1

Ch. 8

DIFFICULTIES ON THE THEORY OF NATURAL SELECTION IN RELATION TO PASSAGES FROM FORM TO FORM 2

[Completed September 29, 1857]

In the sixth chapter I briefly alluded to many grave difficulties, enough at first sight to overwhelm our theory of natural selection. In this chapter we will consider those connected with the absolute necessity of all passages having been extremely gradual from one living being into another, or of one part or organ into another. That this is absolutely necessary follows from all causes of variation apparently acting only slowly, & more especially from our paramount principle of natural selection inevitably thus acting by the addition of numerous slight modifications,—each modification profiting the selected or preserved individual. I fully admit that the difficulties in relation to these passages are many, & in some instances extremely great; & I wish here only to consider whether the difficulties are actually insuperable & enough by themselves to overthrow our theory, whether or not other facts support it.

1 CD. MSS., 114, letter no. 210.

2 [Darwin noted in pencil at top of this sheet:' Be careful in use of word transitional as = intermediate']

2

We will commence with cases of intermediate & possibly transitional habits. It has been asked how for instance a land carnivorous quadruped could be converted into an otter; for how could it have subsisted during its transitional state? A far more difficult question would have been, how could a Bat have lived, before its wings had become perfected? <by long continued natural selection?> I may premise that it is immaterial to us, whether slight changes of structure supervene first leading to changed habits, or whether habits & instinct change slightly first, & the animal is benefitted in subsequent generations by slight selected modifications of structure in relation to the already slightly changed habits. In the chapter on instincts we shall see that mental attributes vary slightly, can be selected & are inherited, like corporeal structures.

Let us take the case of the Bat, which is one of the most difficult, that has

3

occurred to me. What were the stages by which probably an insectivorous & terrestrial animal could have acquired the capacious wings of the Bat;—every single, slight, intermediate grade being so useful to the animal in that state, that it was enabled to conquer in the struggle for life, to which it must inevitably have been exposed? We cannot answer this question even by conjectures. The earliest known, Eocene Bat apparently was as perfect, as one of the present day: if our geological records really make any approach to perfection, this would be a fatal objection. But the whole subject of the value of the evidence from fossil remains will have to be discussed in a separate chapter; & I will here pass by this apparently fatal objection, applicable in many other instances, & consider our more immediate subject, namely the possibility of transitional stages between a Bat & an animal not capable of flight. On this diffculty

4

I think we ought to be extremely cautious in laying much stress.—

Look how amongst Birds, the most perfectly winged animals, we have the Penguin, which uses its wings exclusively as paddles for diving, & as front legs on the land; (as I have witnessed),—the logger-headed Duck (Micropterus brachypterus Eyton) as flappers on the surface of the water, & never as wings,—the ostrich as sails—& the Apterix is destitute of wings capable of any use. Yet all these Birds are enabled to hold their own place in the great struggle for life; & no one will doubt that their wings are most useful to the Penguin, & Logger-headed Duck, & I presume of some use to the ostrich perhaps in its first start to escape a beast of prey. Many birds use their wings as paddles for diving & for flight, others flap along the surface of the water or run with expanded wings before taking flight: is it not thus conceivable that by continued selection organs used exclusively for diving or flapping or sailing on land might come to serve exclusively for flight? But had some half-dozen genera of Birds become utterly extinct, he would have been a bold man, who would have said

5

that Birds might have flourished on our earth which did not fly, but used their wings solely as sails, flappers, paddles, or front legs. I am, of course, far from pretending to indicate what were the transitional grades, by which Birds came to fly; it might have been through some wholly distinct line of change: the penguin may be, as the logger-headed Duck probably is, the degraded descendant of a perfectly winged Bird. All that I want to show is, that as far as habits are concerned & judging only from now living animals, great transitions are possible.

Seeing that we have flying Birds Mammals, & formerly flying Reptiles, & seeing that so eminently an aquatic animal as a crab can by a contrivance to keep its branchiae moist exist on dry land, it is conceivable, that the so-called flying Fish,—which can glide to such great distances through the air, turning & rising slightly in its rapid course by the aid of its fluttering fin1 —could have been converted into a perfectly flying animal: had this been the case, & our present flying-fish/6/unknown, who would have ventured to have even conjectured that the sole use of an early transitional stage of the pectoral fin was to escape danger in the open ocean. Certain fish use their pectoral fins for splattering over the mud, for jumping & even for climbing trees; if fish had become, like land-crabs & onisci, terrestrial animals, how easily ancient transitional uses of the pectoral fins, might have baffled all conjecture.

Amongst mammals, we have squirrels with the tail forming a flattened brush, & we have others

5v

* Owen's Lecture Fishes, p 170.

6

There are, also, gliding insectivorous opossums with the flank-membrane developed in different degrees. In the Galeopithecus, or flying Lemur, which was formerly ranked amongst Bats, the membrane extends from the corner of the jaw & includes all four legs & the tail: the membrane on the flank has a muscle for extending it: the rather long fingers of all four hands are also connected by skin: its habits are imperfectly known, but it is said to descend trees "par une sorte de vol

6 insert

(a) text

with "a peculiar wideness in the posterior part of the body & a fulness of skin of the flanks being an approach to the forms of a true flying squirrel1 & these latter have a wide membrane connecting the front & hind legs together & in one species a slight fold of skin uniting the base of the tail to the hinder thighs. All these contrivances aiding the animal to glide great distances through the air from the top of one tree to the base of another.

7

retarde."2 <& to be even partly aquatic in its habits.>The fact that each animal lives by a struggle,—that each would increase inordinately if not checked at some period of its life,—is constantly eluding us; so that we find it difficult to realise that in course of thousands of generations the power of gliding a few inches further through the air may make an important difference to an animal in escaping dangers or getting food. For myself I can see no difficulty in the means of gliding through the air in squirrels having been perfected through natural selection from a mere flattened brush-like tail to a wide flank membrane; & amongst Lemurs (though all such supposed intermediate forms are extinct) to the enormously developed membrane of the living Galeopithecus; & <in some other unknown & extinct tribes of animals,> even amongst other extinct animals to the wonderfully perfect wings of the Bat. The graduated structure amongst squirrels & the almost intermediate condition of the Galeopithecus between an aerial & terrestrial animal ought at least to

2 Dict. Class. d'Hist. Nat. Tom. 7. p. 122. [Desmoulins, art.' Galéopithèque Galeopithecus'] Buckland Bridgewater Treatise.

8

make us very cautious in supposing that numerous animals constructed on every intermediate type between a Bat & land quadruped, could not formerly have flourished in the great battle of life. Who would have ever supposed (a)

To return to the objection which has actually been made that a land-carnivorous animal could not be changed into an Otter, for it could not live during the transitional state. The genus Mustela is closely allied to Lutra or the otter, & indeed was made into one by Linnaeus. Some species of Mustela occasionally haunt the water, & the common Polecat has been known to lay up stores of half-killed frogs; the N. American Vison-Weasel (Mustela vison allied to the M. lutreola of N. Europe) has webbed feet, a flattened head, short ears, close fur & a tail all like an otter: it can dive well, & preys on fish: but during the winter when the water is frozen, it hunts mice on the land:2 here then we have an animal allied to the otter, wholly aquatic during part of the year & partly terrestrial during another part. Can it, then, be said that there would be any great difficulty, as far

8v

[a deleted start to folio 8]

Insert (D)

Sir J. Richardson Fauna Boreali-Americana Quadrupeds, p. 49.

8v insert

that at the present day there should be a Bat feeding chiefly on frogs & occasionally on fish; or that the frugivorous Pteropus, when put on a floating raft, should take to the water & "swim pertinaceously after a boat."

* Mr. Blyth gives an account of these habits in the Megaderrna lyra in India, in Annals & Mag. of Nat. History.vol. 15.—1845. p. 463.

* Mr. Lay on a Pteropus from the Bonin Islands in Zoological Journal vol iv. 1829. p 459.

9

as transitional habits are concerned, in converting a polecat into an otter. The possibility will rest on there being a place open in the polity of nature, which would allow of a polecat living & increasing in numbers, if rendered more & more aquatic in habits & structure. On the same principles an otter could be converted into a seal-like animal; not, perhaps, now when seals actually exist & well fill their place in nature, but before a seal had been formed. It might well happen through natural selection, that an aquatic animal should be converted into a terrestrial animal, retaining perhaps a trace of its former webbed feet; & subsequently have some of its descendants refitted to inhabit the waters.

Numerous instances could be given, like that of the Vison-weasel, of diversified habits in the same individuals or in the same species both under the ordinary conditions of its life & under peculiar circumstances; & similar differences could be given amongst different, but closely allied species of the same genus. I will give only a very few cases chiefly amongst birds. In S. America,

10

a tyrant-flycatcher3 may often be seen hovering over one spot & then proceeding to another like a Hawk, but its stoop is very inferior in force: at other times it remains motionless on the edge of a piece of water, & like a Kingfisher dashes at any small fish near the margin. In Tierra del Fuego, there is an Owl (Ulula rufipes) which I have reason to believe preys chiefly on the lower marine animals, & I have seen its stomach gorged with the remains of large crabs. The Woodpecker with its peculiar feet, stiff tail, strong wedge-like beak & long tongue, has often & justly been adduced, as a perfect instance of the adaptation of a bird to prey on insects concealed in the bark & wood of trees: but on the wide grassy plains of La Plata, where not a tree exists, the Picus (Chrysoptilus) campestris Licht. feeds exclusively on the ground: even in its colouring in the peculiar undulating flight, & loud cry it resembles pretty closely our common green species.1 A North American woodpecker2 has the extraordinary propensity of catching

10v

*a Zoology of the Voyage of the Beagle p. 113.

* Picus varius. see Mr. T. Macculloch in the Boston Journal of Nat. History vol 4. p. 406.

11

flies on the wing! & some other N.American species feed largely on fruit. In our own country the Titmouse genus (Parus) are properly insectivorous; yet everyone may continually hear in autumn, the loud hammering, of the Parus major, like that made by the especially adapted Nuthatch, as it breaks with its beak the kernels of the yew-berries held on a branch of a neighbor tree. One more instance,—Hearn[e] states that the black bear fishes for small crustaceans in the sea, by swimming about with its mouth widely open, so that here a terrestrial quadruped almost mocks a whale in its occasional manner of getting food !3

<In such cases, if under changing conditions one of the diversified lines of life were especially favoured, it does not seem very difficult to believe, that the structure of the descendants from a parent form having very different habits might become greatly modified through natural selection.>

12

When an animal or plant is introduced into a new country, and vast numbers have been thoroughily naturalised, some slight changes in its habits of life can hardly fail often to occur; & likewise indirectly in some of the aborigenes of the country. Look at our many introduced quite new plants, how they are preyed on by our native animals, sometimes almost to the exclusion of their original food.4 Within two years after planting Berberis dulcis a bush very unlike the common barberry I found its twigs covered with Aphides on which Coccinellae were preying; its flowers were visited by Bees, visits indispensable, I believe, to its fertilisation, & its fruit was devoured by the Robin, which would disseminate its seed. In Tasmania, I found the dung of the introduced quadrupeds, so different from that of any native animal, supporting numerous beetles; & this was likewise the case even in the island of St. Helena,1 where there was no native quadruped; yet Entomologists know that stercovorous beetles are usually restricted in their habits. Innumerable

12v

* Loudon's Magazine of Nat. History vol. v 1832 p. 154. Thus the caterpillar of the Death's Head sphinx is very rarely found except on the Potato or jasmine.

* Darwin Journal of Researches p. 490.

13

parallel instances could be given.

The changes produced by civilised men in many countries must have sensibly affected the habits of life of the native animals & plants. How many insects there are in Britain, which, as far as known, subsist exclusively on artificial or foreign substances. There are many egregious plants, which are scarcely ever seen except on cultivated land, though probably most of them are foreigners.

