skip to content

Darwin Correspondence Project


From John Scott   [1–11] April [1863]1

Botanic Gardens



You have correctly comprehended, what I imperfectly attempted to lay before you in my last: respecting the sterility of individual plants with own-pollen;2 though, both pollen, and stigma were found perfectly capable of performing their functions, when treat reciprocally with another species. I have not as yet succeeded in proving it with different individuals of a species. I have a case or two 〈four lines excised〉 your cases in point of Passifloras;3 but, I am looking forward for some interesting results from the experiments I will try this season. However I will give you illustrations of the above from my experiments on Orchids. I have made many experiments on our plants of Oncidium sphacelatum with own-pollen; but, in no case have I succeeded in fertilising a single capsule.4 On one plant, I fertilised a hundred flowers and upwards, but every capsule aborted. I examined a number of these flowers, as they dropped off, and found in the style of each an abundance of pollen-tubes, which in most cases, I traced into the ovary. Thus the stigmatic function &c is perfect, abortion, apparently caused, by some inappreciable change, in sexual elements, preventing normal conjunction.

By applying the pollen of this species to Oncidium ornithorynchum I have got a nice plump capsule, though before I had never succeeded in fertilizing the latter; with ‘own-pollen’.— I have now a single capsule thus produced. My experiments with pollen from O. ornithorynchum or sphacelatum were ultimately unsuccessful; though indications were given, of the capsules thus treat swelling, which I never observed or very slightly with ‘own pollen’. I then tried pollen from O. divaricatum〉, from which four capsules are maturing. 〈two words excised〉 from this plant—sphacelatum—applied to 〈one line excised〉 effective 〈four lines excised〉 these.5 There is something, 〈six words excisedO. divaricatum, at least with the plants in the Gardens here.6 I know nothing of it in other places. My experiments upon individuals growing in baskets with ‘own-pollen’ have been numerous, yet all without exception unsuccessful; the ovary scarcely ever give the slightest indications of being affected by the pollen. I have, however, succeeded, in fertilising four flowers with ‘own-pollen’, upon a plant growing in a pot— In the latter state the plants are keeped much damper, and the roots less exposed, which may have some influence, especially in houses like those here in which a mixed collection of plants are grown. However, these capsules, are most abnormal looking productions, which I scarcely expect to contain a single seed.7

Though I have not in the above experiments succeeded in reciprocally crossing any two species, I have, nevertheless, I think satisfactorily shown an individual impotence of organs, capable of elimination by the action of other species. And this in a most complex and capricious manner, which might I think with illustrations you have already given, well convince your many caterers, that sterility is not a special endowment.8 I have yet another case; and this of reciprocal action, with the Maxillarias, atro-rubens and squallens.9 I have made [many] fruitless attempts to fertilise the former 〈with〉 ‘own-pollen’   I accordingly tried it with pollen from 〈the l〉atter; which has proved effective, as I have now 〈several〉 capsules maturing on M. atro-rubens, thus ferti〈lised.〉 Again, the functional efficiency of the latters pollen 〈I〉 have also proved by applying it to flowers of M. squallens, on which I have produced a capsule. I may state that M. squallens, produces capsules freely when fertilised with ‘own pollen’.

I have thus imperfectly given you all the information my experiments have yet afforded me on the above subject. The use to be made of it, I beg to leave entirely to you, and I shall certainly be highly gratified if it merits your recording.10

In respect to fertilising ovules by application of pollen to rostellum, permit me to make a few remarks. In my last I mentioned certain genera of Orchids, whose rostellum still exercised stigmatic function, as evidenced by inducing pollen-grains to emit tubes.11 In making this statement, I did not at all intend its general application: it had reference to those genera mentioned,—only—in which I had never observed the pollen tubes produced in situ.12 Your observations have long since convinced me that the power of inducing germination of the pollen-grains, is not a peculiar prerogative of the stigma. Nevertheless, this power of dependent germination is not at all a common pheno〈menon as〉 far as yet known, I think you will readily a〈llow〉, the great majority of plants have been found 〈to prod〉uce pollen-tubes only, when their pollen is in contact 〈w〉ith stigma or its secretions. It was therefore 〈on this〉 consideration that I mentioned the above statement. It is somewhat astonishing to see how much the degree of excitability of pollen-grains varies in species of a genus or even in individuals of the former. I am somehow inclined, however, to suppose that careful research and experiments will show that this difference in the degree of excitability is more apparent than real.

