skip to content

Darwin Correspondence Project


To Journal of Horticulture and Cottage Gardener   [17–24 March 1863]1


Had Mr. Anderson asked me two days ago for any facts illustrative of his case of unopened flowers of Cattleya crispa and Dendrobium cretaceum producing seed-capsules, I could have given no sort of information;2 nor can I now explain the fact. By an odd coincidence, yesterday I received a very interesting letter from Dr. Hermann Cruger, the Director of the Botanic Garden at Trinidad, who informs me that certain native species, and native species alone, of Cattleya, Epidendrum, and Schomburghkia, “are hardly ever known to open their flowers, but which nearly always set fruit.”3 In answer to Dr. Cruger, I have asked him to look at the seed or send me some, and inform me whether it appears good.4

Will Mr. Anderson have the kindness to send me a few seeds produced by his unopened flowers?5

I further asked Dr. Cruger whether these Orchids in their native haunts never open their flowers.6 I can hardly believe that this can be the case, seeing how manifestly adapted the structure of their organs of fructification is to the action of insects. But it is known that several plants, such as Violets, Campanulas, Oxalis, &c., produce two kinds of flowers: one sort adapted for self-fertilisation, and the other sort for fertilisation by insect agency or other means.7 In some cases the two kinds of flowers differ very little in structure; and it occurs to me as possible that something of this kind may occur with Orchids.

Dr. Cruger further informs me that with certain Orchids, as in those which do not open their flowers, the pollen-masses after a time become pulpy; and though remaining still in situ, emit their pollen-tubes, which reach the stigma, and thus cause fertilisation.8

An excellent observer, Mr. J. Scott, of the Royal Botanic Gardens of Edinburgh, will, I am sure, permit me to state that he has been making similar observations, and has seen the pollen-tubes emitted from the pollen-masses whilst still in their proper positions.9

These facts were all unknown to me when I published my small work on the Fertilisation of Orchids; but I ought, perhaps, to have anticipated their occurrence, for I saw the pollen-tubes emitted from the pollen within the anthers in the Bird’s-nest Orchid, and likewise in monstrous flowers of the Man Orchis.10 This latter fact seems related to Mr. Anderson’s remark, that flowers of an imperfect character, wanting a petal or sepal, had a great tendency to produce seed-capsules.11

These curious observations by Dr. Cruger, Mr.  Anderson, and Mr. Scott, convince me that I have in my work underrated the power of tropical Orchids occasionally to produce seed without the aid of insects; but I am not shaken in my belief that their structure is mainly related to insect agency.12 With most British Orchids this conclusion may be looked on as established.

I will only add that since the publication of my work, a number of persons have set seed-capsules with various tropical Orchids.13

Charles Darwin, Down, Bromley, Kent.


The date range is established by the relationship between this letter and James Anderson’s article in the Journal of Horticulture, 17 March 1863, pp. 206–8 (see n. 2, below), and by CD’s reference to this letter in his letter to John Scott of 24 March [1863].
In an article concerning orchid breeding, published in the Journal of Horticulture, 17 March 1863, pp. 206–8, James Anderson, who was a gardener in Uddingston, Scotland (R. Desmond 1994), asked if CD could tell him why some orchids, such as Cattleya crispa and Dendrobium cretaceum, formed apparently normal seed-capsules from apparently abortive, unopening flowers. CD’s unbound annotated copy of this number of the Journal of Horticulture is in the Darwin Library–CUL; there are also notes on the article in DAR 70: 110–11.
See letter from Hermann Crüger, 23 February 1863 and n. 5.
The letter to Crüger has not been found; however, see the letter from Hermann Crüger, 23 April 1863.
See letter from James Anderson, 1 April 1863.
See n. 4, above.
CD had long been interested by the occurrence in some plants of ‘imperfect’, unopening flowers in which self-pollination occurred (a phenomenon later called cleistogamy). His ‘provisional conclusion’ was that ‘the final object of the imperfect flowers is to produce seed safely, without any crossing … the perfect flowers being adapted for getting an occasional cross’ (see Correspondence vol. 10, letter to Daniel Oliver, 12 [April 1862]).
CD cited this information in ‘Fertilization of orchids’, p. 152 (Collected papers 2: 149).
No letter from John Scott has been found recording observations of pollen-tubes emitted from pollen masses while still in situ; moreover, Scott denied having made such claims in his letter to CD of [1–11] April [1863]. See also letter to John Scott, 24 March [1863]. However, Scott had carried out, at CD’s suggestion, numerous experiments on the emission of pollen-tubes from pollen masses placed on the rostellum of orchids (see, for example, letters from John Scott, 6 January 1863, 16 January 1863, and 21 March [1863]).
See Orchids, p. 324 n.
Journal of Horticulture, 17 March 1863, p. 107.
In Orchids, p. 1, CD stated that the object of the book was: to show that the contrivances by which Orchids are fertilised, are as varied and almost as perfect as any of the most beautiful adaptations in the animal kingdom; and, secondly, to show that these contrivances have for their main object the fertilisation of each flower by the pollen of another flower.
The only individual identified is Scott, who had succeeded in pollinating Acropera loddigesii and Gongora atropurpurea (see Correspondence vol. 10, letter from John Scott, 11 November 1862, and this volume, letter from John Scott, 3 March 1863).


Reports the observations of Hermann Crüger and John Scott that fruit is set by orchids whose flowers never open and that pollen-tubes are emitted from pollen-masses still in their proper position. These cases convince CD that in Orchids he underestimated the power of tropical orchids to produce seed without insect aid but he is not shaken in his belief that the structure of the flowers is mainly related to insect agency.

Letter details

Letter no.
Darwin, C. R.
Journal of Horticulture
Sent from
Source of text
Journal of Horticulture and Cottage Gardener n.s. 4 (1863): 237

Please cite as

Darwin Correspondence Project, “Letter no. 4069,” accessed on 26 July 2016,