skip to content

Darwin Correspondence Project

From G. J. Romanes   10 July 1874

Dunskaith | Parkhill | Ross-shire

July 10th./ 74

Dear Sir,

Knowing that you do not dissuade the more attentive of your readers, from communicating directly to yourself any ideas they may have upon subjects connected with your writings; I take the liberty of sending the enclosed copy of a letter, which I have recently addressed to Mr. Herbert Spencer. You will perceive that the subject dealt with, is the same as that to wh. a letter of mine in last week’s Nature refers, viz., “Disuse as a Reducing Cause in Species.”1 In submitting this more detailed exposition of my views to your consideration, I should like to state again what I stated in Nature some weeks ago, viz., that in propounding the cessation of selection as a reducing cause, I do not suppose that I am suggesting anything which has not occurred to you already. Not only is this principle embodied in the theory set forth in the Article on Rudimentary Organs (Nature Vol. IX.);2 but it is more than once hinted at in the Origin, in the passages where rudimentary organs are said to be more variable than others, because no longer under the restraining influence of natural selection. And still more distinctly is this principle recognized in p. 120.3

Thus in sending you the enclosed letter, I do not imagine that I am bringing any novel suggestions under your notice. As I see that you have already applied the principle in question to the case of artificially-bred structures, I cannot but infer that you have pondered it in connection with naturally-bred structures. What objection, however you can have seen to this principle in this latter connection, I am unable to divine; and so I think the best course for me to pursue is the one I adopt, viz., to send you my considerations in full.

In the absence of express information, the most natural inference is that the reason you refuse to entertain the principle in question, is because you deem the backward tendency of indis-criminate variability, inadequate to contend with the conservative tendency of long inheritance. The converse of this is expressed in the words:—“That the struggle between natural selection on the one hand, and the tendency to reversion and variability on the other hand, will in the course of time cease; and that the most abnormally developed organs may be made constant, I see no reason to doubt.” (Origin. p. 121)4 Certainly not, if, as I doubt not, the word “constant” is intended to bear a relative signification; but to say that constancy can ever become absolute,—i.e. that any term of inheritance could secure to an organ a total immunity from the smallest amount of spontaneous variability,— to say this would be unwarrantable. Suppose, for instance, that for some reason or other, a further increase in the size of a bat’s wing should now suddenly become highly beneficial to that animal; we can scarcely suppose that variations would not be forth-coming for natural selection to seize upon, (unless the limit of possible size has now been reached, which is an altogether distinct matter). And if we suppose that minute variations on the side of increase are thus even now occasionally taking place, much more is it probable that similar variations on the side of decrease are now taking place—i.e., that if the conservative influence of natural selection were removed for a long period of time, more variations would ensue below the present size of bat’s wings, than above it. To this it may be added, that when the influence of “steady selection” is removed, it seems in itself highly probable that the structure would, for this reason, become more variable; for the only reason why it ever ceased to be variable (i.e. after attaining its maximum size), was because of the influence of selection constantly destroying those individuals in which a tendency to vary occurred. When, therefore, this force antagonistic to variability was removed, it seems highly probable that the latter principle would again begin to assert itself, and this in a cumulative manner. Those individuals in which a tendency to vary occurred being no longer cut off, they would have as good a chance of leaving progeny to inherit their fluctuating disposition, as would their more inflexible companions.

For the rest I refer you to the enclosed letter, and trusting that you will accept the assurance of my profoundest esteem, | I remain, dear Sir, | Your obedient and devoted servant, | George J. Romanes.

Charles Darwin M.A. F.R.S. etc. etc.


