taking the case of a country un- other and to their physical conditions of life. Can it, dergoing some physical change, for instance, of cli- then, be thought improbable, seeing that variations mate. The proportional numbers of its inhabitants useful to man have undoubtedly occurred, that other would almost immediately undergo a change, and variations useful in some way to each being in the some species might become extinct. We may con- great and complex battle of life, should sometimes clude, from what we have seen of the intimate and occur in the course of thousands of generations? If complex manner in which the inhabitants of each such do occur, can we doubt (remembering that country are bound together, that any change in the many more individuals are born than can possibly numerical proportions of some of the inhabitants, survive) that individuals having any advantage, independently of the change of climate itself, would however slight, over others, would have the best most seriously affect many of the others. If the coun-

79 On the Origin of Species try were open on its borders, new forms would cer- chapter, that a change in the conditions of life, by tainly immigrate, and this also would seriously dis- specially acting on the reproductive system, causes turb the relations of some of the former inhabitants. or increases variability; and in the foregoing case the Let it be remembered how powerful the influence of conditions of life are supposed to have undergone a a single introduced tree or mammal has been shown change, and this would manifestly be favourable to to be. But in the case of an island, or of a country natural selection, by giving a better chance of profit- partly surrounded by barriers, into which new and able variations occurring; and unless profitable varia- better adapted forms could not freely enter, we tions do occur, natural selection can do nothing. Not should then have places in the economy of nature that, as I believe, any extreme amount of variability which would assuredly be better filled up, if some is necessary; as man can certainly produce great re- of the original inhabitants were in some manner sults by adding up in any given direction mere indi- modified; for, had the area been open to immigra- vidual differences, so could Nature, but far more tion, these same places would have been seized on easily, from having incomparably longer time at her by intruders. In such case, every slight modification, disposal. Nor do I believe that any great physical which in the course of ages chanced to arise, and change, as of climate, or any unusual degree of iso- which in any way favoured the individuals of any of lation to check immigration, is actually necessary to the species, by better adapting them to their altered produce new and unoccupied places for natural se- conditions, would tend to be preserved; and natural lection to fill up by modifying and improving some selection would thus have free scope for the work of of the varying inhabitants. For as all the inhabitants improvement. of each country are struggling together with nicely We have reason to believe, as stated in the first balanced forces, extremely slight modifications in the

80 Charles Darwin structure or habits of one inhabitant would often give every internal organ, on every shade of constitutional it an advantage over others; and still further modifi- difference, on the whole machinery of life. Man se- cations of the same kind would often still further in- lects only for his own good; Nature only for that of crease the advantage. No country can be named in the being which she tends. Every selected character which all the native inhabitants are now so perfectly is fully exercised by her; and the being is placed adapted to each other and to the physical conditions under well-suited conditions of life. Man keeps the under which they live, that none of them could any- natives of many climates in the same country; he sel- how be improved; for in all countries, the natives dom exercises each selected character in some pecu- have been so far conquered by naturalised produc- liar and fitting manner; he feeds a long and a short tions, that they have allowed foreigners to take firm beaked pigeon on the same food; he does not exer- possession of the land. And as foreigners have thus cise a long-backed or long-legged quadruped in any everywhere beaten some of the natives, we may peculiar manner; he exposes sheep with long and safely conclude that the natives might have been short wool to the same climate. He does not allow modified with advantage, so as to have better re- the most vigorous males to struggle for the females. sisted such intruders. He does not rigidly destroy all inferior animals, but As man can produce and certainly has produced a protects during each varying season, as far as lies in great result by his methodical and unconscious his power, all his productions. He often begins his means of selection, what may not nature effect? Man selection by some half-monstrous form; or at least can act only on external and visible characters: na- by some modification prominent enough to catch his ture cares nothing for appearances, except in so far eye, or to be plainly useful to him. Under nature, the as they may be useful to any being. She can act on slightest difference of structure or constitution may

81 On the Origin of Species well turn the nicely-balanced scale in the struggle fect is our view into long past geological ages, that for life, and so be preserved. How fleeting are the we only see that the forms of life are now different wishes and efforts of man! how short his time! and from what they formerly were. consequently how poor will his products be, com- Although natural selection can act only through pared with those accumulated by nature during and for the good of each being, yet characters and whole geological periods. Can we wonder, then, that structures, which we are apt to consider as of very nature’s productions should be far ‘truer’ in charac- trifling importance, may thus be acted on. When we ter than man’s productions; that they should be infi- see leaf-eating insects green, and bark-feeders nitely better adapted to the most complex conditions mottled-grey; the alpine ptarmigan white in winter, of life, and should plainly bear the stamp of far higher the red-grouse the colour of heather, and the black- workmanship? grouse that of peaty earth, we must believe that these It may be said that natural selection is daily and tints are of service to these birds and insects in pre- hourly scrutinising, throughout the world, every serving them from danger. Grouse, if not destroyed variation, even the slightest; rejecting that which is at some period of their lives, would increase in count- bad, preserving and adding up all that is good; si- less numbers; they are known to suffer largely from lently and insensibly working, whenever and wher- birds of prey; and hawks are guided by eyesight to ever opportunity offers, at the improvement of each their prey,—so much so, that on parts of the Conti- organic being in relation to its organic and inorganic nent persons are warned not to keep white pigeons, conditions of life. We see nothing of these slow as being the most liable to destruction. Hence I can changes in progress, until the hand of time has see no reason to doubt that natural selection might marked the long lapse of ages, and then so imper- be most effective in giving the proper colour to each

82 Charles Darwin kind of grouse, and in keeping that colour, when a smooth or downy, a yellow or purple fleshed fruit, once acquired, true and constant. Nor ought we to should succeed. think that the occasional destruction of an animal of In looking at many small points of difference be- any particular colour would produce little effect: we tween species, which, as far as our ignorance permits should remember how essential it is in a flock of us to judge, seem to be quite unimportant, we must white sheep to destroy every lamb with the faintest not forget that climate, food, &c., probably produce trace of black. In plants the down on the fruit and the some slight and direct effect. It is, however, far more colour of the flesh are considered by botanists as necessary to bear in mind that there are many un- characters of the most trifling importance: yet we hear known laws of correlation of growth, which, when one from an excellent horticulturist, Downing, that in the part of the organisation is modified through varia- United States smooth-skinned fruits suffer far more tion, and the modifications are accumulated by natu- from a beetle, a curculio, than those with down; that ral selection for the good of the being, will cause other purple plums suffer far more from a certain disease modifications, often of the most unexpected nature. than yellow plums; whereas another disease attacks As we see that those variations which under do- yellow-fleshed peaches far more than those with mestication appear at any particular period of life, other coloured flesh. If, with all the aids of art, these tend to reappear in the offspring at the same pe- slight differences make a great difference in culti- riod;—for instance, in the seeds of the many variet- vating the several varieties, assuredly, in a state of ies of our culinary and agricultural plants; in the cat- nature, where the trees would have to struggle with erpillar and cocoon stages of the varieties of the silk- other trees and with a host of enemies, such differ- worm; in the eggs of poultry, and in the colour of ences would effectually settle which variety, whether the down of their chickens; in the horns of our sheep

83 On the Origin of Species and cattle when nearly adult;—so in a state of na- natural selection will ensure that modifications con- ture, natural selection will be enabled to act on and sequent on other modifications at a different period modify organic beings at any age, by the accumula- of life, shall not be in the least degree injurious: for tion of profitable variations at that age, and by their if they became so, they would cause the extinction inheritance at a corresponding age. If it profit a plant of the species. to have its seeds more and more widely dissemi- Natural selection will modify the structure of the nated by the wind, I can see no greater difficulty in young in relation to the parent, and of the parent in this being effected through natural selection, than in relation to the young. In social animals it will adapt the cotton-planter increasing and improving by se- the structure of each individual for the benefit of the lection the down in the pods on his cotton-trees. community; if each in consequence profits by the Natural selection may modify and adapt the larva selected change. What natural selection cannot do, of an insect to a score of contingencies, wholly dif- is to modify the structure of one species, without ferent from those which concern the mature insect. giving it any advantage, for the good of another spe- These modifications will no doubt affect, through cies; and though statements to this effect may be the laws of correlation, the structure of the adult; and found in works of natural history, I cannot find one probably in the case of those insects which live only case which will bear investigation. A structure used for a few hours, and which never feed, a large part of only once in an animal’s whole life, if of high impor- their structure is merely the correlated result of suc- tance to it, might be modified to any extent by natu- cessive changes in the structure of their larvae. So, ral selection; for instance, the great jaws possessed conversely, modifications in the adult will probably by certain insects, and used exclusively for opening often affect the structure of the larva; but in all cases the cocoon—or the hard tip to the beak of nestling

84 Charles Darwin birds, used for breaking the egg. It has been asserted, to say a few words on what I call Sexual Selection. that of the best short-beaked tumbler-pigeons more This depends, not on a struggle for existence, but on perish in the egg than are able to get out of it; so that a struggle between the males for possession of the fanciers assist in the act of hatching. Now, if nature females; the result is not death to the unsuccessful had to make the beak of a full-grown pigeon very short competitor, but few or no offspring. Sexual selection for the bird’s own advantage, the process of modifi- is, therefore, less rigorous than natural selection. cation would be very slow, and there would be si- Generally, the most vigorous males, those which are multaneously the most rigorous selection of the best fitted for their places in nature, will leave most young birds within the egg, which had the most pow- progeny. But in many cases, victory will depend not erful and hardest beaks, for all with weak beaks would on general vigour, but on having special weapons, inevitably perish: or, more delicate and more easily confined to the male sex. A hornless stag or spurless broken shells might be selected, the thickness of the cock would have a poor chance of leaving offspring. shell being known to vary like every other structure. Sexual selection by always allowing the victor to Sexual Selection. — Inasmuch as peculiarities of- breed might surely give indomitable courage, length ten appear under domestication in one sex and be- to the spur, and strength to the wing to strike in the come hereditarily attached to that sex, the same fact spurred leg, as well as the brutal cock-fighter, who probably occurs under nature, and if so, natural se- knows well that he can improve his breed by careful lection will be able to modify one sex in its func- selection of the best cocks. How low in the scale of tional relations to the other sex, or in relation to nature this law of battle descends, I know not; male wholly different habits of life in the two sexes, as is alligators have been described as fighting, bellow- sometimes the case with insects. And this leads me ing, and whirling round, like Indians in a war-dance,

85 On the Origin of Species for the possession of the females; male salmons have ing by as spectators, at last choose the most attrac- been seen fighting all day long; male stag-beetles tive partner. Those who have closely attended to often bear wounds from the huge mandibles of other birds in confinement well know that they often take males. The war is, perhaps, severest between the individual preferences and dislikes: thus Sir R. males of polygamous animals, and these seem Heron has described how one pied peacock was oftenest provided with special weapons. The males eminently attractive to all his hen birds. It may ap- of carnivorous animals are already well armed; pear childish to attribute any effect to such appar- though to them and to others, special means of de- ently weak means: I cannot here enter on the details fence may be given through means of sexual selec- necessary to support this view; but if man can in a tion, as the mane to the lion, the shoulder-pad to the short time give elegant carriage and beauty to his boar, and the hooked jaw to the male salmon; for the bantams, according to his standard of beauty, I can shield may be as important for victory, as the sword see no good reason to doubt that female birds, by or spear. selecting, during thousands of generations, the most Amongst birds, the contest is often of a more peace- melodious or beautiful males, according to their ful character. All those who have attended to the standard of beauty, might produce a marked effect. subject, believe that there is the severest rivalry be- I strongly suspect that some well-known laws with tween the males of many species to attract by sing- respect to the plumage of male and female birds, in ing the females. The rock-thrush of Guiana, birds of comparison with the plumage of the young, can be Paradise, and some others, congregate; and succes- explained on the view of plumage having been sive males display their gorgeous plumage and per- chiefly modified by sexual selection, acting when the form strange antics before the females, which stand- birds have come to the breeding age or during the

86 Charles Darwin breeding season; the modifications thus produced key-cock, which can hardly be either useful or orna- being inherited at corresponding ages or seasons, ei- mental to this bird;—indeed, had the tuft appeared ther by the males alone, or by the males and females; under domestication, it would have been called a but I have not space here to enter on this subject. monstrosity. Thus it is, as I believe, that when the males and females of any animal have the same general habits Illustrations of the action of Natural Selection. — of life, but differ in structure, colour, or ornament, In order to make it clear how, as I believe, natural such differences have been mainly caused by sexual selection acts, I must beg permission to give one or selection; that is, individual males have had, in suc- two imaginary illustrations. Let us take the case of a cessive generations, some slight advantage over wolf, which preys on various animals, securing some other males, in their weapons, means of defence, or by craft, some by strength, and some by fleetness; charms; and have transmitted these advantages to and let us suppose that the fleetest prey, a deer for their male offspring. Yet, I would not wish to at- instance, had from any change in the country in- tribute all such sexual differences to this agency: for creased in numbers, or that other prey had decreased we see peculiarities arising and becoming attached in numbers, during that season of the year when the to the male sex in our domestic animals (as the wattle wolf is hardest pressed for food. I can under such in male carriers, horn-like protuberances in the cocks circumstances see no reason to doubt that the swift- of certain fowls, &c.), which we cannot believe to be est and slimmest wolves would have the best chance either useful to the males in battle, or attractive to of surviving, and so be preserved or selected,—pro- the females. We see analogous cases under nature, vided always that they retained strength to master for instance, the tuft of hair on the breast of the tur- their prey at this or at some other period of the year,

87 On the Origin of Species when they might be compelled to prey on other ani- the best chance of surviving and of leaving offspring. mals. I can see no more reason to doubt this, than Some of its young would probably inherit the same that man can improve the fleetness of his greyhounds habits or structure, and by the repetition of this pro- by careful and methodical selection, or by that un- cess, a new variety might be formed which would conscious selection which results from each man try- either supplant or coexist with the parent-form of ing to keep the best dogs without any thought of wolf. Or, again, the wolves inhabiting a mountain- modifying the breed. ous district, and those frequenting the lowlands, Even without any change in the proportional num- would naturally be forced to hunt different prey; and bers of the animals on which our wolf preyed, a cub from the continued preservation of the individuals might be born with an innate tendency to pursue best fitted for the two sites, two varieties might certain kinds of prey. Nor can this be thought very slowly be formed. These varieties would cross and improbable; for we often observe great differences blend where they met; but to this subject of inter- in the natural tendencies of our domestic animals; crossing we shall soon have to return. I may add, one cat, for instance, taking to catch rats, another that, according to Mr. Pierce, there are two varieties mice; one cat, according to Mr. St. John, bringing of the wolf inhabiting the Catskill Mountains in the home winged game, another hares or rabbits, and United States, one with a light greyhound-like form, another hunting on marshy ground and almost which pursues deer, and the other more bulky, with nightly catching woodcocks or snipes. The tendency shorter legs, which more frequently attacks the to catch rats rather than mice is known to be inher- shepherd’s flocks. ited. Now, if any slight innate change of habit or of Let us now take a more complex case. Certain structure benefited an individual wolf, it would have plants excrete a sweet juice, apparently for the sake

88 Charles Darwin of eliminating something injurious from their sap: long-run would gain the upper hand. Those flow- this is effected by glands at the base of the stipules ers, also, which had their stamens and pistils placed, in some Leguminosae, and at the back of the leaf of in relation to the size and habits of the particular in- the common laurel. This juice, though small in quan- sects which visited them, so as to favour in any de- tity, is greedily sought by insects. Let us now sup- gree the transportal of their pollen from flower to pose a little sweet juice or nectar to be excreted by flower, would likewise be favoured or selected. We the inner bases of the petals of a flower. In this case might have taken the case of insects visiting flowers insects in seeking the nectar would get dusted with for the sake of collecting pollen instead of nectar; pollen, and would certainly often transport the pol- and as pollen is formed for the sole object of len from one flower to the stigma of another flower. fertilisation, its destruction appears a simple loss to The flowers of two distinct individuals of the same the plant; yet if a little pollen were carried, at first species would thus get crossed; and the act of cross- occasionally and then habitually, by the pollen-de- ing, we have good reason to believe (as will hereaf- vouring insects from flower to flower, and a cross ter be more fully alluded to), would produce very thus effected, although nine-tenths of the pollen were vigorous seedlings, which consequently would have destroyed, it might still be a great gain to the plant; the best chance of flourishing and surviving. Some and those individuals which produced more and of these seedlings would probably inherit the nec- more pollen, and had larger and larger anthers, tar-excreting power. Those individual flowers which would be selected. had the largest glands or nectaries, and which ex- When our plant, by this process of the continued creted most nectar, would be oftenest visited by in- preservation or natural selection of more and more sects, and would be oftenest crossed; and so in the attractive flowers, had been rendered highly attrac-

89 On the Origin of Species tive to insects, they would, unintentionally on their favourable to bees, nevertheless every female flower part, regularly carry pollen from flower to flower; which I examined had been effectually fertilised by and that they can most effectually do this, I could the bees, accidentally dusted with pollen, having easily show by many striking instances. I will give flown from tree to tree in search of nectar. But to re- only one—not as a very striking case, but as like- turn to our imaginary case: as soon as the plant had wise illustrating one step in the separation of the been rendered so highly attractive to insects that sexes of plants, presently to be alluded to. Some pollen was regularly carried from flower to flower, holly-trees bear only male flowers, which have four another process might commence. No naturalist stamens producing rather a small quantity of pol- doubts the advantage of what has been called the len, and a rudimentary pistil; other holly-trees bear ‘physiological division of labour;’ hence we may only female flowers; these have a full-sized pistil, believe that it would be advantageous to a plant to and four stamens with shrivelled anthers, in which produce stamens alone in one flower or on one whole not a grain of pollen can be detected. Having found plant, and pistils alone in another flower or on an- a female tree exactly sixty yards from a male tree, I other plant. In plants under culture and placed un- put the stigmas of twenty flowers, taken from differ- der new conditions of life, sometimes the male or- ent branches, under the microscope, and on all, with- gans and sometimes the female organs become more out exception, there were pollen-grains, and on some or less impotent; now if we suppose this to occur in a profusion of pollen. As the wind had set for sev- ever so slight a degree under nature, then as pollen eral days from the female to the male tree, the pol- is already carried regularly from flower to flower, len could not thus have been carried. The weather and as a more complete separation of the sexes of had been cold and boisterous, and therefore not our plant would be advantageous on the principle

90 Charles Darwin of the division of labour, individuals with this ten- dants. Its descendants would probably inherit a tendency more and more increased, would be continu- dency to a similar slight deviation of structure. The ally favoured or selected, until at last a complete tubes of the corollas of the common red and incar- separation of the sexes would be effected. nate clovers (Trifolium pratense and incarnatum) do Let us now turn to the nectar-feeding insects in our not on a hasty glance appear to differ in length; yet imaginary case: we may suppose the plant of which the hive-bee can easily suck the nectar out of the in- we have been slowly increasing the nectar by con- carnate clover, but not out of the common red clo- tinued selection, to be a common plant; and that cer- ver, which is visited by humble-bees alone; so that tain insects depended in main part on its nectar for whole fields of the red clover offer in vain an abun- food. I could give many facts, showing how anxious dant supply of precious nectar to the hive-bee. Thus bees are to save time; for instance, their habit of cut- it might be a great advantage to the hive-bee to have ting holes and sucking the nectar at the bases of cer- a slightly longer or differently constructed probos- tain flowers, which they can, with a very little more cis. On the other hand, I have found by experiment trouble, enter by the mouth. Bearing such facts in that the fertility of clover greatly depends on bees mind, I can see no reason to doubt that an accidental visiting and moving parts of the corolla, so as to push deviation in the size and form of the body, or in the the pollen on to the stigmatic surface. Hence, again, curvature and length of the proboscis, &c., far too if humble-bees were to become rare in any country, slight to be appreciated by us, might profit a bee or it might be a great advantage to the red clover to other insect, so that an individual so characterised have a shorter or more deeply divided tube to its would be able to obtain its food more quickly, and corolla, so that the hive-bee could visit its flowers. so have a better chance of living and leaving descen- Thus I can understand how a flower and a bee might

91 On the Origin of Species slowly become, either simultaneously or one after lief of the continued creation of new organic beings, the other, modified and adapted in the most perfect or of any great and sudden modification in their manner to each other, by the continued preservation structure. of individuals presenting mutual and slightly On the Intercrossing of Individuals. — I must here favourable deviations of structure. introduce a short digression. In the case of animals I am well aware that this doctrine of natural selec- and plants with separated sexes, it is of course obvi- tion, exemplified in the above imaginary instances, ous that two individuals must always unite for each is open to the same objections which were at first birth; but in the case of hermaphrodites this is far urged against Sir Charles Lyell’s noble views on ‘the from obvious. Nevertheless I am strongly inclined modern changes of the earth, as illustrative of geol- to believe that with all hermaphrodites two individu- ogy;’ but we now very seldom hear the action, for als, either occasionally or habitually, concur for the instance, of the coast-waves, called a trifling and in- reproduction of their kind. This view, I may add, significant cause, when applied to the excavation of was first suggested by Andrew Knight. We shall gigantic valleys or to the formation of the longest presently see its importance; but I must here treat lines of inland cliffs. Natural selection can act only the subject with extreme brevity, though I have the by the preservation and accumulation of infinitesi- materials prepared for an ample discussion. All ver- mally small inherited modifications, each profitable tebrate animals, all insects, and some other large to the preserved being; and as modern geology has groups of animals, pair for each birth. Modern re- almost banished such views as the excavation of a search has much diminished the number of sup- great valley by a single diluvial wave, so will natu- posed hermaphrodites, and of real hermaphrodites ral selection, if it be a true principle, banish the be- a large number pair; that is, two individuals regu-

92 Charles Darwin larly unite for reproduction, which is all that concerns perhaps at very long intervals—indispensable. us. But still there are many hermaphrodite animals On the belief that this is a law of nature, we can, I which certainly do not habitually pair, and a vast think, understand several large classes of facts, such majority of plants are hermaphrodites. What reason, as the following, which on any other view are inex- it may be asked, is there for supposing in these cases plicable. Every hybridizer knows how unfavourable that two individuals ever concur in reproduction? As exposure to wet is to the fertilisation of a flower, yet it is impossible here to enter on details, I must trust to what a multitude of flowers have their anthers and some general considerations alone. stigmas fully exposed to the weather! but if an occa- In the first place, I have collected so large a body sional cross be indispensable, the fullest freedom for of facts, showing, in accordance with the almost uni- the entrance of pollen from another individual will versal belief of breeders, that with animals and explain this state of exposure, more especially as the plants a cross between different varieties, or be- plant’s own anthers and pistil generally stand so tween individuals of the same variety but of an- close together that self-fertilisation seems almost in- other strain, gives vigour and fertility to the off- evitable. Many flowers, on the other hand, have their spring; and on the other hand, that close interbreed- organs of fructification closely enclosed, as in the ing diminishes vigour and fertility; that these facts great papilionaceous or pea-family; but in several, alone incline me to believe that it is a general law perhaps in all, such flowers, there is a very curious of nature (utterly ignorant though we be of the adaptation between the structure of the flower and meaning of the law) that no organic being self- the manner in which bees suck the nectar; for, in do- fertilises itself for an eternity of generations; but that ing this, they either push the flower’s own pollen on a cross with another individual is occasionally— the stigma, or bring pollen from another flower. So

93 On the Origin of Species necessary are the visits of bees to papilionaceous flow- this end: but, the agency of insects is often required ers, that I have found, by experiments published else- to cause the stamens to spring forward, as Kolreuter where, that their fertility is greatly diminished if these has shown to be the case with the barberry; and curi- visits be prevented. Now, it is scarcely possible that ously in this very genus, which seems to have a spe- bees should fly from flower to flower, and not carry cial contrivance for self-fertilisation, it is well known pollen from one to the other, to the great good, as I that if very closely-allied forms or varieties are believe, of the plant. Bees will act like a camel-hair planted near each other, it is hardly possible to raise pencil, and it is quite sufficient just to touch the an- pure seedlings, so largely do they naturally cross. thers of one flower and then the stigma of another with In many other cases, far from there being any aids the same brush to ensure fertilisation; but it must not for self-fertilisation, there are special contrivances, be supposed that bees would thus produce a multi- as I could show from the writings of C. C. Sprengel tude of hybrids between distinct species; for if you and from my own observations, which effectually bring on the same brush a plant’s own pollen and prevent the stigma receiving pollen from its own pollen from another species, the former will have such flower: for instance, in Lobelia fulgens, there is a re- a prepotent effect, that it will invariably and com- ally beautiful and elaborate contrivance by which pletely destroy, as has been shown by Gartner, any every one of the infinitely numerous pollen-gran- influence from the foreign pollen. ules are swept out of the conjoined anthers of each When the stamens of a flower suddenly spring to- flower, before the stigma of that individual flower is wards the pistil, or slowly move one after the other ready to receive them; and as this flower is never towards it, the contrivance seems adapted solely to visited, at least in my garden, by insects, it never ensure self-fertilisation; and no doubt it is useful for sets a seed, though by placing pollen from one flower

94 Charles Darwin on the stigma of another, I raised plenty of seedlings; seedlings thus raised will turn out mongrels: for in- and whilst another species of Lobelia growing close stance, I raised 233 seedling cabbages from some by, which is visited by bees, seeds freely. In very plants of different varieties growing near each other, many other cases, though there be no special me- and of these only 78 were true to their kind, and some chanical contrivance to prevent the stigma of a flower even of these were not perfectly true. Yet the pistil receiving its own pollen, yet, as C. C. Sprengel has of each cabbage-flower is surrounded not only by shown, and as I can confirm, either the anthers burst its own six stamens, but by those of the many other before the stigma is ready for fertilisation, or the flowers on the same plant. How, then, comes it that stigma is ready before the pollen of that flower is such a vast number of the seedlings are mongrel- ready, so that these plants have in fact separated sexes, ized? I suspect that it must arise from the pollen of a and must habitually be crossed. How strange are these distinct variety having a prepotent effect over a facts! How strange that the pollen and stigmatic sur- flower’s own pollen; and that this is part of the gen- face of the same flower, though placed so close to- eral law of good being derived from the intercross- gether, as if for the very purpose of self-fertilisation, ing of distinct individuals of the same species. When should in so many cases be mutually useless to each distinct species are crossed the case is directly the other! How simply are these facts explained on the reverse, for a plant’s own pollen is always prepo- view of an occasional cross with a distinct individual tent over foreign pollen; but to this subject we shall being advantageous or indispensable! return in a future chapter. If several varieties of the cabbage, radish, onion, In the case of a gigantic tree covered with innu- and of some other plants, be allowed to seed near merable flowers, it may be objected that pollen could each other, a large majority, as I have found, of the seldom be carried from tree to tree, and at most only

95 On the Origin of Species from flower to flower on the same tree, and that flow- land there are some hermaphrodites, as land- ers on the same tree can be considered as distinct mollusca and earth-worms; but these all pair. As yet individuals only in a limited sense. I believe this I have not found a single case of a terrestrial animal objection to be valid, but that nature has largely pro- which fertilises itself. We can understand this re- vided against it by giving to trees a strong tendency markable fact, which offers so strong a contrast with to bear flowers with separated sexes. When the sexes terrestrial plants, on the view of an occasional cross are separated, although the male and female flow- being indispensable, by considering the medium in ers may be produced on the same tree, we can see which terrestrial animals live, and the nature of the that pollen must be regularly carried from flower to fertilising element; for we know of no means, analo- flower; and this will give a better chance of pollen gous to the action of insects and of the wind in the being occasionally carried from tree to tree. That trees case of plants, by which an occasional cross could belonging to all Orders have their sexes more often be effected with terrestrial animals without the con- separated than other plants, I find to be the case in currence of two individuals. Of aquatic animals, this country; and at my request Dr. Hooker tabulated there are many self-fertilising hermaphrodites; but the trees of New Zealand, and Dr. Asa Gray those of here currents in the water offer an obvious means the United States, and the result was as I anticipated. for an occasional cross. And, as in the case of flow- On the other hand, Dr. Hooker has recently informed ers, I have as yet failed, after consultation with one me that he finds that the rule does not hold in Aus- of the highest authorities, namely, Professor Huxley, tralia; and I have made these few remarks on the to discover a single case of an hermaphrodite ani- sexes of trees simply to call attention to the subject. mal with the organs of reproduction so perfectly Turning for a very brief space to animals: on the enclosed within the body, that access from without

96 Charles Darwin and the occasional influence of a distinct individual doms, an occasional intercross with a distinct indi- can be shown to be physically impossible. Cirripedes vidual is a law of nature. I am well aware that there long appeared to me to present a case of very great are, on this view, many cases of difficulty, some of difficulty under this point of view; but I have been which I am trying to investigate. Finally then, we may enabled, by a fortunate chance, elsewhere to prove conclude that in many organic beings, a cross be- that two individuals, though both are self-fertilising tween two individuals is an obvious necessity for hermaphrodites, do sometimes cross. each birth; in many others it occurs perhaps only at It must have struck most naturalists as a strange long intervals; but in none, as I suspect, can self- anomaly that, in the case of both animals and plants, fertilisation go on for perpetuity. species of the same family and even of the same genus, though agreeing closely with each other in al- Circumstances favourable to Natural Selection. — most their whole organisation, yet are not rarely, some This is an extremely intricate subject. A large amount of them hermaphrodites, and some of them unisexual. of inheritable and diversified variability is But if, in fact, all hermaphrodites do occasionally in- favourable, but I believe mere individual differences tercross with other individuals, the difference between suffice for the work. A large number of individuals, hermaphrodites and unisexual species, as far as func- by giving a better chance for the appearance within tion is concerned, becomes very small. any given period of profitable variations, will com- From these several considerations and from the pensate for a lesser amount of variability in each in- many special facts which I have collected, but which dividual, and is, I believe, an extremely important I am not here able to give, I am strongly inclined to element of success. Though nature grants vast peri- suspect that, both in the vegetable and animal king- ods of time for the work of natural selection, she does

97 On the Origin of Species not grant an indefinite period; for as all organic be- tricts will almost certainly present different condi- ings are striving, it may be said, to seize on each place tions of life; and then if natural selection be modify- in the economy of nature, if any one species does ing and improving a species in the several districts, not become modified and improved in a correspond- there will be intercrossing with the other individu- ing degree with its competitors, it will soon be ex- als of the same species on the confines of each. And terminated. in this case the effects of intercrossing can hardly be In man’s methodical selection, a breeder selects for counterbalanced by natural selection always tend- some definite object, and free intercrossing will ing to modify all the individuals in each district in wholly stop his work. But when many men, without exactly the same manner to the conditions of each; intending to alter the breed, have a nearly common for in a continuous area, the conditions will gener- standard of perfection, and all try to get and breed ally graduate away insensibly from one district to from the best animals, much improvement and modi- another. The intercrossing will most affect those ani- fication surely but slowly follow from this uncon- mals which unite for each birth, which wander much, scious process of selection, notwithstanding a large and which do not breed at a very quick rate. Hence amount of crossing with inferior animals. Thus it will in animals of this nature, for instance in birds, vari- be in nature; for within a confined area, with some eties will generally be confined to separated coun- place in its polity not so perfectly occupied as might tries; and this I believe to be the case. In hermaphro- be, natural selection will always tend to preserve all dite organisms which cross only occasionally, and the individuals varying in the right direction, though likewise in animals which unite for each birth, but in different degrees, so as better to fill up the unoc- which wander little and which can increase at a very cupied place. But if the area be large, its several dis- rapid rate, a new and improved variety might be

98 Charles Darwin quickly formed on any one spot, and might there will obviously thus act far more efficiently with those maintain itself in a body, so that whatever intercross- animals which unite for each birth; but I have already ing took place would be chiefly between the indi- attempted to show that we have reason to believe viduals of the same new variety. A local variety when that occasional intercrosses take place with all ani- once thus formed might subsequently slowly spread mals and with all plants. Even if these take place to other districts. On the above principle, nursery- only at long intervals, I am convinced that the young men always prefer getting seed from a large body of thus produced will gain so much in vigour and fer- plants of the same variety, as the chance of intercross- tility over the offspring from long-continued self- ing with other varieties is thus lessened. fertilisation, that they will have a better chance of Even in the case of slow-breeding animals, which surviving and propagating their kind; and thus, in unite for each birth, we must not overrate the effects the long run, the influence of intercrosses, even at of intercrosses in retarding natural selection; for I can rare intervals, will be great. If there exist organic be- bring a considerable catalogue of facts, showing that ings which never intercross, uniformity of character within the same area, varieties of the same animal can be retained amongst them, as long as their con- can long remain distinct, from haunting different sta- ditions of life remain the same, only through the prin- tions, from breeding at slightly different seasons, or ciple of inheritance, and through natural selection from varieties of the same kind preferring to pair destroying any which depart from the proper type; together. but if their conditions of life change and they un- Intercrossing plays a very important part in nature dergo modification, uniformity of character can be in keeping the individuals of the same species, or of given to their modified offspring, solely by natural the same variety, true and uniform in character. It selection preserving the same favourable variations.

99 On the Origin of Species Isolation, also, is an important element in the pro- importance in the production of new species. If, how- cess of natural selection. In a confined or isolated ever, an isolated area be very small, either from be- area, if not very large, the organic and inorganic con- ing surrounded by barriers, or from having very ditions of life will generally be in a great degree peculiar physical conditions, the total number of the uniform; so that natural selection will tend to modify individuals supported on it will necessarily be very all the individuals of a varying species throughout small; and fewness of individuals will greatly retard the area in the same manner in relation to the same the production of new species through natural se- conditions. Intercrosses, also, with the individuals lection, by decreasing the chance of the appearance of the same species, which otherwise would have of favourable variations. inhabited the surrounding and differently circum- If we turn to nature to test the truth of these re- stanced districts, will be prevented. But isolation marks, and look at any small isolated area, such as probably acts more efficiently in checking the immi- an oceanic island, although the total number of the gration of better adapted organisms, after any physi- species inhabiting it, will be found to be small, as cal change, such as of climate or elevation of the land, we shall see in our chapter on geographical distri- &c.; and thus new places in the natural economy of bution; yet of these species a very large proportion the country are left open for the old inhabitants to are endemic,—that is, have been produced there, and struggle for, and become adapted to, through modi- nowhere else. Hence an oceanic island at first sight fications in their structure and constitution. Lastly, seems to have been highly favourable for the pro- isolation, by checking immigration and consequently duction of new species. But we may thus greatly competition, will give time for any new variety to deceive ourselves, for to ascertain whether a small be slowly improved; and this may sometimes be of isolated area, or a large open area like a continent,

100 Charles Darwin has been most favourable for the production of new more new places will be formed, and the competi- organic forms, we ought to make the comparison within tion to fill them will be more severe, on a large than equal times; and this we are incapable of doing. on a small and isolated area. Moreover, great areas, Although I do not doubt that isolation is of consid- though now continuous, owing to oscillations of erable importance in the production of new species, level, will often have recently existed in a broken on the whole I am inclined to believe that largeness condition, so that the good effects of isolation will of area is of more importance, more especially in the generally, to a certain extent, have concurred. Finally, production of species, which will prove capable of I conclude that, although small isolated areas prob- enduring for a long period, and of spreading widely. ably have been in some respects highly favourable Throughout a great and open area, not only will there for the production of new species, yet that the course be a better chance of favourable variations arising of modification will generally have been more rapid from the large number of individuals of the same on large areas; and what is more important, that the species there supported, but the conditions of life new forms produced on large areas, which already are infinitely complex from the large number of al- have been victorious over many competitors, will ready existing species; and if some of these many be those that will spread most widely, will give rise species become modified and improved, others will to most new varieties and species, and will thus play have to be improved in a corresponding degree or an important part in the changing history of the or- they will be exterminated. Each new form, also, as ganic world. soon as it has been much improved, will be able to We can, perhaps, on these views, understand some spread over the open and continuous area, and will facts which will be again alluded to in our chapter thus come into competition with many others. Hence on geographical distribution; for instance, that the

101 On the Origin of Species productions of the smaller continent of Australia fossils, connect to a certain extent orders now widely have formerly yielded, and apparently are now separated in the natural scale. These anomalous yielding, before those of the larger Europaeo-Asi- forms may almost be called living fossils; they have atic area. Thus, also, it is that continental produc- endured to the present day, from having inhabited a tions have everywhere become so largely naturalised confined area, and from having thus been exposed on islands. On a small island, the race for life will to less severe competition. have been less severe, and there will have been less To sum up the circumstances favourable and modification and less extermination. Hence, perhaps, unfavourable to natural selection, as far as the ex- it comes that the flora of Madeira, according to treme intricacy of the subject permits. I conclude, Oswald Heer, resembles the extinct tertiary flora of looking to the future, that for terrestrial productions Europe. All fresh-water basins, taken together, make a large continental area, which will probably un- a small area compared with that of the sea or of the dergo many oscillations of level, and which conse- land; and, consequently, the competition between quently will exist for long periods in a broken con- fresh-water productions will have been less severe dition, will be the most favourable for the produc- than elsewhere; new forms will have been more tion of many new forms of life, likely to endure long slowly formed, and old forms more slowly extermi- and to spread widely. For the area will first have nated. And it is in fresh water that we find seven existed as a continent, and the inhabitants, at this genera of Ganoid fishes, remnants of a once prepon- period numerous in individuals and kinds, will have derant order: and in fresh water we find some of the been subjected to very severe competition. When most anomalous forms now known in the world, as converted by subsidence into large separate islands, the Ornithorhynchus and Lepidosiren, which, like there will still exist many individuals of the same

102 Charles Darwin species on each island: intercrossing on the confines country undergoing modification of some kind. The of the range of each species will thus be checked: existence of such places will often depend on physi- after physical changes of any kind, immigration will cal changes, which are generally very slow, and on be prevented, so that new places in the polity of each the immigration of better adapted forms having been island will have to be filled up by modifications of checked. But the action of natural selection will prob- the old inhabitants; and time will be allowed for the ably still oftener depend on some of the inhabitants varieties in each to become well modified and per- becoming slowly modified; the mutual relations of fected. When, by renewed elevation, the islands shall many of the other inhabitants being thus disturbed. be re-converted into a continental area, there will Nothing can be effected, unless favourable variations again be severe competition: the most favoured or occur, and variation itself is apparently always a very improved varieties will be enabled to spread: there slow process. The process will often be greatly re- will be much extinction of the less improved forms, tarded by free intercrossing. Many will exclaim that and the relative proportional numbers of the vari- these several causes are amply sufficient wholly to ous inhabitants of the renewed continent will again stop the action of natural selection. I do not believe be changed; and again there will be a fair field for so. On the other hand, I do believe that natural se- natural selection to improve still further the inhabit- lection will always act very slowly, often only at long ants, and thus produce new species. intervals of time, and generally on only a very few That natural selection will always act with extreme of the inhabitants of the same region at the same time. slowness, I fully admit. Its action depends on there I further believe, that this very slow, intermittent ac- being places in the polity of nature, which can be tion of natural selection accords perfectly well with better occupied by some of the inhabitants of the what geology tells us of the rate and manner at which

103 On the Origin of Species the inhabitants of this world have changed. as geology tells us, is the precursor to extinction. We Slow though the process of selection may be, if can, also, see that any form represented by few indi- feeble man can do much by his powers of artificial viduals will, during fluctuations in the seasons or in selection, I can see no limit to the amount of change, the number of its enemies, run a good chance of utter to the beauty and infinite complexity of the coadap- extinction. But we may go further than this; for as new tations between all organic beings, one with another forms are continually and slowly being produced, and with their physical conditions of life, which may unless we believe that the number of specific forms be effected in the long course of time by nature’s goes on perpetually and almost indefinitely increas- power of selection. ing, numbers inevitably must become extinct. That the number of specific forms has not indefinitely in- Extinction. — This subject will be more fully dis- creased, geology shows us plainly; and indeed we cussed in our chapter on Geology; but it must be can see reason why they should not have thus in- here alluded to from being intimately connected with creased, for the number of places in the polity of na- natural selection. Natural selection acts solely ture is not indefinitely great,—not that we have any through the preservation of variations in some way means of knowing that any one region has as yet got advantageous, which consequently endure. But as its maximum of species. Probably no region is as yet from the high geometrical powers of increase of all fully stocked, for at the Cape of Good Hope, where organic beings, each area is already fully stocked more species of plants are crowded together than in with inhabitants, it follows that as each selected and any other quarter of the world, some foreign plants favoured form increases in number, so will the less have become naturalised, without causing, as far as favoured forms decrease and become rare. Rarity, we know, the extinction of any natives.

104 Charles Darwin Furthermore, the species which are most numer- which, from having nearly the same structure, con- ous in individuals will have the best chance of pro- stitution, and habits, generally come into the sever- ducing within any given period favourable varia- est competition with each other. Consequently, each tions. We have evidence of this, in the facts given in new variety or species, during the progress of its the second chapter, showing that it is the common formation, will generally press hardest on its near- species which afford the greatest number of recorded est kindred, and tend to exterminate them. We see varieties, or incipient species. Hence, rare species the same process of extermination amongst our do- will be less quickly modified or improved within mesticated productions, through the selection of any given period, and they will consequently be improved forms by man. Many curious instances beaten in the race for life by the modified descen- could be given showing how quickly new breeds of dants of the commoner species. cattle, sheep, and other animals, and varieties of flow- From these several considerations I think it inevi- ers, take the place of older and inferior kinds. In tably follows, that as new species in the course of Yorkshire, it is historically known that the ancient time are formed through natural selection, others will black cattle were displaced by the long-horns, and become rarer and rarer, and finally extinct. The forms that these ‘were swept away by the short-horns’ (I which stand in closest competition with those un- quote the words of an agricultural writer) ‘as if by dergoing modification and improvement, will natu- some murderous pestilence.’ rally suffer most. And we have seen in the chapter on the Struggle for Existence that it is the most Divergence of Character. — The principle, which I closely-allied forms,—varieties of the same species, have designated by this term, is of high importance and species of the same genus or of related genera,— on my theory, and explains, as I believe, several im-

105 On the Origin of Species portant facts. In the first place, varieties, even amount of difference as that between varieties of the strongly-marked ones, though having somewhat of same species and species of the same genus. the character of species—as is shown by the hope- As has always been my practice, let us seek light less doubts in many cases how to rank them—yet on this head from our domestic productions. We certainly differ from each other far less than do good shall here find something analogous. A fancier is and distinct species. Nevertheless, according to my struck by a pigeon having a slightly shorter beak; view, varieties are species in the process of forma- another fancier is struck by a pigeon having a rather tion, or are, as I have called them, incipient species. longer beak; and on the acknowledged principle that How, then, does the lesser difference between vari- ‘fanciers do not and will not admire a medium stan- eties become augmented into the greater difference dard, but like extremes,’ they both go on (as has ac- between species? That this does habitually happen, tually occurred with tumbler-pigeons) choosing and we must infer from most of the innumerable species breeding from birds with longer and longer beaks, throughout nature presenting well-marked differ- or with shorter and shorter beaks. Again, we may ences; whereas varieties, the supposed prototypes suppose that at an early period one man preferred and parents of future well-marked species, present swifter horses; another stronger and more bulky slight and ill-defined differences. Mere chance, as horses. The early differences would be very slight; we may call it, might cause one variety to differ in in the course of time, from the continued selection some character from its parents, and the offspring of of swifter horses by some breeders, and of stronger this variety again to differ from its parent in the very ones by others, the differences would become same character and in a greater degree; but this alone greater, and would be noted as forming two sub- would never account for so habitual and large an breeds; finally, after the lapse of centuries, the sub-

106 Charles Darwin breeds would become converted into two well-es- ruped, of which the number that can be supported tablished and distinct breeds. As the differences in any country has long ago arrived at its full aver- slowly become greater, the inferior animals with in- age. If its natural powers of increase be allowed to termediate characters, being neither very swift nor act, it can succeed in increasing (the country not un- very strong, will have been neglected, and will have dergoing any change in its conditions) only by its tended to disappear. Here, then, we see in man’s varying descendants seizing on places at present productions the action of what may be called the occupied by other animals: some of them, for in- principle of divergence, causing differences, at first stance, being enabled to feed on new kinds of prey, barely appreciable, steadily to increase, and the either dead or alive; some inhabiting new stations, breeds to diverge in character both from each other climbing trees, frequenting water, and some perhaps and from their common parent. becoming less carnivorous. The more diversified in But how, it may be asked, can any analogous prin- habits and structure the descendants of our carnivo- ciple apply in nature? I believe it can and does ap- rous animal became, the more places they would be ply most efficiently, from the simple circumstance enabled to occupy. What applies to one animal will that the more diversified the descendants from any apply throughout all time to all animals—that is, if one species become in structure, constitution, and they vary—for otherwise natural selection can do habits, by so much will they be better enabled to seize nothing. So it will be with plants. It has been experi- on many and widely diversified places in the polity mentally proved, that if a plot of ground be sown of nature, and so be enabled to increase in numbers. with several distinct genera of grasses, a greater num- We can clearly see this in the case of animals with ber of plants and a greater weight of dry herbage simple habits. Take the case of a carnivorous quad- can thus be raised. The same has been found to hold

107 On the Origin of Species good when first one variety and then several mixed of life can be supported by great diversification of varieties of wheat have been sown on equal spaces structure, is seen under many natural circumstances. of ground. Hence, if any one species of grass were to In an extremely small area, especially if freely open go on varying, and those varieties were continually to immigration, and where the contest between in- selected which differed from each other in at all the dividual and individual must be severe, we always same manner as distinct species and genera of find great diversity in its inhabitants. For instance, I grasses differ from each other, a greater number of found that a piece of turf, three feet by four in size, individual plants of this species of grass, including which had been exposed for many years to exactly its modified descendants, would succeed in living the same conditions, supported twenty species of on the same piece of ground. And we well know that plants, and these belonged to eighteen genera and each species and each variety of grass is annually to eight orders, which shows how much these plants sowing almost countless seeds; and thus, as it may differed from each other. So it is with the plants and be said, is striving its utmost to increase its num- insects on small and uniform islets; and so in small bers. Consequently, I cannot doubt that in the course ponds of fresh water. Farmers find that they can raise of many thousands of generations, the most distinct most food by a rotation of plants belonging to the varieties of any one species of grass would always most different orders: nature follows what may be have the best chance of succeeding and of increas- called a simultaneous rotation. Most of the animals ing in numbers, and thus of supplanting the less dis- and plants which live close round any small piece tinct varieties; and varieties, when rendered very of ground, could live on it (supposing it not to be in distinct from each other, take the rank of species. any way peculiar in its nature), and may be said to The truth of the principle, that the greatest amount be striving to the utmost to live there; but, it is seen,

108 Charles Darwin that where they come into the closest competition native genera and species, far more in new genera with each other, the advantages of diversification of than in new species. To give a single instance: in the structure, with the accompanying differences of habit last edition of Dr. Asa Gray’s ‘Manual of the Flora of and constitution, determine that the inhabitants, the Northern United States,’ 260 naturalised plants which thus jostle each other most closely, shall, as a are enumerated, and these belong to 162 genera. We general rule, belong to what we call different genera thus see that these naturalised plants are of a highly and orders. diversified nature. They differ, moreover, to a large The same principle is seen in the naturalisation of extent from the indigenes, for out of the 162 genera, plants through man’s agency in foreign lands. It no less than 100 genera are not there indigenous, and might have been expected that the plants which have thus a large proportional addition is made to the succeeded in becoming naturalised in any land genera of these States. would generally have been closely allied to the in- By considering the nature of the plants or animals digenes; for these are commonly looked at as spe- which have struggled successfully with the indi- cially created and adapted for their own country. It genes of any country, and have there become might, also, perhaps have been expected that naturalised, we can gain some crude idea in what naturalised plants would have belonged to a few manner some of the natives would have had to be groups more especially adapted to certain stations modified, in order to have gained an advantage over in their new homes. But the case is very different; the other natives; and we may, I think, at least safely and Alph. De Candolle has well remarked in his great infer that diversification of structure, amounting to and admirable work, that floras gain by new generic differences, would have been profitable naturalisation, proportionally with the number of the to them.

109 On the Origin of Species The advantage of diversification in the inhabitants we see the process of diversification in an early and of the same region is, in fact, the same as that of the incomplete stage of development. physiological division of labour in the organs of the After the foregoing discussion, which ought to have same individual body—a subject so well elucidated been much amplified, we may, I think, assume that by Milne Edwards. No physiologist doubts that a the modified descendants of any one species will stomach by being adapted to digest vegetable mat- succeed by so much the better as they become more ter alone, or flesh alone, draws most nutriment from diversified in structure, and are thus enabled to en- these substances. So in the general economy of any croach on places occupied by other beings. Now let land, the more widely and perfectly the animals and us see how this principle of great benefit being de- plants are diversified for different habits of life, so rived from divergence of character, combined with will a greater number of individuals be capable of the principles of natural selection and of extinction, there supporting themselves. A set of animals, with will tend to act. their organisation but little diversified, could hardly The accompanying diagram will aid us in under- compete with a set more perfectly diversified in standing this rather perplexing subject. Let A to L structure. It may be doubted, for instance, whether represent the species of a genus large in its own coun- the Australian marsupials, which are divided into try; these species are supposed to resemble each groups differing but little from each other, and fee- other in unequal degrees, as is so generally the case bly representing, as Mr. Waterhouse and others have in nature, and as is represented in the diagram by remarked, our carnivorous, ruminant, and rodent the letters standing at unequal distances. I have said mammals, could successfully compete with these a large genus, because we have seen in the second well-pronounced orders. In the Australian mammals, chapter, that on an average more of the species of

110 Charles Darwin large genera vary than of small genera; and the vary- selection. When a dotted line reaches one of the hori- ing species of the large genera present a greater num- zontal lines, and is there marked by a small number of varieties. We have, also, seen that the species, bered letter, a sufficient amount of variation is sup- which are the commonest and the most widely-dif- posed to have been accumulated to have formed a fused, vary more than rare species with restricted fairly well-marked variety, such as would be thought ranges. Let (A) be a common, widely-diffused, and worthy of record in a systematic work. varying species, belonging to a genus large in its own The intervals between the horizontal lines in the country. The little fan of diverging dotted lines of diagram, may represent each a thousand generations; unequal lengths proceeding from (A), may represent but it would have been better if each had represented its varying offspring. The variations are supposed ten thousand generations. After a thousand genera- to be extremely slight, but of the most diversified tions, species (A) is supposed to have produced two nature; they are not supposed all to appear simulta- fairly well-marked varieties, namely a1 and m1. neously, but often after long intervals of time; nor These two varieties will generally continue to be are they all supposed to endure for equal periods. exposed to the same conditions which made their Only those variations which are in some way profit- parents variable, and the tendency to variability is able will be preserved or naturally selected. And in itself hereditary, consequently they will tend to here the importance of the principle of benefit being vary, and generally to vary in nearly the same man- derived from divergence of character comes in; for ner as their parents varied. Moreover, these two va- this will generally lead to the most different or di- rieties, being only slightly modified forms, will tend vergent variations (represented by the outer dotted to inherit those advantages which made their com- lines) being preserved and accumulated by natural mon parent (A) more numerous than most of the

111 On the Origin of Species other inhabitants of the same country; they will like- modified descendants, proceeding from the common wise partake of those more general advantages which parent (A), will generally go on increasing in number made the genus to which the parent-species be- and diverging in character. In the diagram the prolonged, a large genus in its own country. And these cess is represented up to the ten-thousandth genera- circumstances we know to be favourable to the pro- tion, and under a condensed and simplified form up duction of new varieties. to the fourteen-thousandth generation. If, then, these two varieties be variable, the most di- But I must here remark that I do not suppose that vergent of their variations will generally be preserved the process ever goes on so regularly as is repre- during the next thousand generations. And after this sented in the diagram, though in itself made some- interval, variety a1 is supposed in the diagram to have what irregular. I am far from thinking that the most produced variety a2, which will, owing to the prin- divergent varieties will invariably prevail and mul- ciple of divergence, differ more from (A) than did tiply: a medium form may often long endure, and variety a1. Variety m1 is supposed to have produced may or may not produce more than one modified two varieties, namely m2 and s2, differing from each descendant; for natural selection will always act ac- other, and more considerably from their common cording to the nature of the places which are either parent (A). We may continue the process by similar unoccupied or not perfectly occupied by other be- steps for any length of time; some of the varieties, af- ings; and this will depend on infinitely complex re- ter each thousand generations, producing only a lations. But as a general rule, the more diversified in single variety, but in a more and more modified con- structure the descendants from any one species can dition, some producing two or three varieties, and be rendered, the more places they will be enabled to some failing to produce any. Thus the varieties or seize on, and the more their modified progeny will

112 Charles Darwin be increased. In our diagram the line of succession lines. In some cases I do not doubt that the process of is broken at regular intervals by small numbered modification will be confined to a single line of de- letters marking the successive forms which have be- scent, and the number of the descendants will not be come sufficiently distinct to be recorded as variet- increased; although the amount of divergent modifi- ies. But these breaks are imaginary, and might have cation may have been increased in the successive gen- been inserted anywhere, after intervals long enough erations. This case would be represented in the dia- to have allowed the accumulation of a considerable gram, if all the lines proceeding from (A) were re- amount of divergent variation. moved, excepting that from a1 to a10. In the same way, As all the modified descendants from a common for instance, the English race-horse and English and widely-diffused species, belonging to a large pointer have apparently both gone on slowly diverg- genus, will tend to partake of the same advantages ing in character from their original stocks, without which made their parent successful in life, they will either having given off any fresh branches or races. generally go on multiplying in number as well as After ten thousand generations, species (A) is sup- diverging in character: this is represented in the dia- posed to have produced three forms, a10, f10, and gram by the several divergent branches proceeding m10, which, from having diverged in character dur- from (A). The modified offspring from the later and ing the successive generations, will have come to more highly improved branches in the lines of de- differ largely, but perhaps unequally, from each scent, will, it is probable, often take the place of, and other and from their common parent. If we suppose so destroy, the earlier and less improved branches: the amount of change between each horizontal line this is represented in the diagram by some of the in our diagram to be excessively small, these three lower branches not reaching to the upper horizontal forms may still be only well-marked varieties; or they

113 On the Origin of Species may have arrived at the doubtful category of sub- the letters n14 to z14, are supposed to have been pro- species; but we have only to suppose the steps in duced. In each genus, the species, which are already the process of modification to be more numerous or extremely different in character, will generally tend greater in amount, to convert these three forms into to produce the greatest number of modified descen- well-defined species: thus the diagram illustrates the dants; for these will have the best chance of filling new steps by which the small differences distinguishing and widely different places in the polity of nature: varieties are increased into the larger differences dis- hence in the diagram I have chosen the extreme spe- tinguishing species. By continuing the same process cies (A), and the nearly extreme species (I), as those for a greater number of generations (as shown in the which have largely varied, and have given rise to new diagram in a condensed and simplified manner), we varieties and species. The other nine species (marked get eight species, marked by the letters between a14 by capital letters) of our original genus, may for a long and m14, all descended from (A). Thus, as I believe, period continue transmitting unaltered descendants; species are multiplied and genera are formed. and this is shown in the diagram by the dotted lines In a large genus it is probable that more than one not prolonged far upwards from want of space. species would vary. In the diagram I have assumed But during the process of modification, represented that a second species (I) has produced, by analogous in the diagram, another of our principles, namely steps, after ten thousand generations, either two well- that of extinction, will have played an important part. marked varieties (w10 and z10) or two species, ac- As in each fully stocked country natural selection cording to the amount of change supposed to be rep- necessarily acts by the selected form having some resented between the horizontal lines. After fourteen advantage in the struggle for life over other forms, thousand generations, six new species, marked by there will be a constant tendency in the improved

114 Charles Darwin descendants of any one species to supplant and ex- and (I) will have been replaced by six (n14 to z14) terminate in each stage of descent their predecessors new species. and their original parent. For it should be remem- But we may go further than this. The original spe- bered that the competition will generally be most cies of our genus were supposed to resemble each severe between those forms which are most nearly other in unequal degrees, as is so generally the case related to each other in habits, constitution, and struc- in nature; species (A) being more nearly related to ture. Hence all the intermediate forms between the B, C, and D, than to the other species; and species (I) earlier and later states, that is between the less and more to G, H, K, L, than to the others. These two more improved state of a species, as well as the origi- species (A) and (I), were also supposed to be very nal parent-species itself, will generally tend to be- common and widely diffused species, so that they come extinct. So it probably will be with many whole must originally have had some advantage over most collateral lines of descent, which will be conquered of the other species of the genus. Their modified by later and improved lines of descent. If, however, descendants, fourteen in number at the fourteen- the modified offspring of a species get into some thousandth generation, will probably have inherited distinct country, or become quickly adapted to some some of the same advantages: they have also been quite new station, in which child and parent do not modified and improved in a diversified manner at come into competition, both may continue to exist. each stage of descent, so as to have become adapted If then our diagram be assumed to represent a con- to many related places in the natural economy of siderable amount of modification, species (A) and their country. It seems, therefore, to me extremely all the earlier varieties will have become extinct, hav- probable that they will have taken the places of, and ing been replaced by eight new species (a14 to m14); thus exterminated, not only their parents (A) and (I),

115 On the Origin of Species but likewise some of the original species which were named species; and lastly, o14, e14, and m14, will most nearly related to their parents. Hence very few be nearly related one to the other, but from having of the original species will have transmitted offspring diverged at the first commencement of the process to the fourteen-thousandth generation. We may sup- of modification, will be widely different from the pose that only one (F), of the two species which were other five species, and may constitute a sub-genus least closely related to the other nine original spe- or even a distinct genus. cies, has transmitted descendants to this late stage The six descendants from (I) will form two sub- of descent. genera or even genera. But as the original species (I) The new species in our diagram descended from differed largely from (A), standing nearly at the ex- the original eleven species, will now be fifteen in treme points of the original genus, the six descen- number. Owing to the divergent tendency of natu- dants from (I) will, owing to inheritance, differ con- ral selection, the extreme amount of difference in siderably from the eight descendants from (A); the character between species a14 and z14 will be much two groups, moreover, are supposed to have gone greater than that between the most different of the on diverging in different directions. The intermedi- original eleven species. The new species, moreover, ate species, also (and this is a very important con- will be allied to each other in a widely different man- sideration), which connected the original species (A) ner. Of the eight descendants from (A) the three and (I), have all become, excepting (F), extinct, and marked a14, q14, p14, will be nearly related from have left no descendants. Hence the six new species having recently branched off from a10; b14 and f14, descended from (I), and the eight descended from from having diverged at an earlier period from a5, (A), will have to be ranked as very distinct genera, will be in some degree distinct from the three first- or even as distinct sub-families.

116 Charles Darwin Thus it is, as I believe, that two or more genera are as these two groups have gone on diverging in char- produced by descent, with modification, from two or acter from the type of their parents, the new species more species of the same genus. And the two or more (F14) will not be directly intermediate between them, parent-species are supposed to have descended from but rather between types of the two groups; and ev- some one species of an earlier genus. In our diagram, ery naturalist will be able to bring some such case this is indicated by the broken lines, beneath the capi- before his mind. tal letters, converging in sub-branches downwards In the diagram, each horizontal line has hitherto towards a single point; this point representing a single been supposed to represent a thousand generations, species, the supposed single parent of our several new but each may represent a million or hundred mil- sub-genera and genera. lion generations, and likewise a section of the suc- It is worth while to reflect for a moment on the char- cessive strata of the earth’s crust including extinct acter of the new species F14, which is supposed not remains. We shall, when we come to our chapter on to have diverged much in character, but to have re- Geology, have to refer again to this subject, and I tained the form of (F), either unaltered or altered only think we shall then see that the diagram throws light in a slight degree. In this case, its affinities to the on the affinities of extinct beings, which, though gen- other fourteen new species will be of a curious and erally belonging to the same orders, or families, or circuitous nature. Having descended from a form genera, with those now living, yet are often, in some which stood between the two parent-species (A) and degree, intermediate in character between existing (I), now supposed to be extinct and unknown, it will groups; and we can understand this fact, for the ex- be in some degree intermediate in character between tinct species lived at very ancient epochs when the the two groups descended from these species. But branching lines of descent had diverged less.

117 On the Origin of Species I see no reason to limit the process of modifica- of the larger genera which oftenest present varieties tion, as now explained, to the formation of genera or incipient species. This, indeed, might have been alone. If, in our diagram, we suppose the amount of expected; for as natural selection acts through one change represented by each successive group of di- form having some advantage over other forms in the verging dotted lines to be very great, the forms struggle for existence, it will chiefly act on those marked a14 to p14, those marked b14 and f14, and which already have some advantage; and the large- those marked o14 to m14, will form three very dis- ness of any group shows that its species have inher- tinct genera. We shall also have two very distinct ited from a common ancestor some advantage in genera descended from (I) and as these latter two common. Hence, the struggle for the production of genera, both from continued divergence of charac- new and modified descendants, will mainly lie be- ter and from inheritance from a different parent, will tween the larger groups, which are all trying to in- differ widely from the three genera descended from crease in number. One large group will slowly con- (A), the two little groups of genera will form two quer another large group, reduce its numbers, and distinct families, or even orders, according to the thus lessen its chance of further variation and im- amount of divergent modification supposed to be provement. Within the same large group, the later represented in the diagram. And the two new fami- and more highly perfected sub-groups, from branch- lies, or orders, will have descended from two spe- ing out and seizing on many new places in the pol- cies of the original genus; and these two species are ity of Nature, will constantly tend to supplant and supposed to have descended from one species of a destroy the earlier and less improved sub-groups. still more ancient and unknown genus. Small and broken groups and sub-groups will finally We have seen that in each country it is the species tend to disappear. Looking to the future, we can pre-

118 Charles Darwin dict that the groups of organic beings which are now animal and vegetable kingdoms. Although extremely large and triumphant, and which are least broken few of the most ancient species may now have living up, that is, which as yet have suffered least extinc- and modified descendants, yet at the most remote tion, will for a long period continue to increase. But geological period, the earth may have been as well which groups will ultimately prevail, no man can peopled with many species of many genera, families, predict; for we well know that many groups, for- orders, and classes, as at the present day. merly most extensively developed, have now become extinct. Looking still more remotely to the fu- Summary of Chapter — If during the long course of ture, we may predict that, owing to the continued ages and under varying conditions of life, organic and steady increase of the larger groups, a multi- beings vary at all in the several parts of their tude of smaller groups will become utterly extinct, organisation, and I think this cannot be disputed; if and leave no modified descendants; and conse- there be, owing to the high geometrical powers of quently that of the species living at any one period, increase of each species, at some age, season, or year, extremely few will transmit descendants to a remote a severe struggle for life, and this certainly cannot futurity. I shall have to return to this subject in the be disputed; then, considering the infinite complex- chapter on Classification, but I may add that on this ity of the relations of all organic beings to each other view of extremely few of the more ancient species and to their conditions of existence, causing an infi- having transmitted descendants, and on the view of nite diversity in structure, constitution, and habits, all the descendants of the same species making a to be advantageous to them, I think it would be a class, we can understand how it is that there exist most extraordinary fact if no variation ever had oc- but very few classes in each main division of the curred useful to each being’s own welfare, in the

119 On the Origin of Species same way as so many variations have occurred use- evidence given in the following chapters. But we al- ful to man. But if variations useful to any organic ready see how it entails extinction; and how largely being do occur, assuredly individuals thus extinction has acted in the world’s history, geology characterised will have the best chance of being pre- plainly declares. Natural selection, also, leads to di- served in the struggle for life; and from the strong vergence of character; for more living beings can be principle of inheritance they will tend to produce supported on the same area the more they diverge offspring similarly characterised. This principle of in structure, habits, and constitution, of which we preservation, I have called, for the sake of brevity, see proof by looking at the inhabitants of any small Natural Selection. Natural selection, on the principle spot or at naturalised productions. Therefore dur- of qualities being inherited at corresponding ages, ing the modification of the descendants of any one can modify the egg, seed, or young, as easily as the species, and during the incessant struggle of all spe- adult. Amongst many animals, sexual selection will cies to increase in numbers, the more diversified give its aid to ordinary selection, by assuring to the these descendants become, the better will be their most vigorous and best adapted males the greatest chance of succeeding in the battle of life. Thus the number of offspring. Sexual selection will also give small differences distinguishing varieties of the same characters useful to the males alone, in their struggles species, will steadily tend to increase till they come with other males. to equal the greater differences between species of Whether natural selection has really thus acted in the same genus, or even of distinct genera. nature, in modifying and adapting the various forms We have seen that it is the common, the widely- of life to their several conditions and stations, must diffused, and widely-ranging species, belonging to be judged of by the general tenour and balance of the larger genera, which vary most; and these will

120 Charles Darwin tend to transmit to their modified offspring that su- points, and these round other points, and so on in periority which now makes them dominant in their almost endless cycles. On the view that each species own countries. Natural selection, as has just been has been independently created, I can see no expla- remarked, leads to divergence of character and to nation of this great fact in the classification of all or- much extinction of the less improved and interme- ganic beings; but, to the best of my judgment, it is diate forms of life. On these principles, I believe, the explained through inheritance and the complex ac- nature of the affinities of all organic beings may be tion of natural selection, entailing extinction and di- explained. It is a truly wonderful fact—the wonder vergence of character, as we have seen illustrated in of which we are apt to overlook from familiarity— the diagram. that all animals and all plants throughout all time The affinities of all the beings of the same class have and space should be related to each other in group sometimes been represented by a great tree. I be- subordinate to group, in the manner which we ev- lieve this simile largely speaks the truth. The green erywhere behold—namely, varieties of the same spe- and budding twigs may represent existing species; cies most closely related together, species of the same and those produced during each former year may genus less closely and unequally related together, represent the long succession of extinct species. At forming sections and sub-genera, species of distinct each period of growth all the growing twigs have genera much less closely related, and genera related tried to branch out on all sides, and to overtop and in different degrees, forming sub-families, families, kill the surrounding twigs and branches, in the same orders, sub-classes, and classes. The several subor- manner as species and groups of species have tried dinate groups in any class cannot be ranked in a to overmaster other species in the great battle for life. single file, but seem rather to be clustered round The limbs divided into great branches, and these into

121 On the Origin of Species lesser and lesser branches, were themselves once, Ornithorhynchus or Lepidosiren, which in some when the tree was small, budding twigs; and this small degree connects by its affinities two large connexion of the former and present buds by rami- branches of life, and which has apparently been fying branches may well represent the classification saved from fatal competition by having inhabited a of all extinct and living species in groups subordi- protected station. As buds give rise by growth to nate to groups. Of the many twigs which flourished fresh buds, and these, if vigorous, branch out and when the tree was a mere bush, only two or three, overtop on all sides many a feebler branch, so by now grown into great branches, yet survive and bear generation I believe it has been with the great Tree all the other branches; so with the species which lived of Life, which fills with its dead and broken branches during long-past geological periods, very few now the crust of the earth, and covers the surface with its have living and modified descendants. From the first ever branching and beautiful ramifications. growth of the tree, many a limb and branch has de- cayed and dropped off; and these lost branches of various sizes may represent those whole orders, families, and genera which have now no living representatives, and which are known to us only from having been found in a fossil state. As we here and there see a thin straggling branch springing from a fork low down in a tree, and which by some chance has been favoured and is still alive on its summit, so we occasionally see an animal like the

122 Charles Darwin Chapter V serves to acknowledge plainly our ignorance of the cause of each particular variation. Some authors be- Laws of Variation lieve it to be as much the function of the reproductive system to produce individual differences, or very Effects of external conditions — Use and disuse, com- slight deviations of structure, as to make the child like bined with natural selection; organs of flight and of its parents. But the much greater variability, as well vision — Acclimatisation — Correlation of growth as the greater frequency of monstrosities, under do- — Compensation and economy of growth — False mestication or cultivation, than under nature, leads correlations —Multiple, rudimentary, and lowly me to believe that deviations of structure are in some organised structures variable — Parts developed in way due to the nature of the conditions of life, to which an unusual manner are highly variable: specific char- the parents and their more remote ancestors have been acters more variable than generic: secondary sexual exposed during several generations. I have remarked characters variable —Species of the same genus vary in the first chapter—but a long catalogue of facts which in an analogous manner — Reversions to long lost cannot be here given would be necessary to show the characters — Summary. truth of the remark—that the reproductive system is eminently susceptible to changes in the conditions of I HAVE HITHERTO sometimes spoken as if the varia- life; and to this system being functionally disturbed tions—so common and multiform in organic beings in the parents, I chiefly attribute the varying or plastic under domestication, and in a lesser degree in those condition of the offspring. The male and female sexual in a state of nature—had been due to chance. This, elements seem to be affected before that union takes of course, is a wholly incorrect expression, but it place which is to form a new being. In the case of ‘sport-

123 On the Origin of Species ing’ plants, the bud, which in its earliest condition does or from greater depths. Gould believes that birds of not apparently differ essentially from an ovule, is alone the same species are more brightly coloured under a affected. But why, because the reproductive system is clear atmosphere, than when living on islands or near disturbed, this or that part should vary more or less, the coast. So with insects, Wollaston is convinced that we are profoundly ignorant. Nevertheless, we can here residence near the sea affects their colours. Moquin- and there dimly catch a faint ray of light, and we may Tandon gives a list of plants which when growing near feel sure that there must be some cause for each devia- the sea-shore have their leaves in some degree fleshy, tion of structure, however slight. though not elsewhere fleshy. Several other such cases How much direct effect difference of climate, food, could be given. &c., produces on any being is extremely doubtful. My The fact of varieties of one species, when they range impression is, that the effect is extremely small in the into the zone of habitation of other species, often ac- case of animals, but perhaps rather more in that of quiring in a very slight degree some of the charac- plants. We may, at least, safely conclude that such ters of such species, accords with our view that spe- influences cannot have produced the many striking cies of all kinds are only well-marked and perma- and complex co-adaptations of structure between one nent varieties. Thus the species of shells which are organic being and another, which we see everywhere confined to tropical and shallow seas are generally throughout nature. Some little influence may be at- brighter-coloured than those confined to cold and tributed to climate, food, &c.: thus, E. Forbes speaks deeper seas. The birds which are confined to conti- confidently that shells at their southern limit, and nents are, according to Mr. Gould, brighter-coloured when living in shallow water, are more brightly than those of islands. The insect-species confined to coloured than those of the same species further north sea-coasts, as every collector knows, are often brassy

124 Charles Darwin or lurid. Plants which live exclusively on the sea- produced under conditions of life as different as can side are very apt to have fleshy leaves. He who be- well be conceived; and, on the other hand, of differ- lieves in the creation of each species, will have to ent varieties being produced from the same species say that this shell, for instance, was created with under the same conditions. Such facts show how in- bright colours for a warm sea; but that this other shell directly the conditions of life must act. Again, innu- became bright-coloured by variation when it ranged merable instances are known to every naturalist of into warmer or shallower waters. species keeping true, or not varying at all, although When a variation is of the slightest use to a being, living under the most opposite climates. Such con- we cannot tell how much of it to attribute to the ac- siderations as these incline me to lay very little cumulative action of natural selection, and how weight on the direct action of the conditions of life. much to the conditions of life. Thus, it is well known Indirectly, as already remarked, they seem to play to furriers that animals of the same species have an important part in affecting the reproductive sys- thicker and better fur the more severe the climate is tem, and in thus inducing variability; and natural under which they have lived; but who can tell how selection will then accumulate all profitable varia- much of this difference may be due to the warmest- tions, however slight, until they become plainly de- clad individuals having been favoured and pre- veloped and appreciable by us. served during many generations, and how much to the direct action of the severe climate? for it would Effects of Use and Disuse. — From the facts alluded appear that climate has some direct action on the hair to in the first chapter, I think there can be little doubt of our domestic quadrupeds. that use in our domestic animals strengthens and Instances could be given of the same variety being enlarges certain parts, and disuse diminishes them;

