(1

(Ch. 4. Variability natural Definition of Species.)

In this Chapter we have to discuss the variability of species in a state of nature. It would The first & obvious thing to do would be to give a clear & simple definition of what is meant by a species; but this has been found hopelessly difficult by naturalists, if we may judge by scarcely two having given the same.

I will copy the latest & most laboured definition by Alph. De Candolle (Géographic Botanique p. 1072) who has carefully discussed the subject in relation to plants: he says species are "collections d'individus [See Stauffer for the rest of the quote, pp. 95-6]

M. De Candolle lays stress on f making the element of descent subordinate to that of resemblance, so that the definition may be less hypothetical. But as animals & plants must be here equally considered, I agree with Dr. Carpenter who gave at Glasgow to the British Association an interesting lecture on this subject, that descent does come in as a prominent idea. Although when speaking of the resemblance of two forms, the comparison should of course apply extend to all ages & sexes, yet it is as zoologists

(2

(Ch. 4. Variability natural — Definition)

have often described these stages as specifically distinct, an error instantly corrected when their descent was ascertainedknown, it is very natural that they should havebring this idea prominently forward. Thus if the development of      had not been known, asthe stages through which it passes, as M. Quatrefages* (Revue des deux Mondes     ) has remarked, would have been considered as forming eight distinct genera: I am convinced that a in the cirripede Ibla without knowledge of its descent, the male & female & its two larval stages would have formed four distinct Families in the eyes of most systematic naturalists. Again the most ill-shapen monster is rendered home to its species the instant we know its parents.

Let us test M. de Candolles definition with a plant. We shall presently see that there Assuming for the moment that it was demonstrated (& we shall presently see that the evidence is almost as very strong as can be conceived) that the primrose & cowslip descendcan be produced from the same stock; would they be called by any Botanist distinct species in the ordinary acceptation of the word? Yet thesethe two former causal here do not said to accord in any one respect with M. De Candolles' definition as above given, of one & the same species; for (1')

(3

(Ch. 4. Variability natural — Definition)

the individuals of the cowslip & the individuals of the primrose cowslip & primrose are considered by most accord in every single respect with M. Decandolle's definition of two distinct species: for (1st) the individuals in each theyagree in many important characters, which are constant during many generations under different conditions, for they are found growing distinct in curiousdistinct parts of Europe: (2d) they do not fertilise each other with facility, as the best experimentiser, who ever lived, Gærtner, found after repeated trials during many years: (3d) they do not behave in the same way in regard to temperature & soil, for they have different ranges & inhabit different situations (4th) they cannot be said to resemble each other as much as analogous plants do, which we positively & habitually know to have descended from a common source. Hence I conclude, that the idea of descent is a prominent idea inside the word species as commonly accepted.

The idea of descent almost inevitably leads the mind to the first parent, & consequently to its first appearance, or creation. We see this in Morton's (followed pithy definition of "primordial forms", adopted by Agassiz. ?  The same idea is supreme, & resemblance goes for nothing, with those zoologists, who consider two forms, absolutely similar as far as our senses serve, when inhabiting distant countries, or distant geological

(4

(Ch. 4. Variability natural — Definition)

geological times, as specifically distinct. They argue Having the idea of the first appearance of a form prominently in their minds, they argue logically that as most of the forms in the two countries or times are distinct, in some greatly distinct in others very slightly the distinction being in some great, in others less & less, they naturally ask, why forms apparently distinct absolutely identical should not have been separately created, & which they in consequence would call distinct species.—As we are we have to discuss in this work whether forms considercalled by all naturalists distinct species are not lineal descendants of other forms, this minor question if will fall or rise with the greater question; & is here only alluded to in connection with the definition of the word species.—

Short as this discussion has been, De thinks it will have been seen how many ideas, & those ideas with very

Some authors, as Kölreuter, take the fertility of the offspring of two forms as the sole or leading test of what to consider as species; & however unlike two forms may be, if they produce quite fertile offspring, they consider them as specifically the same. The great importance of this distin difference in fertility in what are ordinarily called varieties

(5

(Ch. 4. Variability natural — Definition)

& species, has in my opinion of late years been much undervalued by some authors. In the chapter on Hybridism we shall fully consider this subject & we shall find that there are great difficulties (I do not mean merely practical ones in its application) in taking in lessened fertility in the offspring as an unerring guide what forms to call species. I will here only remark, that the perfect fertility & utter sterility glide into each other, in so insensible a manner that it is hardly possible to draw any line; hence the two most laborious experimentisers who ever lived, Kolreuter & Gærtner after numerous experiments in regard to certain forms, have come to diametrically opposite conclusions; the one concluding that certain forms are varieties, & the other that they are undoubted species.—

Short as this discussion has been it suffices, I think, to show what how various are the ideas, all & as in that enter into the minds of naturalists when speaking of species. & without different degrees of prominence, (a) At the end of this chapter, it will be seen that according to the views, which we have to discuss in this volume, it is no wonder that there should be difficulty in

[5v]

(a) With some, resemblance is the reigning idea & descent goes for little; with others descent is the infallible unfailing criterion being; with others resemblance goes for almost nothing, & Creation is everything; with others sterility in crossed forms is an unfailing test, whilst with others it is quite ignored regarded of little no value.

(6

(Ch. 4. Variability natural — Definition)

defining a species the difference between a species & a variety;—there being no essential, only an arbitrary difference. In the following pages I mean by species, those collections of individuals, which have commonly been so designated by naturalists. No one doubts Everyone loosely understands what is meant when one speaks of the cabbage, Radish & sea-kale as species; or of the Broccoli, cauliflo & cauliflowers as varieties: between these such extremes there is often a wide neutral territory in which the term species & varieties are bandied about according to the state of our knowledge & our ideas of the term species. —

Botanists in discussing the subject of variation have usually treated of included together that occurring variation which occurs under domestication & that under natural conditions; & which this is probably the best plan, though not for our particular subject object. They have divided* (M. Alp. De Candolle has given a full & exc discussion on this subject. Geograph. Bot. p. 1078) varieties into "variations" in which the varying characters are not fixed even in the individual plant, all the buds produced on the same plant being here considered as one individual. In animals we have very few instances of this class; but but as the black colour in cage birds produced by hemp-seed goes off with change of food; & slight changes in the

(7

(Ch. 4. Varieties classes of natural)

hairy covering of animals when transported into a different climate* have been observed.

In Bernhardi's classification (Ueber den Begriff der Pflanzenart 1834. p. 5) The term "Abanderungen" is applied to many similar cases, or to those, A The term 'Variety' is applied to forms often offering considerable differences, & which can be securely propagated by buds, grafts, cuttings, suckers &c, but which are supposed believed not to be hereditary tre be inheritable by seed. This class nearly corresponds with "Spielarten" "abanderungen" in Bernhardi's classification (Ueber den Begriff der Pflanzenart 1834. p. 5) & likewise in which the form tends to go back is not hereditary or only so in certain soils; & likewise in a lesser degree with his "Spielarten" in which the form tends to go back in one or more generations to the parent type. As we know scarcely anything of the as variation of those lower animals which can be propagated by buds division, the class of "Variety" in the above strictest sense is not applicable to the animal Kingdom; though no doubt in all possibility no doubt, in the less strict sense of being hereditary in only a slight degree, there are very many cases in the amongst animals Kingdom, & some even in a state of nature. Lastly we have the class "Race", corresponding with "Abarten" of Bernhardi

[7v]

*The cat in West Africa

(8

(Ch. 4. Varieties classes of)

& with subspecies of some authors, in which the form is strictly strictly inherited, often even under changed conditions; of this class we know there are plenty under domestication, some known, & more suspected in a state of nature, as in the geographical races of some Zoologists. But the term subspecies is used by some authors, to express define (& corresponds in this sense with "unterart" of Bernhardi) very close species, in which they authors cannot determine whether to call consider them as species or varieties. The existence of these doubtful forms has lately been explicitly admitted by M. Alp. Decandolle in regard to plants, & by implication by Mr. Wollaston (on the Variation of Species, p. 185) in regard to insects: M. Decaisne & Dr. Hooker use the term without expressing more than that the difference between such forms subspecies is slight, yet permanent. As these authors are of the highest authority, this fact admission is important as subspecies fill up a gap, between species, admitted by everyone & varieties admitted by everyone. Between varieties derived & individual differences there seems a gradual passage but to this subject to which we shall recur. And moreover In species we should remember how extremely close some undoubtedly species distinct forms are to each other, as many plants, & as in some of the willow wrens, which are so close that the most experienced ornithologists can hardly distinguish them except by their voice, & the materials with which they line their nests; yet as these wrens inhabit the same country & always exhibit the same

(9

(Ch. 4. Varieties classes of)

difference, no one can doubt that they are good species. So that between individual differences & undoubted species the naturalists have made various short steps.

(In the above classification the of several varieties the main difference rests on the hereditariness of the characters. Though the classes blend insensibly into each other, this classification is are of some use when applied to domestic productions; & no doubt they are applicable to it holds good in varieties in a state of nature, which we are here considering. But it seems to me that we are far too ignorant to apply this in the it to natural varieties under natural conditions, more especially in regard to animals. We have seen in our first chapter that the same character is inherited in very different degrees by different species, & even in different individuals of the same species; we have reason to suppose suspect that a character becomes more fixed by long continued generation; although on the other hand, a character suddenly appearing is sometimes strongly inherited. Who can tell how much the dwarfed character of a plant or the dark colour of an dark colour of an insect on a mountain, or the of a shell in brackish water (a) is due to inheritance & how much to the exposure of the individual from its earliest days to the condition in question. Probably in all such cases, the

[9v]

(a) or of the improved character of the fur of Beavers Martins &c the further we go north (Kalms Travels vol 3. p. 58)

(10

(Ch. 4. Varieties classes of)

form would change when placed in th under other circumstances; & some in fewer generations than others; & others than in others & some in imagination but then it might be argued that this was not a fair test, as many species races or hereditary vars strongly hereditary varieties change in some degree under new conditions. I am inclined to believe that with the rarest exceptions every changed structure is in some degree inheritable. In animals perfectly black individuals are occasions not very rarely born, even in the same litter with ordinary coloured individuals: I know & in cer some cases places these appear much more frequently than in others, thus I am informed by Mr. Crawfurd that black Leopards are far more commonly produced in Java, than elsewhere; & in such cases I know not whether to attribute this to a strange hereditary principle, or to some unknown conditions acting on the parents. Fish of the same species are perfectly well known to present distinguishable differences in different lakes: Sir H. Davy (Salmonia p 53) states that

states that a red-fleshed dark-banded trout were placed taken from one Scotch lake & put into another, where the trout were white-fleshed; but the young were here produced had their flesh less red, & in 20 years the variety was lost.

Laying on one side the probability of crosses having taken place, we see here that the red flesh was in some degree inherited; & some would assert that if the red-flesh trout in their own lake had been inherited

(11

(Ch. 4. Varieties classes of)

transmitted their characters for some additional hundred-thousand generations, the character would have kept truer. From these & similar considerations I have thought it advisable to use only the term "variety", & where it is known or almost known to be strictly inherited "race": and I use the term variety loosely, simply in accordance with common acceptation, as I do the term species. for the same reason in both cases If the distinction could be drawn between hereditary & temporary variation in a state of nature it would be of great importance for my our object; for variations in a state of nature which are not inherited are of little signification, & deserve notice, (perhaps) only as showing the possibility of change in structure.—

Practically the systematic naturalist, without troubling himself more than he can help about descent & creation, considers those forms distinct as one species which he cannot unite by other intermediate & graduated step forms. It is his golden rule.

But those who have not themselves worked, can form little idea of the irksome labour required in its application. For example look at the case of Aquilegia vulgaris, as wh as worked out by

(12

(Ch. 4 variation in nature)

Dr. Hooker in his Flora Indica, who devoted weeks to the examination of specimens from all parts of Asia & Europe, & who ends in uniting about 16 species of other authors into one. I may state, as I know that th similar cases have happened occurred with others, that in certain & species of the garden Lepas anatifera & Balanus tintinnabulum.

I at first wrote out full descriptions of several supposed species; then after getting more specimens from various parts of the world, I was found thought that I ought to run them from all into one species, & tore up my separate descriptions: after an interval of some months I looked over my specimens & could not persuade myself to call such different forms one species & rewrote separate descriptions; but lastly having got still more specimens, I had again to tear up those & finally concluded that it was impossible to separate them! When the naturalist has got the intermediate forms between two supposed species, the work though laborious is generally simple; but he is very often obliged to judge by analogy. And here enters springs up an endless source of doubt. From On how far different from widely distinct groups may he draw for his analogies?

