skip to content

Darwin Correspondence Project

To G. R. Waterhouse   [26 July 1843]

Down, Bromley Kent


My dear Waterhouse

Now for a letter in answer to your two ones on classification—on which I have been often thinking. It has long appeared to me, that the root of the difficulty in settling such questions as yours,—whether number of species &c &c should enter as an element in settling the value or existence of a group—lies in our ignorance of what we are searching after in our natural classifications.—1 Linnæus confesses profound ignorance.—2 Most authors say it is an endeavour to discover the laws according to which the Creator has willed to produce organized beings— But what empty high-sounding sentences these are— it does not mean order in time of creation, nor propinquity to any one type, as man.— in fact it means just nothing—3 According to my opinion, (which I give every one leave to hoot at, like I should have, six years since, hooted at them, for holding like views) classification consists in grouping beings according to their actual relationship, ie their consanguinity, or descent from common stocks— In this view all relations of analogy &c &c &, consist of those resemblances between two forms, which they do not owe to having inherited it, from a common stock.—4 To me, of course, the difficulty of ascertaining true relationship ie a natural classification remains just the same, though I know what I am looking for.— This being the case viz ignorance of a distinct object I think, we ought to look at classification as a simple logical process, i.e. a means of conveying much information through single words— now the laws of classification tryed by this rule—it is clear that neither number of species—nor grade of organization ought to come in, as an element.— For instance, the word Marsupiata implies a great number of facts, common to many animals, and if by introducing another group of animals the definition of Marsupiata is forced to include, any one fact, not applicable to the former animals, then I maintain the second group ought not to have been added to the first.5 But if they have much in common, a third more general term ought to be invented to include them.— Viewing classification, as a logical process, the only cheque to the splitting of groups, appears to be convenience; & if you had objected to making the Monotremes & the Styloptera6 orders, on account of the inconvenience of increasing the number of orders I think, it might have been valid. I repeat, that until you can define your object in classifying, you have no right to introduce number of species, or (arbitary) grades of organization, but are bound to follow simple amount of differences of organization.— This, I think, you will find is the rule amongst Botanists, & they have many families very small in number.— This is clearly the rule, on the view of classification, being a genealogical process, exhibiting literal or actual relationship.

There is one caution, which should not be overlooked, namely the great doubt whether the groups, which are now small, may not have been at some former time abundant: and you will admit fossil & recent beings all come into one system.— In fish, it would appear, that some of the main divisions, which are now least abundant in species, appear formerly to have been most so. It wd take a Chapter to argue, how probable it is that Geology has never revealed & never will reveal, more than one out of a million forms, which have existed. Finally, however, I confess, there is much weight in your argument, that as most orders contain many species, the usual harmony of nature would make you very cautious in admitting a few species to form a group of equal rank with one, containing many species— It is certainly a very curious fact, the existence of only a few groups, apparently of high value & containing only a very few species.— The only doubt, which has ever occurred to me, lies in the rules being so exceedingly arbitary by which the value of groups are judged by. I mean, that perhaps many orders really exist, containing only one or two genera, but that from this very circumstance, they are not viewed as orders but only as families— Perhaps if the Goatsucker & Woodpecker, were varied into very many genera & very many species of each—they would be looked on as orders equal to the Hawks &c &c— (Tell me what you think of this) This is a mere illustration of what I mean.—

—Finally then I protest against number or grades of organization being used as elements in classification, though I believe they have silently been used.—

Have you had patience to read thus far.— Yours ever, C. D.


CD’s Notebooks B–E and other notes (in DAR 205) of the late 1830s and early 1840s contain many entries recording his effort to work out a theory of classification based on descent with modification. Ospovat 1981, pp. 101–14, provides a good account of CD’s efforts to reconcile his theory with what he considered sound in contemporary taxonomical practice.
Possibly a reference to Linnaeus’ statement in a letter to Paul Dietrich Giseke, quoted in Whewell 1837, 3: 322–3: ‘You ask me for the characters of the natural orders: I confess I cannot give them.’
CD makes the same point in the essay of 1844 (DAR 7: 133; Foundations, p. 202): ‘many naturalists have said that the natural system reveals the plan of the Creator: but without it be specified whether order in time or place, or what else is meant by the plan of the Creator, such expressions appear to me to leave the question exactly where it was.’
English naturalists distinguished between affinities, which an organism shares with other members of the same taxonomic group, and analogies, characteristics which other members of its group may not possess but which appear in members of some other group (Ospovat 1981, p. 105). Darwin here redefines the terms in an evolutionary sense.
Waterhouse was evidently proposing to include Monotremata as a subgroup of Marsupialia because only two species were known (see letter to G. R. Waterhouse, [31 July 1843]).
The order Strepsitera (Kirby), consisting of the single family Stylopidae, contained a few species of parasitic insects commonly regarded as ‘the most anomalous annulose animals with which we are acquainted’ (Westwood 1839–40, 2: 288). The order was one of William Sharp MacLeay’s ‘osculant’ groups and was placed by him between Hymenoptera and Coleoptera (MacLeay 1819–21, pp. 371–2).


Ospovat, Dov. 1981. The development of Darwin’s theory. Natural history, natural theology, and natural selection, 1838–1859. Cambridge: Cambridge University Press.

Westwood, John Obadiah. 1839–40. An introduction to the modern classification of insects; founded on the natural habits and corresponding organisation of the different families. 2 vols. London: Longman, Orme, Brown, Green, and Longman.

Whewell, William. 1837. History of the inductive sciences, from the earliest to the present times. 3 vols. London.


Classification consists of grouping beings according to descent from common stocks. Analogies are resemblances between forms not inherited from common stocks. Neither number of species nor grade of organisation should be considered in classification. Admits that caution is necessary in admitting a few species to form a group of rank equal to one containing many species.

Letter details

Letter no.
Charles Robert Darwin
George Robert Waterhouse
Sent from
Source of text
DAR 185: 68
Physical description
ALS 5pp

Please cite as

Darwin Correspondence Project, “Letter no. 684,” accessed on 15 July 2024,

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