125 On the Origin of Species and that such modifications are inherited. Under free progenitor of the ostrich had habits like those of a nature, we can have no standard of comparison, by bustard, and that as natural selection increased in which to judge of the effects of long-continued use successive generations the size and weight of its or disuse, for we know not the parent-forms; but body, its legs were used more, and its wings less, many animals have structures which can be ex- until they became incapable of flight. plained by the effects of disuse. As Professor Owen Kirby has remarked (and I have observed the same has remarked, there is no greater anomaly in nature fact) that the anterior tarsi, or feet, of many male than a bird that cannot fly; yet there are several in dung-feeding beetles are very often broken off; he this state. The logger-headed duck of South America examined seventeen specimens in his own collec- can only flap along the surface of the water, and has tion, and not one had even a relic left. In the Onites its wings in nearly the same condition as the domes- apelles the tarsi are so habitually lost, that the insect tic Aylesbury duck. As the larger ground-feeding has been described as not having them. In some other birds seldom take flight except to escape danger, I genera they are present, but in a rudimentary condi- believe that the nearly wingless condition of several tion. In the Ateuchus or sacred beetle of the Egyp- birds, which now inhabit or have lately inhabited tians, they are totally deficient. There is not suffi- several oceanic islands, tenanted by no beast of prey, cient evidence to induce us to believe that mutila- has been caused by disuse. The ostrich indeed in- tions are ever inherited; and I should prefer explain- habits continents and is exposed to danger from ing the entire absence of the anterior tarsi in which it cannot escape by flight, but by kicking it Ateuchus, and their rudimentary condition in some can defend itself from enemies, as well as any of the other genera, by the long-continued effects of dis- smaller quadrupeds. We may imagine that the early use in their progenitors; for as the tarsi are almost

126 Charles Darwin always lost in many dung-feeding beetles, they must have habits of life almost necessitating frequent be lost early in life, and therefore cannot be much flight;—these several considerations have made me used by these insects. believe that the wingless condition of so many Ma- In some cases we might easily put down to disuse deira beetles is mainly due to the action of natural modifications of structure which are wholly, or selection, but combined probably with disuse. For mainly, due to natural selection. Mr. Wollaston has during thousands of successive generations each discovered the remarkable fact that 200 beetles, out individual beetle which flew least, either from its of the 550 species inhabiting Madeira, are so far de- wings having been ever so little less perfectly deficient in wings that they cannot fly; and that of the veloped or from indolent habit, will have had the twenty-nine endemic genera, no less than twenty- best chance of surviving from not being blown out three genera have all their species in this condition! to sea; and, on the other hand, those beetles which Several facts, namely, that beetles in many parts of most readily took to flight will oftenest have been the world are very frequently blown to sea and per- blown to sea and thus have been destroyed. ish; that the beetles in Madeira, as observed by Mr. The insects in Madeira which are not ground-feed- Wollaston, lie much concealed, until the wind lulls ers, and which, as the flower-feeding coleoptera and and the sun shines; that the proportion of wingless lepidoptera, must habitually use their wings to gain beetles is larger on the exposed Dezertas than in their subsistence, have, as Mr. Wollaston suspects, Madeira itself; and especially the extraordinary fact, their wings not at all reduced, but even enlarged. so strongly insisted on by Mr. Wollaston, of the al- This is quite compatible with the action of natural most entire absence of certain large groups of beetles, selection. For when a new insect first arrived on the elsewhere excessively numerous, and which groups island, the tendency of natural selection to enlarge

127 On the Origin of Species or to reduce the wings, would depend on whether a As frequent inflammation of the eyes must be injuri- greater number of individuals were saved by suc- ous to any animal, and as eyes are certainly not in- cessfully battling with the winds, or by giving up dispensable to animals with subterranean habits, a the attempt and rarely or never flying. As with mari- reduction in their size with the adhesion of the eye- ners shipwrecked near a coast, it would have been lids and growth of fur over them, might in such case better for the good swimmers if they had been able be an advantage; and if so, natural selection would to swim still further, whereas it would have been constantly aid the effects of disuse. better for the bad swimmers if they had not been able It is well known that several animals, belonging to to swim at all and had stuck to the wreck. the most different classes, which inhabit the caves of The eyes of moles and of some burrowing rodents Styria and of Kentucky, are blind. In some of the crabs are rudimentary in size, and in some cases are quite the foot-stalk for the eye remains, though the eye is covered up by skin and fur. This state of the eyes is gone; the stand for the telescope is there, though the probably due to gradual reduction from disuse, but telescope with its glasses has been lost. As it is diffi- aided perhaps by natural selection. In South cult to imagine that eyes, though useless, could be America, a burrowing rodent, the tuco-tuco, or in any way injurious to animals living in darkness, I Ctenomys, is even more subterranean in its habits attribute their loss wholly to disuse. In one of the than the mole; and I was assured by a Spaniard, who blind animals, namely, the cave-rat, the eyes are of had often caught them, that they were frequently immense size; and Professor Silliman thought that it blind; one which I kept alive was certainly in this regained, after living some days in the light, some condition, the cause, as appeared on dissection, hav- slight power of vision. In the same manner as in ing been inflammation of the nictitating membrane. Madeira the wings of some of the insects have been

128 Charles Darwin enlarged, and the wings of others have been reduced world into the deeper and deeper recesses of the by natural selection aided by use and disuse, so in Kentucky caves, as did European animals into the the case of the cave-rat natural selection seems to caves of Europe. We have some evidence of this gra- have struggled with the loss of light and to have in- dation of habit; for, as Schiodte remarks, ‘animals creased the size of the eyes; whereas with all the other not far remote from ordinary forms, prepare the tran- inhabitants of the caves, disuse by itself seems to sition from light to darkness. Next follow those that have done its work. are constructed for twilight; and, last of all, those It is difficult to imagine conditions of life more simi- destined for total darkness.’ By the time that an ani- lar than deep limestone caverns under a nearly simi- mal had reached, after numberless generations, the lar climate; so that on the common view of the blind deepest recesses, disuse will on this view have more animals having been separately created for the or less perfectly obliterated its eyes, and natural se- American and European caverns, close similarity in lection will often have effected other changes, such their organisation and affinities might have been as an increase in the length of the antennae or palpi, expected; but, as Schiodte and others have remarked, as a compensation for blindness. Notwithstanding this is not the case, and the cave-insects of the two such modifications, we might expect still to see in continents are not more closely allied than might the cave-animals of America, affinities to the other have been anticipated from the general resemblance inhabitants of that continent, and in those of Europe, of the other inhabitants of North America and Eu- to the inhabitants of the European continent. And rope. On my view we must suppose that American this is the case with some of the American cave-ani- animals, having ordinary powers of vision, slowly mals, as I hear from Professor Dana; and some of the migrated by successive generations from the outer European cave-insects are very closely allied to those

129 On the Origin of Species of the surrounding country. It would be most diffi- acclimatisation. As it is extremely common for spe- cult to give any rational explanation of the affinities cies of the same genus to inhabit very hot and very of the blind cave-animals to the other inhabitants of cold countries, and as I believe that all the species of the two continents on the ordinary view of their in- the same genus have descended from a single par- dependent creation. That several of the inhabitants ent, if this view be correct, acclimatisation must be of the caves of the Old and New Worlds should be readily effected during long-continued descent. It is closely related, we might expect from the well- notorious that each species is adapted to the climate known relationship of most of their other produc- of its own home: species from an arctic or even from tions. Far from feeling any surprise that some of the a temperate region cannot endure a tropical climate, cave-animals should be very anomalous, as Agassiz or conversely. So again, many succulent plants can- has remarked in regard to the blind fish, the not endure a damp climate. But the degree of adap- Amblyopsis, and as is the case with the blind Pro- tation of species to the climates under which they teus with reference to the reptiles of Europe, I am live is often overrated. We may infer this from our only surprised that more wrecks of ancient life have frequent inability to predict whether or not an im- not been preserved, owing to the less severe compe- ported plant will endure our climate, and from the tition to which the inhabitants of these dark abodes number of plants and animals brought from warmer will probably have been exposed. countries which here enjoy good health. We have Acclimatisation. — Habit is hereditary with plants, reason to believe that species in a state of nature are as in the period of flowering, in the amount of rain limited in their ranges by the competition of other requisite for seeds to germinate, in the time of sleep, organic beings quite as much as, or more than, by &c., and this leads me to say a few words on adaptation to particular climates. But whether or not

130 Charles Darwin the adaptation be generally very close, we have evi- As I believe that our domestic animals were origi- dence, in the case of some few plants, of their be- nally chosen by uncivilised man because they were coming, to a certain extent, naturally habituated to useful and bred readily under confinement, and not different temperatures, or becoming acclimatised: because they were subsequently found capable of thus the pines and rhododendrons, raised from seed far-extended transportation, I think the common and collected by Dr. Hooker from trees growing at dif- extraordinary capacity in our domestic animals of ferent heights on the Himalaya, were found in this not only withstanding the most different climates but country to possess different constitutional powers of being perfectly fertile (a far severer test) under of resisting cold. Mr. Thwaites informs me that he them, may be used as an argument that a large pro- has observed similar facts in Ceylon, and analogous portion of other animals, now in a state of nature, observations have been made by Mr. H. C. Watson could easily be brought to bear widely different cli- on European species of plants brought from the mates. We must not, however, push the foregoing Azores to England. In regard to animals, several au- argument too far, on account of the probable origin thentic cases could be given of species within his- of some of our domestic animals from several wild torical times having largely extended their range stocks: the blood, for instance, of a tropical and arc- from warmer to cooler latitudes, and conversely; but tic wolf or wild dog may perhaps be mingled in our we do not positively know that these animals were domestic breeds. The rat and mouse cannot be con- strictly adapted to their native climate, but in all or- sidered as domestic animals, but they have been dinary cases we assume such to be the case; nor do transported by man to many parts of the world, and we know that they have subsequently become now have a far wider range than any other rodent, acclimatised to their new homes. living free under the cold climate of Faroe in the north

131 On the Origin of Species and of the Falklands in the south, and on many is- ricultural works, even in the ancient Encyclopaedias lands in the torrid zones. Hence I am inclined to look of China, to be very cautious in transposing animals at adaptation to any special climate as a quality from one district to another; for it is not likely that readily grafted on an innate wide flexibility of con- man should have succeeded in selecting so many stitution, which is common to most animals. On this breeds and sub-breeds with constitutions specially view, the capacity of enduring the most different cli- fitted for their own districts: the result must, I think, mates by man himself and by his domestic animals, be due to habit. On the other hand, I can see no rea- and such facts as that former species of the elephant son to doubt that natural selection will continually and rhinoceros were capable of enduring a glacial tend to preserve those individuals which are born climate, whereas the living species are now all tropi- with constitutions best adapted to their native coun- cal or sub-tropical in their habits, ought not to be tries. In treatises on many kinds of cultivated plants, looked at as anomalies, but merely as examples of a certain varieties are said to withstand certain climates very common flexibility of constitution, brought, better than others: this is very strikingly shown in under peculiar circumstances, into play. works on fruit trees published in the United States, How much of the acclimatisation of species to any in which certain varieties are habitually recom- peculiar climate is due to mere habit, and how much mended for the northern, and others for the south- to the natural selection of varieties having different ern States; and as most of these varieties are of re- innate constitutions, and how much to both means cent origin, they cannot owe their constitutional dif- combined, is a very obscure question. That habit or ferences to habit. The case of the Jerusalem artichoke, custom has some influence I must believe, both from which is never propagated by seed, and of which analogy, and from the incessant advice given in ag- consequently new varieties have not been produced,

132 Charles Darwin has even been advanced—for it is now as tender as natural selection of innate differences. ever it was—as proving that acclimatisation cannot Correlation of Growth. — I mean by this expres- be effected! The case, also, of the kidney-bean has sion that the whole organisation is so tied together been often cited for a similar purpose, and with much during its growth and development, that when slight greater weight; but until some one will sow, during variations in any one part occur, and are accumu- a score of generations, his kidney-beans so early that lated through natural selection, other parts become a very large proportion are destroyed by frost, and modified. This is a very important subject, most im- then collect seed from the few survivors, with care perfectly understood. The most obvious case is, that to prevent accidental crosses, and then again get seed modifications accumulated solely for the good of the from these seedlings, with the same precautions, the young or larva, will, it may safely be concluded, af- experiment cannot be said to have been even tried. fect the structure of the adult; in the same manner as Nor let it be supposed that no differences in the con- any malconformation affecting the early embryo, stitution of seedling kidney-beans ever appear, for seriously affects the whole organisation of the adult. an account has been published how much more The several parts of the body which are homologous, hardy some seedlings appeared to be than others. and which, at an early embryonic period, are alike, On the whole, I think we may conclude that habit, seem liable to vary in an allied manner: we see this use, and disuse, have, in some cases, played a con- in the right and left sides of the body varying in the siderable part in the modification of the constitution, same manner; in the front and hind legs, and even in and of the structure of various organs; but that the the jaws and limbs, varying together, for the lower effects of use and disuse have often been largely com- jaw is believed to be homologous with the limbs. bined with, and sometimes overmastered by, the These tendencies, I do not doubt, may be mastered

133 On the Origin of Species more or less completely by natural selection: thus a quently quite obscure. M. Is. Geoffroy St. Hilaire has family of stags once existed with an antler only on forcibly remarked, that certain malconformations one side; and if this had been of any great use to the very frequently, and that others rarely coexist, with- breed it might probably have been rendered perma- out our being able to assign any reason. What can be nent by natural selection. more singular than the relation between blue eyes Homologous parts, as has been remarked by some and deafness in cats, and the tortoise-shell colour authors, tend to cohere; this is often seen in mon- with the female sex; the feathered feet and skin be- strous plants; and nothing is more common than the tween the outer toes in pigeons, and the presence of union of homologous parts in normal structures, as more or less down on the young birds when first the union of the petals of the corolla into a tube. Hard hatched, with the future colour of their plumage; or, parts seem to affect the form of adjoining soft parts; again, the relation between the hair and teeth in the it is believed by some authors that the diversity in naked Turkish dog, though here probably homol- the shape of the pelvis in birds causes the remark- ogy comes into play? With respect to this latter case able diversity in the shape of their kidneys. Others of correlation, I think it can hardly be accidental, that believe that the shape of the pelvis in the human if we pick out the two orders of mammalia which mother influences by pressure the shape of the head are most abnormal in their dermal coverings, viz. of the child. In snakes, according to Schlegel, the Cetacea (whales) and Edentata (armadilloes, scaly shape of the body and the manner of swallowing ant-eaters, &c.), that these are likewise the most ab- determine the position of several of the most impor- normal in their teeth. tant viscera. I know of no case better adapted to show the im- The nature of the bond of correlation is very fre- portance of the laws of correlation in modifying im-

134 Charles Darwin portant structures, independently of utility and, seeds of the outer and inner florets without any dif- therefore, of natural selection, than that of the differ- ference in the corolla. Possibly, these several difference between the outer and inner flowers in some ences may be connected with some difference in the Compositous and Umbelliferous plants. Every one flow of nutriment towards the central and external knows the difference in the ray and central florets of, flowers: we know, at least, that in irregular flowers, for instance, the daisy, and this difference is often those nearest to the axis are oftenest subject to accompanied with the abortion of parts of the flower. peloria, and become regular. I may add, as an in- But, in some Compositous plants, the seeds also dif- stance of this, and of a striking case of correlation, fer in shape and sculpture; and even the ovary itself, that I have recently observed in some garden pelar- with its accessory parts, differs, as has been described goniums, that the central flower of the truss often by Cassini. These differences have been attributed loses the patches of darker colour in the two upper by some authors to pressure, and the shape of the petals; and that when this occurs, the adherent seeds in the ray-florets in some Compositae counte- nectary is quite aborted; when the colour is absent nances this idea; but, in the case of the corolla of the from only one of the two upper petals, the nectary is Umbelliferae, it is by no means, as Dr. Hooker in- only much shortened. forms me, in species with the densest heads that the With respect to the difference in the corolla of the inner and outer flowers most frequently differ. It central and exterior flowers of a head or umbel, I do might have been thought that the development of not feel at all sure that C. C. Sprengel’s idea that the the ray-petals by drawing nourishment from certain ray-florets serve to attract insects, whose agency is other parts of the flower had caused their abortion; highly advantageous in the fertilisation of plants of but in some Compositae there is a difference in the these two orders, is so far-fetched, as it may at first

135 On the Origin of Species appear: and if it be advantageous, natural selection acquired through natural selection some one modi- may have come into play. But in regard to the differ- fication in structure, and, after thousands of gen- ences both in the internal and external structure of erations, some other and independent modification; the seeds, which are not always correlated with any and these two modifications, having been transmit- differences in the flowers, it seems impossible that ted to a whole group of descendants with diverse they can be in any way advantageous to the plant: habits, would naturally be thought to be correlated yet in the Umbelliferae these differences are of such in some necessary manner. So, again, I do not doubt apparent importance—the seeds being in some cases, that some apparent correlations, occurring through- according to Tausch, orthospermous in the exterior out whole orders, are entirely due to the manner flowers and coelospermous in the central flowers,— alone in which natural selection can act. For in- that the elder De Candolle founded his main divi- stance, Alph. De Candolle has remarked that sions of the order on analogous differences. Hence winged seeds are never found in fruits which do we see that modifications of structure, viewed by not open: I should explain the rule by the fact that systematists as of high value, may be wholly due to seeds could not gradually become winged through unknown laws of correlated growth, and without natural selection, except in fruits which opened; so being, as far as we can see, of the slightest service to that the individual plants producing seeds which the species. were a little better fitted to be wafted further, might We may often falsely attribute to correlation of get an advantage over those producing seed less growth, structures which are common to whole fitted for dispersal; and this process could not pos- groups of species, and which in truth are simply due sibly go on in fruit which did not open. to inheritance; for an ancient progenitor may have The elder Geoffroy and Goethe propounded, at

136 Charles Darwin about the same period, their law of compensation or hand, of a part being largely developed through balancement of growth; or, as Goethe expressed it, natural selection and another and adjoining part be- ‘in order to spend on one side, nature is forced to ing reduced by this same process or by disuse, and, economise on the other side.’ I think this holds true on the other hand, the actual withdrawal of nutri- to a certain extent with our domestic productions: if ment from one part owing to the excess of growth in nourishment flows to one part or organ in excess, it another and adjoining part. rarely flows, at least in excess, to another part; thus I suspect, also, that some of the cases of compensa- it is difficult to get a cow to give much milk and to tion which have been advanced, and likewise some fatten readily. The same varieties of the cabbage do other facts, may be merged under a more general not yield abundant and nutritious foliage and a co- principle, namely, that natural selection is continu- pious supply of oil-bearing seeds. When the seeds ally trying to economise in every part of the in our fruits become atrophied, the fruit itself gains organisation. If under changed conditions of life a largely in size and quality. In our poultry, a large structure before useful becomes less useful, any tuft of feathers on the head is generally accompa- diminution, however slight, in its development, will nied by a diminished comb, and a large beard by be seized on by natural selection, for it will profit diminished wattles. With species in a state of nature the individual not to have its nutriment wasted in it can hardly be maintained that the law is of univer- building up an useless structure. I can thus only un- sal application; but many good observers, more es- derstand a fact with which I was much struck when pecially botanists, believe in its truth. I will not, how- examining cirripedes, and of which many other in- ever, here give any instances, for I see hardly any stances could be given: namely, that when a cirripede way of distinguishing between the effects, on the one is parasitic within another and is thus protected, it

137 On the Origin of Species loses more or less completely its own shell or cara- part of the organisation, as soon as it is rendered pace. This is the case with the male Ibla, and in a superfluous, without by any means causing some truly extraordinary manner with the Proteolepas: for other part to be largely developed in a correspond- the carapace in all other cirripedes consists of the ing degree. And, conversely, that natural selection three highly-important anterior segments of the head may perfectly well succeed in largely developing enormously developed, and furnished with great any organ, without requiring as a necessary compen- nerves and muscles; but in the parasitic and pro- sation the reduction of some adjoining part. tected Proteolepas, the whole anterior part of the It seems to be a rule, as remarked by Is. Geoffroy head is reduced to the merest rudiment attached to St. Hilaire, both in varieties and in species, that when the bases of the prehensile antennae. Now the sav- any part or organ is repeated many times in the struc- ing of a large and complex structure, when rendered ture of the same individual (as the vertebrae in superfluous by the parasitic habits of the snakes, and the stamens in polyandrous flowers) the Proteolepas, though effected by slow steps, would number is variable; whereas the number of the same be a decided advantage to each successive indi- part or organ, when it occurs in lesser numbers, is vidual of the species; for in the struggle for life to constant. The same author and some botanists have which every animal is exposed, each individual further remarked that multiple parts are also very Proteolepas would have a better chance of support- liable to variation in structure. Inasmuch as this ‘veg- ing itself, by less nutriment being wasted in devel- etative repetition,’ to use Prof. Owen’s expression, oping a structure now become useless. seems to be a sign of low organisation; the forego- Thus, as I believe, natural selection will always suc- ing remark seems connected with the very general ceed in the long run in reducing and saving every opinion of naturalists, that beings low in the scale of

138 Charles Darwin nature are more variable than those which are higher. having no power to check deviations in their struc- I presume that lowness in this case means that the ture. Thus rudimentary parts are left to the free play several parts of the organisation have been but little of the various laws of growth, to the effects of long- specialised for particular functions; and as long as continued disuse, and to the tendency to reversion. the same part has to perform diversified work, we A part developed in any species in an extraordi- can perhaps see why it should remain variable, that nary degree or manner, in comparison with the same is, why natural selection should have preserved or part in allied species, tends to be highly variable. — rejected each little deviation of form less carefully Several years ago I was much struck with a remark, than when the part has to serve for one special pur- nearly to the above effect, published by Mr. pose alone. In the same way that a knife which has Waterhouse. I infer also from an observation made to cut all sorts of things may be of almost any shape; by Professor Owen, with respect to the length of the whilst a tool for some particular object had better be arms of the ourang-outang, that he has come to a of some particular shape. Natural selection, it should nearly similar conclusion. It is hopeless to attempt never be forgotten, can act on each part of each be- to convince any one of the truth of this proposition ing, solely through and for its advantage. without giving the long array of facts which I have Rudimentary parts, it has been stated by some au- collected, and which cannot possibly be here intro- thors, and I believe with truth, are apt to be highly duced. I can only state my conviction that it is a rule variable. We shall have to recur to the general subject of high generality. I am aware of several causes of of rudimentary and aborted organs; and I will here error, but I hope that I have made due allowance for only add that their variability seems to be owing to them. It should be understood that the rule by no their uselessness, and therefore to natural selection means applies to any part, however unusually de-

139 On the Origin of Species veloped, unless it be unusually developed in com- doubt. But that our rule is not confined to secondary parison with the same part in closely allied species. sexual characters is clearly shown in the case of her- Thus, the bat’s wing is a most abnormal structure in maphrodite cirripedes; and I may here add, that I par- the class mammalia; but the rule would not here ticularly attended to Mr. Waterhouse’s remark, whilst apply, because there is a whole group of bats hav- investigating this Order, and I am fully convinced that ing wings; it would apply only if some one species the rule almost invariably holds good with cirripedes. of bat had its wings developed in some remarkable I shall, in my future work, give a list of the more re- manner in comparison with the other species of the markable cases; I will here only briefly give one, as it same genus. The rule applies very strongly in the illustrates the rule in its largest application. The oper- case of secondary sexual characters, when displayed cular valves of sessile cirripedes (rock barnacles) are, in any unusual manner. The term, secondary sexual in every sense of the word, very important structures, characters, used by Hunter, applies to characters and they differ extremely little even in different gen- which are attached to one sex, but are not directly era; but in the several species of one genus, Pyrgoma, connected with the act of reproduction. The rule ap- these valves present a marvellous amount of diversi- plies to males and females; but as females more fication: the homologous valves in the different spe- rarely offer remarkable secondary sexual characters, cies being sometimes wholly unlike in shape; and the it applies more rarely to them. The rule being so amount of variation in the individuals of several of plainly applicable in the case of secondary sexual the species is so great, that it is no exaggeration to characters, may be due to the great variability of these state that the varieties differ more from each other in characters, whether or not displayed in any unusual the characters of these important valves than do other manner—of which fact I think there can be little species of distinct genera.

140 Charles Darwin As birds within the same country vary in a remark- in the Dorking fowl) or the whole breed will cease to ably small degree, I have particularly attended to have a nearly uniform character. The breed will then them, and the rule seems to me certainly to hold good be said to have degenerated. In rudimentary organs, in this class. I cannot make out that it applies to and in those which have been but little specialised plants, and this would seriously have shaken my for any particular purpose, and perhaps in polymor- belief in its truth, had not the great variability in phic groups, we see a nearly parallel natural case; plants made it particularly difficult to compare their for in such cases natural selection either has not or relative degrees of variability. cannot come into full play, and thus the organisation When we see any part or organ developed in a re- is left in a fluctuating condition. But what here more markable degree or manner in any species, the fair especially concerns us is, that in our domestic ani- presumption is that it is of high importance to that mals those points, which at the present time are un- species; nevertheless the part in this case is eminently dergoing rapid change by continued selection, are liable to variation. Why should this be so? On the also eminently liable to variation. Look at the breeds view that each species has been independently cre- of the pigeon; see what a prodigious amount of dif- ated, with all its parts as we now see them, I can see ference there is in the beak of the different tumblers, no explanation. But on the view that groups of spe- in the beak and wattle of the different carriers, in the cies have descended from other species, and have carriage and tail of our fantails, &c., these being the been modified through natural selection, I think we points now mainly attended to by English fanciers. can obtain some light. In our domestic animals, if Even in the sub-breeds, as in the short-faced tum- any part, or the whole animal, be neglected and no bler, it is notoriously difficult to breed them nearly selection be applied, that part (for instance, the comb to perfection, and frequently individuals are born

141 On the Origin of Species which depart widely from the standard. There may genus, we may conclude that this part has under- be truly said to be a constant struggle going on be- gone an extraordinary amount of modification, since tween, on the one hand, the tendency to reversion to the period when the species branched off from the a less modified state, as well as an innate tendency common progenitor of the genus. This period will to further variability of all kinds, and, on the other seldom be remote in any extreme degree, as species hand, the power of steady selection to keep the breed very rarely endure for more than one geological pe- true. In the long run selection gains the day, and we riod. An extraordinary amount of modification im- do not expect to fail so far as to breed a bird as coarse plies an unusually large and long-continued amount as a common tumbler from a good short-faced strain. of variability, which has continually been accumu- But as long as selection is rapidly going on, there lated by natural selection for the benefit of the spe- may always be expected to be much variability in cies. But as the variability of the extraordinarily-de- the structure undergoing modification. It further veloped part or organ has been so great and long- deserves notice that these variable characters, pro- continued within a period not excessively remote, duced by man’s selection, sometimes become at- we might, as a general rule, expect still to find more tached, from causes quite unknown to us, more to variability in such parts than in other parts of the one sex than to the other, generally to the male sex, organisation, which have remained for a much longer as with the wattle of carriers and the enlarged crop period nearly constant. And this, I am convinced, is of pouters. the case. That the struggle between natural selection Now let us turn to nature. When a part has been on the one hand, and the tendency to reversion and developed in an extraordinary manner in any one variability on the other hand, will in the course of species, compared with the other species of the same time cease; and that the most abnormally developed

142 Charles Darwin organs may be made constant, I can see no reason to genus of plants had blue flowers and some had red, doubt. Hence when an organ, however abnormal it the colour would be only a specific character, and may be, has been transmitted in approximately the no one would be surprised at one of the blue spe- same condition to many modified descendants, as cies varying into red, or conversely; but if all the in the case of the wing of the bat, it must have ex- species had blue flowers, the colour would become isted, according to my theory, for an immense pe- a generic character, and its variation would be a more riod in nearly the same state; and thus it comes to be unusual circumstance. I have chosen this example no more variable than any other structure. It is only because an explanation is not in this case applicable, in those cases in which the modification has been which most naturalists would advance, namely, that comparatively recent and extraordinarily great that specific characters are more variable than generic, we ought to find the generative variability, as it may because they are taken from parts of less physiologi- be called, still present in a high degree. For in this cal importance than those commonly used for class- case the variability will seldom as yet have been fixed ing genera. I believe this explanation is partly, yet by the continued selection of the individuals vary- only indirectly, true; I shall, however, have to return ing in the required manner and degree, and by the to this subject in our chapter on Classification. It continued rejection of those tending to revert to a would be almost superfluous to adduce evidence former and less modified condition. in support of the above statement, that specific char- The principle included in these remarks may be acters are more variable than generic; but I have re- extended. It is notorious that specific characters are peatedly noticed in works on natural history, that more variable than generic. To explain by a simple when an author has remarked with surprise that example what is meant. If some species in a large some important organ or part, which is generally

143 On the Origin of Species very constant throughout large groups of species, might surely expect to find them still often continu- has differed considerably in closely-allied species, ing to vary in those parts of their structure which that it has, also, been variable in the individuals of have varied within a moderately recent period, and some of the species. And this fact shows that a char- which have thus come to differ. Or to state the case acter, which is generally of generic value, when it in another manner:—the points in which all the spe- sinks in value and becomes only of specific value, cies of a genus resemble each other, and in which often becomes variable, though its physiological they differ from the species of some other genus, are importance may remain the same. Something of the called generic characters; and these characters in com- same kind applies to monstrosities: at least Is. mon I attribute to inheritance from a common pro- Geoffroy St. Hilaire seems to entertain no doubt, that genitor, for it can rarely have happened that natural the more an organ normally differs in the different selection will have modified several species, fitted species of the same group, the more subject it is to to more or less widely-different habits, in exactly the individual anomalies. same manner: and as these so-called generic charac- On the ordinary view of each species having been ters have been inherited from a remote period, since independently created, why should that part of the that period when the species first branched off from structure, which differs from the same part in other their common progenitor, and subsequently have not independently-created species of the same genus, varied or come to differ in any degree, or only in a be more variable than those parts which are closely slight degree, it is not probable that they should vary alike in the several species? I do not see that any at the present day. On the other hand, the points in explanation can be given. But on the view of species which species differ from other species of the same being only strongly marked and fixed varieties, we genus, are called specific characters; and as these

144 Charles Darwin specific characters have varied and come to differ can see why these characters should not have been within the period of the branching off of the species rendered as constant and uniform as other parts of from a common progenitor, it is probable that they the organisation; for secondary sexual characters should still often be in some degree variable,—at have been accumulated by sexual selection, which least more variable than those parts of the is less rigid in its action than ordinary selection, as it organisation which have for a very long period re- does not entail death, but only gives fewer offspring mained constant. to the less favoured males. Whatever the cause may In connexion with the present subject, I will make be of the variability of secondary sexual characters, only two other remarks. I think it will be admitted, as they are highly variable, sexual selection will have without my entering on details, that secondary had a wide scope for action, and may thus readily sexual characters are very variable; I think it also will have succeeded in giving to the species of the same be admitted that species of the same group differ group a greater amount of difference in their sexual from each other more widely in their secondary characters, than in other parts of their structure. sexual characters, than in other parts of their It is a remarkable fact, that the secondary sexual organisation; compare, for instance, the amount of differences between the two sexes of the same spe- difference between the males of gallinaceous birds, cies are generally displayed in the very same parts in which secondary sexual characters are strongly of the organisation in which the different species of displayed, with the amount of difference between the same genus differ from each other. Of this fact I their females; and the truth of this proposition will will give in illustration two instances, the first which be granted. The cause of the original variability of happen to stand on my list; and as the differences in secondary sexual characters is not manifest; but we these cases are of a very unusual nature, the relation

145 On the Origin of Species can hardly be accidental. The same number of joints sexes of the same species to each other, or to fit the in the tarsi is a character generally common to very males and females to different habits of life, or the large groups of beetles, but in the Engidae, as males to struggle with other males for the posses- Westwood has remarked, the number varies greatly; sion of the females. and the number likewise differs in the two sexes of Finally, then, I conclude that the greater variabil- the same species: again in fossorial hymenoptera, ity of specific characters, or those which distinguish the manner of neuration of the wings is a character species from species, than of generic characters, or of the highest importance, because common to large those which the species possess in common;—that groups; but in certain genera the neuration differs in the frequent extreme variability of any part which is the different species, and likewise in the two sexes developed in a species in an extraordinary manner of the same species. This relation has a clear mean- in comparison with the same part in its congeners; ing on my view of the subject: I look at all the spe- and the not great degree of variability in a part, how- cies of the same genus as having as certainly de- ever extraordinarily it may be developed, if it be scended from the same progenitor, as have the two common to a whole group of species;—that the great sexes of any one of the species. Consequently, what- variability of secondary sexual characters, and the ever part of the structure of the common progenitor, great amount of difference in these same characters or of its early descendants, became variable; varia- between closely allied species;—that secondary tions of this part would it is highly probable, be taken sexual and ordinary specific differences are gener- advantage of by natural and sexual selection, in or- ally displayed in the same parts of the der to fit the several species to their several places organisation,—are all principles closely connected in the economy of nature, and likewise to fit the two together. All being mainly due to the species of the

146 Charles Darwin same group having descended from a common pro- feathers on the feet,—characters not possessed by the genitor, from whom they have inherited much in aboriginal rock-pigeon; these then are analogous common,—to parts which have recently and largely variations in two or more distinct races. The frequent varied being more likely still to go on varying than presence of fourteen or even sixteen tail-feathers in parts which have long been inherited and have not the pouter, may be considered as a variation repre- varied,—to natural selection having more or less senting the normal structure of another race, the fan- completely, according to the lapse of time, overmas- tail. I presume that no one will doubt that all such tered the tendency to reversion and to further vari- analogous variations are due to the several races of ability,—to sexual selection being less rigid than the pigeon having inherited from a common parent ordinary selection,—and to variations in the same the same constitution and tendency to variation, parts having been accumulated by natural and sexual when acted on by similar unknown influences. In selection, and thus adapted for secondary sexual, the vegetable kingdom we have a case of analogous and for ordinary specific purposes. variation, in the enlarged stems, or roots as com- Distinct species present analogous variations; and monly called, of the Swedish turnip and Ruta baga, a variety of one species often assumes some of the plants which several botanists rank as varieties pro- characters of an allied species, or reverts to some of duced by cultivation from a common parent: if this the characters of an early progenitor. — These propo- be not so, the case will then be one of analogous sitions will be most readily understood by looking variation in two so-called distinct species; and to to our domestic races. The most distinct breeds of these a third may be added, namely, the common pigeons, in countries most widely apart, present sub- turnip. According to the ordinary view of each spe- varieties with reversed feathers on the head and cies having been independently created, we should

147 On the Origin of Species have to attribute this similarity in the enlarged stems beyond the influence of the mere act of crossing on of these three plants, not to the vera causa of com- the laws of inheritance. munity of descent, and a consequent tendency to No doubt it is a very surprising fact that characters vary in a like manner, but to three separate yet closely should reappear after having been lost for many, related acts of creation. perhaps for hundreds of generations. But when a With pigeons, however, we have another case, breed has been crossed only once by some other namely, the occasional appearance in all the breeds, breed, the offspring occasionally show a tendency of slaty-blue birds with two black bars on the wings, to revert in character to the foreign breed for many a white rump, a bar at the end of the tail, with the generations—some say, for a dozen or even a score outer feathers externally edged near their bases with of generations. After twelve generations, the propor- white. As all these marks are characteristic of the tion of blood, to use a common expression, of any parent rock-pigeon, I presume that no one will one ancestor, is only 1 in 2048; and yet, as we see, it doubt that this is a case of reversion, and not of a is generally believed that a tendency to reversion is new yet analogous variation appearing in the sev- retained by this very small proportion of foreign eral breeds. We may I think confidently come to blood. In a breed which has not been crossed, but in this conclusion, because, as we have seen, these which both parents have lost some character which coloured marks are eminently liable to appear in their progenitor possessed, the tendency, whether the crossed offspring of two distinct and differently strong or weak, to reproduce the lost character might coloured breeds; and in this case there is nothing be, as was formerly remarked, for all that we can see in the external conditions of life to cause the reap- to the contrary, transmitted for almost any number pearance of the slaty-blue, with the several marks, of generations. When a character which has been lost

148 Charles Darwin in a breed, reappears after a great number of genera- have an inherited tendency to produce it. tions, the most probable hypothesis is, not that the As all the species of the same genus are supposed, offspring suddenly takes after an ancestor some hun- on my theory, to have descended from a common dred generations distant, but that in each successive parent, it might be expected that they would occa- generation there has been a tendency to reproduce sionally vary in an analogous manner; so that a vari- the character in question, which at last, under un- ety of one species would resemble in some of its known favourable conditions, gains an ascendancy. characters another species; this other species being For instance, it is probable that in each generation of on my view only a well-marked and permanent va- the barb-pigeon, which produces most rarely a blue riety. But characters thus gained would probably be and black-barred bird, there has been a tendency in of an unimportant nature, for the presence of all im- each generation in the plumage to assume this portant characters will be governed by natural se- colour. This view is hypothetical, but could be sup- lection, in accordance with the diverse habits of the ported by some facts; and I can see no more abstract species, and will not be left to the mutual action of improbability in a tendency to produce any charac- the conditions of life and of a similar inherited con- ter being inherited for an endless number of genera- stitution. It might further be expected that the spe- tions, than in quite useless or rudimentary organs cies of the same genus would occasionally exhibit being, as we all know them to be, thus inherited. In- reversions to lost ancestral characters. As, however, deed, we may sometimes observe a mere tendency we never know the exact character of the common to produce a rudiment inherited: for instance, in the ancestor of a group, we could not distinguish these common snapdragon (Antirrhinum) a rudiment of a two cases: if, for instance, we did not know that the fifth stamen so often appears, that this plant must rock-pigeon was not feather-footed or turn-crowned,

149 On the Origin of Species we could not have told, whether these characters in species of the same genus. A considerable catalogue, our domestic breeds were reversions or only analo- also, could be given of forms intermediate between gous variations; but we might have inferred that the two other forms, which themselves must be doubt- blueness was a case of reversion, from the number of fully ranked as either varieties or species; and this the markings, which are correlated with the blue tint, shows, unless all these forms be considered as inde- and which it does not appear probable would all ap- pendently created species, that the one in varying pear together from simple variation. More especially has assumed some of the characters of the other, so we might have inferred this, from the blue colour and as to produce the intermediate form. But the best marks so often appearing when distinct breeds of di- evidence is afforded by parts or organs of an impor- verse colours are crossed. Hence, though under na- tant and uniform nature occasionally varying so as ture it must generally be left doubtful, what cases are to acquire, in some degree, the character of the same reversions to an anciently existing character, and what part or organ in an allied species. I have collected a are new but analogous variations, yet we ought, on long list of such cases; but here, as before, I lie under my theory, sometimes to find the varying offspring a great disadvantage in not being able to give them. of a species assuming characters (either from rever- I can only repeat that such cases certainly do occur, sion or from analogous variation) which already oc- and seem to me very remarkable. cur in some other members of the same group. And I will, however, give one curious and complex case, this undoubtedly is the case in nature. not indeed as affecting any important character, but A considerable part of the difficulty in recognising from occurring in several species of the same genus, a variable species in our systematic works, is due to partly under domestication and partly under nature. its varieties mocking, as it were, some of the other It is a case apparently of reversion. The ass not rarely

150 Charles Darwin has very distinct transverse bars on its legs, like those England of the spinal stripe in horses of the most on the legs of a zebra: it has been asserted that these distinct breeds, and of all colours; transverse bars are plainest in the foal, and from inquiries which I on the legs are not rare in duns, mouse-duns, and in have made, I believe this to be true. It has also been one instance in a chestnut: a faint shoulder-stripe may asserted that the stripe on each shoulder is some- sometimes be seen in duns, and I have seen a trace times double. The shoulder stripe is certainly very in a bay horse. My son made a careful examination variable in length and outline. A white ass, but not and sketch for me of a dun Belgian cart-horse with a an albino, has been described without either spinal double stripe on each shoulder and with leg-stripes; or shoulder-stripe; and these stripes are sometimes and a man, whom I can implicitly trust, has exam- very obscure, or actually quite lost, in dark-coloured ined for me a small dun Welch pony with three short asses. The koulan of Pallas is said to have been seen parallel stripes on each shoulder. with a double shoulder-stripe. The hemionus has no In the north-west part of India the Kattywar breed shoulder-stripe; but traces of it, as stated by Mr. Blyth of horses is so generally striped, that, as I hear from and others, occasionally appear: and I have been in- Colonel Poole, who examined the breed for the In- formed by Colonel Poole that foals of this species dian Government, a horse without stripes is not con- are generally striped on the legs, and faintly on the sidered as purely-bred. The spine is always striped; shoulder. The quagga, though so plainly barred like the legs are generally barred; and the shoulder-stripe, a zebra over the body, is without bars on the legs; which is sometimes double and sometimes treble, but Dr. Gray has figured one specimen with very is common; the side of the face, moreover, is some- distinct zebra-like bars on the hocks. times striped. The stripes are plainest in the foal; and With respect to the horse, I have collected cases in sometimes quite disappear in old horses. Colonel

151 On the Origin of Species Poole has seen both gray and bay Kattywar horses I am not at all satisfied with this theory, and should striped when first foaled. I have, also, reason to sus- be loth to apply it to breeds so distinct as the heavy pect, from information given me by Mr. W. W. Belgian cart-horse, Welch ponies, cobs, the lanky Edwards, that with the English race-horse the spinal Kattywar race, &c., inhabiting the most distant parts stripe is much commoner in the foal than in the full- of the world. grown animal. Without here entering on further de- Now let us turn to the effects of crossing the sev- tails, I may state that I have collected cases of leg eral species of the horse-genus. Rollin asserts, that and shoulder stripes in horses of very different the common mule from the ass and horse is particu- breeds, in various countries from Britain to Eastern larly apt to have bars on its legs. I once saw a mule China; and from Norway in the north to the Malay with its legs so much striped that any one at first Archipelago in the south. In all parts of the world would have thought that it must have been the prod- these stripes occur far oftenest in duns and mouse- uct of a zebra; and Mr. W. C. Martin, in his excellent duns; by the term dun a large range of colour is in- treatise on the horse, has given a figure of a similar cluded, from one between brown and black to a close mule. In four coloured drawings, which I have seen, approach to cream-colour. of hybrids between the ass and zebra, the legs were I am aware that Colonel Hamilton Smith, who has much more plainly barred than the rest of the body; written on this subject, believes that the several and in one of them there was a double shoulder- breeds of the horse have descended from several stripe. In Lord Moreton’s famous hybrid from a chest- aboriginal species—one of which, the dun, was nut mare and male quagga, the hybrid, and even the striped; and that the above-described appearances pure offspring subsequently produced from the are all due to ancient crosses with the dun stock. But mare by a black Arabian sire, were much more

152 Charles Darwin plainly barred across the legs than is even the pure like a zebra, or striped on the shoulders like an ass. quagga. Lastly, and this is another most remarkable In the horse we see this tendency strong whenever a case, a hybrid has been figured by Dr. Gray (and he dun tint appears—a tint which approaches to that of informs me that he knows of a second case) from the the general colouring of the other species of the ge- ass and the hemionus; and this hybrid, though the nus. The appearance of the stripes is not accompa- ass seldom has stripes on its legs and the hemionus nied by any change of form or by any other new char- has none and has not even a shoulder-stripe, never- acter. We see this tendency to become striped most theless had all four legs barred, and had three short strongly displayed in hybrids from between several shoulder-stripes, like those on the dun Welch pony, of the most distinct species. Now observe the case of and even had some zebra-like stripes on the sides of the several breeds of pigeons: they are descended its face. With respect to this last fact, I was so con- from a pigeon (including two or three sub-species vinced that not even a stripe of colour appears from or geographical races) of a bluish colour, with cer- what would commonly be called an accident, that I tain bars and other marks; and when any breed as- was led solely from the occurrence of the face-stripes sumes by simple variation a bluish tint, these bars on this hybrid from the ass and hemionus, to ask Colo- and other marks invariably reappear; but without nel Poole whether such face-stripes ever occur in the any other change of form or character. When the old- eminently striped Kattywar breed of horses, and was, est and truest breeds of various colours are crossed, as we have seen, answered in the affirmative. we see a strong tendency for the blue tint and bars What now are we to say to these several facts? We and marks to reappear in the mongrels. I have stated see several very distinct species of the horse-genus that the most probable hypothesis to account for the becoming, by simple variation, striped on the legs reappearance of very ancient characters, is—that

153 On the Origin of Species there is a tendency in the young of each successive other species of the genus; and that each has been generation to produce the long-lost character, and created with a strong tendency, when crossed with that this tendency, from unknown causes, sometimes species inhabiting distant quarters of the world, to prevails. And we have just seen that in several spe- produce hybrids resembling in their stripes, not their cies of the horse-genus the stripes are either plainer own parents, but other species of the genus. To ad- or appear more commonly in the young than in the mit this view is, as it seems to me, to reject a real for old. Call the breeds of pigeons, some of which have an unreal, or at least for an unknown, cause. It makes bred true for centuries, species; and how exactly the works of God a mere mockery and deception; I parallel is the case with that of the species of the would almost as soon believe with the old and ig- horse-genus! For myself, I venture confidently to norant cosmogonists, that fossil shells had never look back thousands on thousands of generations, lived, but had been created in stone so as to mock and I see an animal striped like a zebra, but perhaps the shells now living on the sea-shore. otherwise very differently constructed, the common parent of our domestic horse, whether or not it be Summary. — Our ignorance of the laws of variation descended from one or more wild stocks, of the ass, is profound. Not in one case out of a hundred can the hemionus, quagga, and zebra. we pretend to assign any reason why this or that part He who believes that each equine species was in- differs, more or less, from the same part in the par- dependently created, will, I presume, assert that each ents. But whenever we have the means of instituting species has been created with a tendency to vary, a comparison, the same laws appear to have acted both under nature and under domestication, in this in producing the lesser differences between variet- particular manner, so as often to become striped like ies of the same species, and the greater differences

154 Charles Darwin between species of the same genus. The external con- been closely checked by natural selection. It is prob- ditions of life, as climate and food, &c., seem to have ably from this same cause that organic beings low in induced some slight modifications. Habit in produc- the scale of nature are more variable than those which ing constitutional differences, and use in strength- have their whole organisation more specialised, and ening, and disuse in weakening and diminishing or- are higher in the scale. Rudimentary organs, from gans, seem to have been more potent in their effects. being useless, will be disregarded by natural selec- Homologous parts tend to vary in the same way, and tion, and hence probably are variable. Specific char- homologous parts tend to cohere. Modifications in acters—that is, the characters which have come to hard parts and in external parts sometimes affect differ since the several species of the same genus softer and internal parts. When one part is largely branched off from a common parent—are more vari- developed, perhaps it tends to draw nourishment able than generic characters, or those which have long from the adjoining parts; and every part of the struc- been inherited, and have not differed within this ture which can be saved without detriment to the same period. In these remarks we have referred to individual, will be saved. Changes of structure at an special parts or organs being still variable, because early age will generally affect parts subsequently they have recently varied and thus come to differ; developed; and there are very many other correla- but we have also seen in the second Chapter that the tions of growth, the nature of which we are utterly same principle applies to the whole individual; for unable to understand. Multiple parts are variable in in a district where many species of any genus are number and in structure, perhaps arising from such found—that is, where there has been much former parts not having been closely specialised to any par- variation and differentiation, or where the manufac- ticular function, so that their modifications have not tory of new specific forms has been actively at

155 On the Origin of Species work—there, on an average, we now find most vari- dants—which on my view must be a very slow pro- eties or incipient species. Secondary sexual charac- cess, requiring a long lapse of time—in this case, ters are highly variable, and such characters differ natural selection may readily have succeeded in giv- much in the species of the same group. Variability ing a fixed character to the organ, in however extraor- in the same parts of the organisation has generally dinary a manner it may be developed. Species in- been taken advantage of in giving secondary sexual heriting nearly the same constitution from a com- differences to the sexes of the same species, and spe- mon parent and exposed to similar influences will cific differences to the several species of the same naturally tend to present analogous variations, and genus. Any part or organ developed to an extraordi- these same species may occasionally revert to some nary size or in an extraordinary manner, in compari- of the characters of their ancient progenitors. Al- son with the same part or organ in the allied species, though new and important modifications may not must have gone through an extraordinary amount arise from reversion and analogous variation, such of modification since the genus arose; and thus we modifications will add to the beautiful and harmo- can understand why it should often still be variable nious diversity of nature. in a much higher degree than other parts; for varia- Whatever the cause may be of each slight differ- tion is a long-continued and slow process, and natu- ence in the offspring from their parents—and a cause ral selection will in such cases not as yet have had for each must exist—it is the steady accumulation, time to overcome the tendency to further variability through natural selection, of such differences, when and to reversion to a less modified state. But when a beneficial to the individual, that gives rise to all the species with any extraordinarily-developed organ more important modifications of structure, by which has become the parent of many modified descen- the innumerable beings on the face of this earth are

156 Charles Darwin enabled to struggle with each other, and the best Chapter VI adapted to survive. Difficulties on Theory

Difficulties on the theory of descent with modification — Transitions —Absence or rarity of transitional varieties — Transitions in habits of life — Diversi- fied habits in the same species — Species with habits widely different from those of their allies — Or- gans of extreme perfection — Means of transition — Cases of difficulty — Natura non facit saltum —Or- gans of small importance — Organs not in all cases absolutely perfect —The law of Unity of Type and of the Conditions of Existence embraced by the theory of Natural Selection.

LONG BEFORE HAVING ARRIVED at this part of my work, a crowd of difficulties will have occurred to the reader. Some of them are so grave that to this day I can never reflect on them without being staggered; but, to the best of my judgment, the greater number are only

157 On the Origin of Species apparent, and those that are real are not, I think, fa- marvellous an instinct as that which leads the bee to tal to my theory. make cells, which have practically anticipated the These difficulties and objections may be classed discoveries of profound mathematicians? under the following heads:- Firstly, why, if species Fourthly, how can we account for species, when have descended from other species by insensibly fine crossed, being sterile and producing sterile offspring, gradations, do we not everywhere see innumerable whereas, when varieties are crossed, their fertility is transitional forms? Why is not all nature in confu- unimpaired? sion instead of the species being, as we see them, The two first heads shall be here discussed—In- well defined? stinct and Hybridism in separate chapters. Secondly, is it possible that an animal having, for On the absence or rarity of transitional varieties. instance, the structure and habits of a bat, could have — As natural selection acts solely by the preserva- been formed by the modification of some animal with tion of profitable modifications, each new form will wholly different habits? Can we believe that natural tend in a fully-stocked country to take the place of, selection could produce, on the one hand, organs of and finally to exterminate, its own less improved trifling importance, such as the tail of a giraffe, which parent or other less-favoured forms with which it serves as a fly-flapper, and, on the other hand, or- comes into competition. Thus extinction and natu- gans of such wonderful structure, as the eye, of which ral selection will, as we have seen, go hand in hand. we hardly as yet fully understand the inimitable Hence, if we look at each species as descended from perfection? some other unknown form, both the parent and all Thirdly, can instincts be acquired and modified the transitional varieties will generally have been ex- through natural selection? What shall we say to so terminated by the very process of formation and per-

158 Charles Darwin fection of the new form. is being deposited, there will be blanks in our geo- But, as by this theory innumerable transitional logical history. The crust of the earth is a vast mu- forms must have existed, why do we not find them seum; but the natural collections have been made embedded in countless numbers in the crust of the only at intervals of time immensely remote. earth? It will be much more convenient to discuss But it may be urged that when several closely-al- this question in the chapter on the Imperfection of lied species inhabit the same territory we surely the geological record; and I will here only state that ought to find at the present time many transitional I believe the answer mainly lies in the record being forms. Let us take a simple case: in travelling from incomparably less perfect than is generally sup- north to south over a continent, we generally meet posed; the imperfection of the record being chiefly at successive intervals with closely allied or repre- due to organic beings not inhabiting profound sentative species, evidently filling nearly the same depths of the sea, and to their remains being embed- place in the natural economy of the land. These rep- ded and preserved to a future age only in masses of resentative species often meet and interlock; and as sediment sufficiently thick and extensive to with- the one becomes rarer and rarer, the other becomes stand an enormous amount of future degradation; more and more frequent, till the one replaces the and such fossiliferous masses can be accumulated other. But if we compare these species where they only where much sediment is deposited on the shal- intermingle, they are generally as absolutely distinct low bed of the sea, whilst it slowly subsides. These from each other in every detail of structure as are contingencies will concur only rarely, and after enor- specimens taken from the metropolis inhabited by mously long intervals. Whilst the bed of the sea is each. By my theory these allied species have de- stationary or is rising, or when very little sediment scended from a common parent; and during the pro-

159 On the Origin of Species cess of modification, each has become adapted to isting in the intermediate zones. By changes in the the conditions of life of its own region, and has sup- form of the land and of climate, marine areas now planted and exterminated its original parent and all continuous must often have existed within recent the transitional varieties between its past and present times in a far less continuous and uniform condition states. Hence we ought not to expect at the present than at present. But I will pass over this way of es- time to meet with numerous transitional varieties in caping from the difficulty; for I believe that many each region, though they must have existed there, perfectly defined species have been formed on and may be embedded there in a fossil condition. strictly continuous areas; though I do not doubt that But in the intermediate region, having intermediate the formerly broken condition of areas now continu- conditions of life, why do we not now find closely- ous has played an important part in the formation of linking intermediate varieties? This difficulty for a new species, more especially with freely-crossing long time quite confounded me. But I think it can be and wandering animals. in large part explained. In looking at species as they are now distributed In the first place we should be extremely cautious over a wide area, we generally find them tolerably in inferring, because an area is now continuous, that numerous over a large territory, then becoming it has been continuous during a long period. Geol- somewhat abruptly rarer and rarer on the confines, ogy would lead us to believe that almost every con- and finally disappearing. Hence the neutral territory tinent has been broken up into islands even during between two representative species is generally nar- the later tertiary periods; and in such islands dis- row in comparison with the territory proper to each. tinct species might have been separately formed We see the same fact in ascending mountains, and without the possibility of intermediate varieties ex- sometimes it is quite remarkable how abruptly, as

160 Charles Darwin Alph. De Candolle has observed, a common alpine blending one into another by insensible gradations, species disappears. The same fact has been noticed the range of any one species, depending as it does by Forbes in sounding the depths of the sea with the on the range of others, will tend to be sharply de- dredge. To those who look at climate and the physi- fined. Moreover, each species on the confines of its cal conditions of life as the all-important elements range, where it exists in lessened numbers, will, of distribution, these facts ought to cause surprise, during fluctuations in the number of its enemies or as climate and height or depth graduate away in- of its prey, or in the seasons, be extremely liable to sensibly. But when we bear in mind that almost ev- utter extermination; and thus its geographical range ery species, even in its metropolis, would increase will come to be still more sharply defined. immensely in numbers, were it not for other com- If I am right in believing that allied or representa- peting species; that nearly all either prey on or serve tive species, when inhabiting a continuous area, are as prey for others; in short, that each organic being is generally so distributed that each has a wide range, either directly or indirectly related in the most im- with a comparatively narrow neutral territory be- portant manner to other organic beings, we must see tween them, in which they become rather suddenly that the range of the inhabitants of any country by rarer and rarer; then, as varieties do not essentially no means exclusively depends on insensibly chang- differ from species, the same rule will probably ap- ing physical conditions, but in large part on the pres- ply to both; and if we in imagination adapt a vary- ence of other species, on which it depends, or by ing species to a very large area, we shall have to which it is destroyed, or with which it comes into adapt two varieties to two large areas, and a third competition; and as these species are already defined variety to a narrow intermediate zone. The interme- objects (however they may have become so), not diate variety, consequently, will exist in lesser num-

161 On the Origin of Species bers from inhabiting a narrow and lesser area; and terminated than one existing in large numbers; and practically, as far as I can make out, this rule holds in this particular case the intermediate form would good with varieties in a state of nature. I have met be eminently liable to the inroads of closely allied with striking instances of the rule in the case of vari- forms existing on both sides of it. But a far more im- eties intermediate between well-marked varieties in portant consideration, as I believe, is that, during the the genus Balanus. And it would appear from infor- process of further modification, by which two vari- mation given me by Mr. Watson, Dr. Asa Gray, and eties are supposed on my theory to be converted and Mr. Wollaston, that generally when varieties inter- perfected into two distinct species, the two which mediate between two other forms occur, they are exist in larger numbers from inhabiting larger areas, much rarer numerically than the forms which they will have a great advantage over the intermediate connect. Now, if we may trust these facts and infer- variety, which exists in smaller numbers in a narrow ences, and therefore conclude that varieties linking and intermediate zone. For forms existing in larger two other varieties together have generally existed numbers will always have a better chance, within in lesser numbers than the forms which they con- any given period, of presenting further favourable nect, then, I think, we can understand why interme- variations for natural selection to seize on, than will diate varieties should not endure for very long peri- the rarer forms which exist in lesser numbers. Hence, ods;—why as a general rule they should be extermi- the more common forms, in the race for life, will tend nated and disappear, sooner than the forms which to beat and supplant the less common forms, for they originally linked together. these will be more slowly modified and improved. For any form existing in lesser numbers would, as It is the same principle which, as I believe, accounts already remarked, run a greater chance of being ex- for the common species in each country, as shown in

162 Charles Darwin the second chapter, presenting on an average a riod present an inextricable chaos of varying and greater number of well-marked varieties than do the intermediate links: firstly, because new varieties are rarer species. I may illustrate what I mean by sup- very slowly formed, for variation is a very slow pro- posing three varieties of sheep to be kept, one cess, and natural selection can do nothing until adapted to an extensive mountainous region; a sec- favourable variations chance to occur, and until a ond to a comparatively narrow, hilly tract; and a third place in the natural polity of the country can be bet- to wide plains at the base; and that the inhabitants ter filled by some modification of some one or more are all trying with equal steadiness and skill to im- of its inhabitants. And such new places will depend prove their stocks by selection; the chances in this on slow changes of climate, or on the occasional im- case will be strongly in favour of the great holders migration of new inhabitants, and, probably, in a still on the mountains or on the plains improving their more important degree, on some of the old inhabit- breeds more quickly than the small holders on the ants becoming slowly modified, with the new forms intermediate narrow, hilly tract; and consequently thus produced and the old ones acting and reacting the improved mountain or plain breed will soon take on each other. So that, in any one region and at any the place of the less improved hill breed; and thus one time, we ought only to see a few species pre- the two breeds, which originally existed in greater senting slight modifications of structure in some numbers, will come into close contact with each degree permanent; and this assuredly we do see. other, without the interposition of the supplanted, Secondly, areas now continuous must often have intermediate hill-variety. existed within the recent period in isolated portions, To sum up, I believe that species come to be toler- in which many forms, more especially amongst the ably well-defined objects, and do not at any one pe- classes which unite for each birth and wander much,

163 On the Origin of Species may have separately been rendered sufficiently dis- fication through natural selection, they will almost tinct to rank as representative species. In this case, certainly be beaten and supplanted by the forms intermediate varieties between the several represen- which they connect; for these from existing in greater tative species and their common parent, must for- numbers will, in the aggregate, present more varia- merly have existed in each broken portion of the tion, and thus be further improved through natural land, but these links will have been supplanted and selection and gain further advantages. exterminated during the process of natural selection, Lastly, looking not to any one time, but to all time, so that they will no longer exist in a living state. if my theory be true, numberless intermediate variet- Thirdly, when two or more varieties have been ies, linking most closely all the species of the same formed in different portions of a strictly continuous group together, must assuredly have existed; but the area, intermediate varieties will, it is probable, at first very process of natural selection constantly tends, as have been formed in the intermediate zones, but they has been so often remarked, to exterminate the par- will generally have had a short duration. For these ent forms and the intermediate links. Consequently intermediate varieties will, from reasons already as- evidence of their former existence could be found only signed (namely from what we know of the actual dis- amongst fossil remains, which are preserved, as we tribution of closely allied or representative species, shall in a future chapter attempt to show, in an ex- and likewise of acknowledged varieties), exist in the tremely imperfect and intermittent record. intermediate zones in lesser numbers than the variet- On the origin and transitions of organic beings with ies which they tend to connect. From this cause alone peculiar habits and structure. — It has been asked the intermediate varieties will be liable to accidental by the opponents of such views as I hold, how, for extermination; and during the process of further modi- instance, a land carnivorous animal could have been

164 Charles Darwin converted into one with aquatic habits; for how could have collected, I can give only one or two instances of the animal in its transitional state have subsisted? It transitional habits and structures in closely allied spe- would be easy to show that within the same group cies of the same genus; and of diversified habits, ei- carnivorous animals exist having every intermedi- ther constant or occasional, in the same species. And ate grade between truly aquatic and strictly terres- it seems to me that nothing less than a long list of such trial habits; and as each exists by a struggle for life, cases is sufficient to lessen the difficulty in any par- it is clear that each is well adapted in its habits to its ticular case like that of the bat. place in nature. Look at the Mustela vison of North Look at the family of squirrels; here we have the America, which has webbed feet and which re- finest gradation from animals with their tails only sembles an otter in its fur, short legs, and form of slightly flattened, and from others, as Sir J. tail; during summer this animal dives for and preys Richardson has remarked, with the posterior part of on fish, but during the long winter it leaves the fro- their bodies rather wide and with the skin on their zen waters, and preys like other polecats on mice flanks rather full, to the so-called flying squirrels; and land animals. If a different case had been taken, and flying squirrels have their limbs and even the and it had been asked how an insectivorous quad- base of the tail united by a broad expanse of skin, ruped could possibly have been converted into a fly- which serves as a parachute and allows them to glide ing bat, the question would have been far more dif- through the air to an astonishing distance from tree ficult, and I could have given no answer. Yet I think to tree. We cannot doubt that each structure is of use such difficulties have very little weight. to each kind of squirrel in its own country, by en- Here, as on other occasions, I lie under a heavy dis- abling it to escape birds or beasts of prey, or to col- advantage, for out of the many striking cases which I lect food more quickly, or, as there is reason to be-

165 On the Origin of Species lieve, by lessening the danger from occasional falls. the limbs and the elongated fingers: the flank mem- But it does not follow from this fact that the struc- brane is, also, furnished with an extensor muscle. ture of each squirrel is the best that it is possible to Although no graduated links of structure, fitted for conceive under all natural conditions. Let the climate gliding through the air, now connect the and vegetation change, let other competing rodents Galeopithecus with the other Lemuridae, yet I can or new beasts of prey immigrate, or old ones become see no difficulty in supposing that such links for- modified, and all analogy would lead us to believe merly existed, and that each had been formed by the that some at least of the squirrels would decrease in same steps as in the case of the less perfectly gliding numbers or become exterminated, unless they also squirrels; and that each grade of structure had been became modified and improved in structure in a cor- useful to its possessor. Nor can I see any insuper- responding manner. Therefore, I can see no difficulty, able difficulty in further believing it possible that more especially under changing conditions of life, the membrane-connected fingers and fore-arm of the in the continued preservation of individuals with Galeopithecus might be greatly lengthened by natu- fuller and fuller flank-membranes, each modification ral selection; and this, as far as the organs of flight being useful, each being propagated, until by the are concerned, would convert it into a bat. In bats accumulated effects of this process of natural selec- which have the wing-membrane extended from the tion, a perfect so-called flying squirrel was produced. top of the shoulder to the tail, including the hind- Now look at the Galeopithecus or flying lemur, legs, we perhaps see traces of an apparatus origi- which formerly was falsely ranked amongst bats. It nally constructed for gliding through the air rather has an extremely wide flank-membrane, stretching than for flight. from the corners of the jaw to the tail, and including If about a dozen genera of birds had become ex-

166 Charles Darwin tinct or were unknown, who would have ventured diversified types, and formerly had flying reptiles, to have surmised that birds might have existed which it is conceivable that flying-fish, which now glide far used their wings solely as flappers, like the logger- through the air, slightly rising and turning by the headed duck (Micropterus of Eyton); as fins in the aid of their fluttering fins, might have been modi- water and front legs on the land, like the penguin; as fied into perfectly winged animals. If this had been sails, like the ostrich; and functionally for no pur- effected, who would have ever imagined that in an pose, like the Apteryx. Yet the structure of each of early transitional state they had been inhabitants of these birds is good for it, under the conditions of life the open ocean, and had used their incipient organs to which it is exposed, for each has to live by a of flight exclusively, as far as we know, to escape struggle; but it is not necessarily the best possible being devoured by other fish? under all possible conditions. It must not be inferred When we see any structure highly perfected for any from these remarks that any of the grades of wing- particular habit, as the wings of a bird for flight, we structure here alluded to, which perhaps may all should bear in mind that animals displaying early have resulted from disuse, indicate the natural steps transitional grades of the structure will seldom con- by which birds have acquired their perfect power of tinue to exist to the present day, for they will have flight; but they serve, at least, to show what diversi- been supplanted by the very process of perfection fied means of transition are possible. through natural selection. Furthermore, we may con- Seeing that a few members of such water-breath- clude that transitional grades between structures fit- ing classes as the Crustacea and Mollusca are ted for very different habits of life will rarely have adapted to live on the land, and seeing that we have been developed at an early period in great numbers flying birds and mammals, flying insects of the most and under many subordinate forms. Thus, to return

167 On the Origin of Species to our imaginary illustration of the flying-fish, it does ture lead to changed habits; both probably often not seem probable that fishes capable of true flight change almost simultaneously. Of cases of changed would have been developed under many subordi- habits it will suffice merely to allude to that of the nate forms, for taking prey of many kinds in many many British insects which now feed on exotic plants, ways, on the land and in the water, until their or- or exclusively on artificial substances. Of diversified gans of flight had come to a high stage of perfection, habits innumerable instances could be given: I have so as to have given them a decided advantage over often watched a tyrant flycatcher (Saurophagus other animals in the battle for life. Hence the chance sulphuratus) in South America, hovering over one of discovering species with transitional grades of spot and then proceeding to another, like a kestrel, structure in a fossil condition will always be less, and at other times standing stationary on the margin from their having existed in lesser numbers, than in of water, and then dashing like a kingfisher at a fish. the case of species with fully developed structures. In our own country the larger titmouse (Parus ma- I will now give two or three instances of diversi- jor) may be seen climbing branches, almost like a fied and of changed habits in the individuals of the creeper; it often, like a shrike, kills small birds by same species. When either case occurs, it would be blows on the head; and I have many times seen and easy for natural selection to fit the animal, by some heard it hammering the seeds of the yew on a branch, modification of its structure, for its changed habits, and thus breaking them like a nuthatch. In North or exclusively for one of its several different habits. America the black bear was seen by Hearne swim- But it is difficult to tell, and immaterial for us, ming for hours with widely open mouth, thus catch- whether habits generally change first and structure ing, like a whale, insects in the water. Even in so ex- afterwards; or whether slight modifications of struc- treme a case as this, if the supply of insects were

168 Charles Darwin constant, and if better adapted competitors did not there is a woodpecker, which in every essential part already exist in the country, I can see no difficulty in of its organisation, even in its colouring, in the harsh a race of bears being rendered, by natural selection, tone of its voice, and undulatory flight, told me more and more aquatic in their structure and habits, plainly of its close blood-relationship to our com- with larger and larger mouths, till a creature was mon species; yet it is a woodpecker which never produced as monstrous as a whale. climbs a tree! As we sometimes see individuals of a species fol- Petrels are the most aerial and oceanic of birds, yet lowing habits widely different from those both of in the quiet Sounds of Tierra del Fuego, the their own species and of the other species of the same Puffinuria berardi, in its general habits, in its aston- genus, we might expect, on my theory, that such in- ishing power of diving, its manner of swimming, and dividuals would occasionally have given rise to new of flying when unwillingly it takes flight, would be species, having anomalous habits, and with their mistaken by any one for an auk or grebe; neverthe- structure either slightly or considerably modified less, it is essentially a petrel, but with many parts of from that of their proper type. And such instances its organisation profoundly modified. On the other do occur in nature. Can a more striking instance of hand, the acutest observer by examining the dead adaptation be given than that of a woodpecker for body of the water-ouzel would never have suspected climbing trees and for seizing insects in the chinks its sub-aquatic habits; yet this anomalous member of the bark? Yet in North America there are wood- of the strictly terrestrial thrush family wholly sub- peckers which feed largely on fruit, and others with sists by diving,—grasping the stones with its feet and elongated wings which chase insects on the wing; using its wings under water. and on the plains of La Plata, where not a tree grows, He who believes that each being has been created

169 On the Origin of Species as we now see it, must occasionally have felt sur- that structure has begun to change. prise when he has met with an animal having habits He who believes in separate and innumerable acts and structure not at all in agreement. What can be of creation will say, that in these cases it has pleased plainer than that the webbed feet of ducks and geese the Creator to cause a being of one type to take the are formed for swimming? yet there are upland geese place of one of another type; but this seems to me with webbed feet which rarely or never go near the only restating the fact in dignified language. He who water; and no one except Audubon has seen the frig- believes in the struggle for existence and in the prin- ate-bird, which has all its four toes webbed, alight ciple of natural selection, will acknowledge that ev- on the surface of the sea. On the other hand, grebes ery organic being is constantly endeavouring to in- and coots are eminently aquatic, although their toes crease in numbers; and that if any one being vary are only bordered by membrane. What seems plainer ever so little, either in habits or structure, and thus than that the long toes of grallatores are formed for gain an advantage over some other inhabitant of the walking over swamps and floating plants, yet the country, it will seize on the place of that inhabitant, water-hen is nearly as aquatic as the coot; and the however different it may be from its own place. landrail nearly as terrestrial as the quail or partridge. Hence it will cause him no surprise that there should In such cases, and many others could be given, hab- be geese and frigate-birds with webbed feet, either its have changed without a corresponding change of living on the dry land or most rarely alighting on structure. The webbed feet of the upland goose may the water; that there should be long-toed corncrakes be said to have become rudimentary in function, living in meadows instead of in swamps; that there though not in structure. In the frigate-bird, the should be woodpeckers where not a tree grows; that deeply-scooped membrane between the toes shows there should be diving thrushes, and petrels with

170 Charles Darwin the habits of auks. originated; but I may remark that several facts make Organs of extreme perfection and complication. — me suspect that any sensitive nerve may be rendered To suppose that the eye, with all its inimitable con- sensitive to light, and likewise to those coarser vi- trivances for adjusting the focus to different dis- brations of the air which produce sound. tances, for admitting different amounts of light, and In looking for the gradations by which an organ in for the correction of spherical and chromatic aberra- any species has been perfected, we ought to look tion, could have been formed by natural selection, exclusively to its lineal ancestors; but this is scarcely seems, I freely confess, absurd in the highest pos- ever possible, and we are forced in each case to look sible degree. Yet reason tells me, that if numerous to species of the same group, that is to the collateral gradations from a perfect and complex eye to one descendants from the same original parent-form, in very imperfect and simple, each grade being useful order to see what gradations are possible, and for to its possessor, can be shown to exist; if further, the the chance of some gradations having been transmit- eye does vary ever so slightly, and the variations be ted from the earlier stages of descent, in an unaltered inherited, which is certainly the case; and if any varia- or little altered condition. Amongst existing Verte- tion or modification in the organ be ever useful to brata, we find but a small amount of gradation in an animal under changing conditions of life, then the the structure of the eye, and from fossil species we difficulty of believing that a perfect and complex eye can learn nothing on this head. In this great class we could be formed by natural selection, though insu- should probably have to descend far beneath the perable by our imagination, can hardly be consid- lowest known fossiliferous stratum to discover the ered real. How a nerve comes to be sensitive to light, earlier stages, by which the eye has been perfected. hardly concerns us more than how life itself first In the Articulata we can commence a series with