(13

(Ch. 4 variation in nature)

it is a remark repeated in almost every systematic work, that the very same organ whether or not of physiological importance will be constant in one group & so afford good specific characters, & will be highly variable in another. His power of drawing analogies will not only obviously depend on his amount of knowledge, but on the tendency frame of his mind to trust the analogy. Is it then surprising that naturalists should differ in the extreme degree in which they do, in determining what forms to call by the various defined & recognised term species?

I have remarked that generally when the naturalist has got intermediate stages he unites with confidence two forms distinct in appearance. But here, also, he sometimes has cause to doubt. The intermediate forms may be hybrids; nor can he by: these he may often recognise by their sterility, but by no means always, at least without counting their seeds & comparing them in number with those of both presumed pure parents; but Gærtner thinks or he may discover that they are not hybrids by one of the supposed parent forms not growing in the neighbourhood;

(14

(Ch. 4 variation in nature)

that a hybrid should be artificially made for comparison. [This line was transposed to p. 13 by Stauffer as a continuity to 'but Gærtner thinks or', see F1583, p. 101]

But independently of this source of doubt, which perhaps has been overrated by some authors, there is another & more important one, namely the probability of one of two forms, or of two forms which would deserve in every sense to be called species, both varying greatly & so running so closely together that the extreme varieties become undistinguishable. This is the more probable probability in this, as we shall afterwards see that [illeg]that, certainly varieties of one form tend to mock the characters of other species in the same group. To give a very few examples: Dr Asa  Drs. Torrey & Gray, & it is impossible to quote a higher authority, speaking of the varieties of two N. American Asters state that in the so run together that it is impossible to dist draw a distinct line of demarcation, & yet he cannot doubt that they should be considered as distinct species. in speaking of the N. American Asters say "that several species, which we cannot but consider as distinct do frequently present very puzzling intermediate forms; & that an apparent transition is not always real."* *(Silliman's American Journal of Science vol. 40. p 280) — Such cases but no so more or less striking case do not seem to be very rare, for even in the small British Flora, Mr. Hewett C. Watson has marked for me 16 16 12 cases, (not including the protean forms in Rubus, Hieracium &c) in which two species & in some cases three species apparently distinct are

[14v]

(a) ⸮ Dr Hooker Two thistle has been considered perfectly distinct by everyone ⸮ since time of Linnæus, yet Dr. Hooker found so perfect ⸮ a blending series, that he takes a line conclude to run them together: the two parents ⸮ are not found there, but no reason can be imagined that these species shd from hybrids been & no where else

(15

(Ch. 4 variation in nature)

united, more or less perfectly, by intermediate forms:*a to give a single example,—Geum urbanum & rivale are universally thought to be distinct; but between them we have the var G. intermedium (considered a distinct species by some authors) & several intermediate forms, which it seems impossible to so breaking down every character between the two types: in this instance Dr. T. Bell Salter has attempted to show has stated (Phytologist vol 4, p. 739) that he produced G. intermedium by by a hybrid, but as find it when one crossing the above two species; but from observations in the Flora 1848 p. 42, in regard to the absence of the two parents did not grow, in a place where G. intermedium was found, we perhaps have here two distinct cause origins of the connecting links, making the confusion doubly confounded.

Mr. Watson, who has attended paid the closest attention to the subject under discussion,*(b) & to whose assistance I am under great obligation, in one a letter, which he has permitted me to publish, has divided the pointed out in a very clear manner the following four categories in our British plants.

[faint pencil note:] in Mr Watson note

[15v]

*a An admirable paper entitled on the Theory of Progressive Development from which I have largely borrowed views & facts by Mr. H. C. Watson on the relations of species to each other & the varieties is given in the Phytologist 1845 p. 140 & 161.

[15vv]

*a Mr. H. C. Watson has given me a list of examples divided into three groups. (1) of varieties passing into two species (2) of two species actually passing graduating into each other by intermediate varieties. (2) of two species closely approximating to each other by intermediate varieties.

(3) & more commonly of one two species varying & its varieties passing assuming some of the characters of the other species, either positive or negative, but without actually passing into that other species.

[15a]

[Enclosure in a letter from H. C. Watson, dated 13 August 1855.]

(1)

Categories of Species

1. Plants distinguishable from each other by positive characters, & generally received as true species.

2. Same as No 1; but so closely resembling each other as to be frequently mistaken one for the other, & by botanists even of some experience.

3. Same as No 1; & not liable to be mistaken in their typical forms; but accompanied by intermediate or transition forms, approximating so much to each or both, as not to be quite satisfactorily assigned to either. (N.B. The primrose & cowslip would be in this category, but it has been there proved that the intermediate produces both the alleged species from the same year's seed)

[15av]

4. Plants deemed true species where their typical & most general forms only are looked at; in but the limit of the species is rendered uncertain by the existence of forms closely allied, deemed varieties of the type by some botanists, distinct species by other botanists.

As with is the case with the intermediates of no. 3, so these varieties or qua sub-species (3,4) being generally of no 4 are usually much more rare or local than the type species. They differ from the intermediates of No 3 only in so far as the as varieties or quasi species clustering around one, instead of linking together two supposed genuine species.

[15b]

Altho' four such categories are easily defined on paper, & illustrated by selected examples, they glide together by other examples; & thus, as groups, they are different in degree rather than in kind. To show this by garden species, &c. give examples of the four categories, familiar to most bo   ,

1. The Apricot, plum, & Cherry are commonly placed under one genus, Prunus; & as species these are very readily distinguished by any body. 2. But there are two Cherries spontaneous in England, an arborescent & a fruticose, which by most botanists are deemed two real though very similar species, & between which in a wild state we can hardly point out any connecting links.

[15bv]

(4)

3. Many botanists deem the wild sloe of England to be quite a distinct species from the cultivated & probably imported plum-tree of the gardens. Nevertheless, between the finest garden plums & the little austere slow plum-tree of the garden & the sloe-bush of the hedges, there exist numerous intermediate forms or links, which render it highly difficult to say, 'here ends the sloe & its varieties, there ends begins the plum & its varieties.' If we hold the intermediate Bullace a good species, it this also passes insensibly down to the sloe, & improves almost as insensibly into the plum, so numerous & fine are the steps or links either way. 4. Linné described the fruticose bramble as a species, under name of Rubus fruticosus; but various modern botanists make out 50 to 100 supposed species of Bramble which others call varieties of R. fruticosus, & [two words illeg] others again group into a small number of species, say half a dozen.

(16

(Ch. 4. Variation in nature) Individual differences)

With respect to the blending of two or more apparently distinct species by intermediate forms,

To the naturalist who looks at species as not essentially differing from varieties, being only more permanent, with the connecting links extinct, the occasional blending by intermedial forms of two or more apparently distinct species, will not be wonderful; indeed the wonder is to us, with our restricted notions of the lapse of time, that many more cases do are not occur on record.—

Individual differences.—Besides the differences which are varieties recorded by naturalists as varieties we have individual differences,

[faint pencil line that spills into the right margin here]

which are not thought worthy of separate notice, either from being so slight, or from being believed to be so little permanent or forms not graduating or blending into each other so that they cannot be reco near divided even into distinct varieties. (a) I think that these

Individual differences as I will call them from being generally very slight compared with the difference between species are sometimes ori sel have not I think not been fully always been sufficiently bore in mind noticed by experienced naturalists when discussing the subject of variation I think one is apt, except in very variable forms after a short preliminary study to forget them; but let any one take up collect specimens in almost any group of beings, about which he is profoundly ignorant, & he

[16v]

as does Mr. Wollaston (Variation of Species. p. 5) when speaking of his "technical" varieties

[16av]

In other cases, when this has not been done, it might be easily effected, especially if a few of the intermediate forms were rendered extinct to become lost; thus as remarked to me by Mr. Watson in regard to Polygonum aviculare.

But on the other hand if we take the extreme case of well marked & permanent varieties, & the difference, just perceptible though hereditary, between a brother & sister organism, some such distinction does exist, as no one would put these differences into the same class. M. Boreau, who has so carefully studied the Flora of France (Flore du centre de la France 1840 p 101) has called attention to this distinction & says "les varieties proprement dite sont plus trancheés",

[16bv]

(p. 16)

(a) Nothing can be looser than this distinction; no doubt a multitude of what perhaps should be called individual variations, with no degree of permanence often figure as recognised varieties; moreover it is quite a common practice with naturalists to pick out of blen a graduated & inextricable mass of forms, a few leading forms types & designate them as varieties ↘ .

[For readability, Darwin's horizontal and vertical deletions are left uncrossed]

 

On the other hand but if we take extreme case, some such distinction does exist, as no one would put in the same class differences though hereditary in a brother & sister only just perceptible, unite strong method & nearly or quite parent varieties.

(17

(Ch. 4. Variation. Individual differences)

will be for a short time, at least I have been, utterly perplexed to giv tell what are individual & what specific characters. This indeed is tacitly acknowledged by every cautious naturalist, by their dislike to define a new species, without it be some strongly marked form, if he possesses only a single specimen.

I have been in the habit of during many years of marking in all careful monographs & works, in which measurements have been given of several individuals, with care taken to mark note sexes & age, & I cannot doubt that individual differences are very often considerable; & no one doubts that this is the case with plants It is impossible to give instances: many cases might be selected from Mr. Waterhouse's excellent work on two great orders of Mammalia, & likewise in Macgillivray's elaborate work on British Birds.

I will refer only to one other instance, as it, also, refers relates to birds, generally considered, & I believe truly, as very fixed in form: Graba (Tagebuch auf eine Reise nach Färo 1830 s. 103: he gives details of measurement of beaks &c of Anthus s. 56 & 67.—of beak & tarsi in Larus s 65 & 80.—& in Colymbus s. 118 &c) who particularly attended to this subject, says that he shot hundreds of seabirds at Faroe & that he seldom omitted to measure every one, & the result that he came to was that rarely did two individuals of the same species agree throughout in their measurements.

(18

(Ch. Individual variations)

These individual differences variation differences differ in amount to a surprising degree in various species & in various groups of species, one part or organ being affected in one species or group, & another the same part being very constant in another set of species Some forms are extremely constant in their whole organisation others as variable, causing to the naturalist an odious amount of perplexity. Generally the characters which individually vary, are of slight physiological importance, but this is by not means always the case; & I will immediately give a table of some of the more important & curious cases of these individual of variation (the slighter ones not being worth notice) which do not seem to be l characteristic of any breed or variety, separately & therefore are not marked as separate varieties by naturalists.

Individual varieties differe

But here arises a perplexing question; are these individual differences of the same order, & have they the same origin as those other differences, either greater, rath more permanent, or less closely linked together, which separate recognised varieties.

[For readability, Darwin's horizontal and vertical deletions are left uncrossed]

Many authors seem to consider that each species was created with a certain fixed amount of variability, or to use an expression in a letter of M Prof. Dana, with "its system of librations under the influences of nature to which it may be subject", & this would include both recognised varieties & individual variations. No one will pretend that any clear line of demarcation can be drawn between these two classes of facts; but some authors as Dr Prosper Lucas think

(19

(Ch. Individual variations)

that the production of slight differences is the normal & invariable function of the reproductive system in all organisms, independently of their conditions of existence; & the generality of the fact universality of some slight individual differences countenance this conclusion; but this view I presume no one would extend to marked varieties, which seems to show that these & thus even a fundamental difference between individual differences varieties seem to be indicated. But to me it seems a simpler view to account for all individual differences, which cannot be explained by differences in the parents or more remote ancestors, by the effects of external varied conditions acting on the parents more remote ancestors ancestral forms & thus affecting indirectly (as we have seen in the last chapter) the reproductive system & consequently its products. in the same manner as others & greater changes affect this system so as even to render it. According to this view if we could start with quite similar organisms & bred them for many generations during their whole lives under absolutely similar conditions, the offspring would be absolutely similar; & as a consequently individuals we should look at all individual differences having (independently of those produced by crossing) as having the same nature & origin with those marked by naturalists as varieties.

