Over the past decade, molecular methods have provided ample evidence that, in most apparently monogamous animal species, it is quite common for a couple to raise offspring that resulted from either the male or the female mating with another partner. This is particularly evident in birds. Yet explanations for why some female birds actively seek copulations with other males, or why males mate with other females and allow them to lay eggs in the male's nest, are speculative. Elsewhere in this issue, however, a group of ten biologists from six countries provide convincing evidence that, in some cases, the phenomenon can be explained by how closely related the male and female of a pair are (D. Blomqvist et al. Nature 419, 613–615; 2002).

The authors found that in three shorebird species, including the western sandpiper (Calidris mauri), pictured here, the birds of a pair that is rearing 'illegitimate' offspring are more closely related to each other than the average relatedness. So it seems plausible that they sought alternative mates to reduce the harmful effects of inbreeding. Relatives often carry the same versions (alleles) of various genes, so there is a strong chance that their offspring will inherit two identical alleles — one from each parent — of those genes. This matters for recessive alleles, which are expressed only if both copies are the same.

Increased expression of recessive alleles depends on the degree of relatedness of the parents and the frequency of the alleles in the population. The effect is greater for rarer alleles — and one would expect alleles with harmful effects to be both rare and recessive. This means that inbreeding is usually associated with increased expression of harmful alleles and a low survival rate of offspring, and therefore that mating with relatives should be avoided.

So there is a logic to Blomqvist and colleagues' finding that shorebirds that are, through some social constraint, paired to a relative then seek alternative partners. The strategy would increase the fitness of offspring and could be evolutionarily selected for. The big question now is, how do the birds know who their mate is? It is likely that many birds can recognize individuals that grew up in the same nest, but we have no idea how they recognize half-siblings from other nests, or cousins. In rodents, smell provides clues, but in most birds the sense of smell is comparatively underdeveloped.

However, Blomqvist et al.'s observation fits with other evidence that birds are cleverer and learn more from past experiences than previously thought. For instance, birds can assess the breeding success of other individuals from the same species, and use that information in deciding where to breed the following year (see, for example, B. Doligez et al. Science 297, 1168–1170; 2002). They can also adjust the time that they reproduce according to past experience (F. Grieco et al. Science 296, 136–138; 2002). All of these findings provide better insight into bird ecology, but also sound a cautionary note to researchers: the idea that birds react to, and compensate for, experimental treatments has become highly plausible. It has long been known that birds are clever and flexible in their behaviour. But we have apparently seriously underestimated them.