171 On the Origin of Species an optic nerve merely coated with pigment, and with- ratus of an optic nerve merely coated with pigment out any other mechanism; and from this low stage, and invested by transparent membrane, into an opti- numerous gradations of structure, branching off in cal instrument as perfect as is possessed by any mem- two fundamentally different lines, can be shown to ber of the great Articulate class. exist, until we reach a moderately high stage of per- He who will go thus far, if he find on finishing this fection. In certain crustaceans, for instance, there is a treatise that large bodies of facts, otherwise inexpli- double cornea, the inner one divided into facets, cable, can be explained by the theory of descent, within each of which there is a lens-shaped swell- ought not to hesitate to go further, and to admit that ing. In other crustaceans the transparent cones which a structure even as perfect as the eye of an eagle are coated by pigment, and which properly act only might be formed by natural selection, although in by excluding lateral pencils of light, are convex at their this case he does not know any of the transitional upper ends and must act by convergence; and at their grades. His reason ought to conquer his imagina- lower ends there seems to be an imperfect vitreous tion; though I have felt the difficulty far too keenly substance. With these facts, here far too briefly and to be surprised at any degree of hesitation in extend- imperfectly given, which show that there is much ing the principle of natural selection to such star- graduated diversity in the eyes of living crustaceans, tling lengths. and bearing in mind how small the number of living It is scarcely possible to avoid comparing the eye animals is in proportion to those which have become to a telescope. We know that this instrument has been extinct, I can see no very great difficulty (not more perfected by the long-continued efforts of the high- than in the case of many other structures) in believing est human intellects; and we naturally infer that the that natural selection has converted the simple appa- eye has been formed by a somewhat analogous pro-

172 Charles Darwin cess. But may not this inference be presumptuous? generation will multiply them almost infinitely, and Have we any right to assume that the Creator works natural selection will pick out with unerring skill by intellectual powers like those of man? If we must each improvement. Let this process go on for mil- compare the eye to an optical instrument, we ought lions on millions of years; and during each year on in imagination to take a thick layer of transparent millions of individuals of many kinds; and may we tissue, with a nerve sensitive to light beneath, and not believe that a living optical instrument might then suppose every part of this layer to be continu- thus be formed as superior to one of glass, as the ally changing slowly in density, so as to separate works of the Creator are to those of man? into layers of different densities and thicknesses, If it could be demonstrated that any complex or- placed at different distances from each other, and gan existed, which could not possibly have been with the surfaces of each layer slowly changing in formed by numerous, successive, slight modifica- form. Further we must suppose that there is a power tions, my theory would absolutely break down. But always intently watching each slight accidental al- I can find out no such case. No doubt many organs teration in the transparent layers; and carefully se- exist of which we do not know the transitional lecting each alteration which, under varied circum- grades, more especially if we look to much-isolated stances, may in any way, or in any degree, tend to species, round which, according to my theory, there produce a distincter image. We must suppose each has been much extinction. Or again, if we look to an new state of the instrument to be multiplied by the organ common to all the members of a large class, million; and each to be preserved till a better be pro- for in this latter case the organ must have been first duced, and then the old ones to be destroyed. In liv- formed at an extremely remote period, since which ing bodies, variation will cause the slight alterations, all the many members of the class have been devel-

173 On the Origin of Species oped; and in order to discover the early transitional branchiae that breathe the air dissolved in the water, grades through which the organ has passed, we at the same time that they breathe free air in their should have to look to very ancient ancestral forms, swimbladders, this latter organ having a ductus long since become extinct. pneumaticus for its supply, and being divided by We should be extremely cautious in concluding highly vascular partitions. In these cases, one of the that an organ could not have been formed by transi- two organs might with ease be modified and per- tional gradations of some kind. Numerous cases fected so as to perform all the work by itself, being could be given amongst the lower animals of the aided during the process of modification by the other same organ performing at the same time wholly dis- organ; and then this other organ might be modified tinct functions; thus the alimentary canal respires, for some other and quite distinct purpose, or be quite digests, and excretes in the larva of the dragon-fly obliterated. and in the fish Cobites. In the Hydra, the animal may The illustration of the swimbladder in fishes is a be turned inside out, and the exterior surface will good one, because it shows us clearly the highly then digest and the stomach respire. In such cases important fact that an organ originally constructed natural selection might easily specialise, if any ad- for one purpose, namely flotation, may be converted vantage were thus gained, a part or organ, which had into one for a wholly different purpose, namely res- performed two functions, for one function alone, and piration. The swimbladder has, also, been worked thus wholly change its nature by insensible steps. in as an accessory to the auditory organs of certain Two distinct organs sometimes perform simulta- fish, or, for I do not know which view is now gener- neously the same function in the same individual; ally held, a part of the auditory apparatus has been to give one instance, there are fish with gills or worked in as a complement to the swimbladder. All

174 Charles Darwin physiologists admit that the swimbladder is homolo- position. But it is conceivable that the now utterly lost gous, or ‘ideally similar,’ in position and structure branchiae might have been gradually worked in by with the lungs of the higher vertebrate animals: hence natural selection for some quite distinct purpose: in there seems to me to be no great difficulty in believ- the same manner as, on the view entertained by some ing that natural selection has actually converted a naturalists that the branchiae and dorsal scales of swimbladder into a lung, or organ used exclusively Annelids are homologous with the wings and wing- for respiration. covers of insects, it is probable that organs which at a I can, indeed, hardly doubt that all vertebrate ani- very ancient period served for respiration have been mals having true lungs have descended by ordinary actually converted into organs of flight. generation from an ancient prototype, of which we In considering transitions of organs, it is so impor- know nothing, furnished with a floating apparatus tant to bear in mind the probability of conversion or swimbladder. We can thus, as I infer from Profes- from one function to another, that I will give one more sor Owen’s interesting description of these parts, un- instance. Pedunculated cirripedes have two minute derstand the strange fact that every particle of food folds of skin, called by me the ovigerous frena, which and drink which we swallow has to pass over the serve, through the means of a sticky secretion, to re- orifice of the trachea, with some risk of falling into tain the eggs until they are hatched within the sack. the lungs, notwithstanding the beautiful contrivance These cirripedes have no branchiae, the whole sur- by which the glottis is closed. In the higher Verte- face of the body and sack, including the small frena, brata the branchiae have wholly disappeared—the serving for respiration. The Balanidae or sessile slits on the sides of the neck and the loop-like course cirripedes, on the other hand, have no ovigerous of the arteries still marking in the embryo their former frena, the eggs lying loose at the bottom of the sack,

175 On the Origin of Species in the well-enclosed shell; but they have large folded of which will be discussed in my future work. branchiae. Now I think no one will dispute that the One of the gravest is that of neuter insects, which ovigerous frena in the one family are strictly homolo- are often very differently constructed from either the gous with the branchiae of the other family; indeed, males or fertile females; but this case will be treated they graduate into each other. Therefore I do not of in the next chapter. The electric organs of fishes doubt that little folds of skin, which originally served offer another case of special difficulty; it is impos- as ovigerous frena, but which, likewise, very slightly sible to conceive by what steps these wondrous or- aided the act of respiration, have been gradually gans have been produced; but, as Owen and others converted by natural selection into branchiae, sim- have remarked, their intimate structure closely re- ply through an increase in their size and the oblit- sembles that of common muscle; and as it has lately eration of their adhesive glands. If all pedunculated been shown that Rays have an organ closely analo- cirripedes had become extinct, and they have already gous to the electric apparatus, and yet do not, as suffered far more extinction than have sessile Matteuchi asserts, discharge any electricity, we must cirripedes, who would ever have imagined that the own that we are far too ignorant to argue that no tran- branchiae in this latter family had originally existed sition of any kind is possible. as organs for preventing the ova from being washed The electric organs offer another and even more out of the sack? serious difficulty; for they occur in only about a Although we must be extremely cautious in con- dozen fishes, of which several are widely remote in cluding that any organ could not possibly have been their affinities. Generally when the same organ ap- produced by successive transitional gradations, yet, pears in several members of the same class, espe- undoubtedly, grave cases of difficulty occur, some cially if in members having very different habits of

176 Charles Darwin life, we may attribute its presence to inheritance from damental difference can generally be detected. I am a common ancestor; and its absence in some of the inclined to believe that in nearly the same way as members to its loss through disuse or natural selec- two men have sometimes independently hit on the tion. But if the electric organs had been inherited from very same invention, so natural selection, working one ancient progenitor thus provided, we might have for the good of each being and taking advantage of expected that all electric fishes would have been analogous variations, has sometimes modified in specially related to each other. Nor does geology at very nearly the same manner two parts in two or- all lead to the belief that formerly most fishes had ganic beings, which owe but little of their structure electric organs, which most of their modified descen- in common to inheritance from the same ancestor. dants have lost. The presence of luminous organs in Although in many cases it is most difficult to con- a few insects, belonging to different families and or- jecture by what transitions an organ could have ar- ders, offers a parallel case of difficulty. Other cases rived at its present state; yet, considering that the could be given; for instance in plants, the very curi- proportion of living and known forms to the extinct ous contrivance of a mass of pollen-grains, borne on and unknown is very small, I have been astonished a foot-stalk with a sticky gland at the end, is the same how rarely an organ can be named, towards which in Orchis and Asclepias,—genera almost as remote no transitional grade is known to lead. The truth of as possible amongst flowering plants. In all these this remark is indeed shown by that old canon in cases of two very distinct species furnished with ap- natural history of ‘Natura non facit saltum.’ We meet parently the same anomalous organ, it should be with this admission in the writings of almost every observed that, although the general appearance and experienced naturalist; or, as Milne Edwards has function of the organ may be the same, yet some fun- well expressed it, nature is prodigal in variety, but

177 On the Origin of Species niggard in innovation. Why, on the theory of Cre- as much difficulty, though of a very different kind, ation, should this be so? Why should all the parts on this head, as in the case of an organ as perfect and and organs of many independent beings, each sup- complex as the eye. posed to have been separately created for its proper In the first place, we are much too ignorant in replace in nature, be so invariably linked together by gard to the whole economy of any one organic be- graduated steps? Why should not Nature have taken ing, to say what slight modifications would be of a leap from structure to structure? On the theory of importance or not. In a former chapter I have given natural selection, we can clearly understand why she instances of most trifling characters, such as the down should not; for natural selection can act only by tak- on fruit and the colour of the flesh, which, from de- ing advantage of slight successive variations; she can termining the attacks of insects or from being corre- never take a leap, but must advance by the shortest lated with constitutional differences, might assur- and slowest steps. edly be acted on by natural selection. The tail of the Organs of little apparent importance. — As natu- giraffe looks like an artificially constructed fly-flap- ral selection acts by life and death,—by the preser- per; and it seems at first incredible that this could vation of individuals with any favourable variation, have been adapted for its present purpose by suc- and by the destruction of those with any cessive slight modifications, each better and better, unfavourable deviation of structure,—I have some- for so trifling an object as driving away flies; yet we times felt much difficulty in understanding the ori- should pause before being too positive even in this gin of simple parts, of which the importance does case, for we know that the distribution and existence not seem sufficient to cause the preservation of suc- of cattle and other animals in South America abso- cessively varying individuals. I have sometimes felt lutely depends on their power of resisting the at-

178 Charles Darwin tacks of insects: so that individuals which could by bladders betray their aquatic origin, may perhaps any means defend themselves from these small en- be thus accounted for. A well-developed tail having emies, would be able to range into new pastures and been formed in an aquatic animal, it might subse- thus gain a great advantage. It is not that the larger quently come to be worked in for all sorts of pur- quadrupeds are actually destroyed (except in some poses, as a fly-flapper, an organ of prehension, or as rare cases) by the flies, but they are incessantly ha- an aid in turning, as with the dog, though the aid rassed and their strength reduced, so that they are must be slight, for the hare, with hardly any tail, can more subject to disease, or not so well enabled in a double quickly enough. coming dearth to search for food, or to escape from In the second place, we may sometimes attribute beasts of prey. importance to characters which are really of very little Organs now of trifling importance have probably importance, and which have originated from quite in some cases been of high importance to an early secondary causes, independently of natural selec- progenitor, and, after having been slowly perfected tion. We should remember that climate, food, &c., at a former period, have been transmitted in nearly probably have some little direct influence on the the same state, although now become of very slight organisation; that characters reappear from the law use; and any actually injurious deviations in their of reversion; that correlation of growth will have had structure will always have been checked by natural a most important influence in modifying various selection. Seeing how important an organ of loco- structures; and finally, that sexual selection will of- motion the tail is in most aquatic animals, its gen- ten have largely modified the external characters of eral presence and use for many purposes in so many animals having a will, to give one male an advan- land animals, which in their lungs or modified swim- tage in fighting with another or in charming the fe-

179 On the Origin of Species males. Moreover when a modification of structure highest service to the plant; but as we see nearly simi- has primarily arisen from the above or other un- lar hooks on many trees which are not climbers, the known causes, it may at first have been of no advan- hooks on the bamboo may have arisen from un- tage to the species, but may subsequently have been known laws of growth, and have been subsequently taken advantage of by the descendants of the spe- taken advantage of by the plant undergoing further cies under new conditions of life and with newly modification and becoming a climber. The naked acquired habits. skin on the head of a vulture is generally looked at To give a few instances to illustrate these latter re- as a direct adaptation for wallowing in putridity; and marks. If green woodpeckers alone had existed, and so it may be, or it may possibly be due to the direct we did not know that there were many black and action of putrid matter; but we should be very cau- pied kinds, I dare say that we should have thought tious in drawing any such inference, when we see that the green colour was a beautiful adaptation to that the skin on the head of the clean-feeding male hide this tree-frequenting bird from its enemies; and turkey is likewise naked. The sutures in the skulls consequently that it was a character of importance of young mammals have been advanced as a beauti- and might have been acquired through natural se- ful adaptation for aiding parturition, and no doubt lection; as it is, I have no doubt that the colour is due they facilitate, or may be indispensable for this act; to some quite distinct cause, probably to sexual se- but as sutures occur in the skulls of young birds and lection. A trailing bamboo in the Malay Archipelago reptiles, which have only to escape from a broken climbs the loftiest trees by the aid of exquisitely con- egg, we may infer that this structure has arisen from structed hooks clustered around the ends of the the laws of growth, and has been taken advantage of branches, and this contrivance, no doubt, is of the in the parturition of the higher animals.

180 Charles Darwin We are profoundly ignorant of the causes produc- often have to struggle for their own subsistence, and ing slight and unimportant variations; and we are would be exposed to a certain extent to natural se- immediately made conscious of this by reflecting on lection, and individuals with slightly different con- the differences in the breeds of our domesticated stitutions would succeed best under different cli- animals in different countries,—more especially in mates; and there is reason to believe that constitu- the less civilized countries where there has been but tion and colour are correlated. A good observer, also, little artificial selection. Careful observers are con- states that in cattle susceptibility to the attacks of flies vinced that a damp climate affects the growth of the is correlated with colour, as is the liability to be poi- hair, and that with the hair the horns are correlated. soned by certain plants; so that colour would be thus Mountain breeds always differ from lowland breeds; subjected to the action of natural selection. But we and a mountainous country would probably affect are far too ignorant to speculate on the relative im- the hind limbs from exercising them more, and pos- portance of the several known and unknown laws of sibly even the form of the pelvis; and then by the variation; and I have here alluded to them only to law of homologous variation, the front limbs and show that, if we are unable to account for the charac- even the head would probably be affected. The teristic differences of our domestic breeds, which shape, also, of the pelvis might affect by pressure nevertheless we generally admit to have arisen the shape of the head of the young in the womb. The through ordinary generation, we ought not to lay too laborious breathing necessary in high regions would, much stress on our ignorance of the precise cause of we have some reason to believe, increase the size of the slight analogous differences between species. I the chest; and again correlation would come into might have adduced for this same purpose the dif- play. Animals kept by savages in different countries ferences between the races of man, which are so

181 On the Origin of Species strongly marked; I may add that some little light can merly had arisen from correlation of growth, or from apparently be thrown on the origin of these differ- other unknown cause, may reappear from the law of ences, chiefly through sexual selection of a particu- reversion, though now of no direct use. The effects lar kind, but without here entering on copious de- of sexual selection, when displayed in beauty to tails my reasoning would appear frivolous. charm the females, can be called useful only in rather The foregoing remarks lead me to say a few words a forced sense. But by far the most important con- on the protest lately made by some naturalists, sideration is that the chief part of the organisation of against the utilitarian doctrine that every detail of every being is simply due to inheritance; and conse- structure has been produced for the good of its pos- quently, though each being assuredly is well fitted sessor. They believe that very many structures have for its place in nature, many structures now have no been created for beauty in the eyes of man, or for direct relation to the habits of life of each species. mere variety. This doctrine, if true, would be abso- Thus, we can hardly believe that the webbed feet of lutely fatal to my theory. Yet I fully admit that many the upland goose or of the frigate-bird are of special structures are of no direct use to their possessors. use to these birds; we cannot believe that the same Physical conditions probably have had some little bones in the arm of the monkey, in the fore leg of the effect on structure, quite independently of any good horse, in the wing of the bat, and in the flipper of the thus gained. Correlation of growth has no doubt seal, are of special use to these animals. We may played a most important part, and a useful modifi- safely attribute these structures to inheritance. But cation of one part will often have entailed on other to the progenitor of the upland goose and of the frig- parts diversified changes of no direct use. So again ate-bird, webbed feet no doubt were as useful as they characters which formerly were useful, or which for- now are to the most aquatic of existing birds. So we

182 Charles Darwin may believe that the progenitor of the seal had not a good of another species; though throughout nature flipper, but a foot with five toes fitted for walking or one species incessantly takes advantage of, and prof- grasping; and we may further venture to believe that its by, the structure of another. But natural selection the several bones in the limbs of the monkey, horse, can and does often produce structures for the direct and bat, which have been inherited from a common injury of other species, as we see in the fang of the progenitor, were formerly of more special use to that adder, and in the ovipositor of the ichneumon, by progenitor, or its progenitors, than they now are to which its eggs are deposited in the living bodies of these animals having such widely diversified hab- other insects. If it could be proved that any part of its. Therefore we may infer that these several bones the structure of any one species had been formed for might have been acquired through natural selection, the exclusive good of another species, it would an- subjected formerly, as now, to the several laws of nihilate my theory, for such could not have been pro- inheritance, reversion, correlation of growth, &c. duced through natural selection. Although many Hence every detail of structure in every living crea- statements may be found in works on natural his- ture (making some little allowance for the direct ac- tory to this effect, I cannot find even one which seems tion of physical conditions) may be viewed, either to me of any weight. It is admitted that the rattle- as having been of special use to some ancestral form, snake has a poison-fang for its own defence and for or as being now of special use to the descendants of the destruction of its prey; but some authors sup- this form—either directly, or indirectly through the pose that at the same time this snake is furnished complex laws of growth. with a rattle for its own injury, namely, to warn its Natural selection cannot possibly produce any prey to escape. I would almost as soon believe that modification in any one species exclusively for the the cat curls the end of its tail when preparing to

183 On the Origin of Species spring, in order to warn the doomed mouse. But I now rapidly yielding before the advancing legions have not space here to enter on this and other such of plants and animals introduced from Europe. Natu- cases. ral selection will not produce absolute perfection, Natural selection will never produce in a being nor do we always meet, as far as we can judge, with anything injurious to itself, for natural selection acts this high standard under nature. The correction for solely by and for the good of each. No organ will be the aberration of light is said, on high authority, not formed, as Paley has remarked, for the purpose of to be perfect even in that most perfect organ, the eye. causing pain or for doing an injury to its possessor. If our reason leads us to admire with enthusiasm a If a fair balance be struck between the good and evil multitude of inimitable contrivances in nature, this caused by each part, each will be found on the whole same reason tells us, though we may easily err on advantageous. After the lapse of time, under chang- both sides, that some other contrivances are less per- ing conditions of life, if any part comes to be injuri- fect. Can we consider the sting of the wasp or of the ous, it will be modified; or if it be not so, the being bee as perfect, which, when used against many at- will become extinct, as myriads have become extinct. tacking animals, cannot be withdrawn, owing to the Natural selection tends only to make each organic backward serratures, and so inevitably causes the being as perfect as, or slightly more perfect than, the death of the insect by tearing out its viscera? other inhabitants of the same country with which it If we look at the sting of the bee, as having origi- has to struggle for existence. And we see that this is nally existed in a remote progenitor as a boring and the degree of perfection attained under nature. The serrated instrument, like that in so many members endemic productions of New Zealand, for instance, of the same great order, and which has been modi- are perfect one compared with another; but they are fied but not perfected for its present purpose, with

184 Charles Darwin the poison originally adapted to cause galls subse- If we admire the several ingenious contrivances, by quently intensified, we can perhaps understand how which the flowers of the orchis and of many other it is that the use of the sting should so often cause plants are fertilised through insect agency, can we the insect’s own death: for if on the whole the power consider as equally perfect the elaboration by our of stinging be useful to the community, it will fulfil fir-trees of dense clouds of pollen, in order that a all the requirements of natural selection, though it few granules may be wafted by a chance breeze on may cause the death of some few members. If we to the ovules? admire the truly wonderful power of scent by which the males of many insects find their females, can we Summary of Chapter — We have in this chapter dis- admire the production for this single purpose of cussed some of the difficulties and objections which thousands of drones, which are utterly useless to the may be urged against my theory. Many of them are community for any other end, and which are ulti- very grave; but I think that in the discussion light mately slaughtered by their industrious and sterile has been thrown on several facts, which on the theory sisters? It may be difficult, but we ought to admire of independent acts of creation are utterly obscure. the savage instinctive hatred of the queen-bee, which We have seen that species at any one period are not urges her instantly to destroy the young queens her indefinitely variable, and are not linked together by daughters as soon as born, or to perish herself in the a multitude of intermediate gradations, partly be- combat; for undoubtedly this is for the good of the cause the process of natural selection will always be community; and maternal love or maternal hatred, very slow, and will act, at any one time, only on a though the latter fortunately is most rare, is all the very few forms; and partly because the very process same to the inexorable principle of natural selection. of natural selection almost implies the continual sup-

185 On the Origin of Species planting and extinction of preceding and intermedi- glide through the air. ate gradations. Closely allied species, now living on We have seen that a species may under new con- a continuous area, must often have been formed ditions of life change its habits, or have diversified when the area was not continuous, and when the habits, with some habits very unlike those of its near- conditions of life did not insensibly graduate away est congeners. Hence we can understand, bearing in from one part to another. When two varieties are mind that each organic being is trying to live wher- formed in two districts of a continuous area, an in- ever it can live, how it has arisen that there are up- termediate variety will often be formed, fitted for an land geese with webbed feet, ground woodpeckers, intermediate zone; but from reasons assigned, the diving thrushes, and petrels with the habits of auks. intermediate variety will usually exist in lesser num- Although the belief that an organ so perfect as the bers than the two forms which it connects; conse- eye could have been formed by natural selection, is quently the two latter, during the course of further more than enough to stagger any one; yet in the case modification, from existing in greater numbers, will of any organ, if we know of a long series of grada- have a great advantage over the less numerous in- tions in complexity, each good for its possessor, then, termediate variety, and will thus generally succeed under changing conditions of life, there is no logical in supplanting and exterminating it. impossibility in the acquirement of any conceivable We have seen in this chapter how cautious we degree of perfection through natural selection. In the should be in concluding that the most different hab- cases in which we know of no intermediate or tran- its of life could not graduate into each other; that a sitional states, we should be very cautious in con- bat, for instance, could not have been formed by natu- cluding that none could have existed, for the homolo- ral selection from an animal which at first could only gies of many organs and their intermediate states

186 Charles Darwin show that wonderful metamorphoses in function are retained (as the tail of an aquatic animal by its ter- at least possible. For instance, a swim-bladder has restrial descendants), though it has become of such apparently been converted into an air-breathing small importance that it could not, in its present state, lung. The same organ having performed simulta- have been acquired by natural selection,—a power neously very different functions, and then having which acts solely by the preservation of profitable been specialised for one function; and two very dis- variations in the struggle for life. tinct organs having performed at the same time the Natural selection will produce nothing in one spe- same function, the one having been perfected whilst cies for the exclusive good or injury of another; aided by the other, must often have largely facili- though it may well produce parts, organs, and ex- tated transitions. cretions highly useful or even indispensable, or We are far too ignorant, in almost every case, to be highly injurious to another species, but in all cases enabled to assert that any part or organ is so unim- at the same time useful to the owner. Natural selec- portant for the welfare of a species, that modification in each well-stocked country, must act chiefly tions in its structure could not have been slowly ac- through the competition of the inhabitants one with cumulated by means of natural selection. But we may another, and consequently will produce perfection, confidently believe that many modifications, wholly or strength in the battle for life, only according to the due to the laws of growth, and at first in no way ad- standard of that country. Hence the inhabitants of vantageous to a species, have been subsequently one country, generally the smaller one, will often taken advantage of by the still further modified de- yield, as we see they do yield, to the inhabitants of scendants of this species. We may, also, believe that another and generally larger country. For in the larger a part formerly of high importance has often been country there will have existed more individuals,

187 On the Origin of Species and more diversified forms, and the competition will embraced by the principle of natural selection. For have been severer, and thus the standard of perfec- natural selection acts by either now adapting the tion will have been rendered higher. Natural selec- varying parts of each being to its organic and inor- tion will not necessarily produce absolute perfec- ganic conditions of life; or by having adapted them tion; nor, as far as we can judge by our limited facul- during long-past periods of time: the adaptations ties, can absolute perfection be everywhere found. being aided in some cases by use and disuse, being On the theory of natural selection we can clearly slightly affected by the direct action of the external understand the full meaning of that old canon in natu- conditions of life, and being in all cases subjected to ral history, ‘Natura non facit saltum.’ This canon, if the several laws of growth. Hence, in fact, the law of we look only to the present inhabitants of the world, the Conditions of Existence is the higher law; as it is not strictly correct, but if we include all those of includes, through the inheritance of former adapta- past times, it must by my theory be strictly true. tions, that of Unity of Type. It is generally acknowledged that all organic beings have been formed on two great laws—Unity of Type, and the Conditions of Existence. By unity of type is meant that fundamental agreement in structure, which we see in organic beings of the same class, and which is quite independent of their habits of life. On my theory, unity of type is explained by unity of descent. The expression of conditions of existence, so often insisted on by the illustrious Cuvier, is fully

188 Charles Darwin Chapter VII mental powers, any more than I have with that of life itself. We are concerned only with the diversi- Instinct ties of instinct and of the other mental qualities of animals within the same class. Instincts comparable with habits, but different in their I will not attempt any definition of instinct. It would origin —Instincts graduated — Aphides and ants — be easy to show that several distinct mental actions Instincts variable — Domestic instincts, their origin are commonly embraced by this term; but every one — Natural instincts of the cuckoo, ostrich, and para- understands what is meant, when it is said that in- sitic bees — Slave-making ants — Hive-bee, its cell- stinct impels the cuckoo to migrate and to lay her making instinct — Difficulties on the theory of the eggs in other birds’ nests. An action, which we our- Natural Selection of instincts —Neuter or sterile in- selves should require experience to enable us to sects — Summary. perform, when performed by an animal, more especially by a very young one, without any experience, THE SUBJECT OF INSTINCT might have been worked into and when performed by many individuals in the the previous chapters; but I have thought that it same way, without their knowing for what purpose would be more convenient to treat the subject sepa- it is performed, is usually said to be instinctive. But rately, especially as so wonderful an instinct as that I could show that none of these characters of instinct of the hive-bee making its cells will probably have are universal. A little dose, as Pierre Huber expresses occurred to many readers, as a difficulty sufficient it, of judgment or reason, often comes into play, even to overthrow my whole theory. I must premise, that in animals very low in the scale of nature. I have nothing to do with the origin of the primary Frederick Cuvier and several of the older metaphy-

189 On the Origin of Species sicians have compared instinct with habit. This com- mock completed up only to the third stage, the cat- parison gives, I think, a remarkably accurate notion erpillar simply re-performed the fourth, fifth, and of the frame of mind under which an instinctive ac- sixth stages of construction. If, however, a caterpil- tion is performed, but not of its origin. How uncon- lar were taken out of a hammock made up, for in- sciously many habitual actions are performed, instance, to the third stage, and were put into one fin- deed not rarely in direct opposition to our conscious ished up to the sixth stage, so that much of its work will! yet they may be modified by the will or reason. was already done for it, far from feeling the benefit Habits easily become associated with other habits, of this, it was much embarrassed, and, in order to and with certain periods of time and states of the complete its hammock, seemed forced to start from body. When once acquired, they often remain con- the third stage, where it had left off, and thus tried to stant throughout life. Several other points of resem- complete the already finished work. If we suppose blance between instincts and habits could be pointed any habitual action to become inherited—and I think out. As in repeating a well-known song, so in in- it can be shown that this does sometimes happen— stincts, one action follows another by a sort of rhythm; then the resemblance between what originally was a if a person be interrupted in a song, or in repeating habit and an instinct becomes so close as not to be anything by rote, he is generally forced to go back to distinguished. If Mozart, instead of playing the pi- recover the habitual train of thought: so P. Huber anoforte at three years old with wonderfully little found it was with a caterpillar, which makes a very practice, had played a tune with no practice at all, complicated hammock; for if he took a caterpillar be might truly be said to have done so instinctively. which had completed its hammock up to, say, the But it would be the most serious error to suppose sixth stage of construction, and put it into a ham- that the greater number of instincts have been ac-

190 Charles Darwin quired by habit in one generation, and then trans- habit are of quite subordinate importance to the ef- mitted by inheritance to succeeding generations. It fects of the natural selection of what may be called can be clearly shown that the most wonderful in- accidental variations of instincts;—that is of varia- stincts with which we are acquainted, namely, those tions produced by the same unknown causes which of the hive-bee and of many ants, could not possibly produce slight deviations of bodily structure. have been thus acquired. No complex instinct can possibly be produced It will be universally admitted that instincts are as through natural selection, except by the slow and important as corporeal structure for the welfare of gradual accumulation of numerous, slight, yet prof- each species, under its present conditions of life. itable, variations. Hence, as in the case of corporeal Under changed conditions of life, it is at least pos- structures, we ought to find in nature, not the actual sible that slight modifications of instinct might be transitional gradations by which each complex in- profitable to a species; and if it can be shown that stinct has been acquired—for these could be found instincts do vary ever so little, then I can see no diffi- only in the lineal ancestors of each species—but we culty in natural selection preserving and continually ought to find in the collateral lines of descent some accumulating variations of instinct to any extent that evidence of such gradations; or we ought at least to may be profitable. It is thus, as I believe, that all the be able to show that gradations of some kind are most complex and wonderful instincts have origi- possible; and this we certainly can do. I have been nated. As modifications of corporeal structure arise surprised to find, making allowance for the instincts from, and are increased by, use or habit, and are di- of animals having been but little observed except in minished or lost by disuse, so I do not doubt it has Europe and North America, and for no instinct be- been with instincts. But I believe that the effects of ing known amongst extinct species, how very gen-

191 On the Origin of Species erally gradations, leading to the most complex in- dozen aphides on a dock-plant, and prevented their stincts, can be discovered. The canon of ‘Natura non attendance during several hours. After this interval, facit saltum’ applies with almost equal force to in- I felt sure that the aphides would want to excrete. I stincts as to bodily organs. Changes of instinct may watched them for some time through a lens, but not sometimes be facilitated by the same species having one excreted; I then tickled and stroked them with a different instincts at different periods of life, or at hair in the same manner, as well as I could, as the different seasons of the year, or when placed under ants do with their antennae; but not one excreted. different circumstances, &c.; in which case either one Afterwards I allowed an ant to visit them, and it im- or the other instinct might be preserved by natural mediately seemed, by its eager way of running selection. And such instances of diversity of instinct about, to be well aware what a rich flock it had dis- in the same species can be shown to occur in nature. covered; it then began to play with its antennae on Again as in the case of corporeal structure, and con- the abdomen first of one aphis and then of another; formably with my theory, the instinct of each spe- and each aphis, as soon as it felt the antennae, im- cies is good for itself, but has never, as far as we can mediately lifted up its abdomen and excreted a lim- judge, been produced for the exclusive good of oth- pid drop of sweet juice, which was eagerly devoured ers. One of the strongest instances of an animal ap- by the ant. Even the quite young aphides behaved parently performing an action for the sole good of in this manner, showing that the action was instinc- another, with which I am acquainted, is that of tive, and not the result of experience. But as the ex- aphides voluntarily yielding their sweet excretion cretion is extremely viscid, it is probably a conve- to ants: that they do so voluntarily, the following facts nience to the aphides to have it removed; and there- show. I removed all the ants from a group of about a fore probably the aphides do not instinctively ex-

192 Charles Darwin crete for the sole good of the ants. Although I do not unknown to us: Audubon has given several remark- believe that any animal in the world performs an able cases of differences in nests of the same species action for the exclusive good of another of a distinct in the northern and southern United States. Fear of species, yet each species tries to take advantage of any particular enemy is certainly an instinctive qual- the instincts of others, as each takes advantage of the ity, as may be seen in nestling birds, though it is weaker bodily structure of others. So again, in some strengthened by experience, and by the sight of fear few cases, certain instincts cannot be considered as of the same enemy in other animals. But fear of man absolutely perfect; but as details on this and other is slowly acquired, as I have elsewhere shown, by such points are not indispensable, they may be here various animals inhabiting desert islands; and we passed over. may see an instance of this, even in England, in the As some degree of variation in instincts under a greater wildness of all our large birds than of our state of nature, and the inheritance of such variations, small birds; for the large birds have been most per- are indispensable for the action of natural selection, secuted by man. We may safely attribute the greater as many instances as possible ought to have been wildness of our large birds to this cause; for in unin- here given; but want of space prevents me. I can only habited islands large birds are not more fearful than assert, that instincts certainly do vary—for instance, small; and the magpie, so wary in England, is tame the migratory instinct, both in extent and direction, in Norway, as is the hooded crow in Egypt. and in its total loss. So it is with the nests of birds, That the general disposition of individuals of the which vary partly in dependence on the situations same species, born in a state of nature, is extremely chosen, and on the nature and temperature of the diversified, can be shown by a multitude of facts. country inhabited, but often from causes wholly Several cases also, could be given, of occasional and

193 On the Origin of Species strange habits in certain species, which might, if ad- stance) will sometimes point and even back other vantageous to the species, give rise, through natural dogs the very first time that they are taken out; re- selection, to quite new instincts. But I am well aware trieving is certainly in some degree inherited by re- that these general statements, without facts given in trievers; and a tendency to run round, instead of at, detail, can produce but a feeble effect on the reader’s a flock of sheep, by shepherd-dogs. I cannot see that mind. I can only repeat my assurance, that I do not these actions, performed without experience by the speak without good evidence. young, and in nearly the same manner by each indi- The possibility, or even probability, of inherited vidual, performed with eager delight by each breed, variations of instinct in a state of nature will be and without the end being known,—for the young strengthened by briefly considering a few cases un- pointer can no more know that he points to aid his der domestication. We shall thus also be enabled to master, than the white butterfly knows why she lays see the respective parts which habit and the selec- her eggs on the leaf of the cabbage,—I cannot see tion of so-called accidental variations have played that these actions differ essentially from true in- in modifying the mental qualities of our domestic stincts. If we were to see one kind of wolf, when animals. A number of curious and authentic in- young and without any training, as soon as it scented stances could be given of the inheritance of all shades its prey, stand motionless like a statue, and then of disposition and tastes, and likewise of the oddest slowly crawl forward with a peculiar gait; and an- tricks, associated with certain frames of mind or pe- other kind of wolf rushing round, instead of at, a riods of time. But let us look to the familiar case of herd of deer, and driving them to a distant point, we the several breeds of dogs: it cannot be doubted that should assuredly call these actions instinctive. Do- young pointers (I have myself seen a striking in- mestic instincts, as they may be called, are certainly

194 Charles Darwin far less fixed or invariable than natural instincts; but tions which have become inherited solely from long- they have been acted on by far less rigorous selec- continued and compulsory habit, but this, I think, is tion, and have been transmitted for an incompara- not true. No one would ever have thought of teach- bly shorter period, under less fixed conditions of life. ing, or probably could have taught, the tumbler-pi- How strongly these domestic instincts, habits, and geon to tumble,—an action which, as I have wit- dispositions are inherited, and how curiously they nessed, is performed by young birds, that have never become mingled, is well shown when different seen a pigeon tumble. We may believe that some breeds of dogs are crossed. Thus it is known that a one pigeon showed a slight tendency to this strange cross with a bull-dog has affected for many genera- habit, and that the long-continued selection of the tions the courage and obstinacy of greyhounds; and best individuals in successive generations made a cross with a greyhound has given to a whole fam- tumblers what they now are; and near Glasgow there ily of shepherd-dogs a tendency to hunt hares. These are house-tumblers, as I hear from Mr. Brent, which domestic instincts, when thus tested by crossing, re- cannot fly eighteen inches high without going head semble natural instincts, which in a like manner be- over heels. It may be doubted whether any one come curiously blended together, and for a long would have thought of training a dog to point, had period exhibit traces of the instincts of either parent: not some one dog naturally shown a tendency in this for example, Le Roy describes a dog, whose great- line; and this is known occasionally to happen, as I grandfather was a wolf, and this dog showed a trace once saw in a pure terrier. When the first tendency of its wild parentage only in one way, by not com- was once displayed, methodical selection and the ing in a straight line to his master when called. inherited effects of compulsory training in each suc- Domestic instincts are sometimes spoken of as ac- cessive generation would soon complete the work;

195 On the Origin of Species and unconscious selection is still at work, as each the cat genus, when kept tame, are most eager to at- man tries to procure, without intending to improve tack poultry, sheep, and pigs; and this tendency has the breed, dogs which will stand and hunt best. On been found incurable in dogs which have been the other hand, habit alone in some cases has suf- brought home as puppies from countries, such as ficed; no animal is more difficult to tame than the Tierra del Fuego and Australia, where the savages young of the wild rabbit; scarcely any animal is tamer do not keep these domestic animals. How rarely, on than the young of the tame rabbit; but I do not sup- the other hand, do our civilised dogs, even when pose that domestic rabbits have ever been selected quite young, require to be taught not to attack poul- for tameness; and I presume that we must attribute try, sheep, and pigs! No doubt they occasionally do the whole of the inherited change from extreme wild- make an attack, and are then beaten; and if not cured, ness to extreme tameness, simply to habit and long- they are destroyed; so that habit, with some degree continued close confinement. of selection, has probably concurred in civilising by Natural instincts are lost under domestication: a inheritance our dogs. On the other hand, young remarkable instance of this is seen in those breeds of chickens have lost, wholly by habit, that fear of the fowls which very rarely or never become ‘broody,’ dog and cat which no doubt was originally instinc- that is, never wish to sit on their eggs. Familiarity tive in them, in the same way as it is so plainly in- alone prevents our seeing how universally and stinctive in young pheasants, though reared under a largely the minds of our domestic animals have been hen. It is not that chickens have lost all fear, but fear modified by domestication. It is scarcely possible to only of dogs and cats, for if the hen gives the dan- doubt that the love of man has become instinctive in ger-chuckle, they will run (more especially young the dog. All wolves, foxes, jackals, and species of turkeys) from under her, and conceal themselves in

196 Charles Darwin the surrounding grass or thickets; and this is evi- out of the several which I shall have to discuss in my dently done for the instinctive purpose of allowing, future work,—namely, the instinct which leads the as we see in wild ground-birds, their mother to fly cuckoo to lay her eggs in other birds’ nests; the slave- away. But this instinct retained by our chickens has making instinct of certain ants; and the comb-making become useless under domestication, for the mother- power of the hive-bee: these two latter instincts have hen has almost lost by disuse the power of flight. generally, and most justly, been ranked by natural- Hence, we may conclude, that domestic instincts ists as the most wonderful of all known instincts. have been acquired and natural instincts have been It is now commonly admitted that the more imme- lost partly by habit, and partly by man selecting and diate and final cause of the cuckoo’s instinct is, that accumulating during successive generations, pecu- she lays her eggs, not daily, but at intervals of two liar mental habits and actions, which at first appeared or three days; so that, if she were to make her own from what we must in our ignorance call an accident. nest and sit on her own eggs, those first laid would In some cases compulsory habit alone has sufficed have to be left for some time unincubated, or there to produce such inherited mental changes; in other would be eggs and young birds of different ages in cases compulsory habit has done nothing, and all the same nest. If this were the case, the process of has been the result of selection, pursued both me- laying and hatching might be inconveniently long, thodically and unconsciously; but in most cases, more especially as she has to migrate at a very early probably, habit and selection have acted together. period; and the first hatched young would probably We shall, perhaps, best understand how instincts in have to be fed by the male alone. But the American a state of nature have become modified by selection, cuckoo is in this predicament; for she makes her own by considering a few cases. I will select only three, nest and has eggs and young successively hatched,

197 On the Origin of Species all at the same time. It has been asserted that the birds’ nests, and thus be successful in rearing their American cuckoo occasionally lays her eggs in other young. By a continued process of this nature, I be- birds’ nests; but I hear on the high authority of Dr. lieve that the strange instinct of our cuckoo could Brewer, that this is a mistake. Nevertheless, I could be, and has been, generated. I may add that, accord- give several instances of various birds which have ing to Dr. Gray and to some other observers, the been known occasionally to lay their eggs in other European cuckoo has not utterly lost all maternal birds’ nests. Now let us suppose that the ancient pro- love and care for her own offspring. genitor of our European cuckoo had the habits of The occasional habit of birds laying their eggs in the American cuckoo; but that occasionally she laid other birds’ nests, either of the same or of a distinct an egg in another bird’s nest. If the old bird profited species, is not very uncommon with the Gallinaceae; by this occasional habit, or if the young were made and this perhaps explains the origin of a singular more vigorous by advantage having been taken of instinct in the allied group of ostriches. For several the mistaken maternal instinct of another bird, than hen ostriches, at least in the case of the American by their own mother’s care, encumbered as she can species, unite and lay first a few eggs in one nest hardly fail to be by having eggs and young of differ- and then in another; and these are hatched by the ent ages at the same time; then the old birds or the males. This instinct may probably be accounted for fostered young would gain an advantage. And anal- by the fact of the hens laying a large number of eggs; ogy would lead me to believe, that the young thus but, as in the case of the cuckoo, at intervals of two reared would be apt to follow by inheritance the oc- or three days. This instinct, however, of the Ameri- casional and aberrant habit of their mother, and in can ostrich has not as yet been perfected; for a sur- their turn would be apt to lay their eggs in other prising number of eggs lie strewed over the plains,

198 Charles Darwin so that in one day’s hunting I picked up no less than sect whose nest and stored food are thus feloniously twenty lost and wasted eggs. appropriated, be not thus exterminated. Many bees are parasitic, and always lay their eggs Slave-making instinct. — This remarkable instinct in the nests of bees of other kinds. This case is more was first discovered in the Formica (Polyerges) remarkable than that of the cuckoo; for these bees rufescens by Pierre Huber, a better observer even have not only their instincts but their structure modi- than his celebrated father. This ant is absolutely de- fied in accordance with their parasitic habits; for they pendent on its slaves; without their aid, the species do not possess the pollen-collecting apparatus which would certainly become extinct in a single year. The would be necessary if they had to store food for their males and fertile females do no work. The workers own young. Some species, likewise, of Sphegidae or sterile females, though most energetic and coura- (wasp-like insects) are parasitic on other species; and geous in capturing slaves, do no other work. They M. Fabre has lately shown good reason for believ- are incapable of making their own nests, or of feeding that although the Tachytes nigra generally makes ing their own larvae. When the old nest is found in- its own burrow and stores it with paralysed prey for convenient, and they have to migrate, it is the slaves its own larvae to feed on, yet that when this insect which determine the migration, and actually carry finds a burrow already made and stored by another their masters in their jaws. So utterly helpless are sphex, it takes advantage of the prize, and becomes the masters, that when Huber shut up thirty of them for the occasion parasitic. In this case, as with the without a slave, but with plenty of the food which supposed case of the cuckoo, I can see no difficulty they like best, and with their larvae and pupae to in natural selection making an occasional habit per- stimulate them to work, they did nothing; they could manent, if of advantage to the species, and if the in- not even feed themselves, and many perished of

199 On the Origin of Species hunger. Huber then introduced a single slave (F. found a few slaves in all. Males and fertile females fusca), and she instantly set to work, fed and saved of the slave-species are found only in their own the survivors; made some cells and tended the lar- proper communities, and have never been observed vae, and put all to rights. What can be more extraor- in the nests of F. sanguinea. The slaves are black and dinary than these well-ascertained facts? If we had not above half the size of their red masters, so that not known of any other slave-making ant, it would the contrast in their appearance is very great. When have been hopeless to have speculated how so won- the nest is slightly disturbed, the slaves occasion- derful an instinct could have been perfected. ally come out, and like their masters are much agi- Formica sanguinea was likewise first discovered tated and defend their nest: when the nest is much by P. Huber to be a slave-making ant. This species is disturbed and the larvae and pupae are exposed, found in the southern parts of England, and its hab- the slaves work energetically with their masters in its have been attended to by Mr. F. Smith, of the Brit- carrying them away to a place of safety. Hence, it is ish Museum, to whom I am much indebted for in- clear, that the slaves feel quite at home. During the formation on this and other subjects. Although fully months of June and July, on three successive years, I trusting to the statements of Huber and Mr. Smith, I have watched for many hours several nests in Sur- tried to approach the subject in a sceptical frame of rey and Sussex, and never saw a slave either leave mind, as any one may well be excused for doubting or enter a nest. As, during these months, the slaves the truth of so extraordinary and odious an instinct are very few in number, I thought that they might as that of making slaves. Hence I will give the obser- behave differently when more numerous; but Mr. vations which I have myself made, in some little Smith informs me that he has watched the nests at detail. I opened fourteen nests of F. sanguinea, and various hours during May, June and August, both

200 Charles Darwin in Surrey and Hampshire, and has never seen the Switzerland than in England. slaves, though present in large numbers in August, One day I fortunately chanced to witness a migra- either leave or enter the nest. Hence he considers them tion from one nest to another, and it was a most in- as strictly household slaves. The masters, on the other teresting spectacle to behold the masters carefully hand, may be constantly seen bringing in materials carrying, as Huber has described, their slaves in their for the nest, and food of all kinds. During the present jaws. Another day my attention was struck by about year, however, in the month of July, I came across a a score of the slave-makers haunting the same spot, community with an unusually large stock of slaves, and evidently not in search of food; they approached and I observed a few slaves mingled with their mas- and were vigorously repulsed by an independent ters leaving the nest, and marching along the same community of the slave species (F. fusca); sometimes road to a tall Scotch-fir-tree, twenty-five yards distant, as many as three of these ants clinging to the legs of which they ascended together, probably in search of the slave-making F. sanguinea. The latter ruthlessly aphides or cocci. According to Huber, who had ample killed their small opponents, and carried their dead opportunities for observation, in Switzerland the bodies as food to their nest, twenty-nine yards dis- slaves habitually work with their masters in making tant; but they were prevented from getting any pu- the nest, and they alone open and close the doors in pae to rear as slaves. I then dug up a small parcel of the morning and evening; and, as Huber expressly the pupae of F. fusca from another nest, and put them states, their principal office is to search for aphides. down on a bare spot near the place of combat; they This difference in the usual habits of the masters and were eagerly seized, and carried off by the tyrants, slaves in the two countries, probably depends merely who perhaps fancied that, after all, they had been on the slaves being captured in greater numbers in victorious in their late combat.

201 On the Origin of Species At the same time I laid on the same place a small earth from the nest of F. flava, and quickly ran away; parcel of the pupae of another species, F. flava, with but in about a quarter of an hour, shortly after all the a few of these little yellow ants still clinging to the little yellow ants had crawled away, they took heart fragments of the nest. This species is sometimes, and carried off the pupae. though rarely, made into slaves, as has been de- One evening I visited another community of F. scribed by Mr. Smith. Although so small a species, it sanguinea, and found a number of these ants enter- is very courageous, and I have seen it ferociously ing their nest, carrying the dead bodies of F. fusca attack other ants. In one instance I found to my sur- (showing that it was not a migration) and numerous prise an independent community of F. flava under a pupae. I traced the returning file burthened with stone beneath a nest of the slave-making F. booty, for about forty yards, to a very thick clump of sanguinea; and when I had accidentally disturbed heath, whence I saw the last individual of F. both nests, the little ants attacked their big sanguinea emerge, carrying a pupa; but I was not neighbours with surprising courage. Now I was cu- able to find the desolated nest in the thick heath. The rious to ascertain whether F. sanguinea could dis- nest, however, must have been close at hand, for two tinguish the pupae of F. fusca, which they habitu- or three individuals of F. fusca were rushing about ally make into slaves, from those of the little and in the greatest agitation, and one was perched mo- furious F. flava, which they rarely capture, and it was tionless with its own pupa in its mouth on the top of evident that they did at once distinguish them: for a spray of heath over its ravaged home. we have seen that they eagerly and instantly seized Such are the facts, though they did not need confir- the pupae of F. fusca, whereas they were much terri- mation by me, in regard to the wonderful instinct of fied when they came across the pupae, or even the making slaves. Let it be observed what a contrast

202 Charles Darwin the instinctive habits of F. sanguinea present with their slaves than they do in Switzerland. those of the F. rufescens. The latter does not build By what steps the instinct of F. sanguinea origi- its own nest, does not determine its own migrations, nated I will not pretend to conjecture. But as ants, does not collect food for itself or its young, and can- which are not slave-makers, will, as I have seen, carry not even feed itself: it is absolutely dependent on its off pupae of other species, if scattered near their numerous slaves. Formica sanguinea, on the other nests, it is possible that pupae originally stored as hand, possesses much fewer slaves, and in the early food might become developed; and the ants thus un- part of the summer extremely few. The masters de- intentionally reared would then follow their proper termine when and where a new nest shall be formed, instincts, and do what work they could. If their pres- and when they migrate, the masters carry the slaves. ence proved useful to the species which had seized Both in Switzerland and England the slaves seem to them—if it were more advantageous to this species have the exclusive care of the larvae, and the mas- to capture workers than to procreate them—the habit ters alone go on slave-making expeditions. In Swit- of collecting pupae originally for food might by natu- zerland the slaves and masters work together, mak- ral selection be strengthened and rendered perma- ing and bringing materials for the nest: both, but nent for the very different purpose of raising slaves. chiefly the slaves, tend, and milk as it may be called, When the instinct was once acquired, if carried out their aphides; and thus both collect food for the com- to a much less extent even than in our British F. munity. In England the masters alone usually leave sanguinea, which, as we have seen, is less aided by the nest to collect building materials and food for its slaves than the same species in Switzerland, I can themselves, their slaves and larvae. So that the mas- see no difficulty in natural selection increasing and ters in this country receive much less service from modifying the instinct—always supposing each

203 On the Origin of Species modification to be of use to the species—until an perceive when they are correctly made. But the dif- ant was formed as abjectly dependent on its slaves ficulty is not nearly so great as it at first appears: all as is the Formica rufescens. this beautiful work can be shown, I think, to follow Cell-making instinct of the Hive-Bee. — I will not from a few very simple instincts. here enter on minute details on this subject, but will I was led to investigate this subject by Mr. merely give an outline of the conclusions at which I Waterhouse, who has shown that the form of the cell have arrived. He must be a dull man who can exam- stands in close relation to the presence of adjoining ine the exquisite structure of a comb, so beautifully cells; and the following view may, perhaps, be con- adapted to its end, without enthusiastic admiration. sidered only as a modification of this theory. Let us We hear from mathematicians that bees have practi- look to the great principle of gradation, and see cally solved a recondite problem, and have made whether Nature does not reveal to us her method of their cells of the proper shape to hold the greatest work. At one end of a short series we have humble- possible amount of honey, with the least possible bees, which use their old cocoons to hold honey, consumption of precious wax in their construction. sometimes adding to them short tubes of wax, and It has been remarked that a skilful workman, with likewise making separate and very irregular fitting tools and measures, would find it very diffi- rounded cells of wax. At the other end of the series cult to make cells of wax of the true form, though we have the cells of the hive-bee, placed in a double this is perfectly effected by a crowd of bees working layer: each cell, as is well known, is an hexagonal in a dark hive. Grant whatever instincts you please, prism, with the basal edges of its six sides bevelled and it seems at first quite inconceivable how they so as to join on to a pyramid, formed of three rhombs. can make all the necessary angles and planes, or even These rhombs have certain angles, and the three

204 Charles Darwin which form the pyramidal base of a single cell on Hence each cell consists of an outer spherical por- one side of the comb, enter into the composition of tion and of two, three, or more perfectly flat surfaces, the bases of three adjoining cells on the opposite side. according as the cell adjoins two, three or more other In the series between the extreme perfection of the cells. When one cell comes into contact with three cells of the hive-bee and the simplicity of those of other cells, which, from the spheres being nearly of the humble-bee, we have the cells of the Mexican the same size, is very frequently and necessarily the Melipona domestica, carefully described and figured case, the three flat surfaces are united into a pyra- by Pierre Huber. The Melipona itself is intermedi- mid; and this pyramid, as Huber has remarked, is ate in structure between the hive and humble bee, manifestly a gross imitation of the three-sided pyra- but more nearly related to the latter: it forms a nearly midal basis of the cell of the hive-bee. As in the cells regular waxen comb of cylindrical cells, in which the of the hive-bee, so here, the three plane surfaces in young are hatched, and, in addition, some large cells any one cell necessarily enter into the construction of wax for holding honey. These latter cells are nearly of three adjoining cells. It is obvious that the spherical and of nearly equal sizes, and are aggre- Melipona saves wax by this manner of building; for gated into an irregular mass. But the important point the flat walls between the adjoining cells are not to notice, is that these cells are always made at that double, but are of the same thickness as the outer degree of nearness to each other, that they would spherical portions, and yet each flat portion forms a have intersected or broken into each other, if the part of two cells. spheres had been completed; but this is never per- Reflecting on this case, it occurred to me that if the mitted, the bees building perfectly flat walls of wax Melipona had made its spheres at some given dis- between the spheres which thus tend to intersect. tance from each other, and had made them of equal

205 On the Origin of Species sizes and had arranged them symmetrically in a Hence we may safely conclude that if we could double layer, the resulting structure would probably slightly modify the instincts already possessed by have been as perfect as the comb of the hive-bee. the Melipona, and in themselves not very wonder- Accordingly I wrote to Professor Miller, of Cam- ful, this bee would make a structure as wonderfully bridge, and this geometer has kindly read over the perfect as that of the hive-bee. We must suppose the following statement, drawn up from his information, Melipona to make her cells truly spherical, and of and tells me that it is strictly correct:- equal sizes; and this would not be very surprising, If a number of equal spheres be described with their seeing that she already does so to a certain extent, centres placed in two parallel layers; with the centre of and seeing what perfectly cylindrical burrows in each sphere at the distance of radius x sqrt(2) or radius wood many insects can make, apparently by turn- x 1.41421 (or at some lesser distance), from the centres ing round on a fixed point. We must suppose the of the six surrounding spheres in the same layer; and Melipona to arrange her cells in level layers, as she at the same distance from the centres of the adjoining already does her cylindrical cells; and we must fur- spheres in the other and parallel layer; then, if planes ther suppose, and this is the greatest difficulty, that of intersection between the several spheres in both lay- she can somehow judge accurately at what distance ers be formed, there will result a double layer of hex- to stand from her fellow-labourers when several are agonal prisms united together by pyramidal bases making their spheres; but she is already so far en- formed of three rhombs; and the rhombs and the sides abled to judge of distance, that she always describes of the hexagonal prisms will have every angle identi- her spheres so as to intersect largely; and then she cally the same with the best measurements which have unites the points of intersection by perfectly flat sur- been made of the cells of the hive-bee. faces. We have further to suppose, but this is no dif-

206 Charles Darwin ficulty, that after hexagonal prisms have been formed had begun to excavate these basins near together, by the intersection of adjoining spheres in the same they had begun their work at such a distance from layer, she can prolong the hexagon to any length req- each other, that by the time the basins had acquired uisite to hold the stock of honey; in the same way as the above stated width (i.e. about the width of an the rude humble-bee adds cylinders of wax to the ordinary cell), and were in depth about one sixth of circular mouths of her old cocoons. By such modifi- the diameter of the sphere of which they formed a cations of instincts in themselves not very wonder- part, the rims of the basins intersected or broke into ful,—hardly more wonderful than those which guide each other. As soon as this occurred, the bees ceased a bird to make its nest,—I believe that the hive-bee to excavate, and began to build up flat walls of wax has acquired, through natural selection, her inimi- on the lines of intersection between the basins, so table architectural powers. that each hexagonal prism was built upon the fes- But this theory can be tested by experiment. Fol- tooned edge of a smooth basin, instead of on the lowing the example of Mr. Tegetmeier, I separated straight edges of a three-sided pyramid as in the case two combs, and put between them a long, thick, of ordinary cells. square strip of wax: the bees instantly began to exca- I then put into the hive, instead of a thick, square vate minute circular pits in it; and as they deepened piece of wax, a thin and narrow, knife-edged ridge, these little pits, they made them wider and wider coloured with vermilion. The bees instantly began until they were converted into shallow basins, ap- on both sides to excavate little basins near to each pearing to the eye perfectly true or parts of a sphere, other, in the same way as before; but the ridge of and of about the diameter of a cell. It was most inter- wax was so thin, that the bottoms of the basins, if esting to me to observe that wherever several bees they had been excavated to the same depth as in the

207 On the Origin of Species former experiment, would have broken into each Considering how flexible thin wax is, I do not see other from the opposite sides. The bees, however, that there is any difficulty in the bees, whilst at work did not suffer this to happen, and they stopped their on the two sides of a strip of wax, perceiving when excavations in due time; so that the basins, as soon they have gnawed the wax away to the proper thin- as they had been a little deepened, came to have flat ness, and then stopping their work. In ordinary bottoms; and these flat bottoms, formed by thin little combs it has appeared to me that the bees do not plates of the vermilion wax having been left always succeed in working at exactly the same rate ungnawed, were situated, as far as the eye could from the opposite sides; for I have noticed half-com- judge, exactly along the planes of imaginary inter- pleted rhombs at the base of a just-commenced cell, section between the basins on the opposite sides of which were slightly concave on one side, where I the ridge of wax. In parts, only little bits, in other suppose that the bees had excavated too quickly, and parts, large portions of a rhombic plate had been left convex on the opposed side, where the bees had between the opposed basins, but the work, from the worked less quickly. In one well-marked instance, I unnatural state of things, had not been neatly per- put the comb back into the hive, and allowed the formed. The bees must have worked at very nearly bees to go on working for a short time, and again the same rate on the opposite sides of the ridge of examined the cell, and I found that the rhombic plate vermilion wax, as they circularly gnawed away and had been completed, and had become perfectly flat: deepened the basins on both sides, in order to have it was absolutely impossible, from the extreme thin- succeeded in thus leaving flat plates between the ness of the little rhombic plate, that they could have basins, by stopping work along the intermediate effected this by gnawing away the convex side; and planes or planes of intersection. I suspect that the bees in such cases stand in the op-

208 Charles Darwin posed cells and push and bend the ductile and warm menced. Some of these statements differ from those wax (which as I have tried is easily done) into its made by the justly celebrated elder Huber, but I am proper intermediate plane, and thus flatten it. convinced of their accuracy; and if I had space, I could From the experiment of the ridge of vermilion wax, show that they are conformable with my theory. we can clearly see that if the bees were to build for Huber’s statement that the very first cell is exca- themselves a thin wall of wax, they could make their vated out of a little parallel-sided wall of wax, is not, cells of the proper shape, by standing at the proper as far as I have seen, strictly correct; the first com- distance from each other, by excavating at the same mencement having always been a little hood of wax; rate, and by endeavouring to make equal spherical but I will not here enter on these details. We see how hollows, but never allowing the spheres to break into important a part excavation plays in the construc- each other. Now bees, as may be clearly seen by ex- tion of the cells; but it would be a great error to sup- amining the edge of a growing comb, do make a pose that the bees cannot build up a rough wall of rough, circumferential wall or rim all round the wax in the proper position—that is, along the plane comb; and they gnaw into this from the opposite of intersection between two adjoining spheres. I have sides, always working circularly as they deepen each several specimens showing clearly that they can do cell. They do not make the whole three-sided pyra- this. Even in the rude circumferential rim or wall of midal base of any one cell at the same time, but only wax round a growing comb, flexures may sometimes the one rhombic plate which stands on the extreme be observed, corresponding in position to the planes growing margin, or the two plates, as the case may of the rhombic basal plates of future cells. But the be; and they never complete the upper edges of the rough wall of wax has in every case to be finished rhombic plates, until the hexagonal walls are com- off, by being largely gnawed away on both sides.

209 On the Origin of Species The manner in which the bees build is curious; they It seems at first to add to the difficulty of under- always make the first rough wall from ten to twenty standing how the cells are made, that a multitude of times thicker than the excessively thin finished wall bees all work together; one bee after working a short of the cell, which will ultimately be left. We shall time at one cell going to another, so that, as Huber understand how they work, by supposing masons has stated, a score of individuals work even at the first to pile up a broad ridge of cement, and then to commencement of the first cell. I was able practically begin cutting it away equally on both sides near the to show this fact, by covering the edges of the hex- ground, till a smooth, very thin wall is left in the agonal walls of a single cell, or the extreme margin middle; the masons always piling up the cut-away of the circumferential rim of a growing comb, with cement, and adding fresh cement, on the summit of an extremely thin layer of melted vermilion wax; and the ridge. We shall thus have a thin wall steadily I invariably found that the colour was most delicately growing upward; but always crowned by a gigantic diffused by the bees—as delicately as a painter could coping. From all the cells, both those just commenced have done with his brush—by atoms of the coloured and those completed, being thus crowned by a strong wax having been taken from the spot on which it had coping of wax, the bees can cluster and crawl over been placed, and worked into the growing edges of the comb without injuring the delicate hexagonal the cells all round. The work of construction seems walls, which are only about one four-hundredth of to be a sort of balance struck between many bees, all an inch in thickness; the plates of the pyramidal ba- instinctively standing at the same relative distance sis being about twice as thick. By this singular man- from each other, all trying to sweep equal spheres, ner of building, strength is continually given to the and then building up, or leaving ungnawed, the comb, with the utmost ultimate economy of wax. planes of intersection between these spheres. It was

210 Charles Darwin really curious to note in cases of difficulty, as when of laying down under certain circumstances a rough two pieces of comb met at an angle, how often the wall in its proper place between two just-commenced bees would entirely pull down and rebuild in dif- cells, is important, as it bears on a fact, which seems ferent ways the same cell, sometimes recurring to a at first quite subversive of the foregoing theory; shape which they had at first rejected. namely, that the cells on the extreme margin of wasp- When bees have a place on which they can stand combs are sometimes strictly hexagonal; but I have in their proper positions for working,—for instance, not space here to enter on this subject. Nor does there on a slip of wood, placed directly under the middle seem to me any great difficulty in a single insect (as of a comb growing downwards so that the comb has in the case of a queen-wasp) making hexagonal cells, to be built over one face of the slip—in this case the if she work alternately on the inside and outside of bees can lay the foundations of one wall of a new two or three cells commenced at the same time, al- hexagon, in its strictly proper place, projecting be- ways standing at the proper relative distance from yond the other completed cells. It suffices that the the parts of the cells just begun, sweeping spheres bees should be enabled to stand at their proper rela- or cylinders, and building up intermediate planes. tive distances from each other and from the walls of It is even conceivable that an insect might, by fixing the last completed cells, and then, by striking imagi- on a point at which to commence a cell, and then nary spheres, they can build up a wall intermediate moving outside, first to one point, and then to five between two adjoining spheres; but, as far as I have other points, at the proper relative distances from seen, they never gnaw away and finish off the angles the central point and from each other, strike the of a cell till a large part both of that cell and of the planes of intersection, and so make an isolated hexa- adjoining cells has been built. This capacity in bees gon: but I am not aware that any such case has been

211 On the Origin of Species observed; nor would any good be derived from a cess of secretion. A large store of honey is indispens- single hexagon being built, as in its construction more able to support a large stock of bees during the win- materials would be required than for a cylinder. ter; and the security of the hive is known mainly to As natural selection acts only by the accumulation depend on a large number of bees being supported. of slight modifications of structure or instinct, each Hence the saving of wax by largely saving honey profitable to the individual under its conditions of must be a most important element of success in any life, it may reasonably be asked, how a long and family of bees. Of course the success of any species graduated succession of modified architectural in- of bee may be dependent on the number of its para- stincts, all tending towards the present perfect plan sites or other enemies, or on quite distinct causes, of construction, could have profited the progenitors and so be altogether independent of the quantity of of the hive-bee? I think the answer is not difficult: it honey which the bees could collect. But let us sup- is known that bees are often hard pressed to get suf- pose that this latter circumstance determined, as it ficient nectar; and I am informed by Mr. Tegetmeier probably often does determine, the numbers of a that it has been experimentally found that no less humble-bee which could exist in a country; and let than from twelve to fifteen pounds of dry sugar are us further suppose that the community lived consumed by a hive of bees for the secretion of each throughout the winter, and consequently required a pound of wax; so that a prodigious quantity of fluid store of honey: there can in this case be no doubt nectar must be collected and consumed by the bees that it would be an advantage to our humble-bee, if in a hive for the secretion of the wax necessary for a slight modification of her instinct led her to make the construction of their combs. Moreover, many bees her waxen cells near together, so as to intersect a little; have to remain idle for many days during the pro- for a wall in common even to two adjoining cells,

212 Charles Darwin would save some little wax. Hence it would continuous, successive, slight modifications of simpler in- ally be more and more advantageous to our humble- stincts; natural selection having by slow degrees, bee, if she were to make her cells more and more more and more perfectly, led the bees to sweep equal regular, nearer together, and aggregated into a mass, spheres at a given distance from each other in a like the cells of the Melipona; for in this case a large double layer, and to build up and excavate the wax part of the bounding surface of each cell would serve along the planes of intersection. The bees, of course, to bound other cells, and much wax would be saved. no more knowing that they swept their spheres at Again, from the same cause, it would be advanta- one particular distance from each other, than they geous to the Melipona, if she were to make her cells know what are the several angles of the hexagonal closer together, and more regular in every way than prisms and of the basal rhombic plates. The motive at present; for then, as we have seen, the spherical power of the process of natural selection having been surfaces would wholly disappear, and would all be economy of wax; that individual swarm which replaced by plane surfaces; and the Melipona would wasted least honey in the secretion of wax, having make a comb as perfect as that of the hive-bee. Be- succeeded best, and having transmitted by inherit- yond this stage of perfection in architecture, natural ance its newly acquired economical instinct to new selection could not lead; for the comb of the hive- swarms, which in their turn will have had the best bee, as far as we can see, is absolutely perfect in chance of succeeding in the struggle for existence. economising wax. No doubt many instincts of very difficult explana- Thus, as I believe, the most wonderful of all known tion could be opposed to the theory of natural selec- instincts, that of the hive-bee, can be explained by tion,—cases, in which we cannot see how an instinct natural selection having taken advantage of numer- could possibly have originated; cases, in which no

213 On the Origin of Species intermediate gradations are known to exist; cases of than that of any other striking modification of struc- instinct of apparently such trifling importance, that ture; for it can be shown that some insects and other they could hardly have been acted on by natural se- articulate animals in a state of nature occasionally lection; cases of instincts almost identically the same become sterile; and if such insects had been social, in animals so remote in the scale of nature, that we and it had been profitable to the community that a cannot account for their similarity by inheritance number should have been annually born capable of from a common parent, and must therefore believe work, but incapable of procreation, I can see no very that they have been acquired by independent acts of great difficulty in this being effected by natural se- natural selection. I will not here enter on these sev- lection. But I must pass over this preliminary diffi- eral cases, but will confine myself to one special dif- culty. The great difficulty lies in the working ants ficulty, which at first appeared to me insuperable, differing widely from both the males and the fertile and actually fatal to my whole theory. I allude to the females in structure, as in the shape of the thorax neuters or sterile females in insect-communities: for and in being destitute of wings and sometimes of these neuters often differ widely in instinct and in eyes, and in instinct. As far as instinct alone is con- structure from both the males and fertile females, cerned, the prodigious difference in this respect be- and yet, from being sterile, they cannot propagate tween the workers and the perfect females, would their kind. have been far better exemplified by the hive-bee. If a The subject well deserves to be discussed at great working ant or other neuter insect had been an ani- length, but I will here take only a single case, that of mal in the ordinary state, I should have working or sterile ants. How the workers have been unhesitatingly assumed that all its characters had rendered sterile is a difficulty; but not much greater been slowly acquired through natural selection;

214 Charles Darwin namely, by an individual having been born with sex; for oxen of certain breeds have longer horns than some slight profitable modification of structure, this in other breeds, in comparison with the horns of the being inherited by its offspring, which again varied bulls or cows of these same breeds. Hence I can see and were again selected, and so onwards. But with no real difficulty in any character having become the working ant we have an insect differing greatly correlated with the sterile condition of certain mem- from its parents, yet absolutely sterile; so that it could bers of insect-communities: the difficulty lies in un- never have transmitted successively acquired modi- derstanding how such correlated modifications of fications of structure or instinct to its progeny. It may structure could have been slowly accumulated by well be asked how is it possible to reconcile this case natural selection. with the theory of natural selection? This difficulty, though appearing insuperable, is First, let it be remembered that we have innumer- lessened, or, as I believe, disappears, when it is re- able instances, both in our domestic productions and membered that selection may be applied to the fam- in those in a state of nature, of all sorts of differences ily, as well as to the individual, and may thus gain of structure which have become correlated to cer- the desired end. Thus, a well-flavoured vegetable is tain ages, and to either sex. We have differences cor- cooked, and the individual is destroyed; but the hor- related not only to one sex, but to that short period ticulturist sows seeds of the same stock, and confi- alone when the reproductive system is active, as in dently expects to get nearly the same variety; breed- the nuptial plumage of many birds, and in the ers of cattle wish the flesh and fat to be well marbled hooked jaws of the male salmon. We have even slight together; the animal has been slaughtered, but the differences in the horns of different breeds of cattle breeder goes with confidence to the same family. I in relation to an artificially imperfect state of the male have such faith in the powers of selection, that I do

215 On the Origin of Species not doubt that a breed of cattle, always yielding oxen males, but from each other, sometimes to an almost with extraordinarily long horns, could be slowly incredible degree, and are thus divided into two or formed by carefully watching which individual bulls even three castes. The castes, moreover, do not gen- and cows, when matched, produced oxen with the erally graduate into each other, but are perfectly well longest horns; and yet no one ox could ever have defined; being as distinct from each other, as are any propagated its kind. Thus I believe it has been with two species of the same genus, or rather as any two social insects: a slight modification of structure, or genera of the same family. Thus in Eciton, there are instinct, correlated with the sterile condition of cer- working and soldier neuters, with jaws and instincts tain members of the community, has been advanta- extraordinarily different: in Cryptocerus, the work- geous to the community: consequently the fertile ers of one caste alone carry a wonderful sort of shield males and females of the same community flour- on their heads, the use of which is quite unknown: ished, and transmitted to their fertile offspring a ten- in the Mexican Myrmecocystus, the workers of one dency to produce sterile members having the same caste never leave the nest; they are fed by the work- modification. And I believe that this process has been ers of another caste, and they have an enormously repeated, until that prodigious amount of difference developed abdomen which secretes a sort of honey, between the fertile and sterile females of the same supplying the place of that excreted by the aphides, species has been produced, which we see in many or the domestic cattle as they may be called, which social insects. our European ants guard or imprison. But we have not as yet touched on the climax of It will indeed be thought that I have an overween- the difficulty; namely, the fact that the neuters of sev- ing confidence in the principle of natural selection, eral ants differ, not only from the fertile females and when I do not admit that such wonderful and well-

216 Charles Darwin established facts at once annihilate my theory. In the be perfectly linked together by individuals taken out simpler case of neuter insects all of one caste or of of the same nest: I have myself compared perfect the same kind, which have been rendered by natu- gradations of this kind. It often happens that the ral selection, as I believe to be quite possible, differ- larger or the smaller sized workers are the most nu- ent from the fertile males and females,—in this case, merous; or that both large and small are numerous, we may safely conclude from the analogy of ordi- with those of an intermediate size scanty in num- nary variations, that each successive, slight, profit- bers. Formica flava has larger and smaller workers, able modification did not probably at first appear in with some of intermediate size; and, in this species, all the individual neuters in the same nest, but in a as Mr. F. Smith has observed, the larger workers have few alone; and that by the long-continued selection simple eyes (ocelli), which though small can be of the fertile parents which produced most neuters plainly distinguished, whereas the smaller workers with the profitable modification, all the neuters ulti- have their ocelli rudimentary. Having carefully dis- mately came to have the desired character. On this sected several specimens of these workers, I can af- view we ought occasionally to find neuter-insects of firm that the eyes are far more rudimentary in the the same species, in the same nest, presenting gra- smaller workers than can be accounted for merely dations of structure; and this we do find, even often, by their proportionally lesser size; and I fully be- considering how few neuter-insects out of Europe lieve, though I dare not assert so positively, that the have been carefully examined. Mr. F. Smith has workers of intermediate size have their ocelli in an shown how surprisingly the neuters of several Brit- exactly intermediate condition. So that we here have ish ants differ from each other in size and sometimes two bodies of sterile workers in the same nest, dif- in colour; and that the extreme forms can sometimes fering not only in size, but in their organs of vision,

217 On the Origin of Species yet connected by some few members in an interme- tration: the difference was the same as if we were to diate condition. I may digress by adding, that if the see a set of workmen building a house of whom smaller workers had been the most useful to the com- many were five feet four inches high, and many six- munity, and those males and females had been con- teen feet high; but we must suppose that the larger tinually selected, which produced more and more workmen had heads four instead of three times as of the smaller workers, until all the workers had big as those of the smaller men, and jaws nearly five come to be in this condition; we should then have times as big. The jaws, moreover, of the working ants had a species of ant with neuters very nearly in the of the several sizes differed wonderfully in shape, same condition with those of Myrmica. For the work- and in the form and number of the teeth. But the im- ers of Myrmica have not even rudiments of ocelli, portant fact for us is, that though the workers can be though the male and female ants of this genus have grouped into castes of different sizes, yet they gradu- well-developed ocelli. ate insensibly into each other, as does the widely- I may give one other case: so confidently did I ex- different structure of their jaws. I speak confidently pect to find gradations in important points of struc- on this latter point, as Mr. Lubbock made drawings ture between the different castes of neuters in the for me with the camera lucida of the jaws which I same species, that I gladly availed myself of Mr. F. had dissected from the workers of the several sizes. Smith’s offer of numerous specimens from the same With these facts before me, I believe that natural nest of the driver ant (Anomma) of West Africa. The selection, by acting on the fertile parents, could form reader will perhaps best appreciate the amount of a species which should regularly produce neuters, difference in these workers, by my giving not the either all of large size with one form of jaw, or all of actual measurements, but a strictly accurate illus- small size with jaws having a widely different struc-

218 Charles Darwin ture; or lastly, and this is our climax of difficulty, ers being sterile; for had they been fertile, they would one set of workers of one size and structure, and si- have intercrossed, and their instincts and structure multaneously another set of workers of a different would have become blended. And nature has, as I size and structure;—a graduated series having been believe, effected this admirable division of labour first formed, as in the case of the driver ant, and then in the communities of ants, by the means of natural the extreme forms, from being the most useful to the selection. But I am bound to confess, that, with all community, having been produced in greater and my faith in this principle, I should never have antici- greater numbers through the natural selection of the pated that natural selection could have been efficient parents which generated them; until none with an in so high a degree, had not the case of these neuter intermediate structure were produced. insects convinced me of the fact. I have, therefore, Thus, as I believe, the wonderful fact of two dis- discussed this case, at some little but wholly insuffi- tinctly defined castes of sterile workers existing in cient length, in order to show the power of natural the same nest, both widely different from each other selection, and likewise because this is by far the most and from their parents, has originated. We can see serious special difficulty, which my theory has en- how useful their production may have been to a so- countered. The case, also, is very interesting, as it cial community of insects, on the same principle that proves that with animals, as with plants, any amount the division of labour is useful to civilised man. As of modification in structure can be effected by the ants work by inherited instincts and by inherited accumulation of numerous, slight, and as we must tools or weapons, and not by acquired knowledge call them accidental, variations, which are in any and manufactured instruments, a perfect division of manner profitable, without exercise or habit having labour could be effected with them only by the work- come into play. For no amount of exercise, or habit,

219 On the Origin of Species or volition, in the utterly sterile members of a com- judgment, annihilate it. On the other hand, the fact munity could possibly have affected the structure that instincts are not always absolutely perfect and or instincts of the fertile members, which alone leave are liable to mistakes;—that no instinct has been pro- descendants. I am surprised that no one has adduced for the exclusive good of other animals, but vanced this demonstrative case of neuter insects, that each animal takes advantage of the instincts of against the well-known doctrine of Lamarck. others;—that the canon in natural history, of ‘natura non facit saltum’ is applicable to instincts as well as Summary. — I have endeavoured briefly in this to corporeal structure, and is plainly explicable on chapter to show that the mental qualities of our do- the foregoing views, but is otherwise inexplicable,— mestic animals vary, and that the variations are in- all tend to corroborate the theory of natural selection. herited. Still more briefly I have attempted to show This theory is, also, strengthened by some few other that instincts vary slightly in a state of nature. No facts in regard to instincts; as by that common case one will dispute that instincts are of the highest im- of closely allied, but certainly distinct, species, when portance to each animal. Therefore I can see no diffi- inhabiting distant parts of the world and living un- culty, under changing conditions of life, in natural der considerably different conditions of life, yet of- selection accumulating slight modifications of in- ten retaining nearly the same instincts. For instance, stinct to any extent, in any useful direction. In some we can understand on the principle of inheritance, cases habit or use and disuse have probably come how it is that the thrush of South America lines its into play. I do not pretend that the facts given in this nest with mud, in the same peculiar manner as does chapter strengthen in any great degree my theory; our British thrush: how it is that the male wrens (Tro- but none of the cases of difficulty, to the best of my glodytes) of North America, build ‘cock-nests,’ to

220 Charles Darwin roost in, like the males of our distinct Kitty-wrens,— Chapter VIII a habit wholly unlike that of any other known bird. Finally, it may not be a logical deduction, but to my Hybridism imagination it is far more satisfactory to look at such instincts as the young cuckoo ejecting its foster-broth- Distinction between the sterility of first crosses and ers,—ants making slaves,—the larvae of of hybrids —Sterility various in degree, not univer- ichneumonidae feeding within the live bodies of cat- sal, affected by close interbreeding, removed by do- erpillars,—not as specially endowed or created in- mestication — Laws governing the sterility of hy- stincts, but as small consequences of one general law, brids — Sterility not a special endowment, but inci- leading to the advancement of all organic beings, dental on other differences — Causes of the sterility namely, multiply, vary, let the strongest live and the of first crosses and of hybrids —Parallelism between weakest die. the effects of changed conditions of life and crossing — Fertility of varieties when crossed and of their mongrel offspring not universal — Hybrids and mongrels compared independently of their fertility — Summary.

