skip to content

Darwin Correspondence Project

From William Herbert   [c. 27 June 1839]

Dr Sir,

I return my best thanks to you for your obliging present of your very interesting journal this moment received. I was employed part of yesterday reading another copy of it with great pleasure. The present moment is always the best & I will give the best answer I can to your queries. The first question I should answer by adding to the words “2 hybrid Hippeastra” the following words of “any cross or crosses however complicated”. I am pretty sure that I mentioned in my letter to you that I was about to institute an experiment to try if the mule pollen from another plant could beat in the same manner the natural pollen on a wild bulb; & the startling fact has followed that it did so. I took a bulb fresh from the Organ Mountains in Brazil1 of a new species allied to H. Aulicum or variety thereof which produced 2 twoflowered scapes the one having a few days precedence of the other. I impregnated the two flrs. of the 1st. scape with its own pollen & one flower of the second; & having removed the anthers I touched the 4th. flower with the pollen of a triple cross raised by pollen of the many flowered-smooth Brazilian H pulverulentum from a cross between H. vittatum & Reginæ which are bearded. At first the three germens impregnated with the natural dust began to swell for seed & the 4th. was stationary. After a few days the 4th began to swell & from that moment the other three stopped short. The progress of the 4th. was then rapid & in a few more days the 3 first died away & a vigorous capsule was produced by the cross bred pollen. My attempts to cross the crocuses continue to fail. There is great difficulty in extracting the anthers of Croci & English heaths before the pollen is shed, the petals being so involved in the bud that they cannot be easily opened, and if the corolla is prematurely damaged the chance of seed is diminished by the injury.

I do not by any means think it impossible for insects to fecundate those flowers by pollen from other individuals, but I think there must be flowers which are always impregnated by their own males. Flowers of Œnothera can scarcely expand from the bud without impregnating the stigmas, but certainly a moth might light of the stigma at the very moment after expansion & bring pollen from the next flower. I stated in my treatise2 that a course of experiments was wanting to ascertain how soon after contact with the pollen, the ovary was fertilized & closed against all further influence, & till that is ascertained I do not see how the question can be answered positively; & probably the speed of the impregnation varies in different genera.

The objection to your theory wh. occurs to me is this. In some genera the facility of breeding crosses is great; in Hippeastrum there seems even a decided preference for foreign pollen: but if the fecundation by another individual was essential, the cases of wild cross-bred plants would be frequent instead of being rare.

We have so little experience concerning the successive generations of crossbred plants, & they are rarely produced in situations where “impregnation from other quarters may not be suspected” that I cannot answer your last question. I do not believe any general law can be laid down. I suspect a greater disposition to revert to the male than the female: but it is a strange fact, that there seem to be plants of which the first or equal cross is fertile, but the next cross of 34 & 14 sterile, but that fact requires further investigation. I hope to publish in the spring or sooner a supplement to my Amaryllidaceæ3 with some further remarks on crossbred plants, but my knowledge is very defective as to much of the mystery.

I may add to my answer to the first question, that the pollen of another individual seedling of the same cross has not the same effect, as the pollen of another individual of a different species or different cross. I am speaking of the genus Hippeastrum only.

I remain Dr Sir | Yrs faithfully | W Herbert 16 Chester St


CD annotations

1.1 I return … that it did so. 1.9] crossed pencil
1.21 anthers of Croci … injury. 1.24] scored pencil
1.24 injury.] ‘As Gärtner’4 added pencil
2.3 Flowers … next flower. 2.6] scored pencil; ‘against intermarriage’added pencil
2.9 & probably … genera. 2.10] ‘is not this the converse fact’ added pencil
3.1 In some genera … rare. 3.4] scored pencil; ‘against intermarriage’added pencil
4.1 We have … mystery. 4.9] crossed pencil
4.7 further investigation] ‘a’ added pencil, circled pencil
Top of first page: ‘4’ brown crayon

CD note:

Mr Herbert considers *without doubt [interl] old variety more likely to reproduce itself than new—but he throws doubt on this, as if not necessarily effect of ages, but of the chances.—

Negatives a relation between facility in sporting and hybridisation.— but says that *stands to reason that [interl] the pollen of plant liable to sport does not produce such uniform effects on offspring—as one that does not.—but believes there would be no difference between a permanent variety and easily convertible species: all facts about the reappearance of character must be inferred from the animal kingdom.