In favour of this view we have the broad facts that there is much more individual variability as well as distinct varieties of all kinds in domestic

(20

(Ch. 4 Individual variations)

productions, than in those under their natural & unchanged conditions. M. Boreau thinks that it is the plants very common plants, which vegetate in all places & under all exposures, which offer innumerable slight differences. It is certain that some species which are extremely constant in one area are extremely variable in another: thus the Helix aspersa one of our most constant land-shells in the South of France, as I am informed by Sir C. Lyell is very variable; & many instances might be given. On the other hand the general impression which I have taken, is that a variable species is in all places & all times variable; but I have not met with careful observation on this subject head. Variable sea-shells are seem to be variable everywhere, but these there is in most of these cases are attached shells, as limpets & oysters & cirripedes & they would everywhere be modified by the surfaces of attachment. In Beetles Coccinella seems everywhere variable in its spots spotted colouring. I applied to Dr. Hooker on this subject & he went through ⸮ ⸮ the Tasmanian & New Zealand Flora with this idea, & he found that those genera which were very variable in Europe were there also very variable; ⸮ but in the Himalaya, the species of Willows, Rubus, (letter)

(21

(Ch. 4 Individual variations)

Senecio Gnaphalium, which are so eminently variable in Europe & in N. America were there not so.

(a separate Page)

These facts, & more especially the existence in every great group of or great division class of organisms of groups of species adapted to varied conditions & growing in different countries eminently variable, as the Brachiopoda & the genera of plants just mentioned & many others as Hieracium breathe &c certainly & as in the Brachiopoda in various geological formations seem to p plainly indicate that the variability is here innate & independent of the conditions of existence, & in insects as such species or that according to the common view, that they have been created with this tendency, with others &c each having its own system of libration to use an expression of Mr Prof. Dana in a letter to me. But this tendency can seldom be predicated of every species in the variable group; thus even in Rubus, the R.     is a very fixed form: in the eminently variable genus of shells, Pleurotomaria M. re Eudes-Deslongchamps (Mem. de la Soc. Linn. de Normandie Tom. 8. 1849. p. 23) states that some are very constant vary hardly at all, some, so to speak without any limit. How variable are the species of Squirrels, yet Dr Bachman who has so carefully studied the N. American species, says some informed me that several some are very true to their characters.

As under cultivation forms are often produced which are characterised for the by being variable, it was

[21a]

If the great variability be looked at

(21

Insert in 9a) between p. 21

(I have applied, also, to Mr. Davidson, whose vast experience in Brachiopodous shells, makes his opinion of the highest value & I find he has specially attended to this subject & is equally puzzled by it equally with myself: he says that certainly some many fossil shells, as Spirifer rostrata of which he has examined vast numbers of sh specimens from various places & periods, present everywhere the same quite extraordinary amount of variability: on the other hand some other shells of this Family same order vary but little either in time or space: innumerable examples could be given of the foregoing cases & this I found was the opinion of all that I could learn on this subject from the late Prof. E. Forbes & f from Mr. Woodward. Under certain conditions the same species, of which Mr. Davidson has given me examples, will be very variable in one space & constant in another: thus, also, Mr. Searles Wood, who is so intimately acquainted with the Crag fossil shells, informs me that several species, from the Mammaliferous stage are remarkably variable more so than the same shells at the present day, & which he is inclined to attribute to the site former estuary conditions of the site: on the other hand Mr Wood has not found the same degree of variability in the Eocene estuary shells of Hampshire.)

(22

(Ch. 4. Individual variation)

is perhaps possible, according to the views we are examining in this work, to account for groups of variable species from by their descent inheriting this tendency from a common parent; but I am not satisfied with this conjecture. (a)

If it could be rendered probable that in the course of time some one or two of the individual forms of an a species individually very eminently variable species could might become fixed, then with the extinction of the intermediate forms we should in some cases see the stages & in some cases better understand the origin of some of of the more permanent varieties. In fact of and between them The fact occurrence of certain constant species in the most variable groups would harmonises with such a view.—M. Lecoq, I presume is of opinion that this would happen, for he speaks & likewise that the more fixed varieties do not differ from would be converted into & deserve to be called species, for he speaks of such genera as Rubus; as for being genera in process of formation. But I must leave the case of these many Protean genera groups an open question; not doubting, however, that in very many instances there is no real distinction in nature or origin between individual variation differences & more strongly marked & permanent varieties.

(23

(Ch. 4. Individual variation)

(small type)

I will now give a few selected examples of indivi individual variation or differences, not known to characterise any a recognised variety; & I shall select them from various motives, some from the physiological importance orga of the part organ affected, or from such part being in the group in question generally constant &c.

But I must here state that bear Several other cases might have been added, & will be subsequently given, illustrating the variability of rudimentary organs, of greatly developed parts, & of sexual characters &c.

One chief object in the following list, is to show that the common remark that organs called important by naturalists never vary is not quite correct, but anyone, [illeg] unacquainted with natural History, who might infer that because this or that part varies in certain species given as examples, it would likewise vary in other groups, would err greatly.—

(24

Ch 4

In Utricularia nelumbifolia, (Silli (Dr. Asa Gray, Silliman's American Journal vol. 45. p 215) in the perfect (sexual) flower, especially where only one stamen is antheriferous the anther is commonly found to be one celled. The lobes of the style are variable in number, as are the scales of corolla & calyx.

In Zannichellia palustris (Sir W. J. Hooker & Arnott's British Flora 1855 p 486) "the form of the stigma the length of the style, the number of anther-cells… the fruits more or less stipitate are very variable."

In the Fag common Beech Fagus sylvestris (Linnæan Transactions vol 5. p. 232) Persoon has described a wild variety individual with extraordinary large leaves & fruit, & another with the bark & manner of branching so precisely like an oak, that the country people consider it a cross.

Prof. Vaucher says that he has found the kind of gemmation (Mem. Soc. Phys. de Geneve Tom. p. 300) is with with one exception always the same in the same species of tree, & that it generally is a generic character; but that in the common Lilac, Syringa vulgaris, he has observed two forms, "bourgeon terminal" & "presentant ruptures".

(25

Ch. 4.

In FD (Papaver bracteatum & orientale (—Decandolle, Mém. Soc. Phys de Genève. Tom 2. Part 2. p 127) present indifferently two sepals & four petals or three sepals & six petals, which is sufficiently rare with the other species of the genus.)

[Origin, p. 155, see also Darwin's abstract: CUL-DAR73.9]

FD (In the Primulaceæ, & in the great class to which this Family belongs (Duby Mém Soc. Phys de Genève Tom. x. p 406) the unilocular ovarium is free, but M. Duby has often found individuals in Cyclamen hederaefolium "ou la base de l'ovaire etait soudee jusque a un tiers de la longeur avec la partie inferieure un peu charnue et dilatee du calice.")

FD (M. Aug. St. Hilaire (Mem. Sur le Gynobase, Mem. du Mus. d'Hist. Nat. Tom x. (1823) p 134) speaking of some bushes of the Gomphia oleaefolia, which he at first thought formed a quite distinct species, says, "Voila done dans un meme individu des loges et un style qui se rattachent tantot a un axe vertical, et tantôt a un gynobase; donc celui-ci n'est qu'un axe véritable; mais cet axe est deprimé au lieu d'etre vertical."He adds (p 151) "Tout ce qui precédè n'

[Faint pencil note:] I forget & I am not sure that this has bearing.

(26

Ch. 4

indiqueroit-il pas que la nature s'est en quelque sorte essayé dans la famille des Rutacees a former d'un seul ovaire multiloculaire, monostylé et symetrique, plusieurs ovaires uniloculaires munis chacun d'un style." For And he subsequently shows (p. 364) that in Zanthoxylum monogynum "il arrive sou vent que sur le même pied, sur la même panicule on trouve des fleurs à un ou deux ovaires." And that this an important character, from the Rutaceæ, to which this ge Zanthoxylum belongs being placed "dans la cohorte (Tom. XI. p. 48) a ovaire solitaire."—The same author (Tom XI. p. 49) referring to this same subject says character differing in the different species of Helianthemum, states that in the H. mutabile "une lame, plus ou moins large, s'etend entre le pericarpe et le placenta.")

(27

FD (De Candolle has divided the Cruciferæ into five sub-orders in accordance with position of radicle & cotyledons, yet M. Monnard J. Gay (Annales des Scien. Naturelles 1. S. Tom 7. p. 389) in found in 16 seeds of Petrocallis Pyrenaica the form of the embryo so uncertain that it he could not tell in which of the two sub-orders it whether it ought to be placed in "pleurorhiée" or "notorhizée": so again (p 400) in Cochlearia saxatilis M. Gay examined 29 embryos & of these 16 were rigorously "pleurorhizées" 9 had character intermediate between pleuro- & notorhizees & 4 were pure notorhizées: of a few other examples are given of variability in a character of great importance in this large Family.)

In the Cruciferæ it is well known, that Bracteæ are generally absent, but these have been observed (Henfrey's. Botanical Gazette vol. 3. p. 82. & vol 1 p. 307) in Car certain individuals of Cardamine pratensis, in Erucastrum Pollichii & in (cultivated) wall-flowers.—In regard to bracts, I may add that W. Herbert (Journal of Hort. Soc. vol. 2 p. 283) says that there are natural & cu varieties natural & arising from cultivation of Crocus aureus, with & without bracts.

[faint pencil note:] Hooker says there are species of Crocus with Bracts.

(28

Ch. 4

The insertion of petals & stamens is a character of high generality; but M. J. Gay (Ann. des des Sci. Nat Tom 3. (1s series) p. 27) found in Arenaria tetraquetra, that in var. uniflora, which is polygamous, that in the hermaphrodite flowers the insertion was ambiguous neither visibly perigynous or hypogynous, whereas in the female individuals, the insertion was perigynous: in var. aggregata (thought by some to be a distinct species) the insertion was ambiguous in all the individuals.

FD (M. Raspail asserts (Annal des Sci. Nat 1s ser. Tom 5. p. 440) that a grass Nastus Borbonicus is so eminently variable in its floral organization, that the varieties might serve to make a Family with sufficiently numerous genera & tribes,— a remark which shows that important organs must be here variable.)

In Globularia nudicaulis (Cambessèdes in Annal. Nat. Scien. 1 ser. Tom. 9. p. 15) the upper lip of the corolla varies remarkably, being sometimes entirely wanting, sometimes very small & divided to the base.

(29

Ch. 4

In some species of Hernaria (Decaisne, in Annal. des Sc. Nat. 1' ser. Tom. 22. p. 97) on the same individual, the ca divisions of the calyx has are regular or irregular with four or five sepals.

In Suæda, the vertical or horizontal position of the seeds in the pericarp has been thought a character of some importance, but M. A. Moquin (Annal. des Soc. Nat. 1 Ser. Tom. 23 p. 274) found in that S. altissima "presente des grains tantot droit, tantôt obliques et quelquefois couchées." With the different position of the seeds the point of attachment position of the umbilicus varies.

(30

FD (M. Milne Edwards (Annal. des Science. Nat. 1 Series. Tom. XV 16, p. 50) has given a curious table of measurements of 14 specimens of Lacerta, & taking the length of the head of standard, he finds, neck, trunk, tail, anterior front & hind legs, second toes of posterior legs, colour & femoral pores all varying wonderfully; & so it is more or less with eleven other species. So apparently trifling a character, as the scales on the head, being affording almost the only constant character.)

Mr. Couch (Linn. Transacts. vol. XIV p. 73) has seen the common ling Gadus molva with two cirri on the throat & G. mustela with five barbs.

The eggs of many Birds, especially of the Crow genus, of Shrikes, & Gulls vary in tint of colour, in spotting & size, even sometimes in the same nest.

(Linn. Transact. vol XV Part I. p. 9.)

See, also, for numerous cases W. C. Hewitsons British Oology where the variations are shown by coloured Plates.

The length Beak of birds, though generally so constant in character that most of the systematic divisions are founded on it, varies sometimes considerably in length; & I was shown in the British Museum by Mr. G. R. Gray three examples of

(31

Ch 4

a Nutcracker (Nucifraga) shot in some forest, with beaks of remarkably different length: he showed me, also, a Himalayan Nuthatch (Sitta) with beaks similarly varying. I observed the same fact in two S. American birds (Zoology of Voyage of Beagle: Birds p. 66, 67) the Uppucerthia & Opetiorhynchus.

