Charles Darwin was fascinated by bees and followed precisely this path. ‘He must be a dull man who can examine the exquisite structure of a comb, so beautifully adapted to its end, without enthusiastic admiration’, he wrote in On the Origin of Species. I enjoy the directness of Victorian writing; if your mind isn’t inquisitive, you are a dullard. In the same seminal work, Darwin describes a series of experiments he conducted in order to understand the cell-making instincts of the hive bee.
‘… it seems at first quite inconceivable how they can make all the necessary angles and planes, or even perceive when they are correctly made. But the difficulty is not nearly so great as it first appears: all this beautiful work can be shown, I think, to follow from a few very simple instincts.’
To identify these simple instincts, Darwin compared the hive-making behaviours of the honeybee with a less architecturally accomplished species of bee, the Mexican Melipona domestica. The Melipona bees construct regular combs of cylindrical cells which Darwin asserted to be a simpler geometrical form, intermediate between no structure at all and the hexagons of the honeybees. ‘We may safely conclude that if we could slightly modify the instincts already possessed by the Melipona, this bee would make a structure as wonderfully perfect as that of the hive bee.’
To test the hypothesis, Darwin conducted a series of experiments in collaboration with his friend and fellow naturalist William Bernhardt Tegetmeier. They added different-coloured dyes to the beeswax, enabling them to create a visual record of the construction process, and were able to conclude that the bees first build cylindrical cells that are subsequently modified to form hexagons. Darwin was able to describe this in terms of natural selection:
‘Thus, as I believe, the most wonderful of all known instincts, that of the hive-bee, can be explained by natural selection having taken advantage of numerous, successive, slight modifications of simpler instincts; natural selection having by slow degrees, more and more perfectly, led the bees to sweep equal spheres at a given distance from each other in a double layer, and to build up and excavate the wax along the planes of intersection. The bees, of course, no more knowing that they swept their spheres at one particular distance from each other, than they know what are the several angles of the hexagonal prisms and of the basal rhombic plates. The motive power of the process of natural selection having been economy of wax; that individual swarm which wasted least honey in the secretion of wax, having succeeded best, and having transmitted by inheritance its newly acquired economical instinct to new swarms, which in their turn will have had the best chance of succeeding in the struggle for existence.’
Darwin concluded that bees build hexagonal honeycombs because they are the most economical way of dividing up their honey storage area. Hexagons use less wax, and the bees that use less wax are more efficient and more likely to survive and pass on their inherited behaviour to the next generation. This makes sense, because building a wax hive is extremely honey-intensive; for every gram of wax a bee produces it has to consume up to eight grams of honey. There is clearly an impetus to build efficiently, since using as little wax as possible maximises the honey available for food – an advantage that will have shaped the behaviour of honeybees over generations.
Is this correct? It’s certainly plausible. If bees used cylinders to build their honeycomb there would be gaps between each cell and the whole structure would be less efficient. Similarly, pentagons and octagons also produce gaps and so cannot be optimal. It is possible to imagine that each cell could be constructed in a bespoke shape by each bee to fit perfectly with its neighbour. In this ‘custom-made’ scenario each cell would be a different shape, but the gaps in the honeycomb could still be minimised. A problem with this strategy might be that one bee has to finish before the next bee can create a cell to fit. That’s an inefficient use of time. A repeatable single shape that leaves no gaps would seem to be preferred. The square, the triangle and the hexagon are the only regular geometrical figures that can fit together in a plane without leaving gaps.2
But why do bees use hexagons? Sometime around 36 BC, the Roman scholar Marcus Terentius Varro wrote down the earliest-known description of the honeycomb conjecture. This states that the most efficient way to divide a surface into regions of equal area (cells) with the least total perimeter (wax) is to use a regular hexagonal grid or honeycomb. No proof was offered, and the assertion remained conjecture for the next 2000 years until, in 1999, a mathematician at the University of Michigan named Thomas Hales found a proof: a hexagonal pattern is the most efficient engineering design. Natural selection, selecting for efficiency and creating structures that are a shadow of an elegant underlying mathematical law. What a beautiful answer to a simple question.
‘BEES, THEN, KNOW JUST THIS FACT WHICH IS USEFUL TO THEM – THAT THE HEXAGON IS GREATER THAN THE SQUARE AND THE TRIANGLE AND WILL HOLD MORE HONEY FOR THE SAME EXPENDITURE OF MATERIAL IN CONSTRUCTING EACH.’
— PAPAS OF ALEXANDRIA, AD 340
Well … possibly, but there may be more to it. In 2013, three engineers – Karihaloo, Zhang and Wang – published an article entitled ‘Honeybee combs: how the circular cells transform into rounded hexagons’. The claim is that honeybees, just like the Melipona bees that Darwin dismissed as inferior architects, make cells that are initially circular in cross section. The hexagons appear because the bees’ body heat softens the wax until it reaches 45 degrees Celsius, a temperature at which wax begins to flow like a viscous fluid. The circular cells of molten wax then act in a similar way to soap bubbles, joining together at an angle of 120 degrees wherever they meet. If all the bubbles or wax cells are identical in size and spacing, the circular cells spontaneously reform into a sheet of hexagons. Karihaloo and his team demonstrated this by using smoke to interrupt honeybees in the process of making a hive, revealing that the most recently built cells were circular, whilst the older ones had developed into hexagons. This transition from cylindrical to hexagonal structure appears to be what Darwin observed, but the explanation for the transition is different.
Natural selection is still the basic explanation for the hexagons, but the bees don’t have to go to the trouble of building the most efficient packing shape because physics will do that for them, given a nice sheet of circular cells of similar size and spacing and some body heat. To me, this is even more elegant and efficient; the bees allow physics to finish their work! As the authors of the study write: ‘We cannot … ignore, nor can we not marvel at the role played by the bees in this process by heating, kneading and thinning the wax exactly where needed.’ Is this the solution to a problem that has fired the imagination of so many for so long? The origin of the hexagons continues to be debated, and Karihaloo et al. will probably not be the last word in the literature.
This is as it should be, and illustrative of something that is often missed in the presentation of science. Scientific results are always preliminary. No good scientist will believe that they have offered the last word on a given subject. A result is published if the authors and a group of their peers consider it to be a valuable contribution to the field. Crucially, this does not mean that it’s correct; it means that it’s not obviously wrong. Rather than closing down a question, publication is intended to be a red flag to bullish colleagues. As one reads in Kepler’s partial, yet evident, delight in not discovering a satisfactory explanation for the structure of a snowflake, there is joy in hearing what you think, my most ingenious colleague.
Knocking on the doors of chemistry
In the final lines of The Six-Cornered Snowflake, Kepler writes with lovely regret