skip to content

Darwin Correspondence Project


From Jeffries Wyman   11 January 1866

Cambridge [Massachusetts]

Jan. 11th. 1866

Dear Sir

Having lately given some attention to the study of the cells of the bee I have been struck with the fact that the deviations from the form alledged to have been measured by Maraldi & which was calculated by Koenig, are much greater than is mentioned in any of the descriptions I have seen.1 The conclusion to which I have been led, is, that whatever the typical form of the cell be, it is rarely, if ever, realised.2 As you have given your own careful attention to this subject, I feel that I run the risk of writing what may be already better known to you than to myself.3 Then it occurs to me that cumulative evidence is worth something. At all events I will state what I have to say briefly.

1st After repeated measurments of hundreds of worker cells I find, as Reaumer did in the drone cells, that the diameters are very frequently variable, so that nine cells measured in the direction of one diameter will sometimes fill a space, which it will require ten to fill if measured in the direction of another.4 2d. The terminal rhombs are the most variable parts of the cell. I have shown this very satisfactorily by means of plaster casts, & even by the coccoons from old brood cells which give a very good model of the base— Sometimes the rhombs are all of different sizes, & at others one is nearly eliminated, the other two making nearly the whole of the partition between two adjoining cells. 3d. The introduction of the fourth side to the base does not always, or even generally depend upon the transition from worker to drone cells, as stated by Huber,5 but either upon an incorrect allinement of the cells on opposite sides of the comb, or upon an increase of the size of the cells on one side not corresponding with that on the other. 4th the most remarkable deviation which I have noticed is a sort of rotation as it were of the cells, so that those of the two sides, differed in position about 30o.


Fig 1 being the normal position, 2 & 3 are the abnormal ones— in 2 the angle of one cell corresponds with the centre of the other, but instead of pointing to one of the angles of the other cell it points to one of the sides. In 3 the cells were fitted together by their whole bases, &, projected on to the same plane, are concentric. In both of these cases the rhombic faces became impracticable, & the bees compromised on a flat bottom to the cell— I found this arrangement in a large part of a worker brood-comb, & in the whole of a piece of drone comb. 5th. Lastly there is no regular mode of transition from worker to drone cells, but this may be effected in several ways, viz by means of irregular polygonal cells, by a gradual change of size, so that no line of separation is noticeable between the two kinds; & by the close approximation of worker & drone cells making an abrupt change; but as the worker cells are one fifth smaller than the drone cells, & therefore fail to fill up the space, a new row is intercalated for this purpose.


These intercalated rows however do come at regular intervals, & the emergency is often met by thickened walls.

I have studied the cells with reference to the hexagonal shape, & it seems to me that the view taken by Mr Waterhouse & yourself is the most tenable one.6 Fearing that you may not have seen the work, I take the liberty of calling your attention to the so called “fossil tadpole nests”, in Prof Hitchcocks report on the Ichnology of Massachusetts. Have you not there exactly the phenomenon in question? A lot of tadpoles get together each wiggling about on a centre, as is is necessary in a crowd, each excavates a place in the mud, but their respective cavities soon come in collision & a flat wall between them results, & in many cases a hexagonal cell almost as true as that of the bee—7

One thing seems entirely clear, that the hexagonal cell results from the cooperation of several bees & is never made by a single bee, & if he have no antagonist in the construction of a wall it is curvilinear.8 In some of my wasps nests from S America it appears that even in working with paste board, the outer walls of the new cells are always curved at first, but when other cells are made outside of them they are flattened so as to conform to those that have preceded them.

Before I close allow me to mention the observations of Mr Putnam on the cells of our humble bees—which he has shown are at first a mass of pollen & honey, on which an egg is placed, & into which the larva eats as soon as hatched.9 The cavity thus formed is the beginning of the cell.— This is increased by additions of pollen, which is gradually consumed; eventually wax is applied to the exterior in a very thin layer.10 I requested Mr Putnam to send you a copy of his paper, which perhaps you will have received before this reaches you.

I fear I have trespassed upon your time, but hope you will excuse me on the score of good intention

With great respect | Truly yours | J. Wyman

Charles Darwin, Esq. | Downe Bromley Kent.