The tooth conspicuous character of the tooth on the upper mandible, varies in some Hawks, as in the Jer Falcon. & in the [illeg] (Sir J. Richardson & Swainson, Fauna Boreali-Americana. p. 27) see 31

x FD (In whole Families of Birds the number of tail feathers is constant; but in some, as in Swans & in some Gallinaceæ the number is variable; & this is the case according to    (Isis    ) in many short-tailed Birds as the King-fisher: in the N. American coot ( Fulica Americana, in Richardson's Fauna Bor. Americ. p. 404) the number varies from 10 to 16.) In some Hawks & Owls, the proportional lengths of the primaries, a character perpetually used to separate species, varies. (Ib. Birds. p. 58, 60, 80, 90)  I have already quoted from Graba instances of variations in length of the

(32

Ch 4

tarsi in several sea-birds & so it is with Anser Canadensis. (Ib. p. 469)

Is. Geoffroy St. Hilaire (    ) has mentioned the case of a monkey with an extra pair of molar teeth. [pencil note:] Owen Ourang Outang

Such cases, I may remark, are often called monstrosities; but if the teeth are well formed, I hardly see that they should be so called without every deviation from the normal structure be so designated. Mr. Bellamy exhibited to Brit. Association in 1841, the head of Arvicola agrestis with fangs to its teeth, a character known to separate two groups of mice. Dr. J. E. Gray has found considerable variability (Proceeding Zoolog. Soc June 12 1849) in the molars of certain seals. The form of the lower jaw seems also to vary (Ib May 8th 1849) considerably in Bra Sloths. So according to M. De Blainville it is with the lower jaws of the Hippopotamus

(33

Ch 4

Dr. Andrew Smith (Illust. Zoolog. of S. Africa. 1849 Pl. 32) in speaking of the antelope Cephalopus Natalensis, "the females are almost always found without horns, yet individuals are occasionally killed in which they exist; hence it would appear that their presence or absence ought not to be highly considered in establishing the generic characters."

FD (In some species of Shrews (Sorex) & in some field-mice Arvicolæ, the Revd L. Jenyns (Annals of Nat. Hist. vol 7. 1841. p. 267. 272.) found the proportional length of the intestines to vary considerably. He found the same variability in the number of the caudal vertebræ. In three specimens of an Arvicola, (Ib. p. 272) he found the Gall-bladder having a very different degree of development, & there is reason to believe it is sometimes absent. Prof. Owen has shown (     ) that this is the case with the gall-bladder of the Giraffe)

(34

Ch 4 Animals

It has been long known that the presence of nails on the posterior thumbs of the Borneo Ourang (Sir J. Brooke in Annals of Nat. Hist. vol. 8 9. 1842 p 58) is variable; & Prof. Owen has shown that with the nail goes there is an additional joint & bone. Prof. Owen informs me that he has seen a specimen with having that muscle of the index-finger, which has been thought characteristic of man; but in another specimen it ran to the second finger as well as to the index.

In Spiders, from six cases recorded by Mr. Blackwall (Annals of Nat. Hist Vol XI. 1843. p 166) the more or less complete & absence of some pairs of the eyes, & even the presence of a symmetrical superpernumerary one does not seem to be so rare a variation, as might have been anticipated in so important an organ.

In the sea-urchins (Clypeastroidæ) the position of the anal orifice is highly variable, being even in the same undoubted species, sometimes above, sometimes below, & sometimes on the border of the shell. (Agassiz & Desor in Annal. des Scienc. Nat. 3 series. Tom. 6. p. 318)

[34v]

In many insects of several widely different classes, the presence of wings, is extremely variable within the limits of the same undoubted species; as in one British beetle Calathus mollis, in some Hymenoptera, as in the bed bug & & in the several aquatic hemiptera. (Westwood Modern Classification of Insects Vol. 2. p. 431. & Wollaston Variation of Species. p. 43. 101) In some a rare case described by Mr. Wollaston (p. 96) the connateness of the elytra varied.—

(35

(Ch. 4. Diff in distinguishing species in well-known countries)

It has been remarked by some authors, that the difficulty in determining what forms are really species, is due simply to want of knowledge.

Undoubtedly this is often true, more especially in regard to animals the different stages of growth & sex of animals. But on t But I suppose the Flora of Great Britain may be considered well-known, & yet what how differently is the number of species estimated by different authors! Mr. Hewett C. Watson informs me that in after examining the London Catalogue (4th Edit) for this object, he finds that there are about 1800 names which have been considered by one some Botanists as two species, but that out of this number, about 450 are considered by other Botanists as mere varieties: moreover he has given me curious details, showing how opinions have alternated in successive periods two forms having been considered varieties, then species then varieties & lastly species again; these opinions being probably at no time unanimous. In certain Protean British genera which, however, under one the following table, published by Mr. Watson* (Phytologist. May 1845. p. 143) shows at a glance how unfixed is the criterion of a species.—And it especially particularly deserves notice that most of these genera in our own country

(36

(Ch. 4. Difficult in distinguishing species in well-known countries)

have been the subject of special monographs, sometimes by more than one successive authors, who have devoted the closest attention to these genera*

So again M. Ch. Des Moulins (actes de la Soc. Linn. Tom. 16. 1849 p. 56) in his discourse on the well-known Flora of central France, says that in 2332 phanerograms, there are still 250 forms under litigation.

I suppose no two land-shells are better known than Helix hortensis & nemoralis. Mr. Bean (as quoted in Forbes Report Brit. Assoc. 1839 p. 136) of Scarborough has collected 152 vars. of H. hortensis 58 of H. pullata of some authors or the white-mouthed var. of this species; 236 vars. of H. nemoralis, & 21 of its variety or supposed species H. notabilis. Notwithstanding all this attention, & notwithstanding the fact, as I am informed by Sir C. Lyell, that H. hortensis ranges further north than H. nemoralis & is alone found in Canada, yet some great conchologists, as Deshayes doubt whether H. hortensis & nemoralis are not one the same species.

[36v]

* Atriplex is another protean genus. The Rev. Leighton told me that he had some seeds of several species collected in various places in his garden, & that a mass of plants came up, which defied the powers of the two most skilful botanists in this most skilful in this tribe, to classify.—

[36A]

[Printed]

	Salix.	Mentha.	Rosa.	Rubus.	Saxifraga.	[Darwin's total]
Hudson (1791), Smith (1824-8), Lindley (1835), Hooker (1842), Babington (1843), London Catalogue (1844),	18 64 29 70 57 38	6 13 9 13 8 8	5 22 17 19 19 7	5 14 21 14 24 34	9 25 24 16 20 16	43 138 100 132 128 103

                                                  

(37

(Ch. 4. Difficult in distinguishing species)

To give another example, not so much to show that there is difficulty in deciding what form to call species & what varieties, but that even in very big a class, generally having such fixed characters as Birds, there is some appreciable amount of variation. In Germany, according to most common authors, there are about 282 Birds, but Brehm (Vögel Deutschlands 1831) by dividing species, adds to this number 576 species, making a total of 856 species: thus he divides the tit-lark (Anthus pratensis) into 12 species & the nightingale into 6 &c.—Now I have never met an ornithologist who thought these species worthy of consideration, & it has been asserted th in Germany that many have been formed on single specimens.— On the other hand Brehm was a laborious observer: he collected (Ib. Introduct p. xix) more than 4000 skins, & he positively asserts that his new species are often found paired together, that they can be found on the same spot in successive years, & that they can often be distinguished by sound of their voices & habits; & lastly that Bird catchers practically make similar distinctions. He grounds his distinctions chiefly on slight differences in the shape of the skull, beak, tail & feet. Though it may be very proper to

(38

(Ch. 4. Difficult in distinguishing species)

to ignore these fine differences as specific, I can hardly doubt but that they exist. I believe this the more as our great ornithologist Mr. Gould has lately shown me some of our commonest birds from different districts, certainly presenting a an appreciable difference.—

Lamarck long since remarked that there was not much difficulty in distinguishing species from varieties as long as specimens were brought from one a single country,—not that this can be considered, as we have just seen, as always quite correct—but that the real difficulty begins when specimens poured pour in from every region inhabited by the genus.— Though this may be very true, yet with cautious & sound naturalists, how often do these numerous & widely collected specimens if collected from continuous regions clear away doubts*a; but the doubts are dispelled almost always by generally dispelled by admitting considerable variation;—intermediate forms connecting others which m might have been thought classed as t specifically distinct. Hence apparently it arises that those who study local floras are apt to admit more forms as species, than those who take-a wider field. But the

[38v]

*a p. 38.

I am far from wishing to assert that this always the case: on the contrary I was formerly much struck, when witnessing Mr. Waterhouse (than whom a more cautious accurate naturalist can hardly be found) examining the large collection of Mice, which I made in S. America: when the specimens came all from the same exact locality, or from very distant localities, there was seldom much difficulty in mo distinguishing the species; but when perhaps a single a specimen or two had been collected some at a moderate distance from any other locality, then I repeatedly observed there was much very great doubt & difficulty. As it Exactly the same thing was a very noticeable in the difficult genus of Birds, Synallaxis, of which I collected many specimens.—Perhaps Probably if I had collected still more numerous specimens, from every intermediate station, there would have been less difficulty, but the difficulty would have been removed only by am admitting considerable variations, or by designating every infinitesimal difference as specific.

[39a]

*p. 39

 Instances innumerable could be given in regard to the islands of almost Such cautious nat Every the several great archipelagoes; & even from so small a one as the Galapagos group. Mr. G. R. Gray showed me some small pigeons from se (Peristera Macrodactylus, Brasiliensis, brevipennis &c) from the W. Indian islands & mainland, which certainly differed slightly sensibly in length of wings, toes & plumage; & yet so cautious a naturalist as Mr. Gray, is strongly inclined to believe that they are only local races. I have quoted this instance, because Mr. Blyth has instanced in a letter to me another genus Treron in this family, as offering in the East the very same cause of doubt; but he leans to considering these slight differences as specific; for he remarks if these be given up, where can we stop; & well may he ask this: & the answer in future years, will be, as I believe, no where.—It is known that large rivers in S. America often form an impassible barrier to monkeys, & on their opposite sides the monkeys often differ slightly; these forms have been described by many authors as distinct species; but Dr. Natterer, a most careful observer who resided many years in Brazil (Note by Mr. Waterhouse in Annals of Nat. History 1844. vol. 13. p. 48) was convinced that these forms were only races of the same species See Mr. Wollastons works in regard to the insects of Madeira. See Mr. Layards & Blyths remarks in regard to the Birds of Ceylon.—

[39b]

* note to 39.

Dr. Asa Gray has lately published one a truly admirable paper on the Statistics of the Flora of the Northern United States (N. American (Journal of Science 2 ser. vol 23. p 80) & he gives ge excluding a list of 15 forms varieties of Eur plants common to Europe "which not only have been, but are not unlikely to be again distinguished as species", & another list of 42 N. American species, "almost all of which are more or less liable to be reduced to geographical varieties", of European plants. Had the United States been worked as carefully by local botanists as have the different parts of Europe, there can be no question, that the a number of varieties which had been forms, which Dr. Asa Gray considers identical with European plants, would have been cut off by Botanists having less widely extended knowledge, as distinct North American species.

* For Birds compare Prince Napoleons list with that by Sir J. Richardson & Swainson & other work. For Coleoptera, compare Mr. Murrays excellent remarks on 49 (Proc. Phys. Soc. Edinburgh in Zoologist. vols. 11 & 12. p. 3894) the slight differences, which he (as Kirby likewise) did considers too slight to be specific, with M. Leconte in Agassiz Lake Superior p. — who seems to consider that all, or nearly all, should be specifically separated.

(39

(Ch.  Difficulty in distinguishing species)

difficulty all rises to a climax & indeed seems almost insuperable where very closely allied similar forms are compared coming from islands & from countries apparently now quite separated*b: where I was much struck how entirely arbitrary the distinction is between varieties & species, when I witnessed different naturalists comparing the organic productions which I brought home from the islands, off the coast of S. America.

In such cases there is no intermediate territory for the existence of intermediate forms; & the naturalist must rely wholly on analogy. while & this will depend be governed not only by his knowledge but by the finest conditions of his mind. North America & Europe offer the most striking example of this difficulty: let it be observed to what different conclusions the best naturalists have come to in regard to the many quadrupeds, birds, insects* (z) & all plants*a of these two quarters of the world; Some of the many some calling the f slight differences which can be undoubtedly be observed in nearly all than in animals the animal productions from the old & new world, varieties, & some calling them species.

At present a considerable number of naturalists cut the knot by calling every all forms presenting the slight from distinct regions, distinct species, even if the differences are excessively slight or & even if apparently they are

(40

(Ch.  Difficulty in distinguishing species)

identical. To those who rest on the hypothesis of distinct creation as the criterion of a species, this may be logical; but it rests who can say what regions should be called distinct? Can we say we know all the means of distribution; past & present; on as what part was land & what sea, & what was the exact temperature of either, within comparatively recent geological times? In regard to distance, as Prof Mssr Haldeman & Mr Wollaston (Boston Journ. of Nat Hist vol 4. p 480. Wollaston Variation of Species p. 38.) have well remarked where shall we draw the line; if N. America & Europe are so distant from each other, that we may call their most closely allied inhabitants distinct species; are now are the Azores nearly in the middle or Madeira sufficiently distant fr in regard to Europe to justify the same distinction. Must we extend the same view to Madeira & Porto Santo, so within     miles of each other, but with so many shells & insects so quite distinct, & so many forms presenting marked varieties? Lastly must we extend it to Ireland & England, with some only on very extremely few species distinct, but with some few, as generally considered, well marked varieties? Practically each naturalist settles arbitrarily decides the question for himself, in accordance

(41

(Ch.  Difficulty in distinguishing species)

with his hypothetical idea of the term species, in accordance with the amount of his knowledge what he knows of what he considers the amount of variation witnessed by none during the present time, & according to his tendency to trust in analogy.

