skip to content

Darwin Correspondence Project

To Charles Lyell  [7? December 1849]

Down Farnborough Kent

Friday Even

My dear Lyell

I confess the dikes overflowing on both sides made me at first feel uncomfortable; but I quite agree with you that this case is exceptional: it is nevertheless, I fear, probable that on those volcanos, in which it does occur, it will occur frequently. Dana wd require reconsidering with respect to the cause; it certainly appears connected with liquidity or absence of much gaseous emissions. I do not think I shd be of any service in sending you the pages, for I strongly recommend you to read (it will be quickly done) whole Hawaii Arch. Chapter,1 & perhaps summary of volcanic action in Pacific— The other volcanic chapters have little in them.—

I think you ought to consider what Dana’s evidence is for so many currents proceeding from fissures; it struck me that his lines or zones of clinkers, might possibly have been produced by gases proceeding for some time from a crack, in nearly same way as the numerous great bubbles were produced in the lava-stream, which I also saw at Galapagos.— I shd. not hesitate to contradict E. de B. not so much from number of rectangular junctions, as from Etna having been a scoriæ-producing mountain; there must be cones where there is much scoriæ.—2

Dana gives woodcut of what I think may be a Denudation-crater,3 with two great passages; or exits  this wd. perhaps be worth considering.— Again Dana discusses inclination of lava streams, & disbelieves E. de B; but I fear his inclinations were only estimated. He gives nearly same inclination as I do viz 30o for the tuff=craters, which must be very like those of Galapagos; Dana wd. have made his book more valuable, I think, if he had more compared his results with those of others.— You will be interested by his very curious observations on the different heights of contemporaneous lava-eruptions.— How curious, also, are the columns formed by driblets of lava! I do not know whether these remarks will be of any use to you in calling your attention to these points. From what I have seen, comparing Galapagos & Ascension, the difference in liquidity of lava is immense.—

Dana is dreadfully hypothetical in many parts & often as “d——d cocked sure” as Macaulay.4 He writes, however, so lucidly that he is very persuasive: I am more struck with his remarks on denudation than you seem to be; I came to exactly same conclusion in Tahiti that the wonderful vallies there (on the opposite extreme of scale of wonder with the valleys of N.S. Wales) were formed exclusively by fresh-water.— He underrates the power of sea no doubt,—but read his remarks on vallies in Sandwich group.— I came to conclusion in S. America (p. 67 of vol) that main effect of fresh-water is to deepen valleys & sea to widen them; I now rather doubt whether in valleys or fiords far penetrating country, the sea would deepen rock at its head during elevation of the land..— I shd. like to tour on W. coast of Scotland & attend to this; I forget how far generally the shores of fiords, (not straits) are cliff-formed:— It is a most interesting subject.—

To return once again to coral: I find with some reefs, he does not differ so much in detail with me regarding areas of subsidence:—his map is coloured on some quite unintelligible principle, & he deduces subsidence from vaguest grounds, such as that the N. Marianne isld must have subsided because they are small, though long in volcanic action; & that the Marquesa subsided because they are penetrated by deep Bays, &c &c.— I utterly disbelieve his statements that most of the atolls over whole have been lately raised a foot or two; he does not condescend to notice my explanation for such appearances. He misrepresents me, also, when he states that I deduce without restriction elevation from all fringing reefs & even from isld. without any reefs!! If his facts are true, it is very curious, that the atolls decrease in size, in approaching the vast open ocean S. of the Sandwich islands.— Dana puts me in a passion several times by disputing my conclusions, without condescending to allude to my reasons; thus regarding S. Lorenzo elevation, he is pleased to speak of my “characteristic accuracy” & then gives difficulties as if his own, when they are stated by me & I believe explained by me; whereas he only alludes to a few of the facts: so in Australian valleys he does not allude to my several reasons.— —

But I am forgetting myself & running on about what can only interest myself: He strikes me as a very clever fellow; I wish he was not quite so grand a generaliser.— I see little of interest except on Volcanic action & Denudation, & here & there scattered remarks: some of the later chapters are very bald.—

Farewell— | I have written a very long letter | Yours C. Darwin


In that chapter of Dana 1849a, James Dwight Dana produced evidence that contradicted Jean Baptiste Armand Louis Léonce élie de Beaumont’s theory that lava could not solidify on slopes of more than a few degrees of steepness. This was a crucial point of difference between the ‘craters of elevation’ theory and Lyell’s. Dana’s observation was added to the published version of Lyell’s paper on ‘Craters of denudation’ (C. Lyell 1850a, pp. 232–3).
Dana 1849a. See also letters to Charles Lyell, [2 September 1849], n. 5, and [1 November 1849].
William Lamb, Viscount Melbourne, is supposed to have said of Thomas Babington Macaulay: ‘I wish that I was as cocksure of anything, as Tom Macaulay is of everything’ (Wintle and Kenin eds. 1978).


Continues discussion of Dana’s Geology [1849]. Comments on dikes of Hawaiian volcanoes and Dana’s view of craters of denudation. Compares role of sea and rivers in forming valleys. Criticises Dana’s treatment of CD’s account of coral reefs.

Letter details

Letter no.
Charles Robert Darwin
Charles Lyell (1st baronet)
Sent from
Source of text
American Philosophical Society (88)
Physical description

Please cite as

Darwin Correspondence Project, “Letter no. 1277,” accessed on 18 December 2017,

Also published in The Correspondence of Charles Darwin, vol. 4