In respect to Dr Gray’s observation, permit me to ask a single question.13 Am I right in supposing that he found the tubes in the soft moist tip of rostellum alone? or has he in reality traced them downwards? In his remarks, he speaks of the grains falling upon the naked cellular tip of the narrow process of rostellum, and upon the cellular sumit the process outside of each disc, and sending tubes down freely into their substance, I am somehow, inclined to think that the term “substance” has here a special reference to that portion of the rostellum mentioned—its cellular tip—though certainly it may possibly be intended to include the entire organ. As far as my observations go, the rostellum seems absolutely impervious by pollen-tubes; and certainly one might have expected to have seen, at least some, of the pollen-tubes protruding from pollen grains in contact with rostellum—mentioned in former letter14—penetrating the latter, if it had been possible, instead of being immediately deflected, and extending themselves upon its surface. I am now, indeed, 〈more than〉 ever inclined 〈to a〉ccept your f〈irst su〉ggestion on the absence of stigmatic utriculi— on a〈  〉 〈    〉 tissue, which of course includes the former—probably preventing fertilisation.15 Indeed if it proves otherwise it will be a fatal blow to our attempts at discovering physiological changes in the structure of the styles of sterile hybrids, or at least of giving such if discovered a causal signification.16 But, independent of the above there are other difficulties, of an almost insuperable nature to contend with: for example, how is the Catasetum & Acropera cases to be reconciled with the preceeding view. You have shown that the structure of the female organs in the former is much less abnormal than in the latter; nevertheless, consider the different results of experiments on their fertilisation.17 And even supposing, that Catasetum should ultimately yield a capsule to some assiduous & skilful operator, we will still have the anomalous Acropera! I candidly admit, that my experiments on the latter genus and Gongora—I have yet to state those on the latter, which are curious, I will do so in an early letter—lead me to doubt, (though perhaps, I may happily be wrong) the absolute sterility of Catasetum. 18 Dr. Hooker’s experiments also as you will remember, in fertilising ovules of Meconopsis by〉 direct application of the pollen, and thus entirely dispensing 〈with〉 stigma & style, present a grave difficulty.19five words missing〉, such 〈three words missing〉 rapid extension of the tubes, as well as their passi〈n〉g through matter of a denser consistency will I am incline〈d to〉 think prevent any osmotic action between them and stigmatic utriculi. But I need not further particularize, as all are familiar to you, and I mention them merely as difficulties which I have felt in the following out of your suggestions. I will be glad, if time and health permit you, to hear whether or not, they are really such.

Since I last wrote you I have made an interesting and very opportune observation on Epidendrum cochleatum, which you may like to hear.20 On a plant of the above I observed a somewhat abnormal looking flower, which was unusually long in expanding: this I found on examination to be due to complete adhesion of the outer segments of the perianth, which were much shorter than usual. Under the impression that there might possibly be some interesting abnormalities in the interior structure, I at last cut the flower for examination. I was disappointed in this, however, there being simply an abortion or nearly so of the inner circle of the perianth, and the column seemed perfectly normal. There, however appeared to be an unusual amount of viscous matter ab〈ove the〉 stig〈ma〉tic chamber, which covering 〈in〉 part the rostellum, laterally reached, the sides of an〈thers〉 [illeg] 〈your〉 〈ob〉servations on Neottia and Aceras 〈imme〉diately induced a hope that probably this near approach of stigmatic secretions to the pollen-masses, might have excited the emission of pollen-tubes.21 I accord〈ing〉ly placed sections of the column under the microscope and was much gratified to find that they had protruded; and this not only from below that lateral portion of the anther where the stigmatic secretions, all but touched the pollen-masses; but, likewise, from the centre extending themselves over the anterior surface of rostellum, downwards into the stigma. Thus the simple rostellum of Epidendrum has likewise afforded me no illustration of its penetrability by pollen-tubes. If I have an opportunity this season, I will make a few experimental observations on the British Gymnadenia, and see if I can discover anything analogous to Dr Gray’s observation.22

I have seen your letter in Cottage Gardner.23 I thank you most sincerely for the honour you have done me in referring to my observations.24 Permit me to ask a single question or so in reference to Dr Cruger’s observations on the pollen-masses of unopened flowers becoming 〈pu〉lpy, and then emitting their tubes.25 Does he make no mention of the production of fungi upon this pulpy 〈m〉ass? This I ask in consequence of an ob〈serva〉ti〈on〉 upon polle〈n m〉asses of Blet〈ia〉 〈which〉 assumed a somewhat pulpy form. I placed a little 〈of〉 th〈is〉 under the microscope, and was much pleased to observe, what I then fancied, an abundance of pollen-tubes. Great was my astonishment, however, on a careful examination under a higher power, to find that what I had fancied, pollen-tubes, were in reality the threads of some fungus. (Possibly—but I know nothing about fungi—a species of Aspergillus.) They protruded from the pulpy mass, in the form of elongated jointless filaments, some of which bore at their extremities processes from which neck-laces of spores originate. But for the latter, I candidly admit—though perhaps it is a most unpardonable admission—that I should have considered this as illustrative of the emission of tubes from pollen while still in situ. I have made similar observations on pollen from another flower, but after the most careful scrutiny I failed to observe a single pollen-tube, though I saw an abundance of fertile and sterile threads of a fungus.

Permit me to make a few remarks on Dr Cruger’s suggestion that the emission of pollen-tubes-grains in situ is due to ants carrying stigmatic secretion to pollen,26 and that of Mr. Anderson’s that flowers of an imperfect character have a great tendency to produce seed-capsules.27 Dr Cruger, may possibly have based his suggestion on the observance of stigmatic secretions in contact with anthers. If it should be 〈an observ〉ation of this nature, my experiments on fertilising the rostellum, incline me to suppose, that he possibly mistakes the effect for the cause. I have already made—as you may remember—a somewhat similar mistake in Lælia.28 I then stated that two lateral processes were developed by the stigmas which extended themselves upwards to rostellum, and thus connected the pollen-masses applied to that organ with the stigma. These processes are simply the hardened stigmatic secretions. In all my experiments on this point—excepting of course those in which plaister of Paris was applied to the stigma—I in general observed a superabundance of viscous matter exuded by the stigmas. But this I now believe does not take place until at least some of the pollen-tubes have penetrated the stigma: these somehow concentrate, assimilating or organising forces of the flower, and induce an increased activity in the secretion of viscous matter, which in most cases fills the stigmatic cavity, and even in that of Epidendrum, above referred to—reached the anther. Before the penetration of the stigma by the pollen-tubes the secretion of viscous matter is limited, in consequence of the assimilative forces of the flower being directed somewhat equally to its different parts, and we find that this balance is in general maintained in an unfertilised flower; its various parts decaying together. The penetration of the stigma by pollen-tubes however, turns the balance, the 〈perianth〉al parts quickly fade, and all is directed to the supply of the ovary. The mere circumstance then, of the viscous matter of the stigma reaching the anthers or pollen-masses—does not in my opinion, justify us in regarding it as necessary for the emission of tubes; it must rather in my opinion be regarded as a consequence of the emission and penetration of the stigma by pollen-tubes.

In respect to Mr Anderson’s observations, or rather—as I see he limits it to flower wanting a sepal or petal—Dr Cruger’s on unopened flowers so frequently setting fruit.29 The greater humidity, and equability of temperature, consequent on such conditions, is I believe the probable cause for these abnormally conditioned flowers, so frequently fertilising themselves. A consideration of the extreme delicacy of the pollen-tubes must at once convince us of the greatly increased facility such conditions afford for accomplishing the above phenomenon, than obtained in the normal condition. Indeed; in my opinion, I do not think it improbable that the above phenomenon—the germination of pollen grains in situ—which in the majority of cases occurs only in closed flowers, is a purely accidental occurrence, having no ultimate import in the economy 〈of〉 the spec〈ies.〉 That in fact the latent vitality of the pollen grains is simply, and necessarily excited by the abnormal conditions of the flower. I certainly admit that those cases of Viola and Oxalis, on which the different flowers are produced at distinct periods, renders the above somewhat improbable.30 But are you aware of the strict observance of this law in those cases? do they not also occur together upon the plant? the abnormal, self-fertilising flowers being arrested in their development. I think, but I will study the subject more carefully, that I have observed both flowers together on plants of Oxalis Acetosella in our hot-houses, in some of which it has become a great pest. In the cases of Impatiens and Campanula they seem to occur on plants at the same period and thus I think justify us in regarding the imperfect flowers, as simply due to a casual arrest in the development of the perianth, and having nothing whatever to do with special purposes in the economy of the species. That in fine we have no more right to look for such ends in this case—even though the ultimate end of vegetal life is perfectly accomplished—than we have in the case of other abnormalities in the flower, regularly reproduced—peloria, for example. I may be quite wrong in this view, nevertheless I cannot but think that we might w〈i〉th equal justice suppose that nature had some ultimate end to serve 〈in the〉 latter as well as the former. I have ventured these remarks, under the conviction that the inferences 〈as w〉ell as the demands, I have seen founded on this phenomenon, are alike unwarranted. I cannot see for example how it can possibly be maintained, that this continued self-fertilisation exercises no baneful influence on successive progeny. Now, inasmuch, as no single species, I think, can be instanced, as absolutely dependent on this exclusive mode of self-fertilisation: the “struggle for existence” between the offspring of the perfect and imperfect flowers, must necessarily be brought into action, and thus invalidates the objection, to your views, on the necessity of occasional crosses for the healthy continuation of species.31 But I need not further follow out this subject, I have said sufficient to show you in what light I regard it. And if I am wrong, I have already said more than enough. I may remark that I have commenced experiments to elucidate, if possible the subject in Orchids, by cementing the perianth of perfect flowers shortly before their expansion. I will thus see if I can artificially produce the phenomenon.

The results with homomorphic Cowslips is a little provoking.32 I now think, however, we ought to have anticipated these results, from experiments with such plants, as Cowslips, Primroses & Auricula &c. which present both forms. Inasmuch as the plant from which we save our seed might have heteromorphic parents, and thus would have a greater tendency to produce heteromorphic offspring. An exact experiment, would thus necessarily require a knowledge of a plants lineage for a number of generations. On a consideration of this I think we may look for satisfactory elucidations of this point from those long-cultivated exotic species, that present both forms in their native wilds; but of which one only has been introduced. These I look forward to with more hope; though certainly in the case of homomorphic crosses with Cowslips &c. we ought to have the majority of offspring presenting parent form.33

But I fear I will long ere this have exhausted your time and patience: Pray excuse me,

And now sincerely trusting that your health is still progressing favourably, I remain | Sir | Yours very respectfully | John Scott

I have just had a letter, in reply to one I sent, from one of the most experienced Auricula growers in Scotland.34 He informs me that as soon as a long-styled plant presents itself it is immediately thrown out. And yet from seed saved from these short-styled plants, he states that usually one-fourth are long-styled. They certainly bear a much higher proportion, than I expected.35 | J. S.

He also informs me that he never observed the two forms produced by one plant

CD annotations

2.8 There is … in a pot— 2.14] scored brown crayon
4.3 recording.] before closing square bracket, brown crayon
5.1 In respect] after opening square bracket, brown crayon
6.33 But I need … phenomenon. 10.42] crossed ink
11.1 The results … of which one 11.9] crossed pencil
11.9 only has been … John Scott 13.2] crossed ink
14.1 I have just … one plant 15.2] scored brown crayon
Top of letter: ‘Sterility of Orchids | John Scott.— | Orchid Portfolio— Rostellum’ ink; ‘Crüger’ blue crayon
End of letter: ‘April. 1863’ ink


The date range is established by the relationship between this letter, the letter to Journal of Horticulture and Cottage Gardener, [17–24 March 1863] (published in the issue for 31 March 1863), and the letter to John Scott, 12 April [1863].
See letter from John Scott, 21 March [1863] and n. 4, and letter to John Scott, 24 March [1863] and n. 3.
See letter from John Scott, 21 March [1863] and n. 4, and letter to John Scott, 24 March [1863].
Scott’s experiments in pollinating species of Oncidium with their own pollen are discussed in Scott 1863a, pp. 545–7. See also letter from John Scott, 21 March [1863] and n. 5.
In his account of these experiments prepared for the Botanical Society of Edinburgh, Scott wrote that he found his most successful crossings to be those in which he pollinated six flowers of Oncidium sphacelatum with pollen from O. divaricatum var. cupreum; four fine capsules resulted. He also tried unsuccessfully to fertilise flowers of O. altissimum and O. graminifolium with pollen from O. sphacelatum (Scott 1863a, pp. 547–8).
Scott was a foreman in the propagating department at the Royal Botanic Garden, Edinburgh (R. Desmond 1994).
See Scott 1863a, pp. 548–9.
Scott 1863a, p. 549. Scott discussed the function of sterility and its relation to CD’s theory in Scott 1863a, pp. 543–4. CD discussed his opposition to the notion of sterility being a special endowment in Origin, pp. 260–3. See also letter from John Scott, 6 January 1863 and n. 10, and Correspondence vol. 10, Appendix VI.
Scott’s experiments with species of Maxillaria are described in Scott 1863a, p. 549.
CD encouraged Scott to publish his results (see letter to John Scott, 12 April [1863]). In Variation 2: 133, CD discussed Scott’s work with Oncidium, also stating that ‘Mr. Scott found that Maxillaria atro-rubens was “totally insusceptible of fertilisation with its own pollen,” but fertilised, and was fertilised by, a widely distinct species, viz. M. squalens’. See also Orchids 2d ed., p. 289.
See letter from John Scott, 21 March [1863]. See also letter to John Scott, 24 March [1863], and letter to Journal of Horticulture and Cottage Gardener, [17–24 March 1863] and n. 9.
See letter from John Scott, 21 March [1863] and nn. 7 and 8.
Asa Gray had reported that in Gymnadenia tridentata pollen grains could fall onto ‘the naked cellular tip of narrow process of the rostellum’, germinate, and their pollen tubes would penetrate the rostellum (A. Gray 1862a, p. 426). After CD sent Scott a copy of A. Gray 1862a (see letter to John Scott, 6 March 1863), Scott discussed Gray’s statement, noting that in his own experiments he had never observed pollen tubes ‘insinuating themselves into the tissues’ of the rostellum (see letter from John Scott, 21 March [1863]).
See letter from John Scott, 21 March [1863].
The reference is to Orchids, p. 311 (see letter from John Scott, 16 January 1863 and n. 6).
See Origin, pp. 260–7.
The role of the rostellum, stigma, and style in pollination appeared to be different in Catasetum and Acropera compared with other orchids (see Orchids, pp. 203–10, 231–48, and letter from John Scott, 18 February [1863] and nn. 4 and 5). See also letters to John Scott, 21 January [1863], 20 [February 1863], and 24 March [1863], and letter from John Scott, 3 March 1863.
Scott apparently refers to his pollination experiments with Gongora truncata, the difficulty and ultimate success of which he reported in his letter to CD of [after 12] April [1863]. Scott had previously told CD of his success in pollinating G. atropurpurea (see letter from John Scott, 3 March 1863).
The reference is to experiments described in J. D. Hooker 1854a. See also letter from John Scott, 6 January 1863 and n. 8.
CD compared the stigma and rostellum of Epidendrum cochleatum in Orchids, pp. 310–11, and Orchids 2d ed., pp. 249–50; Scott’s specific observations were not mentioned.
See Orchids, p. 324 n., and letter to John Scott, 24 March [1863].
See n. 13, above.
CD’s letter to the Journal of Horticulture and Cottage Gardener of [17–24 March 1863] appeared in the issue for 31 March 1863.
In his letter to the Journal of Horticulture and Cottage Gardener of [17–24 March 1863], CD referred to Scott as an ‘excellent observer’, and cited his observations of orchid pollen tubes.
See letter to Journal of Horticulture and Cottage Gardener, [17–24 March 1863]. See also letter from Hermann Crüger, 23 February 1863.
See letter to John Scott, 24 March [1863], and letter from Hermann Crüger, 23 February 1863 and n. 5.
See letter to Journal of Horticulture and Cottage Gardener, [17–24 March 1863], and James Anderson’s letter in the Journal of Horticulture, 17 March 1863, pp. 206–8.
See letter from John Scott, 16 January 1863. See also letter from John Scott, 21 March [1863].
See letter to Journal of Horticulture and Cottage Gardener, [17–24 March 1863].
See letter to Journal of Horticulture and Cottage Gardener, [17–24 March 1863] and n. 7. CD began experimenting in 1862 with the unopening, ‘imperfect’ flowers of Viola and Oxalis; these appeared at different times of the year from the larger, opening flowers (see Correspondence vol. 10, letter to John Scott, 12 November [1862] and n. 5, and letter from John Scott, 15 November [1862]).
In the last paragraph of Orchids, p. 359, CD argued that nature ‘abhors perpetual self-fertilisation’. See also Origin, pp. 96–101.
In his letter to John Scott of 24 March [1863], CD stated that cowslips raised from seed resulting from self-pollination of a short-styled parent yielded both short- and long-styled progeny; CD had expected short-styled seedlings only.
With CD’s encouragement, Scott was already undertaking his own experiments with Primula species, including cowslips (see Scott 1864a, pp. 106–10).
Scott may refer to George Lightbody of Falkirk, who raised numerous varieties of Primula auricula (R. Desmond 1994).
Scott later published his own work with Primula auricula (see Scott 1864a, pp. 86–97).


Studying self-sterility, particularly in Oncidium, where abortion occurs consistently but stigma functions normally. His hybrid orchid crosses show sterility occurs capriciously. Thus it is not a "special endowment".

Disputes Asa Gray’s and Hermann Crüger’s view of rostellar germination.

Doubts absolute sterility of Catasetum.

Disappointed by results with homomorphic cowslips.

Letter details

Letter no.
Scott, John
Darwin, C. R.
Sent from
Botanic Gardens, Edinburgh
Source of text
DAR 108: 183, DAR 177: 86 (fragile)
Physical description
14pp †

Please cite as

Darwin Correspondence Project, “Letter no. 4073,” accessed on 30 July 2016,