THE VIEW GENERALLY ENTERTAINED by naturalists is that species, when intercrossed, have been specially endowed with the quality of sterility, in order to prevent the confusion of all organic forms. This view

221 On the Origin of Species certainly seems at first probable, for species within on the other hand, have their reproductive organs the same country could hardly have kept distinct had functionally impotent, as may be clearly seen in the they been capable of crossing freely. The importance state of the male element in both plants and animals; of the fact that hybrids are very generally sterile, has, though the organs themselves are perfect in struc- I think, been much underrated by some late writers. ture, as far as the microscope reveals. In the first case On the theory of natural selection the case is espe- the two sexual elements which go to form the em- cially important, inasmuch as the sterility of hybrids bryo are perfect; in the second case they are either could not possibly be of any advantage to them, and not at all developed, or are imperfectly developed. therefore could not have been acquired by the con- This distinction is important, when the cause of the tinued preservation of successive profitable degrees sterility, which is common to the two cases, has to of sterility. I hope, however, to be able to show that be considered. The distinction has probably been sterility is not a specially acquired or endowed qual- slurred over, owing to the sterility in both cases be- ity, but is incidental on other acquired differences. ing looked on as a special endowment, beyond the In treating this subject, two classes of facts, to a large province of our reasoning powers. extent fundamentally different, have generally been The fertility of varieties, that is of the forms known confounded together; namely, the sterility of two or believed to have descended from common par- species when first crossed, and the sterility of the ents, when intercrossed, and likewise the fertility of hybrids produced from them. their mongrel offspring, is, on my theory, of equal Pure species have of course their organs of repro- importance with the sterility of species; for it seems duction in a perfect condition, yet when intercrossed to make a broad and clear distinction between vari- they produce either few or no offspring. Hybrids, eties and species.

222 Charles Darwin First, for the sterility of species when crossed and duced: a plant to be hybridised must be castrated, of their hybrid offspring. It is impossible to study and, what is often more important, must be secluded the several memoirs and works of those two consci- in order to prevent pollen being brought to it by in- entious and admirable observers, Kolreuter and sects from other plants. Nearly all the plants Gartner, who almost devoted their lives to this sub- experimentised on by Gartner were potted, and ap- ject, without being deeply impressed with the high parently were kept in a chamber in his house. That generality of some degree of sterility. Kolreuter these processes are often injurious to the fertility of makes the rule universal; but then he cuts the knot, a plant cannot be doubted; for Gartner gives in his for in ten cases in which he found two forms, consid- table about a score of cases of plants which he cas- ered by most authors as distinct species, quite fer- trated, and artificially fertilised with their own pol- tile together, he unhesitatingly ranks them as variet- len, and (excluding all cases such as the ies. Gartner, also, makes the rule equally universal; Leguminosae, in which there is an acknowledged and he disputes the entire fertility of Kolreuter’s ten difficulty in the manipulation) half of these twenty cases. But in these and in many other cases, Gartner plants had their fertility in some degree impaired. is obliged carefully to count the seeds, in order to Moreover, as Gartner during several years repeat- show that there is any degree of sterility. He always edly crossed the primrose and cowslip, which we compares the maximum number of seeds produced have such good reason to believe to be varieties, and by two species when crossed and by their hybrid only once or twice succeeded in getting fertile seed; offspring, with the average number produced by as he found the common red and blue pimpernels both pure parent-species in a state of nature. But a (Anagallis arvensis and coerulea), which the best serious cause of error seems to me to be here intro- botanists rank as varieties, absolutely sterile to-

223 On the Origin of Species gether; and as he came to the same conclusion in sev- ity adduced by different hybridisers, or by the same eral other analogous cases; it seems to me that we author, from experiments made during different may well be permitted to doubt whether many other years. It can thus be shown that neither sterility nor species are really so sterile, when intercrossed, as fertility affords any clear distinction between species Gartner believes. and varieties; but that the evidence from this source It is certain, on the one hand, that the sterility of vari- graduates away, and is doubtful in the same degree ous species when crossed is so different in degree and as is the evidence derived from other constitutional graduates away so insensibly, and, on the other hand, and structural differences. that the fertility of pure species is so easily affected In regard to the sterility of hybrids in successive by various circumstances, that for all practical pur- generations; though Gartner was enabled to rear poses it is most difficult to say where perfect fertility some hybrids, carefully guarding them from a cross ends and sterility begins. I think no better evidence with either pure parent, for six or seven, and in one of this can be required than that the two most experi- case for ten generations, yet he asserts positively that enced observers who have ever lived, namely, their fertility never increased, but generally greatly Kolreuter and Gartner, should have arrived at dia- decreased. I do not doubt that this is usually the case, metrically opposite conclusions in regard to the very and that the fertility often suddenly decreases in the same species. It is also most instructive to compare— first few generations. Nevertheless I believe that in but I have not space here to enter on details—the evi- all these experiments the fertility has been dimin- dence advanced by our best botanists on the ques- ished by an independent cause, namely, from close tion whether certain doubtful forms should be ranked interbreeding. I have collected so large a body of as species or varieties, with the evidence from fertil- facts, showing that close interbreeding lessens fer-

224 Charles Darwin tility, and, on the other hand, that an occasional cross as from the anthers of the flower itself which is to be with a distinct individual or variety increases fertil- fertilised; so that a cross between two flowers, though ity, that I cannot doubt the correctness of this almost probably on the same plant, would be thus effected. universal belief amongst breeders. Hybrids are sel- Moreover, whenever complicated experiments are in dom raised by experimentalists in great numbers; progress, so careful an observer as Gartner would and as the parent-species, or other allied hybrids, have castrated his hybrids, and this would have in- generally grow in the same garden, the visits of in- sured in each generation a cross with the pollen from sects must be carefully prevented during the flow- a distinct flower, either from the same plant or from ering season: hence hybrids will generally be another plant of the same hybrid nature. And thus, fertilised during each generation by their own indi- the strange fact of the increase of fertility in the suc- vidual pollen; and I am convinced that this would be cessive generations of artificially fertilised hybrids injurious to their fertility, already lessened by their may, I believe, be accounted for by close interbreed- hybrid origin. I am strengthened in this conviction by ing having been avoided. a remarkable statement repeatedly made by Gartner, Now let us turn to the results arrived at by the third namely, that if even the less fertile hybrids be artifi- most experienced hybridiser, namely, the Hon. and cially fertilised with hybrid pollen of the same kind, Rev. W. Herbert. He is as emphatic in his conclusion their fertility, notwithstanding the frequent ill effects that some hybrids are perfectly fertile—as fertile as of manipulation, sometimes decidedly increases, and the pure parent-species—as are Kolreuter and goes on increasing. Now, in artificial fertilisation pol- Gartner that some degree of sterility between dis- len is as often taken by chance (as I know from my tinct species is a universal law of nature. He own experience) from the anthers of another flower, experimentised on some of the very same species as

225 On the Origin of Species did Gartner. The difference in their results may, I tinct species. So that certain individual plants and think, be in part accounted for by Herbert’s great all the individuals of certain species can actually be horticultural skill, and by his having hothouses at hybridised much more readily than they can be self- his command. Of his many important statements I fertilised! For instance, a bulb of Hippeastrum will here give only a single one as an example, aulicum produced four flowers; three were fertilised namely, that ‘every ovule in a pod of Crinum capense by Herbert with their own pollen, and the fourth was fertilised by C. revolutum produced a plant, which subsequently fertilised by the pollen of a compound (he says) I never saw to occur in a case of its natural hybrid descended from three other and distinct spe- fecundation.’ So that we here have perfect, or even cies: the result was that ‘the ovaries of the three first more than commonly perfect, fertility in a first cross flowers soon ceased to grow, and after a few days between two distinct species. perished entirely, whereas the pod impregnated by This case of the Crinum leads me to refer to a most the pollen of the hybrid made vigorous growth and singular fact, namely, that there are individual plants, rapid progress to maturity, and bore good seed, as with certain species of Lobelia, and with all the which vegetated freely.’ In a letter to me, in 1839, species of the genus Hippeastrum, which can be far Mr. Herbert told me that he had then tried the ex- more easily fertilised by the pollen of another and periment during five years, and he continued to try distinct species, than by their own pollen. For these it during several subsequent years, and always with plants have been found to yield seed to the pollen the same result. This result has, also, been confirmed of a distinct species, though quite sterile with their by other observers in the case of Hippeastrum with own pollen, notwithstanding that their own pollen its sub-genera, and in the case of some other genera, was found to be perfectly good, for it fertilised dis- as Lobelia, Passiflora and Verbascum. Although the

226 Charles Darwin plants in these experiments appeared perfectly certain the degree of fertility of some of the complex healthy, and although both the ovules and pollen of crosses of Rhododendrons, and I am assured that the same flower were perfectly good with respect to many of them are perfectly fertile. Mr. C. Noble, for other species, yet as they were functionally imper- instance, informs me that he raises stocks for graft- fect in their mutual self-action, we must infer that ing from a hybrid between Rhod. Ponticum and the plants were in an unnatural state. Nevertheless Catawbiense, and that this hybrid ‘seeds as freely as these facts show on what slight and mysterious it is possible to imagine.’ Had hybrids, when fairly causes the lesser or greater fertility of species when treated, gone on decreasing in fertility in each suc- crossed, in comparison with the same species when cessive generation, as Gartner believes to be the case, self-fertilised, sometimes depends. the fact would have been notorious to nurserymen. The practical experiments of horticulturists, though Horticulturists raise large beds of the same hybrids, not made with scientific precision, deserve some and such alone are fairly treated, for by insect agency notice. It is notorious in how complicated a manner the several individuals of the same hybrid variety the species of Pelargonium, Fuchsia, Calceolaria, are allowed to freely cross with each other, and the Petunia, Rhododendron, &c., have been crossed, yet injurious influence of close interbreeding is thus many of these hybrids seed freely. For instance, prevented. Any one may readily convince himself Herbert asserts that a hybrid from Calceolaria of the efficiency of insect-agency by examining the integrifolia and plantaginea, species most widely flowers of the more sterile kinds of hybrid rhodo- dissimilar in general habit, ‘reproduced itself as per- dendrons, which produce no pollen, for he will find fectly as if it had been a natural species from the on their stigmas plenty of pollen brought from other mountains of Chile.’ I have taken some pains to as- flowers.

227 On the Origin of Species In regard to animals, much fewer experiments have been raised at the same time from different parents, been carefully tried than with plants. If our system- so as to avoid the ill effects of close interbreeding. atic arrangements can be trusted, that is if the genera On the contrary, brothers and sisters have usually of animals are as distinct from each other, as are the been crossed in each successive generation, in op- genera of plants, then we may infer that animals more position to the constantly repeated admonition of widely separated in the scale of nature can be more every breeder. And in this case, it is not at all sur- easily crossed than in the case of plants; but the hy- prising that the inherent sterility in the hybrids brids themselves are, I think, more sterile. I doubt should have gone on increasing. If we were to act whether any case of a perfectly fertile hybrid animal thus, and pair brothers and sisters in the case of any can be considered as thoroughly well authenticated. pure animal, which from any cause had the least ten- It should, however, be borne in mind that, owing to dency to sterility, the breed would assuredly be lost few animals breeding freely under confinement, few in a very few generations. experiments have been fairly tried: for instance, the Although I do not know of any thoroughly well- canary-bird has been crossed with nine other finches, authenticated cases of perfectly fertile hybrid ani- but as not one of these nine species breeds freely in mals, I have some reason to believe that the hybrids confinement, we have no right to expect that the first from Cervulus vaginalis and Reevesii, and from crosses between them and the canary, or that their Phasianus colchicus with P. torquatus and with P. hybrids, should be perfectly fertile. Again, with re- versicolor are perfectly fertile. The hybrids from the spect to the fertility in successive generations of the common and Chinese geese (A. cygnoides), species more fertile hybrid animals, I hardly know of an in- which are so different that they are generally ranked stance in which two families of the same hybrid have in distinct genera, have often bred in this country

228 Charles Darwin with either pure parent, and in one single instance This latter alternative seems to me the most probable, they have bred inter se. This was effected by Mr. and I am inclined to believe in its truth, although it Eyton, who raised two hybrids from the same par- rests on no direct evidence. I believe, for instance, that ents but from different hatches; and from these two our dogs have descended from several wild stocks; birds he raised no less than eight hybrids (grand- yet, with perhaps the exception of certain indigenous children of the pure geese) from one nest. In India, domestic dogs of South America, all are quite fertile however, these cross-bred geese must be far more together; and analogy makes me greatly doubt, fertile; for I am assured by two eminently capable whether the several aboriginal species would at first judges, namely Mr. Blyth and Capt. Hutton, that have freely bred together and have produced quite whole flocks of these crossed geese are kept in vari- fertile hybrids. So again there is reason to believe that ous parts of the country; and as they are kept for our European and the humped Indian cattle are quite profit, where neither pure parent-species exists, they fertile together; but from facts communicated to me must certainly be highly fertile. by Mr. Blyth, I think they must be considered as dis- A doctrine which originated with Pallas, has been tinct species. On this view of the origin of many of largely accepted by modern naturalists; namely, that our domestic animals, we must either give up the most of our domestic animals have descended from belief of the almost universal sterility of distinct spe- two or more aboriginal species, since commingled cies of animals when crossed; or we must look at ste- by intercrossing. On this view, the aboriginal spe- rility, not as an indelible characteristic, but as one ca- cies must either at first have produced quite fertile pable of being removed by domestication. hybrids, or the hybrids must have become in subse- Finally, looking to all the ascertained facts on the quent generations quite fertile under domestication. intercrossing of plants and animals, it may be con-

229 On the Origin of Species cluded that some degree of sterility, both in first fertility, both of first crosses and of hybrids, gradu- crosses and in hybrids, is an extremely general re- ates from zero to perfect fertility. It is surprising in sult; but that it cannot, under our present state of how many curious ways this gradation can be shown knowledge, be considered as absolutely universal. to exist; but only the barest outline of the facts can here be given. When pollen from a plant of one fam- Laws governing the Sterility of first Crosses and of ily is placed on the stigma of a plant of a distinct Hybrids. — We will now consider a little more in family, it exerts no more influence than so much in- detail the circumstances and rules governing the ste- organic dust. From this absolute zero of fertility, the rility of first crosses and of hybrids. Our chief object pollen of different species of the same genus applied will be to see whether or not the rules indicate that to the stigma of some one species, yields a perfect species have specially been endowed with this qual- gradation in the number of seeds produced, up to ity, in order to prevent their crossing and blending nearly complete or even quite complete fertility; and, together in utter confusion. The following rules and as we have seen, in certain abnormal cases, even to conclusions are chiefly drawn up from Gartner’s an excess of fertility, beyond that which the plant’s admirable work on the hybridisation of plants. I have own pollen will produce. So in hybrids themselves, taken much pains to ascertain how far the rules ap- there are some which never have produced, and ply to animals, and considering how scanty our probably never would produce, even with the pol- knowledge is in regard to hybrid animals, I have len of either pure parent, a single fertile seed: but in been surprised to find how generally the same rules some of these cases a first trace of fertility may be apply to both kingdoms. detected, by the pollen of one of the pure parent- It has been already remarked, that the degree of species causing the flower of the hybrid to wither

230 Charles Darwin earlier than it otherwise would have done; and the The fertility, both of first crosses and of hybrids, is early withering of the flower is well known to be a more easily affected by unfavourable conditions, sign of incipient fertilisation. From this extreme de- than is the fertility of pure species. But the degree of gree of sterility we have self-fertilised hybrids pro- fertility is likewise innately variable; for it is not al- ducing a greater and greater number of seeds up to ways the same when the same two species are perfect fertility. crossed under the same circumstances, but depends Hybrids from two species which are very difficult in part upon the constitution of the individuals which to cross, and which rarely produce any offspring, are happen to have been chosen for the experiment. So generally very sterile; but the parallelism between it is with hybrids, for their degree of fertility is often the difficulty of making a first cross, and the sterility found to differ greatly in the several individuals of the hybrids thus produced—two classes of facts raised from seed out of the same capsule and ex- which are generally confounded together—is by no posed to exactly the same conditions. means strict. There are many cases, in which two pure By the term systematic affinity is meant, the resem- species can be united with unusual facility, and pro- blance between species in structure and in constitu- duce numerous hybrid-offspring, yet these hybrids tion, more especially in the structure of parts which are remarkably sterile. On the other hand, there are are of high physiological importance and which dif- species which can be crossed very rarely, or with fer little in the allied species. Now the fertility of first extreme difficulty, but the hybrids, when at last pro- crosses between species, and of the hybrids produced, are very fertile. Even within the limits of the duced from them, is largely governed by their sys- same genus, for instance in Dianthus, these two op- tematic affinity. This is clearly shown by hybrids posite cases occur. never having been raised between species ranked

231 On the Origin of Species by systematists in distinct families; and on the other could be given. hand, by very closely allied species generally unit- No one has been able to point out what kind, or ing with facility. But the correspondence between what amount, of difference in any recognisable char- systematic affinity and the facility of crossing is by acter is sufficient to prevent two species crossing. It no means strict. A multitude of cases could be given can be shown that plants most widely different in of very closely allied species which will not unite, habit and general appearance, and having strongly or only with extreme difficulty; and on the other hand marked differences in every part of the flower, even of very distinct species which unite with the utmost in the pollen, in the fruit, and in the cotyledons, can facility. In the same family there may be a genus, as be crossed. Annual and perennial plants, deciduous Dianthus, in which very many species can most and evergreen trees, plants inhabiting different sta- readily be crossed; and another genus, as Silene, in tions and fitted for extremely different climates, can which the most persevering efforts have failed to often be crossed with ease. produce between extremely close species a single By a reciprocal cross between two species, I mean hybrid. Even within the limits of the same genus, the case, for instance, of a stallion-horse being first we meet with this same difference; for instance, the crossed with a female-ass, and then a male-ass with many species of Nicotiana have been more largely a mare: these two species may then be said to have crossed than the species of almost any other genus; been reciprocally crossed. There is often the widest but Gartner found that N. acuminata, which is not a possible difference in the facility of making recipro- particularly distinct species, obstinately failed to cal crosses. Such cases are highly important, for they fertilise, or to be fertilised by, no less than eight other prove that the capacity in any two species to cross is species of Nicotiana. Very many analogous facts often completely independent of their systematic

232 Charles Darwin affinity, or of any recognisable difference in their a remarkable fact, that hybrids raised from recipro- whole organisation. On the other hand, these cases cal crosses, though of course compounded of the clearly show that the capacity for crossing is con- very same two species, the one species having first nected with constitutional differences imperceptible been used as the father and then as the mother, gen- by us, and confined to the reproductive system. This erally differ in fertility in a small, and occasionally difference in the result of reciprocal crosses between in a high degree. the same two species was long ago observed by Several other singular rules could be given from Kolreuter. To give an instance: Mirabilis jalappa can Gartner: for instance, some species have a remark- easily be fertilised by the pollen of M. longiflora, able power of crossing with other species; other spe- and the hybrids thus produced are sufficiently fer- cies of the same genus have a remarkable power of tile; but Kolreuter tried more than two hundred impressing their likeness on their hybrid offspring; times, during eight following years, to fertilise re- but these two powers do not at all necessarily go ciprocally M. longiflora with the pollen of M. jalappa, together. There are certain hybrids which instead of and utterly failed. Several other equally striking having, as is usual, an intermediate character be- cases could be given. Thuret has observed the same tween their two parents, always closely resemble one fact with certain sea-weeds or Fuci. Gartner, more- of them; and such hybrids, though externally so like over, found that this difference of facility in making one of their pure parent-species, are with rare ex- reciprocal crosses is extremely common in a lesser ceptions extremely sterile. So again amongst hybrids degree. He has observed it even between forms so which are usually intermediate in structure between closely related (as Matthiola annua and glabra) that their parents, exceptional and abnormal individu- many botanists rank them only as varieties. It is also als sometimes are born, which closely resemble one

233 On the Origin of Species of their pure parents; and these hybrids are almost affinity or degree of resemblance to each other. This always utterly sterile, even when the other hybrids latter statement is clearly proved by reciprocal crosses raised from seed from the same capsule have a con- between the same two species, for according as the siderable degree of fertility. These facts show how one species or the other is used as the father or the completely fertility in the hybrid is independent of mother, there is generally some difference, and occa- its external resemblance to either pure parent. sionally the widest possible difference, in the facility Considering the several rules now given, which of effecting an union. The hybrids, moreover, pro- govern the fertility of first crosses and of hybrids, duced from reciprocal crosses often differ in fertility. we see that when forms, which must be considered Now do these complex and singular rules indicate as good and distinct species, are united, their fertil- that species have been endowed with sterility sim- ity graduates from zero to perfect fertility, or even ply to prevent their becoming confounded in nature? to fertility under certain conditions in excess. That I think not. For why should the sterility be so ex- their fertility, besides being eminently susceptible tremely different in degree, when various species to favourable and unfavourable conditions, is in- are crossed, all of which we must suppose it would nately variable. That it is by no means always the be equally important to keep from blending to- same in degree in the first cross and in the hybrids gether? Why should the degree of sterility be in- produced from this cross. That the fertility of hybrids nately variable in the individuals of the same spe- is not related to the degree in which they resemble cies? Why should some species cross with facility, in external appearance either parent. And lastly, that and yet produce very sterile hybrids; and other spe- the facility of making a first cross between any two cies cross with extreme difficulty, and yet produce species is not always governed by their systematic fairly fertile hybrids? Why should there often be so

234 Charles Darwin great a difference in the result of a reciprocal cross the capacity of one plant to be grafted or budded on between the same two species? Why, it may even be another is so entirely unimportant for its welfare in asked, has the production of hybrids been permit- a state of nature, I presume that no one will suppose ted? to grant to species the special power of produc- that this capacity is a specially endowed quality, but ing hybrids, and then to stop their further propaga- will admit that it is incidental on differences in the tion by different degrees of sterility, not strictly re- laws of growth of the two plants. We can sometimes lated to the facility of the first union between their see the reason why one tree will not take on another, parents, seems to be a strange arrangement. from differences in their rate of growth, in the hard- The foregoing rules and facts, on the other hand, ness of their wood, in the period of the flow or na- appear to me clearly to indicate that the sterility both ture of their sap, &c.; but in a multitude of cases we of first crosses and of hybrids is simply incidental or can assign no reason whatever. Great diversity in the dependent on unknown differences, chiefly in the size of two plants, one being woody and the other reproductive systems, of the species which are herbaceous, one being evergreen and the other de- crossed. The differences being of so peculiar and lim- ciduous, and adaptation to widely different climates, ited a nature, that, in reciprocal crosses between two does not always prevent the two grafting together. species the male sexual element of the one will of- As in hybridisation, so with grafting, the capacity is ten freely act on the female sexual element of the limited by systematic affinity, for no one has been other, but not in a reversed direction. It will be ad- able to graft trees together belonging to quite dis- visable to explain a little more fully by an example tinct families; and, on the other hand, closely allied what I mean by sterility being incidental on other species, and varieties of the same species, can usu- differences, and not a specially endowed quality. As ally, but not invariably, be grafted with ease. But this

235 On the Origin of Species capacity, as in hybridisation, is by no means abso- We have seen that the sterility of hybrids, which lutely governed by systematic affinity. Although have their reproductive organs in an imperfect con- many distinct genera within the same family have dition, is a very different case from the difficulty of been grafted together, in other cases species of the uniting two pure species, which have their repro- same genus will not take on each other. The pear ductive organs perfect; yet these two distinct cases can be grafted far more readily on the quince, which run to a certain extent parallel. Something analogous is ranked as a distinct genus, than on the apple, which occurs in grafting; for Thouin found that three spe- is a member of the same genus. Even different vari- cies of Robinia, which seeded freely on their own eties of the pear take with different degrees of facil- roots, and which could be grafted with no great dif- ity on the quince; so do different varieties of the apri- ficulty on another species, when thus grafted were cot and peach on certain varieties of the plum. rendered barren. On the other hand, certain species As Gartner found that there was sometimes an in- of Sorbus, when grafted on other species, yielded nate difference in different individuals of the same twice as much fruit as when on their own roots. We two species in crossing; so Sagaret believes this to are reminded by this latter fact of the extraordinary be the case with different individuals of the same case of Hippeastrum, Lobelia, &c., which seeded two species in being grafted together. As in recipro- much more freely when fertilised with the pollen of cal crosses, the facility of effecting an union is often distinct species, than when self-fertilised with their very far from equal, so it sometimes is in grafting; own pollen. the common gooseberry, for instance, cannot be We thus see, that although there is a clear and fun- grafted on the currant, whereas the currant will take, damental difference between the mere adhesion of though with difficulty, on the gooseberry. grafted stocks, and the union of the male and female

236 Charles Darwin elements in the act of reproduction, yet that there Causes of the Sterility of first Crosses and of Hy- is a rude degree of parallelism in the results of graft- brids. — We may now look a little closer at the prob- ing and of crossing distinct species. And as we must able causes of the sterility of first crosses and of hy- look at the curious and complex laws governing brids. These two cases are fundamentally different, the facility with which trees can be grafted on each for, as just remarked, in the union of two pure spe- other as incidental on unknown differences in their cies the male and female sexual elements are per- vegetative systems, so I believe that the still more fect, whereas in hybrids they are imperfect. Even in complex laws governing the facility of first crosses, first crosses, the greater or lesser difficulty in effect- are incidental on unknown differences, chiefly in ing a union apparently depends on several distinct their reproductive systems. These differences, in causes. There must sometimes be a physical impos- both cases, follow to a certain extent, as might have sibility in the male element reaching the ovule, as been expected, systematic affinity, by which every would be the case with a plant having a pistil too kind of resemblance and dissimilarity between or- long for the pollen-tubes to reach the ovarium. It has ganic beings is attempted to be expressed. The facts also been observed that when pollen of one species by no means seem to me to indicate that the greater is placed on the stigma of a distantly allied species, or lesser difficulty of either grafting or crossing to- though the pollen-tubes protrude, they do not pen- gether various species has been a special endow- etrate the stigmatic surface. Again, the male element ment; although in the case of crossing, the difficulty may reach the female element, but be incapable of is as important for the endurance and stability of causing an embryo to be developed, as seems to have specific forms, as in the case of grafting it is unim- been the case with some of Thuret’s experiments on portant for their welfare. Fuci. No explanation can be given of these facts, any

237 On the Origin of Species more than why certain trees cannot be grafted on oth- early period; more especially as all very young be- ers. Lastly, an embryo may be developed, and then ings seem eminently sensitive to injurious or unnatu- perish at an early period. This latter alternative has ral conditions of life. not been sufficiently attended to; but I believe, from In regard to the sterility of hybrids, in which the observations communicated to me by Mr. Hewitt, sexual elements are imperfectly developed, the case who has had great experience in hybridising galli- is very different. I have more than once alluded to a naceous birds, that the early death of the embryo is large body of facts, which I have collected, showing a very frequent cause of sterility in first crosses. I that when animals and plants are removed from their was at first very unwilling to believe in this view; as natural conditions, they are extremely liable to have hybrids, when once born, are generally healthy and their reproductive systems seriously affected. This, long-lived, as we see in the case of the common mule. in fact, is the great bar to the domestication of ani- Hybrids, however, are differently circumstanced mals. Between the sterility thus superinduced and before and after birth: when born and living in a coun- that of hybrids, there are many points of similarity. try where their two parents can live, they are gener- In both cases the sterility is independent of general ally placed under suitable conditions of life. But a health, and is often accompanied by excess of size hybrid partakes of only half of the nature and con- or great luxuriance. In both cases, the sterility oc- stitution of its mother, and therefore before birth, as curs in various degrees; in both, the male element is long as it is nourished within its mother’s womb or the most liable to be affected; but sometimes the fe- within the egg or seed produced by the mother, it male more than the male. In both, the tendency goes may be exposed to conditions in some degree un- to a certain extent with systematic affinity, or whole suitable, and consequently be liable to perish at an groups of animals and plants are rendered impo-

238 Charles Darwin tent by the same unnatural conditions; and whole species, the reproductive system, independently of groups of species tend to produce sterile hybrids. On the general state of health, is affected by sterility in a the other hand, one species in a group will sometimes very similar manner. In the one case, the conditions resist great changes of conditions with unimpaired of life have been disturbed, though often in so slight fertility; and certain species in a group will produce a degree as to be inappreciable by us; in the other unusually fertile hybrids. No one can tell, till he tries, case, or that of hybrids,the external conditions have whether any particular animal will breed under con- remained the same, but the organisation has been finement or any plant seed freely under culture; nor disturbed by two different structures and constitu- can he tell, till he tries, whether any two species of a tions having been blended into one. For it is scarcely genus will produce more or less sterile hybrids. possible that two organisations should be com- Lastly, when organic beings are placed during sev- pounded into one, without some disturbance occur- eral generations under conditions not natural to them, ring in the development, or periodical action, or they are extremely liable to vary, which is due, as I mutual relation of the different parts and organs one believe, to their reproductive systems having been to another, or to the conditions of life. When hybrids specially affected, though in a lesser degree than when are able to breed inter se, they transmit to their off- sterility ensues. So it is with hybrids, for hybrids in spring from generation to generation the same com- successive generations are eminently liable to vary, pounded organisation, and hence we need not be as every experimentalist has observed. surprised that their sterility, though in some degree Thus we see that when organic beings are placed variable, rarely diminishes. under new and unnatural conditions, and when hy- It must, however, be confessed that we cannot un- brids are produced by the unnatural crossing of two derstand, excepting on vague hypotheses, several

239 On the Origin of Species facts with respect to the sterility of hybrids; for in- tubers, &c., from one soil or climate to another, and stance, the unequal fertility of hybrids produced back again. During the convalescence of animals, we from reciprocal crosses; or the increased sterility in plainly see that great benefit is derived from almost those hybrids which occasionally and exceptionally any change in the habits of life. Again, both with plants resemble closely either pure parent. Nor do I pre- and animals, there is abundant evidence, that a cross tend that the foregoing remarks go to the root of the between very distinct individuals of the same spe- matter: no explanation is offered why an organism, cies, that is between members of different strains or when placed under unnatural conditions, is rendered sub-breeds, gives vigour and fertility to the offspring. sterile. All that I have attempted to show, is that in I believe, indeed, from the facts alluded to in our two cases, in some respects allied, sterility is the com- fourth chapter, that a certain amount of crossing is in- mon result,—in the one case from the conditions of dispensable even with hermaphrodites; and that close life having been disturbed, in the other case from interbreeding continued during several generations the organisation having been disturbed by two between the nearest relations, especially if these be organisations having been compounded into one. kept under the same conditions of life, always induces It may seem fanciful, but I suspect that a similar par- weakness and sterility in the progeny. allelism extends to an allied yet very different class Hence it seems that, on the one hand, slight changes of facts. It is an old and almost universal belief, in the conditions of life benefit all organic beings, founded, I think, on a considerable body of evidence, and on the other hand, that slight crosses, that is that slight changes in the conditions of life are benefi- crosses between the males and females of the same cial to all living things. We see this acted on by farm- species which have varied and become slightly dif- ers and gardeners in their frequent exchanges of seed, ferent, give vigour and fertility to the offspring. But

240 Charles Darwin we have seen that greater changes, or changes of a we are immediately involved in hopeless difficul- particular nature, often render organic beings in ties; for if two hitherto reputed varieties be found in some degree sterile; and that greater crosses, that is any degree sterile together, they are at once ranked crosses between males and females which have be- by most naturalists as species. For instance, the blue come widely or specifically different, produce hy- and red pimpernel, the primrose and cowslip, which brids which are generally sterile in some degree. I are considered by many of our best botanists as va- cannot persuade myself that this parallelism is an rieties, are said by Gartner not to be quite fertile when accident or an illusion. Both series of facts seem to crossed, and he consequently ranks them as un- be connected together by some common but undoubted species. If we thus argue in a circle, the fer- known bond, which is essentially related to the prin- tility of all varieties produced under nature will as- ciple of life. suredly have to be granted. Fertility of Varieties when crossed, and of their If we turn to varieties, produced, or supposed to Mongrel off-spring. — It may be urged, as a most have been produced, under domestication, we are forcible argument, that there must be some essential still involved in doubt. For when it is stated, for in- distinction between species and varieties, and that stance, that the German Spitz dog unites more eas- there must be some error in all the foregoing remarks, ily than other dogs with foxes, or that certain South inasmuch as varieties, however much they may dif- American indigenous domestic dogs do not readily fer from each other in external appearance, cross with cross with European dogs, the explanation which perfect facility, and yield perfectly fertile offspring. will occur to everyone, and probably the true one, is I fully admit that this is almost invariably the case. that these dogs have descended from several aborigi- But if we look to varieties produced under nature, nally distinct species. Nevertheless the perfect fer-

241 On the Origin of Species tility of so many domestic varieties, differing widely are produced under domestication by man’s me- from each other in appearance, for instance of the thodical and unconscious power of selection, for his pigeon or of the cabbage, is a remarkable fact; more own use and pleasure: he neither wishes to select, especially when we reflect how many species there nor could select, slight differences in the reproduc- are, which, though resembling each other most tive system, or other constitutional differences cor- closely, are utterly sterile when intercrossed. Sev- related with the reproductive system. He supplies eral considerations, however, render the fertility of his several varieties with the same food; treats them domestic varieties less remarkable than at first ap- in nearly the same manner, and does not wish to al- pears. It can, in the first place, be clearly shown that ter their general habits of life. Nature acts uniformly mere external dissimilarity between two species and slowly during vast periods of time on the whole does not determine their greater or lesser degree of organisation, in any way which may be for each sterility when crossed; and we may apply the same creature’s own good; and thus she may, either di- rule to domestic varieties. In the second place, some rectly, or more probably indirectly, through correla- eminent naturalists believe that a long course of do- tion, modify the reproductive system in the several mestication tends to eliminate sterility in the suc- descendants from any one species. Seeing this dif- cessive generations of hybrids, which were at first ference in the process of selection, as carried on by only slightly sterile; and if this be so, we surely ought man and nature, we need not be surprised at some not to expect to find sterility both appearing and dis- difference in the result. appearing under nearly the same conditions of life. I have as yet spoken as if the varieties of the same Lastly, and this seems to me by far the most impor- species were invariably fertile when intercrossed. tant consideration, new races of animals and plants But it seems to me impossible to resist the evidence

242 Charles Darwin of the existence of a certain amount of sterility in the Girou de Buzareingues crossed three varieties of few following cases, which I will briefly abstract. The gourd, which like the maize has separated sexes, and evidence is at least as good as that from which we he asserts that their mutual fertilisation is by so much believe in the sterility of a multitude of species. The the less easy as their differences are greater. How far evidence is, also, derived from hostile witnesses, who these experiments may be trusted, I know not; but in all other cases consider fertility and sterility as safe the forms experimentised on, are ranked by Sagaret, criterions of specific distinction. Gartner kept during who mainly founds his classification by the test of several years a dwarf kind of maize with yellow seeds, infertility, as varieties. and a tall variety with red seeds, growing near each The following case is far more remarkable, and other in his garden; and although these plants have seems at first quite incredible; but it is the result of separated sexes, they never naturally crossed. He then an astonishing number of experiments made during fertilised thirteen flowers of the one with the pollen many years on nine species of Verbascum, by so of the other; but only a single head produced any seed, good an observer and so hostile a witness, as Gartner: and this one head produced only five grains. Manipu- namely, that yellow and white varieties of the same lation in this case could not have been injurious, as species of Verbascum when intercrossed produce the plants have separated sexes. No one, I believe, less seed, than do either coloured varieties when has suspected that these varieties of maize are dis- fertilised with pollen from their own coloured flow- tinct species; and it is important to notice that the hy- ers. Moreover, he asserts that when yellow and white brid plants thus raised were themselves perfectly fer- varieties of one species are crossed with yellow and tile; so that even Gartner did not venture to consider white varieties of a distinct species, more seed is the two varieties as specifically distinct. produced by the crosses between the same coloured

243 On the Origin of Species flowers, than between those which are differently rieties when crossed with N. glutinosa. Hence the coloured. Yet these varieties of Verbascum present reproductive system of this one variety must have no other difference besides the mere colour of the been in some manner and in some degree modified. flower; and one variety can sometimes be raised from From these facts; from the great difficulty of ascer- the seed of the other. taining the infertility of varieties in a state of nature, From observations which I have made on certain for a supposed variety if infertile in any degree varieties of hollyhock, I am inclined to suspect that would generally be ranked as species; from man se- they present analogous facts. lecting only external characters in the production of Kolreuter, whose accuracy has been confirmed by the most distinct domestic varieties, and from not every subsequent observer, has proved the remark- wishing or being able to produce recondite and func- able fact, that one variety of the common tobacco is tional differences in the reproductive system; from more fertile, when crossed with a widely distinct these several considerations and facts, I do not think species, than are the other varieties. He that the very general fertility of varieties can be experimentised on five forms, which are commonly proved to be of universal occurrence, or to form a reputed to be varieties, and which he tested by the fundamental distinction between varieties and spe- severest trial, namely, by reciprocal crosses, and he cies. The general fertility of varieties does not seem found their mongrel offspring perfectly fertile. But to me sufficient to overthrow the view which I have one of these five varieties, when used either as fa- taken with respect to the very general, but not in- ther or mother, and crossed with the Nicotiana variable, sterility of first crosses and of hybrids, glutinosa, always yielded hybrids not so sterile as namely, that it is not a special endowment, but is those which were produced from the four other va- incidental on slowly acquired modifications, more

244 Charles Darwin especially in the reproductive systems of the forms brids between very closely allied species are more which are crossed. variable than those from very distinct species; and Hybrids and Mongrels compared, independently this shows that the difference in the degree of vari- of their fertility. —Independently of the question of ability graduates away. When mongrels and the fertility, the offspring of species when crossed and more fertile hybrids are propagated for several gen- of varieties when crossed may be compared in sev- erations an extreme amount of variability in their eral other respects. Gartner, whose strong wish was offspring is notorious; but some few cases both of to draw a marked line of distinction between spe- hybrids and mongrels long retaining uniformity of cies and varieties, could find very few and, as it character could be given. The variability, however, seems to me, quite unimportant differences between in the successive generations of mongrels is, perhaps, the so-called hybrid offspring of species, and the so- greater than in hybrids. called mongrel offspring of varieties. And, on the This greater variability of mongrels than of hybrids other hand, they agree most closely in very many does not seem to me at all surprising. For the par- important respects. ents of mongrels are varieties, and mostly domestic I shall here discuss this subject with extreme brev- varieties (very few experiments having been tried ity. The most important distinction is, that in the first on natural varieties), and this implies in most cases generation mongrels are more variable than hybrids; that there has been recent variability; and therefore but Gartner admits that hybrids from species which we might expect that such variability would often have long been cultivated are often variable in the continue and be super-added to that arising from first generation; and I have myself seen striking in- the mere act of crossing. The slight degree of vari- stances of this fact. Gartner further admits that hy- ability in hybrids from the first cross or in the first

245 On the Origin of Species generation, in contrast with their extreme variability though most closely allied to each other, are crossed in the succeeding generations, is a curious fact and with a third species, the hybrids are widely different deserves attention. For it bears on and corroborates from each other; whereas if two very distinct variet- the view which I have taken on the cause of ordinary ies of one species are crossed with another species, variability; namely, that it is due to the reproduc- the hybrids do not differ much. But this conclusion, tive system being eminently sensitive to any change as far as I can make out, is founded on a single ex- in the conditions of life, being thus often rendered periment; and seems directly opposed to the results either impotent or at least incapable of its proper of several experiments made by Kolreuter. function of producing offspring identical with the These alone are the unimportant differences, which parent-form. Now hybrids in the first generation are Gartner is able to point out, between hybrid and descended from species (excluding those long culti- mongrel plants. On the other hand, the resemblance vated) which have not had their reproductive sys- in mongrels and in hybrids to their respective par- tems in any way affected, and they are not variable; ents, more especially in hybrids produced from but hybrids themselves have their reproductive sys- nearly related species, follows according to Gartner tems seriously affected, and their descendants are the same laws. When two species are crossed, one highly variable. has sometimes a prepotent power of impressing its But to return to our comparison of mongrels and likeness on the hybrid; and so I believe it to be with hybrids: Gartner states that mongrels are more liable varieties of plants. With animals one variety certainly than hybrids to revert to either parent-form; but this, often has this prepotent power over another variety. if it be true, is certainly only a difference in degree. Hybrid plants produced from a reciprocal cross, gen- Gartner further insists that when any two species, al- erally resemble each other closely; and so it is with

246 Charles Darwin mongrels from a reciprocal cross. Both hybrids and supposed fact, that mongrel animals alone are born mongrels can be reduced to either pure parent-form, closely like one of their parents; but it can be shown by repeated crosses in successive generations with that this does sometimes occur with hybrids; yet I either parent. grant much less frequently with hybrids than with These several remarks are apparently applicable mongrels. Looking to the cases which I have collected to animals; but the subject is here excessively com- of cross-bred animals closely resembling one par- plicated, partly owing to the existence of secondary ent, the resemblances seem chiefly confined to char- sexual characters; but more especially owing to pre- acters almost monstrous in their nature, and which potency in transmitting likeness running more have suddenly appeared—such as albinism, melan- strongly in one sex than in the other, both when one ism, deficiency of tail or horns, or additional fingers species is crossed with another, and when one vari- and toes; and do not relate to characters which have ety is crossed with another variety. For instance, I been slowly acquired by selection. Consequently, think those authors are right, who maintain that the sudden reversions to the perfect character of either ass has a prepotent power over the horse, so that parent would be more likely to occur with mongrels, both the mule and the hinny more resemble the ass which are descended from varieties often suddenly than the horse; but that the prepotency runs more produced and semi-monstrous in character, than strongly in the male-ass than in the female, so that with hybrids, which are descended from species the mule, which is the offspring of the male-ass and slowly and naturally produced. On the whole I en- mare, is more like an ass, than is the hinny, which is tirely agree with Dr. Prosper Lucas, who, after ar- the offspring of the female-ass and stallion. ranging an enormous body of facts with respect to Much stress has been laid by some authors on the animals, comes to the conclusion, that the laws of

247 On the Origin of Species resemblance of the child to its parents are the same, innately variable in individuals of the same species, whether the two parents differ much or little from and is eminently susceptible of favourable and each other, namely in the union of individuals of the unfavourable conditions. The degree of sterility does same variety, or of different varieties, or of distinct not strictly follow systematic affinity, but is governed species. by several curious and complex laws. It is generally Laying aside the question of fertility and sterility, different, and sometimes widely different, in recip- in all other respects there seems to be a general and rocal crosses between the same two species. It is not close similarity in the offspring of crossed species, always equal in degree in a first cross and in the hy- and of crossed varieties. If we look at species as hav- brid produced from this cross. ing been specially created, and at varieties as hav- In the same manner as in grafting trees, the capac- ing been produced by secondary laws, this similarity of one species or variety to take on another, is ity would be an astonishing fact. But it harmonises incidental on generally unknown differences in their perfectly with the view that there is no essential dis- vegetative systems, so in crossing, the greater or less tinction between species and varieties. facility of one species to unite with another, is inci- Summary of Chapter — First crosses between forms dental on unknown differences in their reproductive sufficiently distinct to be ranked as species, and their systems. There is no more reason to think that spe- hybrids, are very generally, but not universally, ster- cies have been specially endowed with various de- ile. The sterility is of all degrees, and is often so slight grees of sterility to prevent them crossing and blend- that the two most careful experimentalists who have ing in nature, than to think that trees have been spe- ever lived, have come to diametrically opposite con- cially endowed with various and somewhat analo- clusions in ranking forms by this test. The sterility is gous degrees of difficulty in being grafted together

248 Charles Darwin in order to prevent them becoming inarched in our respond, though due to distinct causes; for both de- forests. pend on the amount of difference of some kind be- The sterility of first crosses between pure species, tween the species which are crossed. Nor is it sur- which have their reproductive systems perfect, prising that the facility of effecting a first cross, the seems to depend on several circumstances; in some fertility of the hybrids produced, and the capacity of cases largely on the early death of the embryo. The being grafted together—though this latter capacity sterility of hybrids, which have their reproductive evidently depends on widely different circum- systems imperfect, and which have had this system stances—should all run, to a certain extent, parallel and their whole organisation disturbed by being with the systematic affinity of the forms which are compounded of two distinct species, seems closely subjected to experiment; for systematic affinity at- allied to that sterility which so frequently affects pure tempts to express all kinds of resemblance between species, when their natural conditions of life have all species. been disturbed. This view is supported by a paral- First crosses between forms known to be varieties, lelism of another kind;—namely, that the crossing or sufficiently alike to be considered as varieties, and of forms only slightly different is favourable to the their mongrel offspring, are very generally, but not vigour and fertility of their offspring; and that slight quite universally, fertile. Nor is this nearly general changes in the conditions of life are apparently and perfect fertility surprising, when we remember favourable to the vigour and fertility of all organic how liable we are to argue in a circle with respect to beings. It is not surprising that the degree of diffi- varieties in a state of nature; and when we remem- culty in uniting two species, and the degree of ste- ber that the greater number of varieties have been rility of their hybrid-offspring should generally cor- produced under domestication by the selection of

249 On the Origin of Species mere external differences, and not of differences in Chapter IX the reproductive system. In all other respects, excluding fertility, there is a close general resemblance be- On the Imperfection of the Geological tween hybrids and mongrels. Finally, then, the facts Record briefly given in this chapter do not seem to me opposed to, but even rather to support the view, that On the absence of intermediate varieties at the there is no fundamental distinction between species present day — On the nature of extinct intermediate and varieties. varieties; on their number — On the vast lapse of time, as inferred from the rate of deposition and of denudation —On the poorness of our palaeontological collections — On the intermittence of geological formations — On the absence of intermediate varieties in any one formation — On the sudden appearance of groups of species — On their sudden appearance in the lowest known fossiliferous strata.

IN THE SIXTH CHAPTER I enumerated the chief objections which might be justly urged against the views maintained in this volume. Most of them have now been discussed. One, namely the distinctness of specific

250 Charles Darwin forms, and their not being blended together by in- places of and exterminate their parent-forms. But just numerable transitional links, is a very obvious diffi- in proportion as this process of extermination has culty. I assigned reasons why such links do not com- acted on an enormous scale, so must the number of monly occur at the present day, under the circum- intermediate varieties, which have formerly existed stances apparently most favourable for their pres- on the earth, be truly enormous. Why then is not ence, namely on an extensive and continuous area every geological formation and every stratum full with graduated physical conditions. I endeavoured of such intermediate links? Geology assuredly does to show, that the life of each species depends in a not reveal any such finely graduated organic chain; more important manner on the presence of other al- and this, perhaps, is the most obvious and gravest ready defined organic forms, than on climate; and, objection which can be urged against my theory. The therefore, that the really governing conditions of life explanation lies, as I believe, in the extreme imper- do not graduate away quite insensibly like heat or fection of the geological record. moisture. I endeavoured, also, to show that inter- In the first place it should always be borne in mind mediate varieties, from existing in lesser numbers what sort of intermediate forms must, on my theory, than the forms which they connect, will generally be have formerly existed. I have found it difficult, when beaten out and exterminated during the course of looking at any two species, to avoid picturing to further modification and improvement. The main myself, forms directly intermediate between them. cause, however, of innumerable intermediate links But this is a wholly false view; we should always not now occurring everywhere throughout nature look for forms intermediate between each species depends on the very process of natural selection, and a common but unknown progenitor; and the through which new varieties continually take the progenitor will generally have differed in some re-

251 On the Origin of Species spects from all its modified descendants. To give a an unknown common parent. The common parent simple illustration: the fantail and pouter pigeons will have had in its whole organisation much gen- have both descended from the rock-pigeon; if we pos- eral resemblance to the tapir and to the horse; but in sessed all the intermediate varieties which have ever some points of structure may have differed consid- existed, we should have an extremely close series erably from both, even perhaps more than they dif- between both and the rock-pigeon; but we should fer from each other. Hence in all such cases, we have no varieties directly intermediate between the should be unable to recognise the parent-form of any fantail and pouter; none, for instance, combining a tail two or more species, even if we closely compared somewhat expanded with a crop somewhat enlarged, the structure of the parent with that of its modified the characteristic features of these two breeds. These descendants, unless at the same time we had a nearly two breeds, moreover, have become so much modi- perfect chain of the intermediate links. fied, that if we had no historical or indirect evidence It is just possible by my theory, that one of two regarding their origin, it would not have been pos- living forms might have descended from the other; sible to have determined from a mere comparison of for instance, a horse from a tapir; and in this case their structure with that of the rock-pigeon, whether direct intermediate links will have existed between they had descended from this species or from some them. But such a case would imply that one form other allied species, such as C. oenas. had remained for a very long period unaltered, So with natural species, if we look to forms very whilst its descendants had undergone a vast amount distinct, for instance to the horse and tapir, we have of change; and the principle of competition between no reason to suppose that links ever existed directly organism and organism, between child and parent, intermediate between them, but between each and will render this a very rare event; for in all cases the

252 Charles Darwin new and improved forms of life will tend to sup- me even to recall to the reader, who may not be a prac- plant the old and unimproved forms. tical geologist, the facts leading the mind feebly to By the theory of natural selection all living species comprehend the lapse of time. He who can read Sir have been connected with the parent-species of each Charles Lyell’s grand work on the Principles of Geol- genus, by differences not greater than we see be- ogy, which the future historian will recognise as hav- tween the varieties of the same species at the present ing produced a revolution in natural science, yet does day; and these parent-species, now generally extinct, not admit how incomprehensibly vast have been the have in their turn been similarly connected with more past periods of time, may at once close this volume. ancient species; and so on backwards, always con- Not that it suffices to study the Principles of Geol- verging to the common ancestor of each great class. ogy, or to read special treatises by different observ- So that the number of intermediate and transitional ers on separate formations, and to mark how each links, between all living and extinct species, must author attempts to give an inadequate idea of the have been inconceivably great. But assuredly, if this duration of each formation or even each stratum. A theory be true, such have lived upon this earth. man must for years examine for himself great piles of superimposed strata, and watch the sea at work grind- On the lapse of Time. — Independently of our not ing down old rocks and making fresh sediment, be- finding fossil remains of such infinitely numerous fore he can hope to comprehend anything of the lapse connecting links, it may be objected, that time will of time, the monuments of which we see around us. not have sufficed for so great an amount of organic It is good to wander along lines of sea-coast, when change, all changes having been effected very slowly formed of moderately hard rocks, and mark the pro- through natural selection. It is hardly possible for cess of degradation. The tides in most cases reach

253 On the Origin of Species the cliffs only for a short time twice a day, and the on our shores, will, I believe, be most deeply im- waves eat into them only when they are charged with pressed with the slowness with which rocky coasts sand or pebbles; for there is reason to believe that are worn away. The observations on this head by pure water can effect little or nothing in wearing Hugh Miller, and by that excellent observer Mr. away rock. At last the base of the cliff is undermined, Smith of Jordan Hill, are most impressive. With the huge fragments fall down, and these remaining fixed, mind thus impressed, let any one examine beds of have to be worn away, atom by atom, until reduced conglomerate many thousand feet in thickness, in size they can be rolled about by the waves, and which, though probably formed at a quicker rate than then are more quickly ground into pebbles, sand, or many other deposits, yet, from being formed of worn mud. But how often do we see along the bases of and rounded pebbles, each of which bears the stamp retreating cliffs rounded boulders, all thickly clothed of time, are good to show how slowly the mass has by marine productions, showing how little they are been accumulated. Let him remember Lyell’s pro- abraded and how seldom they are rolled about! found remark, that the thickness and extent of sedi- Moreover, if we follow for a few miles any line of mentary formations are the result and measure of rocky cliff, which is undergoing degradation, we find the degradation which the earth’s crust has elsewhere that it is only here and there, along a short length or suffered. And what an amount of degradation is round a promontory, that the cliffs are at the present implied by the sedimentary deposits of many coun- time suffering. The appearance of the surface and tries! Professor Ramsay has given me the maximum the vegetation show that elsewhere years have thickness, in most cases from actual measurement, elapsed since the waters washed their base. in a few cases from estimate, of each formation in He who most closely studies the action of the sea different parts of Great Britain; and this is the result:-

254 Charles Darwin ment is transported by the currents of the sea, the Feet process of accumulation in any one area must be ex- Palaeozoic strata (not including igneous beds)..57,154 tremely slow. Secondary strata...... 13,190 But the amount of denudation which the strata have Tertiary strata...... 2,240 in many places suffered, independently of the rate of accumulation of the degraded matter, probably —making altogether 72,584 feet; that is, very nearly offers the best evidence of the lapse of time. I remem- thirteen and three-quarters British miles. Some of ber having been much struck with the evidence of these formations, which are represented in England denudation, when viewing volcanic islands, which by thin beds, are thousands of feet in thickness on have been worn by the waves and pared all round the Continent. Moreover, between each successive into perpendicular cliffs of one or two thousand feet formation, we have, in the opinion of most geolo- in height; for the gentle slope of the lava-streams, gists, enormously long blank periods. So that the due to their formerly liquid state, showed at a glance lofty pile of sedimentary rocks in Britain, gives but how far the hard, rocky beds had once extended into an inadequate idea of the time which has elapsed the open ocean. The same story is still more plainly during their accumulation; yet what time this must told by faults,—those great cracks along which the have consumed! Good observers have estimated that strata have been upheaved on one side, or thrown sediment is deposited by the great Mississippi river down on the other, to the height or depth of thou- at the rate of only 600 feet in a hundred thousand sands of feet; for since the crust cracked, the surface years. This estimate may be quite erroneous; yet, of the land has been so completely planed down by considering over what wide spaces very fine sedi- the action of the sea, that no trace of these vast dislo-

255 On the Origin of Species cations is externally visible. an admirable lesson to stand on the North Downs The Craven fault, for instance, extends for upwards and to look at the distant South Downs; for, remem- of 30 miles, and along this line the vertical displace- bering that at no great distance to the west the north- ment of the strata has varied from 600 to 3000 feet. ern and southern escarpments meet and close, one Prof. Ramsay has published an account of a can safely picture to oneself the great dome of rocks downthrow in Anglesea of 2300 feet; and he informs which must have covered up the Weald within so me that he fully believes there is one in limited a period as since the latter part of the Chalk Merionethshire of 12,000 feet; yet in these cases there formation. The distance from the northern to the is nothing on the surface to show such prodigious southern Downs is about 22 miles, and the thickness movements; the pile of rocks on the one or other side of the several formations is on an average about 1100 having been smoothly swept away. The consider- feet, as I am informed by Prof. Ramsay. But if, as ation of these facts impresses my mind almost in the some geologists suppose, a range of older rocks un- same manner as does the vain endeavour to grapple derlies the Weald, on the flanks of which the overly- with the idea of eternity. ing sedimentary deposits might have accumulated I am tempted to give one other case, the well- in thinner masses than elsewhere, the above estimate known one of the denudation of the Weald. Though would be erroneous; but this source of doubt prob- it must be admitted that the denudation of the Weald ably would not greatly affect the estimate as applied has been a mere trifle, in comparison with that which to the western extremity of the district. If, then, we has removed masses of our palaeozoic strata, in parts knew the rate at which the sea commonly wears away ten thousand feet in thickness, as shown in Prof. a line of cliff of any given height, we could measure Ramsay’s masterly memoir on this subject. Yet it is the time requisite to have denuded the Weald. This,

256 Charles Darwin of course, cannot be done; but we may, in order to be an ample allowance. At this rate, on the above form some crude notion on the subject, assume that data, the denudation of the Weald must have re- the sea would eat into cliffs 500 feet in height at the quired 306,662,400 years; or say three hundred mil- rate of one inch in a century. This will at first appear lion years. much too small an allowance; but it is the same as if The action of fresh water on the gently inclined we were to assume a cliff one yard in height to be Wealden district, when upraised, could hardly have eaten back along a whole line of coast at the rate of been great, but it would somewhat reduce the above one yard in nearly every twenty-two years. I doubt estimate. On the other hand, during oscillations of whether any rock, even as soft as chalk, would yield level, which we know this area has undergone, the at this rate excepting on the most exposed coasts; surface may have existed for millions of years as though no doubt the degradation of a lofty cliff land, and thus have escaped the action of the sea: would be more rapid from the breakage of the fallen when deeply submerged for perhaps equally long fragments. On the other hand, I do not believe that periods, it would, likewise, have escaped the action any line of coast, ten or twenty miles in length, ever of the coast-waves. So that in all probability a far suffers degradation at the same time along its whole longer period than 300 million years has elapsed indented length; and we must remember that almost since the latter part of the Secondary period. all strata contain harder layers or nodules, which I have made these few remarks because it is highly from long resisting attrition form a breakwater at the important for us to gain some notion, however im- base. Hence, under ordinary circumstances, I con- perfect, of the lapse of years. During each of these clude that for a cliff 500 feet in height, a denudation years, over the whole world, the land and the water of one inch per century for the whole length would has been peopled by hosts of living forms. What an

257 On the Origin of Species infinite number of generations, which the mind can- nearly the whole bed of the sea, at a rate sufficiently not grasp, must have succeeded each other in the quick to embed and preserve fossil remains. long roll of years! Now turn to our richest geologi- Throughout an enormously large proportion of the cal museums, and what a paltry display we behold! ocean, the bright blue tint of the water bespeaks its On the poorness of our Palaeontological collec- purity. The many cases on record of a formation con- tions. — That our palaeontological collections are formably covered, after an enormous interval of time, very imperfect, is admitted by every one. The remark by another and later formation, without the under- of that admirable palaeontologist, the late Edward lying bed having suffered in the interval any wear Forbes, should not be forgotten, namely, that num- and tear, seem explicable only on the view of the bers of our fossil species are known and named from bottom of the sea not rarely lying for ages in an un- single and often broken specimens, or from a few altered condition. The remains which do become specimens collected on some one spot. Only a small embedded, if in sand or gravel, will when the beds portion of the surface of the earth has been geologi- are upraised generally be dissolved by the percola- cally explored, and no part with sufficient care, as tion of rain-water. I suspect that but few of the very the important discoveries made every year in Eu- many animals which live on the beach between high rope prove. No organism wholly soft can be pre- and low watermark are preserved. For instance, the served. Shells and bones will decay and disappear several species of the Chthamalinae (a sub-family of when left on the bottom of the sea, where sediment sessile cirripedes) coat the rocks all over the world is not accumulating. I believe we are continually tak- in infinite numbers: they are all strictly littoral, with ing a most erroneous view, when we tacitly admit to the exception of a single Mediterranean species, ourselves that sediment is being deposited over which inhabits deep water and has been found fos-

258 Charles Darwin sil in Sicily, whereas not one other species has hith- belonging to the age of our secondary or palaeozoic erto been found in any tertiary formation: yet it is formations. now known that the genus Chthamalus existed dur- But the imperfection in the geological record ing the chalk period. The molluscan genus Chiton mainly results from another and more important offers a partially analogous case. cause than any of the foregoing; namely, from the With respect to the terrestrial productions which several formations being separated from each other lived during the Secondary and Palaeozoic periods, by wide intervals of time. When we see the forma- it is superfluous to state that our evidence from fos- tions tabulated in written works, or when we follow sil remains is fragmentary in an extreme degree. For them in nature, it is difficult to avoid believing that instance, not a land shell is known belonging to ei- they are closely consecutive. But we know, for in- ther of these vast periods, with one exception dis- stance, from Sir R. Murchison’s great work on Rus- covered by Sir C. Lyell in the carboniferous strata of sia, what wide gaps there are in that country between North America. In regard to mammiferous remains, the superimposed formations; so it is in North a single glance at the historical table published in America, and in many other parts of the world. The the Supplement to Lyell’s Manual, will bring home most skilful geologist, if his attention had been ex- the truth, how accidental and rare is their preserva- clusively confined to these large territories, would tion, far better than pages of detail. Nor is their rar- never have suspected that during the periods which ity surprising, when we remember how large a pro- were blank and barren in his own country, great piles portion of the bones of tertiary mammals have been of sediment, charged with new and peculiar forms discovered either in caves or in lacustrine deposits; of life, had elsewhere been accumulated. And if in and that not a cave or true lacustrine bed is known each separate territory, hardly any idea can be

259 On the Origin of Species formed of the length of time which has elapsed be- faunas will probably be preserved to a distant age. tween the consecutive formations, we may infer that A little reflection will explain why along the rising this could nowhere be ascertained. The frequent and coast of the western side of South America, no ex- great changes in the mineralogical composition of tensive formations with recent or tertiary remains can consecutive formations, generally implying great anywhere be found, though the supply of sediment changes in the geography of the surrounding lands, must for ages have been great, from the enormous whence the sediment has been derived, accords with degradation of the coast-rocks and from muddy the belief of vast intervals of time having elapsed streams entering the sea. The explanation, no doubt, between each formation. is, that the littoral and sub-littoral deposits are con- But we can, I think, see why the geological forma- tinually worn away, as soon as they are brought up tions of each region are almost invariably intermit- by the slow and gradual rising of the land within the tent; that is, have not followed each other in close grinding action of the coast-waves. sequence. Scarcely any fact struck me more when We may, I think, safely conclude that sediment examining many hundred miles of the South Ameri- must be accumulated in extremely thick, solid, or can coasts, which have been upraised several hun- extensive masses, in order to withstand the inces- dred feet within the recent period, than the absence sant action of the waves, when first upraised and of any recent deposits sufficiently extensive to last during subsequent oscillations of level. Such thick for even a short geological period. Along the whole and extensive accumulations of sediment may be west coast, which is inhabited by a peculiar marine formed in two ways; either, in profound depths of fauna, tertiary beds are so scantily developed, that the sea, in which case, judging from the researches no record of several successive and peculiar marine of E. Forbes, we may conclude that the bottom will

260 Charles Darwin be inhabited by extremely few animals, and the mass degradation as it has as yet suffered, but which will when upraised will give a most imperfect record of hardly last to a distant geological age, was certainly the forms of life which then existed; or, sediment may deposited during a downward oscillation of level, be accumulated to any thickness and extent over a and thus gained considerable thickness. shallow bottom, if it continue slowly to subside. In All geological facts tell us plainly that each area this latter case, as long as the rate of subsidence and has undergone numerous slow oscillations of level, supply of sediment nearly balance each other, the and apparently these oscillations have affected wide sea will remain shallow and favourable for life, and spaces. Consequently formations rich in fossils and thus a fossiliferous formation thick enough, when sufficiently thick and extensive to resist subsequent upraised, to resist any amount of degradation, may degradation, may have been formed over wide be formed. spaces during periods of subsidence, but only where I am convinced that all our ancient formations, the supply of sediment was sufficient to keep the which are rich in fossils, have thus been formed dur- sea shallow and to embed and preserve the remains ing subsidence. Since publishing my views on this before they had time to decay. On the other hand, as subject in 1845, I have watched the progress of Geol- long as the bed of the sea remained stationary, thick ogy, and have been surprised to note how author deposits could not have been accumulated in the after author, in treating of this or that great forma- shallow parts, which are the most favourable to life. tion, has come to the conclusion that it was accumu- Still less could this have happened during the alter- lated during subsidence. I may add, that the only nate periods of elevation; or, to speak more accu- ancient tertiary formation on the west coast of South rately, the beds which were then accumulated will America, which has been bulky enough to resist such have been destroyed by being upraised and brought

261 On the Origin of Species within the limits of the coast-action. ods of subsidence, that our great deposits rich in fos- Thus the geological record will almost necessarily sils have been accumulated. Nature may almost be be rendered intermittent. I feel much confidence in said to have guarded against the frequent discovery the truth of these views, for they are in strict accor- of her transitional or linking forms. dance with the general principles inculcated by Sir From the foregoing considerations it cannot be C. Lyell; and E. Forbes independently arrived at a doubted that the geological record, viewed as a similar conclusion. whole, is extremely imperfect; but if we confine our One remark is here worth a passing notice. During attention to any one formation, it becomes more dif- periods of elevation the area of the land and of the ficult to understand, why we do not therein find adjoining shoal parts of the sea will be increased, and closely graduated varieties between the allied spe- new stations will often be formed;—all circumstances cies which lived at its commencement and at its close. most favourable, as previously explained, for the for- Some cases are on record of the same species pre- mation of new varieties and species; but during such senting distinct varieties in the upper and lower parts periods there will generally be a blank in the geo- of the same formation, but, as they are rare, they may logical record. On the other hand, during subsidence, be here passed over. Although each formation has the inhabited area and number of inhabitants will indisputably required a vast number of years for its decrease (excepting the productions on the shores of deposition, I can see several reasons why each a continent when first broken up into an archipelago), should not include a graduated series of links be- and consequently during subsidence, though there tween the species which then lived; but I can by no will be much extinction, fewer new varieties or spe- means pretend to assign due proportional weight to cies will be formed; and it is during these very peri- the following considerations.