We have seen that in the best known countries in the world there is that in many of their inhabitants must very true, at least & that in many same there is much uncertainty in deciding what to call species & what varieties. And further it seems to me that if there are very few exceptions to the following remark, that whatever organism is found in very generally if an animal or plant inhabits different districts or even if very common in one district, if it be conspicuous for any quality, or if it be valuable or in any way attracts man's notice, so as to be thoroughily well studied varieties will have been observed, & the more striking varieties will often have been considered by some authors as distinct species.

Look to the King of beasts, as popularly called, how naturalists have doubted whether or not the Maneless Lion of Persia* (a) is a distinct species: Some few think that of Nubia also distinct; & the great lion-slaughterer believe Mr. Gordon

[41v]

* Capt Smee in Zoolog. Transacts. vol III p. 165 concludes that the Maneless lion of Guzerat is only a variety; I believe many naturalists now think it distinct. The Hyæna of Persia (Harlan's Researches p. 535) is, also, said to be without mane in Persia. differ from that of Morocco only in wanting a mane.

(42

(Ch. 4. var. of well known conspicuous organisms)

Cumming thinks is convinced that there is more than one even in the Cape district.* (*Lichtenstein in his Travels vol 2. p. 31. says the country people distinguish three different sorts of Lions at the Cape.) or look to the Elephant in India, but the variation in this animal is so curious that I shall presently enter into some de little detail on the subject; as I shall on the presen well-known & persecuted Fox of Europe.

What disputes there have been in regard to the Bears which of Scandinavia, which there is so ardently purseued them, hunted, whether then there var se be one or more species. How many moles may a person casually examine without perceiving the slightest difference, yet being a thoroughly well known, animal, we hear from Mr. Bell, in his excellent history of British Quadrupeds (p. 106) that there are several remarkable varieties.*

(* The Rev R. Sheppard in Linn. Transact. vol. XIV. p 587, describes a remarkable variety with a white snout, & white line down the back on the head, belly orange, forming a line on the chest; tail covered with long white hairs, & with the tip quite white.)

The Sportsman* (See W. Scrope's Art of Deer Stalking,—a most interesting work.) can distinguish the Red Deer (Cervus elaphus)

(43

(Ch. 4. var. of conspicuous organisms)

of the different Scotch forests; "the Braemar deer are allowed to be quite different from those of Atholl, they stand higher & are in general of greater weight": those of Corrichebar are again different & have larger head larger in proport than those of Atholl: the red deer of the outer Hebrides are very small (    )

So in Germany three varieties of this deer are distinguished in accordance with the & inhabit different localities.* (Bechstein Naturgesch. Deutschlands. 1801. p 458.)

Other instances could be given as with the common Hare. So with Fish, it is certain that the salmon of many different rivers can be distinguished by fishermen; & the Herring which has been so closely studied, is found to present a very vast range great amount of variation.*

(Wilson's Voyage round Scotland vol 2. p. 206. The Herring Fishery was one of the points especially attended to in this voyage.) To descend lower in the scale; Fishmongers can distinguish whence their oysters come, & so they can on the coast of N. America with the clam, of which they distinguish five varieties.*

(* Venus mercenaria. Dr. Mitchill in Sillimans Journal of Sciences. vol 10. p. 287.)

(44

(Ch. 4. var. in conspicuous organisms)

In plants most of those useful or much noticed by man are cultivated, & therefore do not come in here, as their variations may be all due to cultivation. To begin with a humble example; varieties of the water-cress (nasturtium officinale) are hardly noticed by botanists, but those who cultivate acres of this wil wild plant for the London Market (not seedlings raised under cultivation) for the London market distinguish three varieties, which are not caused by any difference in the quality of the water, for they can may be seen growing together; but differ, but yet they differ in hardiness & other qualities; & the large brown-leaved variety is the bas only one which will grow well, when the water is not very shallow.* (* Mr. Bradbery account of the cultivation of the water-cress. Transact. Hort. Soc. Vol IV. p. 537.) What is the tree, which ought to be best known in Britain? assuredly the Oak; yet I see that Mr. Babington, Hooker & Arnott with Dr Greville in their last Edition, treat Quercus robur & sessiliflora as varieties, whereas Dr Lindley in the Gardeners' Chronicle speaks decisively of them as if distinct species, & Sir James Smith seems to entertain no doubt about their distinctness on this subject. Every forester can distinguish the two forms: it is asserted that they come true to seed* (Gardeners' Chronicle   )

(45

(Ch. 4. var. of conspicuous organisms)

though this has been denied: the quality of their timber is said to be different (Sir J Smith English Flora. vol IV. p 149 & Gardeners' Chronicle   )

; & Quercus sessiliflora is said to be hardier & ascends the Scotch mountains higher than Q. robur.* (* Mr. Farquarson in Hooker's Bot. Misc. vol 3. p 127) on the other hand the existence of a perfect gradation of intermediate forms is admitted by everyone & Dr. Bromfield quotes (Phytologist Vol. 3. p. 883) with approval the remarks of another most careful observer Mr. Bree that "though there are sessile oaks bearing fruit on peduncles & pedunculated oaks bearing almost sessile fruit, there is yet a certain indescribable something about the trees, by means of which I can always distinguish each, without minutely examining either the acorns or the leaf-stalks." So that according to these two excellent observers the distinction of the two varieties or two species, (& the highest possible authority can be quoted for either term) of our one most conspicuous Tree can be best recognised, like a man's face like a man's face, by "a certain indescribable something."

It would be superfluous to give other examples; but parallel ones could be given in regard to our Elms, to our Birches, & most striking ones

(46

(Ch. 4. var. of conspicuous organisms)

in regard to one the Scotch Fir, in which the varieties or species, call them which you please & you will have high authority for doing so, are adapted to different situations, produce different kinds of timber & are hereditary in their quality.*

* (In regard to the Elm, see Dr. Bromfields remarks in Phytologist. vol 3. p. 837. Mr. H. C. Watson exhibited before the Bot. Soc. of London (Annals of Nat History. vol 12. 1843 p 450) specimens showing that Betula alba, pendula, glutinosa, & pubescens are all mere fleeting varieties of the common Birch. For the Pinus sylvestris, see Loudon's Arboretum p. 2189 & 2150 & Gardeners Chronicle    )

In the Yew, the highest authority Dr. Asa Gray thinks the Canadian form perhaps only a variety; & this seems the general opinion of Botanists in regard to that most remarkable bush, the Irish yew, found growing wild in Ireland with its upright dwarf habit, & large scattered leaves; but I presume, if this plant had been found growing of both sexes, growing abundantly in some distant region, no botanist would have hesitated to call name it as a distinct species. The last example which I will give is that of the noble Cedar of Lebanon: it appears in our gardens most

(47

(Ch. 4. var. of conspicuous organisms)

distinct from the Deodar, yet when old, Botanists cannot point out any good character between these two forms & the Cedar of the Atlas, & as they vary the seedlings vary hence are inclined to consider them as varieties, a conclusion indignantly repudiated by other Botanists. (Hooker — Gardeners Chronicle   ) The question in these several cases, is not whether these forms deserve a name, popular usage has settled that point, but whether they should be designated by the undefined name title of Species.—

Incidentally several cases of variation in a state of nature have now been given, & incidentally others will appear be hereafter given. It would be tedious to give as easy as useless to quote the almost numberless instances of forms, which have been considered on good authorities as permanent varieties having much of the character of species; & I will conclude this chapter by giving from various motives only, a few additional instances of variation, in whi in which the evidence is rather better than in most cases.

(48

(Ch. 4. marked varieties)

Indian Elephants. Dr Falconer informs who has had great experience in Elephants, & who has seen as many as 1200 at one a fair, informs me that they differ considerably, more than horses of the same breed, in size, pr general proportions, manner of carrying the head, form of tusks, shape of feet & in the absence of the nail on one toe: Mr Corse has given a nearly similar account (Philosoph. Transacts 1799 p. 206.) & says that the different castes have their proper names. In the Ayeen Akbery, publ written in about the year 1600, four kinds of Elephants are specified. Most of these differences probably come under our class of merely individual differences; but other both Dr. Falconer & Mr. Corse believe that some of the breeds inhabit different, & sometimes adjoining districts; fa what & animals which are thought to be cross-bred animals, are occasionally seen caught. As far as size is concerned, climate appears influential; at least, as I am informed by Mr. Crawfurd, elephants northward of a certain latitude are excluded by the government contracts. Dr. Falconer tells me that there are two marked breeds, one thicker in its general proportions, more courageous, & with short tusks directed downwards; in the other breed, the tusks are upturned, & the

(49

(Ch 4. Elephants marked var)

animals when attacked by a tiger tries to fling pitch his opponent up into the air; whereas the breed with the downward directed tusks, when attacked, falls as if instinctively on its knees, & tries endeavours to crush & pin the tiger to the ground; this breed is consequently more dangerous to ride, as sometimes even experienced hunters are thrown off its back on to the tiger. If the Elephant now such differences in structure & habits, I think all zoologists, would will agree, would in most cases be thought of specific value; but I believe no one has even suspected that there are two species in India.

In Ceylon, there is, also, a distinct breed, but this has by some (Mr. Hodgson in Asit. Soc. of Bengal vol I (1832) p. 345.) been thought to form another species. Quite Until quite lately the Elephant of Sumatra, has been was thought to be the same, (Crawfurd Descriptive Diet, of Indian Islands. 1856 p. 136.) has been shown to be deserve but now from differences in its skeleton it is thought to deserve to being called be a distinct species.

(50

(Ch 4. Marked var. Foxes)

Foxes. These are well known to be variable animals & all over the world the species are discriminated with difficulty. British sportsmen speak (Encyclop. of Rural Sports. p. 448.)

of three kinds, but it is doubtful whether these are anything but individual indifference. In Scotland the accurate Macgillivray (Transacts of Wernerian Soc. vol VII. p. 481) describes four kinds in Scotland, but he uses besides general proportions as one character the tail being tipped with white, which Bechstein (Naturgesch. Deutschlands. B I. s. 627.) has shown is a quite variable point. But the Highland or mountain Fox of Scotland seems certainly to form a distinct race: Mr Colquhoun (The Moor & the Loch. p. 97. Ch. St. John, Wild Sports & Nat. History of the Highlands p 232), a very good observer says any one can distinguish this animal even from at a distance from the small fox of the low grounds; he stands higher, his head broad, nose not so pointed, his coat more shaggy & mixed with white hairs: he is much more powerful & preys on young sheep, & rears his young, not in holes, but in clefts in the rocks; is less nocturnal in his habits,

(31

(Ch 4. Foxes)

& altogether, as Mr. St. John remarks, is more like a wolf, than a lowland fox. In Scandinavia it has been a dispu question disputed both by naturalists & hunters, whether the common red, the black & crucigerous Foxes are distinct species or only varieties. So in N. America a parallel series occurs & it has been disputed whether the red Fox, (thought to be ranked as a different species from that of Europe) the black & silver & crucigerous (ie with a dorsal stripe & a transverse one on the shoulders) foxes are distinct or not: Sir J. Richardson (Frauna Boreali-america p 93.) strongly inclines to rank consider them all as varieties.

So much interest has this question excited in Scandinavia that a set of gentlemen as the differences are said not to be confined to colour alone that near Stockholm a fox colony was established by some gentlemen near Stockholm (L. Lloyd Scandinavian Adventures vol II. (1854) p. 52.) & in it two crossed crucigerous foxes produced in the course of four years 19 cubs; of these 9 were crucigerous; 8 were black (including those with white tipped tails), 9 crucigerous & 2 red: two of the black cubs, also, produced young & these, six in number, were all black. Mr. Lloyd infers from these experiments that the crucigerous fox is a cross from the black & red, which he seems to consider, with

(52

(Ch 4. Foxes)

many of the inhabitants, as distinct species, producing, as it thus seems perfectly fertile hybrids, but Prof Nillson's conclusion that they are proved by these experiments proved to deserve only the name of varieties, seems to me the most probable. It is clear that these variations are in some degree hereditary, & the whole case is interesting as showing how difficult it is to decide what to call species & what varieties. The production occurrence, also, of strictly analogous varieties in N. America from the generally received distinct American C. fulnus or & Arctic Foxes C. fulnus & likewise (C. fulvus & lagopus) from C. lagopus is an interesting fact; & the more interesting from these varieties forms, not appearing being produced in Great Britain, though they de are, according to Bechstein, in Germany.