Omitting preliminaries, the letter is as follows:—

The issue is clearly one of “direct equilibration,” versus “indirect equilibration”5   It is to be observed, however, that the more [ultimate] question as to the comparative degrees of influence exerted by these two principles respectively, is a question only partly involved in the present issue; for it is manifest that if indirect equilibration is supposed to have the greater share in the reducing of a disused organ, it does not follow that it shd. also have the greater share in the increasing of a used one. The two cases are not obverse, but converse; and so far from the one involving the other, it seems to me that the one does not even imply the other. Additional causes of the indirectly equilibrating kind are at work in the case of disuse—viz., economy of growth and cessation of selection,—while no corresponding causes are at work in the case of use; and this one element of difference is clearly sufficient to destroy any relation of constant proportionality—much less of equality—between the two cases. Hence, while agreeing with much you say regarding use, considered as a directly-equilibrating principle, I may without inconsistency proceed to question the supremacy wh. you assign to disuse, similarly considered.

Premising that the words use and disuse will throughout be used in an extended signification of the presence and absence of function, I may begin by supplying one or two notes on my letter to Nature.

(The following is an abstract of these notes, the less important being omitted. The four exceptional cases alluded to in Nature are, 1st. an Angora rabbit, brain decreased 46 per cent! 2d. Indian frizzled fowl, wing-bones 33 per cent too light. 3d. Dorking fowl same bones 30 per cent too light. 4th. Silk fowl, depth of sternum reduced 34 per cent.6 Regarding 1st. I can only suggest that “the characteristic narrowness of the skull” may be admired by those who breed these rabbits: if so, the extraordinary degree of reduction in this case is probably due to selection. 2d. and 3d. left out of calculation. (see ref. Nature. For same reason table on p. 285 is left out).7 4th. is hardest to account for, since without any assignable cause (other than disuse), degree of reduction has exceeded by 9 per cent the lowest average expected. According to doctrine of chances, however (upon wh. cessation of selection depends), it is not improbable that in one or two per-cent of cases, the reduction shd. fall considerably below the lowest ordinary average. For cessation of selection depends upon indis-criminate variation, and it is not improbable that in one or two per cent of such variable things as are the degrees of congenital variabilities, several unusually high degrees of reduction shd. have followed one another in the same line of descent. Hence, on doctrine of chances, it is not improbable that in a small per centage of cases, the degree of reduction shd. fall below the lowest ordinary average. (Cf. Biology I. p. 268.)).8

From all this it follows, that unless we had happened to possess independent reasons for believing in the potency of disuse, we might rationally have ascribed all the results observed by Mr. Darwin to the influence of indiscriminate variability. But as we do possess such independent reasons, the question to be decided is, How far in our estimation are we to allow the influence of indiscriminate variability, to grow at the expense of the influence of disuse? This question, I believe, cannot be decided in the present meagre state of our knowledge; and hence the only position I stipulate for is that of suspended judgment. At the same time, for the sake of shewing how much may be said for the cessation of selection, I refer you to the two letters upon this subject in Nature,9 and also to the following supplementary remarks.

In the first place, you will allow that all à priori considerations must yield before facts, if the latter do not support the former. Now it seems to me that so far as the question of disuse (as distinguished from that of use) is concerned, we possess unexceptionable à posteriori evidence in the comparatively unused organs of our domestic animals: they afford, as it were, an experiment on a large and varied scale, and with so strict an adherence to the “scientific methods,” that I do not think we could have improved upon this experiment, had we had the devising of it ourselves. Yet, even upon the supposition that in this experiment disuse has been the sole reducing cause, (wh. I strongly deny), disuse has here shewn itself a feeble influence, as measured by the expectation to wh. your à priori reasoning would conduce. I am aware that this difficulty has been anticipated in Biology II pp. 358–60;10 but I am unable to see that it has been disposed of. If you will have the kindness to refer to this passage, you will see that the only argument to be gathered is, that because a certain amount of muscular tissue is required, for example, to sustain flight; therefore the pectoral muscles, etc., of our domestic birds “must be kept up to a certain level of power,”— i.e. altho. these muscles may deteriorate as to their quality under the influence of disuse, it is not to be expected that they shd. diminish much in quantity; seeing that if they did, flight, even for a short distance, wd. be impossible. Now it seems to me that until some necessary reason is shewn why flight shd. not have become impossible to our domestic birds, there is no relevancy in pointing to the fact that a certain bulk of muscular tissue is essential to its performance. No doubt the alleged fact is a fact, but why shd. we therefore expect our poultry to possess the requisite bulk of muscular tissue? Had we been dealing with wild species, where the influence of natural selection might have cut off all the non-flying birds, then, no doubt, the fact wd. have been crushing; but I fail to understand why an inability to fly wd. have been so serious a detriment to our domestic birds, that the tendency of disuse, or of any other cause, to effect this inability, must necessarily have been neutralized by some other influence. Fully admitting that “if these parts dwindle much, the creature will be unable to lift itself from the ground,”11 I cannot perceive the implied necessity of the creature being able so to lift itself.