262 Charles Darwin Although each formation may mark a very long infer a large amount of migration during climatal lapse of years, each perhaps is short compared with and other changes; and when we see a species first the period requisite to change one species into an- appearing in any formation, the probability is that it other. I am aware that two palaeontologists, whose only then first immigrated into that area. It is well opinions are worthy of much deference, namely known, for instance, that several species appeared Bronn and Woodward, have concluded that the av- somewhat earlier in the palaeozoic beds of North erage duration of each formation is twice or thrice as America than in those of Europe; time having ap- long as the average duration of specific forms. But parently been required for their migration from the insuperable difficulties, as it seems to me, prevent American to the European seas. In examining the us coming to any just conclusion on this head. When latest deposits of various quarters of the world, it we see a species first appearing in the middle of any has everywhere been noted, that some few still ex- formation, it would be rash in the extreme to infer isting species are common in the deposit, but have that it had not elsewhere previously existed. So again become extinct in the immediately surrounding sea; when we find a species disappearing before the up- or, conversely, that some are now abundant in the permost layers have been deposited, it would be neighbouring sea, but are rare or absent in this par- equally rash to suppose that it then became wholly ticular deposit. It is an excellent lesson to reflect on extinct. We forget how small the area of Europe is the ascertained amount of migration of the inhabit- compared with the rest of the world; nor have the ants of Europe during the Glacial period, which several stages of the same formation throughout Eu- forms only a part of one whole geological period; rope been correlated with perfect accuracy. and likewise to reflect on the great changes of level, With marine animals of all kinds, we may safely on the inordinately great change of climate, on the

263 On the Origin of Species prodigious lapse of time, all included within this far greater, that is extending from before the glacial same glacial period. Yet it may be doubted whether epoch to the present day. in any quarter of the world, sedimentary deposits, In order to get a perfect gradation between two including fossil remains, have gone on accumulat- forms in the upper and lower parts of the same for- ing within the same area during the whole of this mation, the deposit must have gone on accumulat- period. It is not, for instance, probable that sediment ing for a very long period, in order to have given was deposited during the whole of the glacial pe- sufficient time for the slow process of variation; hence riod near the mouth of the Mississippi, within that the deposit will generally have to be a very thick limit of depth at which marine animals can flourish; one; and the species undergoing modification will for we know what vast geographical changes oc- have had to live on the same area throughout this curred in other parts of America during this space of whole time. But we have seen that a thick fossilifer- time. When such beds as were deposited in shallow ous formation can only be accumulated during a water near the mouth of the Mississippi during some period of subsidence; and to keep the depth approxi- part of the glacial period shall have been upraised, mately the same, which is necessary in order to en- organic remains will probably first appear and dis- able the same species to live on the same space, the appear at different levels, owing to the migration of supply of sediment must nearly have counterbal- species and to geographical changes. And in the dis- anced the amount of subsidence. But this same move- tant future, a geologist examining these beds, might ment of subsidence will often tend to sink the area be tempted to conclude that the average duration of whence the sediment is derived, and thus diminish life of the embedded fossils had been less than that the supply whilst the downward movement contin- of the glacial period, instead of having been really ues. In fact, this nearly exact balancing between the

264 Charles Darwin supply of sediment and the amount of subsidence yet no one ignorant of this fact would have suspected is probably a rare contingency; for it has been ob- the vast lapse of time represented by the thinner for- served by more than one palaeontologist, that very mation. Many cases could be given of the lower beds thick deposits are usually barren of organic remains, of a formation having been upraised, denuded, sub- except near their upper or lower limits. merged, and then re-covered by the upper beds of It would seem that each separate formation, like the same formation,—facts, showing what wide, yet the whole pile of formations in any country, has gen- easily overlooked, intervals have occurred in its ac- erally been intermittent in its accumulation. When cumulation. In other cases we have the plainest evi- we see, as is so often the case, a formation composed dence in great fossilised trees, still standing upright of beds of different mineralogical composition, we as they grew, of many long intervals of time and may reasonably suspect that the process of deposi- changes of level during the process of deposition, tion has been much interrupted, as a change in the which would never even have been suspected, had currents of the sea and a supply of sediment of a not the trees chanced to have been preserved: thus, different nature will generally have been due to geo- Messrs. Lyell and Dawson found carboniferous beds graphical changes requiring much time. Nor will the 1400 feet thick in Nova Scotia, with ancient root-bear- closest inspection of a formation give any idea of ing strata, one above the other, at no less than sixty- the time which its deposition has consumed. Many eight different levels. Hence, when the same species instances could be given of beds only a few feet in occur at the bottom, middle, and top of a formation, thickness, representing formations, elsewhere thou- the probability is that they have not lived on the same sands of feet in thickness, and which must have re- spot during the whole period of deposition, but have quired an enormous period for their accumulation; disappeared and reappeared, perhaps many times,

265 On the Origin of Species during the same geological period. So that if such with either one or both forms by intermediate variet- species were to undergo a considerable amount of ies. Nor should it be forgotten, as before explained, modification during any one geological period, a that A might be the actual progenitor of B and C, and section would not probably include all the fine in- yet might not at all necessarily be strictly intermedi- termediate gradations which must on my theory have ate between them in all points of structure. So that we existed between them, but abrupt, though perhaps might obtain the parent-species and its several modi- very slight, changes of form. fied descendants from the lower and upper beds of a It is all-important to remember that naturalists have formation, and unless we obtained numerous transi- no golden rule by which to distinguish species and tional gradations, we should not recognise their rela- varieties; they grant some little variability to each tionship, and should consequently be compelled to species, but when they meet with a somewhat greater rank them all as distinct species. amount of difference between any two forms, they It is notorious on what excessively slight differ- rank both as species, unless they are enabled to con- ences many palaeontologists have founded their nect them together by close intermediate gradations. species; and they do this the more readily if the speci- And this from the reasons just assigned we can sel- mens come from different sub-stages of the same for- dom hope to effect in any one geological section. mation. Some experienced conchologists are now Supposing B and C to be two species, and a third, A, sinking many of the very fine species of D’Orbigny to be found in an underlying bed; even if A were and others into the rank of varieties; and on this view strictly intermediate between B and C, it would sim- we do find the kind of evidence of change which on ply be ranked as a third and distinct species, unless my theory we ought to find. Moreover, if we look to at the same time it could be most closely connected rather wider intervals, namely, to distinct but con-

266 Charles Darwin secutive stages of the same great formation, we find that with shells and other marine animals, it is prob- that the embedded fossils, though almost universally ably those which have had the widest range, far ex- ranked as specifically different, yet are far more ceeding the limits of the known geological forma- closely allied to each other than are the species found tions of Europe, which have oftenest given rise, first in more widely separated formations; but to this sub- to local varieties and ultimately to new species; and ject I shall have to return in the following chapter. this again would greatly lessen the chance of our One other consideration is worth notice: with ani- being able to trace the stages of transition in any one mals and plants that can propagate rapidly and are geological formation. not highly locomotive, there is reason to suspect, as It should not be forgotten, that at the present day, we have formerly seen, that their varieties are gen- with perfect specimens for examination, two forms erally at first local; and that such local varieties do can seldom be connected by intermediate varieties not spread widely and supplant their parent-forms and thus proved to be the same species, until many until they have been modified and perfected in some specimens have been collected from many places; considerable degree. According to this view, the and in the case of fossil species this could rarely be chance of discovering in a formation in any one coun- effected by palaeontologists. We shall, perhaps, best try all the early stages of transition between any two perceive the improbability of our being enabled to forms, is small, for the successive changes are sup- connect species by numerous, fine, intermediate, posed to have been local or confined to some one fossil links, by asking ourselves whether, for in- spot. Most marine animals have a wide range; and stance, geologists at some future period will be able we have seen that with plants it is those which have to prove, that our different breeds of cattle, sheep, the widest range, that oftenest present varieties; so horses, and dogs have descended from a single stock

267 On the Origin of Species or from several aboriginal stocks; or, again, whether under an imaginary illustration. The Malay Archi- certain sea-shells inhabiting the shores of North pelago is of about the size of Europe from the North America, which are ranked by some conchologists Cape to the Mediterranean, and from Britain to Rus- as distinct species from their European representa- sia; and therefore equals all the geological forma- tives, and by other conchologists as only varieties, tions which have been examined with any accuracy, are really varieties or are, as it is called, specifically excepting those of the United States of America. I distinct. This could be effected only by the future fully agree with Mr. Godwin-Austen, that the present geologist discovering in a fossil state numerous in- condition of the Malay Archipelago, with its numer- termediate gradations; and such success seems to ous large islands separated by wide and shallow me improbable in the highest degree. seas, probably represents the former state of Europe, Geological research, though it has added numer- when most of our formations were accumulating. The ous species to existing and extinct genera, and has Malay Archipelago is one of the richest regions of made the intervals between some few groups less the whole world in organic beings; yet if all the spe- wide than they otherwise would have been, yet has cies were to be collected which have ever lived there, done scarcely anything in breaking down the dis- how imperfectly would they represent the natural tinction between species, by connecting them to- history of the world! gether by numerous, fine, intermediate varieties; and But we have every reason to believe that the ter- this not having been effected, is probably the grav- restrial productions of the archipelago would be pre- est and most obvious of all the many objections served in an excessively imperfect manner in the for- which may be urged against my views. Hence it will mations which we suppose to be there accumulat- be worth while to sum up the foregoing remarks, ing. I suspect that not many of the strictly littoral

268 Charles Darwin animals, or of those which lived on naked subma- be least perfect. rine rocks, would be embedded; and those embed- It may be doubted whether the duration of any one ded in gravel or sand, would not endure to a distant great period of subsidence over the whole or part of epoch. Wherever sediment did not accumulate on the archipelago, together with a contemporaneous the bed of the sea, or where it did not accumulate at accumulation of sediment, would exceed the aver- a sufficient rate to protect organic bodies from de- age duration of the same specific forms; and these cay, no remains could be preserved. contingencies are indispensable for the preservation In our archipelago, I believe that fossiliferous for- of all the transitional gradations between any two or mations could be formed of sufficient thickness to more species. If such gradations were not fully pre- last to an age, as distant in futurity as the secondary served, transitional varieties would merely appear formations lie in the past, only during periods of as so many distinct species. It is, also, probable that subsidence. These periods of subsidence would be each great period of subsidence would be inter- separated from each other by enormous intervals, rupted by oscillations of level, and that slight during which the area would be either stationary or climatal changes would intervene during such rising; whilst rising, each fossiliferous formation lengthy periods; and in these cases the inhabitants would be destroyed, almost as soon as accumulated, of the archipelago would have to migrate, and no by the incessant coast-action, as we now see on the closely consecutive record of their modifications shores of South America. During the periods of sub- could be preserved in any one formation. sidence there would probably be much extinction of Very many of the marine inhabitants of the archi- life; during the periods of elevation, there would be pelago now range thousands of miles beyond its much variation, but the geological record would then confines; and analogy leads me to believe that it

269 On the Origin of Species would be chiefly these far-ranging species which found in different stages of the same formation, would oftenest produce new varieties; and the vari- would, by most palaeontologists, be ranked as dis- eties would at first generally be local or confined to tinct species. But I do not pretend that I should ever one place, but if possessed of any decided advan- have suspected how poor a record of the mutations tage, or when further modified and improved, they of life, the best preserved geological section pre- would slowly spread and supplant their parent- sented, had not the difficulty of our not discovering forms. When such varieties returned to their ancient innumerable transitional links between the species homes, as they would differ from their former state, which appeared at the commencement and close of in a nearly uniform, though perhaps extremely slight each formation, pressed so hardly on my theory. degree, they would, according to the principles followed by many palaeontologists, be ranked as new On the sudden appearance of whole groups of Al- and distinct species. lied Species. — The abrupt manner in which whole If then, there be some degree of truth in these re- groups of species suddenly appear in certain for- marks, we have no right to expect to find in our geo- mations, has been urged by several palaeontologists, logical formations, an infinite number of those fine for instance, by Agassiz, Pictet, and by none more transitional forms, which on my theory assuredly forcibly than by Professor Sedgwick, as a fatal objec- have connected all the past and present species of tion to the belief in the transmutation of species. If the same group into one long and branching chain numerous species, belonging to the same genera or of life. We ought only to look for a few links, some families, have really started into life all at once, the more closely, some more distantly related to each fact would be fatal to the theory of descent with slow other; and these links, let them be ever so close, if modification through natural selection. For the de-

270 Charles Darwin velopment of a group of forms, all of which have and in the succeeding formation such species will descended from some one progenitor, must have appear as if suddenly created. been an extremely slow process; and the progeni- I may here recall a remark formerly made, namely tors must have lived long ages before their modi- that it might require a long succession of ages to fied descendants. But we continually over-rate the adapt an organism to some new and peculiar line of perfection of the geological record, and falsely infer, life, for instance to fly through the air; but that when because certain genera or families have not been this had been effected, and a few species had thus found beneath a certain stage, that they did not exist acquired a great advantage over other organisms, a before that stage. We continually forget how large comparatively short time would be necessary to pro- the world is, compared with the area over which our duce many divergent forms, which would be able to geological formations have been carefully examined; spread rapidly and widely throughout the world. we forget that groups of species may elsewhere have I will now give a few examples to illustrate these long existed and have slowly multiplied before they remarks; and to show how liable we are to error in invaded the ancient archipelagoes of Europe and of supposing that whole groups of species have sud- the United States. We do not make due allowance denly been produced. I may recall the well-known for the enormous intervals of time, which have prob- fact that in geological treatises, published not many ably elapsed between our consecutive formations,— years ago, the great class of mammals was always longer perhaps in some cases than the time required spoken of as having abruptly come in at the com- for the accumulation of each formation. These inter- mencement of the tertiary series. And now one of vals will have given time for the multiplication of the richest known accumulations of fossil mammals species from some one or some few parent-forms; belongs to the middle of the secondary series; and

271 On the Origin of Species one true mammal has been discovered in the new that, from the number of existing and extinct tertiary red sandstone at nearly the commencement of this species; from the extraordinary abundance of the great series. Cuvier used to urge that no monkey individuals of many species all over the world, from occurred in any tertiary stratum; but now extinct the Arctic regions to the equator, inhabiting various species have been discovered in India, South zones of depths from the upper tidal limits to 50 fath- America, and in Europe even as far back as the oms; from the perfect manner in which specimens eocene stage. The most striking case, however, is that are preserved in the oldest tertiary beds; from the of the Whale family; as these animals have huge ease with which even a fragment of a valve can be bones, are marine, and range over the world, the fact recognised; from all these circumstances, I inferred of not a single bone of a whale having been discov- that had sessile cirripedes existed during the sec- ered in any secondary formation, seemed fully to ondary periods, they would certainly have been pre- justify the belief that this great and distinct order had served and discovered; and as not one species had been suddenly produced in the interval between the been discovered in beds of this age, I concluded that latest secondary and earliest tertiary formation. But this great group had been suddenly developed at now we may read in the Supplement to Lyell’s the commencement of the tertiary series. This was a ‘Manual,’ published in 1858, clear evidence of the sore trouble to me, adding as I thought one more existence of whales in the upper greensand, some instance of the abrupt appearance of a great group time before the close of the secondary period. of species. But my work had hardly been published, I may give another instance, which from having when a skilful palaeontologist, M. Bosquet, sent me passed under my own eyes has much struck me. In a drawing of a perfect specimen of an unmistakeable a memoir on Fossil Sessile Cirripedes, I have stated sessile cirripede, which he had himself extracted

272 Charles Darwin from the chalk of Belgium. And, as if to make the the fact would certainly be highly remarkable; but I case as striking as possible, this sessile cirripede was cannot see that it would be an insuperable difficulty a Chthamalus, a very common, large, and ubiqui- on my theory, unless it could likewise be shown that tous genus, of which not one specimen has as yet the species of this group appeared suddenly and si- been found even in any tertiary stratum. Hence we multaneously throughout the world at this same now positively know that sessile cirripedes existed period. It is almost superfluous to remark that hardly during the secondary period; and these cirripedes any fossil-fish are known from south of the equator; might have been the progenitors of our many ter- and by running through Pictet’s Palaeontology it will tiary and existing species. be seen that very few species are known from sev- The case most frequently insisted on by eral formations in Europe. Some few families of fish palaeontologists of the apparently sudden appear- now have a confined range; the teleostean fish might ance of a whole group of species, is that of the teleo- formerly have had a similarly confined range, and stean fishes, low down in the Chalk period. This after having been largely developed in some one sea, group includes the large majority of existing spe- might have spread widely. Nor have we any right to cies. Lately, Professor Pictet has carried their exist- suppose that the seas of the world have always been ence one sub-stage further back; and some so freely open from south to north as they are at palaeontologists believe that certain much older present. Even at this day, if the Malay Archipelago fishes, of which the affinities are as yet imperfectly were converted into land, the tropical parts of the known, are really teleostean. Assuming, however, Indian Ocean would form a large and perfectly en- that the whole of them did appear, as Agassiz be- closed basin, in which any great group of marine lieves, at the commencement of the chalk formation, animals might be multiplied; and here they would

273 On the Origin of Species remain confined, until some of the species became which have convinced me that all the existing spe- adapted to a cooler climate, and were enabled to cies of the same group have descended from one double the southern capes of Africa or Australia, and progenitor, apply with nearly equal force to the ear- thus reach other and distant seas. liest known species. For instance, I cannot doubt that From these and similar considerations, but chiefly all the Silurian trilobites have descended from some from our ignorance of the geology of other countries one crustacean, which must have lived long before beyond the confines of Europe and the United States; the Silurian age, and which probably differed greatly and from the revolution in our palaeontological from any known animal. Some of the most ancient ideas on many points, which the discoveries of even Silurian animals, as the Nautilus, Lingula, &c., do the last dozen years have effected, it seems to me to not differ much from living species; and it cannot on be about as rash in us to dogmatize on the succes- my theory be supposed, that these old species were sion of organic beings throughout the world, as it the progenitors of all the species of the orders to would be for a naturalist to land for five minutes on which they belong, for they do not present charac- some one barren point in Australia, and then to dis- ters in any degree intermediate between them. If, cuss the number and range of its productions. moreover, they had been the progenitors of these On the sudden appearance of groups of Allied Spe- orders, they would almost certainly have been long cies in the lowest known fossiliferous strata. — There ago supplanted and exterminated by their numer- is another and allied difficulty, which is much graver. ous and improved descendants. I allude to the manner in which numbers of species Consequently, if my theory be true, it is indisput- of the same group, suddenly appear in the lowest able that before the lowest Silurian stratum was known fossiliferous rocks. Most of the arguments deposited, long periods elapsed, as long as, or prob-

274 Charles Darwin ably far longer than, the whole interval from the Sil- of understanding the absence of vast piles of fossil- urian age to the present day; and that during these iferous strata, which on my theory no doubt were vast, yet quite unknown, periods of time, the world somewhere accumulated before the Silurian epoch, swarmed with living creatures. is very great. If these most ancient beds had been To the question why we do not find records of these wholly worn away by denudation, or obliterated by vast primordial periods, I can give no satisfactory metamorphic action, we ought to find only small answer. Several of the most eminent geologists, with remnants of the formations next succeeding them in Sir R. Murchison at their head, are convinced that age, and these ought to be very generally in a meta- we see in the organic remains of the lowest Silurian morphosed condition. But the descriptions which we stratum the dawn of life on this planet. Other highly now possess of the Silurian deposits over immense competent judges, as Lyell and the late E. Forbes, territories in Russia and in North America, do not dispute this conclusion. We should not forget that support the view, that the older a formation is, the only a small portion of the world is known with ac- more it has suffered the extremity of denudation and curacy. M. Barrande has lately added another and metamorphism. lower stage to the Silurian system, abounding with The case at present must remain inexplicable; and new and peculiar species. Traces of life have been may be truly urged as a valid argument against the detected in the Longmynd beds beneath Barrande’s views here entertained. To show that it may hereaf- so-called primordial zone. The presence of phos- ter receive some explanation, I will give the follow- phatic nodules and bituminous matter in some of ing hypothesis. From the nature of the organic re- the lowest azoic rocks, probably indicates the former mains, which do not appear to have inhabited pro- existence of life at these periods. But the difficulty found depths, in the several formations of Europe

275 On the Origin of Species and of the United States; and from the amount of sedi- would in all probability have been accumulated from ment, miles in thickness, of which the formations are sediment derived from their wear and tear; and composed, we may infer that from first to last large would have been at least partially upheaved by the islands or tracts of land, whence the sediment was oscillations of level, which we may fairly conclude derived, occurred in the neighbourhood of the ex- must have intervened during these enormously long isting continents of Europe and North America. But periods. If then we may infer anything from these we do not know what was the state of things in the facts, we may infer that where our oceans now ex- intervals between the successive formations; whether tend, oceans have extended from the remotest pe- Europe and the United States during these intervals riod of which we have any record; and on the other existed as dry land, or as a submarine surface near hand, that where continents now exist, large tracts of land, on which sediment was not deposited, or again land have existed, subjected no doubt to great oscil- as the bed of an open and unfathomable sea. lations of level, since the earliest silurian period. The Looking to the existing oceans, which are thrice as coloured map appended to my volume on Coral extensive as the land, we see them studded with Reefs, led me to conclude that the great oceans are many islands; but not one oceanic island is as yet still mainly areas of subsidence, the great archipela- known to afford even a remnant of any palaeozoic goes still areas of oscillations of level, and the conti- or secondary formation. Hence we may perhaps in- nents areas of elevation. But have we any right to fer, that during the palaeozoic and secondary peri- assume that things have thus remained from eter- ods, neither continents nor continental islands ex- nity? Our continents seem to have been formed by a isted where our oceans now extend; for had they preponderance, during many oscillations of level, existed there, palaeozoic and secondary formations of the force of elevation; but may not the areas of

276 Charles Darwin preponderant movement have changed in the lapse areas, the many formations long anterior to the silurian of ages? At a period immeasurably antecedent to the epoch in a completely metamorphosed condition. silurian epoch, continents may have existed where The several difficulties here discussed, namely our oceans are now spread out; and clear and open not finding in the successive formations infinitely oceans may have existed where our continents now numerous transitional links between the many spe- stand. Nor should we be justified in assuming that cies which now exist or have existed; the sudden if, for instance, the bed of the Pacific Ocean were now manner in which whole groups of species appear in converted into a continent, we should there find for- our European formations; the almost entire absence, mations older than the silurian strata, supposing as at present known, of fossiliferous formations be- such to have been formerly deposited; for it might neath the Silurian strata, are all undoubtedly of the well happen that strata which had subsided some gravest nature. We see this in the plainest manner miles nearer to the centre of the earth, and which had by the fact that all the most eminent palaeontologists, been pressed on by an enormous weight of namely Cuvier, Owen, Agassiz, Barrande, Falconer, superincumbent water, might have undergone far E. Forbes, &c., and all our greatest geologists, as more metamorphic action than strata which have al- Lyell, Murchison, Sedgwick, &c., have unanimously, ways remained nearer to the surface. The immense often vehemently, maintained the immutability of areas in some parts of the world, for instance in South species. But I have reason to believe that one great America, of bare metamorphic rocks, which must have authority, Sir Charles Lyell, from further reflexion been heated under great pressure, have always entertains grave doubts on this subject. I feel how seemed to me to require some special explanation; rash it is to differ from these great authorities, to and we may perhaps believe that we see in these large whom, with others, we owe all our knowledge. Those

277 On the Origin of Species who think the natural geological record in any de- Chapter X gree perfect, and who do not attach much weight to the facts and arguments of other kinds given in this On the Geological Succession of Organic volume, will undoubtedly at once reject my theory. Beings For my part, following out Lyell’s metaphor, I look at the natural geological record, as a history of the On the slow and successive appearance of new spe- world imperfectly kept, and written in a changing cies — On their different rates of change — Species dialect; of this history we possess the last volume once lost do not reappear — Groups of species fol- alone, relating only to two or three countries. Of this low the same general rules in their appearance and volume, only here and there a short chapter has been disappearance as do single species — On Extinction preserved; and of each page, only here and there a — On simultaneous changes in the forms of life few lines. Each word of the slowly-changing lan- throughout the world — On the affinities of extinct guage, in which the history is supposed to be writ- species to each other and to living species — On the ten, being more or less different in the interrupted state of development of ancient forms — On the suc- succession of chapters, may represent the apparently cession of the same types within the same areas — abruptly changed forms of life, entombed in our con- Summary of preceding and present chapters. secutive, but widely separated formations. On this view, the difficulties above discussed are greatly di- LET US NOW SEE whether the several facts and rules minished, or even disappear. relating to the geological succession of organic beings, better accord with the common view of the immutability of species, or with that of their slow and

278 Charles Darwin gradual modification, through descent and natural Species of different genera and classes have not selection. changed at the same rate, or in the same degree. In New species have appeared very slowly, one after the oldest tertiary beds a few living shells may still another, both on the land and in the waters. Lyell be found in the midst of a multitude of extinct forms. has shown that it is hardly possible to resist the evi- Falconer has given a striking instance of a similar dence on this head in the case of the several tertiary fact, in an existing crocodile associated with many stages; and every year tends to fill up the blanks strange and lost mammals and reptiles in the sub- between them, and to make the percentage system Himalayan deposits. The Silurian Lingula differs but of lost and new forms more gradual. In some of the little from the living species of this genus; whereas most recent beds, though undoubtedly of high an- most of the other Silurian Molluscs and all the Crus- tiquity if measured by years, only one or two spe- taceans have changed greatly. The productions of cies are lost forms, and only one or two are new the land seem to change at a quicker rate than those forms, having here appeared for the first time, either of the sea, of which a striking instance has lately been locally, or, as far as we know, on the face of the earth. observed in Switzerland. There is some reason to If we may trust the observations of Philippi in Sic- believe that organisms, considered high in the scale ily, the successive changes in the marine inhabitants of nature, change more quickly than those that are of that island have been many and most gradual. The low: though there are exceptions to this rule. The secondary formations are more broken; but, as Bronn amount of organic change, as Pictet has remarked, has remarked, neither the appearance nor disappear- does not strictly correspond with the succession of ance of their many now extinct species has been si- our geological formations; so that between each two multaneous in each separate formation. consecutive formations, the forms of life have sel-

279 On the Origin of Species dom changed in exactly the same degree. Yet if we variations be accumulated to a greater or lesser compare any but the most closely related formations, amount, thus causing a greater or lesser amount of all the species will be found to have undergone some modification in the varying species, depends on change. When a species has once disappeared from many complex contingencies,—on the variability the face of the earth, we have reason to believe that being of a beneficial nature, on the power of inter- the same identical form never reappears. The stron- crossing, on the rate of breeding, on the slowly chang- gest apparent exception to this latter rule, is that of ing physical conditions of the country, and more es- the so-called ‘colonies’ of M. Barrande, which intrude pecially on the nature of the other inhabitants with for a period in the midst of an older formation, and which the varying species comes into competition. then allow the pre-existing fauna to reappear; but Hence it is by no means surprising that one species Lyell’s explanation, namely, that it is a case of tem- should retain the same identical form much longer porary migration from a distinct geographical prov- than others; or, if changing, that it should change less. ince, seems to me satisfactory. We see the same fact in geographical distribution; These several facts accord well with my theory. I for instance, in the land-shells and coleopterous in- believe in no fixed law of development, causing all sects of Madeira having come to differ considerably the inhabitants of a country to change abruptly, or from their nearest allies on the continent of Europe, simultaneously, or to an equal degree. The process whereas the marine shells and birds have remained of modification must be extremely slow. The vari- unaltered. We can perhaps understand the appar- ability of each species is quite independent of that ently quicker rate of change in terrestrial and in more of all others. Whether such variability be taken ad- highly organised productions compared with ma- vantage of by natural selection, and whether the rine and lower productions, by the more complex

280 Charles Darwin relations of the higher beings to their organic and sils embedded in consecutive formations is not inorganic conditions of life, as explained in a former equal. Each formation, on this view, does not mark a chapter. When many of the inhabitants of a country new and complete act of creation, but only an occa- have become modified and improved, we can un- sional scene, taken almost at hazard, in a slowly derstand, on the principle of competition, and on that changing drama. of the many all-important relations of organism to We can clearly understand why a species when organism, that any form which does not become in once lost should never reappear, even if the very some degree modified and improved, will be liable same conditions of life, organic and inorganic, to be exterminated. Hence we can see why all the should recur. For though the offspring of one spe- species in the same region do at last, if we look to cies might be adapted (and no doubt this has oc- wide enough intervals of time, become modified; for curred in innumerable instances) to fill the exact those which do not change will become extinct. place of another species in the economy of nature, In members of the same class the average amount and thus supplant it; yet the two forms—the old and of change, during long and equal periods of time, the new—would not be identically the same; for both may, perhaps, be nearly the same; but as the accu- would almost certainly inherit different characters mulation of long-enduring fossiliferous formations from their distinct progenitors. For instance, it is just depends on great masses of sediment having been possible, if our fantail-pigeons were all destroyed, deposited on areas whilst subsiding, our formations that fanciers, by striving during long ages for the have been almost necessarily accumulated at wide same object, might make a new breed hardly distin- and irregularly intermittent intervals; consequently guishable from our present fantail; but if the parent the amount of organic change exhibited by the fos- rock-pigeon were also destroyed, and in nature we

281 On the Origin of Species have every reason to believe that the parent-form will have descended from some one species, it is clear generally be supplanted and exterminated by its that as long as any species of the group have ap- improved offspring, it is quite incredible that a fan- peared in the long succession of ages, so long must tail, identical with the existing breed, could be raised its members have continuously existed, in order to from any other species of pigeon, or even from the have generated either new and modified or the same other well-established races of the domestic pigeon, old and unmodified forms. Species of the genus Lin- for the newly-formed fantail would be almost sure gula, for instance, must have continuously existed to inherit from its new progenitor some slight char- by an unbroken succession of generations, from the acteristic differences. lowest Silurian stratum to the present day. Groups of species, that is, genera and families, fol- We have seen in the last chapter that the species of low the same general rules in their appearance and a group sometimes falsely appear to have come in disappearance as do single species, changing more abruptly; and I have attempted to give an explana- or less quickly, and in a greater or lesser degree. A tion of this fact, which if true would have been fatal group does not reappear after it has once disap- to my views. But such cases are certainly exceptional; peared; or its existence, as long as it lasts, is continu- the general rule being a gradual increase in number, ous. I am aware that there are some apparent excep- till the group reaches its maximum, and then, sooner tions to this rule, but the exceptions are surprisingly or later, it gradually decreases. If the number of the few, so few, that E. Forbes, Pictet, and Woodward species of a genus, or the number of the genera of a (though all strongly opposed to such views as I main- family, be represented by a vertical line of varying tain) admit its truth; and the rule strictly accords with thickness, crossing the successive geological forma- my theory. For as all the species of the same group tions in which the species are found, the line will

282 Charles Darwin sometimes falsely appear to begin at its lower end, the extinction of old forms and the production of new not in a sharp point, but abruptly; it then gradually and improved forms are intimately connected to- thickens upwards, sometimes keeping for a space gether. The old notion of all the inhabitants of the of equal thickness, and ultimately thins out in the earth having been swept away at successive periods upper beds, marking the decrease and final extinc- by catastrophes, is very generally given up, even by tion of the species. This gradual increase in number those geologists, as Elie de Beaumont, Murchison, of the species of a group is strictly conformable with Barrande, &c., whose general views would naturally my theory; as the species of the same genus, and the lead them to this conclusion. On the contrary, we genera of the same family, can increase only slowly have every reason to believe, from the study of the and progressively; for the process of modification tertiary formations, that species and groups of spe- and the production of a number of allied forms must cies gradually disappear, one after another, first from be slow and gradual,—one species giving rise first one spot, then from another, and finally from the to two or three varieties, these being slowly con- world. Both single species and whole groups of spe- verted into species, which in their turn produce by cies last for very unequal periods; some groups, as equally slow steps other species, and so on, like the we have seen, having endured from the earliest branching of a great tree from a single stem, till the known dawn of life to the present day; some having group becomes large. disappeared before the close of the palaeozoic period. No fixed law seems to determine the length of On Extinction. — We have as yet spoken only inci- time during which any single species or any single dentally of the disappearance of species and of genus endures. There is reason to believe that the groups of species. On the theory of natural selection complete extinction of the species of a group is gen-

283 On the Origin of Species erally a slower process than their production: if the late geological period, I was filled with astonishment; appearance and disappearance of a group of spe- for seeing that the horse, since its introduction by cies be represented, as before, by a vertical line of the Spaniards into South America, has run wild over varying thickness, the line is found to taper more the whole country and has increased in numbers at gradually at its upper end, which marks the progress an unparalleled rate, I asked myself what could so of extermination, than at its lower end, which marks recently have exterminated the former horse under the first appearance and increase in numbers of the conditions of life apparently so favourable. But how species. In some cases, however, the extermination utterly groundless was my astonishment! Professor of whole groups of beings, as of ammonites towards Owen soon perceived that the tooth, though so like the close of the secondary period, has been wonder- that of the existing horse, belonged to an extinct spe- fully sudden. cies. Had this horse been still living, but in some The whole subject of the extinction of species has degree rare, no naturalist would have felt the least been involved in the most gratuitous mystery. Some surprise at its rarity; for rarity is the attribute of a authors have even supposed that as the individual vast number of species of all classes, in all countries. has a definite length of life, so have species a defi- If we ask ourselves why this or that species is rare, nite duration. No one I think can have marvelled we answer that something is unfavourable in its con- more at the extinction of species, than I have done. ditions of life; but what that something is, we can When I found in La Plata the tooth of a horse em- hardly ever tell. On the supposition of the fossil horse bedded with the remains of Mastodon, still existing as a rare species, we might have felt Megatherium, Toxodon, and other extinct monsters, certain from the analogy of all other mammals, even which all co-existed with still living shells at a very of the slow-breeding elephant, and from the history

284 Charles Darwin of the naturalisation of the domestic horse in South which have been exterminated, either locally or America, that under more favourable conditions it wholly, through man’s agency. I may repeat what I would in a very few years have stocked the whole con- published in 1845, namely, that to admit that spe- tinent. But we could not have told what the cies generally become rare before they become ex- unfavourable conditions were which checked its in- tinct—to feel no surprise at the rarity of a species, crease, whether some one or several contingencies, and and yet to marvel greatly when it ceases to exist, is at what period of the horse’s life, and in what degree, much the same as to admit that sickness in the indi- they severally acted. If the conditions had gone on, vidual is the forerunner of death—to feel no surprise however slowly, becoming less and less favourable, at sickness, but when the sick man dies, to wonder we assuredly should not have perceived the fact, yet and to suspect that he died by some unknown deed the fossil horse would certainly have become rarer and of violence. rarer, and finally extinct;—its place being seized on by The theory of natural selection is grounded on the some more successful competitor. belief that each new variety, and ultimately each new It is most difficult always to remember that the in- species, is produced and maintained by having some crease of every living being is constantly being advantage over those with which it comes into com- checked by unperceived injurious agencies; and that petition; and the consequent extinction of less- these same unperceived agencies are amply suffi- favoured forms almost inevitably follows. It is the cient to cause rarity, and finally extinction. We see same with our domestic productions: when a new in many cases in the more recent tertiary formations, and slightly improved variety has been raised, it at that rarity precedes extinction; and we know that this first supplants the less improved varieties in the has been the progress of events with those animals same neighbourhood; when much improved it is

285 On the Origin of Species transported far and near, like our short-horn cattle, the nearest allies of that species, i.e. the species of and takes the place of other breeds in other coun- the same genus, will be the most liable to extermi- tries. Thus the appearance of new forms and the dis- nation. Thus, as I believe, a number of new species appearance of old forms, both natural and artificial, descended from one species, that is a new genus, are bound together. In certain flourishing groups, comes to supplant an old genus, belonging to the the number of new specific forms which have been same family. But it must often have happened that a produced within a given time is probably greater new species belonging to some one group will have than that of the old forms which have been extermi- seized on the place occupied by a species belonging nated; but we know that the number of species has to a distinct group, and thus caused its extermina- not gone on indefinitely increasing, at least during tion; and if many allied forms be developed from the later geological periods, so that looking to later the successful intruder, many will have to yield their times we may believe that the production of new places; and it will generally be allied forms, which forms has caused the extinction of about the same will suffer from some inherited inferiority in com- number of old forms. mon. But whether it be species belonging to the same The competition will generally be most severe, as or to a distinct class, which yield their places to other formerly explained and illustrated by examples, species which have been modified and improved, a between the forms which are most like each other in few of the sufferers may often long be preserved, all respects. Hence the improved and modified de- from being fitted to some peculiar line of life, or from scendants of a species will generally cause the ex- inhabiting some distant and isolated station, where termination of the parent-species; and if many new they have escaped severe competition. For instance, forms have been developed from any one species, a single species of Trigonia, a great genus of shells

286 Charles Darwin in the secondary formations, survives in the Austra- Thus, as it seems to me, the manner in which single lian seas; and a few members of the great and al- species and whole groups of species become extinct, most extinct group of Ganoid fishes still inhabit our accords well with the theory of natural selection. We fresh waters. Therefore the utter extinction of a group need not marvel at extinction; if we must marvel, let is generally, as we have seen, a slower process than it be at our presumption in imagining for a moment its production. that we understand the many complex contingen- With respect to the apparently sudden extermina- cies, on which the existence of each species depends. tion of whole families or orders, as of Trilobites at If we forget for an instant, that each species tends to the close of the palaeozoic period and of Ammonites increase inordinately, and that some check is always at the close of the secondary period, we must remem- in action, yet seldom perceived by us, the whole ber what has been already said on the probable wide economy of nature will be utterly obscured. When- intervals of time between our consecutive forma- ever we can precisely say why this species is more tions; and in these intervals there may have been abundant in individuals than that; why this species much slow extermination. Moreover, when by sud- and not another can be naturalised in a given coun- den immigration or by unusually rapid develop- try; then, and not till then, we may justly feel sur- ment, many species of a new group have taken pos- prise why we cannot account for the extinction of session of a new area, they will have exterminated this particular species or group of species. in a correspondingly rapid manner many of the old On the Forms of Life changing almost simulta- inhabitants; and the forms which thus yield their neously throughout the World. — Scarcely any places will commonly be allied, for they will par- palaeontological discovery is more striking than the take of some inferiority in common. fact, that the forms of life change almost simulta-

287 On the Origin of Species neously throughout the world. Thus our European has been observed by several authors: so it is, ac- Chalk formation can be recognised in many distant cording to Lyell, with the several European and parts of the world, under the most different climates, North American tertiary deposits. Even if the few where not a fragment of the mineral chalk itself can fossil species which are common to the Old and New be found; namely, in North America, in equatorial Worlds be kept wholly out of view, the general par- South America, in Tierra del Fuego, at the Cape of allelism in the successive forms of life, in the stages Good Hope, and in the peninsula of India. For at of the widely separated palaeozoic and tertiary pe- these distant points, the organic remains in certain riods, would still be manifest, and the several for- beds present an unmistakeable degree of resem- mations could be easily correlated. blance to those of the Chalk. It is not that the same These observations, however, relate to the marine species are met with; for in some cases not one spe- inhabitants of distant parts of the world: we have cies is identically the same, but they belong to the not sufficient data to judge whether the productions same families, genera, and sections of genera, and of the land and of fresh water change at distant points sometimes are similarly characterised in such trifling in the same parallel manner. We may doubt whether points as mere superficial sculpture. Moreover other they have thus changed: if the Megatherium, forms, which are not found in the Chalk of Europe, Mylodon, Macrauchenia, and Toxodon had been but which occur in the formations either above or brought to Europe from La Plata, without any infor- below, are similarly absent at these distant points of mation in regard to their geological position, no one the world. In the several successive palaeozoic for- would have suspected that they had coexisted with mations of Russia, Western Europe and North still living sea-shells; but as these anomalous mon- America, a similar parallelism in the forms of life sters coexisted with the Mastodon and Horse, it

288 Charles Darwin might at least have been inferred that they had lived if this be so, it is evident that fossiliferous beds de- during one of the latter tertiary stages. posited at the present day on the shores of North When the marine forms of life are spoken of as hav- America would hereafter be liable to be classed with ing changed simultaneously throughout the world, somewhat older European beds. Nevertheless, look- it must not be supposed that this expression relates ing to a remotely future epoch, there can, I think, be to the same thousandth or hundred-thousandth year, little doubt that all the more modern marine forma- or even that it has a very strict geological sense; for tions, namely, the upper pliocene, the pleistocene if all the marine animals which live at the present and strictly modern beds, of Europe, North and day in Europe, and all those that lived in Europe South America, and Australia, from containing fos- during the pleistocene period (an enormously re- sil remains in some degree allied, and from not in- mote period as measured by years, including the cluding those forms which are only found in the whole glacial epoch), were to be compared with older underlying deposits, would be correctly those now living in South America or in Australia, ranked as simultaneous in a geological sense. the most skilful naturalist would hardly be able to The fact of the forms of life changing simulta- say whether the existing or the pleistocene inhabit- neously, in the above large sense, at distant parts of ants of Europe resembled most closely those of the the world, has greatly struck those admirable observ- southern hemisphere. So, again, several highly com- ers, MM. de Verneuil and d’Archiac. After referring petent observers believe that the existing produc- to the parallelism of the palaeozoic forms of life in tions of the United States are more closely related to various parts of Europe, they add, ‘If struck by this those which lived in Europe during certain later ter- strange sequence, we turn our attention to North tiary stages, than to those which now live here; and America, and there discover a series of analogous

289 On the Origin of Species phenomena, it will appear certain that all these modi- tage over older forms; and those forms, which are fications of species, their extinction, and the intro- already dominant, or have some advantage over the duction of new ones, cannot be owing to mere other forms in their own country, would naturally changes in marine currents or other causes more or oftenest give rise to new varieties or incipient spe- less local and temporary, but depend on general laws cies; for these latter must be victorious in a still higher which govern the whole animal kingdom.’ M. degree in order to be preserved and to survive. We Barrande has made forcible remarks to precisely the have distinct evidence on this head, in the plants same effect. It is, indeed, quite futile to look to which are dominant, that is, which are commonest changes of currents, climate, or other physical con- in their own homes, and are most widely diffused, ditions, as the cause of these great mutations in the having produced the greatest number of new variet- forms of life throughout the world, under the most ies. It is also natural that the dominant, varying, and different climates. We must, as Barrande has re- far-spreading species, which already have invaded marked, look to some special law. We shall see this to a certain extent the territories of other species, more clearly when we treat of the present distribu- should be those which would have the best chance tion of organic beings, and find how slight is the re- of spreading still further, and of giving rise in new lation between the physical conditions of various countries to new varieties and species. The process countries, and the nature of their inhabitants. of diffusion may often be very slow, being depen- This great fact of the parallel succession of the forms dent on climatal and geographical changes, or on of life throughout the world, is explicable on the strange accidents, but in the long run the dominant theory of natural selection. New species are formed forms will generally succeed in spreading. The dif- by new varieties arising, which have some advan- fusion would, it is probable, be slower with the ter-

290 Charles Darwin restrial inhabitants of distinct continents than with those in the waters of the sea. If two great regions the marine inhabitants of the continuous sea. We had been for a long period favourably circumstanced might therefore expect to find, as we apparently do in an equal degree, whenever their inhabitants met, find, a less strict degree of parallel succession in the the battle would be prolonged and severe; and some productions of the land than of the sea. from one birthplace and some from the other might Dominant species spreading from any region might be victorious. But in the course of time, the forms encounter still more dominant species, and then their dominant in the highest degree, wherever produced, triumphant course, or even their existence, would would tend everywhere to prevail. As they pre- cease. We know not at all precisely what are all the vailed, they would cause the extinction of other and conditions most favourable for the multiplication of inferior forms; and as these inferior forms would be new and dominant species; but we can, I think, allied in groups by inheritance, whole groups would clearly see that a number of individuals, from giv- tend slowly to disappear; though here and there a ing a better chance of the appearance of favourable single member might long be enabled to survive. variations, and that severe competition with many Thus, as it seems to me, the parallel, and, taken in already existing forms, would be highly favourable, a large sense, simultaneous, succession of the same as would be the power of spreading into new terri- forms of life throughout the world, accords well with tories. A certain amount of isolation, recurring at long the principle of new species having been formed by intervals of time, would probably be also favourable, dominant species spreading widely and varying; the as before explained. One quarter of the world may new species thus produced being themselves domi- have been most favourable for the production of new nant owing to inheritance, and to having already had and dominant species on the land, and another for some advantage over their parents or over other spe-

291 On the Origin of Species cies; these again spreading, varying, and producing tion from other parts of the world. As we have rea- new species. The forms which are beaten and which son to believe that large areas are affected by the yield their places to the new and victorious forms, same movement, it is probable that strictly contem- will generally be allied in groups, from inheriting poraneous formations have often been accumulated some inferiority in common; and therefore as new over very wide spaces in the same quarter of the and improved groups spread throughout the world, world; but we are far from having any right to con- old groups will disappear from the world; and the clude that this has invariably been the case, and that succession of forms in both ways will everywhere large areas have invariably been affected by the same tend to correspond. movements. When two formations have been depos- There is one other remark connected with this sub- ited in two regions during nearly, but not exactly the ject worth making. I have given my reasons for be- same period, we should find in both, from the causes lieving that all our greater fossiliferous formations explained in the foregoing paragraphs, the same gen- were deposited during periods of subsidence; and eral succession in the forms of life; but the species that blank intervals of vast duration occurred dur- would not exactly correspond; for there will have been ing the periods when the bed of the sea was either a little more time in the one region than in the other stationary or rising, and likewise when sediment was for modification, extinction, and immigration. not thrown down quickly enough to embed and pre- I suspect that cases of this nature have occurred in serve organic remains. During these long and blank Europe. Mr. Prestwich, in his admirable Memoirs intervals I suppose that the inhabitants of each re- on the eocene deposits of England and France, is able gion underwent a considerable amount of modifica- to draw a close general parallelism between the suc- tion and extinction, and that there was much migra- cessive stages in the two countries; but when he com-

292 Charles Darwin pares certain stages in England with those in France, the two regions could be arranged in the same or- although he finds in both a curious accordance in der, in accordance with the general succession of the the numbers of the species belonging to the same form of life, and the order would falsely appear to genera, yet the species themselves differ in a man- be strictly parallel; nevertheless the species would ner very difficult to account for, considering the prox- not all be the same in the apparently corresponding imity of the two areas,—unless, indeed, it be as- stages in the two regions. sumed that an isthmus separated two seas inhab- On the Affinities of extinct Species to each other, ited by distinct, but contemporaneous, faunas. Lyell and to living forms. —Let us now look to the mu- has made similar observations on some of the later tual affinities of extinct and living species. They all tertiary formations. Barrande, also, shows that there fall into one grand natural system; and this fact is at is a striking general parallelism in the successive once explained on the principle of descent. The more Silurian deposits of Bohemia and Scandinavia; nev- ancient any form is, the more, as a general rule, it ertheless he finds a surprising amount of difference differs from living forms. But, as Buckland long ago in the species. If the several formations in these re- remarked, all fossils can be classed either in still ex- gions have not been deposited during the same existing groups, or between them. That the extinct act periods,—a formation in one region often corre- forms of life help to fill up the wide intervals be- sponding with a blank interval in the other,—and if tween existing genera, families, and orders, cannot in both regions the species have gone on slowly be disputed. For if we confine our attention either to changing during the accumulation of the several for- the living or to the extinct alone, the series is far less mations and during the long intervals of time be- perfect than if we combine both into one general sys- tween them; in this case, the several formations in tem. With respect to the Vertebrata, whole pages

293 On the Origin of Species could be filled with striking illustrations from our ate in all its characters between two living forms, the great palaeontologist, Owen, showing how extinct objection is probably valid. But I apprehend that in animals fall in between existing groups. Cuvier a perfectly natural classification many fossil species ranked the Ruminants and Pachyderms, as the two would have to stand between living species, and most distinct orders of mammals; but Owen has dis- some extinct genera between living genera, even covered so many fossil links, that he has had to alter between genera belonging to distinct families. The the whole classification of these two orders; and has most common case, especially with respect to very placed certain pachyderms in the same sub-order distinct groups, such as fish and reptiles, seems to with ruminants: for example, he dissolves by fine be, that supposing them to be distinguished at the gradations the apparently wide difference between present day from each other by a dozen characters, the pig and the camel. In regard to the Invertebrata, the ancient members of the same two groups would Barrande, and a higher authority could not be named, be distinguished by a somewhat lesser number of asserts that he is every day taught that palaeozoic characters, so that the two groups, though formerly animals, though belonging to the same orders, fami- quite distinct, at that period made some small ap- lies, or genera with those living at the present day, proach to each other. were not at this early epoch limited in such distinct It is a common belief that the more ancient a form groups as they now are. is, by so much the more it tends to connect by some Some writers have objected to any extinct species of its characters groups now widely separated from or group of species being considered as intermedi- each other. This remark no doubt must be restricted ate between living species or groups. If by this term to those groups which have undergone much change it is meant that an extinct form is directly intermedi- in the course of geological ages; and it would be dif-

294 Charles Darwin ficult to prove the truth of the proposition, for every and f14 a closely allied family or sub-family; and now and then even a living animal, as the o14, e14, m14, a third family. These three families, Lepidosiren, is discovered having affinities directed together with the many extinct genera on the several towards very distinct groups. Yet if we compare the lines of descent diverging from the parent-form A, older Reptiles and Batrachians, the older Fish, the will form an order; for all will have inherited some- older Cephalopods, and the eocene Mammals, with thing in common from their ancient and common pro- the more recent members of the same classes, we genitor. On the principle of the continued tendency must admit that there is some truth in the remark. to divergence of character, which was formerly il- Let us see how far these several facts and infer- lustrated by this diagram, the more recent any form ences accord with the theory of descent with modifi- is, the more it will generally differ from its ancient cation. As the subject is somewhat complex, I must progenitor. Hence we can understand the rule that request the reader to turn to the diagram in the fourth the most ancient fossils differ most from existing chapter. We may suppose that the numbered letters forms. We must not, however, assume that diver- represent genera, and the dotted lines diverging from gence of character is a necessary contingency; it de- them the species in each genus. The diagram is much pends solely on the descendants from a species be- too simple, too few genera and too few species being thus enabled to seize on many and different ing given, but this is unimportant for us. The hori- places in the economy of nature. Therefore it is quite zontal lines may represent successive geological for- possible, as we have seen in the case of some Sil- mations, and all the forms beneath the uppermost urian forms, that a species might go on being slightly line may be considered as extinct. The three existing modified in relation to its slightly altered conditions genera, a14, q14, p14, will form a small family; b14 of life, and yet retain throughout a vast period the

295 On the Origin of Species same general characteristics. This is represented in the living genera of three families together, interme- the diagram by the letter F14. diate in character, would be justified, as they are in- All the many forms, extinct and recent, descended termediate, not directly, but only by a long and cir- from A, make, as before remarked, one order; and cuitous course through many widely different forms. this order, from the continued effects of extinction If many extinct forms were to be discovered above and divergence of character, has become divided into one of the middle horizontal lines or geological for- several sub-families and families, some of which are mations—for instance, above No. VI.—but none from supposed to have perished at different periods, and beneath this line, then only the two families on the some to have endured to the present day. left hand (namely, a14, &c., and b14, &c.) would have By looking at the diagram we can see that if many to be united into one family; and the two other fami- of the extinct forms, supposed to be embedded in lies (namely, a14 to f14 now including five genera, the successive formations, were discovered at sev- and o14 to m14) would yet remain distinct. These eral points low down in the series, the three existing two families, however, would be less distinct from families on the uppermost line would be rendered each other than they were before the discovery of less distinct from each other. If, for instance, the gen- the fossils. If, for instance, we suppose the existing era a1, a5, a10, f8, m3, m6, m9 were disinterred, these genera of the two families to differ from each other three families would be so closely linked together by a dozen characters, in this case the genera, at the that they probably would have to be united into one early period marked VI., would differ by a lesser great family, in nearly the same manner as has oc- number of characters; for at this early stage of de- curred with ruminants and pachyderms. Yet he who scent they have not diverged in character from the objected to call the extinct genera, which thus linked common progenitor of the order, nearly so much as

296 Charles Darwin they subsequently diverged. Thus it comes that an- best palaeontologists seems frequently to be the case. cient and extinct genera are often in some slight de- Thus, on the theory of descent with modification, the gree intermediate in character between their modi- main facts with respect to the mutual affinities of the fied descendants, or between their collateral relations. extinct forms of life to each other and to living forms, In nature the case will be far more complicated than seem to me explained in a satisfactory manner. And is represented in the diagram; for the groups will they are wholly inexplicable on any other view. have been more numerous, they will have endured On this same theory, it is evident that the fauna of for extremely unequal lengths of time, and will have any great period in the earth’s history will be inter- been modified in various degrees. As we possess mediate in general character between that which pre- only the last volume of the geological record, and ceded and that which succeeded it. Thus, the spe- that in a very broken condition, we have no right to cies which lived at the sixth great stage of descent in expect, except in very rare cases, to fill up wide in- the diagram are the modified offspring of those tervals in the natural system, and thus unite distinct which lived at the fifth stage, and are the parents of families or orders. All that we have a right to expect, those which became still more modified at the sev- is that those groups, which have within known geo- enth stage; hence they could hardly fail to be nearly logical periods undergone much modification, intermediate in character between the forms of life should in the older formations make some slight above and below. We must, however, allow for the approach to each other; so that the older members entire extinction of some preceding forms, and for should differ less from each other in some of their the coming in of quite new forms by immigration, characters than do the existing members of the same and for a large amount of modification, during the groups; and this by the concurrent evidence of our long and blank intervals between the successive for-

297 On the Origin of Species mations. Subject to these allowances, the fauna of ter are not the oldest, or the most recent; nor are those each geological period undoubtedly is intermedi- which are intermediate in character, intermediate in ate in character, between the preceding and succeed- age. But supposing for an instant, in this and other ing faunas. I need give only one instance, namely, such cases, that the record of the first appearance and the manner in which the fossils of the Devonian sys- disappearance of the species was perfect, we have tem, when this system was first discovered, were at no reason to believe that forms successively pro- once recognised by palaeontologists as intermedi- duced necessarily endure for corresponding lengths ate in character between those of the overlying car- of time: a very ancient form might occasionally last boniferous, and underlying Silurian system. But each much longer than a form elsewhere subsequently fauna is not necessarily exactly intermediate, as un- produced, especially in the case of terrestrial pro- equal intervals of time have elapsed between con- ductions inhabiting separated districts. To compare secutive formations. small things with great: if the principal living and It is no real objection to the truth of the statement, extinct races of the domestic pigeon were arranged that the fauna of each period as a whole is nearly as well as they could be in serial affinity, this arrange- intermediate in character between the preceding and ment would not closely accord with the order in time succeeding faunas, that certain genera offer excep- of their production, and still less with the order of tions to the rule. For instance, mastodons and el- their disappearance; for the parent rock-pigeon now ephants, when arranged by Dr. Falconer in two se- lives; and many varieties between the rock-pigeon ries, first according to their mutual affinities and then and the carrier have become extinct; and carriers according to their periods of existence, do not ac- which are extreme in the important character of cord in arrangement. The species extreme in charac- length of beak originated earlier than short-beaked

298 Charles Darwin tumblers, which are at the opposite end of the series sea, have changed almost simultaneously through- in this same respect. out the world, and therefore under the most differ- Closely connected with the statement, that the or- ent climates and conditions. Consider the prodigious ganic remains from an intermediate formation are in vicissitudes of climate during the pleistocene period, some degree intermediate in character, is the fact, which includes the whole glacial period, and note insisted on by all palaeontologists, that fossils from how little the specific forms of the inhabitants of the two consecutive formations are far more closely re- sea have been affected. lated to each other, than are the fossils from two re- On the theory of descent, the full meaning of the mote formations. Pictet gives as a well-known in- fact of fossil remains from closely consecutive for- stance, the general resemblance of the organic re- mations, though ranked as distinct species, being mains from the several stages of the chalk formation, closely related, is obvious. As the accumulation of though the species are distinct in each stage. This each formation has often been interrupted, and as fact alone, from its generality, seems to have shaken long blank intervals have intervened between suc- Professor Pictet in his firm belief in the immutabil- cessive formations, we ought not to expect to find, ity of species. He who is acquainted with the distri- as I attempted to show in the last chapter, in any one bution of existing species over the globe, will not or two formations all the intermediate varieties be- attempt to account for the close resemblance of the tween the species which appeared at the commence- distinct species in closely consecutive formations, ment and close of these periods; but we ought to by the physical conditions of the ancient areas hav- find after intervals, very long as measured by years, ing remained nearly the same. Let it be remembered but only moderately long as measured geologically, that the forms of life, at least those inhabiting the closely allied forms, or, as they have been called by

299 On the Origin of Species some authors, representative species; and these we process of improvement has affected in a marked assuredly do find. We find, in short, such evidence and sensible manner the organisation of the more of the slow and scarcely sensible mutation of spe- recent and victorious forms of life, in comparison cific forms, as we have a just right to expect to find. with the ancient and beaten forms; but I can see no On the state of Development of Ancient Forms. — way of testing this sort of progress. Crustaceans, for There has been much discussion whether recent instance, not the highest in their own class, may have forms are more highly developed than ancient. I will beaten the highest molluscs. From the extraordinary not here enter on this subject, for naturalists have manner in which European productions have re- not as yet defined to each other’s satisfaction what is cently spread over New Zealand, and have seized meant by high and low forms. But in one particular on places which must have been previously occu- sense the more recent forms must, on my theory, be pied, we may believe, if all the animals and plants higher than the more ancient; for each new species is of Great Britain were set free in New Zealand, that formed by having had some advantage in the in the course of time a multitude of British forms struggle for life over other and preceding forms. If would become thoroughly naturalized there, and under a nearly similar climate, the eocene inhabit- would exterminate many of the natives. On the other ants of one quarter of the world were put into com- hand, from what we see now occurring in New petition with the existing inhabitants of the same or Zealand, and from hardly a single inhabitant of the some other quarter, the eocene fauna or flora would southern hemisphere having become wild in any part certainly be beaten and exterminated; as would a of Europe, we may doubt, if all the productions of secondary fauna by an eocene, and a palaeozoic New Zealand were set free in Great Britain, whether fauna by a secondary fauna. I do not doubt that this any considerable number would be enabled to seize

300 Charles Darwin on places now occupied by our native plants and whilst it leaves the embryo almost unaltered, con- animals. Under this point of view, the productions tinually adds, in the course of successive generations, of Great Britain may be said to be higher than those more and more difference to the adult. of New Zealand. Yet the most skilful naturalist from Thus the embryo comes to be left as a sort of pic- an examination of the species of the two countries ture, preserved by nature, of the ancient and less modi- could not have foreseen this result. fied condition of each animal. This view may be true, Agassiz insists that ancient animals resemble to a and yet it may never be capable of full proof. Seeing, certain extent the embryos of recent animals of the for instance, that the oldest known mammals, reptiles, same classes; or that the geological succession of and fish strictly belong to their own proper classes, extinct forms is in some degree parallel to the em- though some of these old forms are in a slight degree bryological development of recent forms. I must fol- less distinct from each other than are the typical mem- low Pictet and Huxley in thinking that the truth of bers of the same groups at the present day, it would this doctrine is very far from proved. Yet I fully ex- be vain to look for animals having the common em- pect to see it hereafter confirmed, at least in regard bryological character of the Vertebrata, until beds far to subordinate groups, which have branched off from beneath the lowest Silurian strata are discovered—a each other within comparatively recent times. For discovery of which the chance is very small. this doctrine of Agassiz accords well with the theory On the Succession of the same Types within the of natural selection. In a future chapter I shall attempt same areas, during the later tertiary periods. — Mr. to show that the adult differs from its embryo, ow- Clift many years ago showed that the fossil mam- ing to variations supervening at a not early age, and mals from the Australian caves were closely allied being inherited at a corresponding age. This process, to the living marsupials of that continent. In South

301 On the Origin of Species America, a similar relationship is manifest, even to not well displayed by them. Other cases could be an uneducated eye, in the gigantic pieces of armour added, as the relation between the extinct and living like those of the armadillo, found in several parts of land-shells of Madeira; and between the extinct and La Plata; and Professor Owen has shown in the most living brackish-water shells of the Aralo-Caspian Sea. striking manner that most of the fossil mammals, Now what does this remarkable law of the succes- buried there in such numbers, are related to South sion of the same types within the same areas mean? American types. This relationship is even more He would be a bold man, who after comparing the clearly seen in the wonderful collection of fossil present climate of Australia and of parts of South bones made by MM. Lund and Clausen in the caves America under the same latitude, would attempt to of Brazil. I was so much impressed with these facts account, on the one hand, by dissimilar physical con- that I strongly insisted, in 1839 and 1845, on this ‘law ditions for the dissimilarity of the inhabitants of these of the succession of types,’—on ‘this wonderful re- two continents, and, on the other hand, by similarity lationship in the same continent between the dead of conditions, for the uniformity of the same types and the living.’ Professor Owen has subsequently in each during the later tertiary periods. Nor can it extended the same generalisation to the mammals be pretended that it is an immutable law that mar- of the Old World. We see the same law in this supials should have been chiefly or solely produced author’s restorations of the extinct and gigantic birds in Australia; or that Edentata and other American of New Zealand. We see it also in the birds of the types should have been solely produced in South caves of Brazil. Mr. Woodward has shown that the America. For we know that Europe in ancient times same law holds good with sea-shells, but from the was peopled by numerous marsupials; and I have wide distribution of most genera of molluscs, it is shown in the publications above alluded to, that in

302 Charles Darwin America the law of distribution of terrestrial mam- the same manner and degree. But after very long in- mals was formerly different from what it now is. North tervals of time and after great geographical changes, America formerly partook strongly of the present char- permitting much inter-migration, the feebler will acter of the southern half of the continent; and the yield to the more dominant forms, and there will be southern half was formerly more closely allied, than nothing immutable in the laws of past and present it is at present, to the northern half. In a similar man- distribution. ner we know from Falconer and Cautley’s discover- It may be asked in ridicule, whether I suppose that ies, that northern India was formerly more closely re- the megatherium and other allied huge monsters lated in its mammals to Africa than it is at the present have left behind them in South America the sloth, time. Analogous facts could be given in relation to armadillo, and anteater, as their degenerate descen- the distribution of marine animals. dants. This cannot for an instant be admitted. These On the theory of descent with modification, the huge animals have become wholly extinct, and have great law of the long enduring, but not immutable, left no progeny. But in the caves of Brazil, there are succession of the same types within the same areas, many extinct species which are closely allied in size is at once explained; for the inhabitants of each quar- and in other characters to the species still living in ter of the world will obviously tend to leave in that South America; and some of these fossils may be the quarter, during the next succeeding period of time, actual progenitors of living species. It must not be closely allied though in some degree modified de- forgotten that, on my theory, all the species of the scendants. If the inhabitants of one continent formerly same genus have descended from some one species; differed greatly from those of another continent, so so that if six genera, each having eight species, be will their modified descendants still differ in nearly found in one geological formation, and in the next

303 On the Origin of Species succeeding formation there be six other allied or rep- ber both of specimens and of species, preserved in resentative genera with the same number of species, our museums, is absolutely as nothing compared then we may conclude that only one species of each with the incalculable number of generations which of the six older genera has left modified descendants, must have passed away even during a single forma- constituting the six new genera. The other seven spe- tion; that, owing to subsidence being necessary for cies of the old genera have all died out and have left the accumulation of fossiliferous deposits thick no progeny. Or, which would probably be a far com- enough to resist future degradation, enormous in- moner case, two or three species of two or three alone tervals of time have elapsed between the successive of the six older genera will have been the parents of formations; that there has probably been more ex- the six new genera; the other old species and the other tinction during the periods of subsidence, and more whole genera having become utterly extinct. In fail- variation during the periods of elevation, and during orders, with the genera and species decreasing ing the latter the record will have been least perfectly in numbers, as apparently is the case of the Edentata kept; that each single formation has not been con- of South America, still fewer genera and species will tinuously deposited; that the duration of each for- have left modified blood-descendants. mation is, perhaps, short compared with the aver- Summary of the preceding and present Chapters age duration of specific forms; that migration has — I have attempted to show that the geological played an important part in the first appearance of record is extremely imperfect; that only a small por- new forms in any one area and formation; that widely tion of the globe has been geologically explored with ranging species are those which have varied most, care; that only certain classes of organic beings have and have oftenest given rise to new species; and that been largely preserved in a fossil state; that the num- varieties have at first often been local. All these

304 Charles Darwin causes taken conjointly, must have tended to make the Silurian system was deposited: I can answer this the geological record extremely imperfect, and will latter question only hypothetically, by saying that to a large extent explain why we do not find intermi- as far as we can see, where our oceans now extend nable varieties, connecting together all the extinct and they have for an enormous period extended, and existing forms of life by the finest graduated steps. where our oscillating continents now stand they have He who rejects these views on the nature of the stood ever since the Silurian epoch; but that long geological record, will rightly reject my whole before that period, the world may have presented a theory. For he may ask in vain where are the num- wholly different aspect; and that the older continents, berless transitional links which must formerly have formed of formations older than any known to us, connected the closely allied or representative spe- may now all be in a metamorphosed condition, or cies, found in the several stages of the same great may lie buried under the ocean. formation. He may disbelieve in the enormous in- Passing from these difficulties, all the other great tervals of time which have elapsed between our con- leading facts in palaeontology seem to me simply to secutive formations; he may overlook how impor- follow on the theory of descent with modification tant a part migration must have played, when the through natural selection. We can thus understand formations of any one great region alone, as that of how it is that new species come in slowly and suc- Europe, are considered; he may urge the apparent, cessively; how species of different classes do not nec- but often falsely apparent, sudden coming in of essarily change together, or at the same rate, or in whole groups of species. He may ask where are the the same degree; yet in the long run that all undergo remains of those infinitely numerous organisms modification to some extent. The extinction of old which must have existed long before the first bed of forms is the almost inevitable consequence of the

305 On the Origin of Species production of new forms. We can understand why will in the long run tend to people the world with when a species has once disappeared it never reap- allied, but modified, descendants; and these will pears. Groups of species increase in numbers slowly, generally succeed in taking the places of those and endure for unequal periods of time; for the pro- groups of species which are their inferiors in the cess of modification is necessarily slow, and depends struggle for existence. Hence, after long intervals of on many complex contingencies. The dominant spe- time, the productions of the world will appear to cies of the larger dominant groups tend to leave have changed simultaneously. many modified descendants, and thus new sub- We can understand how it is that all the forms of groups and groups are formed. As these are formed, life, ancient and recent, make together one grand the species of the less vigorous groups, from their system; for all are connected by generation. We can inferiority inherited from a common progenitor, tend understand, from the continued tendency to diver- to become extinct together, and to leave no modi- gence of character, why the more ancient a form is, fied offspring on the face of the earth. But the utter the more it generally differs from those now living. extinction of a whole group of species may often be Why ancient and extinct forms often tend to fill up a very slow process, from the survival of a few de- gaps between existing forms, sometimes blending scendants, lingering in protected and isolated situa- two groups previously classed as distinct into one; tions. When a group has once wholly disappeared, but more commonly only bringing them a little closer it does not reappear; for the link of generation has together. The more ancient a form is, the more often, been broken. apparently, it displays characters in some degree We can understand how the spreading of the domi- intermediate between groups now distinct; for the nant forms of life, which are those that oftenest vary, more ancient a form is, the more nearly it will be

306 Charles Darwin related to, and consequently resemble, the common ligible. The succession of the same types of struc- progenitor of groups, since become widely diver- ture within the same areas during the later geologi- gent. Extinct forms are seldom directly intermediate cal periods ceases to be mysterious, and is simply between existing forms; but are intermediate only explained by inheritance. by a long and circuitous course through many ex- If then the geological record be as imperfect as I tinct and very different forms. We can clearly see why believe it to be, and it may at least be asserted that the organic remains of closely consecutive forma- the record cannot be proved to be much more per- tions are more closely allied to each other, than are fect, the main objections to the theory of natural se- those of remote formations; for the forms are more lection are greatly diminished or disappear. On the closely linked together by generation: we can clearly other hand, all the chief laws of palaeontology see why the remains of an intermediate formation plainly proclaim, as it seems to me, that species have are intermediate in character. been produced by ordinary generation: old forms The inhabitants of each successive period in the having been supplanted by new and improved forms world’s history have beaten their predecessors in the of life, produced by the laws of variation still acting race for life, and are, in so far, higher in the scale of round us, and preserved by Natural Selection. nature; and this may account for that vague yet ill- defined sentiment, felt by many palaeontologists, that organisation on the whole has progressed. If it should hereafter be proved that ancient animals resemble to a certain extent the embryos of more recent animals of the same class, the fact will be intel-

307 On the Origin of Species Chapter XI circumpolar land is almost continuous, all authors agree that one of the most fundamental divisions in Geographical Distribution geographical distribution is that between the New and Old Worlds; yet if we travel over the vast Ameri- Present distribution cannot be accounted for by dif- can continent, from the central parts of the United ferences in physical conditions — Importance of bar- States to its extreme southern point, we meet with riers — Affinity of the productions of the same con- the most diversified conditions; the most humid dis- tinent — Centres of creation — Means of dispersal, tricts, arid deserts, lofty mountains, grassy plains, by changes of climate and of the level of the land, forests, marshes, lakes, and great rivers, under al- and by occasional means — Dispersal during the most every temperature. There is hardly a climate Glacial period co-extensive with the world. or condition in the Old World which cannot be par- alleled in the New—at least as closely as the same IN CONSIDERING the distribution of organic beings over species generally require; for it is a most rare case to the face of the globe, the first great fact which strikes find a group of organisms confined to any small spot, us is, that neither the similarity nor the dissimilarity having conditions peculiar in only a slight degree; of the inhabitants of various regions can be accounted for instance, small areas in the Old World could be for by their climatal and other physical conditions. pointed out hotter than any in the New World, yet Of late, almost every author who has studied the these are not inhabited by a peculiar fauna or flora. subject has come to this conclusion. The case of Notwithstanding this parallelism in the conditions America alone would almost suffice to prove its of the Old and New Worlds, how widely different truth: for if we exclude the northern parts where the are their living productions!