(53

(Ch. 4 Corvus)

Raven. It has long been known that at the pied Ravens are found at the little islands of Faroe. & have are not known else where else.  This bird is white somewhat symmetrically marked with black, & as the beak "is much larger being not higher only higher at the base, but more elongated, & in form more attenuated at the end"* than that of the Ravens, (Macgillivray History of British Birds. vol 3 p. 745.) it has been admitted by Brisson, Viellot, Wagler, Temminck, & others the most distinguished ornithologists, as a distinct species under the name of Corvus leucophæus. (a) But when When, however the ornithologist Graba when visiting visits these islands, when he fin he & investigates the case he finds that one of the pied ravens (at first quite white, the black feathers appearing with age), are produced in the same nest with ordinary ravens; & he detects gives an the instance of a & that in one case when black & pied a being were mated together & some most which often years producing the all either only exclusively young even produced black birds & often or one pied white bird with the the others black. birds were produced. (Tagebuch auf eine Reise nach Färo 1830 p. 51.) Graba's description of the beak a nearly agrees with that of Macgillvray. I may add that Landt in 1810 in his Description of Feroe p 220, says that black & speckled ravens are sometimes seen paired & that both kinds are sometimes found in the same nest.)

The fact of the black & pied ravens being sometimes

[53v]

as this particular race is known no where else, & is endemic in these islands (though partial albino ravens do occur elsewhere) this fact has been used as an argument that it is a distinct species; but perhaps the argument might be reversed with equal force, as not one other bird or indeed other production is endemic in is peculiar to this small spot.

(54

(Ch. 4. Corvus)

paired mated & producing either black or white young, is not, as we shall immediately see in the case of the Hooded crow, not so conclusive as Graba seems to think; but combined with the white appearing in the nest of common ravens, & more especially with the fact of the two birds described by Graba, the one by Macgillvray, & that by Temminck, differing very considerably in their colouring, even sometimes on opposite sides of the same individual, I think this can leave no doubt that the C. leucophæus is only a variety. Graba states says that they are not very rare, & he states the interesting fact of which he was a witness that the pied birds are persecuted & driven away by the common ravens (p. 51, 54.); & are not willingly allowed; & Macgillvray who seem once once saw on the Hebrides an individual a bird of this kind, apparently a wanderer, which he describes as "a neglected & persecuted stranger". Now what suppose whatever the cause is may be, which gives rise to this variety in Faroe were to act with rather more intensity, so that pied ravens were alone were to hold possession of these islands, how utterly impossible it would be ever to ascertain whether it was right to call this form this was a variety or species.

It is probable that the pied colouring & other characters would become here fixed in the course of many

(55

(Ch. 4. Corvus)

No doubt any chance wandering black raven would be persecuted & driven away by the pied majority, as these latter now are by the black birds; & crosses being thus [illeg]prevented, it is probable that the pied character colouring & other characters would become in the course of many generations more fixed & constant.

Now let us turn to the Carrion & Hooded crows (Corvus corone & comix): these birds are so much alike that as Magillvray observes (History of British Birds vol I. p. 529.) "were the colours the same in both it would be almost impossible to distinguish them". "The extent & tint of the grey-coloured space varies greatly in the Hooded crow (Ib. p. 534) & Bechstein asserts that in Siberia some are quite black, but how these can be distinguished from carrion crows I know not. The eggs of the two species are undistinguishable as are their digestive organs & their general habits are alike. Numerous cases are on record of the two in Germany, England, Scotland & Ireland of these two forms being seen paired, & the young are either

(56

(Ch. 4. Genus Corvus)

quite like one of the parents or intermediate in colour. (Bechstein, refers to three cases in his Naturgeschichte B. 2. s. 1170; Mr. Slater informs me that he has known of a case in Hampshire W. Thompson gives cases

Macgillivray vol 3 p. 721 gives cases in Yorkshire & Scotland.) Hence several respectable ornithologists have looked at these birds as varieties; yet, as their voice is slightly different & the carrion corvus is rather & as they inhabit one species different districts are often inhabited separately by either one or the other form; & as when occurring together they keep separate; & as the carrion crow seems to have a more southern range than the Hooded crow & more especially as typical ordinary specimens of both can be distinguished with the utmost facility, I must agree with Mr. Macgillvray that, in common parlance, "the two species are perfectly distinct".

Lastly let us consider one other case; we have in Britain one single well-known bird, the red Grouse, (Tetrao Scoticus) which has been almost universally ranked as a distinct species, & is confined to our the British islands. On the other hand the Tetrao saliceti of Scandinavia, is a bird which we might have expected to inhabit Great Britain, but is not found here.

(57

(Ch. 4. Grouse)

Gloger alone, as I believe, has argued at length (Das abandern der Vögel 1833. p. 121 117) that they are certainly only local varieties of the same species.—Mr. Gould in after studying T. saliceti in Scandinavia tells me that they agree perfectly in eggs, in the plumage whilst quite young immature slate plumage, in habits, in voice & in summer plumage, with the exception of the white primary feathers & that he cannot avoid the suspicion of them being that they may possibly be varieties. (a) The pied Grouse is very variable in plumage I apprehend if these birds had been found together, & it does not seem very improbable that colonies of the one might now be established in the territory of the other; no ornithologist whatever would have suspected thrown a suspicion on their specific distinctness; hence their geographical distribution separation & consequent exposure to a different climate is the chief argument seems to have been the sole cause of their specific difference diversity having been suspected; & undoubtedly as Britain has no other endemic bird this is an argument of some apparent weight in favour of the two forms being identical: on the other hand if we had possessed a few more endemic species the argument might have been reversed, notwithstanding it might most truly be said that every gradation exists in the proportional number of endemic forms possessed

[57v]

(a) The Red Grouse is very variable in plumage, but & easily runs into sub-local races. (Macgillivray British Birds. vol I p. 174 & p. 186) &: MacGillivray says that it differs from T. saliceti in having a lesser beak; but Nilson, as quoted by Gloger says he examined 30 specimens of T. saliceti, & the beak was not scarcely alike in two.

(58

(Ch 4. Grouse)

by a country, & why should not insular Britain possess its our own single endemic Bird?

I have entered into these three last cases at some little length in order to show how difficult it is to determine what to call a species & what a variety, even with using all sorts of collateral evidence in well-known Birds, which are amongst the least varying animals. The series seems to me an interesting one, from the pied & black ravens which must be considered as varieties, though hitherto esteemed by most ornithologists as distinct & which inhabit the same little island, but with some tendency to keep separate—to the carrion & Hooded crow, considered by a vast majority of ornithologists as distinct, often inhabiting distinct regions, but when mingling, often crossing—to the red & willow grouse, almost universally considered as distinct & inhabiting quite distinct countries, but yet with a strong taint of suspicion hanging over them.

(59

(Ch 4. Uria)

As it is so rare that varieties este of Birds, sufficiently distinct to have been esteemed species by first rate naturalists can be proved to be not distinct that I will give one more case. The common & ring eyed Guillemot (Uria to troile & U. ringvia or lacrymans) have been by about an equal number of ornithologists esteemed as good or or or doubtful species or as mere varieties. The ring-eyed form inhabits the northern islands & is generally rare; but Graba found that the Faroe islands (Tagebuch. p. 106, p 150) were its home, one out of about five existing as this form; at first Graba thought it was specifically distinct, for besides the conspicuous ring of white round the eyes & from the eye backwards, it differed in other respects; but these differences Graba found were not constant, & he subsequently ascertained that himself twice saw it paired with the common Guillemot; & the inhabitants affirm that sometimes from the two eggs of the common Guillemot, one will be ring-eyed.

Madeira In Madeira, there is only one endemic bird, but some of the European birds are slightly smaller, & some are slightly duskier; & the Redpole (Fringilla cannabina) retains

(60

(Ch. 4. marked vars)

its crimson breast throughout the year. (E. Vernon Harcourt Annals & Mag. of Nat. History. June 1855 and Sketch of Island of Madeira 1851.) The black-cap (Sylvia atricapilla) besides being sometimes duskier, som presents a variety, in which the black colour extends from the cap to the shoulders & occasionally even over all the under parts of the body; this has been described by so good an ornithologist as Sir W. Jardine as a distinct species; but as the inhabitants believe that it is produced from the same nest as the common forms black cap, there cannot be much doubt that Dr. Heineken & Mr. Harcourt are right in esteeming it as a variety.

I will now give a single case in Fish taken from Bronn; (Gesichte der Natur. B. 2. s. 106): the Cyprinus gibelio & carassius have generally been considered distinct species, for they differ in almost every proportion part in proportion, as shown by the table given by Bronn; but Eckstrom found narrates that fish the offspring of the C. carassius removed from a large lake into a small pond, assumed an intermediate form; & on the other hand the offspring from C. gibelio from a small pond turned into a large lake 40-50 years before, had become changed into C. carassius.

(61

(Ch. 4. insects)

In I have selected the foregoing instances, from being able to adduce some other evidence besides the mere existence of a graduated series of intermediate forms. But I will now give two or three instances from Mr. Wollaston's works of Variation deduced from intermediate forms observed with especial care by this excellent entomologist in the confined locality of Madeira. Harpalus vividus (Insecta Maderensia p. 54: The Variation of Species. p. 67.) is perhaps the best example; if very many specimens from many sites had not been collected, clearly showing a perfectly graduating series, its form might would have been described as several the varieties would have been described as forming several species; those from th the lowland & the wooded mountain slopes appearing "altogether distinct".

It is an interesting fact, that it attains its maximum of sculpture & minimum of size at about the elevation of 3000 to 4000 feet; both above & below which height, "as it recedes from the upper & lower limits of the sylvan districts, it becomes gradually modified, & almost in a similar manner". It varies greatly in colour

(62

(Ch. 4. insects)

shape, & in puncturing & in striation & what is even more important in the degree to which the elytra are soldered together: the latter structure united elytra are found only rarely in the sylvan districts. This insect beetle, also offers an instance, of which very many could be cited in the most distinct genera, namely of the individuals inhabiting the rocky islet called the Deserta grande, attaining a larger size than elsewhere. To take another very different genus of beetles, namely in the genus Ptinus*; have many of which some of in which some species of which "do not attain half the bulk on many of the adjacent rocks, that they do in the more sheltered districts; & so marvellously is this verified in one a particular instance, that I have but little doubt that five or six species, so called, might have been recorded out of one". Ptinus albo albopictus has a separate radiating form on every islet of the group, but all merge together by innumerable intermediate links. Moreover Very many other examples might have been adduced of each islet & even rock of different altitude having its separate variety variety. That these forms are varieties

[62v]

* Insecta Maderensia p. 260, 267. For other cases see p. 11, 30, 36, & 78.—

(63

(Ch. 4 Marked vars.)

Plants

Centaurea nigrescens has been separated by some botanists from C. niga, (the common hard head knap-weed) by several characters, of which the most conspicuous is that it has the heads are rayed. The Rev. Prof. Henslow informs me that this form kept true for two generations in his garden, but that in the third or fourth year it was clearly reduced to C. nigra. I mention this case, because, the var. C. nigrescens, as I am informed by Prof. Henslow occupies Ham parts nearly the whole of Hampshire to the exclusion of the common forms; so that and here we have the argument from range, on a small scale, as with the Red Grouse of Britain, which may be used on either side.

Koch raised the ensuing year from seeds of the a dandelion (Taraxicum officinale & palustri) in the next year (Annal. des Scienc. Nat. 2d Series. Bot. Tom 2. p. 119) T. palustri, T. officinale, T erectum, T nigricans, & T corniculatum,—forms which have been admitted by some Botanists as species, & two of which were first named by De Candolle. Yet Prof. Henslow on the other hand, though not doubting that has found that T. palustre, is a variety, has found it

(64

(Ch. 4 Marked vars.)

Plants

sowing itself has come up true for three or four generations when self-sown in his garden Koch has, also, raised from seed of one species of Isatis (Annal des Scienc. Nat. Bot. 2 series. Tom. 3. p. 375) T. tinctoria, campestris, praecox, dasycarpa,— species forms as species by De Candolle, Ledebour & other distinguished Botanists: most of these forms inhabit different parts of Europe & Asia Siberia.