Thus I cannot help feeling that the fact of disuse having effected so little reduction in the bulk of the unused organs of tame species, is a serious objection to our accepting adverse à priori conclusions with regard to wild ones. And this objection is very much increased, if we find reason to suspect, (as I think we do), that some equilibrating cause, other than disuse, has also been at work in the case of tame species. Indeed, that there is a high à priori probability of some such other cause having been at work, may, I think, be gathered even from the Biology itself. For in vol. I. p. 441,12 it is truly observed, that increased use (at least on the theory of strains) cannot be supposed to increase the lengths of bones, etc.; and the cogent arguments there employed would apply equally well to the case of disuse. Now these arguments, coupled with the admission referred to in the last paragraph—viz., that the effects of disuse are primarily interstitial,—go far to prepare the way for the reception of other reducing causes,— the issue being with regard to the bulk, and not to the constitution, of the affected organs. I wd. also refer you to the admirable dis-cussion in § 304, and ask whether you do not think the following a fair corollary upon it. The effects of use are, within limits, of more rapid production than those of disuse; for the former are immediately consequent upon use, while the latter are only indirectly consequent upon disuse—depend upon the unused organ being in competition with the used ones. If you think this a fair corollary, it not only confirms what I said at the beginning of this letter, viz., that there need be no equality between the effects of use and those of disuse; but it also gives us a rough measure of the effects of disuse, for we know that the effects of use in species are not of rapid manifestation.

Now, as I pointed out in Nature, the whole question turns upon the rapidity with wh. disuse operates. Unquestionably, if it operates at all, it must in time be able to reduce unused organs to rudiments; but as it is in competition with the economy of growth and the cessation of selection, the only issue is as to the comparative speeds with wh. these several causes act. And I think we are very liable to error, when estimating the rapidity with wh. disuse operates. We know that its action on individual organisms is exceedingly marked, and are hence prepared to expect the same of it in the case of species—neglecting to attach sufficient importance to the fact, that the laws under wh. it operates in the two cases must be quite different. Our judgment may also be vitiated by choosing, as typical examples, cases of rapid structural change following immediately upon an alteration in the conditions of life—forgetting that the strength of inheritance (i.e., the opposition to change) increases with the increase of such change. Hence, it seems to me, that our estimation shd. be guided solely by the results observable in our domestic animals; and, as already observed, the effects of disuse have here shewn themselves of slow development. In what proportion the rate of reduction due to disuse here stands to the rate of reduction due to indiscriminate variation, it is manifestly impossible to determine with any approach to certainty; but, looking to the great variability of our domestic productions, I cannot help feeling it probable that the action of the cause I am enunciating has been the more rapid of the two.

It may be presumed, then, that a very fair case has been made out in favour of the hypothesis, that the cessation of selection is a true cause of reduction in the case of all unused organs. Indeed, “unless we deny the persistence of force,” the existence of this cause seems a deductive necessity; altho. opinions may legitimately vary as to the potency of its influence.