308 Charles Darwin In the southern hemisphere, if we compare large parts, where the land almost joins, and where, under tracts of land in Australia, South Africa, and western a slightly different climate, there might have been free South America, between latitudes 25 deg and 35 deg, migration for the northern temperate forms, as there we shall find parts extremely similar in all their con- now is for the strictly arctic productions. We see the ditions, yet it would not be possible to point out three same fact in the great difference between the inhabit- faunas and floras more utterly dissimilar. Or again ants of Australia, Africa, and South America under we may compare the productions of South America the same latitude: for these countries are almost as south of lat. 35 deg with those north of 25 deg, which much isolated from each other as is possible. On each consequently inhabit a considerably different cli- continent, also, we see the same fact; for on the oppo- mate, and they will be found incomparably more site sides of lofty and continuous mountain-ranges, closely related to each other, than they are to the pro- and of great deserts, and sometimes even of large riv- ductions of Australia or Africa under nearly the same ers, we find different productions; though as moun- climate. Analogous facts could be given with respect tain chains, deserts, &c., are not as impassable, or to the inhabitants of the sea. likely to have endured so long as the oceans separat- A second great fact which strikes us in our general ing continents, the differences are very inferior in de- review is, that barriers of any kind, or obstacles to gree to those characteristic of distinct continents. free migration, are related in a close and important Turning to the sea, we find the same law. No two manner to the differences between the productions marine faunas are more distinct, with hardly a fish, of various regions. We see this in the great differ- shell, or crab in common, than those of the eastern ence of nearly all the terrestrial productions of the and western shores of South and Central America; New and Old Worlds, excepting in the northern yet these great faunas are separated only by the nar-

309 On the Origin of Species row, but impassable, isthmus of Panama. Westward the eastern Pacific islands, yet many fish range from of the shores of America, a wide space of open ocean the Pacific into the Indian Ocean, and many shells extends, with not an island as a halting-place for are common to the eastern islands of the Pacific and emigrants; here we have a barrier of another kind, the eastern shores of Africa, on almost exactly oppo- and as soon as this is passed we meet in the eastern site meridians of longitude. islands of the Pacific, with another and totally dis- A third great fact, partly included in the foregoing tinct fauna. So that here three marine faunas range statements, is the affinity of the productions of the far northward and southward, in parallel lines not same continent or sea, though the species themselves far from each other, under corresponding climates; are distinct at different points and stations. It is a but from being separated from each other by impass- law of the widest generality, and every continent able barriers, either of land or open sea, they are offers innumerable instances. Nevertheless the natu- wholly distinct. On the other hand, proceeding still ralist in travelling, for instance, from north to south further westward from the eastern islands of the never fails to be struck by the manner in which suc- tropical parts of the Pacific, we encounter no impass- cessive groups of beings, specifically distinct, yet able barriers, and we have innumerable islands as clearly related, replace each other. He hears from halting-places, until after travelling over a hemi- closely allied, yet distinct kinds of birds, notes nearly sphere we come to the shores of Africa; and over this similar, and sees their nests similarly constructed, vast space we meet with no well-defined and dis- but not quite alike, with eggs coloured in nearly the tinct marine faunas. Although hardly one shell, crab same manner. The plains near the Straits of Magellan or fish is common to the above-named three approxi- are inhabited by one species of Rhea (American os- mate faunas of Eastern and Western America and trich), and northward the plains of La Plata by an-

310 Charles Darwin other species of the same genus; and not by a true of land and water, and independent of their physi- ostrich or emeu, like those found in Africa and Aus- cal conditions. The naturalist must feel little curios- tralia under the same latitude. On these same plains ity, who is not led to inquire what this bond is. of La Plata, we see the agouti and bizcacha, animals This bond, on my theory, is simply inheritance, that having nearly the same habits as our hares and rab- cause which alone, as far as we positively know, pro- bits and belonging to the same order of Rodents, but duces organisms quite like, or, as we see in the case they plainly display an American type of structure. of varieties nearly like each other. The dissimilarity We ascend the lofty peaks of the Cordillera and we of the inhabitants of different regions may be attrib- find an alpine species of bizcacha; we look to the uted to modification through natural selection, and waters, and we do not find the beaver or musk-rat, in a quite subordinate degree to the direct influence but the coypu and capybara, rodents of the Ameri- of different physical conditions. The degree of dis- can type. Innumerable other instances could be similarity will depend on the migration of the more given. If we look to the islands off the American dominant forms of life from one region into another shore, however much they may differ in geological having been effected with more or less ease, at peri- structure, the inhabitants, though they may be all ods more or less remote;—on the nature and num- peculiar species, are essentially American. We may ber of the former immigrants;—and on their action look back to past ages, as shown in the last chapter, and reaction, in their mutual struggles for life;—the and we find American types then prevalent on the relation of organism to organism being, as I have al- American continent and in the American seas. We ready often remarked, the most important of all re- see in these facts some deep organic bond, prevail- lations. Thus the high importance of barriers comes ing throughout space and time, over the same areas into play by checking migration; as does time for

311 On the Origin of Species the slow process of modification through natural se- species will be no uniform quantity. If, for instance, lection. Widely-ranging species, abounding in indi- a number of species, which stand in direct competi- viduals, which have already triumphed over many tion with each other, migrate in a body into a new competitors in their own widely-extended homes and afterwards isolated country, they will be little will have the best chance of seizing on new places, liable to modification; for neither migration nor iso- when they spread into new countries. In their new lation in themselves can do anything. These prin- homes they will be exposed to new conditions, and ciples come into play only by bringing organisms will frequently undergo further modification and into new relations with each other, and in a lesser improvement; and thus they will become still fur- degree with the surrounding physical conditions. As ther victorious, and will produce groups of modi- we have seen in the last chapter that some forms have fied descendants. On this principle of inheritance retained nearly the same character from an enor- with modification, we can understand how it is that mously remote geological period, so certain species sections of genera, whole genera, and even families have migrated over vast spaces, and have not be- are confined to the same areas, as is so commonly come greatly modified. and notoriously the case. On these views, it is obvious, that the several spe- I believe, as was remarked in the last chapter, in cies of the same genus, though inhabiting the most no law of necessary development. As the variability distant quarters of the world, must originally have of each species is an independent property, and will proceeded from the same source, as they have de- be taken advantage of by natural selection, only so scended from the same progenitor. In the case of far as it profits the individual in its complex struggle those species, which have undergone during whole for life, so the degree of modification in different geological periods but little modification, there is

312 Charles Darwin not much difficulty in believing that they may have the same species could possibly have migrated from migrated from the same region; for during the vast some one point to the several distant and isolated geographical and climatal changes which will have points, where now found. Nevertheless the simplic- supervened since ancient times, almost any amount ity of the view that each species was first produced of migration is possible. But in many other cases, in within a single region captivates the mind. He who which we have reason to believe that the species of a rejects it, rejects the vera causa of ordinary genera- genus have been produced within comparatively tion with subsequent migration, and calls in the recent times, there is great difficulty on this head. It agency of a miracle. It is universally admitted, that is also obvious that the individuals of the same spe- in most cases the area inhabited by a species is con- cies, though now inhabiting distant and isolated re- tinuous; and when a plant or animal inhabits two gions, must have proceeded from one spot, where points so distant from each other, or with an interval their parents were first produced: for, as explained of such a nature, that the space could not be easily in the last chapter, it is incredible that individuals passed over by migration, the fact is given as some- identically the same should ever have been pro- thing remarkable and exceptional. The capacity of duced through natural selection from parents spe- migrating across the sea is more distinctly limited cifically distinct. in terrestrial mammals, than perhaps in any other We are thus brought to the question which has been organic beings; and, accordingly, we find no inex- largely discussed by naturalists, namely, whether plicable cases of the same mammal inhabiting dis- species have been created at one or more points of tant points of the world. No geologist will feel any the earth’s surface. Undoubtedly there are very many difficulty in such cases as Great Britain having been cases of extreme difficulty, in understanding how formerly united to Europe, and consequently pos-

313 On the Origin of Species sessing the same quadrupeds. But if the same spe- most natural genera, or those genera in which the cies can be produced at two separate points, why do species are most closely related to each other, are we not find a single mammal common to Europe generally local, or confined to one area. What a and Australia or South America? The conditions of strange anomaly it would be, if, when coming one life are nearly the same, so that a multitude of Euro- step lower in the series, to the individuals of the same pean animals and plants have become naturalised species, a directly opposite rule prevailed; and spe- in America and Australia; and some of the aborigi- cies were not local, but had been produced in two or nal plants are identically the same at these distant more distinct areas! points of the northern and southern hemispheres? Hence it seems to me, as it has to many other natu- The answer, as I believe, is, that mammals have not ralists, that the view of each species having been pro- been able to migrate, whereas some plants, from their duced in one area alone, and having subsequently varied means of dispersal, have migrated across the migrated from that area as far as its powers of mi- vast and broken interspace. The great and striking gration and subsistence under past and present con- influence which barriers of every kind have had on ditions permitted, is the most probable. Undoubt- distribution, is intelligible only on the view that the edly many cases occur, in which we cannot explain great majority of species have been produced on one how the same species could have passed from one side alone, and have not been able to migrate to the point to the other. But the geographical and climatal other side. Some few families, many sub-families, changes, which have certainly occurred within re- very many genera, and a still greater number of sec- cent geological times, must have interrupted or ren- tions of genera are confined to a single region; and it dered discontinuous the formerly continuous range has been observed by several naturalists, that the of many species. So that we are reduced to consider

314 Charles Darwin whether the exceptions to continuity of range are so plained on the view of each species having migrated numerous and of so grave a nature, that we ought to from a single birthplace; then, considering our igno- give up the belief, rendered probable by general rance with respect to former climatal and geographi- considerations, that each species has been produced cal changes and various occasional means of trans- within one area, and has migrated thence as far as it port, the belief that this has been the universal law, could. It would be hopelessly tedious to discuss all seems to me incomparably the safest. the exceptional cases of the same species, now liv- In discussing this subject, we shall be enabled at ing at distant and separated points; nor do I for a the same time to consider a point equally important moment pretend that any explanation could be of- for us, namely, whether the several distinct species fered of many such cases. But after some prelimi- of a genus, which on my theory have all descended nary remarks, I will discuss a few of the most strik- from a common progenitor, can have migrated (un- ing classes of facts; namely, the existence of the same dergoing modification during some part of their mi- species on the summits of distant mountain-ranges, gration) from the area inhabited by their progenitor. and at distant points in the arctic and antarctic re- If it can be shown to be almost invariably the case, gions; and secondly (in the following chapter), the that a region, of which most of its inhabitants are wide distribution of freshwater productions; and closely related to, or belong to the same genera with thirdly, the occurrence of the same terrestrial spe- the species of a second region, has probably received cies on islands and on the mainland, though sepa- at some former period immigrants from this other rated by hundreds of miles of open sea. If the exist- region, my theory will be strengthened; for we can ence of the same species at distant and isolated points clearly understand, on the principle of modification, of the earth’s surface, can in many instances be ex- why the inhabitants of a region should be related to

315 On the Origin of Species those of another region, whence it has been stocked. lied question,—namely whether all the individuals A volcanic island, for instance, upheaved and formed of the same species have descended from a single at the distance of a few hundreds of miles from a pair, or single hermaphrodite, or whether, as some continent, would probably receive from it in the authors suppose, from many individuals simulta- course of time a few colonists, and their descendants, neously created. With those organic beings which though modified, would still be plainly related by never intercross (if such exist), the species, on my inheritance to the inhabitants of the continent. Cases theory, must have descended from a succession of of this nature are common, and are, as we shall here- improved varieties, which will never have blended after more fully see, inexplicable on the theory of with other individuals or varieties, but will have sup- independent creation. This view of the relation of planted each other; so that, at each successive stage species in one region to those in another, does not of modification and improvement, all the individu- differ much (by substituting the word variety for als of each variety will have descended from a single species) from that lately advanced in an ingenious parent. But in the majority of cases, namely, with all paper by Mr. Wallace, in which he concludes, that organisms which habitually unite for each birth, or ‘every species has come into existence coincident which often intercross, I believe that during the slow both in space and time with a pre-existing closely process of modification the individuals of the spe- allied species.’ And I now know from correspon- cies will have been kept nearly uniform by intercross- dence, that this coincidence he attributes to genera- ing; so that many individuals will have gone on sition with modification. multaneously changing, and the whole amount of The previous remarks on ‘single and multiple cen- modification will not have been due, at each stage, tres of creation’ do not directly bear on another al- to descent from a single parent. To illustrate what I

316 Charles Darwin mean: our English racehorses differ slightly from the merly have been submerged, and the two faunas will horses of every other breed; but they do not owe their now blend or may formerly have blended: where the difference and superiority to descent from any single sea now extends, land may at a former period have pair, but to continued care in selecting and training connected islands or possibly even continents to- many individuals during many generations. gether, and thus have allowed terrestrial productions Before discussing the three classes of facts, which I to pass from one to the other. No geologist will dis- have selected as presenting the greatest amount of pute that great mutations of level have occurred difficulty on the theory of ‘single centres of creation,’ within the period of existing organisms. Edward I must say a few words on the means of dispersal. Forbes insisted that all the islands in the Atlantic must recently have been connected with Europe or Means of Dispersal. — Sir C. Lyell and other au- Africa, and Europe likewise with America. Other thors have ably treated this subject. I can give here authors have thus hypothetically bridged over ev- only the briefest abstract of the more important facts. ery ocean, and have united almost every island to Change of climate must have had a powerful influ- some mainland. If indeed the arguments used by ence on migration: a region when its climate was Forbes are to be trusted, it must be admitted that different may have been a high road for migration, scarcely a single island exists which has not recently but now be impassable; I shall, however, presently been united to some continent. This view cuts the have to discuss this branch of the subject in some Gordian knot of the dispersal of the same species to detail. Changes of level in the land must also have the most distant points, and removes many a diffi- been highly influential: a narrow isthmus now sepa- culty: but to the best of any judgment we are not rates two marine faunas; submerge it, or let it for- authorized in admitting such enormous geographi-

317 On the Origin of Species cal changes within the period of existing species. It with each other, and with the many existing oceanic seems to me that we have abundant evidence of great islands. Several facts in distribution,—such as the oscillations of level in our continents; but not of such great difference in the marine faunas on the oppo- vast changes in their position and extension, as to site sides of almost every continent,—the close rela- have united them within the recent period to each tion of the tertiary inhabitants of several lands and other and to the several intervening oceanic islands. even seas to their present inhabitants,—a certain I freely admit the former existence of many islands, degree of relation (as we shall hereafter see) between now buried beneath the sea, which may have served the distribution of mammals and the depth of the as halting places for plants and for many animals sea,—these and other such facts seem to me opposed during their migration. In the coral-producing oceans to the admission of such prodigious geographical such sunken islands are now marked, as I believe, revolutions within the recent period, as are necessi- by rings of coral or atolls standing over them. When- tated on the view advanced by Forbes and admitted ever it is fully admitted, as I believe it will some day by his many followers. The nature and relative pro- be, that each species has proceeded from a single portions of the inhabitants of oceanic islands like- birthplace, and when in the course of time we know wise seem to me opposed to the belief of their former something definite about the means of distribution, continuity with continents. Nor does their almost we shall be enabled to speculate with security on universally volcanic composition favour the admis- the former extension of the land. But I do not believe sion that they are the wrecks of sunken continents;— that it will ever be proved that within the recent pe- if they had originally existed as mountain-ranges on riod continents which are now quite separate, have the land, some at least of the islands would have been continuously, or almost continuously, united been formed, like other mountain-summits, of gran-

318 Charles Darwin ite, metamorphic schists, old fossiliferous or other and some of these floated for a long time. It is well such rocks, instead of consisting of mere piles of known what a difference there is in the buoyancy of volcanic matter. green and seasoned timber; and it occurred to me I must now say a few words on what are called that floods might wash down plants or branches, and accidental means, but which more properly might that these might be dried on the banks, and then by be called occasional means of distribution. I shall a fresh rise in the stream be washed into the sea. here confine myself to plants. In botanical works, this Hence I was led to dry stems and branches of 94 or that plant is stated to be ill adapted for wide dis- plants with ripe fruit, and to place them on sea wa- semination; but for transport across the sea, the ter. The majority sank quickly, but some which whilst greater or less facilities may be said to be almost green floated for a very short time, when dried wholly unknown. Until I tried, with Mr. Berkeley’s floated much longer; for instance, ripe hazel-nuts aid, a few experiments, it was not even known how sank immediately, but when dried, they floated for far seeds could resist the injurious action of sea-wa- 90 days and afterwards when planted they germi- ter. To my surprise I found that out of 87 kinds, 64 nated; an asparagus plant with ripe berries floated germinated after an immersion of 28 days, and a few for 23 days, when dried it floated for 85 days, and survived an immersion of 137 days. For convenience the seeds afterwards germinated: the ripe seeds of sake I chiefly tried small seeds, without the capsule Helosciadium sank in two days, when dried they or fruit; and as all of these sank in a few days, they floated for above 90 days, and afterwards germi- could not be floated across wide spaces of the sea, nated. Altogether out of the 94 dried plants, 18 whether or not they were injured by the salt-water. floated for above 28 days, and some of the 18 floated Afterwards I tried some larger fruits, capsules, &c., for a very much longer period. So that as 64/87 seeds

319 On the Origin of Species germinated after an immersion of 28 days; and as 18/ likewise seeds from plants which live near the sea; 94 plants with ripe fruit (but not all the same species and this would have favoured the average length of as in the foregoing experiment) floated, after being their flotation and of their resistance to the injurious dried, for above 28 days, as far as we may infer any- action of the salt-water. On the other hand he did thing from these scanty facts, we may conclude that not previously dry the plants or branches with the the seeds of 14/100 plants of any country might be fruit; and this, as we have seen, would have caused floated by sea-currents during 28 days, and would some of them to have floated much longer. The re- retain their power of germination. In Johnston’s Physi- sult was that 18/98 of his seeds floated for 42 days, cal Atlas, the average rate of the several Atlantic cur- and were then capable of germination. But I do not rents is 33 miles per diem (some currents running at doubt that plants exposed to the waves would float the rate of 60 miles per diem); on this average, the for a less time than those protected from violent seeds of 14/100 plants belonging to one country might movement as in our experiments. Therefore it would be floated across 924 miles of sea to another country; perhaps be safer to assume that the seeds of about and when stranded, if blown to a favourable spot by 10/100 plants of a flora, after having been dried, an inland gale, they would germinate. could be floated across a space of sea 900 miles in Subsequently to my experiments, M. Martens tried width, and would then germinate. The fact of the similar ones, but in a much better manner, for he larger fruits often floating longer than the small, is placed the seeds in a box in the actual sea, so that interesting; as plants with large seeds or fruit could they were alternately wet and exposed to the air like hardly be transported by any other means; and Alph. really floating plants. He tried 98 seeds, mostly dif- de Candolle has shown that such plants generally ferent from mine; but he chose many large fruits and have restricted ranges.

320 Charles Darwin But seeds may be occasionally transported in an- out of the crop of a pigeon, which had floated on other manner. Drift timber is thrown up on most is- artificial salt-water for 30 days, to my surprise nearly lands, even on those in the midst of the widest all germinated. oceans; and the natives of the coral-islands in the Living birds can hardly fail to be highly effective Pacific, procure stones for their tools, solely from the agents in the transportation of seeds. I could give roots of drifted trees, these stones being a valuable many facts showing how frequently birds of many royal tax. I find on examination, that when irregu- kinds are blown by gales to vast distances across the larly shaped stones are embedded in the roots of ocean. We may I think safely assume that under such trees, small parcels of earth are very frequently en- circumstances their rate of flight would often be 35 closed in their interstices and behind them,—so per- miles an hour; and some authors have given a far fectly that not a particle could be washed away in higher estimate. I have never seen an instance of the longest transport: out of one small portion of nutritious seeds passing through the intestines of a earth thus completely enclosed by wood in an oak bird; but hard seeds of fruit will pass uninjured about 50 years old, three dicotyledonous plants ger- through even the digestive organs of a turkey. In the minated: I am certain of the accuracy of this observa- course of two months, I picked up in my garden 12 tion. Again, I can show that the carcasses of birds, kinds of seeds, out of the excrement of small birds, when floating on the sea, sometimes escape being and these seemed perfect, and some of them, which immediately devoured; and seeds of many kinds in I tried, germinated. But the following fact is more the crops of floating birds long retain their vitality: important: the crops of birds do not secrete gastric peas and vetches, for instance, are killed by even a juice, and do not in the least injure, as I know by few days’ immersion in sea-water; but some taken trial, the germination of seeds; now after a bird has

321 On the Origin of Species found and devoured a large supply of food, it is quently devoured by birds, and thus the seeds might positively asserted that all the grains do not pass be transported from place to place. I forced many into the gizzard for 12 or even 18 hours. A bird in kinds of seeds into the stomachs of dead fish, and this interval might easily be blown to the distance of then gave their bodies to fishing-eagles, storks, and 500 miles, and hawks are known to look out for tired pelicans; these birds after an interval of many hours, birds, and the contents of their torn crops might thus either rejected the seeds in pellets or passed them in readily get scattered. Mr. Brent informs me that a their excrement; and several of these seeds retained friend of his had to give up flying carrier-pigeons their power of germination. Certain seeds, however, from France to England, as the hawks on the English were always killed by this process. coast destroyed so many on their arrival. Some Although the beaks and feet of birds are generally hawks and owls bolt their prey whole, and after an quite clean, I can show that earth sometimes adheres interval of from twelve to twenty hours, disgorge to them: in one instance I removed twenty-two grains pellets, which, as I know from experiments made in of dry argillaceous earth from one foot of a partridge, the Zoological Gardens, include seeds capable of and in this earth there was a pebble quite as large as germination. Some seeds of the oat, wheat, millet, the seed of a vetch. Thus seeds might occasionally canary, hemp, clover, and beet germinated after hav- be transported to great distances; for many facts ing been from twelve to twenty-one hours in the could be given showing that soil almost everywhere stomachs of different birds of prey; and two seeds of is charged with seeds. Reflect for a moment on the beet grew after having been thus retained for two millions of quails which annually cross the Mediter- days and fourteen hours. Freshwater fish, I find, eat ranean; and can we doubt that the earth adhering to seeds of many land and water plants: fish are fre- their feet would sometimes include a few minute

322 Charles Darwin seeds? But I shall presently have to recur to this sub- archipelago. Hence we may safely infer that icebergs ject. formerly landed their rocky burthens on the shores As icebergs are known to be sometimes loaded of these mid-ocean islands, and it is at least possible with earth and stones, and have even carried brush- that they may have brought thither the seeds of north- wood, bones, and the nest of a land-bird, I can hardly ern plants. doubt that they must occasionally have transported Considering that the several above means of trans- seeds from one part to another of the arctic and ant- port, and that several other means, which without arctic regions, as suggested by Lyell; and during the doubt remain to be discovered, have been in action Glacial period from one part of the now temperate year after year, for centuries and tens of thousands regions to another. In the Azores, from the large num- of years, it would I think be a marvellous fact if many ber of the species of plants common to Europe, in plants had not thus become widely transported. comparison with the plants of other oceanic islands These means of transport are sometimes called acci- nearer to the mainland, and (as remarked by Mr. H. dental, but this is not strictly correct: the currents of C. Watson) from the somewhat northern character of the sea are not accidental, nor is the direction of the flora in comparison with the latitude, I suspected prevalent gales of wind. It should be observed that that these islands had been partly stocked by ice- scarcely any means of transport would carry seeds borne seeds, during the Glacial epoch. At my request for very great distances; for seeds do not retain their Sir C. Lyell wrote to M. Hartung to inquire whether vitality when exposed for a great length of time to he had observed erratic boulders on these islands, the action of seawater; nor could they be long car- and he answered that he had found large fragments ried in the crops or intestines of birds. These means, of granite and other rocks, which do not occur in the however, would suffice for occasional transport

323 On the Origin of Species across tracts of sea some hundred miles in breadth, ain, has not, as far as is known (and it would be very or from island to island, or from a continent to a difficult to prove this), received within the last few neighbouring island, but not from one distant conti- centuries, through occasional means of transport, nent to another. The floras of distant continents immigrants from Europe or any other continent, that would not by such means become mingled in any a poorly-stocked island, though standing more re- great degree; but would remain as distinct as we now mote from the mainland, would not receive colonists see them to be. The currents, from their course, by similar means. I do not doubt that out of twenty would never bring seeds from North America to Brit- seeds or animals transported to an island, even if far ain, though they might and do bring seeds from the less well-stocked than Britain, scarcely more than one West Indies to our western shores, where, if not killed would be so well fitted to its new home, as to be- by so long an immersion in salt-water, they could come naturalised. But this, as it seems to me, is no not endure our climate. Almost every year, one or valid argument against what would be effected by two land-birds are blown across the whole Atlantic occasional means of transport, during the long lapse Ocean, from North America to the western shores of of geological time, whilst an island was being up- Ireland and England; but seeds could be transported heaved and formed, and before it had become fully by these wanderers only by one means, namely, in stocked with inhabitants. On almost bare land, with dirt sticking to their feet, which is in itself a rare acci- few or no destructive insects or birds living there, dent. Even in this case, how small would the chance nearly every seed, which chanced to arrive, would be of a seed falling on favourable soil, and coming be sure to germinate and survive. to maturity! But it would be a great error to argue that because a well-stocked island, like Great Brit- Dispersal during the Glacial period. — The iden-

324 Charles Darwin tity of many plants and animals, on mountain-sum- facts. We have evidence of almost every conceivable mits, separated from each other by hundreds of miles kind, organic and inorganic, that within a very re- of lowlands, where the Alpine species could not pos- cent geological period, central Europe and North sibly exist, is one of the most striking cases known America suffered under an Arctic climate. The ruins of the same species living at distant points, without of a house burnt by fire do not tell their tale more the apparent possibility of their having migrated plainly, than do the mountains of Scotland and from one to the other. It is indeed a remarkable fact Wales, with their scored flanks, polished surfaces, to see so many of the same plants living on the snowy and perched boulders, of the icy streams with which regions of the Alps or Pyrenees, and in the extreme their valleys were lately filled. So greatly has the cli- northern parts of Europe; but it is far more remark- mate of Europe changed, that in Northern Italy, gi- able, that the plants on the White Mountains, in the gantic moraines, left by old glaciers, are now clothed United States of America, are all the same with those by the vine and maize. Throughout a large part of of Labrador, and nearly all the same, as we hear from the United States, erratic boulders, and rocks scored Asa Gray, with those on the loftiest mountains of by drifted icebergs and coast-ice, plainly reveal a Europe. Even as long ago as 1747, such facts led former cold period. Gmelin to conclude that the same species must have The former influence of the glacial climate on the been independently created at several distinct distribution of the inhabitants of Europe, as ex- points; and we might have remained in this same plained with remarkable clearness by Edward belief, had not Agassiz and others called vivid at- Forbes, is substantially as follows. But we shall fol- tention to the Glacial period, which, as we shall im- low the changes more readily, by supposing a new mediately see, affords a simple explanation of these glacial period to come slowly on, and then pass

325 On the Origin of Species away, as formerly occurred. As the cold came on, may suppose that the Glacial period came on a little and as each more southern zone became fitted for earlier or later in North America than in Europe, so arctic beings and ill-fitted for their former more tem- will the southern migration there have been a little perate inhabitants, the latter would be supplanted earlier or later; but this will make no difference in and arctic productions would take their places. The the final result. inhabitants of the more temperate regions would at As the warmth returned, the arctic forms would the same time travel southward, unless they were retreat northward, closely followed up in their re- stopped by barriers, in which case they would per- treat by the productions of the more temperate re- ish. The mountains would become covered with gions. And as the snow melted from the bases of the snow and ice, and their former Alpine inhabitants mountains, the arctic forms would seize on the would descend to the plains. By the time that the cleared and thawed ground, always ascending cold had reached its maximum, we should have a higher and higher, as the warmth increased, whilst uniform arctic fauna and flora, covering the central their brethren were pursuing their northern journey. parts of Europe, as far south as the Alps and Hence, when the warmth had fully returned, the Pyrenees, and even stretching into Spain. The now same arctic species, which had lately lived in a body temperate regions of the United States would like- together on the lowlands of the Old and New Worlds, wise be covered by arctic plants and animals, and would be left isolated on distant mountain-summits these would be nearly the same with those of Eu- (having been exterminated on all lesser heights) and rope; for the present circumpolar inhabitants, which in the arctic regions of both hemispheres. we suppose to have everywhere travelled south- Thus we can understand the identity of many ward, are remarkably uniform round the world. We plants at points so immensely remote as on the

326 Charles Darwin mountains of the United States and of Europe. We migration across the low intervening tracts, since be- can thus also understand the fact that the Alpine come too warm for their existence. plants of each mountain-range are more especially If the climate, since the Glacial period, has ever related to the arctic forms living due north or nearly been in any degree warmer than at present (as some due north of them: for the migration as the cold came geologists in the United States believe to have been on, and the re-migration on the returning warmth, the case, chiefly from the distribution of the fossil will generally have been due south and north. The Gnathodon), then the arctic and temperate produc- Alpine plants, for example, of Scotland, as remarked tions will at a very late period have marched a little by Mr. H. C. Watson, and those of the Pyrenees, as further north, and subsequently have retreated to remarked by Ramond, are more especially allied to their present homes; but I have met with no satisfac- the plants of northern Scandinavia; those of the tory evidence with respect to this intercalated slightly United States to Labrador; those of the mountains of warmer period, since the Glacial period. Siberia to the arctic regions of that country. These The arctic forms, during their long southern mi- views, grounded as they are on the perfectly well- gration and re-migration northward, will have been ascertained occurrence of a former Glacial period, exposed to nearly the same climate, and, as is espe- seem to me to explain in so satisfactory a manner cially to be noticed, they will have kept in a body the present distribution of the Alpine and Arctic pro- together; consequently their mutual relations will not ductions of Europe and America, that when in other have been much disturbed, and, in accordance with regions we find the same species on distant moun- the principles inculcated in this volume, they will tain-summits, we may almost conclude without other not have been liable to much modification. But with evidence, that a colder climate permitted their former our Alpine productions, left isolated from the mo-

327 On the Origin of Species ment of the returning warmth, first at the bases and In illustrating what, as I believe, actually took place ultimately on the summits of the mountains, the case during the Glacial period, I assumed that at its com- will have been somewhat different; for it is not likely mencement the arctic productions were as uniform that all the same arctic species will have been left on round the polar regions as they are at the present mountain ranges distant from each other, and have day. But the foregoing remarks on distribution ap- survived there ever since; they will, also, in all prob- ply not only to strictly arctic forms, but also to many ability have become mingled with ancient Alpine sub-arctic and to some few northern temperate forms, species, which must have existed on the mountains for some of these are the same on the lower moun- before the commencement of the Glacial epoch, and tains and on the plains of North America and Eu- which during its coldest period will have been tem- rope; and it may be reasonably asked how I account porarily driven down to the plains; they will, also, for the necessary degree of uniformity of the sub- have been exposed to somewhat different climatal arctic and northern temperate forms round the world, influences. Their mutual relations will thus have been at the commencement of the Glacial period. At the in some degree disturbed; consequently they will present day, the sub-arctic and northern temperate have been liable to modification; and this we find has productions of the Old and New Worlds are sepa- been the case; for if we compare the present Alpine rated from each other by the Atlantic Ocean and by plants and animals of the several great European the extreme northern part of the Pacific. During the mountain-ranges, though very many of the species Glacial period, when the inhabitants of the Old and are identically the same, some present varieties, some New Worlds lived further southwards than at are ranked as doubtful forms, and some few are dis- present, they must have been still more completely tinct yet closely allied or representative species. separated by wider spaces of ocean. I believe the

328 Charles Darwin above difficulty may be surmounted by looking to Believing, from reasons before alluded to, that our still earlier changes of climate of an opposite nature. continents have long remained in nearly the same We have good reason to believe that during the relative position, though subjected to large, but par- newer Pliocene period, before the Glacial epoch, and tial oscillations of level, I am strongly inclined to whilst the majority of the inhabitants of the world extend the above view, and to infer that during some were specifically the same as now, the climate was earlier and still warmer period, such as the older warmer than at the present day. Hence we may sup- Pliocene period, a large number of the same plants pose that the organisms now living under the cli- and animals inhabited the almost continuous circum- mate of latitude 60 deg, during the Pliocene period polar land; and that these plants and animals, both lived further north under the Polar Circle, in latitude in the Old and New Worlds, began slowly to mi- 66 deg-67 deg; and that the strictly arctic productions grate southwards as the climate became less warm, then lived on the broken land still nearer to the pole. long before the commencement of the Glacial period. Now if we look at a globe, we shall see that under We now see, as I believe, their descendants, mostly the Polar Circle there is almost continuous land from in a modified condition, in the central parts of Eu- western Europe, through Siberia, to eastern America. rope and the United States. On this view we can un- And to this continuity of the circumpolar land, and derstand the relationship, with very little identity, to the consequent freedom for intermigration under between the productions of North America and Eu- a more favourable climate, I attribute the necessary rope,—a relationship which is most remarkable, con- amount of uniformity in the sub-arctic and northern sidering the distance of the two areas, and their sepa- temperate productions of the Old and New Worlds, ration by the Atlantic Ocean. We can further under- at a period anterior to the Glacial epoch. stand the singular fact remarked on by several ob-

329 On the Origin of Species servers, that the productions of Europe and America more modification than with the Alpine productions, during the later tertiary stages were more closely re- left isolated, within a much more recent period, on lated to each other than they are at the present time; the several mountain-ranges and on the arctic lands for during these warmer periods the northern parts of the two Worlds. Hence it has come, that when we of the Old and New Worlds will have been almost compare the now living productions of the temper- continuously united by land, serving as a bridge, ate regions of the New and Old Worlds, we find very since rendered impassable by cold, for the inter-mi- few identical species (though Asa Gray has lately gration of their inhabitants. shown that more plants are identical than was for- During the slowly decreasing warmth of the merly supposed), but we find in every great class Pliocene period, as soon as the species in common, many forms, which some naturalists rank as geo- which inhabited the New and Old Worlds, migrated graphical races, and others as distinct species; and a south of the Polar Circle, they must have been com- host of closely allied or representative forms which pletely cut off from each other. This separation, as are ranked by all naturalists as specifically distinct. far as the more temperate productions are concerned, As on the land, so in the waters of the sea, a slow took place long ages ago. And as the plants and ani- southern migration of a marine fauna, which during mals migrated southward, they will have become the Pliocene or even a somewhat earlier period, was mingled in the one great region with the native nearly uniform along the continuous shores of the American productions, and have had to compete Polar Circle, will account, on the theory of modifica- with them; and in the other great region, with those tion, for many closely allied forms now living in ar- of the Old World. Consequently we have here ev- eas completely sundered. Thus, I think, we can un- erything favourable for much modification,—for far derstand the presence of many existing and tertiary

330 Charles Darwin representative forms on the eastern and western view may be largely extended. In Europe we have shores of temperate North America; and the still more the plainest evidence of the cold period, from the striking case of many closely allied crustaceans (as western shores of Britain to the Oural range, and described in Dana’s admirable work), of some fish and southward to the Pyrenees. We may infer, from the other marine animals, in the Mediterranean and in the frozen mammals and nature of the mountain veg- seas of Japan,—areas now separated by a continent etation, that Siberia was similarly affected. Along the and by nearly a hemisphere of equatorial ocean. Himalaya, at points 900 miles apart, glaciers have These cases of relationship, without identity, of the left the marks of their former low descent; and in inhabitants of seas now disjoined, and likewise of Sikkim, Dr. Hooker saw maize growing on gigantic the past and present inhabitants of the temperate ancient moraines. South of the equator, we have some lands of North America and Europe, are inexplicable direct evidence of former glacial action in New on the theory of creation. We cannot say that they Zealand; and the same plants, found on widely sepa- have been created alike, in correspondence with the rated mountains in this island, tell the same story. If nearly similar physical conditions of the areas; for if one account which has been published can be we compare, for instance, certain parts of South trusted, we have direct evidence of glacial action in America with the southern continents of the Old the south-eastern corner of Australia. World, we see countries closely corresponding in Looking to America; in the northern half, ice-borne all their physical conditions, but with their inhabit- fragments of rock have been observed on the east- ants utterly dissimilar. ern side as far south as lat. 36 deg-37 deg, and on the But we must return to our more immediate sub- shores of the Pacific, where the climate is now so ject, the Glacial period. I am convinced that Forbes’s different, as far south as lat. 46 deg; erratic boulders

331 On the Origin of Species have, also, been noticed on the Rocky Mountains. In temporaneous in a geological sense, it seems to me the Cordillera of Equatorial South America, glaciers probable that it was, during a part at least of the pe- once extended far below their present level. In central riod, actually simultaneous throughout the world. Chile I was astonished at the structure of a vast mound Without some distinct evidence to the contrary, we of detritus, about 800 feet in height, crossing a valley may at least admit as probable that the glacial action of the Andes; and this I now feel convinced was a gi- was simultaneous on the eastern and western sides gantic moraine, left far below any existing glacier. Fur- of North America, in the Cordillera under the equa- ther south on both sides of the continent, from lat. 41 tor and under the warmer temperate zones, and on deg to the southernmost extremity, we have the clearest both sides of the southern extremity of the continent. evidence of former glacial action, in huge boulders If this be admitted, it is difficult to avoid believing transported far from their parent source. that the temperature of the whole world was at this We do not know that the Glacial epoch was strictly period simultaneously cooler. But it would suffice simultaneous at these several far distant points on for my purpose, if the temperature was at the same opposite sides of the world. But we have good evi- time lower along certain broad belts of longitude. dence in almost every case, that the epoch was in- On this view of the whole world, or at least of broad cluded within the latest geological period. We have, longitudinal belts, having been simultaneously also, excellent evidence, that it endured for an enor- colder from pole to pole, much light can be thrown mous time, as measured by years, at each point. The on the present distribution of identical and allied cold may have come on, or have ceased, earlier at species. In America, Dr. Hooker has shown that be- one point of the globe than at another, but seeing tween forty and fifty of the flowering plants of Tierra that it endured for long at each, and that it was con- del Fuego, forming no inconsiderable part of its

332 Charles Darwin scanty flora, are common to Europe, enormously the same or representing each other, and at the same remote as these two points are; and there are many time representing plants of Europe, not found in the closely allied species. On the lofty mountains of intervening hot lowlands. A list of the genera col- equatorial America a host of peculiar species belong- lected on the loftier peaks of Java raises a picture of ing to European genera occur. On the highest moun- a collection made on a hill in Europe! Still more strik- tains of Brazil, some few European genera were ing is the fact that southern Australian forms are found by Gardner, which do not exist in the wide clearly represented by plants growing on the sum- intervening hot countries. So on the Silla of Caraccas mits of the mountains of Borneo. Some of these Aus- the illustrious Humboldt long ago found species tralian forms, as I hear from Dr. Hooker, extend along belonging to genera characteristic of the Cordillera. the heights of the peninsula of Malacca, and are thinly On the mountains of Abyssinia, several European scattered, on the one hand over India and on the other forms and some few representatives of the peculiar as far north as Japan. flora of the Cape of Good Hope occur. At the Cape On the southern mountains of Australia, Dr. F. of Good Hope a very few European species, believed Muller has discovered several European species; not to have been introduced by man, and on the other species, not introduced by man, occur on the mountains, some few representative European forms lowlands; and a long list can be given, as I am in- are found, which have not been discovered in the formed by Dr. Hooker, of European genera, found intertropical parts of Africa. On the Himalaya, and in Australia, but not in the intermediate torrid re- on the isolated mountain-ranges of the peninsula of gions. In the admirable ‘Introduction to the Flora of India, on the heights of Ceylon, and on the volcanic New Zealand,’ by Dr. Hooker, analogous and strik- cones of Java, many plants occur, either identically ing facts are given in regard to the plants of that large

333 On the Origin of Species island. Hence we see that throughout the world, the It should be observed that the northern species and plants growing on the more lofty mountains, and on forms found in the southern parts of the southern the temperate lowlands of the northern and south- hemisphere, and on the mountain-ranges of the inter- ern hemispheres, are sometimes identically the same; tropical regions, are not arctic, but belong to the north- but they are much oftener specifically distinct, ern temperate zones. As Mr. H. C. Watson has recently though related to each other in a most remarkable remarked, ‘In receding from polar towards equato- manner. rial latitudes, the Alpine or mountain floras really This brief abstract applies to plants alone: some become less and less arctic.’ Many of the forms living strictly analogous facts could be given on the distri- on the mountains of the warmer regions of the earth bution of terrestrial animals. In marine productions, and in the southern hemisphere are of doubtful value, similar cases occur; as an example, I may quote a being ranked by some naturalists as specifically dis- remark by the highest authority, Prof. Dana, that ‘it tinct, by others as varieties; but some are certainly is certainly a wonderful fact that New Zealand identical, and many, though closely related to north- should have a closer resemblance in its crustacea to ern forms, must be ranked as distinct species. Great Britain, its antipode, than to any other part of Now let us see what light can be thrown on the the world.’ Sir J. Richardson, also, speaks of the re- foregoing facts, on the belief, supported as it is by a appearance on the shores of New Zealand, Tasma- large body of geological evidence, that the whole nia, &c., of northern forms of fish. Dr. Hooker informs world, or a large part of it, was during the Glacial me that twenty-five species of Algae are common to period simultaneously much colder than at present. New Zealand and to Europe, but have not been The Glacial period, as measured by years, must have found in the intermediate tropical seas. been very long; and when we remember over what

334 Charles Darwin vast spaces some naturalised plants and animals have been placed under somewhat new conditions, have spread within a few centuries, this period will will have suffered less. And it is certain that many have been ample for any amount of migration. As temperate plants, if protected from the inroads of the cold came slowly on, all the tropical plants and competitors, can withstand a much warmer climate other productions will have retreated from both sides than their own. Hence, it seems to me possible, bear- towards the equator, followed in the rear by the tem- ing in mind that the tropical productions were in a perate productions, and these by the arctic; but with suffering state and could not have presented a firm the latter we are not now concerned. The tropical front against intruders, that a certain number of the plants probably suffered much extinction; how much more vigorous and dominant temperate forms might no one can say; perhaps formerly the tropics sup- have penetrated the native ranks and have reached ported as many species as we see at the present day or even crossed the equator. The invasion would, of crowded together at the Cape of Good Hope, and in course, have been greatly favoured by high land, and parts of temperate Australia. As we know that many perhaps by a dry climate; for Dr. Falconer informs tropical plants and animals can withstand a consid- me that it is the damp with the heat of the tropics erable amount of cold, many might have escaped which is so destructive to perennial plants from a extermination during a moderate fall of temperature, temperate climate. On the other hand, the most hu- more especially by escaping into the warmest spots. mid and hottest districts will have afforded an asy- But the great fact to bear in mind is, that all tropical lum to the tropical natives. The mountain-ranges productions will have suffered to a certain extent. north-west of the Himalaya, and the long line of the On the other hand, the temperate productions, after Cordillera, seem to have afforded two great lines of migrating nearer to the equator, though they will invasion: and it is a striking fact, lately communi-

335 On the Origin of Species cated to me by Dr. Hooker, that all the flowering tions, migrated during the Glacial period from the plants, about forty-six in number, common to Tierra northern and southern temperate zones into the in- del Fuego and to Europe still exist in North America, tertropical regions, and some even crossed the equa- which must have lain on the line of march. But I do tor. As the warmth returned, these temperate forms not doubt that some temperate productions entered would naturally ascend the higher mountains, be- and crossed even the lowlands of the tropics at the ing exterminated on the lowlands; those which had period when the cold was most intense,—when arc- not reached the equator, would re-migrate northward tic forms had migrated some twenty-five degrees of or southward towards their former homes; but the latitude from their native country and covered the forms, chiefly northern, which had crossed the equa- land at the foot of the Pyrenees. At this period of tor, would travel still further from their homes into extreme cold, I believe that the climate under the the more temperate latitudes of the opposite hemi- equator at the level of the sea was about the same sphere. Although we have reason to believe from with that now felt there at the height of six or seven geological evidence that the whole body of arctic thousand feet. During this the coldest period, I sup- shells underwent scarcely any modification during pose that large spaces of the tropical lowlands were their long southern migration and re-migration clothed with a mingled tropical and temperate veg- northward, the case may have been wholly different etation, like that now growing with strange luxuri- with those intruding forms which settled themselves ance at the base of the Himalaya, as graphically de- on the intertropical mountains, and in the southern scribed by Hooker. hemisphere. These being surrounded by strangers Thus, as I believe, a considerable number of plants, will have had to compete with many new forms of a few terrestrial animals, and some marine produc- life; and it is probable that selected modifications in

336 Charles Darwin their structure, habits, and constitutions will have during the Glacial period, the northern forms were profited them. Thus many of these wanderers, enabled to beat the less powerful southern forms. though still plainly related by inheritance to their Just in the same manner as we see at the present day, brethren of the northern or southern hemispheres, that very many European productions cover the now exist in their new homes as well-marked variet- ground in La Plata, and in a lesser degree in Austra- ies or as distinct species. lia, and have to a certain extent beaten the natives; It is a remarkable fact, strongly insisted on by whereas extremely few southern forms have become Hooker in regard to America, and by Alph. de naturalised in any part of Europe, though hides, Candolle in regard to Australia, that many more wool, and other objects likely to carry seeds have identical plants and allied forms have apparently been largely imported into Europe during the last migrated from the north to the south, than in a re- two or three centuries from La Plata, and during the versed direction. We see, however, a few southern last thirty or forty years from Australia. Something vegetable forms on the mountains of Borneo and of the same kind must have occurred on the inter- Abyssinia. I suspect that this preponderant migra- tropical mountains: no doubt before the Glacial pe- tion from north to south is due to the greater extent riod they were stocked with endemic Alpine forms; of land in the north, and to the northern forms hav- but these have almost everywhere largely yielded ing existed in their own homes in greater numbers, to the more dominant forms, generated in the larger and having consequently been advanced through areas and more efficient workshops of the north. In natural selection and competition to a higher stage many islands the native productions are nearly of perfection or dominating power, than the south- equalled or even outnumbered by the naturalised; ern forms. And thus, when they became commingled and if the natives have not been actually extermi-

337 On the Origin of Species nated, their numbers have been greatly reduced, and plain such facts, until we can say why one species this is the first stage towards extinction. A mountain and not another becomes naturalised by man’s is an island on the land; and the intertropical moun- agency in a foreign land; why one ranges twice or tains before the Glacial period must have been com- thrice as far, and is twice or thrice as common, as pletely isolated; and I believe that the productions another species within their own homes. of these islands on the land yielded to those pro- I have said that many difficulties remain to be duced within the larger areas of the north, just in the solved: some of the most remarkable are stated with same way as the productions of real islands have admirable clearness by Dr. Hooker in his botanical everywhere lately yielded to continental forms, works on the antarctic regions. These cannot be here naturalised by man’s agency. discussed. I will only say that as far as regards the I am far from supposing that all difficulties are re- occurrence of identical species at points so enor- moved on the view here given in regard to the range mously remote as Kerguelen Land, New Zealand, and affinities of the allied species which live in the and Fuegia, I believe that towards the close of the northern and southern temperate zones and on the Glacial period, icebergs, as suggested by Lyell, have mountains of the intertropical regions. Very many been largely concerned in their dispersal. But the difficulties remain to be solved. I do not pretend to existence of several quite distinct species, belong- indicate the exact lines and means of migration, or ing to genera exclusively confined to the south, at the reason why certain species and not others have these and other distant points of the southern hemi- migrated; why certain species have been modified sphere, is, on my theory of descent with modifica- and have given rise to new groups of forms, and oth- tion, a far more remarkable case of difficulty. For ers have remained unaltered. We cannot hope to ex- some of these species are so distinct, that we cannot

338 Charles Darwin suppose that there has been time since the com- Sir C. Lyell in a striking passage has speculated, in mencement of the Glacial period for their migration, language almost identical with mine, on the effects and for their subsequent modification to the neces- of great alternations of climate on geographical dis- sary degree. The facts seem to me to indicate that tribution. I believe that the world has recently felt peculiar and very distinct species have migrated in one of his great cycles of change; and that on this radiating lines from some common centre; and I am view, combined with modification through natural inclined to look in the southern, as in the northern selection, a multitude of facts in the present distri- hemisphere, to a former and warmer period, before bution both of the same and of allied forms of life the commencement of the Glacial period, when the can be explained. The living waters may be said to antarctic lands, now covered with ice, supported a have flowed during one short period from the north highly peculiar and isolated flora. I suspect that be- and from the south, and to have crossed at the equa- fore this flora was exterminated by the Glacial ep- tor; but to have flowed with greater force from the och, a few forms were widely dispersed to various north so as to have freely inundated the south. As points of the southern hemisphere by occasional the tide leaves its drift in horizontal lines, though means of transport, and by the aid, as halting-places, rising higher on the shores where the tide rises high- of existing and now sunken islands, and perhaps at est, so have the living waters left their living drift on the commencement of the Glacial period, by icebergs. our mountain-summits, in a line gently rising from By these means, as I believe, the southern shores of the arctic lowlands to a great height under the equa- America, Australia, New Zealand have become tor. The various beings thus left stranded may be slightly tinted by the same peculiar forms of veg- compared with savage races of man, driven up and etable life. surviving in the mountain-fastnesses of almost ev-

339 On the Origin of Species ery land, which serve as a record, full of interest to Chapter XII us, of the former inhabitants of the surrounding lowlands. Geographical Distribution—continued

Distribution of fresh-water productions — On the inhabitants of oceanic islands — Absence of Batra- chians and of terrestrial Mammals — On the relation of the inhabitants of islands to those of the nearest mainland —On colonisation from the nearest source with subsequent modification —Summary of the last and present chapters.

AS LAKES AND RIVER-SYSTEMS are separated from each other by barriers of land, it might have been thought that fresh-water productions would not have ranged widely within the same country, and as the sea is apparently a still more impassable barrier, that they never would have extended to distant countries. But the case is exactly the reverse. Not only have many fresh-water species, belonging to quite different classes, an enormous range, but allied species pre-

340 Charles Darwin vail in a remarkable manner throughout the world. I that of the live fish not rarely dropped by whirlwinds well remember, when first collecting in the fresh in India, and the vitality of their ova when removed waters of Brazil, feeling much surprise at the simi- from the water. But I am inclined to attribute the dis- larity of the fresh-water insects, shells, &c., and at persal of fresh-water fish mainly to slight changes the dissimilarity of the surrounding terrestrial be- within the recent period in the level of the land, hav- ings, compared with those of Britain. ing caused rivers to flow into each other. Instances, But this power in fresh-water productions of rang- also, could be given of this having occurred during ing widely, though so unexpected, can, I think, in floods, without any change of level. We have evi- most cases be explained by their having become fit- dence in the loess of the Rhine of considerable ted, in a manner highly useful to them, for short and changes of level in the land within a very recent geo- frequent migrations from pond to pond, or from logical period, and when the surface was peopled stream to stream; and liability to wide dispersal by existing land and fresh-water shells. The wide would follow from this capacity as an almost neces- difference of the fish on opposite sides of continu- sary consequence. We can here consider only a few ous mountain-ranges, which from an early period cases. In regard to fish, I believe that the same spe- must have parted river-systems and completely pre- cies never occur in the fresh waters of distant conti- vented their inosculation, seems to lead to this same nents. But on the same continent the species often conclusion. With respect to allied fresh-water fish range widely and almost capriciously; for two river- occurring at very distant points of the world, no systems will have some fish in common and some doubt there are many cases which cannot at present different. A few facts seem to favour the possibility be explained: but some fresh-water fish belong to of their occasional transport by accidental means; like very ancient forms, and in such cases there will have

341 On the Origin of Species been ample time for great geographical changes, and to be observed—throw some light on this subject. consequently time and means for much migration. When a duck suddenly emerges from a pond cov- In the second place, salt-water fish can with care be ered with duck-weed, I have twice seen these little slowly accustomed to live in fresh water; and, ac- plants adhering to its back; and it has happened to cording to Valenciennes, there is hardly a single me, in removing a little duck-weed from one group of fishes confined exclusively to fresh water, aquarium to another, that I have quite unintention- so that we may imagine that a marine member of a ally stocked the one with fresh-water shells from the fresh-water group might travel far along the shores other. But another agency is perhaps more effectual: of the sea, and subsequently become modified and I suspended a duck’s feet, which might represent adapted to the fresh waters of a distant land. those of a bird sleeping in a natural pond, in an Some species of fresh-water shells have a very wide aquarium, where many ova of fresh-water shells range, and allied species, which, on my theory, are were hatching; and I found that numbers of the ex- descended from a common parent and must have tremely minute and just hatched shells crawled on proceeded from a single source, prevail throughout the feet, and clung to them so firmly that when taken the world. Their distribution at first perplexed me out of the water they could not be jarred off, though much, as their ova are not likely to be transported at a somewhat more advanced age they would vol- by birds, and they are immediately killed by sea untarily drop off. These just hatched molluscs, water, as are the adults. I could not even understand though aquatic in their nature, survived on the how some naturalised species have rapidly spread duck’s feet, in damp air, from twelve to twenty hours; throughout the same country. But two facts, which I and in this length of time a duck or heron might fly have observed—and no doubt many others remain at least six or seven hundred miles, and would be

342 Charles Darwin sure to alight on a pool or rivulet, if blown across edges of ponds, if suddenly flushed, would be the sea to an oceanic island or to any other distant point. most likely to have muddy feet. Birds of this order I Sir Charles Lyell also informs me that a Dyticus has can show are the greatest wanderers, and are occa- been caught with an Ancylus (a fresh-water shell like sionally found on the most remote and barren islands a limpet) firmly adhering to it; and a water-beetle of in the open ocean; they would not be likely to alight the same family, a Colymbetes, once flew on board on the surface of the sea, so that the dirt would not the ‘Beagle,’ when forty-five miles distant from the be washed off their feet; when making land, they nearest land: how much farther it might have flown would be sure to fly to their natural fresh-water with a favouring gale no one can tell. haunts. I do not believe that botanists are aware how With respect to plants, it has long been known what charged the mud of ponds is with seeds: I have tried enormous ranges many fresh-water and even marsh- several little experiments, but will here give only species have, both over continents and to the most the most striking case: I took in February three table- remote oceanic islands. This is strikingly shown, as spoonfuls of mud from three different points, be- remarked by Alph. de Candolle, in large groups of neath water, on the edge of a little pond; this mud terrestrial plants, which have only a very few aquatic when dry weighed only 6 3/4 ounces; I kept it cov- members; for these latter seem immediately to ac- ered up in my study for six months, pulling up and quire, as if in consequence, a very wide range. I think counting each plant as it grew; the plants were of favourable means of dispersal explain this fact. I have many kinds, and were altogether 537 in number; and before mentioned that earth occasionally, though yet the viscid mud was all contained in a breakfast rarely, adheres in some quantity to the feet and beaks cup! Considering these facts, I think it would be an of birds. Wading birds, which frequent the muddy inexplicable circumstance if water-birds did not

343 On the Origin of Species transport the seeds of fresh-water plants to vast dis- southern water-lily (probably, according to Dr. tances, and if consequently the range of these plants Hooker, the Nelumbium luteum) in a heron’s stom- was not very great. The same agency may have come ach; although I do not know the fact, yet analogy into play with the eggs of some of the smaller fresh- makes me believe that a heron flying to another pond water animals. and getting a hearty meal of fish, would probably re- Other and unknown agencies probably have also ject from its stomach a pellet containing the seeds of played a part. I have stated that fresh-water fish eat the Nelumbium undigested; or the seeds might be some kinds of seeds, though they reject many other dropped by the bird whilst feeding its young, in the kinds after having swallowed them; even small fish same way as fish are known sometimes to be dropped. swallow seeds of moderate size, as of the yellow In considering these several means of distribution, water-lily and Potamogeton. Herons and other birds, it should be remembered that when a pond or stream century after century, have gone on daily devouring is first formed, for instance, on a rising islet, it will fish; they then take flight and go to other waters, or be unoccupied; and a single seed or egg will have a are blown across the sea; and we have seen that seeds good chance of succeeding. Although there will al- retain their power of germination, when rejected in ways be a struggle for life between the individuals pellets or in excrement, many hours afterwards. of the species, however few, already occupying any When I saw the great size of the seeds of that fine pond, yet as the number of kinds is small, compared water-lily, the Nelumbium, and remembered Alph. with those on the land, the competition will prob- de Candolle’s remarks on this plant, I thought that ably be less severe between aquatic than between its distribution must remain quite inexplicable; but terrestrial species; consequently an intruder from the Audubon states that he found the seeds of the great waters of a foreign country, would have a better

344 Charles Darwin chance of seizing on a place, than in the case of ter- and drops them in another equally well fitted for restrial colonists. We should, also, remember that them. some, perhaps many, fresh-water productions are On the Inhabitants of Oceanic Islands. — We now low in the scale of nature, and that we have reason come to the last of the three classes of facts, which I to believe that such low beings change or become have selected as presenting the greatest amount of modified less quickly than the high; and this will difficulty, on the view that all the individuals both give longer time than the average for the migration of the same and of allied species have descended of the same aquatic species. We should not forget from a single parent; and therefore have all pro- the probability of many species having formerly ceeded from a common birthplace, notwithstanding ranged as continuously as fresh-water productions that in the course of time they have come to inhabit ever can range, over immense areas, and having sub- distant points of the globe. I have already stated that sequently become extinct in intermediate regions. I cannot honestly admit Forbes’s view on continen- But the wide distribution of fresh-water plants and tal extensions, which, if legitimately followed out, of the lower animals, whether retaining the same would lead to the belief that within the recent pe- identical form or in some degree modified, I believe riod all existing islands have been nearly or quite mainly depends on the wide dispersal of their seeds joined to some continent. This view would remove and eggs by animals, more especially by fresh-wa- many difficulties, but it would not, I think, explain ter birds, which have large powers of flight, and natu- all the facts in regard to insular productions. In the rally travel from one to another and often distant following remarks I shall not confine myself to the piece of water. Nature, like a careful gardener, thus mere question of dispersal; but shall consider some takes her seeds from a bed of a particular nature, other facts, which bear on the truth of the two theo-

345 On the Origin of Species ries of independent creation and of descent with Zealand and on every other oceanic island which can modification. be named. In St. Helena there is reason to believe The species of all kinds which inhabit oceanic is- that the naturalised plants and animals have nearly lands are few in number compared with those on or quite exterminated many native productions. He equal continental areas: Alph. de Candolle admits who admits the doctrine of the creation of each sepa- this for plants, and Wollaston for insects. If we look rate species, will have to admit, that a sufficient num- to the large size and varied stations of New Zealand, ber of the best adapted plants and animals have not extending over 780 miles of latitude, and compare been created on oceanic islands; for man has unin- its flowering plants, only 750 in number, with those tentionally stocked them from various sources far on an equal area at the Cape of Good Hope or in more fully and perfectly than has nature. Australia, we must, I think, admit that something Although in oceanic islands the number of kinds quite independently of any difference in physical of inhabitants is scanty, the proportion of endemic conditions has caused so great a difference in num- species (i.e. those found nowhere else in the world) ber. Even the uniform county of Cambridge has 847 is often extremely large. If we compare, for instance, plants, and the little island of Anglesea 764, but a the number of the endemic land-shells in Madeira, few ferns and a few introduced plants are included or of the endemic birds in the Galapagos Archi- in these numbers, and the comparison in some other pelago, with the number found on any continent, and respects is not quite fair. We have evidence that the then compare the area of the islands with that of the barren island of Ascension aboriginally possessed continent, we shall see that this is true. This fact might under half-a-dozen flowering plants; yet many have have been expected on my theory, for, as already become naturalised on it, as they have on New explained, species occasionally arriving after long

346 Charles Darwin intervals in a new and isolated district, and having during their great annual migrations, visit either to compete with new associates, will be eminently periodically or occasionally this island. Madeira liable to modification, and will often produce groups does not possess one peculiar bird, and many Euro- of modified descendants. But it by no means follows, pean and African birds are almost every year blown that, because in an island nearly all the species of there, as I am informed by Mr. E. V. Harcourt. So one class are peculiar, those of another class, or of that these two islands of Bermuda and Madeira have another section of the same class, are peculiar; and been stocked by birds, which for long ages have this difference seems to depend on the species which struggled together in their former homes, and have do not become modified having immigrated with become mutually adapted to each other; and when facility and in a body, so that their mutual relations settled in their new homes, each kind will have been have not been much disturbed. Thus in the kept by the others to their proper places and habits, Galapagos Islands nearly every land-bird, but only and will consequently have been little liable to modi- two out of the eleven marine birds, are peculiar; and fication. Madeira, again, is inhabited by a wonder- it is obvious that marine birds could arrive at these ful number of peculiar land-shells, whereas not one islands more easily than land-birds. Bermuda, on the species of sea-shell is confined to its shores: now, other hand, which lies at about the same distance though we do not know how seashells are dispersed, from North America as the Galapagos Islands do yet we can see that their eggs or larvae, perhaps at- from South America, and which has a very peculiar tached to seaweed or floating timber, or to the feet soil, does not possess one endemic land bird; and of wading-birds, might be transported far more eas- we know from Mr. J. M. Jones’s admirable account ily than land-shells, across three or four hundred of Bermuda, that very many North American birds, miles of open sea. The different orders of insects in

347 On the Origin of Species Madeira apparently present analogous facts. difficulty on my view, for a hooked seed might be Oceanic islands are sometimes deficient in certain transported to an island by some other means; and classes, and their places are apparently occupied by the plant then becoming slightly modified, but still the other inhabitants; in the Galapagos Islands rep- retaining its hooked seeds, would form an endemic tiles, and in New Zealand gigantic wingless birds, species, having as useless an appendage as any ru- take the place of mammals. In the plants of the dimentary organ,—for instance, as the shrivelled Galapagos Islands, Dr. Hooker has shown that the wings under the soldered elytra of many insular proportional numbers of the different orders are very beetles. Again, islands often possess trees or bushes different from what they are elsewhere. Such cases belonging to orders which elsewhere include only are generally accounted for by the physical condi- herbaceous species; now trees, as Alph. de Candolle tions of the islands; but this explanation seems to has shown, generally have, whatever the cause may me not a little doubtful. Facility of immigration, I be, confined ranges. Hence trees would be little likely believe, has been at least as important as the nature to reach distant oceanic islands; and an herbaceous of the conditions. plant, though it would have no chance of success- Many remarkable little facts could be given with fully competing in stature with a fully developed respect to the inhabitants of remote islands. For in- tree, when established on an island and having to stance, in certain islands not tenanted by mammals, compete with herbaceous plants alone, might readily some of the endemic plants have beautifully hooked gain an advantage by growing taller and taller and seeds; yet few relations are more striking than the overtopping the other plants. If so, natural selection adaptation of hooked seeds for transportal by the would often tend to add to the stature of herbaceous wool and fur of quadrupeds. This case presents no plants when growing on an island, to whatever or-

348 Charles Darwin der they belonged, and thus convert them first into in their transportal across the sea, and therefore why bushes and ultimately into trees. they do not exist on any oceanic island. But why, on With respect to the absence of whole orders on oce- the theory of creation, they should not have been cre- anic islands, Bory St. Vincent long ago remarked that ated there, it would be very difficult to explain. Batrachians (frogs, toads, newts) have never been Mammals offer another and similar case. I have found on any of the many islands with which the carefully searched the oldest voyages, but have not great oceans are studded. I have taken pains to verify finished my search; as yet I have not found a single this assertion, and I have found it strictly true. I have, instance, free from doubt, of a terrestrial mammal however, been assured that a frog exists on the moun- (excluding domesticated animals kept by the natives) tains of the great island of New Zealand; but I sus- inhabiting an island situated above 300 miles from a pect that this exception (if the information be cor- continent or great continental island; and many is- rect) may be explained through glacial agency. This lands situated at a much less distance are equally general absence of frogs, toads, and newts on so barren. The Falkland Islands, which are inhabited many oceanic islands cannot be accounted for by by a wolf-like fox, come nearest to an exception; but their physical conditions; indeed it seems that islands this group cannot be considered as oceanic, as it lies are peculiarly well fitted for these animals; for frogs on a bank connected with the mainland; moreover, have been introduced into Madeira, the Azores, and icebergs formerly brought boulders to its western Mauritius, and have multiplied so as to become a shores, and they may have formerly transported nuisance. But as these animals and their spawn are foxes, as so frequently now happens in the arctic re- known to be immediately killed by sea-water, on my gions. Yet it cannot be said that small islands will view we can see that there would be great difficulty not support small mammals, for they occur in many

349 On the Origin of Species parts of the world on very small islands, if close to a mote islands? On my view this question can easily continent; and hardly an island can be named on be answered; for no terrestrial mammal can be trans- which our smaller quadrupeds have not become ported across a wide space of sea, but bats can fly naturalised and greatly multiplied. It cannot be said, across. Bats have been seen wandering by day far on the ordinary view of creation, that there has not over the Atlantic Ocean; and two North American been time for the creation of mammals; many volca- species either regularly or occasionally visit Ber- nic islands are sufficiently ancient, as shown by the muda, at the distance of 600 miles from the main- stupendous degradation which they have suffered land. I hear from Mr. Tomes, who has specially stud- and by their tertiary strata: there has also been time ied this family, that many of the same species have for the production of endemic species belonging to enormous ranges, and are found on continents and other classes; and on continents it is thought that on far distant islands. Hence we have only to sup- mammals appear and disappear at a quicker rate pose that such wandering species have been modi- than other and lower animals. Though terrestrial fied through natural selection in their new homes in mammals do not occur on oceanic islands, aerial relation to their new position, and we can understand mammals do occur on almost every island. New the presence of endemic bats on islands, with the Zealand possesses two bats found nowhere else in absence of all terrestrial mammals. the world: Norfolk Island, the Viti Archipelago, the Besides the absence of terrestrial mammals in re- Bonin Islands, the Caroline and Marianne Archipela- lation to the remoteness of islands from continents, goes, and Mauritius, all possess their peculiar bats. there is also a relation, to a certain extent indepen- Why, it may be asked, has the supposed creative dent of distance, between the depth of the sea sepa- force produced bats and no other mammals on re- rating an island from the neighbouring mainland,

350 Charles Darwin and the presence in both of the same mammiferous same on both sides; we meet with analogous facts species or of allied species in a more or less modi- on many islands separated by similar channels from fied condition. Mr. Windsor Earl has made some Australia. The West Indian Islands stand on a deeply striking observations on this head in regard to the submerged bank, nearly 1000 fathoms in depth, and great Malay Archipelago, which is traversed near here we find American forms, but the species and Celebes by a space of deep ocean; and this space even the genera are distinct. As the amount of modi- separates two widely distinct mammalian faunas. On fication in all cases depends to a certain degree on either side the islands are situated on moderately the lapse of time, and as during changes of level it is deep submarine banks, and they are inhabited by obvious that islands separated by shallow channels closely allied or identical quadrupeds. No doubt are more likely to have been continuously united some few anomalies occur in this great archipelago, within a recent period to the mainland than islands and there is much difficulty in forming a judgment separated by deeper channels, we can understand in some cases owing to the probable naturalisation the frequent relation between the depth of the sea of certain mammals through man’s agency; but we and the degree of affinity of the mammalian inhabit- shall soon have much light thrown on the natural ants of islands with those of a neighbouring conti- history of this archipelago by the admirable zeal and nent,—an inexplicable relation on the view of inde- researches of Mr. Wallace. I have not as yet had time pendent acts of creation. to follow up this subject in all other quarters of the All the foregoing remarks on the inhabitants of oce- world; but as far as I have gone, the relation gener- anic islands,—namely, the scarcity of kinds—the ally holds good. We see Britain separated by a shal- richness in endemic forms in particular classes or low channel from Europe, and the mammals are the sections of classes,—the absence of whole groups,

351 On the Origin of Species as of batrachians, and of terrestrial mammals not- halting-places, of which not a wreck now remains, withstanding the presence of aerial bats,—the sin- must not be overlooked. I will here give a single in- gular proportions of certain orders of plants,—her- stance of one of the cases of difficulty. Almost all baceous forms having been developed into trees, oceanic islands, even the most isolated and small- &c.,—seem to me to accord better with the view of est, are inhabited by land-shells, generally by en- occasional means of transport having been largely demic species, but sometimes by species found else- efficient in the long course of time, than with the view where. Dr. Aug. A. Gould has given several interest- of all our oceanic islands having been formerly con- ing cases in regard to the land-shells of the islands nected by continuous land with the nearest conti- of the Pacific. Now it is notorious that land-shells nent; for on this latter view the migration would are very easily killed by salt; their eggs, at least such probably have been more complete; and if modifi- as I have tried, sink in sea-water and are killed by it. cation be admitted, all the forms of life would have Yet there must be, on my view, some unknown, but been more equally modified, in accordance with the highly efficient means for their transportal. Would paramount importance of the relation of organism the just-hatched young occasionally crawl on and to organism. adhere to the feet of birds roosting on the ground, I do not deny that there are many and grave diffi- and thus get transported? It occurred to me that land- culties in understanding how several of the inhabit- shells, when hybernating and having a membranous ants of the more remote islands, whether still retain- diaphragm over the mouth of the shell, might be ing the same specific form or modified since their floated in chinks of drifted timber across moderately arrival, could have reached their present homes. But wide arms of the sea. And I found that several spe- the probability of many islands having existed as cies did in this state withstand uninjured an immer-

352 Charles Darwin sion in sea-water during seven days: one of these yet the close affinity of most of these birds to Ameri- shells was the Helix pomatia, and after it had again can species in every character, in their habits, ges- hybernated I put it in sea-water for twenty days, and tures, and tones of voice, was manifest. So it is with it perfectly recovered. As this species has a thick cal- the other animals, and with nearly all the plants, as careous operculum, I removed it, and when it had shown by Dr. Hooker in his admirable memoir on formed a new membranous one, I immersed it for the Flora of this archipelago. The naturalist, looking fourteen days in sea-water, and it recovered and at the inhabitants of these volcanic islands in the Pa- crawled away: but more experiments are wanted on cific, distant several hundred miles from the conti- this head. nent, yet feels that he is standing on American land. The most striking and important fact for us in re- Why should this be so? why should the species gard to the inhabitants of islands, is their affinity to which are supposed to have been created in the those of the nearest mainland, without being actu- Galapagos Archipelago, and nowhere else, bear so ally the same species. Numerous instances could be plain a stamp of affinity to those created in America? given of this fact. I will give only one, that of the There is nothing in the conditions of life, in the geo- Galapagos Archipelago, situated under the equator, logical nature of the islands, in their height or cli- between 500 and 600 miles from the shores of South mate, or in the proportions in which the several America. Here almost every product of the land and classes are associated together, which resembles water bears the unmistakeable stamp of the Ameri- closely the conditions of the South American coast: can continent. There are twenty-six land birds, and in fact there is a considerable dissimilarity in all these twenty-five of these are ranked by Mr. Gould as dis- respects. On the other hand, there is a considerable tinct species, supposed to have been created here; degree of resemblance in the volcanic nature of the

353 On the Origin of Species soil, in climate, height, and size of the islands, be- nearer to Africa than to America, are related, and that tween the Galapagos and Cape de Verde Archipela- very closely, as we know from Dr. Hooker’s account, gos: but what an entire and absolute difference in to those of America: but on the view that this island their inhabitants! The inhabitants of the Cape de has been mainly stocked by seeds brought with earth Verde Islands are related to those of Africa, like those and stones on icebergs, drifted by the prevailing of the Galapagos to America. I believe this grand fact currents, this anomaly disappears. New Zealand in can receive no sort of explanation on the ordinary its endemic plants is much more closely related to view of independent creation; whereas on the view Australia, the nearest mainland, than to any other here maintained, it is obvious that the Galapagos region: and this is what might have been expected; Islands would be likely to receive colonists, whether but it is also plainly related to South America, which, by occasional means of transport or by formerly con- although the next nearest continent, is so enormously tinuous land, from America; and the Cape de Verde remote, that the fact becomes an anomaly. But this Islands from Africa; and that such colonists would difficulty almost disappears on the view that both be liable to modification;—the principle of inherit- New Zealand, South America, and other southern ance still betraying their original birthplace. lands were long ago partially stocked from a nearly Many analogous facts could be given: indeed it is intermediate though distant point, namely from the an almost universal rule that the endemic produc- antarctic islands, when they were clothed with veg- tions of islands are related to those of the nearest etation, before the commencement of the Glacial pe- continent, or of other near islands. The exceptions riod. The affinity, which, though feeble, I am assured are few, and most of them can be explained. Thus by Dr. Hooker is real, between the flora of the south- the plants of Kerguelen Land, though standing western corner of Australia and of the Cape of Good

