Huxley on species

I bought a fine copy of the Life and Letters of Thomas Henry Huxley in 3 volumes. On page 343f of volume I I find his thoughts on speciation, addressed to the Rev. Charles Kingsley (author of the Water Babies) on April 30, 1863.

My dear Kingsley

With respect to the sterility question [between species], I do not think there is much doubt as to the effect of breeding in and in [we would now use the neologism “inbreeding”] in destroying fertility. But the sterility which must be obtained by the selective breeder in order to convert his morphological species into physiological species – such as we have in nature – must be quite irrespective of breeding in and in.

There is no question of breeding in and in between a horse and an ass, and yet their produce is usually a sterile hybrid.

So if Carrier and Tumbler [varieties of pigeon], e.g., were physiological species equivalent to Horse and Ass, their progeny ought to be sterile or semi-sterile. So far as experience has gone, on the contrary, it is perfectly fertile – as fertile as the progeny of Carrier and Carrier or Tumbler and Tumbler.

From the first time that I wrote about Darwin’s book in the Times and the Westminster [Review] until now, it has been obvious to me that this is the weak point of Darwin’s doctrine. He has shown that selective breeding is a vera causa [Whewell’s term for a true cause, taken from Newton] for physiological species.

But I entertain little doubt that a carefully devised system of experimentation would produce physiological species by selection – only the feat has not been performed yet.

150 odd years later, it still has not. Most biologists would agree with Huxley that sterility is not the outcome of selection, but of drift and hitchhiking upon selection for ecological effects. [In animals, at any rate. Most of the time.]

7 thoughts on “Huxley on species

  1. E. Paterniani, Selection for Reproductive Isolation between Two Populations of Maize, Zea mays L., Evolution 23(4): 534-547 (1969).

    Verne Grant (Plant Speciation 181 (1981)) summarizes this by “In plants a comparable experiment has been carrried out by Paterniana (1969). Paterniani used two varieties of Zea mays differing in kernel colour. The white and yellow varieties were interplanted and allowed to cross-pollinate in an open-field. Artificial selection was practiced in favor of ears containing mainly non-hybrid kernels. In five generations of such selection, the level of spontaneous intervarietal hybridization decreased from 47% to 3% in the yellow variety and from 36% to %% in the white variety. The enhancement of reproductive isolation was due partly to greater differences between the varieties in flowering time, and partly due to the the development of prefertilisation incompatibility barriers.


  2. In an allopatric speciation event, there can be no selection for isolating mechanisms as such. So any isolating mechanism which happens to occur in isolation can become of selective importance only if and when the two putative species come into syntopic contact. One of the features, usually unspoken, of the biological species concept is the idea that individuals who hybridize are of lower fitness than individuals who do not hybridize. If so, then isolating mechanisms, premating or post mating, will be selected for and reinforced. If premating isolating mechanisms are sufficiently efficient, then there is no selection for postmating mechanisms.


  3. This is a really interesting letter, for many reasons – not least that Huxley is discussing the matter with his good friend, Charles Kingsley, who was – amongst much else – Chaplain to the Queen of England. What I found interesting here though, – rather than your own observation about present understandings about possible pathways for the evolution of sterility (interesting though it is), – is Huxley’s expectation of experimental proof of speciation. Darwin, of course, was – as Michael Ruse has long since pointed out, – strongly under the influence of the Cambridge philosopher of science, William Whewell, especially when it came to inductive reasoning and what made a good scientific explanation. In his 1840 work, The Philosophy of the Inductive Sciences, Whewell held to what he called the ‘concilience of inductions’ – ie: if the hypothesis expalined enough facts then it was likley to be near the mark, if it explained enough facts, and especially those one was not looking to make the hypothesis explain, then one could be really confident that it was a true explanation, or vera causa. – Thus, following Whewell, Darwin didn’t need to actually see speciation to be convinced that he was on the right track, – common ancestry explained so much from so many diverse fileds of study, and speciation was a logical outcome of this view in light of the present day diversity of life. Malthus just gave him the mechanism – competition and extinction. The issue of sterility was less of an immediate issue for Darwin, even though he did acknowledge the problem.

    Huxley, on the other hand, was much more of the empiricist school of John Stuart Mill, who wrote in the philosophy of science as well as poltical economy. The most convincing – and indeed for Huxley – the necessary evidence had to be observed, and preferably observed under experimental conditions. This was always going to be a problem for him, especially where speciation was concerned – which left Darwin perplexed. Darwin realised that the geological time involved and the selective nature of the fossil record made it unlikley that speciation would ever be demonstrated in such a way as to satisfy Huxley’s criteria.

    Kingsley, of course, had written to Darwin on the eve of the publication of Origin to say that he was ready to be convinced. Following Darwin’s analogy from domestic animals, he wrote to say that he could readily follow Darwin’s line of reasoning from long familiarity with the breeding of dogs and horses – he also added, almost as an after thought that a God who made things make themseleves was a much more noble conception of the deity than a God who had to interfere all the time. It was Kingsley’s expression of this last sentiment that Darwin included in the second edition of Origin, suggesting that on such authority there were no grounds for thinking that the views he put forward in the book should offend the religious feelings of anyone.


    1. One of the things I prefer about the 19thC discussion is the distinction between morphological and physiological species. Everybody knew that appearances were deceiving (Huxley’s discussion of Darwin’s favourite study organisms, the pigeons, indicates this clearly). It was the vera causae that mattered (a distinction that I reckon goes back toLocke’s Real and Nominal Essence distinction, very influential amongst British – and French – biologists).


  4. A comment regarding ‘species’ …

    John S. Wilkins

    Hull was also an early adopter of Richard Dawkins?’ views on evolution, enthusiastically taking up the notion of a replicator, an entity that exactly or very nearly so copies its structure,…

    Von Neumann considered this problem in the ’40s with his universal constructor

    Regarding heuristics of antiscience, subjectivity and an unspeakable reconstruction of ‘space time’ geometry, you might wish to consider the scope of Von Neumann’s , mathematical contributions.

    No I don’t understand it either.


Leave a Reply