CD annotations

1.4 The conclusion … realised. 1.6] scored pencil
3.3 attention … Massachusetts. 3.5] scored pencil
4.1 One … single bee, 4.2] double scored pencil
5.1 Before … bees— 5.2] double scored pencil
On cover: ‘Size [above del ‘Thickness’] of Cells & irregularities—good—Wyman’ ink


Giacomo Filippo Maraldi’s geometrical measurements of hive-bee (honey-bee) cells (Maraldi 1712, pp. 304–9) and Johann Samuel Koenig’s calculations (Koenig 1739) were cited in Réaumur 1734–42, 5: 382, 388–9. The base of each cell, a hexagonal prism, is composed of three parallelograms (rhombs), for which Koenig calculated the internal angles required to achieve the most efficient relationship between volume and the amount of wax required to build the cell. As Koenig’s theoretical figures corresponded closely with Maraldi’s empirical measurements (Réaumur 1734–42, 5: 388–90), the notion arose that the dimensions of hive-bee cells were more or less standard.
Wyman’s conclusion is cited in Origin 4th ed., pp. 271–2: ‘I hear from Prof. Wyman, who has made numerous careful measurements, that the accuracy of the workmanship of the bee has been greatly exaggerated; so much so, that, as he adds, whatever the typical form of the cell may be, it is rarely, if ever, realised.’
Differences in the shapes of the cells made by various kinds of bees presented a potential problem for natural selection (see, for example, Correspondence vol. 7, and Correspondence vol. 11; see also Origin, p. 224). The issue continued to interest CD after the publication of Origin (see, for example, Correspondence vol. 8, letter to W. H. Miller 1 December [1860], and Correspondence vol. 12, letter to A. R. Wallace, 1 January 1864 and n. 2). CD’s notes on bees’ cells are in DAR 48B. For a detailed examination of CD’s effort to demonstrate that the complex geometry of the cell could be achieved through the gradual evolution of complex instinct, see Prete 1990.
Brood combs include cells of different sizes. The smallest and most abundant cells of hive-bee combs are occupied by the larvae of workers, while the larvae of drones occupy larger cells (Stanek 1969, p. 177, and Seeley 1995, pp. 177–8). In his measurements of hive-bee cells, René Antoine Ferchault de Réaumur recorded greater variability among the diameters of drone cells than among those of worker cells (Réaumur 1734–42, 5: 397–400).
François Huber contended that the transition from worker cells to drone cells required that the pyramidal base of each cell be made up of four rhombs rather than three, as found typically in the worker cell (see F. Huber 1814, 2: 222–4). In his annotated copy of F. Huber 1814 in the Darwin Library–CUL, CD scored parts of this passage (see Marginalia 1: 412). Further detail on structural aspects of the transition from worker to drone cells is given in the appendix to F. Huber 1814, written by his son, Jean Pierre Huber (ibid., 2: 231–51).
In contrast to François Huber, who held that the bee cell was the product of the bee’s ability to construct geometrical forms, George Robert Waterhouse considered that the hexagonal cell of the hive-bee, and the rhombs of its base, originated in the form of a simple cylinder with a curved base. In his theoretical model, a number of pliable wax cylinders when packed tightly together under even pressure became hexagonal ([Waterhouse] 1835, pp. 153–5). In Origin, p. 225, CD wrote: ‘I was led to investigate this subject by Mr. Waterhouse, who has shown that the form of the cell stands in close relation to the presence of adjoining cells; and the following view may, perhaps, be considered only as a modification of his theory.’ CD went on to propose an underlying instinct by which bees build cell walls according to a scheme of intersecting circles, adding that the hive-bee’s hexagonal cells resulted from the natural selection of successive slight modifications in instinct that conferred increasing economy in the use of wax (Origin, p. 235). See also Correspondence vol. 7, letter from G. R. Waterhouse, 13 February 1858, and letter to W. B. Tegetmeier, 17 [October 1858] and nn. 3 and 4.
Edward Hitchcock established the science of ichnology (Proceedings of the American Philosophical Society 9 (1864): 445). Hitchcock observed both mud and fossils containing cavities that were hexagonal in form like bee cells; he concluded that in both cases the cavities had been formed by tadpoles (Hitchcock 1858, pp. 121–3 and plates 21 and 50). See also Correspondence vol. 13, letter from Edward Cresy, 10 September 1865 and n. 11.
In preparing the fourth edition of Origin during the spring of 1866 (see ‘Journal’, Appendix II), CD deleted a sentence in which he had cautiously entertained the possibility that hive-bees construct each cell as ‘an isolated hexagon’ (see Origin, p. 233, and Peckham ed. 1959, p. 410).
The reference is to Frederic Ward Putnam and Putnam 1863a, an annotated copy of which is in the Darwin Pamphlet Collection–CUL.
CD’s annotations on Putnam’s paper included the scoring of passages on the formation by adult humble-bees (Bombus) of a thin wax wall around the pupa, and Putnam’s rejection of the view that eggs were laid in pre-formed cells (Putnam 1863a, pp. 101–4).

Letter details

Letter no.
Wyman, Jeffries
Darwin, C. R.
Sent from
Cambridge, Mass.
Source of text
DAR 181: 191
Physical description


Has made observations on bees’ cells. Their dimensions are not constant, nor do single bees make single cells; each one is a result of co-operation.

Please cite as

Darwin Correspondence Project, “Letter no. 4974,” accessed on 13 February 2016,