354 Charles Darwin Hope, is a far more remarkable case, and is at present views; for it may be asked, how has it happened in inexplicable: but this affinity is confined to the plants, the several islands situated within sight of each other, and will, I do not doubt, be some day explained. having the same geological nature, the same height, The law which causes the inhabitants of an archi- climate, &c., that many of the immigrants should pelago, though specifically distinct, to be closely al- have been differently modified, though only in a lied to those of the nearest continent, we sometimes small degree. This long appeared to me a great dif- see displayed on a small scale, yet in a most inter- ficulty: but it arises in chief part from the deeply- esting manner, within the limits of the same archi- seated error of considering the physical conditions pelago. Thus the several islands of the Galapagos of a country as the most important for its inhabit- Archipelago are tenanted, as I have elsewhere shown, ants; whereas it cannot, I think, be disputed that the in a quite marvellous manner, by very closely re- nature of the other inhabitants, with which each has lated species; so that the inhabitants of each sepa- to compete, is at least as important, and generally a rate island, though mostly distinct, are related in an far more important element of success. Now if we incomparably closer degree to each other than to the look to those inhabitants of the Galapagos Archi- inhabitants of any other part of the world. And this pelago which are found in other parts of the world is just what might have been expected on my view, (laying on one side for the moment the endemic spe- for the islands are situated so near each other that cies, which cannot be here fairly included, as we are they would almost certainly receive immigrants from considering how they have come to be modified the same original source, or from each other. But this since their arrival), we find a considerable amount dissimilarity between the endemic inhabitants of the of difference in the several islands. This difference islands may be used as an argument against my might indeed have been expected on the view of the

355 On the Origin of Species islands having been stocked by occasional means of some analogous instances, is that the new species transport—a seed, for instance, of one plant having formed in the separate islands have not quickly been brought to one island, and that of another plant spread to the other islands. But the islands, though to another island. Hence when in former times an in sight of each other, are separated by deep arms of immigrant settled on any one or more of the islands, the sea, in most cases wider than the British Chan- or when it subsequently spread from one island to nel, and there is no reason to suppose that they have another, it would undoubtedly be exposed to dif- at any former period been continuously united. The ferent conditions of life in the different islands, for it currents of the sea are rapid and sweep across the would have to compete with different sets of organ- archipelago, and gales of wind are extraordinarily isms: a plant, for instance, would find the best-fitted rare; so that the islands are far more effectually sepa- ground more perfectly occupied by distinct plants rated from each other than they appear to be on a in one island than in another, and it would be ex- map. Nevertheless a good many species, both those posed to the attacks of somewhat different enemies. found in other parts of the world and those confined If then it varied, natural selection would probably to the archipelago, are common to the several islands, favour different varieties in the different islands. and we may infer from certain facts that these have Some species, however, might spread and yet retain probably spread from some one island to the others. the same character throughout the group, just as we But we often take, I think, an erroneous view of the see on continents some species spreading widely and probability of closely allied species invading each remaining the same. other’s territory, when put into free intercommuni- The really surprising fact in this case of the cation. Undoubtedly if one species has any advan- Galapagos Archipelago, and in a lesser degree in tage whatever over another, it will in a very brief

356 Charles Darwin time wholly or in part supplant it; but if both are infer that Charles Island is well stocked with its own equally well fitted for their own places in nature, species, for annually more eggs are laid there than both probably will hold their own places and keep can possibly be reared; and we may infer that the separate for almost any length of time. Being famil- mocking-thrush peculiar to Charles Island is at least iar with the fact that many species, naturalised as well fitted for its home as is the species peculiar through man’s agency, have spread with astonish- to Chatham Island. Sir C. Lyell and Mr. Wollaston ing rapidity over new countries, we are apt to infer have communicated to me a remarkable fact bear- that most species would thus spread; but we should ing on this subject; namely, that Madeira and the ad- remember that the forms which become naturalised joining islet of Porto Santo possess many distinct but in new countries are not generally closely allied to representative land-shells, some of which live in the aboriginal inhabitants, but are very distinct spe- crevices of stone; and although large quantities of cies, belonging in a large proportion of cases, as stone are annually transported from Porto Santo to shown by Alph. de Candolle, to distinct genera. In Madeira, yet this latter island has not become the Galapagos Archipelago, many even of the birds, colonised by the Porto Santo species: nevertheless though so well adapted for flying from island to is- both islands have been colonised by some European land, are distinct on each; thus there are three closely- land-shells, which no doubt had some advantage allied species of mocking-thrush, each confined to over the indigenous species. From these consider- its own island. Now let us suppose the mocking- ations I think we need not greatly marvel at the en- thrush of Chatham Island to be blown to Charles Is- demic and representative species, which inhabit the land, which has its own mocking-thrush: why should several islands of the Galapagos Archipelago, not it succeed in establishing itself there? We may safely having universally spread from island to island. In

357 On the Origin of Species many other instances, as in the several districts of och, are related to those of the surrounding low- the same continent, pre-occupation has probably lands;—thus we have in South America, Alpine hum- played an important part in checking the commin- ming-birds, Alpine rodents, Alpine plants, &c., all gling of species under the same conditions of life. of strictly American forms, and it is obvious that a Thus, the south-east and south-west corners of Aus- mountain, as it became slowly upheaved, would tralia have nearly the same physical conditions, and naturally be colonised from the surrounding low- are united by continuous land, yet they are inhab- lands. So it is with the inhabitants of lakes and ited by a vast number of distinct mammals, birds, marshes, excepting in so far as great facility of trans- and plants. port has given the same general forms to the whole The principle which determines the general char- world. We see this same principle in the blind ani- acter of the fauna and flora of oceanic islands, mals inhabiting the caves of America and of Europe. namely, that the inhabitants, when not identically the Other analogous facts could be given. And it will, I same, yet are plainly related to the inhabitants of that believe, be universally found to be true, that wher- region whence colonists could most readily have ever in two regions, let them be ever so distant, many been derived,—the colonists having been subse- closely allied or representative species occur, there quently modified and better fitted to their new will likewise be found some identical species, show- homes,—is of the widest application throughout ing, in accordance with the foregoing view, that at nature. We see this on every mountain, in every lake some former period there has been intercommuni- and marsh. For Alpine species, excepting in so far as cation or migration between the two regions. And the same forms, chiefly of plants, have spread widely wherever many closely-allied species occur, there throughout the world during the recent Glacial ep- will be found many forms which some naturalists

358 Charles Darwin rank as distinct species, and some as varieties; these that they have on an average a wide range; but only doubtful forms showing us the steps in the process that some of the species range very widely; for the of modification. facility with which widely-ranging species vary and This relation between the power and extent of mi- give rise to new forms will largely determine their gration of a species, either at the present time or at average range. For instance, two varieties of the same some former period under different physical condi- species inhabit America and Europe, and the spe- tions, and the existence at remote points of the world cies thus has an immense range; but, if the variation of other species allied to it, is shown in another and had been a little greater, the two varieties would have more general way. Mr. Gould remarked to me long been ranked as distinct species, and the common ago, that in those genera of birds which range over range would have been greatly reduced. Still less is the world, many of the species have very wide it meant, that a species which apparently has the ca- ranges. I can hardly doubt that this rule is generally pacity of crossing barriers and ranging widely, as in true, though it would be difficult to prove it. the case of certain powerfully-winged birds, will Amongst mammals, we see it strikingly displayed necessarily range widely; for we should never for- in Bats, and in a lesser degree in the Felidae and get that to range widely implies not only the power Canidae. We see it, if we compare the distribution of crossing barriers, but the more important power of butterflies and beetles. So it is with most fresh- of being victorious in distant lands in the struggle water productions, in which so many genera range for life with foreign associates. But on the view of all over the world, and many individual species have the species of a genus having descended from a enormous ranges. It is not meant that in world-rang- single parent, though now distributed to the most ing genera all the species have a wide range, or even remote points of the world, we ought to find, and I

359 On the Origin of Species believe as a general rule we do find, that some at least of ranging widely and of still retaining the same spe- of the species range very widely; for it is necessary cific character. This fact, together with the seeds and that the unmodified parent should range widely, un- eggs of many low forms being very minute and bet- dergoing modification during its diffusion, and ter fitted for distant transportation, probably ac- should place itself under diverse conditions counts for a law which has long been observed, and favourable for the conversion of its offspring, firstly which has lately been admirably discussed by Alph. into new varieties and ultimately into new species. de Candolle in regard to plants, namely, that the In considering the wide distribution of certain gen- lower any group of organisms is, the more widely it era, we should bear in mind that some are extremely is apt to range. ancient, and must have branched off from a common The relations just discussed,—namely, low and parent at a remote epoch; so that in such cases there slowly-changing organisms ranging more widely will have been ample time for great climatal and than the high,—some of the species of widely-rang- geographical changes and for accidents of transport; ing genera themselves ranging widely,—such facts, and consequently for the migration of some of the as alpine, lacustrine, and marsh productions being species into all quarters of the world, where they may related (with the exceptions before specified) to those have become slightly modified in relation to their on the surrounding low lands and dry lands, though new conditions. There is, also, some reason to be- these stations are so different—the very close rela- lieve from geological evidence that organisms low tion of the distinct species which inhabit the islets of in the scale within each great class, generally change the same archipelago,—and especially the striking at a slower rate than the higher forms; and conse- relation of the inhabitants of each whole archipelago quently the lower forms will have had a better chance or island to those of the nearest mainland,—are, I

360 Charles Darwin think, utterly inexplicable on the ordinary view of cated, have descended from the same parents, are the independent creation of each species, but are not insuperable. And we are led to this conclusion, explicable on the view of colonisation from the near- which has been arrived at by many naturalists un- est and readiest source, together with the subsequent der the designation of single centres of creation, by modification and better adaptation of the colonists some general considerations, more especially from to their new homes. the importance of barriers and from the analogical Summary of last and present Chapters — In these distribution of sub-genera, genera, and families. chapters I have endeavoured to show, that if we make With respect to the distinct species of the same ge- due allowance for our ignorance of the full effects of nus, which on my theory must have spread from one all the changes of climate and of the level of the land, parent-source; if we make the same allowances as which have certainly occurred within the recent pe- before for our ignorance, and remember that some riod, and of other similar changes which may have forms of life change most slowly, enormous periods occurred within the same period; if we remember of time being thus granted for their migration, I do how profoundly ignorant we are with respect to the not think that the difficulties are insuperable; though many and curious means of occasional transport,— they often are in this case, and in that of the indi- a subject which has hardly ever been properly viduals of the same species, extremely grave. experimentised on; if we bear in mind how often a As exemplifying the effects of climatal changes species may have ranged continuously over a wide on distribution, I have attempted to show how im- area, and then have become extinct in the intermedi- portant has been the influence of the modern Gla- ate tracts, I think the difficulties in believing that all cial period, which I am fully convinced simulta- the individuals of the same species, wherever lo- neously affected the whole world, or at least great

361 On the Origin of Species meridional belts. As showing how diversified are linked to the extinct beings which formerly inhab- the means of occasional transport, I have discussed ited the same continent. Bearing in mind that the at some little length the means of dispersal of fresh- mutual relations of organism to organism are of the water productions. highest importance, we can see why two areas hav- If the difficulties be not insuperable in admitting ing nearly the same physical conditions should of- that in the long course of time the individuals of the ten be inhabited by very different forms of life; for same species, and likewise of allied species, have according to the length of time which has elapsed proceeded from some one source; then I think all since new inhabitants entered one region; according the grand leading facts of geographical distribution to the nature of the communication which allowed are explicable on the theory of migration (generally certain forms and not others to enter, either in greater of the more dominant forms of life), together with or lesser numbers; according or not, as those which subsequent modification and the multiplication of entered happened to come in more or less direct com- new forms. We can thus understand the high impor- petition with each other and with the aborigines; and tance of barriers, whether of land or water, which according as the immigrants were capable of vary- separate our several zoological and botanical proving more or less rapidly, there would ensue in dif- inces. We can thus understand the localisation of sub- ferent regions, independently of their physical con- genera, genera, and families; and how it is that un- ditions, infinitely diversified conditions of life,— der different latitudes, for instance in South America, there would be an almost endless amount of organic the inhabitants of the plains and mountains, of the action and reaction,—and we should find, as we do forests, marshes, and deserts, are in so mysterious a find, some groups of beings greatly, and some only manner linked together by affinity, and are likewise slightly modified,—some developed in great force,

362 Charles Darwin some existing in scanty numbers—in the different but less closely, to those of the nearest continent or great geographical provinces of the world. other source whence immigrants were probably de- On these same principles, we can understand, as I rived. We can see why in two areas, however distant have endeavoured to show, why oceanic islands from each other, there should be a correlation, in the should have few inhabitants, but of these a great presence of identical species, of varieties, of doubtful number should be endemic or peculiar; and why, in species, and of distinct but representative species. relation to the means of migration, one group of be- As the late Edward Forbes often insisted, there is a ings, even within the same class, should have all its striking parallelism in the laws of life throughout species endemic, and another group should have all time and space: the laws governing the succession its species common to other quarters of the world. of forms in past times being nearly the same with We can see why whole groups of organisms, as batra- those governing at the present time the differences chians and terrestrial mammals, should be absent in different areas. We see this in many facts. The en- from oceanic islands, whilst the most isolated islands durance of each species and group of species is con- possess their own peculiar species of aerial mam- tinuous in time; for the exceptions to the rule are so mals or bats. We can see why there should be some few, that they may fairly be attributed to our not hav- relation between the presence of mammals, in a more ing as yet discovered in an intermediate deposit the or less modified condition, and the depth of the sea forms which are therein absent, but which occur between an island and the mainland. We can clearly above and below: so in space, it certainly is the gen- see why all the inhabitants of an archipelago, though eral rule that the area inhabited by a single species, specifically distinct on the several islets, should be or by a group of species, is continuous; and the ex- closely related to each other, and likewise be related, ceptions, which are not rare, may, as I have attempted

363 On the Origin of Species to show, be accounted for by migration at some quarter of the world, or to those which have changed former period under different conditions or by oc- after having migrated into distant quarters, in both casional means of transport, and by the species hav- cases the forms within each class have been con- ing become extinct in the intermediate tracts. Both nected by the same bond of ordinary generation; and in time and space, species and groups of species have the more nearly any two forms are related in blood, their points of maximum development. Groups of the nearer they will generally stand to each other in species, belonging either to a certain period of time, time and space; in both cases the laws of variation or to a certain area, are often characterised by tri- have been the same, and modifications have been fling characters in common, as of sculpture or colour. accumulated by the same power of natural selection. In looking to the long succession of ages, as in now looking to distant provinces throughout the world, we find that some organisms differ little, whilst others belonging to a different class, or to a different order, or even only to a different family of the same order, differ greatly. In both time and space the lower members of each class generally change less than the higher; but there are in both cases marked exceptions to the rule. On my theory these several relations throughout time and space are intelligible; for whether we look to the forms of life which have changed during successive ages within the same

364 Charles Darwin Chapter XIII found to resemble each other in descending degrees, so that they can be classed in groups under groups. Mutual Affinities of Organic Beings: This classification is evidently not arbitrary like the Morphology: Embryology: Rudimentary grouping of the stars in constellations. The existence Organs of groups would have been of simple signification, if one group had been exclusively fitted to inhabit Classification, groups subordinate to groups — the land, and another the water; one to feed on flesh, Natural system — Rules and difficulties in classifi- another on vegetable matter, and so on; but the case cation, explained on the theory of descent with modi- is widely different in nature; for it is notorious how fication — Classification of varieties — Descent al- commonly members of even the same subgroup ways used in classification — Analogical or adap- have different habits. In our second and fourth chap- tive characters — Affinities, general, complex and ters, on Variation and on Natural Selection, I have radiating — Extinction separates and defines groups attempted to show that it is the widely ranging, the —Morphology, between members of the same class, much diffused and common, that is the dominant between parts of the same individual — Embryol- species belonging to the larger genera, which vary ogy, laws of, explained by variations not superven- most. The varieties, or incipient species, thus pro- ing at an early age, and being inherited at a corre- duced ultimately become converted, as I believe, into sponding age — Rudimentary Organs; their origin new and distinct species; and these, on the principle explained — Summary. of inheritance, tend to produce other new and dominant species. Consequently the groups which are FROM THE FIRST DAWN OF LIFE, all organic beings are now large, and which generally include many domi-

365 On the Origin of Species nant species, tend to go on increasing indefinitely in resent a genus including several species; and all the size. I further attempted to show that from the vary- genera on this line form together one class, for all ing descendants of each species trying to occupy as have descended from one ancient but unseen par- many and as different places as possible in the ent, and, consequently, have inherited something in economy of nature, there is a constant tendency in common. But the three genera on the left hand have, their characters to diverge. This conclusion was sup- on this same principle, much in common, and form ported by looking at the great diversity of the forms a sub-family, distinct from that including the next of life which, in any small area, come into the closest two genera on the right hand, which diverged from competition, and by looking to certain facts in a common parent at the fifth stage of descent. These naturalisation. five genera have also much, though less, in common; I attempted also to show that there is a constant and they form a family distinct from that including tendency in the forms which are increasing in num- the three genera still further to the right hand, which ber and diverging in character, to supplant and ex- diverged at a still earlier period. And all these gen- terminate the less divergent, the less improved, and era, descended from (A), form an order distinct from preceding forms. I request the reader to turn to the the genera descended from (I). So that we here have diagram illustrating the action, as formerly ex- many species descended from a single progenitor plained, of these several principles; and he will see grouped into genera; and the genera are included that the inevitable result is that the modified descen- in, or subordinate to, sub-families, families, and or- dants proceeding from one progenitor become bro- ders, all united into one class. Thus, the grand fact ken up into groups subordinate to groups. In the in natural history of the subordination of group un- diagram each letter on the uppermost line may rep- der group, which, from its familiarity, does not al-

366 Charles Darwin ways sufficiently strike us, is in my judgment fully thus added to our knowledge. Such expressions as explained. that famous one of Linnaeus, and which we often meet Naturalists try to arrange the species, genera, and with in a more or less concealed form, that the charac- families in each class, on what is called the Natural ters do not make the genus, but that the genus gives System. But what is meant by this system? Some the characters, seem to imply that something more is authors look at it merely as a scheme for arranging included in our classification, than mere resemblance. together those living objects which are most alike, I believe that something more is included; and that and for separating those which are most unlike; or propinquity of descent,—the only known cause of the as an artificial means for enunciating, as briefly as similarity of organic beings,—is the bond, hidden as possible, general propositions,—that is, by one sen- it is by various degrees of modification, which is par- tence to give the characters common, for instance, to tially revealed to us by our classifications. all mammals, by another those common to all car- Let us now consider the rules followed in classifi- nivora, by another those common to the dog-genus, cation, and the difficulties which are encountered on and then by adding a single sentence, a full descrip- the view that classification either gives some un- tion is given of each kind of dog. The ingenuity and known plan of creation, or is simply a scheme for utility of this system are indisputable. But many enunciating general propositions and of placing to- naturalists think that something more is meant by gether the forms most like each other. It might have the Natural System; they believe that it reveals the been thought (and was in ancient times thought) that plan of the Creator; but unless it be specified whether those parts of the structure which determined the order in time or space, or what else is meant by the habits of life, and the general place of each being in plan of the Creator, it seems to me that nothing is the economy of nature, would be of very high im-

367 On the Origin of Species portance in classification. Nothing can be more false. whereas the organs of reproduction, with their prod- No one regards the external similarity of a mouse to uct the seed, are of paramount importance! a shrew, of a dugong to a whale, of a whale to a fish, We must not, therefore, in classifying, trust to re- as of any importance. These resemblances, though semblances in parts of the organisation, however im- so intimately connected with the whole life of the portant they may be for the welfare of the being in being, are ranked as merely ‘adaptive or analogical relation to the outer world. Perhaps from this cause characters;’ but to the consideration of these resem- it has partly arisen, that almost all naturalists lay the blances we shall have to recur. It may even be given greatest stress on resemblances in organs of high vi- as a general rule, that the less any part of the tal or physiological importance. No doubt this view organisation is concerned with special habits, the of the classificatory importance of organs which are more important it becomes for classification. As an important is generally, but by no means always, true. instance: Owen, in speaking of the dugong, says, ‘The But their importance for classification, I believe, de- generative organs being those which are most re- pends on their greater constancy throughout large motely related to the habits and food of an animal, I groups of species; and this constancy depends on have always regarded as affording very clear indi- such organs having generally been subjected to less cations of its true affinities. We are least likely in the change in the adaptation of the species to their con- modifications of these organs to mistake a merely ditions of life. That the mere physiological impor- adaptive for an essential character.’ So with plants, tance of an organ does not determine its classifica- how remarkable it is that the organs of vegetation, tory value, is almost shown by the one fact, that in on which their whole life depends, are of little signi- allied groups, in which the same organ, as we have fication, excepting in the first main divisions; every reason to suppose, has nearly the same physi-

368 Charles Darwin ological value, its classificatory value is widely dif- and the differences are of quite subordinate value in ferent. No naturalist can have worked at any group classification; yet no one probably will say that the without being struck with this fact; and it has been antennae in these two divisions of the same order most fully acknowledged in the writings of almost are of unequal physiological importance. Any num- every author. It will suffice to quote the highest au- ber of instances could be given of the varying im- thority, Robert Brown, who in speaking of certain portance for classification of the same important or- organs in the Proteaceae, says their generic impor- gan within the same group of beings. tance, ‘like that of all their parts, not only in this but, Again, no one will say that rudimentary or atro- as I apprehend, in every natural family, is very un- phied organs are of high physiological or vital im- equal, and in some cases seems to be entirely lost.’ portance; yet, undoubtedly, organs in this condition Again in another work he says, the genera of the are often of high value in classification. No one will Connaraceae ‘differ in having one or more ovaria, in dispute that the rudimentary teeth in the upper jaws the existence or absence of albumen, in the imbri- of young ruminants, and certain rudimentary bones cate or valvular aestivation. Any one of these char- of the leg, are highly serviceable in exhibiting the acters singly is frequently of more than generic im- close affinity between Ruminants and Pachyderms. portance, though here even when all taken together Robert Brown has strongly insisted on the fact that they appear insufficient to separate Cnestis from the rudimentary florets are of the highest importance Connarus.’ To give an example amongst insects, in in the classification of the Grasses. one great division of the Hymenoptera, the anten- Numerous instances could be given of characters nae, as Westwood has remarked, are most constant derived from parts which must be considered of very in structure; in another division they differ much, trifling physiological importance, but which are uni-

369 On the Origin of Species versally admitted as highly serviceable in the defi- evident in natural history. Hence, as has often been nition of whole groups. For instance, whether or not remarked, a species may depart from its allies in there is an open passage from the nostrils to the several characters, both of high physiological impor- mouth, the only character, according to Owen, which tance and of almost universal prevalence, and yet absolutely distinguishes fishes and reptiles—the leave us in no doubt where it should be ranked. inflection of the angle of the jaws in Marsupials— Hence, also, it has been found, that a classification the manner in which the wings of insects are folded— founded on any single character, however important mere colour in certain Algae—mere pubescence on that may be, has always failed; for no part of the parts of the flower in grasses—the nature of the der- organisation is universally constant. The importance mal covering, as hair or feathers, in the Vertebrata. If of an aggregate of characters, even when none are the Ornithorhynchus had been covered with feath- important, alone explains, I think, that saying of ers instead of hair, this external and trifling charac- Linnaeus, that the characters do not give the genus, ter would, I think, have been considered by natural- but the genus gives the characters; for this saying ists as important an aid in determining the degree of seems founded on an appreciation of many trifling affinity of this strange creature to birds and reptiles, points of resemblance, too slight to be defined. Cer- as an approach in structure in any one internal and tain plants, belonging to the Malpighiaceae, bear important organ. perfect and degraded flowers; in the latter, as A. de The importance, for classification, of trifling char- Jussieu has remarked, ‘the greater number of the acters, mainly depends on their being correlated with characters proper to the species, to the genus, to the several other characters of more or less importance. family, to the class, disappear, and thus laugh at our The value indeed of an aggregate of characters is very classification.’ But when Aspicarpa produced in

370 Charles Darwin France, during several years, only degraded flow- be discovered between them, especial value is set on ers, departing so wonderfully in a number of the them. As in most groups of animals, important or- most important points of structure from the proper gans, such as those for propelling the blood, or for type of the order, yet M. Richard sagaciously saw, aerating it, or those for propagating the race, are found as Jussieu observes, that this genus should still be nearly uniform, they are considered as highly service- retained amongst the Malpighiaceae. This case seems able in classification; but in some groups of animals to me well to illustrate the spirit with which our clas- all these, the most important vital organs, are found sifications are sometimes necessarily founded. to offer characters of quite subordinate value. Practically when naturalists are at work, they do We can see why characters derived from the em- not trouble themselves about the physiological value bryo should be of equal importance with those de- of the characters which they use in defining a group, rived from the adult, for our classifications of course or in allocating any particular species. If they find a include all ages of each species. But it is by no means character nearly uniform, and common to a great obvious, on the ordinary view, why the structure of number of forms, and not common to others, they the embryo should be more important for this pur- use it as one of high value; if common to some lesser pose than that of the adult, which alone plays its full number, they use it as of subordinate value. This part in the economy of nature. Yet it has been strongly principle has been broadly confessed by some natu- urged by those great naturalists, Milne Edwards and ralists to be the true one; and by none more clearly Agassiz, that embryonic characters are the most im- than by that excellent botanist, Aug. St. Hilaire. If portant of any in the classification of animals; and certain characters are always found correlated with this doctrine has very generally been admitted as others, though no apparent bond of connexion can true. The same fact holds good with flowering plants,

371 On the Origin of Species of which the two main divisions have been founded more especially in very large groups of closely al- on characters derived from the embryo,—on the lied forms. Temminck insists on the utility or even number and position of the embryonic leaves or coty- necessity of this practice in certain groups of birds; ledons, and on the mode of development of the plu- and it has been followed by several entomologists mule and radicle. In our discussion on embryology, and botanists. we shall see why such characters are so valuable, on Finally, with respect to the comparative value of the the view of classification tacitly including the idea various groups of species, such as orders, sub-orders, of descent. families, sub-families, and genera, they seem to be, at Our classifications are often plainly influenced by least at present, almost arbitrary. Several of the best chains of affinities. Nothing can be easier than to botanists, such as Mr. Bentham and others, have define a number of characters common to all birds; strongly insisted on their arbitrary value. Instances but in the case of crustaceans, such definition has could be given amongst plants and insects, of a group hitherto been found impossible. There are crusta- of forms, first ranked by practised naturalists as only ceans at the opposite ends of the series, which have a genus, and then raised to the rank of a sub-family or hardly a character in common; yet the species at both family; and this has been done, not because further ends, from being plainly allied to others, and these research has detected important structural differences, to others, and so onwards, can be recognised as un- at first overlooked, but because numerous allied spe- equivocally belonging to this, and to no other class cies, with slightly different grades of difference, have of the Articulata. been subsequently discovered. Geographical distribution has often been used, All the foregoing rules and aids and difficulties in though perhaps not quite logically, in classification, classification are explained, if I do not greatly de-

372 Charles Darwin ceive myself, on the view that the natural system is ferent genera, families, sections, or orders. The reader founded on descent with modification; that the char- will best understand what is meant, if he will take acters which naturalists consider as showing true the trouble of referring to the diagram in the fourth affinity between any two or more species, are those chapter. We will suppose the letters A to L to repre- which have been inherited from a common parent, sent allied genera, which lived during the Silurian and, in so far, all true classification is genealogical; epoch, and these have descended from a species that community of descent is the hidden bond which which existed at an unknown anterior period. Spe- naturalists have been unconsciously seeking, and not cies of three of these genera (A, F, and I) have trans- some unknown plan of creation, or the enunciation mitted modified descendants to the present day, rep- of general propositions, and the mere putting to- resented by the fifteen genera (a14 to z14) on the gether and separating objects more or less alike. uppermost horizontal line. Now all these modified But I must explain my meaning more fully. I be- descendants from a single species, are represented lieve that the arrangement of the groups within each as related in blood or descent to the same degree; class, in due subordination and relation to the other they may metaphorically be called cousins to the groups, must be strictly genealogical in order to be same millionth degree; yet they differ widely and in natural; but that the amount of difference in the sev- different degrees from each other. The forms de- eral branches or groups, though allied in the same scended from A, now broken up into two or three degree in blood to their common progenitor, may families, constitute a distinct order from those de- differ greatly, being due to the different degrees of scended from I, also broken up into two families. modification which they have undergone; and this Nor can the existing species, descended from A, be is expressed by the forms being ranked under dif- ranked in the same genus with the parent A; or those

373 On the Origin of Species from I, with the parent I. But the existing genus F14 dants of the genus F, along its whole line of descent, may be supposed to have been but slightly modi- are supposed to have been but little modified, and fied; and it will then rank with the parent-genus F; they yet form a single genus. But this genus, though just as some few still living organic beings belong to much isolated, will still occupy its proper interme- Silurian genera. So that the amount or value of the diate position; for F originally was intermediate in differences between organic beings all related to each character between A and I, and the several genera other in the same degree in blood, has come to be descended from these two genera will have inher- widely different. Nevertheless their genealogical ited to a certain extent their characters. This natural arrangement remains strictly true, not only at the arrangement is shown, as far as is possible on pa- present time, but at each successive period of de- per, in the diagram, but in much too simple a man- scent. All the modified descendants from A will have ner. If a branching diagram had not been used, and inherited something in common from their common only the names of the groups had been written in a parent, as will all the descendants from I; so will it linear series, it would have been still less possible be with each subordinate branch of descendants, at to have given a natural arrangement; and it is noto- each successive period. If, however, we choose to riously not possible to represent in a series, on a flat suppose that any of the descendants of A or of I have surface, the affinities which we discover in nature been so much modified as to have more or less com- amongst the beings of the same group. Thus, on the pletely lost traces of their parentage, in this case, their view which I hold, the natural system is genealogi- places in a natural classification will have been more cal in its arrangement, like a pedigree; but the de- or less completely lost,—as sometimes seems to have grees of modification which the different groups occurred with existing organisms. All the descen- have undergone, have to be expressed by ranking

374 Charles Darwin them under different so-called genera, sub-families, cal; and this would be strictly natural, as it would families, sections, orders, and classes. connect together all languages, extinct and modern, It may be worth while to illustrate this view of clas- by the closest affinities, and would give the filiation sification, by taking the case of languages. If we pos- and origin of each tongue. sessed a perfect pedigree of mankind, a genealogi- In confirmation of this view, let us glance at the cal arrangement of the races of man would afford classification of varieties, which are believed or the best classification of the various languages now known to have descended from one species. These spoken throughout the world; and if all extinct lan- are grouped under species, with sub-varieties un- guages, and all intermediate and slowly changing der varieties; and with our domestic productions, dialects, had to be included, such an arrangement several other grades of difference are requisite, as would, I think, be the only possible one. Yet it might we have seen with pigeons. The origin of the exist- be that some very ancient language had altered little, ence of groups subordinate to groups, is the same and had given rise to few new languages, whilst oth- with varieties as with species, namely, closeness of ers (owing to the spreading and subsequent isola- descent with various degrees of modification. Nearly tion and states of civilisation of the several races, de- the same rules are followed in classifying varieties, scended from a common race) had altered much, and as with species. Authors have insisted on the neces- had given rise to many new languages and dialects. sity of classing varieties on a natural instead of an The various degrees of difference in the languages artificial system; we are cautioned, for instance, not from the same stock, would have to be expressed by to class two varieties of the pine-apple together, groups subordinate to groups; but the proper or even merely because their fruit, though the most impor- only possible arrangement would still be genealogi- tant part, happens to be nearly identical; no one puts

375 On the Origin of Species the swedish and common turnips together, though the subject, these tumblers are kept in the same the esculent and thickened stems are so similar. group, because allied in blood and alike in some Whatever part is found to be most constant, is used other respects. If it could be proved that the Hottentot in classing varieties: thus the great agriculturist had descended from the Negro, I think he would be Marshall says the horns are very useful for this pur- classed under the Negro group, however much he pose with cattle, because they are less variable than might differ in colour and other important charac- the shape or colour of the body, &c.; whereas with ters from negroes. sheep the horns are much less serviceable, because With species in a state of nature, every naturalist less constant. In classing varieties, I apprehend if we has in fact brought descent into his classification; for had a real pedigree, a genealogical classification he includes in his lowest grade, or that of a species, would be universally preferred; and it has been at- the two sexes; and how enormously these sometimes tempted by some authors. For we might feel sure, differ in the most important characters, is known to whether there had been more or less modification, every naturalist: scarcely a single fact can be predi- the principle of inheritance would keep the forms cated in common of the males and hermaphrodites together which were allied in the greatest number of of certain cirripedes, when adult, and yet no one points. In tumbler pigeons, though some sub-vari- dreams of separating them. The naturalist includes eties differ from the others in the important charac- as one species the several larval stages of the same ter of having a longer beak, yet all are kept together individual, however much they may differ from each from having the common habit of tumbling; but the other and from the adult; as he likewise includes the short-faced breed has nearly or quite lost this habit; so-called alternate generations of Steenstrup, which nevertheless, without any reasoning or thinking on can only in a technical sense be considered as the

376 Charles Darwin same individual. He includes monsters; he includes but then assuredly all the other species of the kan- varieties, not solely because they closely resemble garoo family would have to be classed under the the parent-form, but because they are descended bear genus. The whole case is preposterous; for from it. He who believes that the cowslip is de- where there has been close descent in common, there scended from the primrose, or conversely, ranks will certainly be close resemblance or affinity. them together as a single species, and gives a single As descent has universally been used in classing definition. As soon as three Orchidean forms together the individuals of the same species, though (Monochanthus, Myanthus, and Catasetum), which the males and females and larvae are sometimes ex- had previously been ranked as three distinct gen- tremely different; and as it has been used in classing era, were known to be sometimes produced on the varieties which have undergone a certain, and some- same spike, they were immediately included as a times a considerable amount of modification, may single species. But it may be asked, what ought we not this same element of descent have been uncon- to do, if it could be proved that one species of kan- sciously used in grouping species under genera, and garoo had been produced, by a long course of modi- genera under higher groups, though in these cases fication, from a bear? Ought we to rank this one spe- the modification has been greater in degree, and has cies with bears, and what should we do with the other taken a longer time to complete? I believe it has thus species? The supposition is of course preposterous; been unconsciously used; and only thus can I un- and I might answer by the argumentum ad hominem, derstand the several rules and guides which have and ask what should be done if a perfect kangaroo been followed by our best systematists. We have no were seen to come out of the womb of a bear? Ac- written pedigrees; we have to make out community cording to all analogy, it would be ranked with bears; of descent by resemblances of any kind. Therefore

377 On the Origin of Species we choose those characters which, as far as we can that such correlated or aggregated characters have judge, are the least likely to have been modified in especial value in classification. relation to the conditions of life to which each spe- We can understand why a species or a group of species has been recently exposed. Rudimentary struc- cies may depart, in several of its most important char- tures on this view are as good as, or even sometimes acteristics, from its allies, and yet be safely classed better than, other parts of the organisation. We care with them. This may be safely done, and is often done, not how trifling a character may be—let it be the mere as long as a sufficient number of characters, let them inflection of the angle of the jaw, the manner in which be ever so unimportant, betrays the hidden bond of an insect’s wing is folded, whether the skin be cov- community of descent. Let two forms have not a single ered by hair or feathers—if it prevail throughout character in common, yet if these extreme forms are many and different species, especially those having connected together by a chain of intermediate groups, very different habits of life, it assumes high value; we may at once infer their community of descent, and for we can account for its presence in so many forms we put them all into the same class. As we find or- with such different habits, only by its inheritance gans of high physiological importance—those which from a common parent. We may err in this respect in serve to preserve life under the most diverse condi- regard to single points of structure, but when sev- tions of existence—are generally the most constant, eral characters, let them be ever so trifling, occur to- we attach especial value to them; but if these same gether throughout a large group of beings having organs, in another group or section of a group, are different habits, we may feel almost sure, on the found to differ much, we at once value them less in theory of descent, that these characters have been our classification. We shall hereafter, I think, clearly inherited from a common ancestor. And we know see why embryological characters are of such high

378 Charles Darwin classificatory importance. Geographical distribution greyhound and racehorse is hardly more fanciful may sometimes be brought usefully into play in class- than the analogies which have been drawn by some ing large and widely-distributed genera, because all authors between very distinct animals. On my view the species of the same genus, inhabiting any distinct of characters being of real importance for classifica- and isolated region, have in all probability descended tion, only in so far as they reveal descent, we can from the same parents. clearly understand why analogical or adaptive char- We can understand, on these views, the very im- acter, although of the utmost importance to the wel- portant distinction between real affinities and ana- fare of the being, are almost valueless to the system- logical or adaptive resemblances. Lamarck first atist. For animals, belonging to two most distinct called attention to this distinction, and he has been lines of descent, may readily become adapted to ably followed by Macleay and others. The resem- similar conditions, and thus assume a close external blance, in the shape of the body and in the fin-like resemblance; but such resemblances will not re- anterior limbs, between the dugong, which is a veal—will rather tend to conceal their blood-relation- pachydermatous animal, and the whale, and be- ship to their proper lines of descent. We can also tween both these mammals and fishes, is analogical. understand the apparent paradox, that the very same Amongst insects there are innumerable instances: characters are analogical when one class or order is thus Linnaeus, misled by external appearances, ac- compared with another, but give true affinities when tually classed an homopterous insect as a moth. We the members of the same class or order are compared see something of the same kind even in our domes- one with another: thus the shape of the body and tic varieties, as in the thickened stems of the com- fin-like limbs are only analogical when whales are mon and swedish turnip. The resemblance of the compared with fishes, being adaptations in both

379 On the Origin of Species classes for swimming through the water; but the and ternary classifications have probably arisen. shape of the body and fin-like limbs serve as charac- As the modified descendants of dominant species, ters exhibiting true affinity between the several mem- belonging to the larger genera, tend to inherit the bers of the whale family; for these cetaceans agree in advantages, which made the groups to which they so many characters, great and small, that we cannot belong large and their parents dominant, they are doubt that they have inherited their general shape almost sure to spread widely, and to seize on more of body and structure of limbs from a common an- and more places in the economy of nature. The larger cestor. So it is with fishes. and more dominant groups thus tend to go on in- As members of distinct classes have often been creasing in size; and they consequently supplant adapted by successive slight modifications to live many smaller and feebler groups. Thus we can ac- under nearly similar circumstances,—to inhabit for count for the fact that all organisms, recent and ex- instance the three elements of land, air, and water,— tinct, are included under a few great orders, under we can perhaps understand how it is that a numeri- still fewer classes, and all in one great natural sys- cal parallelism has sometimes been observed be- tem. As showing how few the higher groups are in tween the sub-groups in distinct classes. A natural- number, and how widely spread they are through- ist, struck by a parallelism of this nature in any one out the world, the fact is striking, that the discovery class, by arbitrarily raising or sinking the value of of Australia has not added a single insect belonging the groups in other classes (and all our experience to a new order; and that in the vegetable kingdom, shows that this valuation has hitherto been arbitrary), as I learn from Dr. Hooker, it has added only two or could easily extend the parallelism over a wide three orders of small size. range; and thus the septenary, quinary, quaternary, In the chapter on geological succession I attempted

380 Charles Darwin to show, on the principle of each group having gen- investigation, does not commonly fall to the lot of erally diverged much in character during the long- aberrant genera. We can, I think, account for this fact continued process of modification, how it is that the only by looking at aberrant forms as failing groups more ancient forms of life often present characters in conquered by more successful competitors, with a some slight degree intermediate between existing few members preserved by some unusual coinci- groups. A few old and intermediate parent-forms dence of favourable circumstances. having occasionally transmitted to the present day Mr. Waterhouse has remarked that, when a mem- descendants but little modified, will give to us our ber belonging to one group of animals exhibits an so-called osculant or aberrant groups. The more ab- affinity to a quite distinct group, this affinity in most errant any form is, the greater must be the number cases is general and not special: thus, according to of connecting forms which on my theory have been Mr. Waterhouse, of all Rodents, the bizcacha is most exterminated and utterly lost. And we have some nearly related to Marsupials; but in the points in evidence of aberrant forms having suffered severely which it approaches this order, its relations are gen- from extinction, for they are generally represented eral, and not to any one marsupial species more than by extremely few species; and such species as do to another. As the points of affinity of the bizcacha to occur are generally very distinct from each other, Marsupials are believed to be real and not merely which again implies extinction. The genera adaptive, they are due on my theory to inheritance Ornithorhynchus and Lepidosiren, for example, in common. Therefore we must suppose either that would not have been less aberrant had each been all Rodents, including the bizcacha, branched off represented by a dozen species instead of by a single from some very ancient Marsupial, which will have one; but such richness in species, as I find after some had a character in some degree intermediate with

381 On the Origin of Species respect to all existing Marsupials; or that both Ro- On the principle of the multiplication and gradual dents and Marsupials branched off from a common divergence in character of the species descended progenitor, and that both groups have since under- from a common parent, together with their retention gone much modification in divergent directions. On by inheritance of some characters in common, we can either view we may suppose that the bizcacha has understand the excessively complex and radiating retained, by inheritance, more of the character of its affinities by which all the members of the same fam- ancient progenitor than have other Rodents; and ily or higher group are connected together. For the therefore it will not be specially related to any one common parent of a whole family of species, now existing Marsupial, but indirectly to all or nearly all broken up by extinction into distinct groups and sub- Marsupials, from having partially retained the char- groups, will have transmitted some of its characters, acter of their common progenitor, or of an early mem- modified in various ways and degrees, to all; and ber of the group. On the other hand, of all Marsupi- the several species will consequently be related to als, as Mr. Waterhouse has remarked, the each other by circuitous lines of affinity of various phascolomys resembles most nearly, not any one lengths (as may be seen in the diagram so often re- species, but the general order of Rodents. In this case, ferred to), mounting up through many predecessors. however, it may be strongly suspected that the re- As it is difficult to show the blood-relationship be- semblance is only analogical, owing to the tween the numerous kindred of any ancient and phascolomys having become adapted to habits like noble family, even by the aid of a genealogical tree, those of a Rodent. The elder De Candolle has made and almost impossible to do this without this aid, nearly similar observations on the general nature of we can understand the extraordinary difficulty which the affinities of distinct orders of plants. naturalists have experienced in describing, without

382 Charles Darwin the aid of a diagram, the various affinities which they denly to reappear, though it would be quite impos- perceive between the many living and extinct mem- sible to give definitions by which each group could bers of the same great natural class. be distinguished from other groups, as all would Extinction, as we have seen in the fourth chapter, blend together by steps as fine as those between the has played an important part in defining and wid- finest existing varieties, nevertheless a natural clas- ening the intervals between the several groups in sification, or at least a natural arrangement, would each class. We may thus account even for the dis- be possible. We shall see this by turning to the dia- tinctness of whole classes from each other—for in- gram: the letters, A to L, may represent eleven Sil- stance, of birds from all other vertebrate animals— urian genera, some of which have produced large by the belief that many ancient forms of life have groups of modified descendants. Every intermedi- been utterly lost, through which the early progeni- ate link between these eleven genera and their pri- tors of birds were formerly connected with the early mordial parent, and every intermediate link in each progenitors of the other vertebrate classes. There has branch and sub-branch of their descendants, may be been less entire extinction of the forms of life which supposed to be still alive; and the links to be as fine once connected fishes with batrachians. There has as those between the finest varieties. In this case it been still less in some other classes, as in that of the would be quite impossible to give any definition by Crustacea, for here the most wonderfully diverse which the several members of the several groups forms are still tied together by a long, but broken, could be distinguished from their more immediate chain of affinities. Extinction has only separated parents; or these parents from their ancient and un- groups: it has by no means made them; for if every known progenitor. Yet the natural arrangement in form which has ever lived on this earth were sud- the diagram would still hold good; and, on the prin-

383 On the Origin of Species ciple of inheritance, all the forms descended from A, of character in the many descendants from one domi- or from I, would have something in common. In a nant parent-species, explains that great and univer- tree we can specify this or that branch, though at the sal feature in the affinities of all organic beings, actual fork the two unite and blend together. We namely, their subordination in group under group. could not, as I have said, define the several groups; We use the element of descent in classing the indi- but we could pick out types, or forms, representing viduals of both sexes and of all ages, although hav- most of the characters of each group, whether large ing few characters in common, under one species; or small, and thus give a general idea of the value of we use descent in classing acknowledged varieties, the differences between them. This is what we should however different they may be from their parent; and be driven to, if we were ever to succeed in collecting I believe this element of descent is the hidden bond all the forms in any class which have lived through- of connexion which naturalists have sought under out all time and space. We shall certainly never suc- the term of the Natural System. On this idea of the ceed in making so perfect a collection: nevertheless, natural system being, in so far as it has been per- in certain classes, we are tending in this direction; fected, genealogical in its arrangement, with the and Milne Edwards has lately insisted, in an able grades of difference between the descendants from paper, on the high importance of looking to types, a common parent, expressed by the terms genera, whether or not we can separate and define the groups families, orders, &c., we can understand the rules to which such types belong. which we are compelled to follow in our classifica- Finally, we have seen that natural selection, which tion. We can understand why we value certain re- results from the struggle for existence, and which semblances far more than others; why we are per- almost inevitably induces extinction and divergence mitted to use rudimentary and useless organs, or

384 Charles Darwin others of trifling physiological importance; why, in are homologous. The whole subject is included un- comparing one group with a distinct group, we sum- der the general name of Morphology. This is the most marily reject analogical or adaptive characters, and interesting department of natural history, and may yet use these same characters within the limits of the be said to be its very soul. What can be more curi- same group. We can clearly see how it is that all liv- ous than that the hand of a man, formed for grasping and extinct forms can be grouped together in one ing, that of a mole for digging, the leg of the horse, great system; and how the several members of each the paddle of the porpoise, and the wing of the bat, class are connected together by the most complex should all be constructed on the same pattern, and and radiating lines of affinities. We shall never, prob- should include the same bones, in the same relative ably, disentangle the inextricable web of affinities positions? Geoffroy St. Hilaire has insisted strongly between the members of any one class; but when we on the high importance of relative connexion in ho- have a distinct object in view, and do not look to mologous organs: the parts may change to almost some unknown plan of creation, we may hope to any extent in form and size, and yet they always re- make sure but slow progress. main connected together in the same order. We never find, for instance, the bones of the arm and forearm, Morphology. — We have seen that the members of or of the thigh and leg, transposed. Hence the same the same class, independently of their habits of life, names can be given to the homologous bones in resemble each other in the general plan of their widely different animals. We see the same great law organisation. This resemblance is often expressed by in the construction of the mouths of insects: what can the term ‘unity of type;’ or by saying that the several be more different than the immensely long spiral parts and organs in the different species of the class proboscis of a sphinx-moth, the curious folded one

385 On the Origin of Species of a bee or bug, and the great jaws of a beetle?—yet changes of this nature, there will be little or no ten- all these organs, serving for such different purposes, dency to modify the original pattern, or to transpose are formed by infinitely numerous modifications of parts. The bones of a limb might be shortened and an upper lip, mandibles, and two pairs of maxillae. widened to any extent, and become gradually en- Analogous laws govern the construction of the veloped in thick membrane, so as to serve as a fin; mouths and limbs of crustaceans. So it is with the or a webbed foot might have all its bones, or certain flowers of plants. bones, lengthened to any extent, and the membrane Nothing can be more hopeless than to attempt to connecting them increased to any extent, so as to explain this similarity of pattern in members of the serve as a wing: yet in all this great amount of modi- same class, by utility or by the doctrine of final fication there will be no tendency to alter the frame- causes. The hopelessness of the attempt has been work of bones or the relative connexion of the sev- expressly admitted by Owen in his most interesting eral parts. If we suppose that the ancient progenitor, work on the ‘Nature of Limbs.’ On the ordinary view the archetype as it may be called, of all mammals, of the independent creation of each being, we can had its limbs constructed on the existing general only say that so it is;—that it has so pleased the Cre- pattern, for whatever purpose they served, we can ator to construct each animal and plant. at once perceive the plain signification of the homolo- The explanation is manifest on the theory of the gous construction of the limbs throughout the whole natural selection of successive slight modifica- class. So with the mouths of insects, we have only to tions,—each modification being profitable in some suppose that their common progenitor had an up- way to the modified form, but often affecting by cor- per lip, mandibles, and two pair of maxillae, these relation of growth other parts of the organisation. In parts being perhaps very simple in form; and then

386 Charles Darwin natural selection will account for the infinite diver- ticulate classes are plainly homologous. We see the sity in structure and function of the mouths of insects. same law in comparing the wonderfully complex Nevertheless, it is conceivable that the general pat- jaws and legs in crustaceans. It is familiar to almost tern of an organ might become so much obscured as every one, that in a flower the relative position of to be finally lost, by the atrophy and ultimately by the sepals, petals, stamens, and pistils, as well as the complete abortion of certain parts, by the solder- their intimate structure, are intelligible on the view ing together of other parts, and by the doubling or that they consist of metamorphosed leaves, arranged multiplication of others,—variations which we know in a spire. In monstrous plants, we often get direct to be within the limits of possibility. In the paddles of evidence of the possibility of one organ being trans- the extinct gigantic sea-lizards, and in the mouths of formed into another; and we can actually see in em- certain suctorial crustaceans, the general pattern seems bryonic crustaceans and in many other animals, and to have been thus to a certain extent obscured. in flowers, that organs, which when mature become There is another and equally curious branch of the extremely different, are at an early stage of growth present subject; namely, the comparison not of the exactly alike. same part in different members of a class, but of the How inexplicable are these facts on the ordinary different parts or organs in the same individual. Most view of creation! Why should the brain be enclosed physiologists believe that the bones of the skull are in a box composed of such numerous and such ex- homologous with—that is correspond in number and traordinarily shaped pieces of bone? As Owen has in relative connexion with—the elemental parts of a remarked, the benefit derived from the yielding of certain number of vertebrae. The anterior and poste- the separate pieces in the act of parturition of mam- rior limbs in each member of the vertebrate and ar- mals, will by no means explain the same construc-

387 On the Origin of Species tion in the skulls of birds. Why should similar bones believe that the unknown progenitor of the vertebrata have been created in the formation of the wing and possessed many vertebrae; the unknown progenitor leg of a bat, used as they are for such totally differ- of the articulata, many segments; and the unknown ent purposes? Why should one crustacean, which progenitor of flowering plants, many spiral whorls of has an extremely complex mouth formed of many leaves. We have formerly seen that parts many times parts, consequently always have fewer legs; or con- repeated are eminently liable to vary in number and versely, those with many legs have simpler mouths? structure; consequently it is quite probable that natu- Why should the sepals, petals, stamens, and pistils ral selection, during a long-continued course of modi- in any individual flower, though fitted for such fication, should have seized on a certain number of widely different purposes, be all constructed on the the primordially similar elements, many times re- same pattern? peated, and have adapted them to the most diverse On the theory of natural selection, we can satisfac- purposes. And as the whole amount of modification torily answer these questions. In the vertebrata, we will have been effected by slight successive steps, we see a series of internal vertebrae bearing certain pro- need not wonder at discovering in such parts or or- cesses and appendages; in the articulata, we see the gans, a certain degree of fundamental resemblance, body divided into a series of segments, bearing ex- retained by the strong principle of inheritance. ternal appendages; and in flowering plants, we see a In the great class of molluscs, though we can series of successive spiral whorls of leaves. An indefi- homologise the parts of one species with those of nite repetition of the same part or organ is the com- another and distinct species, we can indicate but few mon characteristic (as Owen has observed) of all low serial homologies; that is, we are seldom enabled to or little-modified forms; therefore we may readily say that one part or organ is homologous with an-

388 Charles Darwin other in the same individual. And we can understand guage having this plain signification. On my view this fact; for in molluscs, even in the lowest members these terms may be used literally; and the wonder- of the class, we do not find nearly so much indefinite ful fact of the jaws, for instance, of a crab retaining repetition of any one part, as we find in the other great numerous characters, which they would probably classes of the animal and vegetable kingdoms. have retained through inheritance, if they had really Naturalists frequently speak of the skull as formed been metamorphosed during a long course of de- of metamorphosed vertebrae: the jaws of crabs as scent from true legs, or from some simple append- metamorphosed legs; the stamens and pistils of flow- age, is explained. ers as metamorphosed leaves; but it would in these cases probably be more correct, as Professor Huxley Embryology. — It has already been casually re- has remarked, to speak of both skull and vertebrae, marked that certain organs in the individual, which both jaws and legs, &c.,—as having been metamor- when mature become widely different and serve for phosed, not one from the other, but from some com- different purposes, are in the embryo exactly alike. mon element. Naturalists, however, use such lan- The embryos, also, of distinct animals within the guage only in a metaphorical sense: they are far from same class are often strikingly similar: a better proof meaning that during a long course of descent, pri- of this cannot be given, than a circumstance men- mordial organs of any kind—vertebrae in the one tioned by Agassiz, namely, that having forgotten to case and legs in the other—have actually been modi- ticket the embryo of some vertebrate animal, he can- fied into skulls or jaws. Yet so strong is the appear- not now tell whether it be that of a mammal, bird, or ance of a modification of this nature having occurred, reptile. The vermiform larvae of moths, flies, beetles, that naturalists can hardly avoid employing lan- &c., resemble each other much more closely than do

389 On the Origin of Species the mature insects; but in the case of larvae, the em- slits are related to similar conditions,—in the young bryos are active, and have been adapted for special mammal which is nourished in the womb of its lines of life. A trace of the law of embryonic resem- mother, in the egg of the bird which is hatched in a blance, sometimes lasts till a rather late age: thus nest, and in the spawn of a frog under water. We birds of the same genus, and of closely allied gen- have no more reason to believe in such a relation, era, often resemble each other in their first and sec- than we have to believe that the same bones in the ond plumage; as we see in the spotted feathers in hand of a man, wing of a bat, and fin of a porpoise, the thrush group. In the cat tribe, most of the species are related to similar conditions of life. No one will are striped or spotted in lines; and stripes can be suppose that the stripes on the whelp of a lion, or plainly distinguished in the whelp of the lion. We the spots on the young blackbird, are of any use to occasionally though rarely see something of this kind these animals, or are related to the conditions to in plants: thus the embryonic leaves of the ulex or which they are exposed. furze, and the first leaves of the phyllodineous The case, however, is different when an animal dur- acaceas, are pinnate or divided like the ordinary ing any part of its embryonic career is active, and leaves of the leguminosae. has to provide for itself. The period of activity may The points of structure, in which the embryos of come on earlier or later in life; but whenever it comes widely different animals of the same class resemble on, the adaptation of the larva to its conditions of each other, often have no direct relation to their con- life is just as perfect and as beautiful as in the adult ditions of existence. We cannot, for instance, sup- animal. From such special adaptations, the similar- pose that in the embryos of the vertebrata the pecu- ity of the larvae or active embryos of allied animals liar loop-like course of the arteries near the branchial is sometimes much obscured; and cases could be

390 Charles Darwin given of the larvae of two species, or of two groups as with certain parasitic crustaceans. To refer once of species, differing quite as much, or even more, again to cirripedes: the larvae in the first stage have from each other than do their adult parents. In most three pairs of legs, a very simple single eye, and a cases, however, the larvae, though active, still obey probosciformed mouth, with which they feed largely, more or less closely the law of common embryonic for they increase much in size. In the second stage, resemblance. Cirripedes afford a good instance of answering to the chrysalis stage of butterflies, they this: even the illustrious Cuvier did not perceive that have six pairs of beautifully constructed natatory a barnacle was, as it certainly is, a crustacean; but a legs, a pair of magnificent compound eyes, and ex- glance at the larva shows this to be the case in an tremely complex antennae; but they have a closed unmistakeable manner. So again the two main divi- and imperfect mouth, and cannot feed: their func- sions of cirripedes, the pedunculated and sessile, tion at this stage is, to search by their well-devel- which differ widely in external appearance, have lar- oped organs of sense, and to reach by their active vae in all their several stages barely distinguishable. powers of swimming, a proper place on which to The embryo in the course of development gener- become attached and to undergo their final meta- ally rises in organisation: I use this expression, morphosis. When this is completed they are fixed though I am aware that it is hardly possible to de- for life: their legs are now converted into prehensile fine clearly what is meant by the organisation being organs; they again obtain a well-constructed mouth; higher or lower. But no one probably will dispute but they have no antennae, and their two eyes are that the butterfly is higher than the caterpillar. In now reconverted into a minute, single, and very some cases, however, the mature animal is gener- simple eye-spot. In this last and complete state, ally considered as lower in the scale than the larva, cirripedes may be considered as either more highly

391 On the Origin of Species or more lowly organised than they were in the lar- period differ widely from the adult: thus Owen has val condition. But in some genera the larvae become remarked in regard to cuttle-fish, ‘there is no meta- developed either into hermaphrodites having the morphosis; the cephalopodic character is manifested ordinary structure, or into what I have called comple- long before the parts of the embryo are completed;’ mental males: and in the latter, the development has and again in spiders, ‘there is nothing worthy to be assuredly been retrograde; for the male is a mere called a metamorphosis.’ The larvae of insects, sack, which lives for a short time, and is destitute of whether adapted to the most diverse and active hab- mouth, stomach, or other organ of importance, ex- its, or quite inactive, being fed by their parents or cepting for reproduction. placed in the midst of proper nutriment, yet nearly We are so much accustomed to see differences in all pass through a similar worm-like stage of devel- structure between the embryo and the adult, and opment; but in some few cases, as in that of Aphis, if likewise a close similarity in the embryos of widely we look to the admirable drawings by Professor different animals within the same class, that we might Huxley of the development of this insect, we see no be led to look at these facts as necessarily contingent trace of the vermiform stage. in some manner on growth. But there is no obvious How, then, can we explain these several facts in reason why, for instance, the wing of a bat, or the fin embryology,—namely the very general, but not uni- of a porpoise, should not have been sketched out versal difference in structure between the embryo with all the parts in proper proportion, as soon as and the adult;—of parts in the same individual em- any structure became visible in the embryo. And in bryo, which ultimately become very unlike and serve some whole groups of animals and in certain mem- for diverse purposes, being at this early period of bers of other groups, the embryo does not at any growth alike;—of embryos of different species within

392 Charles Darwin the same class, generally, but not universally, resem- what its precise features will be. The question is not, bling each other;—of the structure of the embryo not at what period of life any variation has been caused, being closely related to its conditions of existence, but at what period it is fully displayed. The cause except when the embryo becomes at any period of may have acted, and I believe generally has acted, life active and has to provide for itself;—of the em- even before the embryo is formed; and the variation bryo apparently having sometimes a higher may be due to the male and female sexual elements organisation than the mature animal, into which it is having been affected by the conditions to which ei- developed. I believe that all these facts can be ex- ther parent, or their ancestors, have been exposed. plained, as follows, on the view of descent with Nevertheless an effect thus caused at a very early modification. period, even before the formation of the embryo, may It is commonly assumed, perhaps from monstrosi- appear late in life; as when an hereditary disease, ties often affecting the embryo at a very early pe- which appears in old age alone, has been communi- riod, that slight variations necessarily appear at an cated to the offspring from the reproductive element equally early period. But we have little evidence on of one parent. Or again, as when the horns of cross- this head—indeed the evidence rather points the bred cattle have been affected by the shape of the other way; for it is notorious that breeders of cattle, horns of either parent. For the welfare of a very young horses, and various fancy animals, cannot positively animal, as long as it remains in its mother’s womb, tell, until some time after the animal has been born, or in the egg, or as long as it is nourished and pro- what its merits or form will ultimately turn out. We tected by its parent, it must be quite unimportant see this plainly in our own children; we cannot al- whether most of its characters are fully acquired a ways tell whether the child will be tall or short, or little earlier or later in life. It would not signify, for

393 On the Origin of Species instance, to a bird which obtained its food best by peared earlier or later in life, tend to appear at a cor- having a long beak, whether or not it assumed a beak responding age in the offspring and parent. I am far of this particular length, as long as it was fed by its from meaning that this is invariably the case; and I parents. Hence, I conclude, that it is quite possible, could give a good many cases of variations (taking that each of the many successive modifications, by the word in the largest sense) which have supervened which each species has acquired its present struc- at an earlier age in the child than in the parent. ture, may have supervened at a not very early pe- These two principles, if their truth be admitted, riod of life; and some direct evidence from our do- will, I believe, explain all the above specified lead- mestic animals supports this view. But in other cases ing facts in embryology. But first let us look at a few it is quite possible that each successive modification, analogous cases in domestic varieties. Some authors or most of them, may have appeared at an extremely who have written on Dogs, maintain that the grey- early period. hound and bulldog, though appearing so different, I have stated in the first chapter, that there is some are really varieties most closely allied, and have evidence to render it probable, that at whatever age probably descended from the same wild stock; hence any variation first appears in the parent, it tends to I was curious to see how far their puppies differed reappear at a corresponding age in the offspring. Cer- from each other: I was told by breeders that they dif- tain variations can only appear at corresponding ages, fered just as much as their parents, and this, judging for instance, peculiarities in the caterpillar, cocoon, by the eye, seemed almost to be the case; but on ac- or imago states of the silk-moth; or, again, in the horns tually measuring the old dogs and their six-days old of almost full-grown cattle. But further than this, varia- puppies, I found that the puppies had not nearly tions which, for all that we can see, might have ap- acquired their full amount of proportional difference.

394 Charles Darwin So, again, I was told that the foals of cart and race- had they been natural productions. But when the horses differed as much as the full-grown animals; nestling birds of these several breeds were placed and this surprised me greatly, as I think it probable in a row, though most of them could be distinguished that the difference between these two breeds has been from each other, yet their proportional differences wholly caused by selection under domestication; but in the above specified several points were incompa- having had careful measurements made of the dam rably less than in the full-grown birds. Some charac- and of a three-days old colt of a race and heavy cart- teristic points of difference—for instance, that of the horse, I find that the colts have by no means acquired width of mouth—could hardly be detected in the their full amount of proportional difference. young. But there was one remarkable exception to As the evidence appears to me conclusive, that the this rule, for the young of the short-faced tumbler several domestic breeds of Pigeon have descended differed from the young of the wild rock-pigeon and from one wild species, I compared young pigeons of the other breeds, in all its proportions, almost ex- of various breeds, within twelve hours after being actly as much as in the adult state. hatched; I carefully measured the proportions (but The two principles above given seem to me to ex- will not here give details) of the beak, width of plain these facts in regard to the later embryonic mouth, length of nostril and of eyelid, size of feet stages of our domestic varieties. Fanciers select their and length of leg, in the wild stock, in pouters, fan- horses, dogs, and pigeons, for breeding, when they tails, runts, barbs, dragons, carriers, and tumblers. are nearly grown up: they are indifferent whether Now some of these birds, when mature, differ so the desired qualities and structures have been ac- extraordinarily in length and form of beak, that they quired earlier or later in life, if the full-grown ani- would, I cannot doubt, be ranked in distinct genera, mal possesses them. And the cases just given, more

395 On the Origin of Species especially that of pigeons, seem to show that the char- slight successive steps of variation having super- acteristic differences which give value to each breed, vened at a rather late age, and having been inherited and which have been accumulated by man’s selec- at a corresponding age, the young of the new spe- tion, have not generally first appeared at an early cies of our supposed genus will manifestly tend to period of life, and have been inherited by the off- resemble each other much more closely than do the spring at a corresponding not early period. But the adults, just as we have seen in the case of pigeons. case of the short-faced tumbler, which when twelve We may extend this view to whole families or even hours old had acquired its proper proportions, classes. The fore-limbs, for instance, which served proves that this is not the universal rule; for here the as legs in the parent-species, may become, by a long characteristic differences must either have appeared course of modification, adapted in one descendant at an earlier period than usual, or, if not so, the dif- to act as hands, in another as paddles, in another as ferences must have been inherited, not at the corre- wings; and on the above two principles—namely of sponding, but at an earlier age. each successive modification supervening at a rather Now let us apply these facts and the above two late age, and being inherited at a corresponding late principles—which latter, though not proved true, can age—the fore-limbs in the embryos of the several be shown to be in some degree probable—to spe- descendants of the parent-species will still resemble cies in a state of nature. Let us take a genus of birds, each other closely, for they will not have been modi- descended on my theory from some one parent-spe- fied. But in each individual new species, the embry- cies, and of which the several new species have be- onic fore-limbs will differ greatly from the fore-limbs come modified through natural selection in accor- in the mature animal; the limbs in the latter having dance with their diverse habits. Then, from the many undergone much modification at a rather late period

396 Charles Darwin of life, and having thus been converted into hands, of the young in these cases not undergoing any meta- or paddles, or wings. Whatever influence long-con- morphosis, or closely resembling their parents from tinued exercise or use on the one hand, and disuse their earliest age, we can see that this would result on the other, may have in modifying an organ, such from the two following contingencies; firstly, from influence will mainly affect the mature animal, which the young, during a course of modification carried has come to its full powers of activity and has to gain on for many generations, having to provide for their its own living; and the effects thus produced will be own wants at a very early stage of development, and inherited at a corresponding mature age. Whereas secondly, from their following exactly the same hab- the young will remain unmodified, or be modified its of life with their parents; for in this case, it would in a lesser degree, by the effects of use and disuse. be indispensable for the existence of the species, that In certain cases the successive steps of variation the child should be modified at a very early age in might supervene, from causes of which we are the same manner with its parents, in accordance with wholly ignorant, at a very early period of life, or each their similar habits. Some further explanation, how- step might be inherited at an earlier period than that ever, of the embryo not undergoing any metamor- at which it first appeared. In either case (as with the phosis is perhaps requisite. If, on the other hand, it short-faced tumbler) the young or embryo would profited the young to follow habits of life in any de- closely resemble the mature parent-form. We have gree different from those of their parent, and conse- seen that this is the rule of development in certain quently to be constructed in a slightly different man- whole groups of animals, as with cuttle-fish and spi- ner, then, on the principle of inheritance at corre- ders, and with a few members of the great class of sponding ages, the active young or larvae might eas- insects, as with Aphis. With respect to the final cause ily be rendered by natural selection different to any

397 On the Origin of Species conceivable extent from their parents. Such differ- modified state; and in so far it reveals the structure ences might, also, become correlated with succes- of its progenitor. In two groups of animal, however sive stages of development; so that the larvae, in the much they may at present differ from each other in first stage, might differ greatly from the larvae in the structure and habits, if they pass through the same second stage, as we have seen to be the case with or similar embryonic stages, we may feel assured cirripedes. The adult might become fitted for sites that they have both descended from the same or or habits, in which organs of locomotion or of the nearly similar parents, and are therefore in that de- senses, &c., would be useless; and in this case the gree closely related. Thus, community in embryonic final metamorphosis would be said to be retrograde. structure reveals community of descent. It will re- As all the organic beings, extinct and recent, which veal this community of descent, however much the have ever lived on this earth have to be classed to- structure of the adult may have been modified and gether, and as all have been connected by the finest obscured; we have seen, for instance, that cirripedes gradations, the best, or indeed, if our collections were can at once be recognised by their larvae as belong- nearly perfect, the only possible arrangement, would ing to the great class of crustaceans. As the embry- be genealogical. Descent being on my view the hid- onic state of each species and group of species par- den bond of connexion which naturalists have been tially shows us the structure of their less modified seeking under the term of the natural system. On ancient progenitors, we can clearly see why ancient this view we can understand how it is that, in the and extinct forms of life should resemble the em- eyes of most naturalists, the structure of the embryo bryos of their descendants,—our existing species. is even more important for classification than that of Agassiz believes this to be a law of nature; but I am the adult. For the embryo is the animal in its less bound to confess that I only hope to see the law here-

398 Charles Darwin after proved true. It can be proved true in those cases in interest, when we thus look at the embryo as a alone in which the ancient state, now supposed to picture, more or less obscured, of the common par- be represented in many embryos, has not been oblit- ent-form of each great class of animals. erated, either by the successive variations in a long course of modification having supervened at a very Rudimentary, atrophied, or aborted organs. — Or- early age, or by the variations having been inherited gans or parts in this strange condition, bearing the at an earlier period than that at which they first ap- stamp of inutility, are extremely common through- peared. It should also be borne in mind, that the sup- out nature. For instance, rudimentary mammae are posed law of resemblance of ancient forms of life to very general in the males of mammals: I presume the embryonic stages of recent forms, may be true, that the ‘bastard-wing’ in birds may be safely con- but yet, owing to the geological record not extend- sidered as a digit in a rudimentary state: in very ing far enough back in time, may remain for a long many snakes one lobe of the lungs is rudimentary; period, or for ever, incapable of demonstration. in other snakes there are rudiments of the pelvis and Thus, as it seems to me, the leading facts in embry- hind limbs. Some of the cases of rudimentary organs ology, which are second in importance to none in are extremely curious; for instance, the presence of natural history, are explained on the principle of teeth in foetal whales, which when grown up have slight modifications not appearing, in the many de- not a tooth in their heads; and the presence of teeth, scendants from some one ancient progenitor, at a which never cut through the gums, in the upper jaws very early period in the life of each, though perhaps of our unborn calves. It has even been stated on good caused at the earliest, and being inherited at a corre- authority that rudiments of teeth can be detected in sponding not early period. Embryology rises greatly the beaks of certain embryonic birds. Nothing can

399 On the Origin of Species be plainer than that wings are formed for flight, yet times occur as mere rudiments, and sometimes in a in how many insects do we see wings so reduced in well-developed state. In plants with separated sexes, size as to be utterly incapable of flight, and not rarely the male flowers often have a rudiment of a pistil; lying under wing-cases, firmly soldered together! and Kolreuter found that by crossing such male The meaning of rudimentary organs is often quite plants with an hermaphrodite species, the rudiment unmistakeable: for instance there are beetles of the of the pistil in the hybrid offspring was much in- same genus (and even of the same species) resem- creased in size; and this shows that the rudiment and bling each other most closely in all respects, one of the perfect pistil are essentially alike in nature. which will have full-sized wings, and another mere An organ serving for two purposes, may become rudiments of membrane; and here it is impossible rudimentary or utterly aborted for one, even the to doubt, that the rudiments represent wings. Rudi- more important purpose; and remain perfectly effi- mentary organs sometimes retain their potentiality, cient for the other. Thus in plants, the office of the and are merely not developed: this seems to be the pistil is to allow the pollen-tubes to reach the ovules case with the mammae of male mammals, for many protected in the ovarium at its base. The pistil con- instances are on record of these organs having be- sists of a stigma supported on the style; but in some come well developed in full-grown males, and hav- Compositae, the male florets, which of course can- ing secreted milk. So again there are normally four not be fecundated, have a pistil, which is in a rudi- developed and two rudimentary teats in the udders mentary state, for it is not crowned with a stigma; of the genus Bos, but in our domestic cows the two but the style remains well developed, and is clothed sometimes become developed and give milk. In in- with hairs as in other compositae, for the purpose of dividual plants of the same species the petals some- brushing the pollen out of the surrounding anthers.

400 Charles Darwin Again, an organ may become rudimentary for its members of a class, nothing is more common, or proper purpose, and be used for a distinct object: in more necessary, than the use and discovery of rudi- certain fish the swim-bladder seems to be rudimen- ments. This is well shown in the drawings given by tary for its proper function of giving buoyancy, but Owen of the bones of the leg of the horse, ox, and has become converted into a nascent breathing or- rhinoceros. gan or lung. Other similar instances could be given. It is an important fact that rudimentary organs, such Rudimentary organs in the individuals of the same as teeth in the upper jaws of whales and ruminants, species are very liable to vary in degree of develop- can often be detected in the embryo, but afterwards ment and in other respects. Moreover, in closely al- wholly disappear. It is also, I believe, a universal lied species, the degree to which the same organ has rule, that a rudimentary part or organ is of greater been rendered rudimentary occasionally differs size relatively to the adjoining parts in the embryo, much. This latter fact is well exemplified in the state than in the adult; so that the organ at this early age is of the wings of the female moths in certain groups. less rudimentary, or even cannot be said to be in any Rudimentary organs may be utterly aborted; and this degree rudimentary. Hence, also, a rudimentary or- implies, that we find in an animal or plant no trace gan in the adult, is often said to have retained its of an organ, which analogy would lead us to expect embryonic condition. to find, and which is occasionally found in monstrous I have now given the leading facts with respect to individuals of the species. Thus in the snapdragon rudimentary organs. In reflecting on them, every one (antirrhinum) we generally do not find a rudiment must be struck with astonishment: for the same rea- of a fifth stamen; but this may sometimes be seen. In soning power which tells us plainly that most parts tracing the homologies of the same part in different and organs are exquisitely adapted for certain pur-

401 On the Origin of Species poses, tells us with equal plainness that these rudi- man’s fingers have been amputated, imperfect nails mentary or atrophied organs, are imperfect and use- sometimes appear on the stumps: I could as soon less. In works on natural history rudimentary organs believe that these vestiges of nails have appeared, not are generally said to have been created ‘for the sake from unknown laws of growth, but in order to excrete of symmetry,’ or in order ‘to complete the scheme of horny matter, as that the rudimentary nails on the fin nature;’ but this seems to me no explanation, merely of the manatee were formed for this purpose. a restatement of the fact. Would it be thought suffi- On my view of descent with modification, the ori- cient to say that because planets revolve in elliptic gin of rudimentary organs is simple. We have plenty courses round the sun, satellites follow the same of cases of rudimentary organs in our domestic pro- course round the planets, for the sake of symmetry, ductions,—as the stump of a tail in tailless breeds,— and to complete the scheme of nature? An eminent the vestige of an ear in earless breeds,—the reap- physiologist accounts for the presence of rudimen- pearance of minute dangling horns in hornless tary organs, by supposing that they serve to excrete breeds of cattle, more especially, according to Youatt, matter in excess, or injurious to the system; but can in young animals,—and the state of the whole flower we suppose that the minute papilla, which often rep- in the cauliflower. We often see rudiments of vari- resents the pistil in male flowers, and which is ous parts in monsters. But I doubt whether any of formed merely of cellular tissue, can thus act? Can these cases throw light on the origin of rudimentary we suppose that the formation of rudimentary teeth organs in a state of nature, further than by showing which are subsequently absorbed, can be of any ser- that rudiments can be produced; for I doubt whether vice to the rapidly growing embryonic calf by the species under nature ever undergo abrupt changes. excretion of precious phosphate of lime? When a I believe that disuse has been the main agency; that

402 Charles Darwin it has led in successive generations to the gradual ever period of life disuse or selection reduces an reduction of various organs, until they have become organ, and this will generally be when the being has rudimentary,—as in the case of the eyes of animals come to maturity and to its full powers of action, the inhabiting dark caverns, and of the wings of birds principle of inheritance at corresponding ages will inhabiting oceanic islands, which have seldom been reproduce the organ in its reduced state at the same forced to take flight, and have ultimately lost the age, and consequently will seldom affect or reduce power of flying. Again, an organ useful under cer- it in the embryo. Thus we can understand the greater tain conditions, might become injurious under oth- relative size of rudimentary organs in the embryo, ers, as with the wings of beetles living on small and and their lesser relative size in the adult. But if each exposed islands; and in this case natural selection step of the process of reduction were to be inher- would continue slowly to reduce the organ, until it ited, not at the corresponding age, but at an extremely was rendered harmless and rudimentary. early period of life (as we have good reason to be- Any change in function, which can be effected by lieve to be possible) the rudimentary part would insensibly small steps, is within the power of natu- tend to be wholly lost, and we should have a case of ral selection; so that an organ rendered, during complete abortion. The principle, also, of economy, changed habits of life, useless or injurious for one explained in a former chapter, by which the materi- purpose, might easily be modified and used for an- als forming any part or structure, if not useful to the other purpose. Or an organ might be retained for possessor, will be saved as far as is possible, will one alone of its former functions. An organ, when probably often come into play; and this will tend to rendered useless, may well be variable, for its varia- cause the entire obliteration of a rudimentary organ. tions cannot be checked by natural selection. At what- As the presence of rudimentary organs is thus due

403 On the Origin of Species to the tendency in every part of the organisation, the relationship, by which all living and extinct be- which has long existed, to be inherited—we can un- ings are united by complex, radiating, and circui- derstand, on the genealogical view of classification, tous lines of affinities into one grand system; the how it is that systematists have found rudimentary rules followed and the difficulties encountered by parts as useful as, or even sometimes more useful naturalists in their classifications; the value set upon than, parts of high physiological importance. Rudi- characters, if constant and prevalent, whether of high mentary organs may be compared with the letters in vital importance, or of the most trifling importance, a word, still retained in the spelling, but become or, as in rudimentary organs, of no importance; the useless in the pronunciation, but which serve as a wide opposition in value between analogical or clue in seeking for its derivation. On the view of adaptive characters, and characters of true affinity; descent with modification, we may conclude that the and other such rules;—all naturally follow on the existence of organs in a rudimentary, imperfect, and view of the common parentage of those forms which useless condition, or quite aborted, far from present- are considered by naturalists as allied, together with ing a strange difficulty, as they assuredly do on the their modification through natural selection, with its ordinary doctrine of creation, might even have been contingencies of extinction and divergence of char- anticipated, and can be accounted for by the laws of acter. In considering this view of classification, it inheritance. should be borne in mind that the element of descent has been universally used in ranking together the Summary. — In this chapter I have attempted to sexes, ages, and acknowledged varieties of the same show, that the subordination of group to group in species, however different they may be in structure. all organisms throughout all time; that the nature of If we extend the use of this element of descent,—the

404 Charles Darwin only certainly known cause of similarity in organic in different species of a class of the homologous parts beings,—we shall understand what is meant by the or organs, though fitted in the adult members for pur- natural system: it is genealogical in its attempted poses as different as possible. Larvae are active em- arrangement, with the grades of acquired difference bryos, which have become specially modified in remarked by the terms varieties, species, genera, fami- lation to their habits of life, through the principle of lies, orders, and classes. modifications being inherited at corresponding ages. On this same view of descent with modification, On this same principle—and bearing in mind, that all the great facts in Morphology become intelli- when organs are reduced in size, either from disuse gible,—whether we look to the same pattern dis- or selection, it will generally be at that period of life played in the homologous organs, to whatever pur- when the being has to provide for its own wants, and pose applied, of the different species of a class; or to bearing in mind how strong is the principle of inher- the homologous parts constructed on the same pat- itance—the occurrence of rudimentary organs and tern in each individual animal and plant. their final abortion, present to us no inexplicable dif- On the principle of successive slight variations, not ficulties; on the contrary, their presence might have necessarily or generally supervening at a very early been even anticipated. The importance of embryologi- period of life, and being inherited at a correspond- cal characters and of rudimentary organs in classifi- ing period, we can understand the great leading facts cation is intelligible, on the view that an arrangement in Embryology; namely, the resemblance in an indi- is only so far natural as it is genealogical. vidual embryo of the homologous parts, which when Finally, the several classes of facts which have been matured will become widely different from each considered in this chapter, seem to me to proclaim other in structure and function; and the resemblance so plainly, that the innumerable species, genera, and

405 On the Origin of Species families of organic beings, with which this world is Chapter XIV peopled, have all descended, each within its own class or group, from common parents, and have all Recapitulation and Conclusion been modified in the course of descent, that I should without hesitation adopt this view, even if it were Recapitulation of the difficulties on the theory of unsupported by other facts or arguments. Natural Selection —Recapitulation of the general and special circumstances in its favour —Causes of the general belief in the immutability of species — How far the theory of natural selection may be extended — Effects of its adoption on the study of Natural history — Concluding remarks.

AS THIS WHOLE VOLUME is one long argument, it may be convenient to the reader to have the leading facts and inferences briefly recapitulated. That many and grave objections may be advanced against the theory of descent with modification through natural selection, I do not deny. I have endeavoured to give to them their full force. Nothing at first can appear more difficult to believe than that the more complex organs and instincts should have

406 Charles Darwin been perfected, not by means superior to, though extremely cautious in saying that any organ or in- analogous with, human reason, but by the accumu- stinct, or any whole being, could not have arrived at lation of innumerable slight variations, each good its present state by many graduated steps. There are, for the individual possessor. Nevertheless, this dif- it must be admitted, cases of special difficulty on ficulty, though appearing to our imagination insu- the theory of natural selection; and one of the most perably great, cannot be considered real if we admit curious of these is the existence of two or three de- the following propositions, namely,—that gradations fined castes of workers or sterile females in the same in the perfection of any organ or instinct, which we community of ants; but I have attempted to show may consider, either do now exist or could have ex- how this difficulty can be mastered. isted, each good of its kind,—that all organs and in- With respect to the almost universal sterility of spe- stincts are, in ever so slight a degree, variable,—and, cies when first crossed, which forms so remarkable lastly, that there is a struggle for existence leading a contrast with the almost universal fertility of vari- to the preservation of each profitable deviation of eties when crossed, I must refer the reader to the re- structure or instinct. The truth of these propositions capitulation of the facts given at the end of the eighth cannot, I think, be disputed. chapter, which seem to me conclusively to show that It is, no doubt, extremely difficult even to conjec- this sterility is no more a special endowment than is ture by what gradations many structures have been the incapacity of two trees to be grafted together, but perfected, more especially amongst broken and fail- that it is incidental on constitutional differences in ing groups of organic beings; but we see so many the reproductive systems of the intercrossed species. strange gradations in nature, as is proclaimed by the We see the truth of this conclusion in the vast differ- canon, ‘Natura non facit saltum,’ that we ought to be ence in the result, when the same two species are

407 On the Origin of Species crossed reciprocally; that is, when one species is first surprise at hybrids being in some degree sterile, for used as the father and then as the mother. their constitutions can hardly fail to have been dis- The fertility of varieties when intercrossed and of turbed from being compounded of two distinct their mongrel offspring cannot be considered as uni- organisations. This parallelism is supported by an- versal; nor is their very general fertility surprising other parallel, but directly opposite, class of facts; when we remember that it is not likely that either namely, that the vigour and fertility of all organic their constitutions or their reproductive systems beings are increased by slight changes in their con- should have been profoundly modified. Moreover, ditions of life, and that the offspring of slightly modi- most of the varieties which have been fied forms or varieties acquire from being crossed experimentised on have been produced under do- increased vigour and fertility. So that, on the one mestication; and as domestication apparently tends hand, considerable changes in the conditions of life to eliminate sterility, we ought not to expect it also and crosses between greatly modified forms, lessen to produce sterility. fertility; and on the other hand, lesser changes in the The sterility of hybrids is a very different case from conditions of life and crosses between less modified that of first crosses, for their reproductive organs are forms, increase fertility. more or less functionally impotent; whereas in first Turning to geographical distribution, the difficul- crosses the organs on both sides are in a perfect con- ties encountered on the theory of descent with modi- dition. As we continually see that organisms of all fication are grave enough. All the individuals of the kinds are rendered in some degree sterile from their same species, and all the species of the same genus, constitutions having been disturbed by slightly dif- or even higher group, must have descended from ferent and new conditions of life, we need not feel common parents; and therefore, in however distant

408 Charles Darwin and isolated parts of the world they are now found, sentative species throughout the world. We are as yet they must in the course of successive generations profoundly ignorant of the many occasional means have passed from some one part to the others. We of transport. With respect to distinct species of the are often wholly unable even to conjecture how this same genus inhabiting very distant and isolated re- could have been effected. Yet, as we have reason to gions, as the process of modification has necessarily believe that some species have retained the same been slow, all the means of migration will have been specific form for very long periods, enormously long possible during a very long period; and consequently as measured by years, too much stress ought not to the difficulty of the wide diffusion of species of the be laid on the occasional wide diffusion of the same same genus is in some degree lessened. species; for during very long periods of time there As on the theory of natural selection an intermi- will always be a good chance for wide migration by nable number of intermediate forms must have ex- many means. A broken or interrupted range may isted, linking together all the species in each group often be accounted for by the extinction of the spe- by gradations as fine as our present varieties, it may cies in the intermediate regions. It cannot be denied be asked, Why do we not see these linking forms all that we are as yet very ignorant of the full extent of around us? Why are not all organic beings blended the various climatal and geographical changes which together in an inextricable chaos? With respect to have affected the earth during modern periods; and existing forms, we should remember that we have such changes will obviously have greatly facilitated no right to expect (excepting in rare cases) to dis- migration. As an example, I have attempted to show cover directly connecting links between them, but how potent has been the influence of the Glacial pe- only between each and some extinct and supplanted riod on the distribution both of the same and of repre- form. Even on a wide area, which has during a long

409 On the Origin of Species period remained continuous, and of which the cli- why is not every geological formation charged with mate and other conditions of life change insensibly such links? Why does not every collection of fossil in going from a district occupied by one species into remains afford plain evidence of the gradation and another district occupied by a closely allied species, mutation of the forms of life? We meet with no such we have no just right to expect often to find interme- evidence, and this is the most obvious and forcible diate varieties in the intermediate zone. For we have of the many objections which may be urged against reason to believe that only a few species are under- my theory. Why, again, do whole groups of allied going change at any one period; and all changes are species appear, though certainly they often falsely slowly effected. I have also shown that the interme- appear, to have come in suddenly on the several diate varieties which will at first probably exist in geological stages? Why do we not find great piles of the intermediate zones, will be liable to be sup- strata beneath the Silurian system, stored with the planted by the allied forms on either hand; and the remains of the progenitors of the Silurian groups of latter, from existing in greater numbers, will gener- fossils? For certainly on my theory such strata must ally be modified and improved at a quicker rate than somewhere have been deposited at these ancient and the intermediate varieties, which exist in lesser num- utterly unknown epochs in the world’s history. bers; so that the intermediate varieties will, in the I can answer these questions and grave objections long run, be supplanted and exterminated. only on the supposition that the geological record is On this doctrine of the extermination of an infini- far more imperfect than most geologists believe. It tude of connecting links, between the living and ex- cannot be objected that there has not been time suffi- tinct inhabitants of the world, and at each successive cient for any amount of organic change; for the lapse period between the extinct and still older species, of time has been so great as to be utterly inappre-

410 Charles Darwin ciable by the human intellect. The number of speci- of certain classes can be preserved in a fossil condi- mens in all our museums is absolutely as nothing tion, at least in any great number. Widely ranging compared with the countless generations of count- species vary most, and varieties are often at first lo- less species which certainly have existed. We should cal,—both causes rendering the discovery of inter- not be able to recognise a species as the parent of mediate links less likely. Local varieties will not any one or more species if we were to examine them spread into other and distant regions until they are ever so closely, unless we likewise possessed many considerably modified and improved; and when of the intermediate links between their past or par- they do spread, if discovered in a geological forma- ent and present states; and these many links we could tion, they will appear as if suddenly created there, hardly ever expect to discover, owing to the imper- and will be simply classed as new species. Most for- fection of the geological record. Numerous existing mations have been intermittent in their accumula- doubtful forms could be named which are probably tion; and their duration, I am inclined to believe, has varieties; but who will pretend that in future ages so been shorter than the average duration of specific many fossil links will be discovered, that naturalists forms. Successive formations are separated from each will be able to decide, on the common view, whether other by enormous blank intervals of time; for fos- or not these doubtful forms are varieties? As long as siliferous formations, thick enough to resist future most of the links between any two species are un- degradation, can be accumulated only where much known, if any one link or intermediate variety be sediment is deposited on the subsiding bed of the discovered, it will simply be classed as another and sea. During the alternate periods of elevation and of distinct species. Only a small portion of the world stationary level the record will be blank. During these has been geologically explored. Only organic beings latter periods there will probably be more variabil-

411 On the Origin of Species ity in the forms of life; during periods of subsidence, many years to doubt their weight. But it deserves more extinction. especial notice that the more important objections With respect to the absence of fossiliferous forma- relate to questions on which we are confessedly ig- tions beneath the lowest Silurian strata, I can only norant; nor do we know how ignorant we are. We recur to the hypothesis given in the ninth chapter. do not know all the possible transitional gradations That the geological record is imperfect all will ad- between the simplest and the most perfect organs; it mit; but that it is imperfect to the degree which I re- cannot be pretended that we know all the varied quire, few will be inclined to admit. If we look to means of Distribution during the long lapse of years, long enough intervals of time, geology plainly de- or that we know how imperfect the Geological clares that all species have changed; and they have Record is. Grave as these several difficulties are, in changed in the manner which my theory requires, my judgment they do not overthrow the theory of for they have changed slowly and in a graduated descent with modification. manner. We clearly see this in the fossil remains from Now let us turn to the other side of the argument. consecutive formations invariably being much more Under domestication we see much variability. This closely related to each other, than are the fossils from seems to be mainly due to the reproductive system formations distant from each other in time. being eminently susceptible to changes in the con- Such is the sum of the several chief objections and ditions of life; so that this system, when not rendered difficulties which may justly be urged against my impotent, fails to reproduce offspring exactly like theory; and I have now briefly recapitulated the an- the parent-form. Variability is governed by many swers and explanations which can be given to them. complex laws,—by correlation of growth, by use and I have felt these difficulties far too heavily during disuse, and by the direct action of the physical con-

412 Charles Darwin ditions of life. There is much difficulty in ascertain- may do it unconsciously by preserving the individu- ing how much modification our domestic productions als most useful to him at the time, without any have undergone; but we may safely infer that the thought of altering the breed. It is certain that he can amount has been large, and that modifications can be largely influence the character of a breed by select- inherited for long periods. As long as the conditions ing, in each successive generation, individual dif- of life remain the same, we have reason to believe ferences so slight as to be quite inappreciable by an that a modification, which has already been inherited uneducated eye. This process of selection has been for many generations, may continue to be inherited the great agency in the production of the most dis- for an almost infinite number of generations. On the tinct and useful domestic breeds. That many of the other hand we have evidence that variability, when it breeds produced by man have to a large extent the has once come into play, does not wholly cease; for character of natural species, is shown by the inextri- new varieties are still occasionally produced by our cable doubts whether very many of them are variet- most anciently domesticated productions. ies or aboriginal species. Man does not actually produce variability; he only There is no obvious reason why the principles unintentionally exposes organic beings to new con- which have acted so efficiently under domestication ditions of life, and then nature acts on the should not have acted under nature. In the preser- organisation, and causes variability. But man can and vation of favoured individuals and races, during the does select the variations given to him by nature, constantly-recurrent Struggle for Existence, we see and thus accumulate them in any desired manner. the most powerful and ever-acting means of selec- He thus adapts animals and plants for his own ben- tion. The struggle for existence inevitably follows efit or pleasure. He may do this methodically, or he from the high geometrical ratio of increase which is

413 On the Origin of Species common to all organic beings. This high rate of in- With animals having separated sexes there will in crease is proved by calculation, by the effects of a most cases be a struggle between the males for pos- succession of peculiar seasons, and by the results of session of the females. The most vigorous individu- naturalisation, as explained in the third chapter. als, or those which have most successfully struggled More individuals are born than can possibly survive. with their conditions of life, will generally leave most A grain in the balance will determine which indi- progeny. But success will often depend on having vidual shall live and which shall die,—which vari- special weapons or means of defence, or on the ety or species shall increase in number, and which charms of the males; and the slightest advantage will shall decrease, or finally become extinct. As the in- lead to victory. dividuals of the same species come in all respects As geology plainly proclaims that each land has into the closest competition with each other, the undergone great physical changes, we might have struggle will generally be most severe between them; expected that organic beings would have varied it will be almost equally severe between the variet- under nature, in the same way as they generally have ies of the same species, and next in severity between varied under the changed conditions of domestica- the species of the same genus. But the struggle will tion. And if there be any variability under nature, it often be very severe between beings most remote in would be an unaccountable fact if natural selection the scale of nature. The slightest advantage in one had not come into play. It has often been asserted, being, at any age or during any season, over those but the assertion is quite incapable of proof, that the with which it comes into competition, or better ad- amount of variation under nature is a strictly lim- aptation in however slight a degree to the surround- ited quantity. Man, though acting on external char- ing physical conditions, will turn the balance. acters alone and often capriciously, can produce

414 Charles Darwin within a short period a great result by adding up tions of life, to her living products? What limit can mere individual differences in his domestic produc- be put to this power, acting during long ages and tions; and every one admits that there are at least rigidly scrutinising the whole constitution, structure, individual differences in species under nature. But, and habits of each creature,—favouring the good and besides such differences, all naturalists have admit- rejecting the bad? I can see no limit to this power, in ted the existence of varieties, which they think suffi- slowly and beautifully adapting each form to the ciently distinct to be worthy of record in systematic most complex relations of life. The theory of natural works. No one can draw any clear distinction be- selection, even if we looked no further than this, tween individual differences and slight varieties; or seems to me to be in itself probable. I have already between more plainly marked varieties and sub-spe- recapitulated, as fairly as I could, the opposed diffi- cies, and species. Let it be observed how naturalists culties and objections: now let us turn to the special differ in the rank which they assign to the many rep- facts and arguments in favour of the theory. resentative forms in Europe and North America. On the view that species are only strongly marked If then we have under nature variability and a pow- and permanent varieties, and that each species first erful agent always ready to act and select, why existed as a variety, we can see why it is that no line should we doubt that variations in any way useful of demarcation can be drawn between species, com- to beings, under their excessively complex relations monly supposed to have been produced by special of life, would be preserved, accumulated, and in- acts of creation, and varieties which are acknowl- herited? Why, if man can by patience select varia- edged to have been produced by secondary laws. tions most useful to himself, should nature fail in On this same view we can understand how it is that selecting variations useful, under changing condi- in each region where many species of a genus have

415 On the Origin of Species been produced, and where they now flourish, these become more diversified in habits and structure, so same species should present many varieties; for as to be enabled to seize on many and widely differ- where the manufactory of species has been active, ent places in the economy of nature, there will be a we might expect, as a general rule, to find it still in constant tendency in natural selection to preserve action; and this is the case if varieties be incipient the most divergent offspring of any one species. species. Moreover, the species of the large genera, Hence during a long-continued course of modifica- which afford the greater number of varieties or in- tion, the slight differences, characteristic of varieties cipient species, retain to a certain degree the charac- of the same species, tend to be augmented into the ter of varieties; for they differ from each other by a greater differences characteristic of species of the less amount of difference than do the species of same genus. New and improved varieties will in- smaller genera. The closely allied species also of the evitably supplant and exterminate the older, less larger genera apparently have restricted ranges, and improved and intermediate varieties; and thus spe- they are clustered in little groups round other species are rendered to a large extent defined and dis- cies—in which respects they resemble varieties. tinct objects. Dominant species belonging to the These are strange relations on the view of each spe- larger groups tend to give birth to new and domi- cies having been independently created, but are in- nant forms; so that each large group tends to become telligible if all species first existed as varieties. still larger, and at the same time more divergent in As each species tends by its geometrical ratio of character. But as all groups cannot thus succeed in reproduction to increase inordinately in number; and increasing in size, for the world would not hold them, as the modified descendants of each species will be the more dominant groups beat the less dominant. enabled to increase by so much the more as they This tendency in the large groups to go on increas-

416 Charles Darwin ing in size and diverging in character, together with form of woodpecker, should have been created to prey the almost inevitable contingency of much extinc- on insects on the ground; that upland geese, which tion, explains the arrangement of all the forms of life, never or rarely swim, should have been created with in groups subordinate to groups, all within a few webbed feet; that a thrush should have been created great classes, which we now see everywhere around to dive and feed on sub-aquatic insects; and that a us, and which has prevailed throughout all time. This petrel should have been created with habits and struc- grand fact of the grouping of all organic beings seems ture fitting it for the life of an auk or grebe! and so on to me utterly inexplicable on the theory of creation. in endless other cases. But on the view of each spe- As natural selection acts solely by accumulating cies constantly trying to increase in number, with natu- slight, successive, favourable variations, it can pro- ral selection always ready to adapt the slowly vary- duce no great or sudden modification; it can act only ing descendants of each to any unoccupied or ill-oc- by very short and slow steps. Hence the canon of cupied place in nature, these facts cease to be strange, ‘Natura non facit saltum,’ which every fresh addi- or perhaps might even have been anticipated. tion to our knowledge tends to make more strictly As natural selection acts by competition, it adapts correct, is on this theory simply intelligible. We can the inhabitants of each country only in relation to plainly see why nature is prodigal in variety, though the degree of perfection of their associates; so that niggard in innovation. But why this should be a law we need feel no surprise at the inhabitants of any of nature if each species has been independently cre- one country, although on the ordinary view sup- ated, no man can explain. posed to have been specially created and adapted Many other facts are, as it seems to me, explicable for that country, being beaten and supplanted by the on this theory. How strange it is that a bird, under the naturalised productions from another land. Nor

417 On the Origin of Species ought we to marvel if all the contrivances in nature to that zone. In both varieties and species, use and be not, as far as we can judge, absolutely perfect; disuse seem to have produced some effect; for it is and if some of them be abhorrent to our ideas of fit- difficult to resist this conclusion when we look, for ness. We need not marvel at the sting of the bee caus- instance, at the logger-headed duck, which has wings ing the bee’s own death; at drones being produced incapable of flight, in nearly the same condition as in such vast numbers for one single act, and being in the domestic duck; or when we look at the bur- then slaughtered by their sterile sisters; at the aston- rowing tucutucu, which is occasionally blind, and ishing waste of pollen by our fir-trees; at the instinc- then at certain moles, which are habitually blind and tive hatred of the queen bee for her own fertile daugh- have their eyes covered with skin; or when we look ters; at ichneumonidae feeding within the live bod- at the blind animals inhabiting the dark caves of ies of caterpillars; and at other such cases. The won- America and Europe. In both varieties and species der indeed is, on the theory of natural selection, that correlation of growth seems to have played a most more cases of the want of absolute perfection have important part, so that when one part has been modi- not been observed. fied other parts are necessarily modified. In both The complex and little known laws governing varieties and species reversions to long-lost charac- variation are the same, as far as we can see, with the ters occur. How inexplicable on the theory of cre- laws which have governed the production of so- ation is the occasional appearance of stripes on the called specific forms. In both cases physical condi- shoulder and legs of the several species of the horse- tions seem to have produced but little direct effect; genus and in their hybrids! How simply is this fact yet when varieties enter any zone, they occasionally explained if we believe that these species have de- assume some of the characters of the species proper scended from a striped progenitor, in the same man-

418 Charles Darwin ner as the several domestic breeds of pigeon have enormous period. It is inexplicable on the theory of descended from the blue and barred rock-pigeon! creation why a part developed in a very unusual On the ordinary view of each species having been manner in any one species of a genus, and therefore, independently created, why should the specific char- as we may naturally infer, of great importance to the acters, or those by which the species of the same ge- species, should be eminently liable to variation; but, nus differ from each other, be more variable than the on my view, this part has undergone, since the sev- generic characters in which they all agree? Why, for eral species branched off from a common progeni- instance, should the colour of a flower be more likely tor, an unusual amount of variability and modifica- to vary in any one species of a genus, if the other tion, and therefore we might expect this part gener- species, supposed to have been created indepen- ally to be still variable. But a part may be developed dently, have differently coloured flowers, than if all in the most unusual manner, like the wing of a bat, the species of the genus have the same coloured flow- and yet not be more variable than any other struc- ers? If species are only well-marked varieties, of ture, if the part be common to many subordinate which the characters have become in a high degree forms, that is, if it has been inherited for a very long permanent, we can understand this fact; for they have period; for in this case it will have been rendered already varied since they branched off from a com- constant by long-continued natural selection. mon progenitor in certain characters, by which they Glancing at instincts, marvellous as some are, they have come to be specifically distinct from each other; offer no greater difficulty than does corporeal struc- and therefore these same characters would be more ture on the theory of the natural selection of succes- likely still to be variable than the generic characters sive, slight, but profitable modifications. We can thus which have been inherited without change for an understand why nature moves by graduated steps

419 On the Origin of Species in endowing different animals of the same class with rieties, we can at once see why their crossed offspring their several instincts. I have attempted to show how should follow the same complex laws in their de- much light the principle of gradation throws on the grees and kinds of resemblance to their parents,—in admirable architectural powers of the hive-bee. Habit being absorbed into each other by successive crosses, no doubt sometimes comes into play in modifying and in other such points,—as do the crossed off- instincts; but it certainly is not indispensable, as we spring of acknowledged varieties. On the other hand, see, in the case of neuter insects, which leave no prog- these would be strange facts if species have been eny to inherit the effects of long-continued habit. On independently created, and varieties have been pro- the view of all the species of the same genus having duced by secondary laws. descended from a common parent, and having in- If we admit that the geological record is imperfect herited much in common, we can understand how it in an extreme degree, then such facts as the record is that allied species, when placed under consider- gives, support the theory of descent with modifica- ably different conditions of life, yet should follow tion. New species have come on the stage slowly and nearly the same instincts; why the thrush of South at successive intervals; and the amount of change, America, for instance, lines her nest with mud like after equal intervals of time, is widely different in our British species. On the view of instincts having different groups. The extinction of species and of been slowly acquired through natural selection we whole groups of species, which has played so con- need not marvel at some instincts being apparently spicuous a part in the history of the organic world, not perfect and liable to mistakes, and at many in- almost inevitably follows on the principle of natural stincts causing other animals to suffer. selection; for old forms will be supplanted by new If species be only well-marked and permanent va- and improved forms. Neither single species nor

420 Charles Darwin groups of species reappear when the chain of ordi- mediate between existing and allied groups. Recent nary generation has once been broken. The gradual forms are generally looked at as being, in some diffusion of dominant forms, with the slow modifi- vague sense, higher than ancient and extinct forms; cation of their descendants, causes the forms of life, and they are in so far higher as the later and more after long intervals of time, to appear as if they had improved forms have conquered the older and less changed simultaneously throughout the world. The improved organic beings in the struggle for life. fact of the fossil remains of each formation being in Lastly, the law of the long endurance of allied forms some degree intermediate in character between the on the same continent,—of marsupials in Australia, fossils in the formations above and below, is simply of edentata in America, and other such cases,—is in- explained by their intermediate position in the chain telligible, for within a confined country, the recent of descent. The grand fact that all extinct organic be- and the extinct will naturally be allied by descent. ings belong to the same system with recent beings, Looking to geographical distribution, if we admit falling either into the same or into intermediate that there has been during the long course of ages groups, follows from the living and the extinct be- much migration from one part of the world to an- ing the offspring of common parents. As the groups other, owing to former climatal and geographical which have descended from an ancient progenitor changes and to the many occasional and unknown have generally diverged in character, the progenitor means of dispersal, then we can understand, on the with its early descendants will often be intermedi- theory of descent with modification, most of the great ate in character in comparison with its later descen- leading facts in Distribution. We can see why there dants; and thus we can see why the more ancient a should be so striking a parallelism in the distribu- fossil is, the oftener it stands in some degree inter- tion of organic beings throughout space, and in their

421 On the Origin of Species geological succession throughout time; for in both surprise at their inhabitants being widely different, if cases the beings have been connected by the bond of they have been for a long period completely sepa- ordinary generation, and the means of modification rated from each other; for as the relation of organism have been the same. We see the full meaning of the to organism is the most important of all relations, and wonderful fact, which must have struck every trav- as the two areas will have received colonists from eller, namely, that on the same continent, under the some third source or from each other, at various peri- most diverse conditions, under heat and cold, on ods and in different proportions, the course of modi- mountain and lowland, on deserts and marshes, most fication in the two areas will inevitably be different. of the inhabitants within each great class are plainly On this view of migration, with subsequent modi- related; for they will generally be descendants of the fication, we can see why oceanic islands should be same progenitors and early colonists. On this same inhabited by few species, but of these, that many principle of former migration, combined in most should be peculiar. We can clearly see why those cases with modification, we can understand, by the animals which cannot cross wide spaces of ocean, as aid of the Glacial period, the identity of some few frogs and terrestrial mammals, should not inhabit plants, and the close alliance of many others, on the oceanic islands; and why, on the other hand, new most distant mountains, under the most different and peculiar species of bats, which can traverse the climates; and likewise the close alliance of some of ocean, should so often be found on islands far dis- the inhabitants of the sea in the northern and south- tant from any continent. Such facts as the presence of ern temperate zones, though separated by the whole peculiar species of bats, and the absence of all other intertropical ocean. Although two areas may present mammals, on oceanic islands, are utterly inexpli- the same physical conditions of life, we need feel no cable on the theory of independent acts of creation.

422 Charles Darwin The existence of closely allied or representative The fact, as we have seen, that all past and present species in any two areas, implies, on the theory of organic beings constitute one grand natural system, descent with modification, that the same parents for- with group subordinate to group, and with extinct merly inhabited both areas; and we almost invari- groups often falling in between recent groups, is in- ably find that wherever many closely allied spe- telligible on the theory of natural selection with its cies inhabit two areas, some identical species com- contingencies of extinction and divergence of char- mon to both still exist. Wherever many closely al- acter. On these same principles we see how it is, that lied yet distinct species occur, many doubtful forms the mutual affinities of the species and genera within and varieties of the same species likewise occur. It each class are so complex and circuitous. We see why is a rule of high generality that the inhabitants of certain characters are far more serviceable than oth- each area are related to the inhabitants of the near- ers for classification;—why adaptive characters, est source whence immigrants might have been though of paramount importance to the being, are derived. We see this in nearly all the plants and of hardly any importance in classification; why char- animals of the Galapagos archipelago, of Juan acters derived from rudimentary parts, though of no Fernandez, and of the other American islands be- service to the being, are often of high classificatory ing related in the most striking manner to the plants value; and why embryological characters are the and animals of the neighbouring American main- most valuable of all. The real affinities of all organic land; and those of the Cape de Verde archipelago beings are due to inheritance or community of de- and other African islands to the African mainland. scent. The natural system is a genealogical arrange- It must be admitted that these facts receive no ex- ment, in which we have to discover the lines of de- planation on the theory of creation. scent by the most permanent characters, however

423 On the Origin of Species slight their vital importance may be. and arteries running in loops, like those in a fish The framework of bones being the same in the hand which has to breathe the air dissolved in water, by of a man, wing of a bat, fin of the porpoise, and leg the aid of well-developed branchiae. of the horse,—the same number of vertebrae form- Disuse, aided sometimes by natural selection, will ing the neck of the giraffe and of the elephant,—and often tend to reduce an organ, when it has become innumerable other such facts, at once explain them- useless by changed habits or under changed condi- selves on the theory of descent with slow and slight tions of life; and we can clearly understand on this successive modifications. The similarity of pattern view the meaning of rudimentary organs. But dis- in the wing and leg of a bat, though used for such use and selection will generally act on each creature, different purpose,—in the jaws and legs of a crab,— when it has come to maturity and has to play its full in the petals, stamens, and pistils of a flower, is like- part in the struggle for existence, and will thus have wise intelligible on the view of the gradual modifi- little power of acting on an organ during early life; cation of parts or organs, which were alike in the early hence the organ will not be much reduced or ren- progenitor of each class. On the principle of succes- dered rudimentary at this early age. The calf, for in- sive variations not always supervening at an early stance, has inherited teeth, which never cut through age, and being inherited at a corresponding not early the gums of the upper jaw, from an early progenitor period of life, we can clearly see why the embryos of having well-developed teeth; and we may believe, mammals, birds, reptiles, and fishes should be so that the teeth in the mature animal were reduced, closely alike, and should be so unlike the adult during successive generations, by disuse or by the forms. We may cease marvelling at the embryo of an tongue and palate having been fitted by natural se- air-breathing mammal or bird having branchial slits lection to browse without their aid; whereas in the

424 Charles Darwin calf, the teeth have been left untouched by selection in a state of nature are subject to no variation; it can- or disuse, and on the principle of inheritance at corre- not be proved that the amount of variation in the sponding ages have been inherited from a remote course of long ages is a limited quantity; no clear period to the present day. On the view of each or- distinction has been, or can be, drawn between spe- ganic being and each separate organ having been species and well-marked varieties. It cannot be main- cially created, how utterly inexplicable it is that parts, tained that species when intercrossed are invariably like the teeth in the embryonic calf or like the shriv- sterile, and varieties invariably fertile; or that steril- elled wings under the soldered wing-covers of some ity is a special endowment and sign of creation. The beetles, should thus so frequently bear the plain stamp belief that species were immutable productions was of inutility! Nature may be said to have taken pains almost unavoidable as long as the history of the to reveal, by rudimentary organs and by homologous world was thought to be of short duration; and now structures, her scheme of modification, which it seems that we have acquired some idea of the lapse of time, that we wilfully will not understand. we are too apt to assume, without proof, that the I have now recapitulated the chief facts and con- geological record is so perfect that it would have af- siderations which have thoroughly convinced me forded us plain evidence of the mutation of species, that species have changed, and are still slowly chang- if they had undergone mutation. ing by the preservation and accumulation of succes- But the chief cause of our natural unwillingness to sive slight favourable variations. Why, it may be admit that one species has given birth to other and asked, have all the most eminent living naturalists distinct species, is that we are always slow in admit- and geologists rejected this view of the mutability ting any great change of which we do not see the of species? It cannot be asserted that organic beings intermediate steps. The difficulty is the same as that

425 On the Origin of Species felt by so many geologists, when Lyell first insisted ralists, endowed with much flexibility of mind, and that long lines of inland cliffs had been formed, and who have already begun to doubt on the immuta- great valleys excavated, by the slow action of the bility of species, may be influenced by this volume; coast-waves. The mind cannot possibly grasp the full but I look with confidence to the future, to young meaning of the term of a hundred million years; it and rising naturalists, who will be able to view both cannot add up and perceive the full effects of many sides of the question with impartiality. Whoever is slight variations, accumulated during an almost in- led to believe that species are mutable will do good finite number of generations. service by conscientiously expressing his conviction; Although I am fully convinced of the truth of the for only thus can the load of prejudice by which this views given in this volume under the form of an ab- subject is overwhelmed be removed. stract, I by no means expect to convince experienced Several eminent naturalists have of late published naturalists whose minds are stocked with a multi- their belief that a multitude of reputed species in tude of facts all viewed, during a long course of each genus are not real species; but that other spe- years, from a point of view directly opposite to mine. cies are real, that is, have been independently cre- It is so easy to hide our ignorance under such ex- ated. This seems to me a strange conclusion to ar- pressions as the ‘plan of creation,’ ‘unity of design,’ rive at. They admit that a multitude of forms, which &c., and to think that we give an explanation when till lately they themselves thought were special cre- we only restate a fact. Any one whose disposition ations, and which are still thus looked at by the ma- leads him to attach more weight to unexplained dif- jority of naturalists, and which consequently have ficulties than to the explanation of a certain number every external characteristic feature of true species,— of facts will certainly reject my theory. A few natu- they admit that these have been produced by varia-

426 Charles Darwin tion, but they refuse to extend the same view to other ralists very properly demand a full explanation of and very slightly different forms. Nevertheless they every difficulty from those who believe in the muta- do not pretend that they can define, or even conjec- bility of species, on their own side they ignore the ture, which are the created forms of life, and which whole subject of the first appearance of species in are those produced by secondary laws. They admit what they consider reverent silence. variation as a vera causa in one case, they arbitrarily It may be asked how far I extend the doctrine of reject it in another, without assigning any distinc- the modification of species. The question is difficult tion in the two cases. The day will come when this to answer, because the more distinct the forms are will be given as a curious illustration of the blind- which we may consider, by so much the arguments ness of preconceived opinion. These authors seem fall away in force. But some arguments of the great- no more startled at a miraculous act of creation than est weight extend very far. All the members of whole at an ordinary birth. But do they really believe that classes can be connected together by chains of affini- at innumerable periods in the earth’s history certain ties, and all can be classified on the same principle, elemental atoms have been commanded suddenly in groups subordinate to groups. Fossil remains to flash into living tissues? Do they believe that at sometimes tend to fill up very wide intervals be- each supposed act of creation one individual or tween existing orders. Organs in a rudimentary con- many were produced? Were all the infinitely numer- dition plainly show that an early progenitor had the ous kinds of animals and plants created as eggs or organ in a fully developed state; and this in some seed, or as full grown? and in the case of mammals, instances necessarily implies an enormous amount were they created bearing the false marks of nour- of modification in the descendants. Throughout ishment from the mother’s womb? Although natu- whole classes various structures are formed on the

427 On the Origin of Species same pattern, and at an embryonic age the species closely When the views entertained in this volume on the resemble each other. Therefore I cannot doubt that the origin of species, or when analogous views are gen- theory of descent with modification embraces all the erally admitted, we can dimly foresee that there will members of the same class. I believe that animals have be a considerable revolution in natural history. Sys- descended from at most only four or five progenitors, tematists will be able to pursue their labours as at and plants from an equal or lesser number. present; but they will not be incessantly haunted by Analogy would lead me one step further, namely, the shadowy doubt whether this or that form be in to the belief that all animals and plants have de- essence a species. This I feel sure, and I speak after scended from some one prototype. But analogy may experience, will be no slight relief. The endless dis- be a deceitful guide. Nevertheless all living things putes whether or not some fifty species of British have much in common, in their chemical composi- brambles are true species will cease. Systematists tion, their germinal vesicles, their cellular structure, will have only to decide (not that this will be easy) and their laws of growth and reproduction. We see whether any form be sufficiently constant and dis- this even in so trifling a circumstance as that the same tinct from other forms, to be capable of definition; poison often similarly affects plants and animals; or and if definable, whether the differences be suffi- that the poison secreted by the gall-fly produces ciently important to deserve a specific name. This monstrous growths on the wild rose or oak-tree. latter point will become a far more essential consid- Therefore I should infer from analogy that probably eration than it is at present; for differences, however all the organic beings which have ever lived on this slight, between any two forms, if not blended by in- earth have descended from some one primordial termediate gradations, are looked at by most natu- form, into which life was first breathed. ralists as sufficient to raise both forms to the rank of

428 Charles Darwin species. Hereafter we shall be compelled to acknowl- The other and more general departments of natu- edge that the only distinction between species and ral history will rise greatly in interest. The terms used well-marked varieties is, that the latter are known, or by naturalists of affinity, relationship, community believed, to be connected at the present day by inter- of type, paternity, morphology, adaptive characters, mediate gradations, whereas species were formerly rudimentary and aborted organs, &c., will cease to thus connected. Hence, without quite rejecting the con- be metaphorical, and will have a plain signification. sideration of the present existence of intermediate When we no longer look at an organic being as a gradations between any two forms, we shall be led to savage looks at a ship, as at something wholly be- weigh more carefully and to value higher the actual yond his comprehension; when we regard every pro- amount of difference between them. It is quite pos- duction of nature as one which has had a history; sible that forms now generally acknowledged to be when we contemplate every complex structure and merely varieties may hereafter be thought worthy of instinct as the summing up of many contrivances, specific names, as with the primrose and cowslip; and each useful to the possessor, nearly in the same way in this case scientific and common language will come as when we look at any great mechanical invention into accordance. In short, we shall have to treat spe- as the summing up of the labour, the experience, the cies in the same manner as those naturalists treat gen- reason, and even the blunders of numerous work- era, who admit that genera are merely artificial com- men; when we thus view each organic being, how binations made for convenience. This may not be a far more interesting, I speak from experience, will cheering prospect; but we shall at least be freed from the study of natural history become! the vain search for the undiscovered and undiscover- A grand and almost untrodden field of inquiry will able essence of the term species. be opened, on the causes and laws of variation, on

429 On the Origin of Species correlation of growth, on the effects of use and dis- ture, in some degree obscured, of the prototypes of use, on the direct action of external conditions, and each great class. so forth. The study of domestic productions will rise When we can feel assured that all the individuals immensely in value. A new variety raised by man of the same species, and all the closely allied spe- will be a far more important and interesting subject cies of most genera, have within a not very remote for study than one more species added to the infini- period descended from one parent, and have mi- tude of already recorded species. Our classifications grated from some one birthplace; and when we bet- will come to be, as far as they can be so made, gene- ter know the many means of migration, then, by the alogies; and will then truly give what may be called light which geology now throws, and will continue the plan of creation. The rules for classifying will no to throw, on former changes of climate and of the doubt become simpler when we have a definite ob- level of the land, we shall surely be enabled to trace ject in view. We possess no pedigrees or armorial in an admirable manner the former migrations of the bearings; and we have to discover and trace the many inhabitants of the whole world. Even at present, by diverging lines of descent in our natural genealo- comparing the differences of the inhabitants of the gies, by characters of any kind which have long been sea on the opposite sides of a continent, and the na- inherited. Rudimentary organs will speak infallibly ture of the various inhabitants of that continent in with respect to the nature of long-lost structures. relation to their apparent means of immigration, Species and groups of species, which are called ab- some light can be thrown on ancient geography. errant, and which may fancifully be called living fos- The noble science of Geology loses glory from the sils, will aid us in forming a picture of the ancient extreme imperfection of the record. The crust of the forms of life. Embryology will reveal to us the struc- earth with its embedded remains must not be looked

430 Charles Darwin at as a well-filled museum, but as a poor collection tion of others; it follows, that the amount of organic made at hazard and at rare intervals. The accumula- change in the fossils of consecutive formations prob- tion of each great fossiliferous formation will be ably serves as a fair measure of the lapse of actual recognised as having depended on an unusual con- time. A number of species, however, keeping in a currence of circumstances, and the blank intervals body might remain for a long period unchanged, between the successive stages as having been of vast whilst within this same period, several of these spe- duration. But we shall be able to gauge with some cies, by migrating into new countries and coming security the duration of these intervals by a compari- into competition with foreign associates, might be- son of the preceding and succeeding organic forms. come modified; so that we must not overrate the ac- We must be cautious in attempting to correlate as curacy of organic change as a measure of time. Dur- strictly contemporaneous two formations, which in- ing early periods of the earth’s history, when the clude few identical species, by the general succes- forms of life were probably fewer and simpler, the sion of their forms of life. As species are produced rate of change was probably slower; and at the first and exterminated by slowly acting and still existing dawn of life, when very few forms of the simplest causes, and not by miraculous acts of creation and structure existed, the rate of change may have been by catastrophes; and as the most important of all slow in an extreme degree. The whole history of the causes of organic change is one which is almost in- world, as at present known, although of a length dependent of altered and perhaps suddenly altered quite incomprehensible by us, will hereafter be physical conditions, namely, the mutual relation of recognised as a mere fragment of time, compared organism to organism,—the improvement of one with the ages which have elapsed since the first crea- being entailing the improvement or the extermina- ture, the progenitor of innumerable extinct and liv-

431 On the Origin of Species ing descendants, was created. of the species now living very few will transmit In the distant future I see open fields for far more progeny of any kind to a far distant futurity; for the important researches. Psychology will be based on a manner in which all organic beings are grouped, new foundation, that of the necessary acquirement of shows that the greater number of species of each each mental power and capacity by gradation. Light genus, and all the species of many genera, have left will be thrown on the origin of man and his history. no descendants, but have become utterly extinct. Authors of the highest eminence seem to be fully We can so far take a prophetic glance into futurity satisfied with the view that each species has been as to foretel that it will be the common and widely- independently created. To my mind it accords bet- spread species, belonging to the larger and domi- ter with what we know of the laws impressed on nant groups, which will ultimately prevail and pro- matter by the Creator, that the production and ex- create new and dominant species. As all the living tinction of the past and present inhabitants of the forms of life are the lineal descendants of those world should have been due to secondary causes, which lived long before the Silurian epoch, we may like those determining the birth and death of the feel certain that the ordinary succession by genera- individual. When I view all beings not as special tion has never once been broken, and that no cata- creations, but as the lineal descendants of some few clysm has desolated the whole world. Hence we beings which lived long before the first bed of the may look with some confidence to a secure future Silurian system was deposited, they seem to me to of equally inappreciable length. And as natural se- become ennobled. Judging from the past, we may lection works solely by and for the good of each safely infer that not one living species will trans- being, all corporeal and mental endowments will mit its unaltered likeness to a distant futurity. And tend to progress towards perfection.

432 Charles Darwin It is interesting to contemplate an entangled bank, breathed by the Creator into a few forms or into one; clothed with many plants of many kinds, with birds and that, whilst this planet has gone cycling on ac- singing on the bushes, with various insects flitting cording to the fixed law of gravity, from so simple a about, and with worms crawling through the damp beginning endless forms most beautiful and most earth, and to reflect that these elaborately constructed wonderful have been, and are being, evolved. forms, so different from each other, and dependent on each other in so complex a manner, have all been produced by laws acting around us. These laws, taken in the largest sense, being Growth with Re- production; Inheritance which is almost implied by reproduction; Variability from the indirect and direct action of the external conditions of life, and from use and disuse; a Ratio of Increase so high as to lead to a Struggle for Life, and as a consequence to Natu- ral Selection, entailing Divergence of Character and the Extinction of less-improved forms. Thus, from the war of nature, from famine and death, the most exalted object which we are capable of conceiving, namely, the production of the higher animals, directly follows. There is grandeur in this view of life, with its several powers, having been originally

433 On the Origin of Species Subject Index acclimatisation of; of Australia; with thicker fur in cold climates; blind, in caves; extinct, of Australia Aberrant groups Anomma Abyssinia, plants of Antarctic islands, ancient flora of Acclimatisation Antirrhinum Affinities of extinct species; of organic beings Ants attending aphides; slave-making instinct Agassiz on Amblyopsis; on groups of species suddenly appearing; on embryological succes- Ants, neuter, structure of sion; on the glacial period; on embryological Aphides attended by ants characters; on the embryos of vertebrata; on Aphis, development of parallelism of embryological development and Apteryx geological succession Arab horses Algae of New Zealand Aralo-Caspian Sea Alligators, males, fighting Archiac, M. de, on the succession of species Amblyopsis, blind fish America, North, productions allied to those of Artichoke, Jerusalem Europe; boulders and glaciers of; South, no Ascension, plants of modern formations on west coast Asclepias, pollen of Ammonites, sudden extinction of Asparagus Anagallis, sterility of Aspicarpa Analog of variations Asses, striped Ancylus Ateuchus Animals, not domesticated from being variable; Audubon, on habits of frigate-bird; on variation in domestic; descended from several stocks; 434 Charles Darwin birds’-nests; on heron eating seeds of Madeira, Bermuda, and Galapagos; song of Australia, animals of; dogs of; extinct animals of; males; transporting seeds; waders; wingless; European plants in with traces of embryonic teeth Azara on flies destroying cattle Bizcacha, affinities of Azores, flora of Bladder for swimming in fish Blindness of cave animals Babington, Mr., on British plants Blyth, Mr., on distinctness of Indian cattle; on Balancement of growth striped Hemionus; on crossed geese Bamboo with hooks Boar; shoulder-pad of Barberry, flowers of Borrow, Mr., on the Spanish pointer Barrande, M., on Silurian colonies; on the succes- Bory St. Vincent, on Batrachians sion of species; on parallelism of palaeozoic Bosquet, M., on fossil Chthamalus formations; on affinities of ancient species Boulders, erratic, on the Azores Barriers, importance of Branchiae Batrachians on islands Brent, Mr., on house-tumblers; on hawks killing Bats, how structure acquired; distribution of pigeons Bear, catching water-insects Brewer, Dr., on American cuckoo Bee, sting of; queen, killing rivals Britain, mammals of Bees fertilizing flowers; hive; not sucking the red Bronn, on duration of specific forms clover; cell-making instinct; humble, cells of; parasitic Brown, Robert, on classification Beetles, wingless, in Madeira; with deficient tarsi Buckman, on variation in plants Bentham, Mr., on British plants; on classification Buzareingues on sterility of varieties Berkeley, Mr., on seeds in salt-water Bermuda, birds of Birds acquiring fear; annually cross the Atlantic; Cabbage, varieties of, crossed colour of, on continents; fossil, in caves of Brazil; Calceolaria 435 On the Origin of Species Canary-birds, sterility of hybrids Cirripedes capable of crossing; carapace aborted; Cape de Verde islands their ovigerous frena; fossil; larvae of Clift, Mr., on the succession of types Cape of Good Hope, plants of Climate, effects of, in checking increase of beings; Carrier-pigeons killed by hawks adaptation of, to organisms Cassini on flowers of composita Cobites, intestine of Catasetum Cockroach Cats, with blue eyes, deaf; variation in habits of; Collections, palaeontological, poor curlng tail when going to spring Colour, influenced by climate; in relation to attacks Cattle destroying fir-trees; destroyed by flies in La by flies Plata; breeds of, locally extinct; fertility of Indian Columba livia, parent of domestic pigeons and European breeds Colymbetes Cave, inhabitants of, blind Compositae, outer and inner florets of; male flow- Centres of creation ers of Cephalopodae, development of Conclusion, general Cervulus Conditions, slight changes in, favourable to fertil- Cetacea, teeth and hair ity Ceylon, plants of Coot Chalk formation Coral-islands, seeds drifted to; reefs, indicating Charlock movements of earth Checks to increase; mutual Corn-crake Chickens, instinctive tameness of Correlation of growth in domestic productions; of Chthamalinae growth Chthamalus, cretacean species of Cowslip Circumstances favourable to selection of domestic Creation, single centres of products; to natural selection Crinum Crosses, reciprocal 436 Charles Darwin Crossing of domestic animals, importance in Denudation, rate of; of oldest rocks altering breeds; advantages of; unfavourable to Development of ancient forms selection Devonian system Crustacea of New Zealand Dianthus, fertility of crosses Crustacean, blind Dirt on feet of birds Cryptocerus Division, physiological, of labour Dogs, hairless, with imperfect teeth; descended Ctenomys, blind from several wild stocks; domestic instincts of; Cuckoo, instinct of inherited civilisation of; fertility of breeds to- Currants, grafts of gether; of crosses; proportions of, when young Currents of sea, rate of Downing, Mr., on fruit-trees in America Downs, North and South Cuvier on conditions of existence; on fossil mon- Dragon-flies, intestines of keys; Fred., on instinct Drift-timber Driver-ant Dana, Prof., on blind cave-animals; on relations of Drones killed by other bees crustaceans of Japan, on crustaceans of New Duck, domestic, wings of, reduced; logger-headed Zealand Duckweed De Candolle on struggle for existence; on Dugong, affinities of umbelliferae; on general affinities; Alph., on low Dung-beetles with deficient tarsi plants, widely dispersed; on widely-ranging plants being variable; on naturalisation; on Dyticus winged seeds; on Alpine species suddenly becoming rare; on distribution of plants with Earl, Mr., W., on the Malay Archipelago large seeds; on vegetation of Australia; on fresh- Ears, drooping, in domestic animals; rudimentary water plants; on insular plants Degradation of coast-rocks Earth, seeds in roots of trees

437 On the Origin of Species Eciton Fishes, ganoid, now confined to fresh water; elec- Edentata, teeth and hair; fossil species of tric organs of; of southern hemisphere Flight, powers of, how acquired Edwards, Milne, on physiological divisions of Flowers, structure of; in relation to crossing; of labour; on gradations of structure; on embryo- composite and umbelliferae logical characters Forbes, E., on colours of shells; on abrupt range of Eggs, young birds escaping from shells in depth; on poorness of palaeontological Electric organs collections; on continuous succession of genera; Elephant, rate increase; of glacial period on continental extensions; on distribution during Existence, struggle for; conditions of glacial period; on parallelism in time and space Eye, structure of; correction for aberration Forests, changes in, in America Eyes reduced in moles Formation, Devonian Formations, thickness of, in Britain; intermittent Fabre, M., on parasitic sphex Formica rufescens; sanguinea; flava, neuter of Falconer, Dr., on naturalisation of plants in India; Frena, ovigerous, of cirripedes on fossil crocodile; on elephants and mastodons; Fresh-water productions, dispersal of and Cautley on mammals of sub-Himalayan Fries on species in large genera being closely beds allied to other species Falkland Island, wolf of Frigate-bird Faults Frogs on islands Fear, instinctive, in birds Fruit-trees, gradual improvement of; in United Feet of birds, young molluscs adhering to States; varieties of, acclimatised in United States Fertility of hybrids, from slight changes in condi- Fuci, crossed tions; of crossed varieties Fur, thicker in cold climates Fir-trees destroyed by cattle; pollen of Furze Fish, flying; teleostean, sudden appearance of; eating seeds; fresh-water, distribution of Galapagos Archipelago, birds of; productions of 438 Charles Darwin Galeopithecus naturalised plants in the United States; on rarity Game, increase of, checked by vermin of intermediate varieties; on Alpine plants; Dr. Gartner on sterility of hybrids; on reciprocal J.E., on striped mule crosses; on crossed maize and verbascum; on Grebe comparison of hybrids and mongrels Grouse, colours of; red, a doubtful species Geese, fertility when crossed; upland Growth, compensation of; correlation of, in domes- Genealogy important in classification tic products; correlation of Geoffroy St. Hilaire on balancement; on homologous organs; Isidore, on variability of repeated Habit, effect of, under domestication; effect of, parts; on correlation in monstrosities; on correla- under nature; diversified, of same species tion; on variable parts being often monstrous Hair and teeth, correlated Geographical distribution Harcourt, Mr. E.V., on the birds of Madeira Geology, future progress of Hartung, M., on boulders in the Azores Giraffe, tail of Hazel-nuts Glacial period Hearne on habits of bears Gmelin on Distribution Heath, changes in vegetation Gnathodon, fossil Heer, O., on plants of Madeira Godwin-Austin, Mr., on the Malay Archipelago Helix pomatia Goethe on compensation of growth Helosciadium Gooseberry, grafts of Hemionus, striped Gould, Dr. A., on land-shells; Mr., on colours of Herbert, W., on struggle for existence; on sterility birds; on birds of the Galapagos; on distribution of hybrids of genera of birds Hermaphrodites crossing Gourds, crossed Heron eating seed Grafts, capacity of Heron, Sir R., on peacocks Grasses, varieties of Heusinger, on white animals not poisoned by Gray, Dr. Asa, on trees of United States; on certain plants 439 On the Origin of Species Hewitt, Mr., on sterility of first crosses embryological succession; on homologous Himalaya, glaciers of; plants of organs; on the development of aphis Hippeastrum Hybrids and mongrels compared Holly-trees, sexes of Hybridism Hollyhock, varieties of, crossed Hydra, structure of Hooker, Dr., on trees of New Zealand Hooker, Dr., on acclimatisation of Himalayan trees; Ibla on flowers of umbelliferae; on glaciers of Icebergs transporting seeds Himalaya; on algae of New Zealand; on vegeta- Individuals, numbers favourable to selection; tion at the base of the Himalaya; on plants of many, whether simultaneously created Tierra del Fuego; on Australian plants; on rela- Insects, colour of, fitted for habitation; sea-side tions of flora of South America; on flora of the colours of; blind, in caves; luminous; neuter Antarctic lands; on the plants of the Galapagos Instincts, domestic Hooks on bamboos; to seeds on islands Intercrossing, advantages of Horner, Mr., on the antiquity of Egyptians Horns, rudimentary Japan, productions of Horse, fossil, in La Plata Java, plants of Horses destroyed by flies in La Plata; striped; proportions of, when young Jones, Mr. J.M., on the birds of Bermuda Horticulturists, selection applied by Jussieu on classification Huber on cells of bees; P., on reason blended with instinct; on habitual nature of instincts; on slave- Kentucky, caves of making ants; on Melipona domestica Humble-bees; cells of Kerguelen-land, flora of Hunter, J., on secondary sexual characters Kidney-bean, acclimatisation of Hutton, Captain, on crossed geese Kidneys of birds Huxley, Prof., on structure of hermaphrodites; on 440 Charles Darwin Kirby on tarsi deficient in beetles Lubbock, Mr., on the nerves of coccus Knight, Andrew, on cause of variation Lucas, Dr. P., on inheritance; on resemblance of child to parent Kolreuter on the barberry; on sterility of hybrids; Lund and Clausen on fossils of Brazil on reciprocal crosses; on crossed varieties of Lyell, Sir C., on the struggle for existence; on nicotiana; on crossing male and hermaphrodite modern changes of the earth; on measure of flowers denudation; on a carboniferous land-shell; on fossil whales; on strata beneath Silurian system; Land-shells, distribution of, of Madeira, on the imperfection of the geological record; on naturalised the appearance of species; on Barrande’s colo- Languages, classification of nies; on tertiary formations of Europe and North Lapse, great, of time America; on parallelism of tertiary formations; Larvae on transport of seeds by icebergs; on great Laurel, nectar secreted by the leaves alternations of climate; on the distribution of Laws of variation fresh-water shells; on land-shells of Madeira Leech, varieties of Lyell and Dawson on fossilized trees in Nova Leguminosae, nectar secreted by glands Scotia Lepidosiren Life, struggle for Macleay on analogical characters Lingula, Silurian Madeira, plants of; beetles of, wingless; fossil land- Linnaeus, aphorism of shells of; birds of Lion, mane of; young of, striped Lobelia fulgens Magpie tame in Norway Lobelia, sterility of crosses Maize, crossed Loess of the Rhine Malay Archipelago compared with Europe; mam- Lowness of structure connected with variability mals of Lowness, related to wide distribution Malpighiaceae

441 On the Origin of Species Mammae, rudimentary Moles, blind Mammals, fossil, in secondary formation; insular Mongrels, fertility and sterility of; and hybrids Man, origin of races of compared Monkeys, fossil Manatee, rudimentary nails of Monocanthus Marsupials of Australia, fossil species of Mons, Van, on the origin of fruit-trees Martens, M., experiment on seeds Mozart, musical powers of Martin, Mr. W.C., on striped mules Mud, seeds in Mules, striped Matteuchi on the electric organs of rays Muller, Dr. F., on Alpine Australian plants Matthiola, reciprocal crosses of Murchison, Sir, R., on the formations of Russia; on Means of dispersal azoic formations; on extinction Melipona domestica Mustela vison Myanthus Metamorphism of oldest rocks Myrmecocystus Mice destroying bees; acclimatisation of Myrmica, eyes of Migration, bears on first appearance of fossils Miller, Prof., on the cells of bees Nails, rudimentary Natural history, future progress of Mirabilis, crosses of Naturalisation of forms distinct from the indig- Missel-thrush enous species; in New Zealand Misseltoe, complex relations of Nautilus, Silurian Mississippi, rate of deposition at mouth Nectar of plants Mocking-thrush of the Galapagos Nectaries, how formed Modification of species, how far applicable Nelumbium luteum

442 Charles Darwin Nests, variation in organs; on the metamorphosis of cephalopods Neuter insects and spiders New Zealand, productions of, not perfect; Pacific Ocean, faunas of naturalised products of; fossil birds of; glacial Paley, on no organ formed to give pain action in; crustaceans of; algae of; number of Pallas, on the fertility of the wild stocks of domes- plants of; flora of tic animals Nicotiana, crossed varieties of; certain species very Paraguay, cattle destroyed by flies sterile Parasites Noble, Mr., on fertility of Rhododendron Partridge, dirt on feet Nodules, phosphatic, in azoic rocks Parts greatly developed, variable, degrees of utility of Oak, varieties of Parus major Onites apelles Passiflora Orchis, pollen of Peaches in United States Ornithorhynchus Pear, grafts of Ostrich not capable of flight; habit of laying eggs Pelargonium, flowers of, sterility of together; American, two species of Otter, habits of, how acquired Peloria Ouzel, water Pelvis of women Owen, Prof., on birds not flying; on vegetative Period, glacial repetition; on variable length of arms in ourang- Petrels, habits of outang; on the swim-bladder of fishes; on electric organs; on fossil horse of La Plata; on rela- Phasianus, fertility of hybrids tions of ruminants and pachyderms; on fossil Pheasant, young, wild birds of New Zealand; on succession of types; Pictet, Prof., on groups of species suddenly ap- on affinities of the dugong; on homologous pearing; on rate of organic change; on continu-

443 On the Origin of Species ous succession of genera; on close alliance of Primula, varieties of fossils in consecutive formations; on embryo- Proteolepas logical succession Proteus Pierce, Mr., on varieties of wolves Psychology, future progress of Pigeons with feathered feet and skin between toes; breeds described, and origin of; breeds of, how Quagga, striped produced; tumbler, not being able to get out of Quince, grafts of egg; reverting to blue colour; instinct of tum- Rabbit, disposition of young bling; carriers, killed by hawks; young of Race-horses, Arab; English Pistil, rudimentary Races, domestic, characters of Plants, poisonous, not affecting certain coloured Ramond on plants of Pyrenees animals; selection applied to; gradual improve- Ramsay, Prof., on thickness of the British forma- ment of; not improved in barbarous countries; tions; on faults destroyed by insects; in midst of range, have to Ratio of increase struggle with other plants; nectar of; fleshy, on Rats, supplanting each other; acclimatisation of; sea-shores; fresh-water, distribution of; low in blind in cave scale; widely distributed Rattle-snake Plumage, laws of change in sexes of birds Reason and instinct Plums in the United States Recapitulation, general Pointer dog, origin of; habits of Reciprocity of crosses Poison not affecting certain coloured animals Poison, similar effect of, on animals and plants Record, geological, imperfect Pollen of fir-trees Rengger on flies destroying cattle Poole, Col., on striped hemionus Resemblance to parents in mongrels and hybrids Prestwich, Mr., on English and French eocene Reversion, law of inheritance formation Primrose; sterility of Rhododendron, sterility of 444 Charles Darwin Richard, Prof., on Aspicarpa Sexual characters variable; selection Richardson, Sir J., on structure of squirrels; on Sheep, Merino, their selection; two sub-breeds fishes of the southern hemisphere unintentionally produced; mountain, varieties of Robinia, grafts of Shells, colours of, littoral, seldom embedded; Rodents, blind fresh-water, dispersal of; of Madeira; land, Rudimentary organs distribution of Rudiments important for classification Silene, fertility of crosses Silliman, Prof., on blind rat Sageret on grafts Skulls of young mammals Salmons, males fighting, and hooked jaws of Slave-making instinct Salt-water, how far injurious to seeds Smith, Col. Hamilton, on striped horses; Mr. Fred., Saurophagus sulphuratus on slave-making ants; Mr., of Jordan Hill, on Schiodte on blind insects degradation of coast-rocks Schlegel on snakes Snap-dragon Sea-water, how far injurious to seeds Somerville, Lord, on selection of sheep Sebright, Sir J., on crossed animals; on selection of Sorbus, grafts of pigeons Spaniel, King Charles’s breed Sedgwick, Prof., on groups of species suddenly Species, groups of, suddenly appearing; beneath appearing Silurian formations; successively appearing; Seedlings destroyed by insects changing simultaneously throughout the world Seeds, nutriment in; winged; power of resisting Spencer, Lord, on increase in size of cattle salt-water; in crops and intestines of birds; eaten Sphex, parasitic by fish; in mud; hooked, on islands Selection of domestic products, principle not of Spiders, development of recent origin; unconscious; natural; sexual; Spitz-dog crossed with fox natural, circumstances favourable to Sports in plants Sexes, relations of 445 On the Origin of Species Sprengel, C.C., on crossing; on ray-florets Thuret, M., on crossed fuci Squirrels, gradations in structure Thwaites, Mr., on acclimatisation Staffordshire, heath, changes in Tierra del Fuego, dogs of; plants of Stag-beetles, fighting Timber-drift Sterility from changed conditions of life; of hy- Time, lapse of brids; laws of; causes of; from unfavourable Titmouse conditions; of certain varieties Toads on islands St. Helena, productions of Tobacco, crossed varieties of St. Hilaire, Aug., on classification Tomes, Mr., on the distribution of bats St. John, Mr., on habits of cats Transitions in varieties Sting of bee Trees on islands belong to peculiar orders; with Stocks, aboriginal, of domestic animal separated sexes Strata, thickness of, in Britain Trifolium pratense; incarnatum Stripes of horses Trigonia Swallow, one species supplanting another Trilobites; sudden extinction of Swim-bladder Troglodytes Tucutucu, blind Tail of giraffe; of aquatic animals; rudimentary Tarsi deficient Tumbler pigeons, habits of, hereditary; young of Tausch on umbelliferous flowers Turkey-cock, brush of hair on breast Teeth and hair correlated; embryonic, traces of, in Turkey, naked skin on head; young, wild birds; rudimentary, in embryonic calf Turnip and cabbage, analogous variations of Tegetmeier, Mr., on cells of bees Temminck on distribution aiding classification Type, unity of Thouin on grafts Types, succession of, in same areas Thrush, aquatic species of; mocking, of the Galapagos; young of, spotted; nest of Udders enlarged by use; rudimentary 446 Charles Darwin Ulex, young leaves of Waterhouse, Mr., on Australian marsupials; on Umbelliferae, outer and inner florets of greatly developed parts being variable; on the Unity of type cells of bees; on general affinities Use, effects of, under domestication Water-ouzel Utility, how far important in the construction of Watson, Mr. H.C., on range of varieties of British each part plants; on acclimatisation; on flora of Azores; on Alpine plants; on rarity of intermediate varieties Valenciennes on fresh-water fish Weald, denudation of Variability of mongrels and hybrids Web of feet in water-birds Variations appear at corresponding ages; analo- West Indian islands, mammals of gous in distinct species Westwood on species in large genera being closely Varieties, natural; struggle between; domestic, allied to others; on the tarsi of Engidae; on the extinction of; transitional; when crossed, fertile; antennae of hymenopterous insects when crossed, sterile; classification of Whales, fossil Verbascum, sterility of; varieties of, crossed Wheat, varieties of Verneuil, M. de, on the succession of species White Mountains, flora of Viola tricolor Wings, reduction of size Volcanic islands, denudation of Wings of insects homologous with branchiae; Vulture, naked skin on head rudimentary, in insects Wolf crossed with dog; of Falkland Isles Wading-birds Wollaston, Mr., on varieties of insects; on fossil Wallace, Mr., on origin of species; on law of geo- varieties of land-shells in Madeira; on colours of graphical distribution; on the Malay Archi- insects on sea-shore; on wingless beetles; on pelago rarity of intermediate varieties; on insular in- Wasp, sting of sects; on land-shells of Madeira, naturalised Water, fresh, productions of Wolves, varieties of Water-hen Woodpecker, habits of; green colour of 447 On the Origin of Species Woodward, Mr., on the duration of specific forms; on the continuous succession of genera; on the succession of types World, species changing simultaneously throughout To return to the Wrens, nest of Youatt, Mr., on selection, on sub-breeds of sheep; Charles Darwin page, on rudimentary horns in young cattle go to

Zebra, stripes on http://www.hn.psu.edu/ faculty/jmanis/darwin.htm

To return to the Electronic Classics Series page, go to

http://www.hn.psu.edu/ faculty/jmanis/jimspdf.htm

448