skip to content

Darwin Correspondence Project

From William Herbert to J. S. Henslow1  5 April 1839

Dear Sir

I wish it was in my power to give definitive answers to Mr Darwin’s questions. They are all fit points to be investigated, but they require a longer course of experience than the life of one man, especially of one who attends to the subject only incidentally, can furnish.

I can entertain no doubt whatsoever as to the first question, that a long established variety or species must be more likely to reproduce itself uniformly from seed than one of recent origin. You have experience of the reproduction in the first case, and no certainty at all in the 2d. It certainly appears to me that if a seedling is obtained from a mule plant exactly conformable to the hybrid parent you have a much stronger prospect of an established & uniform race than you had before the reproduction of the hybrid type.

It stands to reason that the pollen of a plant which is naturally disposed to produce varieties must be less likely to produce hybrids of uniform appearance, than that of a plant which is not disposed to sport; but I consider the pollen of a permanent garden variety just as likely to produce an uniform effect as that of any easily convertible natural species. Take for instance the hollyhocks of the garden, without question cultivated varieties of one plant, yet steadily reproducing their respective colors by seed. I apprehend that the pollen of one variety of hollyhock on another would be neither more nor less decisive than that of one wild species of Calceolaria on another, at least taking the species wh. have similar constitutions & intermingle easily. These answers refer to Qy 1 & 2

Qy 3. I think crossbred vegetables lean more to the type of the father than of the mother; but I think the offspring of the mule disposed to lean to the parent of wh. the constitution is most congenial to the climate & soil in wh. it is grown.

4th. I am not aware that the genera which it is difficult to hybridize are slow to sport. Up to this day, tho’ I am still trying, I have failed in all attempts to cross Crocuses, yet there is a different either species or vary of Crocus in almost every part of S. Europe, & the garden varieties of Crocus vernus & versicolor are very numerous. I think conformity of constitution a great point to facilitate intermixtures; but I cannot conceive why the genus Hippeastrum courts hybridiza-tion, while it is produced with difficulty in the genera closely allied to it, Zephyranthes & Habranthus.

5th I have not observed that it is more easy to hybridize two cultivated plants than two wild ones. The Calceolaria raised from wild seed are willing to intermingle. The fresh imported bulbs quite as willing as those raised from seed obtained in a cultivated state.

6th. I do not think the genera in wh. hybridization is difficult stand isolated. I have made many attempts to cross species of Iris & have not yet obtained a cross; yet there is no genus in which it is more difficult to discriminate the numerous species, & to decide whether the numerous wild local forms are to be accounted as varieties of common species or not; and there is no order of plants in which it is more difficult to determine the limits of the genera which are in an almost hopeless state of confusion.

7th. I think species from remote parts are only thus far more likely to breed together than those from neighbouring localities, that in the latter case there is greater probability that they have been already approximated & have not bred together. I do not think the distance of their natural location can facilitate the disposition to interbreed.

8. I have never heard of any particular instance of the character wh. first appeared in a given plant disappearing in the first produce & reappearing in the grandchild, but I should think it of common occurrence in garden monsters if they are fertile, but in such cases there is generally a probability that the pollen of the monster may have had access to the produce that had lost the feature & reproduces it. I should think a plain seedling from a hose in hose primrose more likely to produce a hose in hose seedling, than a primrose, which had no such ancestry; a blue offspring of a wild white vary of the common squill more likely to reproduce a white one, that one of thorough bred blue pedigree.

9th. I do not think cross breeding of garden varieties inimical to fertility I did not at all mean that fruit trees might not be improved freely by crossing the varieties; but that if you cross two species of fruit trees, you are more likely to get the advantage of a new flower, than of a new fruit. For instance I obtained one fruit of which the seeds vegetated from the mule Passiflora cæruleâ-racemosa, but I have only obtained one, & that was quite deficient in juicy pulp.

10th. I have not attended to ferns or fungi.

The experience of 4 seasons has now shewn that it is certain that if, taking two hybrid Hippeastra wh. have (say) each a 4-flowered stem, 3 flowers on each are set with the dust of the plant itself & one on each with the dust of the other plant, that one on each of them will take the lead & ripen abundant seed, & the other 3 either fail or proceed more tardily & produce an inferior capsule of seed. I am at this moment trying the experiment in a bulb of that genus fresh from the Organ Mountains in Brazil to see whether its own natural pollen or that of a mule will take the lead. 2

I have answered the questions as well as I am able; but a much longer course of experiments in various genera is necessary to their perfect solution

I remain Dr Sir | Yrs faithfully | W Herbert Spofforth

April 5 1839.

CD annotations

2.1 I can … 2d. 2.4] double scored brown crayon over pencil
3.3 but I consider … natural species. 3.5] partially underl pencil, square brackets added ink
4.1 Qy 3… . grown. 4.3] double scored pencil
4.3 constitution … congenial] underl pencil
5.2 Up to … Crocuses, 5.3] scored brown crayon
5.2 cross Crocuses 5.3] underl brown crayon
5.4 & the garden … numerous. 5.5] underl ink
5.5 numerous.] ‘only colour???’ added ink
5.8 Zephyranthes & Habranthus] underl ink; ‘are there varieties of these??’ added ink
6.1 I have not … state. 6.4] double scored both margins, brown crayon
7.1 I do not … local forms 7.4] scored brown crayon
7.2 Iris] ‘Iris’ added pencil
8.1 I think … pulp. 10.6] crossed pencil
12.1 The experience … abundant seed, 12.4] double scored brown crayon
12.1 it is certain] enclosed in pencil quotation marks
12.1 taking … capsule of seed. 12.5] enclosed in pencil quotation marks
12.1 two] underl pencil; ‘[backwards?] between same parents?’ added pencil
12.2 Hippeastra] “‘any cross or crosses however complicated” [’The‘ del] another individual will not do so well.’ added ink
12.5 I am … lead. 12.8] ‘It did act so; account given in letter’added ink
15.2 1839] double underl brown crayon
Top of first page: ‘2’brown crayon
Verso of last page: ‘About Crocusses & Heaths not crossing | Progeny of hybrids dissimilar?? Important—Walker’s Law’ pencil 31839 | Only thing in this letter, 5 season had tried the Crinum Experiment’ pencil


CD had sent his questions to Herbert via Henslow.
CD later used the information on Herbert’s experiments on Hippeastrum in Natural selection, p. 399 and Variation 2: 138–9.
A. Walker 1838. In his dedication (pp. iii–iv) Alexander Walker defines his ‘law of nature’ as: ‘one parent gives to progeny the forehead and organs of sense, together with the nutritive organs contained within the trunk of the body; while the other parent gives the backhead and cerebel or organ of the will, together with the locomotive organs composing the exterior of the trunk and the whole of the limbs.’


Replies to CD’s questions on plant hybridisation and laws of inheritance. Rejects predominant transmission of characters by established forms. Males show predominance, but congeniality of parents’ constitution to climate and soil more important. No correlation between hybridisation and variability, cultivation, and geographical distribution. Rejects reversion.

Describes experiments in Hippeastrum in which pollen from another species proved more fertile than plant’s own pollen.

Did not intend to say that crossing is inimical to fertility.

Letter details

Letter no.
William Herbert (Dean of Manchester)
John Stevens Henslow
Sent from
Source of text
DAR 185: 63
Physical description
3pp ††(by CD), CD note 4pp

Please cite as

Darwin Correspondence Project, “Letter no. 503,” accessed on 26 April 2017,

Also published in The Correspondence of Charles Darwin, vol. 2