I may now state to you what I could not easily have made clear to the readers of Nature, viz., that the principle I call the cessation of selection admits of being extended, so as likewise to include all cases of the cessation of function.13 This term is very objectionable, because liable to confusion with that of “disuse”—the two principles being, as you will see, quite distinct. I cannot, however, at present think of a better term; so it must serve the purposes of this letter. “Indiscriminate variability”, or “retrogressive variability,” seem well qualified to express the joint action of the cessation of selection and the cessation of function. An example of each case will render this distinction clear. In the glenoid fossa of the human skull there is “a small tubercle, the post-glenoid process, the representative of a prominent tubercle wh., in some of the mammalia, descends behind the condyle of the jaw, and prevents its being displaced backwards during mastication.”14 Now we can scarcely suppose that the presence of this tubercle is, in any animal, due to direct equilibration; seeing that the strains to wh. it is subjected can never, in any animal, have been either frequent or great. On the other hand, dislocation of the jaw in a wild animal means death by starvation; so that the prominent tubercle in such mammalia as possess it, may safely be attributed to the influence of natural selection. Hence its decline into a rudiment when no longer required, must also be attributed to indirect equilibration. And how greatly the economy of growth wd. in this case have been assisted by the cessation of selection, is sufficiently obvious—the affected structure being originally of but small size. Take now a case illustrating the cessation of function. The temporal and zygomatic fossæ vary in size, according to the power of crunching enjoyed by different species. Presumably, therefore, in man these fossæ are in a dwindled condition, as compared with the same parts in his simian ancestor. How was the dwindling caused? Partly, no doubt, by the economy of growth, but chiefly by disuse. Now, as in the last case the economy of growth must have been assisted by indiscriminate variation, continually reducing the average size by free intercrossing; so, in the present case, disuse must have been assisted by the same influence. For, granting that the large size of the fossæ in apes, was directly caused by increase of function leading to the origination and, thro. inheritance, continuance of variations on the side of increase; and it follows that, unless we deny the persistence of force, we cannot but conclude that when this cause of variations on the side of increase was removed, and when, as a consequence, indiscriminate variation supervened, then the affected structures must, for this reason, have begun to diminish. Or thus, not only were the affected structures directly equilibrated by the action of disuse: they were also indirectly equilibrated, by the fact of indiscriminate variation supervening upon the withdrawal of a hitherto restraining influence, viz., that of use.

I have chosen these two complementary examples from the structures concerned in mastication, because in the Biology you have pointed to these structures, as affording a crucial test of efficiency of disuse as a sole reducing cause. But altho. the examples there adduced and the arguments founded upon them are unexceptionable, if we close our eyes to the existence of the principle I am pleading for; these examples and arguments are far from unexceptionable, if we recognize this principle. In the last paragraph of § 166 it is said:— “There are some modifications in the sizes of parts, wh. cannot have been aided by natural selection; but wh. must have resulted wholly from the inheritance of functionally-produced alterations. The dwindling away of organs of wh. the undue sizes entail no appreciable evils, furnishes the best evidence of this. Take, for an example, that diminution of the jaws and teeth wh. characterizes the civilized races, as contrasted with the savage races.”15 And after shewing that this cannot be attributed to the economy of growth, the section closes thus:— “Here, therefore, the decreased action of these parts wh. has accompanied the growth of civilized habits, must have been the sole cause at work … Through direct equilibration, diminished external stress on these parts, has resulted in diminution of the internal forces by wh. this stress is met … And since the survival of individuals must always have been determined by more important structural traits, this trait can neither have been facilitated nor retarded by natural selection.”16 Clearly not by the presence of natural selection; but no less clearly it may have been affected, if not wholly effected, by the absence of natural selection. Indeed, this case furnishes an admirable example of the probable action of retrogressive variability; for it unites the action of the cessation of selection with that of the cessation of function. On the one hand, the powerful character of our ancestors’ jaws must have been to them a “trait” of a sufficiently advantageous kind, to have come under the conservative influence of natural selection: as civilization advanced, however, this powerful character became less and less requisite, and so was allowed ever proportionally to decline,—the positive influence of natural selection continuously waning, the negative influence of indiscriminate variation began correspondingly to dominate. On the other hand, the function of the jaws having decreased with the advance of civilization, the originally directly-equilibrating action was correspondingly lessened, and so, for this reason likewise, the size of the jaws was permitted to decrease, thro. spontaneous variations continually diminishing the average size by means of free intercrossing. Of course I do not mean to say that in the case of this interesting example, subsequent direct equilibration may not have had a share in the accomplishment of the observable result; but merely that the instance is not so crucial a test of the efficiency of such equilibration, as it wd. have been in the absence of the principle now under consideration.

You will observe, then, that whatever we may think as to the intensity of this principle, there can be no question as to the extent of its application; for the cessation of selection and the cessation of function together afford to this principle of retrogressive variability, a field for action larger than that open to disuse, by as much as unused organs are not amenable to directly-equilibrating causes, and by as much as unessential organs are still in a state of moderate functional activity. The former of these contingencies may, I think, be realized in a great number of cases— that is, in all cases where the unused organs in question cannot have been originally built up by the agency of directly-equilibrating causes. In stating the latter of these contingencies, it is necessary to guard the proposition by introducing the word “moderate”; for even if an organ were so far unessential as to be totally exempt from the sustaining influence of natural selection, still if it were in a state of great functional activity, we shd. antecedently expect to find that the negative influence of the cessation of selection, wd. not be enough to counteract the positive influence of such activity: continued use wd. prevent the occurrence of variations on the side of decrease, and so the average size of the organ wd. remain unaffected. It is important to elucidate this point, because altho. the contingency is one that can scarcely ever arise in wild species, it may do so in tame; and in at least one case actually has done so, i.e., in the case of the “call-duck” mentioned in the Variation.17

I may now say a few words on the special examples, wh. have been adduced to illustrate the uncompounded influence of disuse. The case of the jaws has already been commented upon. Another instance noticed by you is, to my mind, the best that has been adduced. I allude to the comparatively small size of the hands and feet of the upper classes. (Biology I p. 248).18 Here there seems little doubt that disuse must have had a large part in causing the diminution   It must be remembered, however, that the cessation of function is more or less in operation, in every case when disuse is so; and further, that in the present case sexual selection must have come largely into play. The blind cave-animals are usually regarded as affording the best demonstration of the effects of disuse in wild species. The demonstration, however, appears to me defective for two reasons; first, because the eye is a highly albuminoid structure, and so ought certainly to have come under the influence of the economy of growth; and second, because it is difficult to imagine how disuse could have affected such merely passive structures as are the transparent and sclerotic tissues. On the other hand, these structures might easily have fallen under the influence of retrogressive variability, assisted by economy of growth; for, other things equal, the greater the complexity of an organ, the greater is its liability to vary. It may be asked, Why in the case of the Crustaceans, shd. the peduncles be left? I answer, first, because they are of a much less complex structure than are the eyes of other stalk-eyed crustacea; second, we do not know how long these peduncles originally were; and third, it is not improbable that they now subsume the purpose of antennæ, and, if so, functional activity and natural selection wd. have combined to prevent their disappearance. To this it may be added that the difficulty is no greater in the case of the present hypothesis, than in that wh. supposes disuse to have been the sole cause of obliterating the eyes. Of course I have no doubt that disuse has been actively operative in this case, but as, in the case of the human jaws before alluded to, I cannot see sufficient reason to conclude that disuse has been the only cause at work. A somewhat better test of the potency of disuse is, I think, supplied to us by the web of the Upland Goose; for this, altho. tolerably vascular, is semi-cartilaginous, and has presumably been in a state of comparative disuse for a long time. Yet I find from casts wh. I have taken from the feet of these birds now living in the Gardens, that the proportion of web to the length of the toes has only decreased a very few per cent, as compared with the same parts in other species. Why retrogressive variability shd. not have taken place in this bird to a greater extent than it appears to have done, I cannot say; but you will perceive from the letter in this week’s Nature, that the fact of its not having done so, is no objection to my belief in the reducing influence of retrogressive variability in other cases.

If you have taken the trouble to follow my remarks thus far, you will perceive that the reducing influence of indiscriminate variability must necessarily be present in every case when decrement of living tissue in species is going on; whether such decrement be due to direct or to indirect equilibration. It has also been observed, that in the case of such an extensively operating principle, it wd. be very desirable to ascertain the proportion wh. its influence bears to that of other reducing causes; altho. a moment’s reflection will shew that this cannot be done with any approach to certainty,—the mere fact of its necessary association with other reducing causes, being enough to prevent our taking quantitative cognizance of its action. The following à priori considerations, however, are worth setting down.

The influence of retrogressive variability as a reducing cause, must vary inversely as the power of inheritance; and since the power of inheritance, other things equal, varies directly as the time during wh. it operates, this is tantamount to saying that the longer a structure has endured, the less rapidly will it be reduced by indiscriminate variation, at whatever time this may set in. And this, indeed, we find to be an observable fact; for artificially-bred structures—i.e. those wh. we know to have been but recently built up—succumb to indiscriminate variation much more rapidly than do other parts of the same organism (Cf. Origin. p. 120/1874).19 A corollary upon this is, that as a very rough and general rule, the influence of retrogressive variability must be more potent to reduce an organ characteristic of a species, than one characteristic of a genus; and one characteristic of a genus, than one characteristic of a family; and so on. However, as the laws of variability are so obscure, it wd. perhaps be better to throw this doctrine into a yet more general form, and say— If from any reason an organ is variable, it will, when useless, fall under the influence of retrogressive variability; while, contrariwise, if from any reason it is not variable, it will not fall under such influence; and proportionally: only in this form the proposition is little better than a truism. Nevertheless, the fact of its being so only serves to shew its necessary truth; and this truth must be regarded as one of high doctrinal importance, unless it can be shewn that, as a rule, the influence of inheritance is so predominant, that practically Inheritance : Variability : : Unity : Zero. That this formula cannot represent the truth, we have clear evidence, from the fact that, if it did, natural selection, no less than the cessation of selection, wd. be unable to act; for natural selection, no less than the cessation of selection, depends upon the power of inheritance not being absolutely constant,—i.e., upon the more or less frequent occurrence of spontaneous variations. The only difference between the two principles in their relation to inheritance lies in this, viz., the rapidity with wh. natural selection acts in the construction of an organ, is greater than that with wh. the cessation of selection acts in its destruction; for the former principle kills off all the unfavourable variations, while the latter allows them to survive. Nevertheless, these considerations tend to shew, that the basis on wh. retrogressive variability has to operate, is not such that this principle shd. be wholly disregarded.

An objection may here occur to you. Variability has throughout been treated of as tho. it were an independent principle; whereas every variation must have some determining causes, and such determining causes must all belong either to the category of direct, or to that of indirect equilibrating influences: hence, to speak of “retrogressive variability,” is merely to make a summum genus20 of all such equilibrating causes as tend to the decrement of organs. This, no doubt, is true; but as these causes are for the most part obscure, nothing is lost by speaking of variability as tho. it really were “spontaneous.”

Upon the whole, then, I think that in subsequent editions of the Biology, some mention ought certainly to be made of the reducing influence of what I have provisionally called retrogressive variability. I also think that the probability of a difference existing between the potency of use and that of disuse might be stated; and lastly, that other portions of the work shd. be altered, so as to be brought into conformity with these doctrines.21 In conclusion, I may say, that my only object in writing is to bring my views upon this subject under your notice, shd. they not have occurred to you already; in order that whatever truth they may contain, may be sifted out by the prodigious power of analyzation to wh. they are thus submitted.

I remain, etc.

P.S. I have sent a copy of this letter to Mr. Darwin.

CD annotations

10.31 unless … disuse: 10.35] scored blue crayon
12.1 You … left? 13.18] ‘Perhaps he wd say that there wd be some tendency to vary about the mean without some existing cause’ in margin pencil
13.13 because … left? 13.18] double scored pencil
14.1 you will … operating principle, 14.5] triple scored pencil


Romanes’s letter appeared in Nature, 2 July 1874, p. 164 (Romanes 1874c). He argued that disuse of an organ could cause variations that would lead to reduction in size of the organ.
Romanes refers to his earlier letter to Nature, 9 April 1874, pp. 440–1 (Romanes 1874b).
See Origin 6th ed., p. 120. CD had argued that where natural selection had not come into full play, rudimentary organs were left in a fluctuating condition.
See Origin 6th ed., p. 121.
In Principles of biology (Spencer 1864–7, 1: 466), Spencer had written: What is ordinarily called adaptation, is, when translated into mechanical terms, direct equilibration. And that process which, under the name of natural selection, Mr Darwin has shown to be an ever-acting means of fitting the structures of organisms to their circumstances, we find, on analysis, to be expressible in mechanical terms as indirect equilibration.
In Romanes 1874c, Romanes had written that the maximum reduction of organs attributable to cessation of selection was 20 to 25 per cent and referred to tables in Variation that confirmed this view; he added the tables showed there were only four exceptions, but did not elaborate on these. The tables referred to are in Variation 1: 127, 271, 273.
In Romanes 1874c, Romanes had written that the deformity of the sternum in fowls pointed to the cessation of selection rather than to disuse (the table referred to for these cases is in Variation 1: 271). The table in Variation 1: 285 shows the relative difference in the weight of bones of the leg and wing in wild and domestic ducks.
Romanes refers to Romanes 1874b and 1874c.
Spencer 1864–7, 2: 358–60.
In Romanes 1874c, Romanes claimed that the rate of reduction when an organ was disused for several generations was much greater than it ought to be if disuse were the main cause of atrophy.
Romanes quotes from H. Gray 1869, p. 30.
See Variation 1: 286. CD noted that there was not only no decrease in the weight of wing-bones in the call-duck relative to the wild duck, but an actual increase. CD had already pointed out that the call-duck, unlike other domestic breeds, was noted for its flying ability (Variation 1: 285).
In Origin 6th ed., pp. 119–20, CD elaborated the proposition that a part developed in any species in an extraordinary degree, in comparison with the same part in allied species, tended to be highly variable.
Summum genus: highest kind (Latin); the highest or most comprehensive division in a classification (OED).
No obvious changes in response to Romanes’s suggestions appear to have been made in later editions of Principles of biology, nor did Spencer refer to Romanes.


Gray, Henry. 1869. Anatomy, descriptive and surgical. 5th edition. with an introduction on general anatomy and development by T. Holmes. London: Longmans, Green.

OED: The Oxford English dictionary. Being a corrected re-issue with an introduction, supplement and bibliography of a new English dictionary. Edited by James A. H. Murray, et al. 12 vols. and supplement. Oxford: Clarendon Press. 1970. A supplement to the Oxford English dictionary. 4 vols. Edited by R. W. Burchfield. Oxford: Clarendon Press. 1972–86. The Oxford English dictionary. 2d edition. 20 vols. Prepared by J. A. Simpson and E. S. C. Weiner. Oxford: Clarendon Press. 1989. Oxford English dictionary additional series. 3 vols. Edited by John Simpson et al. Oxford: Clarendon Press. 1993–7.

Origin 6th ed.: The origin of species by means of natural selection, or the preservation of favoured races in the struggle for life. 6th edition, with additions and corrections. By Charles Darwin. London: John Murray. 1872.

Spencer, Herbert. 1864–7. The principles of biology. 2 vols. London: Williams & Norgate.

Variation: The variation of animals and plants under domestication. By Charles Darwin. 2 vols. London: John Murray. 1868.


Sets out some of his ideas on the effects of disuse on an organ. Disuse as a cause of reduction.

Letter details

Letter no.
George John Romanes
Charles Robert Darwin
Sent from
Dunskaith, Parkhill, Ross-shire
Source of text
DAR 52: D1–2, 10–14
Physical description
4pp, encl 6pp †

Please cite as

Darwin Correspondence Project, “Letter no. 9543,” accessed on 26 September 2020,

Also published in The Correspondence of Charles Darwin, vol. 22