Thoroughily wild animals of the same species, when inhabiting districts of a very different nature can hardly fail to differ somewhat in habits. Some land birds are common to Tierra del Fuego, absolutely covered by inpenetrable forests, & the Falkland Islands, where not a tree exists. In our own country, the Scotch Highland Fox has considerably different habits from the lowland Fox of England. Even with our domestic animals, individuals show a tendency to different habits of life,—of which I/

14

have already given instances in the Cat. Very many (a) In all cases of naturally diversified or of changed habits, I can see no great difficulty in structure being modified through the natural selection of variations better fitted to some one of the diverse habits, or to a new & changed habit of life. If our reason tells us that the structure of the woodpecker is admirably adapted for its insectivorous life on trees, & that another structure is generally best for the capture of insects on the ground or in the air, then I can see no reason why the ground or aerial woodpecker might not have their structure still more modified, than it actually is, from the typical structure of the genus Picus, & so be improved for their aberrant habits of life; or on the other hand why birds having in some degree the habits of these aberrant woodpeckers might not be modified in structure till they acquired the strictly arboreal habits of the typical woodpeckers. To take one more extreme case, that of the Black Bear seen by Hearn: if its habit of catching small crustaceans by swimming with widely open mouth

14 insert

cases are on record of the inhabitants of the sea gradually becoming accustomed to brackish & even fresh water; & of the reverse change; which circumstance, considering how fatal an abrupt change of this nature is to most animals is surprising.*

* See Bronn. Ges[ch]ichte der Natur B 2. p 55-58.—Macculloch Highlands & W. Islands, [IV] p. 377. Many other references could be easily given.

15

became, from the crustaceans being always present with the loss perhaps of other prey, highly important for its sustenance, then as slight variations in the size & shape of the mouth would almost certainly occur during millions of generations, some of these would almost certainly aid ever so little individuals in this strange way of fishing; & these individuals, as all cannot possibly live, would have a better chance of living, & thus such slight variations would be continually added up through natural selection, till an animal, which we should think monstrous, was produced, thoroughily aquatic & with a mouth perhaps proportionally as large as that of a whale. Who would not think it monstrously improbable, if he had never heard of a whale, that so gigantic an animal could subsist by sifting with its huge mouth the minutest animals from the waters of the sea!

Facts do not tell us, as far as I can see, whether habits generally change first, corresponding structures being

16

subsequently selected; or whether structures modified through variation, generally first leads to perhaps <nearly simultaneously> changed habits. In the case of no organic being can we pretend to conjecture through what exact lines of life its progenitors have passed. We may use our knowledge of the habits of existing animals as a guide to conjecture; and somewhat further, in as much as it is probable that amongst the many living & greatly diversified descendants of some ancient & extinct form, some would retain the habits not greatly modified of their several progenitors at different stages of descent. Really to know the transitional states, by which the habits & structure of any one animal have been acquired, we ought to study the long line of its direct ancestors alone, & neglect all collateral branches. How little chance there can be, of one ever knowing, even very imperfectly, all the lineal ancestors of any one form will be seen, when we come to consider the real poverty, under the guise of richness, of our geological records.

It may, however, be here noticed, that when the habits of life of

17

any species or group of animals, is undergoing any great change, as from swimming or gliding to flying, although assuredly every single transitional form has to subsist under a most severe competition for life, yet we ought not to expect the new habit with its corresponding structures, to be developed under many subordinate forms, each with numerous individuals, until it had arrived at a high degree of perfection. To arrive at this, according to our principles of natural selection an enormous period of time would be required. On our former imaginary case of Fish being rendered true flyers, we could hardly expect that they would give rise to a whole class of subordinate groups, fitted for various subordinate stations, until they had obtained through the slow action of natural selection the power of flying perfectly; the intermediate & transitional states, we might expect, would be comparatively few in number at any one period, as we see at the present day with our so-called flying-fish. <So it would probably have been with Lemurs,—the

18

Galeopithecus being looked at as a transitional form—if Bats had not existed, & Lemurs had been developed by natural selection into true flyers.> Hence it seems probable that mere transitional states between very different lines of life, would seldom be largely developed at any one period; for this would not happen till the changing form—changing from having some advantage over its compatriots—could fill its new place in nature with a high degree of perfection. The perfected descendants would generally cause, by the very principle of natural selection or the struggle for life, the extinction of their less perfect progenitors. Hence, also, the chance of finding fossil remains of the progenitors of any organism, during its transitional & less perfect state, would be so much less in proportion as they had been developed under fewer subordinate forms & under fewer individual numbers.

19

We have as yet considered only the possibility of transitional habits, & the difficulties which they seem to oppose to our theory; but some of these same facts may be fairly viewed as supporting our theory. He who believes that each species has been independently created, must feel surprise, at least I remember formerly having felt great surprise, at an animal manifestly adapted for one line of life, following another & very different line. I will take again my illustration from Birds. It cannot be doubted that the general configuration of a Goose is for an aquatic life; & the meaning of webbed feet is unmistakeable; but there are long-legged geese, (a) in S.America the Upland Goose (Chloephaga Magellanica Eyton)1 never frequents the water, except for a short time after hatching for the protection of its young; the feet of this goose are well webbed. The long-legged Flamingo (Phoenicopterus) has webbed feet, but lives on marshes & is said seldom even to wade except in very shallow water. The Frigate-bird (Fregata aquila) with its extremely short legs, never1

19v

*xx Zoology of the Voyage of the Beagle Birds p. 134.—Capt. Sulivan has given me further particulars on the habits of these birds.—

19 insert

(a) text
which run like gallinaceous birds, & seldom or never enter the water: thus Mr. Gould informs me he believes that the Cereopsis goose of Australia is perfectly terrestrial, & I am told at the Zoological Gardens that this bird & the Sandwich Island goose seem quite awkward on the water:

20

alights on the water, but picks up its prey from the surface with wondrous skill; yet its four toes are all united by a web.2 The web, however, is considerably hollowed out between the toes; & tends to be rudimentary: as is the case likewise with the Cereopsis goose: I notice this, because it connects our present class of facts with rudimentary organs, hereafter to be discussed: in the fore-going cases the function of the webbed feet may be said to have become rudimentary without a corresponding change in structure.

On the other hand there does not exist a more thoroughily aquatic Bird, than the Grebe (Podiceps), but its toes are only widely bordered by membrane. The Water-Hen (Gallinula chloropus) may be constantly seen swimming about & diving with perfect ease; yet its long toes are bordered by the narrowest fringe of membrane; other closely allied birds belonging to genera Crex, Parra &c can swim well & yet have scarcely any trace of web; & their extremely long toes seem admirably adapted to walk over the softest swamp & floating plants; yet the common corncrake (Crex pratensis)

20v

Mr. Westwood, (Modern Class, of Insects. vol 2. p. 272) has remarked with surprise that certain parasitic Bees, which have no use for their jaws, have these tools as strong, as in working species. Again Mr. F. Smith showed me specimens of another hymenopterus insect, Scolia, in which the legs are eminently & typically fossorial, but which from being parasitic certainly does not use its legs for burrowing. [Query added in pencil: 'How in the Drone Hive Bee?']

21

belonging to one of these very genera, having the same structure of feet, haunts meadows & is scarcely more aquatic than a quail or partridge.

Several of the cases already given, such as that of the Ground-woodpecker may be looked at under our present point of view: I may add that on the plains, inhabited by these woodpeckers, I saw parrots (Conurus Patachonicus) which build in banks & which can never alight on a tree as one does not exist, & yet have their feet, strictly adapted for perching with two toes in front & two behind: on these plains there were also tree-frogs,3 with their toes enlarged into little sucking discs for climbing. Numerous other instances could be given; in the S. United States, the Swallow-tailed Hawk, (Nauclerus furcatus), a true hawk, has very long wings & a forked tail, & it lives, by catching whilst on the wing, insects. In the quiet creeks of Tierra del Fuego, I was particularly struck

21v

* Hyla agrestis: Bell, in Zoology of Beagle Reptiles. p. 46.

22

with the habits of the Puffinuria (Pelecanoides) Berardi: this bird in its manner of swimming, of flying, though rarely in a straight line by the rapid beating of its short wings, then drop-ping suddenly as if struck dead & diving to a suprising distance, would by anyone be mistaken for an auk or grebe: but the structure of its nostrils & beak & other characters show that it undoubtedly is one of the Petrels,—those most aerial of Birds which hunt the surface of the wide ocean for their prey: here then we see a bird taken from a family having most widely different habits, adapted to fill the place of the Auk of the Northern hemisphere, which are not found in the south.

All such facts must seem strange, as long as we look at each species as independently created: it will be said that a bird belonging to one type of structure has been adapted by the Creator to another line of life; but this seems to me only

22v

In this case there has been a considerable change in the form of the body & power of respiration; the wings have been greatly shortened, the tail altered in shape & hind toe lost. I will now give one more instance of an entire change in habits, with no sensible correlated change in structure;—I refer to the Water-Owzel (Cinclus aquaticus), a member of the common Thrush family; it is sub-aquatic in its habits, using its wings for diving, & its feet for grasping stones under the water; & yet the acutest observer would never have foretold this singular manner of life from the most careful examination of its structure.

23

restating the case in dignified language. The theory of natural selection implies that every single animal in each region tends to increase in number with a geometrical power, & so may be said to strive to gain subsistence anyhow it can, & to fill any place in the economy of nature which it can seize: bearing in mind the many & complicated contingencies to which each animal must be exposed in the long course of its existence, & remembering that the world is not open from end to end for immigration, (as we see proved by the many productions naturalised by man's aid) it seems to me perfectly natural on our theory, that occasionally an animal of a wholly different class or occupation should intrude into that of another species or group of species which laboured under some disadvantage however slight. How far its structure would become modified in relation to its new habits, would depend on how far any change would be advantageous to it, & whether variations in the right direction had occurred, & on how long a time selection had been at work accumulating such variations. On these principles, it

24

is not surprising that there should be webbed geese living & run-ning on the dry land, & webbed Frigate-birds never alighting on the water,—that there should be woodpeckers & tree-frogs where there is not a tree,—that corn-crakes should live in meadows instead of swamps—that there should be diving thrushes, & petrels with the habits of auks.—

Differing Organs of extreme perfection & complication. Although in our consideration of the possibility of great changes in habits corresponding changes in the whole bodily structure have generally been implied; yet it will be adviseable to look at some special cases of particular organs. What shall we say of the eye? Is it conceivable that this transcendant organ with its power of adjusting its focus to different distances & of letting in more or less light,—with its nearly perfect correction for chromatic & spherical aberration, could have been formed by the accumulation, through natural selection, of infinitesimally

25

slight variations, each useful to its possessor. I confess that no language at first seems too strong to condemn the absurdity of such a notion.—

To judge our theory according to its own principles, we ought not to compare the variously perfected eyes in any one group, one with another; but the eye of each species only with the eyes of all its lineal progenitors; so that if we look to the eyes of many species in the group, we should have to look to many lines of ascent converging up to one common parent. This is impossible; & all that we can do, is to look at the eyes of all existing animals within each great class, as a guide for judging how far a transition from one stage of perfection to another stage is possible; at the same time never forgetting how small a fraction the living are compared with the extinct,—almost infinitely small, as I believe. Let us briefly consider the eyes of the Articulata: we have as the

26

lowest grade, an optic nerve, coated with pigment sometimes having a kind of pupil, but without a lens or any other optical mechanism. I need hardly say that we have in this work nothing to do with the origin of nervous sensibility to light, any more than with the origin of life. From this rudimentary eye, <as it must be called,> which cannot possibly distinguish figures, & can only perceive light from darkness, there is an advance towards perfection by two fundamentally different contrivances. Firstly, stemmata (or the so-called "simple eyes") which have a crystalline lens, with a cornea & more or less perfect vitreous body—that is the essential parts of the eyes of the higher animals—, & which act by the rays from each point of the object viewed converging on different points of the retina. Secondly "compound eyes", formed of numerous, diverging, transparent, narrow cones, separated by pigment, & which act simply by excluding all the rays from each point of the object viewed, except the pencil which comes in a line perpendicular to the convex retina; so that

27

a separate & distinct image of each separate point of the object is made at the base of each separate transparent cone. Hence Müller1 the discover[er] of this principle of vision, calls these compound eyes, a mosaic dioptric instrument. In the Articulata we have numerous grades of perfection in the eyes: besides endless difference in form proportion & position of the transparent cones, & in number up to 20,000 in a single eye, there are cases, as in Meloe, in which the facets of the cornea are "slightly convex both externally & internally, that is lens-shaped": in many crustaceae there are two corneae, the external smooth, & the internal divided into facets, within the substance of which, "renflemens lenticulaires paraissent s'etre developpés"; but sometimes these lenses can be detached in a layer distinct from the cornea. The transparent cones are usually attached to the cornea, but not rarely they are detached & have their free ends rounded, &

28

in this case they must act, I presume, as converging lenses, & not simply as tubes excluding all oblique rays. Prof. Milne Edwards thinks that the transparent cones of the compound eyes are homologous with the crystalline lenses of the stemmata or simple eyes; & that behind the transparent cones there is apparently a vitreous substance: on this view the lenses in & beneath the cornea of the compound eye is a structure superadded to that observed in the stemmata. Altogether Muller divides the compound eyes into the three main classes with seven sub-divisions of structure: he makes a fourth main class of "aggregates" of stemmata or simple eyes each of which contains the essential parts of the simple stemmata, namely a lens & globular vitreous humour; & he adds "this is the transition form between mosaic-like compound eyes, unprovided with concentrating apparatus, & the organ of vision with such apparatus."

