News & Views | Published:

Evolutionary biology

To work or not to work

Naturevolume 444pages4243 (2006) | Download Citation


Coercion, not kinship, often determines who acts altruistically in an insect colony. But underlying affinities for kin emerge when coercion is removed: kin selection is what turns suppressed individuals into altruists.

When Shakespeare's Prince Hamlet remarked that his uncle Claudius was “A little more than kin...” he was referring to the added relationship of stepfather that came after Claudius killed Hamlet's father and married his mother. “A little more than kin” could also describe a feature of many social insects. Owing to their odd genetic system, called haplodiploidy, full sisters in ants, bees and wasps are related by 3/4, more than a little above the standard value of 1/2. Historically, this extra kinship figured prominently in the acceptance of W. D. Hamilton's theory of kin selection1, which holds that workers evolved to altruistically forgo reproduction because they can pass on more of their genes by raising siblings. But a series of papers2,3,4,5, including one on page 50 of this issue3, shows that the other half of Hamlet's description — “...and less than kind” — may be more apt. Workers are less than kind both because they must be coerced into their 'altruistic' roles and because workers are often also the ones doing the coercing.

Of course, kin selection is not just about relatedness; a little more or less kinship can matter less than larger differences in the costs and benefits of altruism6,7. A study of a Malaysian hover wasp by Field et al.2 provides an elegant demonstration of this point. Some workers work harder than others, and relatedness to the queen does not explain the difference. In this species, rank comes with age. The second-oldest female is the heir apparent, and she reduces her risky foraging. Field and colleagues' experiments showed that this is causal. Removing the second-ranked female causes the third-ranked female to reduce her foraging in line with her improved prospects. Similarly, the amount of prospective gain also mattered; experimentally reducing colony size (adults and brood) caused second-ranked females to increase altruistic foraging, consistent with the diminished value of inheriting the queenship.

Thus, workers slacken off when they may become queens and work hard when that path is blocked. In social insects that, like the hover wasp, have small colonies and morphologically similar queens and workers, it is usually the queen that does the blocking by her dominance behaviour. For social insects with larger colonies, queen dominance is often replaced by other forms of control. First, there is nutritional coercion. Poorly fed females become small workers and well-fed ones become large queens. This limits the ability of workers to reproduce, but in most species it does not eliminate it fully. Given an opportunity, workers often will lay eggs. In a large colony, the queen could not successfully police all such behaviour and often ignores it. Instead, other workers do the policing, destroying the eggs of their co-workers8.

A comparative study of ten policing species by Wenseleers and Ratnieks3 again shows that altruism is modulated more by constraints on worker reproduction than by relatedness. The species with the highest fraction of fully committed altruistic workers — those that do not lay eggs — tend to be those with lower relatedness, contrary to simple expectation. Instead, more workers are fully committed when policing is most effective, as measured by the fraction of worker-laid eggs eaten by either queens or other workers. Workers are not leaping at every opportunity to be altruistic; they are coerced into that role, often by their fellow workers.

Does this mean that Hamilton's kin-selection theory is dead? The answer is no. One could invoke the efforts social insects make to exclude non-relatives from colonies, or the astonishing sex ratios that reflect kin-selected preferences for sisters over brothers7. But let us stick with the issue of direct reproduction. For nine of the ten species studied3, there are data on worker reproduction in queenless colonies where there can be no policing in favour of the queen's eggs. Two things change as a result. First, freed from policing, a higher fraction of workers opt to lay eggs. But, remarkably, many individuals still help instead of reproducing, and the fraction that help is now positively correlated with relatedness. Relatedness does matter, and this must be the reason that coercion can induce workers to help.

Wenseleers and Ratnieks5 earlier found a similar result when nutritional coercion is absent. In honeybees and their relatives the stingless bees, most species use nutritional coercion to limit queen production to a few at a time. Few are needed, because new queens can reproduce only by usurping the mother queen, or by acquiring a colony on the rare occasions the colony splits into two. The exception is the stingless bee genus Melipona. Here, all brood cells are pre-provisioned equally, and then sealed. A developing larva can therefore choose for herself whether to develop as a queen, with a larger abdomen, or as a worker, with larger fore parts (Fig. 1). Many females choose to develop as queens, showing their preference in the absence of constraint. This can result in hundreds of surplus young queens in a colony, so a new level of control has evolved. The workers slaughter the excess queens, so that the nest evokes the climactic scene of Hamlet, with royal corpses littering the stage. This is a great waste, but shows that it is nutritional coercion that normally keeps queen numbers in check. Yet this case, too, reveals that coercion is not everything and relatedness is important. More than 75% of females still choose the altruistic worker role, and the proportion is higher in species with higher relatedness5.

Figure 1: Queen control.
Figure 1


A comb of the stingless bee Melipona with the cell caps removed to show female larvae. Each female receives the same food, and therefore can choose whether to develop as a worker (identifiable here as those with a large head and eyes) or as a queen (small head and eyes). So many females opt to become queens that the workers kill off the surplus3, a form of control that in this species replaces queen selection through nutrition.

Finally, it should be remembered that Hamilton's kinship theory is not just about altruism per se, but about how all traits of altruistic workers evolve. When honeybee ancestors first evolved sociality, the workers could not waggle dance to convey information to each other, or suicidally detach their stings to better repel enemies, or police each other. These features, and all specialized features of workers, had to evolve by kin selection, through their indirect effects on relatives who could pass on genes for these traits. For example, the surprising positive correlation between relatedness and worker laying3, which has been confirmed in a much larger comparative study4, is expected under policing. Low relatedness among workers favours workers policing each other. Thus, although policing keeps suppressed workers from fully expressing their kin-related interests, policing is itself kin selected.

Many social conflicts create winners and losers. But only kinship allows evolution to make creative use of the social losers, turning them into reproductive police, exquisite communicators and heroic defenders. When Hamlet suffered the slings and arrows of outrageous fortune, he debated putting an end to himself. Social insect workers do sometimes choose suicide but, because of kinship, this hamiltonian choice is profoundly different from the hamletian dilemma. The stinging honeybee worker commits suicide when her sting is torn out, but this saves her kin. She is not making an escape from outrageous fortune, but making the best of it — not fearful of what dreams may come, but hopeful for what genes may come. However socially constrained her life may have been, her last action makes her own clear statement: long live the kin!


  1. 1

    Hamilton, W. D. J. Theor. Biol. 7, 1–52 (1964).

  2. 2

    Field, J., Cronin, A. & Bridge, C. Nature 441, 214–217 (2006).

  3. 3

    Wenseleers, T. & Ratnieks, F. L. W. Nature 444, 50 (2006).

  4. 4

    Wenseleers, T. & Ratnieks, F. L. W. Am. Nat.

  5. 5

    Wenseleers, T. & Ratnieks, F. L. W. Proc. R. Soc. Lond. B 271, S310–S312 (2004).

  6. 6

    West-Eberhard, M. J. Q. Rev. Biol. 50, 1–33 (1975).

  7. 7

    Queller, D. C. & Strassmann, J. E. BioScience 48, 165–175 (1998).

  8. 8

    Ratnieks, F. L. W. Am. Nat. 132, 217–236 (1988).

Download references

Author information


  1. the Department of Ecology and Evolutionary Biology, Rice University, PO Box 1892, Houston, 77251-1892, Texas, USA

    • David C. Queller


  1. Search for David C. Queller in:

About this article

Publication history


Issue Date


Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing