skip to content

Darwin Correspondence Project

From C. B. Clarke   12 May 1878

Herbarium, Kew

12 May 1878

Dear Sir,

It has been suggested to me here that I should communicate to you the following case of extreme dimorphism—

Order Rubiaceæ Genus Adenosacme Bth & Hk. f. Gen. Pl. II. 69.1

The best-known species is Adenosacme longifolia Wall. Cat. 6280 = Rondeletia longifolia, Wall. in Roxb. Fl. Ind. ed. Carey & Wall. II 137.2

The only figure is Adenosacme (Lawia) acuminata, Wight Icon. t. 10703

In the Gen. Pl. l.c. Hooker says (of the genus) “stamina 4–6 medio vel supra medium corollæ tubi inserta”4

Wallich in Roxb Fl. Ind l.c. says (of species longifolia) that the stamens are inserted at the very base of the tube.5

Wight in his Icon shews that the stamens (in species acuminata) are not merely at the base of the corolla tube but he depicts the corolla tube removed leaving the stamens standing on the ovary !! On the slip of paper enclosed I have copied (badly) the two poor pictures of Wight.

On the same piece of paper I have copied four old field notes made by myself from fresh examples of Adenosacme longifolia: collected in distant places and times. I also enclose buds in numbered covers cut from the dried specimens from which these sketches were originally taken. It is possible of course that different forms of flowers may occur on the same shrub but among the Coffeaceæ my experience is that this rarely happens.6

In my No 26947 the stamens are very nearly if not quite free from the corolla (as Wight has depicted in the other species).

The peculiarity of the case is not the difference in length of the style and stamens in the two forms but the extreme difference in the point of insertion of the stamens. The note made from No 14591 shews a distinctly intermediate position of the stamens; the shrub may be trimorphic— I may add that I have no doubt that the four figures referred to A. longifolia are all strictly of one “species”, a marked one that I know well.

Wallich has repeatedly noticed the common occurrence of dimorphism in the Coffeaceæ: it is so common indeed that the length of the style and insertion of the stamens usually given as part of the generic character in this order are (in my opinion) almost useless for determining species.— You are doubtless aware that the cultivated Cinchonas are all dimorphous—7

The following may interest you as it occurs in a genus not much observed in Europe

In Combretum acuminatum Roxb.8 before the calyx expands, a minute round hole appears at the top of the bud. The long simple style (the stigma is not broader than the style) protrudes through this fitting it closely, not a grain even of pollen could get out— After the style has withered the calyx (and small petals) expands, the anthers ripen—

Believe me | Yours with deep respect | C. B. Clarke

C. Darwin Esqre.

[Enclosure 1]



[Enclosure 2]


[Folded piece of paper labelled ‘9134’ containing flower bud]

[Enclosure 3]


[Folded piece of paper labelled ‘26947’ containing flower bud]

[Enclosure 4]


[Folded piece of paper labelled ‘14591’ containing flower bud]

CD annotations

On cover: ‘Curious case of Dimorphism.— Rubiaceæ’ ink


Clarke was on leave from the Bengal civil service and since 1877 had been working at the Royal Botanic Gardens, Kew, on descriptions for Joseph Dalton Hooker’s Flora of British India (Hooker 1872–97). The genus Adenosacme was described in Genera plantarum (Bentham and Hooker 1862–83, 2: 69). Orders in Genera plantarum are analogous to families in modern systematics.
The Wallich catalogue, compiled between 1828 and 1847 by Nathaniel Wallich, listed dried specimens of plants in the collection of the East India Company Museum; Adenosacme longifolia (Rondeletia longifolia is a synonym; both names are synonyms of Mycetia longifolia) is entry number 6280 (ibid., p. 213). Clarke also refers to the edited version of William Roxburgh’s Flora indica, which had additions by Wallich (including the entry for Rondeletia longifolia; Roxburgh 1820–4, 2: 137–9). In his description of Adenosacme longifolia, Hooker noted that the flowers, as observed by Wallich and Clarke, were dimorphic or trimorphic (Hooker 1872–97, 3: 95).
There is a woodcut of Lawia acuminata (a synonym of Mycetia acuminata) in Robert Wight’s Icones plantarum Indiae Orientalis (Wight 1840–53, 3: pl. 1070).
Stamina 4–6 medio vel supra medium corollae tubi inserta: 4–6 stamens inserted at or above the middle of the corolla tube (Latin).
Roxburgh 1820–4, 2: 138 (see also n. 2, above).
The buds are reproduced at QQQQ of their original size. The buds were folded into numbered pieces of paper; the sketches have corresponding numbers. The family name Coffeaceae is a synonym of Rubiaceae.
Cinchona is a genus in the family Rubiaceae; the trees Cinchona officinalis and C. pubescens were cultivated for their bark, a source of quinine. The flowers are heterostyled.
Combretum acuminatum is a species in the family Combretaceae (bushwillow).


Hooker, Joseph Dalton. 1872–97. The flora of British India. Assisted by various botanists. 7 vols. London: L. Reeve & Co.

Roxburgh, William. 1820–4. Flora Indica; or descriptions of Indian plants … to which are added descriptions of plants more recently discovered by Nathaniel Wallich. Edited by William Carey. 2 vols. Serampore: Mission Press.

Wallich catalogue: A numerical list of dried specimens of plants in the East India Company’s Museum, collected under the superintendence of Dr. Wallich of the Company’s Botanic Garden at Calcutta London. [1828–49.]

Wight, Robert. 1840–53. Icones plantarum Indiae Orientalis: or figures of Indian plants. 6 vols. Madras: the author.


Reports curious case of dimorphism in Rubiaceae. Encloses envelope containing bud samples.

Letter details

Letter no.
Charles Baron Clarke
Charles Robert Darwin
Sent from
Source of text
DAR 161: 160
Physical description
ALS 2pp, sketch, encl †

Please cite as

Darwin Correspondence Project, “Letter no. 11505,” accessed on 13 July 2024,