Seeing the numerous gradations & diversity in the eyes of the Articulata, numbering probably at least a hundred thousand in kind, & that

29

the eye of each is good for its habits, then if the eye varies ever so little, & I know of no reason to doubt this, I can see no great difficulty in believing that amongst the Articulata, the eye might be perfected through natural selection from a simple optic nerve to the most complex of compound eyes having numerous transparent cones, a double cornea &, the inner one having both facets & lenses. If we here encounter no greater difficulty than in the case of other structures, then if we look even at the transcendantly perfect eye of the eagle, though we have hardly any guide for judging on the probable transitional stages, yet I think the difficulty is not actually fatal to our theory of natural selection. Nor ought we feel much surprise at our entire ignorance of the transitional stages by which the eye of the higher animals has been perfected; in as much as the links by which the Vertebrates have reached their present position at the head of the animal Kingdom seem to be wholly lost; and amongst existing animals, we

30

are far from understanding the contrivances by which some of the highest perfections of the eye are gained, as for instance the adaptation of the focus to different distances.

A large part of the great difficulty which I have felt in persuading myself that so inimitable an organ as the eye could be perfected by natural selection, has arisen from our constant & almost involuntary habit of comparing the eye with the microscope or telescope. We know that these beautiful instruments have been produced by the long-continued efforts of the highest human intellects; & we naturally infer that the eye has been formed by a somewhat analogous process. But may not this inference be presumptuous? Have we any right to suppose that the Creator works by the same means as man? If we must compare the eye to an optical instrument, we ought <according to our theory,> to take a thick layer of transparent tissue, with nerve sensitive to light beneath, & then suppose

31

<that from external causes,> every part of this layer to be continually changing slowly in density, so as to separate into layers of different densities & thicknesses, placed at different distances from each other & with the surfaces of each changing from flat to various degree of convexity; & further we must suppose that there is a power always intently watching each slight accidental alteration in the transparent layer, & carefully selecting each, which may in any way or degree tend to produce a distincter image <at one end> under the circumstances in which the instrument is used; each of the many new states of the instrument being multiplied by the million; & each preserved till a better is produced, the old being then destroyed. In living bodies, variation will cause the slight alterations, generation will multiply them almost infinitely, and natural selection or the struggle for life will pick out with unerring skill each improvement. Let this process go on for millions on millions of years, & during each year on millions of individuals of many kinds, & may we not believe that a living optical instrument might be formed, as much superior to one of glass, as the works of the Creator <Nature> are to those of Man <Art>.

32

In regard to other organs of extreme perfection, for instance the ear, analogous remarks may be made as on the eye. No doubt there will be in all such cases, many & wider gaps in the known transitional stages, which we cannot bridge over even conjecturally; but the question here is, are the gaps so wide & impassable that they are fatal to our theory, whatever other evidence can be advanced in its favour. In the case of hearing to give one instance of a great difficulty: in a genus of little pelagic Crustaceans, called Mysis, the auditory organs are seated in the caudal plates or swimmers at the posterior extremity of the body.1 According to our theory it would at first appear necessary that these organs should have been moved by infinitely small & numerous variation from the front of the head, where the auditory organs occur in other crustaceans, to the end of the tail; & the possibility of this might be thought to be in some

33

favoured, as I found the case in cirripedes (a sub-class of Crustaceans) placed on the sides of the body about half way between the posterior & anterior extremities of the body; but this would be a false view, as Prof. Huxley has shown that in Mysis the acoustic nerves run to the posterior abdominal ganglion, whereas in other crustaceans they run, as far as known, to the first cephalic ganglion; & it seems impossible to effect such a change by slight transitions. This difficult case can apparently be got over only by hypothetically supposing that hearing is nothing but sensibility to a common vibration carried to an extreme pitch; & hence that a nerve of common sensation might in any part of the body be perfected so as to perceive the finest & most rapid vibrations in air or water. An analogous case occurs in vision: that excellent observer, Quatrefages has shown that some Annelids, which can swim & crawl tail first have eyes at both extremities of the body; & there is another annelid with a pair on each segment of its body. Now it has long been known that some lowly organised animals, have no eyes, yet seem to distinguish

34

light from darkness; & it has been supposed that their bodies are generally sensitive to light; but Müller2 has well remarked that this is quite hypothetical that these animals may perceive only the heat or other influence concomitant with the light.—But our theory almost requires that in low animals, like the annelids & planariae with eyes in diverse parts of their bodies, that an ordinary nerve of sensation may be rendered specially sensitive to light.—

Elements of Physiology Vol 2. p. 1123.

35

Transitions in organs: several kinds of transitions possible. <Organs without known transition states & changes of function in organs.> As natural selection can act only by the accumulation of slight variations, it may naturally be asked, do not absolutely new organs appear in species, or in small groups of species within a class? That this is of rare occurrence is shown by that old saying <of Linnaeus?> become a canon in natural history, "Natura non facit saltum." In recent days one of our highest authorities has insisted how prodigal nature is in variety, how niggard in innovation.1 And in the same spirit, a great Botanist says "Nature, as we have seen a thousand times always proceeds by transitions."2

Before giving a few cases of real & apparent difficulty from the absence of transitional states in understanding how an organ could have reached its present condition, I must make a few remarks on the kinds of possible transition. But first I must admit that those naturalists who speak so strongly

36

of nature not moving by leaps, seldom, probably never, mean to go as far as our theory requires, namely the existence at some period of transitional states <between the same organ in any two members of the same class> as fine as those between an admitted variety & its parent species. Thus to give a single instance, if we look at the family of Humming birds, we shall find a pretty close gradation in the length & form of beak, & although there are considerable gaps in the series, most naturalists would say that nature had here pro-ceeded by transition; but, as Mr. Gould showed me, there are very many forms, for instance in one strange form with the beak bent almost rectangularly downwards, & another with it upturned; & in these two cases there are hardly any transitional forms. Such cases, however, do not seem to me to offer any real difficulty, that is if we admit that the living members of a group bear but a very small proportion to the extinct, as follows inevitably from the working of natural selection. Those naturalists who would lay much stress on so simple a case as this, will long ago have rejected our theory, so no more need be said on it.

37

In considering the possibility of transitions of an organ from one state to another, we should bear in mind that a part having a nearly similar structure may perform in the same individual or in two individuals functions wholly different. Secondly that two widely different organs may perform simultaneously in the same individual the same function; so that whilst one of these organs was continued or perfected through natural selection in its function, the other might come to be used for some quite distinct purpose. Thirdly that organs & the use of parts change normally in the same species with age, or when placed under different conditions; & in these cases it does not seem difficult for the organ to retain throughout life either one or the other of its states.

I will now give a few facts illustrating these remarks; & they will show how cautious we ought to be in assuming in almost any case that a passage from one state of an organ to another is impossible; or

38

that an organ apparently quite new in its class is not some other part changed in function. In such cases the extinction of a few forms would often utterly baffle us in con-jecturing through what stages an organ has apparently passed.

Prof. Milne Edwards has often insisted1 how frequently in the lowest animals, the same fluid & apparently the same tissues serve for digestion, nutrition & respiration;2 thus the Hydra has been turned inside out; the outer surface then serving for digestion, & the inner ceasing to digest & no doubt respiring. This same naturalist, as well as others often insist on the advantages of a division of physiological labour; for instance that a surface will digest better if it has not at the same time to act as lungs, or that a stomach will digest vegetable matter more effectually, if it has not, also, to digest flesh; thus (B)

Owing to this advantage from division of labour natural selection will always tend, where habits permit, to specialise organs. In such cases as the Hydra & many lower animals, the/39/same tissue perform multiple functions;—thus also in many crustaceans, the limbs act as swimmers & branchiae:3

38v

* Annales des Scien. Nat. 3 series Zoolog. Tom 3. p. 264 and Introduct. Zoolog. Generale.

*(B) Dr. Carpenter in his admirable Principles of Comparative Physiology 4th Edit. p. 131. shows that 'in cases where the different functions are highly specialised, the general structure retains, more or less, the primitive community of function which originally characterized it."

(B) it presupposes says von Baer mere prejudice not to rank the stomach of a Ruminant above that of a man./ 38/

39

But we have, also, many instances of distinct organs in the same individual simultaneously performing the same function. Many articulate animals have stemmata & compound eyes which are organs constructed on a fundamentally different principle; the compound eyes not necessarily having any lens or concentrating apparatus. In respiration double organs are common: many animals breathe wholly by their skin, or as in Nereids by a highly vascular portion near their legs,1 aided more or less by branchiae. Even in frogs it has been experimentally proved that the skin largely aids the lungs. The Proteus & other perennibranchiate reptiles, have at the same time both lungs & branchiae. Certain spiders have both a pulmonary sack & tracheae:2 <one species of Nemoura (an insect allied to May flies) has branchiae & tracheae;> tracheae act by carrying air to the diffused blood, & branchiae or lungs by bringing the blood to the water or air, & so are fundamentally different. A few genera of Gasteropod Mollusca have a pulmonary sac combined with branchial organs.3

39v

* (3) Dugès in Annal. des Sc. Nat. 3 [actually 2] series Zoolog. Tom 6. p. 182.—

 

39 insert

(a) text

in the Loach (Cobitis) the whole alimentary canal acts of course for its proper end, but likewise in aid of the lungs, "as this fish swallows air & voids car-bonic acid"4 *: in the larva of the Dragon-fly, water is taken into the intestine by the anus & its oxygen absorbed for respiration; & I may add by the violent expulsion of this water, the animal progresses.

* Owen, Hunterian Lectures, Invertebrata 2nd Edit. p. 560. The Ampullaria of which I was shown drawings by Mr. Woodward offers an excellent instance of these double organs.

40

The proper function of the swim-bladder in fish is explained by its name, but in some fish it becomes divided by vascular partitions & has an air passage or ductus pneumaticus into the oesophagus, & certainly aids respiration4 but these fish have, also, branchiae. There can be no doubt5 that the lungs of the higher vertebrata are homologous or "ideally similar" with the swim-bladder of fish; & according to our theory the progenitor of all the vertebrate animals having lungs, had a swim-bladder, <& that the transition was effected by> the swim bladder having been perfected for respiration through natural selection;—the ductus pneumaticus having become the wind-pipe or trachea,—whilst the branchiae have been atrophied. As the Branchiae became useless for respiration, they might have been slowly converted for some other purpose. Thus the wings of all insects are believed by many entomologists to be homologous with the dorsal branchiae & scales of aquatic annelids; & therefore according to our views there has been, during a long course of descent & a great change of habits, an actual conversion of branchiae into organs of flight though we cannot even conjecture what were the transitional stages. On the other hand the ramified aquiferous

40 insert

* I may just allude to Webers curious discovery of the swim bladder in certain fish being brought into connection with the organ of hearing by a chain of little bones & cavities, & so aiding this function. Indeed in some fish (Owen, Hunterian Lectures. Fish p. 210) as the Cobitis barbatula, the swim-bladder apparently subserves no other function.—

41 insert

Ch. 8 Huxley says he is inclined to think that aquiferous tubes are homologous of tracheae, but from no other reason except they carry circumambient fluid through body.—Thinks mucus-sacks may be same—I doubt whether I had better quote at all—<He disbelieves that Branchiae of Squillae can be considered as a new organ.>

41

vessels of Annelids, by which the circumambient water circulates through their bodies, are believed by Prof. Huxley to be homologous with the airtracheae or respiratory organs of insects.1 We see, perhaps, in the double row of lateral sacs in the leech & earth-worm a transitional stage, for these sacks are considered by Prof. Owen2 as the first morphological step towards tracheae; but their chief office is to excrete mucus.

I will give only one more instance of a <very perfect> morphological or homological transition between organs quite distinct in function; indicating, as I believe in all such cases, a real conversion through natural selection during a long line of descent. In most pedunculated cirripedes there are on the inside of the sack, two small simple folds of skin (called by me ovigerous fraena) with a row of minute glands on their edges. These glands secrete a substance which becomes attached to the ova & thus prevents their being washed out of the sack; but in some few cirripedes of this family which either live embedded or have a more perfect shell, this safe-guard does not seem to be required, & the fraena have no glands, but are

42

larger. In another closely allied family, with perfectly enclosed shells, the fraena have no glands, but are very much larger, & are plicated & sub-plicated, so as to expose a vast surface to the constantly renewed water of the sack; & here these folds of skin have been considered by everyone, as branchiae, <which undoubtedly they are,> although we have but to look a very short distance in the same sub-class to see them serving exclusively as a bridle to retain the eggs.

In the several examples now given, we have seen in respiration alone, the whole skin, or a part, the legs, alimentary canal, mucus-sacks, ovigerous fraena & the swim-bladder either aiding or actually converted into true breathing organs; and in the case of insects, branchiae probably converted into wings. In several of the instances we have also seen two distinct organs simultaneously serving the same office in the same individual.

In all the vast number of animals undergoing metamorphoses, in which the organs at two periods of life are extremely different, it seems quite possible through natural selection to carry on the state during either term of life into the other term. Field mice (Arvicolae)

43

differ from true mice by their molar teeth having fangs; but Mr. Waterhouse tells me that the teeth of old field-mice have been so often observed with fangs that this structure seems almost normally to supervene with age in some species. The two broods in certain annually double-brooded butterflies & moths, differ sensibly in size & colour of which fact Mr. H. Doubleday has given me striking instances. Most parasitic plants are parasites, & most climbing plants are climbers from their earliest days; but the Cuscuta or dodder germinates in the ground becomes parasitic, & its roots then perish: certain shrubs1 become climbing lianas only after having grown to a height.

Certain grasses5 have fibrous roots when growing in moist soils & bulbous when in dry: the immersed & surface leaves of Ranunculus aquatilis differ in a surprising manner: the common Holly when old generally has its lower leaves prickly & the upper smooth. Although these facts do not throw the least light on how a particular state at a certain age, or time of year or under certain conditions is acquired; yet they are worth notice as showing the possibility of a kind of transition by the loss of one of states, different from

5 Alopecums geniculatus & Phleum pratense, Hopkirks Flora anomala p. 22.

43v

* A. de Jussieu. Archives du Mus. d'Hist Nat. Tom in 1843. p. 102. Monographie de la Fain. Malpighiacées.

43 insert

(A text) Some species of Atriplex2 bear on different flowers on the same plant seeds of very different size, colour & smoothness. The same thing occurs, though in definite positions in the flower & seeds of the ray & center in some Umbelliferae & Compositae. The position of the ovule is an im-portant character generally uniform in large groups of plants, but in Buttneria & a few other cases, the same ovarium has one ascendant & the other suspended.3 Moreover Al. Brongniart gives a case4 of an erect ovule becoming, during maturation, suspended.

44

ordinary transition.

That an organ should acquire a particular state at one time of life if useful to the species, presents no particular difficulty, as we have seen that there is a tendency for a variation, or accumulated amount of variation at any period of life to be hereditary at a corresponding period; & we may perhaps hypothetically extend an analogous view to a variation in connexion with some peculiar conditions;

Individual plants raised by florists, as certain <Hollyocks> Dahlias?, have been noticed always to produce flowers of two colours; a variety called the heterophyllous oak produces leaves of several shapes; these tendencies might become strictly hereditary, especially if aided by selection from the two forms being in any way useful to the plant. Seedling Hollies differ greatly in prickliness & the tendency is known to be hereditary; suppose the natural conditions tended to make all the leaves smoother (& luxuriance & starving seem to have a direct action on thorns & prickles); then any natural seedling with all its leaves smooth would be unprotected from grazing animals & would be destroyed <& would not reproduce its kind>, but if smooth only in its upper leaves it might perfectly reproduce its kind, & thus a variety or species be produced with leaves of two kinds, owing to natural selection caring only about the lower leaves.

44v

(a) text

thus the presence or absence of wings in certain insects is believed by several entomologists to stand in relation to the temperature of the season.

45

Real & apparent cases of difficulty in the transition of organs.— By the foregoing cases it will have been seen that an organ may pass through the most extraordinary changes in function & form; this having been apparently facilitated, sometimes, by one organ performing two or more functions, & being then specialised & modified for one function; sometimes by two distinct organs performing the same function, the one being continued for the same & the other being either atrophied or transferred to another office; & sometimes by an organ normally having two states at different ages or under different circumstances, one of the states being preserved & the other lost. Probably many examples might be collected of a part or organ, which, from our not knowing of any intermediate grade, we should be very naturally led to look at as created for some new & special end. But, considering how small a proportion the living bear to the extinct, I have been much surprised at the difficulty, which I have found in collecting many good examples of such

46

cases. It should, however, be here noticed that if we look to an organ in a very isolated being, as the duck-like bill of the ornithorhynchus; or to an organ common to the greater part of a great class, as to the swim-bladder in fishes, the web-secreting organs in Spiders & a thousand such cases, we are very seldom able to indicate intermediate states, & therefore are not able even to conjecture how such structures could have been produced through natural selection. But this on our theory could hardly be expected for isolated beings are supposed to be isolated by extinction; and in the case of an organ common to the whole or greater part of a larger class, in order to find its intermediate stages we should generally have to ascend far in time (the natural selection of many diverse forms always implying a vast lapse of time, & the extermination of numerous less perfect forms) to about the period when the whole or greater part of the class branched off & inherited from a common parent the organ in question. And to ascend very far back in time & to find the inter-mediate stages, by which an organ common to a whole large class was produced, would require infinitely more perfect geological records, than we can hope to acquire. We can only hope to do this, when intermedial states happen to have been handed down by inheritance to the present day.—

47

But cases of extreme difficulty, judged even by the principles of our theory, undoubtedly do occur. Prof. Milne Edwards1 who admits that new organs are occasionally though rarely created, adduces the branchiae of the higher Crustaceans as an instance of an organ, not formed by the modification of any preexistent part. But I must think the case lately given of the branchiae of cirripedes, a sub-class of Crustacea, ought to teach us extreme caution: there has been much extinction amongst the Lepadidae or pedunculated cirripedes & if a few more forms had become extinct, no one could have ever told, that the branchiae of the Balanidae were not a new & special creation. Most naturalists look at the poisonous glands in venomous snakes as specially created organs, & not as modified salivary glands, which their position would indicate; for their intimate structure is wholly different. Here then, apparently, is a case in point. But as we know that many innocuous snakes have channelled or grooved fangs, which convey into the wounds made by them a copious supply of saliva from the large glands at their bases;2

2 H. Schlegel, Essay on Serpents, translated by Dr. Trail 1843, p. 42, 47.

47v

* Introduct. a la Zoolog. Generale. p. 61, 65 &c.

48

and as I have been informed by Dr. Andrew Smith, that a bite of such snake <(Coluber rhombeatus)> caused him immediate pain more than could be accounted for the mere prick, I must believe that saliva3 now in some degree injurious & no doubt useful to the even so-called innocuous snakes could by natural selection be slowly converted into a poison, as deadly as that of the most venomous snakes, entailing with it a change in the intimate structure of the gland.

In another snake (Tropidonotus rudis),1 we have an extraordinary structure, namely the points of certain processes of the vertebrae are tipped with enamel, & penetrating the oesophagus, apparently serve as teeth.

I had thought that the case of the Surinam Toad (Rana pipa Linn.) was quite isolated, here the male glues the eggs of the female on her back;

1 Schlegel on Serpents p. 45.

48v

* Dr. Smith, also, informs me that all the Dutch Colonists assert that the Boomslange (Bucephalus Capensis) a snake without any proper poison gland, causes the death of small animals, which it bites.

48 insert

(a) text

No intermediate structure is known; but, Prof Owen tells me that by.passing the finger down the gullet of other snakes, homologous process can be felt, pointing downwards; & he thinks is quite possible that they may aid in forcing prey down the oesophagous. (The point of the ribs, I may add in serpents, certainly aid in their progression)

49

the skin of which swells & rises so as to form cells. In these cells the eggs are hatched & the young pass their tadpole state. But I find that in a common French Toad, well called Bufo obstetricans, the male helps to deliver the female, & then attaches them to his own thighs. Moreover lately in the same quarter of the globe inhabited by the Pipa, a
Amongst insects I think it likely that instances of apparently quite isolated structures might be found:

wonderful musical instrument of the male Cicadae with its double membrane, powerful muscles & two apertures like those of a violin?3 The Bombardier Beetle (Brachinus) in England & as I have seen on the banks of the Plata, curiously defends itself by crepitations of an acrid fluid & smoke-like gas; but many other allied beetles squirt from their tails

3 Westwood Modern Classification of Insects. vol. 2. p. 422.

49 insert

(a) text

although the highest authority, Kirby & Spence,2 say "there is a regular & measured transition from one form to another, not only with respect to beings themselves, but, also, to their organs —no new organ being produced, without a gradual approach to it." Can a regular transition be shown in the case of the

2 Introduct to Entomology. vol. 3. p 474.

50

an intensely acrid fluid (as I know for I have received a discharge from Cychrus in my eyes), but not so volatile as to turn into gas, & therefore not accompanied by a crepitation. The sting of a Bee or wasp is an admirable weapon, but homologous4 (or rather identical as in the Hive Bee the eggs pass through it) with the ovipositor of other Hymenoptera; & the ovipositor in the Ichneumonidae is known to be occasionally used as an organ of defence, causing "a painful irritation", & for driving prey out of concealed places.

(Kirby confirms the fact of Ichneumons stinging.)

The separation of the two sexes at first seems a difficult case. On our theory it requires that the early progenitors of every class should have been hermaphrodites,—a view countenanced even in regard to mammals, by the rudimentary mammae & womb2 of the males. In plants we can trace numerous

50 insert

Wasps & Bees & Ants use their stings solely as an organ of defence & battle; whereas the Fossorial Hymenoptera almost exclusively use theirs for half-killing insects, & storing their nests with semi-animate prey. And here we see in existing Ichneumons both uses of the sting shadowed forth in an ovipositor: but I shall return to this subject.—

* Westwood Modern Classification of Insects vol. 2. p 77, 117,141. See also [Lacaze-] Duthiers in Annales des Sciences Nat. 3 series. Zoologie Tom. [ ].

51

intermediate steps between hermaphrodite & unisexual flowers. In animals very few such stages are known; but in all the many cases in which hermaphrodites couple or are mutually necessary to each other, as in the oyster in which the male & female elements are matured at different times, it is not very difficult to believe that in some individuals the male & in some the female power might become less & less potent, so as ultimately to abort. And in the Hydra & certain corals one individual is sometimes exclusively male, sometimes exclusively female, but generally hermaphrodite.3

I have given this case of the sexes, that I might allude to the Complementary males of cirripedes, which show in how unexpected a manner nature can effect a transition. Nearly all cirripedes are hermaphrodites, though belonging to the great class of Crustacea, in which the sexes almost universally are distinct; but in two genera4 I found the sexes quite

52

distinct;—several minute males, fourteen in one instance, being attached parasitically on to one female. It may be asked how was the separation of the sexes in cirripedes effected? I venture to assert, if two other small genera had become extinct, this question could never have been answered. In these two genera, some of the very closely allied species have the sexes distinct (in one instance the female carrying in two pouches a pair of minute, mouthless, short-lived males, which when dead are succeeded by another pair), whereas other species are hermaphrodite, but with the male organs rather feebly developed; and these hermaphrodites are aided by a succession of minute short-lived males. From these males being paired not with females, as in every other known case in the animal kingdom, but with hermaphrodites, I have called them Complementary Males. How easily in these two genera a separation of the sexes could be effected: we have but to make the male organs in the hermaphrodite already feebly developed, still

53

more feebly de-veloped so as to abort; & fhe males are already parasitic on the females, & will then, ceasing to be complementary, assume the full dignity of the male sex.

As chemical compounds are definite, it seems at first almost impossible that a substance in one plant should change by gradual transition into a chemically different compound in another plant. That the proportions of different compounds in the same species change most readily under culture is well known; as in the case of wheat & the opium-poppy. But Prof. Christison has shown1 that Oenanthe crocata produces a virulent poison in England, but is innocuous in Scotland: that Hemp yields a peculiar gum-resin only when grown in hot countries; and Dr. Stenhouse2 has shown that the same species of lichen from different regions of the world, contains somewhat different chemical & crystalline substances, which are used for dying.—

54

The electric organs of Fish,—those wonderful organs which, as Owen says, "wield at will the artillery of the skies"—offer a special difficulty. Their intimate structure is closely similar to that of muscle;3 but it is most difficult to imagine by what grades they could have arrived at their present state.4 Nevertheless the fact, <recently discovered> that Rays5 which have never been observed to discharge the feeblest shock, yet have organs closely similar to those of true electric fishes, shows that we are at present too ignorant to speculate on the stages by which these organs, now affording such a powerful means of defence to the Torpedo & Gymnotus, may have been acquired. But the special difficulty in this case lies in the fact that the Electric fishes, only about a dozen in number, belong to two or three of the most distinct orders or better sub-classes of Fish.6 This curious

54v

* Dr. Carpenter in his Principles of Comparative Physiology (IV. Edit.) has an interesting discussion on the Electric organs of fishes: compare p. 465-470, & 471.

+ Valenciennes in Archives du Mus. d'Hist. Nat. Tom 2. 1841. p. 44.

55

subject of closely similar organs occurring in organic beings, which are remote, in the scale of nature, that is, according to our theory, which have branched off from a common progenitor at an immensely remote period & therefore can hardly owe this similar anomalous organ to community of descent will be hereafter considered.

Neuter Insects.—We now come to our last & by far most difficult case of transition, namely the existence amongst wasps, bees, ants & termites of neuters, or sterile females, which often differ in structure & instincts from their parents, & which cannot themselves propagate their kind. We here encounter an accumulation of difficulties. I shall be compelled incidentally to allude to the subject of Instincts, which will hereafter be treated of in a separate chapter, but I shall here as far as possible confine myself to corporeal structure; the remarks, which follow are, however, all applicable to instincts. Of the difficulties, firstly, we have the fact of Neuters occurring in Bees &c, belonging to the Hymenoptera, & in White-ants

56

or termites, belonging to the Neuroptera, that is to a distinct order of insects: this case is parallel with that of electric organs in fishes of distinct orders & will presently be discussed. Secondly, how could the females have been rendered sterile by the agency of natural selection? Thirdly, when formed, how could they possibly come to have a different structure & instincts from their parents? This latter most curious difficulty will best be understood by an example. In certain Ants the neuters consist of two kinds, as in Eciton,1 the soldier-neuters have enormous, peculiarly curved jaws & instincts, greatly different from the jaws & instincts of the other working neuters, and of the fertile females & males: in another species of Eciton, the soldierneuters have large heads & instincts likewise different from the three other occupants of the same community. Now supposing that these soldier-neuters had been ordinary male or female insects, I should have boldly said, that first a slight enlargement of the jaws or head had been favourable to an individual, that this had consequently flourished & propagated its kind; that of its offspring, those with the largest jaws or heads had <been selected> survived, & that this

56v

* I am greatly indebted to Mr. F. Smith of the British Museum, one of the highest authorities on Hymenoptera, for much valuable information on all points in the following discussion.

57

process had been continued, until great protruding jaws or heads had been attained. But in neuter ants, which are absolutely sterile, how is this possible? Granting that in an individual neuter a very slight enlargement of the jaws or more bellicose instincts had been of use to it or its community, & that it or they had in consequence benefitted by the better chance of surviving; yet the neuter could leave no offspring to inherit the peculiarity,—to vary again, & again have the favourable variation selected & propagated. How then could the great jaws & peculiar instinct have been produced by the accumulating power of natural selection? I confess that when this case first occurred to me, I thought that it was actually fatal to my theory; but we shall presently see, that though a very grave difficulty, it cannot in my opinion be considered as absolutely fatal. The case, moreover, is of great interest, for it clearly shows that the Lamarckian doctrine of all modifications of structure being acquired through habits, & being then propagated, is false; for whatever may have been the habits of life

58

of our neuters, they never leave, (at least in ants) offspring to inherit the effects of habit or practice. For my own part, though I do not doubt that use & disuse may affect structures & be inherited, yet long before thinking of this case of neuter-insects I had concluded that the effects of habit were of quite subordinate importance.

First we will consider the simple fact how it is possible that communities of insects should come to possess sterile females or neuters: this is not a special difficulty, but only one of the same class, as that of any organ in a highly peculiar condition. There is some gradation not only in the numbers of the neuters in different species, but likewise, apparently in the degree of sterility: some ants have but few neuters, whereas in the Hive Bee the neuters have been estimated at up even to 40 or 50 thousand to one Queen. In ants there is no reason to suppose that the neuters ever lay eggs, though Huber1 has seen them coupled with males, an act which always causes their death.

59

In the Hive-Bee, the neuters occasionally, though rarely lay eggs, which invariably produce only males. In Wasps & Humble-Bees the early-born females are small, <do not differ in structure from the Neuters> do not survive to winter like the large females, & the eggs they lay yield only males. These small females are said by Huber2 to be attended by a small number of males, & therefore it might be thought that they had been fertilised; but this, probably is not the case anymore than with those neuters of the Hive-Bee, which have been known to lay eggs.3

In our third Chapter numerous facts were given showing how readily organic beings under changed conditions, not un-favourable to life & health were rendered sterile. When a cow produces twin calves, & one is female, she is a free-martin & always sterile. In the male Lucanidae, or stag-beetles we have seen that collectors are not satisfied till they possess a series

59 insert

a text)

Lastly in one of our common Wasps4 the workers produced late in the autumn are larger than those produced earlier in the summer & almost seem to be graduating into the state of fertile females.

4 Vespa Germanica. F. Smith in Zoologist [ —See I (1843) 161-6.]

60

from mandibles developed to an enormous size to mandibles differing little from those of the female: this I believe to be caused by the amount & kind of food which the larva have obtained, & one must suspect that it stands in some relation to the virile power of the males. Male & female Brachyourous Crustacea differ in the width of the abdominal segments; but in some species intermediate individuals are not rarely found1 & these females are believed to have been rendered sterile from some unknown cause

Now let us suppose that these Crustacean, Lepidoptera or Lucanidae were truly social; many males & females (as with wasps & humble-bees) living & working together for the common good: in this case it seems not improbable, owing to the vast fecundity of the lower animals, that a certain number of females, working like the others, but without any waste of time or vital force from breeding, might be of immense service to the community. If this were so, & we see it is so with social insects, then natural selection would favour those communities, in which some of the individuals

60 insert

a text)

In Lepidoptera Mr. Newman2 has given good reason for believing that the females in the autumnal broods, when two broods are not normally produced, are utterly sterile: this has been observed in France & England; & the high authority & experience of Mr. Doubleday is adduced in support of this remark-able fact.

61

had been exposed to conditions, or eaten food which had rendered them in some slight degree less fertile than the other individuals. In the social Hymenoptera we have to suppose that in long past ages some of the larvae were fed in the early part of the summer on some peculiar food or otherwise treated so as to have been rendered slightly less fertile than the other larvae. Then natural selection or the struggle for life, would ensure the con-tinuance or the increase of the same treatment, so that the degree of sterility or the number of sterile individuals might be increased.

Now for our great difficulty of how neuters, not having progeny can be modified through natural selection, so as to fit their various offices in nature. With wasps & humble-bees the large females alone survive the winter; & in the spring in their solitary state they perform all the duties of their neuters which are subsequently produced, & do not differ from them essentially in structure; Hence the neuters of the different species of wasps & Humble bees might be modified by inheriting any selected modifications in the females

61v

(a) text.

: slight differences, as in size, might be accounted for on the same principle that a bullock differs from a bull; or capon from a cock.—

[This note is not transcribed in Stauffer's Natural selection. JvW]

61 insert

(a) text

just as bullocks of the different breeds of cattle, <& capons of different breeds of poultry,> differ from each other & slightly from the perfect males of their own breeds.

62

In the Hive-bee, the queen differs greatly from the neuters in instincts & in many important points of structure, as in the mouth, shape of sting, absence of wax-secreting pockets, & of the several curious contrivances for collecting & carrying pollen.1 Now in most of these respects the Queen differs not only from her own neuters, but from the typical character of most social Bees; & it might be argued on our theory that the Hive neuters have retained by inheritance from an early progenitor certain normal characters which the Queen had lost. It deserves notice that the cuckoo-like species of bees, which lay their eggs in other bee's nests, have lost similar points of structure, either through disuse, or through natural selection, or both combined. That in the Queen Hive-Bee the loss of certain parts or any modification of structure should become attached to the female sex alone, is not at all surprising as we have in previous chapters seen; but here we have the truly astonishing fact that these hereditary losses are correlated with a particular treatment of the larva: this we know from

1 Kirby & Spence Entomology vol 2. p. 131.

63

the fact that larvae which would certainly have become neuters & therefore would have had wax-pockets & the corbicula on their hind-legs &c can by a particular line of treatment be turned into Queens without wax-pockets & pollen-collecting instruments. According then to our theory, the neuter or sterile females of the Hive-bees owe those characters which they have in common with other social Apidae to ancient inheritance; the fertile female having lost them either by disuse or through natural selection, but always in correlation with a certain line of treatment during the larval state.2 This seems a very bold hypothesis; but then there are two kinds of neuter Hive-bees; one larger, with a more capacious stomach, much greater power of secreting wax & which does not build; the other smaller, a nurse & builder.1 How can we make this fact accord with our theory of natural selection? But first let us take the case of neuter-ants, in which analogous facts are more strongly displayed.—

1 Kirby & Spence vol I. p. 492.

63v

2 It is remarkable that the males when fed on royal jelly are not affected; but when they are hatched in workers-cells, they are believed to be rendered smaller: Kirby & Spence Entomology. vol. 2. p 126, [actually 127] 161.—

64

In Ants the case is reversed as compared with Hive-Bees, for their neuters, in differing from the fertile females, differ in an extraordinary manner from the typical structure of their sub-order, —namely in being always wingless, in the very peculiar shape of their thorax, in the frequent absence or very rudimentary condition of the ocelli & indeed in some genera (as Ponera) in being destitute both of ocelli & compound eyes.2 But it deserves notice that in one allied family of non-social Hymenoptera, the Mutillidae, the females are wingless, destitute of ocelli, & the thorax is often singularly like that of a neuter ant:3 hence, perhaps, it may be inferred that there is some correlation4 between these points, so that if one were modified, the other points would tend to follow.

Considering the terrestrial & sub-terrestrial habits of ants, the Lamarckian doctrine that they have lost their wings & ocelli by inherited disuse seems very tempting, but how utterly false; as it is just the wingless individuals which can never leave offspring! As queen-ants, like the large female wasps & Humble-bees, at the first foundation of a

2 Westwood Modern Class, vol 2. p 218, 235 and F. Smith on British Formicidae Transact. Ent. Soc. vol 3. Part 3. p 110, 113, 115.

3 Westwood Modern. Class, vol 2. p. 213.

64v

4 See remarks on this subject by Mr. Westwood in Annals & Mag. of Nat. History. vol 6. (1841) p. 81.

64 insert

(a) text (p. 64)

The neuters, also, differ from the fertile females in size, in the shape of head, & of the mandibles, sometimes in the number of joints (Pseudomyrma) of the antennae, & in the form (Cryptocerus) of the abdomen.5

5 F. Smith in Entomolog. Transacts vol 2. p. 215 & vol.3 p 156. [See Appendix for letter from Smith about forms of Cryptocerus workers.]

65

community do all the work, any selected modification in them would be transmitted to their neuter offspring; but how could these neuters have acquired, through natural selection, a structure so widely different from that of their mothers? Moreover the neuters in closely allied species of the same genus, which we by our theory believe to have all descended from a common ancestor, also, of course, differ from each other. But the difficulty comes to a climax when we remember that amongst the neuter-ants of the same identical species we have in several genera two kinds extraordinarily different in structure & instincts; as in the case of Eciton already alluded to, in which the soldier neuters have enormous jaws & the working neuters, whom they guard ordinary jaws; & in another species of Eciton & likewise in Atta the soldier neuters have heads twice as big as those of the working neuters; In a Mexican genus1 there are ordinary working-neuters & others never quitting the nest, with the abdomen swollen into a great, almost diaphanous sphere nearly five times as large in diameter as that of the common workers: these inactive neuters serve as mere distillers of a sac-charine fluid, which is stored up in a sort of comb. Lastly in the Driver ants of W. Africa (Anomma arcens, Westwood) there are, according to the Revd T. Savage, three classes

1 Myrmecocystus Mexicanus M. Wesmael in Bull. Acad: Royales: Bruxelles. Tom. 5. p. 766.

65 insert

of workers, differing in size, in their manner of biting, & in the work which they perform: and all are absolutely blind:2

2 Transactions of Entomological Soc. vol 5. p. 9 & 16.

66

Grave as these several difficulties are, do they overwhelm our theory <of natural selection>? Let us turn to our best guide the process of selection by man in our domestic productions. Man almost invariably selects from external appearances & breeds from the individual which he approves of: but let us suppose that he cooks & tastes a cabbage or radish & finds it very fine flavoured: that individual plant is utterly destroyed; but let him sow seed from several plants of the same stock in separate beds; of these seedlings let him cook & taste some out of each bed; & let him again save seed from the bed which produced the best-flavoured plants; & so repeat the process: in time, I cannot doubt he would get his desired variety true without ever having bred from a selected individual, only from a selected family. Breeders of cattle, like the famous Bakewell3 who have attended to the grain of the muscle & to the fat & lean being well marbled together must have followed this plan, of breeding from the family to which the slaughtered animal belonged. To give another hypothetical illustration: the oxen or castrated animals of the Craven

3 See, Marshall's account of Bakewell's proceedings in Youatt on Cattle. p 191.

67

cattle have horns not only much longer than those of the Bull, but even than of the cow: now I have such confidence in the principles of inheritance & in man's power of selection, that I fully believe by carefully noticing which families produced oxen with the longest horns, a stock might be reared, which not having themselves very long horns, yet when castrated would invariably produce oxen with extraordinarily long horns: this seems to have been effected, I presume accidentally, as far as size of body is concerned, in the oxen of the Devonshire & Herefordshire breeds,4 which oxen are of an extraordinary size. This principle of selection, namely not of the individual which cannot breed, but of the family which produced such individual, has I believe been followed by nature in regard to the neuters amongst social insects; the selected characters being attached exclusively not only to one sex, which is a circumstance of the commonest occurrences, but to a peculiar & sterile state of one sex.

Now to take the case of neuter ants, which neuters differ more from their parents than in other social insects; if the absence of wings

3 See, Marshall's account of Bakewell's proceedings in Youatt on Cattle. p 191.

67v

4 Youatt on Cattle, p. 17. The oxen of the Devonshire cattle are much larger in body than the Bulls; & the Bull is very much larger than the cow. In the Hereford-shire cattle, the ox is, also, a very large animal; & this cannot be simply accounted for by the effects of castration; as in the Durham or short-horn oxen, I am assured, there is no such inequality of size compared with the bulls & cows. In regard to the Horns of the Craven oxen seep. 197 of Youatt's Work.—

68

was any advantage to them, & we may fairly suppose that it would be so, seeing that the queen ants & termites tear off their own wings as soon as they found a colony, then I believe that those males & females which happened to produce neuters with their wings aborted, or with them ever so little less, would have a slight advantage; & of the myriads every year born these males & females would have a better chance of surviving & pro-creating neuters with less & less wings, till wingless neuters were produced. Judging from the Mutillidae, with the loss of wings, the thorax, owing to the laws of correlation of growth, would be modified, & possibly even the head & ocelli.

No doubt the process of selection would be retarded in an extreme degree by its action being indirect,—that is on the family alone; the individuals themselves born with any useful variation never leaving offspring. Had it not been for neuter insects, I am bound to confess that I never should have supposed that this process could have been as efficient, as our theory requires it to have been in the case of neuter insects. Extremely slow as such selection must

69

be; we have, at least for Bees & Termites, a super-abundance of time, for fossil Bees have been found in Jurassic strata & Termites actually in beds of the Carboniferous age!1

According to all analogy, neuters from the same parents would not all present the same variation, or the same in the same degree. In the case of Ants, for instance, analogy would lead us to infer that some few might be born with wings slightly smaller or the jaws slightly larger but that the other neuters in the same nest would retain their uniform character. Therefore, it may be urged we ought to have, or have had, communities presenting inter-mediate grades; more especially as there can here have been no crossing between the several neuters to keep them all uniform in structure. But I believe, first taking the case in which all the neuters have been altered from the maternal type, that variation is generally so insensibly slow, that without a comparison could be instituted between

1 Lyell's Manual of Geology. 1855. p. 389: Pictet's Paleontologie 1846 Tom. IV. p. 109.

70

the neuters at the present day with those belonging to the same species, a thousand or more generations ago, no difference could be perceived: I suppose that social communities are profited, first by a few neuters having wings, jaws or other organs different from the same parts in the fertile females in so slight a degree as to be imperceptible by us; & that the tendency in the parent to produce such neuters is increased by natural selection, until all the neuters are thus characterised;

But if it were an advantage to the community that only certain proportion of the neuters should have, for instance, larger bodies or jaws than the other neuters, then I can see no insuperable difficulty in believing that by selection parents could be formed, which would produce a certain number of sterile females with big bodies or jaws & a certain number retaining their former small bodies or jaws. To give an instance from the vegetable kingdom, in which an analogous difference has appeared suddenly, & with long continued selection might perhaps be rendered hereditary by seed: there is a grape.1

1 Count Odart. Ampélographie 1849. p. 71.

70 insert

and that subsequently the amount of difference is augmented first in a few neuters, & then again extended to all the neuters; & so onwards.

71

which produces almost regularly on the same bunch small round and large oval berries,—a character I may add, considered by Odart as usually amongst the most constant in the vine. In the case of the two kinds of neuters in the same nest,—the acme of difficulty on our principle of natural selection—there must have been communities presenting during a long period grades between the large bodied or jawed individuals & the small: it must have taken an extraordinary length of time for selection acting only on the parents to produce a defined line of demarcation between the two sets of neuters in the same nest,—between the warriors & workers, for instance, in Eciton. Considering how very few social insects are well known I am surprised that I am able to adduce on the highest authority some instances of intermediate grades in the same nests. Mr. F. Smith informs me that in the nests of Formica flava, though there are large & small neuters, they so graduate into each other that it is impossible to separate them into two distinct bodies. In F. sanguinea, the neuters if viewed in mass may be divided into two bodies, differing considerably from each other in size & colour 1

1 Transact Entomol. Soc. vol 3. P. 3. p. 102. In F. fusca, I may add, (p. 105) the difference in colour between the two-sized neuters is not invariable; the smaller neuter having "usually much paler legs & antennae" than the larger neuter.

72

; & their instincts are slightly different, for if on a hot midday when all are in their nest, the bank be struck, the large neuters alone come to the surface as defenders. But Mr. Smith tells me that if all the neuters in a nest be carefully examined, a considerable number will be found graduating from one extreme to the other. Again in the Driver-ant of Africa

In these cases, we have intermediate forms between the neuters in the same nest; but if we compare the amount of difference between the two classes of neuters in the nest of different social insects, even within the limits of the same genus as in Eciton, we, also, find a gradation, as might have been expected from selection having in some communities produced a far greater difference between the neuters, than in others. In Humble-bees the neuters seem generally to vary more in size but not in colour than the males or females: Mr. Smith3 states that in B. muscorum, which includes seven reputed species of Kirby, the several varieties sometime coexist in the same nest, but generally some one is preponderant: Mr. Newport

3 Catalogue of British Hymenoptera: Apidae—1855. p. 213. Compare size of the neuters & others in the several species of Bombus.

72 insert

which presents the unique case, as far as I know, of three classes of neuters in the same nest; the largest warriors are thrice as large as the least workers, & differ in some other trifling respects; but Mr Westood2 expressly states that "there seems indeed to be a regular gradation in the size from the largest to the smallest," & "I must confess that I can discover no distinct character to separate the largest individuals from the others.

2 On Anomma arcens: Entomological Transactions. vol. 5. p. 16.

73

moreover, seems to have found4 in some species two classes of neuters like those in the Hive Bee; but the difference must be very slight, as this does not seem to have been noticed by other observers. In the Hive-Bee5 the two kinds of neuters differ very slightly in size, considerably in instincts & in the development of the wax-pockets; but in the latter respect the difference is not strongly defined, as the nursing-neuters "do secrete wax, but in very small quantities:1 <occasionally, what apiarians call Captain or black-bees appear in a hive; & here we have the groundwork for the production through natural selection of a third class of neuters, should such Captains prove in any way useful to the community2 > Even in British Ants alone we have some with all the neuters in the same nest quite uniform in size & structure; others with neuters of two classes differing slightly in size & not apparently in habits3 as in Formica flava; others in size & colour & somewhat in habits as in F. sanguinea, and ac-companied by slight differences in shape of thorax as in F. nigra. In many ants the neuters are quite destitute of the ocelli, which are present in the perfect sexes; & in the smaller neuters of F. flava the ocelli are "only distinguishable under a high microscopic power", whereas in the larger neuters of this species they are "distinctly visible",4 but yet far smaller than in the males or females.

73v

5 The Italian Bee (Apis Ligustica) is now considered by capable judges as only a variety of the common Hive-Bee, & this view is rendered extremely probable by their perfect fertility together. They differ considerably in colour, & the Italian Bee is more industrious, flies quieter so that von Berlepsch (Bienen-Zeitung 1856 p. 4 and Dzierzon's article p 61) says he could distinguish hives of the two kinds with his eyes shut, & it stings much seldomer: but as it is expressly stated that the Queens differ in colour & in seldomer stinging, this case, though very interesting as showing that strongly-marked varieties can arise in Bees, does not throw light on the difference between the workers & the Queens, or between the two classes of workers. It seems, however, that in the ancient Roman times (True Parth[en]-ogenesis by von Siebold Engl. translat. p 71. and Bienenzeitung 1856 p. 4) and at the present day common dark-coloured neuter-bees appear amongst the golden Bees even in Italy: this, however, may possibly be due to crosses owing to the common Bee having been anciently introduced into Italy.

1 Kirby & Spence Entomology vol i. p. 493.

74

In some of the species, also, of Eciton the two classes of neuters differ only slightly in size, whereas in other species of this genus, & of Atta & of Myrmecocystus & of Cryptocerus we find the most astonishing differences, in heads, jaws & abdomen.5

According to our theory it might easily happen that parent-ants after having produced two forms of neuters, should through natural selection come to produce more & more of one form till none of the other were left. I infer that this .has actually come to pass with the Polyergus rufescens6 which from making slaves of the neuters of other species (how this truly wonderful instinct could have been acquired, will be discussed in a future chapter) has no working neuters, but only warriors, or slave-takers, which have jaws incapable of building a nest.

75

I have discussed this case of neuter social insects at great length, for it

(as far as corporeal structure is concerned) of the differences of the neuters from the fertile females, & of the two classes of neuters from each other;—namely by the continued selection, not of individuals which have varied in some way profitable to themselves, but of the stock which has produced any profitable variation; the variation having become correlated to a certain state of one sex <favourable to the community>. A division of labour is possible in communities of man, through his intellect, his traditions & artificial instruments; but in communities of insects, which have almost unvarying instincts & for instruments, only their jaws & limbs, a division of labour seems to be possible only by the production of sterile individuals; for had the different workmen been capable of breeding together, the several castes would have been blended together & so lost.

75 insert

is by far the gravest difficulty, which I have encountered; so grave, that to anyone less fully convinced than I am of the strength of the principle of inheritance, & of the slowly accumulating action of natural selection, I do not doubt that the difficulty will appear insuperable. But I have now done my best to show how I reconcile with our theory, the facts

76

I have now finished, as far as seems necessary, the subject of the transitions of organs. We have seen some cases, as that of the eye, most difficult from its transcendant perfection; some from no transitional stages being known, and some from our not seeing as with Electric fishes, how any transition is possible; but I think facts enough have been given to show how extremely cautious we ought to be in ever admitting that a transition is not possible. Considering the number of forms which undoubtedly have been <exterminated> utterly lost, I am much surprised that we have not encountered very many more cases of extreme difficulty in attempting to show how one organ or part may be slowly converted into an apparently quite distinct organ.

Finally it seems to me highly important to bear in mind that he who believes that each species has been independently created, can only say that it has so pleased the Creator never or most rarely to introduce a new organ. Or he may mask his ignorance, & say with Milne Edwards1 that the "law of economy"

77

is almost as paramount in nature, as the law of "the diversity of products". But on our theory of gradual modifications through natural selection, the law of economy is only the law of descent, the canon "Natura non facit saltum" becomes scientifically explicable.—

Similar & peculiar organs in beings far remote in the scale of nature. — I have already alluded to the remarkable case of Electric organs occurring in genera of fish, as in the Torpedo & Gymnotus almost as remote as possible from each other: but the organs differ not only in position, & in the plates being horizontal in one & vertical in the other, but in the far more important circumstance of their nerves proceeding from widely different sources.1 I have also alluded to another very remarkable case, namely both ants, belonging to the Hymenoptera & termites belonging to the Neuroptera having communities, served by sterile females; the fertile females, I may add in both cases, losing their wings, as soon

78

a new community is founded: but, according to the prevalent belief, there is a wide difference in the two communities in the larvae of the termites being the workmen. The luminous power of certain insects is a rare & curious property; but in the Lampyridae it is the under surface of the abdominal segments, in Elater two spots on the hind part of the thorax, which shine.

The eye of the cuttle fish contains all the essential parts of the same organ to the two main divisions of phanerogamic plants, yet they present a curious resemblance in their means of fertilisation; in both, & in no other plants, the pollen-mass is attached by a footstalk to a sticky gland4 which when touched by an insect

78v

in the Vertebrata, belonging to a different King-dom: a cornea, crystalline lens, & vitreous humour, corpus ciliare & retina are said to exist,2 but it seems that neither the cornea, or the iris, are homologous, that is different parts are worked in for the same end; & the structure of the retina is extremely different.3 To give a case of parts of little importance; in the Echidna, one of the most aberrant of the Marsupialia, & in the placental Hedge-hog, we see the body protected by very similar spines.

In the Vegetable Kingdom, Orchis & Asclepias belong

78 insert

in the Vertebrata, belonging to a different Kingdom: a cornea, crystalline lens, & vitreous humour, corpus ciliare & retina are said to exist,2 but it seems that neither the cornea, or the iris, are homologous, that is different parts are worked in for the same end; & the structure of the retina is extremely different.3 To give a case of parts of little importance; in the Echidna, one of the most aberrant of the Marsupialia, & in the placental Hedge-hog, we see the body protected by very similar spines.

In the Vegetable Kingdom, Orchis & Asclepias belong

2 Müller Physiology p. 1117.

3 Carpenter Principles of Comp. Physiology 4th Edit. p. 730.

79

is drawn out, & is thus carried on to the stigmatic surface: moreover, according Aug. St. Hilaire5 the sticky gland with its footstalk, which becomes during growth united to the anther, are developed in both cases in a similar manner. The leaves converted into pitchers in Sarracenia & in Nepenthes is another instance of a nearly similar structure in plants far from closely allied.

According to our theory when we see similar organs in allied beings we attribute the similarity to common descent. But it is impossible to extend this doctrine to such cases, as those just given of the Orchis & Asclepias, the Torpedo & Gymnotus, the Echnida & Hedgehog &c,—excepting in so far that community of descent, however remote the common ancestor may have been, would give something in common to the general organisation. Just in the same way as in our last Chapter we have seen that the occurrence of similar monsters in the most diverse members of the same great class may be attributed to a like organisation from common descent, being acted on by like abnormal

5 Lecons de Botanique p. 448.

80

causes of change. In the case of the eye of the molluscan Cephalopod & of the vertebrate animal, I do not pretend that we have one single fact (without it be the resemblance of the germinal vesicle) to induce us to believe that the members of these two great Kingdoms have had a common descent. It is not, I think, at all surprising that natural selection should have gradually given a fish & a whale something of the same forms, from fitting them to move through the same element; just as man in a small degree has given by his selection something in common to the form of the grey-hound & race-horse. A similar doctrine, I infer, must be extended to the above given remarkable cases of similar, though very peculiar & complex structures, in beings remote in the scale of nature. Such cases are not common; & in some of them the parallelism, as we have seen in the electric organs of fishes & in the eye of Cephalopod & mammal is not absolutely strict. Men, without communication have sometimes simultaneously hit upon the same curious invention: here man's intellect, which is nearly the same for all, may be compared with the power of selection which is the same throughout nature; & the general state of knowledge, the groundwork of all man's inventions, may perhaps be compared to that degree of general resemblance in organisation, which the members of the same great class have derived from common, but immensely remote, ancestors.

81

Organs of little importance modified by Natural Selection.—As natural selection acts solely through life & death by the preservation of slight favourable variations & the destruction of less favourable ones, the formation or modification of organs of apparently extremely little importance to the life of the individual has often seemed to me fully as great a difficulty, as the formation through such means of the most perfect & complex organs.

really know the entire economy of any one being, we may sometimes attribute importance to characters which are of little or no service to the individual; sometimes we may place to the account of natural selection that which is wholly due to the laws of growth; & probably still oftener we think that of little importance, which in truth is of the greatest in the struggle for life.

Thus if we had known only the green woodpecker, we might have said that its colour was of service to it in escaping dangers in the woods, but the many black, white & crimson woodpeckers show that probably this would have been a false view: seeing how over the whole world Kingfishers, both male & female, are bril-liantly coloured, we might naturally attach some importance

81 insert

In the case, however, of those animals, which possess will & choice, we must not forget "sexual selection", which may modify parts of little general importance, namely such as favour the struggle between male & male, or such as serve to charm the females; & characters gained by sexual selection amongst the males seem not rarely to be transferred to some extent, as shown in our sixth chapter, & possibly sometimes to a large extent, to the females. In as much as assuredly we do not

82

to their colours in relation to their fish-taking habits; but a closely allied & similarly coloured bird, the Dacelo Jagoensis inhabits deserts, far from water & preys on lizards & grasshoppers.

Seeing how absolutely necessary whiteness is in the snow-covered Arctic regions to the prey-seizers & the preyed, we might attribute the absence of colour to a long course of selection; but it may be that whiteness is the direct effect of intense cold; & that the struggle for life has only so far come into play that coloured animals would in the arctic regions live under a great disadvantage. So again, the curious recurved hooks on the tips of the branches of the Java Palms (), which are so strong & effective that the natives use a branch as a thief-taker, are quite necessary to this trailing plant that it may climb the lofty forest-trees; & hence we might attribute (& perhaps truly) the formation of these hooks to a long course of selection; but the many curiously formed thorns & hooks on trees, which can apparently be of no use to them from their height, may lead to the conclusion, that such hooks are simply due to unknown laws of growth; & that in the Java

83

palm the plant has become a trailer so as to take advantage of the already formed hooks, & not the hooks slowly formed to suit the changing habits of the plant.—The open sutures in skull of the just-born mammal

Probably we oftenest err in attributing too little importance to slight points of structure in the struggle for life. Looking at the tail of the Giraffe, which seems quite like an artificially con-structed fly-flapper I thought at first that surely this instrument could never have been modified & adapted for its humble end, through natural selection; but when I remembered Bruce's account of the torments suffered from flies by the largest & thickest-skinned pachydermata in Abyssinia; & when I remembered that the extension of the introduced quadrupeds in S. America, is in many cases <absolutely> governed by insects, I felt that it would be rash in this case to put limits to the powers of long-continued selection. Again I doubted whether the form or size of the external ear could be modified by natural selection; but how all-important is hearing to the Hare, & we know in domestic rabbits how prodigiously the ears have been increased by the fancier's selection, so that rabbits have been exhibited, with the two ears from tip to tip [ ] inches in length:

83 insert

which allow the bones to close together so as to facilitate birth, have often been advanced as a case of special adaptation; but as the sutures are equally open in the skull of the young bird or reptile, which has to come only out of an egg, we see that this structure must be due to some quite independent cause; & being present has only been taken advantage of in the birth of mammals.

84

sportsmen, also, know how injurious it is to crop the ears of terriers, which have to enter burrows; & cruel gamekeepers crop the ears of cats, for when this is done they will hardly enter a wood. Again I thought that such an apparently small point of structure, as the eye-lashes, could never have been formed or modified by selection: yet at times when the struggle for food is most severe, what a momentary difference in vision must often determine which shall survive & which perish; what a trifling difference may often determine which individual shall escape some beast of prey or other danger. But why nocturnal marsupials should not have eye-lashes would, I suppose puzzle anyone to account for.1 Vultures which wallow in putridity have the skin of their head naked; whether this adaptation is due to selection, I will not pretend to conjecture; & we should remember that the head of the clean-feeding Turkey-cock is naked like that of the Turkey-buzzard.—

In all cases of organs of apparently trifling importance, we should bear in mind that selection may act on them from their concurrence in a more or less perfect state with other advantages or disadvantages; for when the chance of life is

1 [Pencil note:] ?? Owen cannot remember having made any such statement.

85

trembling in the balance from some quite distinct cause, an extremely slight difference, as more or less protection from insects or temporarily better vision, might well determine which way the well-poised beam should strike; for of those annually born a few alone can leave offspring. Moreover a part or organ, though of secondary importance to most animals, may be of the highest to some having particular habits, as the external ear or eye-lashes to a burrowing animal; & under such conditions the organ might readily be perfected by natural selection, & subsequently inherited by numerous descendants modified in other respects, to whom the organ was of less importance but yet useful in its perfected state. Even in this latter case natural selection might be enabled to check any decidedly injurious deviations from the perfected state; as for instance the eye-lashes growing inwards, which causes to man much suffering & weakness of vision; quite enough almost to ensure the destruction of an animal which had to provide for itself in a time of dearth.—

86

Several distinguished writers1 have of late protested against the utilitarian doctrine that every part of every organic being is of use to it: they seem to think that nature plays with her work for mere variety sake or for beauty. Are we to believe that infusoria are exquisitely sculptured for man to admire them through the microscope? This protest against utilitarianism seems to me rather rash, as assuredly we do not know the whole life, its dangers & advantages, of any one single being; if we did, we could say why one is rarer & one commoner in any country. In the structure of each being, very much must be attributed to the correlation of growth,—that is when one part is modified for the good of the organism, other parts will in consequence be likewise in some degree modified: very much, also, must be attributed to inheritance from ancient progenitors, as we see in an exaggerated degree in rudimentary

87

organs. But in every case, according to our theory, the structure of the ancient progenitor could <must> have been modified or acquired, solely through its own good. So that all structures in all beings, making allowance for the correlation of growth to a larger but unknown extent, & making some allowance for the direct action of food & climate, must either have been useful to a progenitor or be now useful to the present descendant. The doctrine that structure is developed for variety or beauty sake would, if proved, be fatal to our theory.—

Looking again, not to the separate parts or organs, but to the whole individual, one is sometimes tempted to conclude, falsely as I believe, that nature has worked for mere variety: thus when we hear2 that Mr. Bates collected within a day's journey, in a quite uniform part of the valley of the Amazons, 600 different species of Butterflies (Gre[a]t Britain has about 70 species), one may at first doubt whether each is adapted to its own peculiar & different line of life; but from what we know of our own British Lepidoptera

2 A. Wallace, Narrative of Travels on the Amazons. 1853. p. 469.

88

we may confidently believe that most of the 600 caterpillars would have different habits, or be exposed to different dangers from birds & hymenopterous insects. Mr. Wallace in his interesting Travels1 seems to doubt the strict adaptation even of very differently constructed birds; for he lays much stress on the fact of having repeatedly seen the ibis, spoon-bill & heron feeding together on precisely the same food; & so with pigeons, parrots, toucans &c. But until it can be shown that these birds feed throughout the year on exactly the same food, & are throughout their lives from the nest upwards exposed to the same dangers,—for to want or danger each must be sometimes exposed, otherwise each would increase inordinately—the fact of their feeding together for a time or even for the whole year, seems to me to tell as nothing against the strictest adaptation of their whole structure to their conditions of existence.

On the other hand natural selection will produce nothing on the whole injurious to the species; no part or organ, though subject to the acutest suffering, will be actually formed, as Paley has remarked, to give pain. But

1 Id. p. 84.

89

natural selection will not necessarily produce absolute perfection, as judged of by our poor reason. Each organism must be sufficiently perfect in all its parts to struggle with all its competitors in the same country; but by no means with all existing beings, as we see in the lessened numbers & even extinction of indigenous animals when others are introduced. We may err greatly, but can we call the sting of the Bee or Wasp perfect, when its use causes the insect's death by the tearing out of its viscera; the Bee, as I am informed by an apiarian, seeming conscious of its fate & never returning to its hive.2 But if this fatal power of stinging, though it causes the loss of one member (but a member which does not breed) be of use to the community, it satisfies the requirements of the principle of natural selection. If we look at a Bee as an independent creation, this fact of death ensuing from the instinctive use of its own weapon must appear, as was long ago remarked

2 Bevan, Honey Bee. 1827. p 278. gives the best account of the act of stinging of the Bee which I have met with.

90

by John Hunter,3 very singular: but on the principle of inheritance we can perhaps, understand how the two barbs came to be retro-serrated, so that their withdrawal is so difficult; for the two very same organs are serrated in the same manner in very many members of the same order, for the sake of sawing or boring holes for their eggs, in a manner which has mostly justly excited the admiration of every observer1 Hence I infer that the ancient progenitors of Bees & Wasps used their ovipositors as boring instruments & that their eggs were laid with an acrid secretion; the boring instrument having since been con-verted in the neuter bees & wasps exclusively into an organ of defence; the acrid fluid having been intensified into a virulent poison. If in any member of the order, the sting gradually came to be habitually used for any purpose, then I do not doubt that natural selection, by always

3 Philosophical Transactions 1792. p. 191.

91

favouring those individuals, which could easiest withdraw their less-strongly barbed stings, could make the spicula as smooth as in the Sphegidae,2 which require their frequent use in order to half-kill their prey as a store for their larvae.

If we admire the female tiger savagely defending her young, or the hen-bird facing a hawk even to her own destruction, can we equally admire the Queen-bee always trying with the utmost fury to sting to death her own just born rival daughters:3 We are accustomed to maternal love, but here we have instinctive, in-veterate maternal hate; but both are the same, if useful to. the community, to the unconscious & unpitying power of natural selection. We may err greatly, but can we call the drone or male Hive bee a perfect creation, whose sole function is to unite with the female; this union inevitably causing its death? If in most insects, we admire the means by which the male finds the female, —as by that almost incredible power of scent in moths which so often leads the male even down a chimney into a chamber in which the female is confined—or which leads some other moths to fmd

2 M. Fabre in Annal. des Scienc. Nat. 3 series Zoolog. Tom VI. p. 161 [series is 4, not 3] I have examined the sting of Pompilus & I could see no trace of Barbs.

3 Kirby & Spence Introduct to Entomology vol 2. p 142.

91v

If we kill an ant, its brothers immediately seize on its corpse for food, we may admire this [illeg] domestic economy but shall we say the fec only which leads to female spiders to destroy the male

[This note is not transcribed in Stauffer's Natural selection. JvW]

92

& know their females, which never leave their cocoons & remain in a rudimentary & almost monstrous condition; if we justly admire this, can we equally admire the production, in order to fertilise two or three queens, of some 2000 drones, utterly useless in the hive, not even collecting their own food; not even serving as scavengers like the male wasps, & slaughtered before their natural term of life4 by their own nearest relations.

If in very many plants we admire the manner in which insects are tempted to visit the flowers, so as to carry the pollen exactly on to the stigmatic surface,—as for instance in Orchis or Asclepias, or the Kidney Bean, in which latter the Bee always alights on the left side, where the stigma lies exposed,—can we look at this end as attained with equal perfection by the pollen being blown by chance, as in our coniferous trees, on to the ovules; for this is effected by the elaboration of dense clouds of the precious granules which are wasted to such an incalculable degree that buckets-full have been swept off the decks of ships at sea. In the Dionaea we may admire the beautiful contrivance, by which the leaf-appendage

4 Desborough. on the Duration of Life in the Bee: Transactions of Entomolog. Soc. vol. 2. Part v. p. 156.

93

closes like a steel rat-trap & catches insects,—beautiful at least for the plant, if those be right who believe that it is manured by the dead insects. But what shall we say of the terrific waste of insect-life by the varnished & sticky-buds of the Horse-chesnut & other plants; the scales of which are soon blown far away by the wind with the almost innumerable insects sticking on them; on one large tree with thousands & thousands of buds, there seemed to be on an average at least four insects sacrificed on each bud. But in all these cases, if the animal or plant can successfully struggle with its competitors, the principle of natural selection is satisfied.

As in nature selection can act only through the good of the individual, including both sexes, the young, & in social animals the community, no modification can be effected in it for the advantage of other species; & if in any organism structure formed exclusively to profit other species could be shown to exist, it would be fatal to our theory. Yet how often one meets with such statements, as that the fish in the Himalayan rivers are bright-coloured, according

94

[to] an excellent naturalist, that birds may catch them! How the fish came to be bright-coloured I can no more pretend to explain than how the Gold-fish, which Mr. Blyth <informs me he> believes to be a domestic variety of a dull-coloured Chinese fish, has gained its golden tints, or than how the Kingfisher, which preys on these fish, comes to be so brilliantly coloured, without, as far as we can see, any direct relation to its habits. A great physiologist supposes that glow-worms shine that birds may find & devour them! The aphis excretes a sweet fluid, highly useful to ants, & necessary, I presume, to those species which keep the root-feeding aphides in their subterranean nests; but must we infer from this, that aphides were created for the sake of the Ants? An acute observer supposes that the nectar of flowers was created specially for insects; but here there is reason to believe that it serves as an excretion for the plant, & besides in many cases is indispensable by tempting insects for their fertilisation.

How often one sees it stated that insects produce innumerable larvae, & plants innumerable seeds, that animals

95

may feed on them; or that a surprising number of plants, as Wrangell has remarked, bear edible berries in the tundras of arctic Siberia that birds may be there supported: but is it not more reasonable that the innumerable seeds & larvae are produced that some may escape destruction, & in the wretchedly barren Siberian tundras, may not the dung of the birds be almost indispensable to many plants, or at least as good for them, as the pellicle of guano with which some agriculturists coat their seeds, or as the so-called albumen with which nature coats not the outside of the seed, but the embryo within. One author supposes that plants with pitcher-like leaves were created that animals might drink out of their contained water; but the Sarracenia grows in bogs where water abounds. One more of the many instances which could be given, will suffice: it is commonly believed that the Rattle-snake has been created with fangs to destroy & the rattle to warn its prey! In this instance, I may just remark, that in a venomous allied S. American

96

Trigonocephalus, I observed that it constantly vibrated, especially when irritated, the last inch of its tail, with sufficient force, to make a slight noise when gliding amongst dry stalks of grass: I presume that no one would think that this habit was of any more use:1 either to other animals as a warning or to itself for any object, than the vibration of its tongue, or the curling of a cat's tail when angry: now let us suppose that the little bead with which the tail of this snake, like that of many others, is terminated, were not annually moulted with the rest of the skin, but adhered only slightly to the new & larger bead formed with the new skin, we should then have the actual structure, manner of formation & vibratory movement of the rattle in the true rattle-snake; & our new rattle would be of no more use to the Trigonocephalus, & no more created to warn other animals, than its vibrating tongue or the curling of the tail in the cat or enraged lion.

Finally, although within the same class species having a nearly similar structure may be adapted to the most diverse habits, I believe that each single species has had its whole structure formed through natural selection, either in ancient time for the good of its progenitors, or

96v

* If of any use it is more likely to serve to paralyse by fear or fascinate its prey.

97

more recently for its own individual good; every modification, however, having been subjected to the laws of the correlation of growth & to the direct action of <the conditions of existence, as> food & climate. This conclusion seems to me to accord sufficiently well with the famous principle enunciated by Cuvier "celui des conditions d'existence, de la convenance des parties, de leur coordination pour le role que l'animal doit jouer dans la nature."1

Before summing up this chapter, I may remark that if our theory be extended to the utmost limits, which facts of any kind permit, nothing is easier than to make the whole appear to oneself quite ridiculous;—namely by asking whether a rhinoceros & gazelle, an elephant & mouse, a frog & fish, a bird, lizard & mammal could possibly have descended from a common progenitor. In-voluntarily one immediately looks out for a chain of animals directly connecting these extreme forms. One forgets for the moment, that these

1 Quoted from Geoffroy Saint-Hilaire Principes de Philosoph. Zoolog. 1830. p. 65.

98

great groups have been perfectly distinct for enor-mous geological periods; some of them, almost if not quite as distinct at the earliest period of which we possess any fossil records, as at the present day; & therefore if intermediate forms ever did exist, they would all, or nearly all be, assuredly now utterly lost. To lessen in some degree the ridiculous impression of the foregoing question, one ought to think of such animals as the Ornithorhynchus, which though an indisputed mammal, presents in its skeleton & other parts some few plain resemblances to reptiles & birds. When mentally comparing a rhinoceros & gazelle, one ought to bear in mind that Cuvier & all our elder naturalists considered the Pachyderms & ruminants as the two most distinct orders of Mammalia; but now Owen has so connected them by Eocene forms, that he has made them into one great group. Look at the mud-fish (Lepidosiren annectens), which is so intermediate in structure, that although the greatest living authority considers it to be certainly a fish, many highly competent judges class it as a reptile: if then there be any truth in our theory, it would not be ridiculous to suppose that the Lepidosiren could

99

be modified by natural selection into an ordinary fish, or into a reptile. The case is almost parallel with that often encountered by philologists: to one who knew no other language, dead or living, besides French & English, how absurd would the assertion seem, that evêque & bishop had both certainly descended from a common source, & could still be connected by intermediate links, with the extinct word "episcopus". Let it not be supposed that I wish to underrate the extreme difficulty of extending my theory to its utmost limits. I feel it in every sense. The utmost which I wish, is to deprecate mere ridicule,—a tempting but faulty weapon for the discovery of our universal aim, Truth.—

Summary. I think facts enough have been given,—on the unexpected transitions in the ways of life in animals of the same class,—on the diversified habits in the same species or in closely allied species, and on the changes of habit in the same species when placed under new conditions—to show how extremely cautious we should be in admitting that any animal, a bat for instance, could not have been formed by the modification

100

of another animal with totally different habits. On our theory of changes in habit or structure, due to the struggle for life common to every species, we can understand such cases, as birds with webbed feet never haunting the water, which must seem strange if every different species is viewed as an independent creation. So, also, with separate organs, I think facts enough have been given to show, what extraordinary changes in function may be effected; these changes being often facilitated by the same organ performing two wholly different functions, or changing its function during growth, or by two organs simultaneously performing the same function. Seeing the gradation in nature even in so perfect an organ as the eye, each stage being useful to its possessor, it does not seem actually impossible that such organs should have been modified & perfected by natural selection. From our ignorance of the entire economy of any one being, we ought to be very cautious in concluding that any part is too insignificant to have been formed by this same principle; seeing that the part might have been perfected during the life of an ancestral species to which it was of the highest importance, & seeing that natural selection might slightly act on the variations

101

of a most insignificant organ, when accidentally concurrent with other advantages & disadvantages. Even the extraordinary difficulty of neuter insects differing in structure from the fertile females, & being divided into castes in the same nest, can hardly be considered actually fatal to our theory, if we consider what man could probably effect & indeed has effected under somewhat analogous circumstances by his feeble powers of selection. Nor can the rare cases of closely similar, but not strictly homologous organs, in organic beings far remote in the scale of nature, be considered as fatal; for the same means of natural selection acting on nearly the same materials, might sometimes hit on the same result. Considering the vast number of extinct forms, it is surprising that far more numerous cases cannot be readily found, of organs without any known transitions serving to indicate the probable steps by which they were formed. The extreme rarity of the appearance of any quite new organs in a class, is an astonishing fact, as long as we look at each of the innumerable

102

living & extinct species, as independent creations, but gives great support to our theory of gradual modification. Organic beings seem to be perfect only in that degree required by our theory, namely to be enabled to struggle with all competitors in their native country. If we trust our reason, which fills us with the most lively admiration for very many adaptive contrivances, others, like the sting of the Bee or the wasted pollen of coniferous trees, can hardly be considered as equally perfect. We have no good reason to believe that any organic being has been created for the good of another species, though so many mutually profit by each other. The doctrine that each part in each species has been formed (subject to the laws of growth) either for the good of its progenitors or for its own good, accords sufficiently well with Cuvier's principle of the "conditions of existence", & seems to fulfil all that we really see in nature.