The non relation of hybridizing power & variation goes far to overturn my views of fixity of character, being dependent on age, & consequently on time after certain period having any further effect on characters.—& if Mr Herbert is right there should be no [‘effect’ del] difference when an Esquimaux dog *& Wolf [interl] is crossed with other dog: yet to be sure, the Esquimaux & common Dog, breed freely whilst the [‘Holly’ del] thorn varieties *of plants [interl] which grow near each other & do not cross, must have some repulsive tendency.—

The greater [interl] fertility of two hybrids very perplexing to my view of *hybrids going back to old type when crossed [underl pencil]5 with one parent & so becoming fertile.—it is new element the offspring reverting to the constitution most congenial to site—

The [‘fixity of character’ del] *resistance to hybridisation [interl] being consequence of [‘olddel] *age of [interl] species *(inexplicable by Mr Herbert) [interl] is so simple an explanation, that I can hardly give it up, though [‘these’ del] some [interl] species varying & yet not hybridising almost overthrows it.— Is there any case of species which can easily be hybridised *& hybrids be fertile [interl] & yet does not vary?—

Negatives isolation *in series [interl] in non-hybridising genera.6

close species of Crinum & Iris will not cross, yet being many & local, probably *according to my theory [interl] recently produced.

In fact at present all that can be said, is that in wild species, some difference of unknown nature, which cannot be predicated, but which is not always present, originates, which causes two not to breed, or if they do breed, causes offspring to be sterile. And this difference whatever it may be has never been produced amongst tame variations. I must state that no relation of hybridising & variability is opposed to the old view, which still seems innately probable.— Quote Herbert.— But then we find equally inexplicable amt of variability in organisms taken out of their proper conditions *(but here offspring never rendered infertile) [square brackets in MS], & we cannot take these cases of infertility, (we cannot any more explain them) as any criterion of what are species, & why shd. we in case of crossing.—

Not more easy to hybridise cultivated than uncultivated Plants7


Herbert published only a half page note describing two species of Amaryllidaceæ (W. Herbert 1840), which does not seem to be what he had in mind.
Gärtner 1849, p. 357.
CD underlined ‘hybrids … crossed’ in pencil and added in pencil ‘wrong view, because fertility increase by cross with another hybrid.’
The text up to this point is in ink, that following is in pencil.
This memorandum was kept by CD with the Herbert correspondence, see questions for William Herbert, [c. 1 April 1839] n. 7. CD subsequently annotated it in brown crayon:

2.1 Negatives a relation … hybridisation.—] scored

3.1 The non relation … my views] underl

4.2 it is new element … to site— 4.3] underl

6.1 Negatives isolation … genera.] double scored

Top of first page: ‘2a’.


Gärtner, Karl Friedrich von. 1849. Versuche und Beobachtungen über die Bastarderzeugung im Pflanzenreich. Mit Hinweisung auf die ähnlichen Erscheinungen im Thierreiche, ganz umgearbeitete und sehr vermehrte Ausgabe der von der Königlich holländischen Akademie der Wissenschaften. Stuttgart: E. Schweizerbart.

Herbert, William. 1837. Amaryllidaceæ; preceded by an attempt to arrange the monocotyledonous orders, and followed by a treatise on cross-bred vegetables, and supplement. London: James Ridgway & Sons.

Herbert, William. 1840. Amaryllidearum species novae. Annals of Natural History 4: 28.


Rejects necessity of outbreeding and any general law of reversion.

Describes further experiments with Hippeastrum showing greater fertility with foreign pollen than with individual’s own pollen or with pollen from another individual of same species.

Does not believe CD’s questions about reversion can be answered in present state of knowledge.

Letter details

Letter no.
William Herbert, dean of Manchester
Charles Robert Darwin
Sent from
London, Chester St, 16
Source of text
DAR 185: 67
Physical description
3pp †

Please cite as

Darwin Correspondence Project, “Letter no. 524,” accessed on 29 November 2021,

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