From cultivating another cruciferous plant Sisymbrium austriacum, Koch concludes that S. eckartbergense, Willd. & taraxacifolium & acutangulum, both of De Candolle, are only varieties.

Mr. Hewett C. Watson is one of the few British botanists who has experimentally tried to reduce test species by cultivation; thus he has succeeded in running together raising on plants of Festuca pratense & loliacea, generally considered as distinct loliacea "stems which assuredly a botanist would assuredly have assigned to F. pratensis";* (Phytologist, June 1845. p. 166); & he almost succeeded in running together the common & Italian Rye grass (Lolium perenne & multiflorum).*— But Mr. Watsons experiments on seeds & living plants which he collected at the Azores, are particularly interesting: thus seeding plants raised from Azorean seed of the Polygonum maritimum "partook much of

[64v]

* I may add to these cases of conversion in Graminea that Bernhardi (Ueber den Begriff der Pflanzenart. 1834. s. 30) that by repeated sowings Panicum ciliare was perfectly changed into P. sanguinale.

(65

(Ch. 4 Marked vars. plants)

of the physical characters of the P. Raii from the shores of Great Britain". Seeds of the Tolpis crinita from the Azores, produced plants undistinguishable from T. umbellata; yet these plants differed in the pappus of the fruit, in a manner on which distinct genera have been founded by some authors. (Phytologist. 1845. p. 167). Again Mr. Watson has found (London Journal of Botany 2d ser. vol. 6. p. 385) that cultivation during four generations in England of the forms of Raphanus raphanistrum found in the Azores has partially obliterated the difference a character in the pods which was at first obvious. The rich deep colour of Myosotis Azorica (ib. p. 388) tends to fail in th our country; & the seedlings have varied so much that Mr. Watson is unable to say which should be referred to M. Azorica & which to M. maritima; & some approximate to the Canary species, M. sylvatica: yet in their wild state they were as easily distinguished as any other

(66

(Ch. 4 Marked vars)

species of the genus. Lastly

Buckman cases

The accurate Kölreuter (Journal de Physique Tom 21. p 291) asserts that he has seen the Digitalis thapsi, when cultivated in northern Europe, & when artificially simp fertilised, so as to preclude any possibility of a cross, after four or five generations takes assume the characters of D. purpurea, & at last was completely converted into it. The hybrid offspring between these two forms from the reciprocal crosses of D. thapsi & purpurea were perfectly fertile. Dr Lindley in his Monograph on Digitalis expresses some doubt whether Kölreuter may not have taken a variety of D. purpurea for D. thapsi, but as he clearly implies that speaks of his D. thapsi, cases from as that of Spain, this seems to impossible he may probably be trusted. These two forms are considerably unlike in many respects, & have generally been received as good species.*

E. von Berg gives a curious account (Flora 1833 Beiblatter & 1835. B. 2. s. 564) of the extreme variability of the seedlings of cultivated plants of Iris so that Dr. Hornsuch (Flora 1848 p 55) asserts that he raised twenty reputed species from Iris sambucina or Germanica; I confess that owing to some other recorded experiments of E. von Berg

[66v]

* In Loudon's Arboretum vol. 3. p 1374 it is stated that Mr. Masters of Canterbury a great raiser of Elms, & therefore one who ought to judge well, is convinced that the Ulmus Americana is identical with the Huntingdon Elms, a variety undoubtedly of English origin. In Bronn's Gesichte de Natur B. 2. p. 85, there is a marvellous account of the change of a plant of Lobelia lutea into L. bellidifolia; & by Link of Ziziphora intermedia from Z. dasyantha, & of a great change in Ribes alpinumi.

[66v]

Mr. Gordon of Birnie in his Flora of Moray-shire says the "Avena pratensis is confined to soils of this description (calcareous), changing its habits, as the proportion of their ingredients differs. Where there is a super abundance of limy matter, the plant often assumes a glaucous rigid appearance, which has probably originated the A. alpina and causes it still to hold a place as a distinct species.'—

(67

(Ch. 4 Marked vars. Plants)

I should not have audited thought that there had been some mistake here, had not his results in the case of the genus Iris been strongly corroborated by quite independent testimony. For M. C. Bouchés (Flora 1833, Nachschrift, Hornsuch. s. 44) by sowing seeds of T. Germanica raised 13 reputed species; & what is important for us, three of these, namely T. florentina, Germanica & pallida are Linnean species & have been found growing in separate districts, & in their own native habitats remain unaltered.

In 1867 Red & Blue vars of A. grandiflora produced both vars & intermediate—see notes on crossing plants.

1861 a new var Eugenia I send Cattell came red and blue

1867 Both vars. extra fertile when crossed

The blue & red pimpernel (Anagallis arvensis & coerulea) have by a good many botanists been considered as distinct species, for besides in the colour of the flower, they differ in some other respects. (a) The Revd . Prof. Henslow's experiments, though nearly (Loudon's Mag. of Nat. Hist. vol. 3. 1830. p. 537, but compare with vol. 5. p. 493) can hardly be considered absolutely (though nearly so) decisive, in showing that one form can be raised from the other: Dr. Bromfield (Phytologist vol. 3. p. 699) has seen bright blue & flesh-coloured one flowers on the actually the same plant, when cultivated in a garden. Dr. Asa Gray says that in the United States whither this species has been introduced all the coloured varieties are met with, having flowers of variable size. Bernhardi

[67v]

(a) It is certain that each kind can be long perpetuated by seed & keep true. (Linnean Transactions vol. 5. p. 44 & Flora 1821. B. 1. s 15.)

On the other hand

(68

 

 

(Ch. 4 Marked vars)

(Begriff der Pflanzenart p 9) says that it is almost certain (& I have received corroborative evidence) that the allied Anagallis collina produces blue & red flowered varieties. Considering these several statements the probability seems to me strong that they the A. coerulea & arvensis should be considered only as varieties. I have alluded to this case chiefly owing to the remarkable fact, that Gærtner with all his experience failed after repeated & reciprocal trials (Bastarderzeugung s. 309) to raise seedlings from a single hybrid between these two forms, whence he concludes that they are distinct: Herbert succeeded with Anagallis collina; & if Gærtner had shown that he could artificially fertilise either variety with its own pollen one would then have had more more confidence in his result.—

The most interesting case on record is that of the Primrose, common oxlip Bardfield oxlip & cowslip (Primula vulgaris elatior & veris).

These plants differ, as everyone knows, in general appearance their flowers foliage & habit; they all three differ, also, in the forms of the capsule & seed: (Revd . W. Leighton in Annals of Nat. History vol 2. 2 series. 1848. p. 164) they the primrose & cowslip have a

[68v]

could Gærtner have by chance tried only a male plant or with female pollen??"

(69

(Ch. 4. Marked var. Primula)

different scent: they flower at somewhat different times: they ordinarily inhabit different stations, the cowslip in th open fields & the primrose on shaded banks & in shaded woods, but they are sometimes mingled; they abound in different districts in different proportions; * (a) Back of Page & l what is more important They have, also, different geographical ranges; Dr. Bromfield has remarked (Phytologist vol. 3. p. 694) "that the primrose is absent from all the interior regions of Northern Europe, where the cowslip is indigenous": & on inquiring Messrs. H Bentham & Hooker inform me that in the East, the the primrose is found only in the Caucasus; that the oxlip ranges from the Caucasus to about the latitude of Moscow & the Cowslip from the Caucasus to four degrees northwards to the latitude of St. Petersburgh.

In Britain see Cybele Brit

Lastly Gærtner laboriously experimentised on these several forms during four years, & actually castrated & crossed no less than 170 flowers, & yet strange to say he only twice succeeded in getting any good yet scanty seed (Bastarderzeugung. s. 721; & s. 178; but the table is not quite correct for a cross is mentioned at p. 247 not introduced into the table.)

[69v]

(a) ; & in Switzerland the P. vulgaris & elatior ascend to different heights the primrose being the more tender.* (Annals of Nat. History, vol. IX. 1842. p. 156. &  p. 515. See, also, Boreau Flore du centre de la France 1840. Tom 2. p 376) and Hooker's & Arnotts British Flora 1855, on the rarity of P. veris in Scotland.)

[69v]

W. C. Watson Cybele Britannica vol. 2. p. 293 says only on range of Primrose & Cowslip that "P. veris would seem to be an uncommon plant in the W. of Scotland".

vol 3. p. 488—doubts Mr. Sidebotham experiments, so I had better not speak so enthusiastically—doubts them from want of general accuracy—& from his want of Botanical knowledge,—relates more especially to P. elatior fr P. vulgaris veris; & to P. vulgaris from P. elatior —Allows they support P. vulgaris P. veris coming from an intermediate form, & reverse case.—

Thinks he did not take sufficient precautions, what I know not—

Express a doubt about P. elatior.—'.

(70

(Ch. 4. Marked var. Primula)

He expressly states that the Primulaceæ offer no mechanical difficulties to crossing,*a but yet it would have been far more satisfactory if he had shown that he could artificially fertilise a Primula with its own pollen. Nevertheless, whether On the supposition which seems to me most probable that this extreme infertility is rea not real, but only apparent, owing & caused by some want of skill or knowledge, we have, nevertheless, as goo as good as, indeed far better evidence than is attainable, in most cases, of the infertility of these forms together,—seeing how perseveringly the experiment was tried by the most skilful practised operator who ever lived, that of the infertility of these forms—

Considering these several statements, it seems to me difficult to imagine better evidence than in this case that the primrose & cowslip deserve to be called distinct species. But now let us look to the other side: it is universally acknowledged*a that in England there are so many intermediate forms found wild that it is most difficult to draw any strict line of demarcation between the two extremes of the primrose & cowslip. And what is the result of the many experiments

[70v]

Leoq—at Maer gardener assured me he had known whole bed of Polyanthus spoilt by P. veris growing near.

* See Phytologist vol. 3. p 43. for some excellent remarks observations on the intermediate states by Mr. Watson.

(71

(Ch. 4. Marked var. Primula)

which have been made? Long Several years ago, the Hon. & Revd . W. Herbert (Transactions of the Horticult. Soc. vol. IV, p 19) so well raised from the seed of a highly manured red cowslip, a primrose, cowslip, oxlips of various colours, a black polyanthus, a hose-in-hose cowslip, & a natural primrose bearing its flowers on a polyanthus stalk: from the seedling alone seed of that very hose-in-hose cowslip, he raised a hose-in-hose primrose. Subsequently the Revd. Prof. Henslow (Loudon's Mag. of Nat. Hist. vol. 3. 1830. p. 409) doubting Mr. Herbert's experiment took the seed of some cowslips growing in a shady part of his garden, & the raised seedlings which varied considerably, approaching more or less closely to some certain wild oxlips which Prof Henslow had observed; "& one was a perfect primrose". Although These experiments were not thought sufficient;*

[For readability, Darwin's horizontal and vertical deletions are left uncrossed]

for though (Some nurserymen (as I have been myself informed) are convinced that such changes take place in their seed beds, others have strongly denied them, as in Gardener's Magazine vol. VII p. 123, 247

& that most critical observer Mr. H. C. Watson raised tried raised at several periods many seedlings, from the cowslip, (P. veris) [illeg] intermediate & oxslip (P. veris & or (P. veris), from a Claygate oxlip, & from an truly intermediate oxlip, truly intermediate in most points, but with

(72

the primrose predominating, & from a Claygate oxlip& the conclusion at which he arrives* (Phytologist Vol. 3. p 43, and vol 2. p. 217. p. 852) is "that seeds of a cowslip can produce cowslips & oxlips; & that seeds of an oxlip can produce cowslips, oxlips & primroses.") (The experiments of Mr. Sidebotham (Phytologist Vol. 3. p 703) are, perhaps, the most important of all, for the plants from which he procured seed, were covered by bell-glasses & so crossing was prevented. He gather transported the wild plants into the garden himself

He performed all the operations with his own hands. Moreover he experimentised on the Bardfield oxlip or (P. Jacquinii or P. elatior of Jacq.), which has very generally been received as a third distinct species from the common oxslips; though in this case, as with the common oxlip, Mr. Watson & Dr. Bromfield have found (Phytologist vol 1. p. 1001 & vol. 3. p. 695) that though have "seen exceptional instances to all the characters, taken singly, by which this plant is distinguished from P. vulgaris & P. veris"; but Dr. Bromfield admits that it certainly has much the air of a distinct species.

[73A]

Names of seedling produced

[Table not transcribed]

(73

(Ch. 4. Marked var. Primula)

Mr. Sidebotham's experiments were as follows, & they are the more important as he was a hostile witness, & confesses that his the experiments "disappointed me greatly & interfered very materially with my previous idea of specific identity".

These experiments bring out clearly the hereditary tendency in all five forms. Both here & in Mr. Watson's experiments there is no direct passage from a true cowslip to a primrose or reversely; but Mr. Herbert experimentised on a cultivated red cowslip, highly manured, & from it he raised "a natural primrose on a polyanthus stalk" & again on the succeeding year from whi his seedling hose-in-hose (calyantha) cowslip he raised a hose-in-hose

[73v]

*Ths form produced From seed of this form Mr Watson (Phytologist vol. 2. p 218. & Dr. Bromfield remarks vol 3. 69) raised 88 seedlings of which 63 were intermediate, 5 were genuine cowslips & 20 true primroses.

(74

primrose. The Revd . Prof Henslow's cowslip, whence he raised "a perfect primrose", was a garden plant & grew in a shady place. It goes for nothing that some authors have planted seeds, especially if gathered from wild plants, (Phytologist vol. 3 p 180) & have found that all the seedlings, have come true to their kind; it only shows how true the kind is, when not disturbed by cultivation.—

No one, I believe, has disputed the accuracy of the statements of these foregoing four Botanists, Messrs. Herbert, Henslow, Watson & Sidebotham; three of whom, I may add, commenced their experiments in a sceptical state frame of mind.

But the results have been attempted to be explained away by the supposition of the intercrossing of the several forms. Now laying on one side Gærtner's laborious & careful experiments, which nearly all failed, & assuming that insects could effect, that which he could not; do the results agree with this view of crossing? It seems to me most decidedly not. Mr. Sidebotham expressly

(75

(Ch. 4. Marked var. Primula)

states that he protected his flowers by bell glasses; & this having been done, it seems quite incredible that there should have been on this view so much crossing in every all his five cases. indeed apparently as much variations in the offspring in most of the experiments. Moreover on the mountains of Switzerland the P. elatior or supposed hybrid between P. veris & vulgaris grows "by thousands in places within many leagues of which the P. vulgaris & in Russia is absolutely unknown". (Annals of Nat. History, vol. IX. 1842. p. 156); so it must be from as we have seen with the oxlip from its Northern range in Russia; so with the Bradfield oxlip (or P. Jacquinii) of Bardfield, round which place "the primrose does not occur for some miles". (Annals of Nat. History, vol. IX. p. 515).

Lastly, & I may venture to say that I speak after a careful study of all known well ascertained facts on Hybridism, there is no known instance of one species fertilised by the pollen of another species producing pure forms of both or either parent as must have occurred on this view with Herbert's & Henslow's cowslips & with Mr. Sidebotham's P. Jacquinii, if they were had been fertilised by the pollen of the primrose.

Nor do I believe Moreover the common oxlip, or supposed Hybrid between the primrose & cowslip, yielded, as we have seen in Mr. Watson's & Sidebotham's

(76

(Ch. 4. Marked var. Primula)

experiments, gave yielded various oxlips & pure primroses & pure cowslips; whether we choose to imagine these hybrids were self-fertilised, or were fertilised by either pure supposed parent, so sudden & absolute a reversion to either or both parent-forms is in the case of species without any known analogy* (a) in carefully recorded experiments on the crossing of species. From these several & combined reasons I think we are justified in absolutely rejecting the view that all the forms produced in the foregoing several recorded experiments, & likewise existing in nature, all can be accounted for by the crossing of two or three aboriginally distinct species; their origin I think must be attributed to variation, but I am far from wishing to sup assert that some or many of the graduated intermediate forms may not likewise be in large part due to their intercrossing having at some time crossed, which no doubt would in increase their variability & probably aid in their tendency to reversion to either one or both of the parent varieties.

[76v]

The well-known & marvellous case of Cytisus adami would be an analogous case in the individual, (though not in seedlings) if it could be shown that this tree was really a Hybrid: some competent judges firmly believe that it was produced by grafting the union of two buds of the two species.—

(77

(Ch. 4. Marked var. Primula)

In all the experiments, the common oxlip seems the most variable form; though the cowslip is sometimes little less so, for in Prof. Henslow's seedlings "not one had the decided characters of the common cowslip". Unfortunately no one, except Mr. Sidebotham seems to have tried the seed of the pure primrose; & it would be very rash to draw any conclusions from the apparent greater trueness of the primrose; but if this one experiment were confirmed, the primrose probably should be looked at as the primordial form, whence has been derived through intermediate oxlip-forms the cowslip, & the Bardfield oxlip. It is, perhaps, the most probable view that the common oxlips are varieties of the cowslip, retrograding sporting easily reverting back towards the primrose; some of the forms being having been complicated by crosses with either the primrose or cowl cowslip. I have entered into this case with great detail because, considering the general habit structure habitat, range in height & latitude, & apparent infertility of the two forms, & the many careful experiments made on them, this seems the most interesting case on record. An able Botanist has remarked (Phytologisr vol. 2. p. 875) that if the primrose & cowslip are proved to be specifically identical, "we may question 20,000 other

(78

(Ch. 4. Marked var. Primula)

presumed species." But If common descent is to enter into the definition of a species, as is almost universally admitted, then I think it is impossible to doubt that the primrose & cowslip are one species. But if, in accordance to the views which we are examining in this work, all the species of the same genus have a common descent; this instance case differs from others ordinary cases, only because we can prove the common descent, some as some of the intermediate forms still exist in a state of nature, & that we are enabled to prove experimentally the common descent.

Hence common practice & common language is right in giving to the primrose & cowslip distinct names.

I will end this long discussion by recalling attention to another statement by Mr. Herbert in regard to the species of Primula, which, though it may seem incredible I think ought not to be lightly rejected, as Mr. Herberts observations on the Primrose & common cowslip & on various other subjects

(79

(Ch. 4. Marked var. Primula)

have stood the test of subsequent observation. Mr. Herbert affirms (Transact. Hort. Soc. Vol. IV p. 19) that he raised a powdered Auricula (P. auricula) from P. nivalis; & that he likewise raised P. Helvetica (described as a species by Don, but treated as a variety of viscosa in Steudel) from P. nivalis; & that thirdly he raised P. Helvetica also likewise from P. viscosa. which would

Hence Mr. Herbert concludes that these Swiss Primulas are only local varieties.

(80

(Ch 4)

Conclusion. From the various facts now given in this chapter, & innumerable others might have been added, I cannot doubt that there is much variability in organic beings in a state of nature.(a) The variation differs greatly in degree; in some some it is scarcely any perceptible, in others much strongly marked; so that we have a graduated series from the finest shades of individual differences, to strongly marked well defined races, distinguishable with great difficulty, if really distinguishable at all, from sub-species & closely allied species. In certain protean genera, the variability may in part be of a different order nature; but on this point [illeg] it seems difficult to arrive at any clear definite conclusion. From what we have seen of the effects of domestication or changed conditions on organisms of all kinds, & which beings, we it has been shown in the second chapter, could not have been originally selected from the plasticity of their organisation, & as the knowing well that the history of the world is emphatically that

[80v]

(a/text)

The widely-ranging, the much diffused & common, in short the vigorous species are those which are the most apt to vary.

(81

(Ch 4)

of Change, it would have been a discordant result if there had been no variability in a state of nature. Judging from the effects of domestication it is indeed surprising that we do not witness clearly see in nature more organic admitted change, in the organic beings than we do; but then if such greatly changed organisms do exist, they would be universally called species & not varieties.

According to the views discussed in this work, species do not differ essentially from varieties;—two closely allied closely allied species of the same genus usually differing more from each other than two varieties, & being much more constant in all their characters. This greater constancy may be looked at as partly due to the the several causes of change not variability having acted for less energetically on one or both of the two species where under comparison than on the one species yielding the two or more varieties; and partly to the characters of the two species having been long inherited unchanged, & by this very cause having become more

(82

(Ch 4)

fixed. The greater amount of difference between the two species than between the two varieties, may be looked at as simply the result of a less greater amount of variation; the intermediate varieties between the two species (either directly between them or between them & a common though distinct parent) or between them & a common parent having become extinct. Hence as a general rule, two species may be looked at as the result of for variation at a former period; & two varieties, as the result of contemporaneous variation.

I look at the term species, as one arbitrarily given arbitrarily for convenience to a set of individuals close.

But the forms generally considered as varieties & those considered as species differ in one other most important respect; namely in the perfect fertility of varieties together & the lessened fertility of the offspring of two species. This subject will be discussed in a separate chapter; & I will here only repeat that the infertility so of species when crossed graduates away so insensibly

(83

(Ch 4.)

that the two most experienced observers in the world who ever lived have come to diametrically opposite results when experimentising on the same forms;—that the infertility does not closely follow systematic go with the general amount of difference between the two forms, but follows laws of its own;—that it is most powerfully affected by the sex in reciprocal or reversed crosses of the very same two species;—and finally that, as that, as we have seen in the last chapter, the reproductive system is eminently subject to disturbance & that infertility of an analogous kind to that resulting from hybridism supervenes from causes other & totally distinct causes. Hence, as it will be attempted to be shown in the chapter devoted to this subject, that there is no overpowering valid reason, why the different "sexual affinity" (to use Gærtner's expression) of different species to each other should be thought a character of overpowering weight, in comparison with the other differences between species to each other when contrasted with the difference between varieties. with each other; as, for example, in the tendency to adhere when grafted together.

(84

(Ch 4)

(It ha seems to me that the term species is one arbitrarily given for convenience sake to a set of individuals closely like each other; &, that it is not essentially different from the term variety, which is given to less distinct & more fluctuating forms. The term variety, in reference to comparison with mere individual differences, is applied, also, arbitrarily & for convenience.

Practically if two forms are tolerably constant in their characters & are not known to be connected by a nearly perfect series of intermediate forms they are called species; & according to the views here given, even there seems no good reason even if shd the two distinct forms are be thus connected, if the intermediate forms are comparatively rare, so as seldom to offer cause much difficulty in naming an individual specimen, there seems no good reason why they should not be called species; & in that case science & common language would accord in giving distinct names of equal value, to the primrose & cowslip,—

the deodar & cedar of Lebanon, the

(85

(Ch 4)

to the deodar & cedar of Lebanon, —to the Durmast and common oak,—as well as to the many fine species distinguished by the naturalists on characters of little more physiological importance.

As the only known cause of close similarity in two organic beings, is descent from in a common parent, it is natural that the idea of descent should have entered into almost every definition of the term species.

Hence let A monster may be abnormal in any degree, but the instant we know its parentage, we do not doubt about referring it to its species.—On the views here discussed, the idea of the common descent of all the individuals of the same variety, & of the same species equally comes into play; but it is not confined, as local in the ordinary definition, to the individuals of the same species, but is extended to the species themselves belonging to the same genus & family, or as far as to whatever higher group our facts will lead us.—

(86

(Ch 4)

According to these views it is not surprising that naturalists should have found such extreme difficulty in defining to each other's satisfaction the term species as distinct from variety. It ceases to be surprising, indeed it is what might have been expected, that there should exist the finest gradation in the differences between organic beings, from individual differences to quite distinct species;—that there should be often the greatest gravest difficulty in knowing what to call species & what varieties in the best known countries, & amongst the most conspicuous & best known organic beings if ranging over a wide territory; & that the difficulty should be hopelessly great in two adjoining regions of but now perfectly, or almost perfectly separated regions. (a text) That there is no distin essential difference, only one of degree & often in the period of variation, between Species & Varieties, seems to me at least as simple an explanation of the many

[86v]

(a) (text)

We can understand why it is that the species in large genera are generally more closely related to each other other than in small genera, & why they are grouped & related in little clusters like satellites around certain other species, — for on our views why they are apparently local often confined in their distribution, & lastly why they oftener present varieties & a greater number of varieties, than do the species in small genera: for, on our views, the species in the larger genera, where are more especially when we consider them confined to same one country, have, where, in any country, many species of a genus have been formed by varieties there has been most in such genus a greater than average amount of modification or varieties within the existing geological period; & hence we might expect that on a average generally the same forms would be still varying above the average the resultant forms would ib so far be most tend to resemble varieties in being closely resembling each other & in being grouped around certain species, like varieties around their parents & in being local. We might moreover, expect, on these views that where there has been lately much specific modification or variation, there generally would be now most variation in progress.

The conclusion that there is no

(87

(Ch. 4)

difficulties by which naturalists are beset, as that each species should have been, independently created with its own system of variability,—the varieties imitating the characters of other species, supposed to have also been independently created, so closely as to defy in many cases the labours of the most experienced naturalists.