African Cuckoos

On egg size mimicry and competition between honeyguide females:

Spottiswoode, C.N. A brood parasite selects for its own eggs traits. Biology Letters 9: 20130573.

Greater Honeyguide eggs mimic host eggs in size (see also here) and this paper shows that, surprisingly, bee-eater hosts are undiscriminating and readily accept mismatched eggs. This study shows that instead, honeyguide egg size adaptation has probably rather evolved to trick other honeyguides, not host parents: honeyguides selectively puncture any mismatched egg already present in little bee-eater nests when they lay their own, lest it be the offspring of another honeyguide female and brutally kill their own chick when it hatches.

See also: News articles and a podcast about this research
An article by Fugo Takasu on F1000 Prime about this research (subscription only) 
Read the full paper on the journal website [Open Access]

On chick-killing by honeyguides:

Spottiswoode, C.N. & Koorevaar, J. 2012 A stab in the dark: chick killing by brood parasitic honeyguides. Biology Letters 8: 241-244.

This paper confirmed that honeyguides have a much darker side as unusually brutal brood parasites. Honeyguides hatch from the egg already equipped with a pair of needle-sharp hooks at the tips of their beaks, and it has long been inferred that they use these to kill the chicks of their hosts since maimed host young had been found alongside honeyguide chicks. However, this behaviour had never been observed under natural conditions in host nests. Infra-red footage [watch video] from cameras buried in the darkness of the bee-eaters’ underground nests revealed how honeyguide chicks repeatedly grasp, bite and shake chicks of their newly hatched foster siblings until they eventually die [see photos].

The killing behaviour is actually the culmination of a sequence of specialised adaptations that ensure that the young honeyguide has sole access to the food the host parents bring to the nest: the honeyguide mother ensures her chick hatches first by internally incubating the egg for an extra day before laying it, so it has a head start in development compared to the host, and she also punctures host eggs when she lays her own. But some host eggs are overlooked or survive puncturing, and it is these that precipitate chick killing by the young honeyguide as soon as they hatch. Because the honeyguide hatches first, it has grown to about three times the weight of a hatchling bee-eater by the time it sets about killing it. Just one to five minutes of active biting time was enough to inflict sufficient injuries to cause host death. However, after maimed chicks stopped moving honeyguides often ceased their attacks and, as a result, hosts sometimes took over seven hours to die. Host parents are apparently blithely unaware of what is happening and, in the darkness of their burrows, even attempted to feed a honeyguide chick busy attacking their own young. We also filmed one instance of the honeyguide biting its foster parent by accident. By the time the honeyguide emerges from the burrow after about a month of assiduous care by its foster parents, however, its bill hook has grown out and there is no trace of its siblicidal beginnings.

This behaviour is exactly analogous to that of young cuckoos, which hoist host eggs or chicks onto their backs and tip them over the rim of the nest. Because honeyguide hosts breed in tree holes or underground burrows, honeyguides cannot eject host chicks, and have instead evolved this highly effective killing behaviour to make sure that they alone monopolise the nest.

See also: News articles and a video podcast about this research
Read the full paper on the journal website [Open Access]

Watch one of our videos of honeyguides killing host chicks (see here for more examples from the Biology Letters website):

On ancient host specialisation by honeyguide females:

Spottiswoode, C.N., Stryjewski, K.F., Quader, S., Colebrook-Robjent, J.F.R. & Sorenson, M.D. 2011 Ancient host-specificity within a single species of brood parasitic bird. Proceedings of the National Academy of Sciences of the USA 108: 17738-17742.

A long-standing conundrum for cuckoo research has been how a single parasitic species can simultaneously mimic multiple host species. In this paper we begin by showing that Greater Honeyguides show host-specific specialisation in egg size and shape rather than egg colour or pattern, since hosts breed in dark holes where visual appearance may be irrelevant. For example, while Greater Honeyguide eggs are invariably white, they are small and round in the nests of Little Bee-eaters, but large and tapered in the nests of Green Woodhoopoes. Thus in appearance honeyguides appear to show specialised host-races just as do cuckoos and cuckoo finches.

However, the fact that a single species of brood parasitic bird can closely mimic several host species is evolutionarily puzzling. How can such specialised adaptations be maintained in the face of interbreeding among parasitic males and females of the same species, but specialised on different hosts? A possible solution is that the genes underpinning such egg adaptations are on the avian W chromosome that is carried only by females, allowing them to be passed on intact from mother to daughter without detrimental genetic mixing from fathers raised by other host species. For such a genetic mechanism to apply, however, lineages of parasitic females would need to remain highly host-faithful over evolutionary time.

In this paper we tracked lineages of Greater Honeyguide females using mitochondrial DNA as a genetic marker since, like the W chromosome, the mitochondrial genome is inherited only from mothers. This revealed two highly distinct female lineages of honeyguides, each associated with separate groups of hosts: those nesting in terrestrial burrows (mostly bee-eaters) and those nesting in tree cavities (hoopoes, woodhoopoes and others). The genetic divergence between the two strains was astonishingly deep, showing that each parasitic lineage has remained perfectly faithful to its specialist hosts for at least 3 million years. The depth of this divergence prompted us to question whether the Greater Honeyguide might in fact be in the process of splitting into two new species, each specialising on different hosts. We investigated this by analysing nuclear DNA, which is inherited from both parents and should therefore show divergence if the two female lineages are speciating. What we found, however, is that Greater Honeyguides’ nuclear DNA showed no host-specific association whatsoever, indicating that host-specialist females mate at random with any conspecific male, irrespective of what host males were raised by. This interbreeding keeps the Greater Honeyguide a single species comprising two remarkably ancient female host-races.

See also: A commentary article by Robert Fleischer in PNAS about this research.
Read the full paper on the journal website [Open Access]

On egg puncturing by female honeyguides, and host counter-adaptations:

Spottiswoode, C.N. & Colebrook-Robjent, J.F.R. 2007 Egg puncturing by the brood parasitic Greater Honeyguide and potential host counteradaptations. Behavioral Ecology 18: 792-799.

Although honeyguides stab any host chicks to death as soon as they hatch, thus monopolising the hosts’ parental care, the female Greater Honeyguide often reduces the likely competition by puncturing the hosts’ eggs when she lays her own in their nest. Using clutches and data collected by John Colebrook-Robject, we studied punctured eggs in parasitised clutches and found that those with more punctures were more likely to contain a dead embryo, confirming that puncturing does work. However, heavily punctured clutches were sometimes deserted, suggesting that conspicuous damage can alert hosts to the fact that they have been parasitised, and that honeyguides should strike a balance between puncturing too much and too little. Accordingly, honeyguides appeared to puncture clutches more heavily (hence taking a greater risk of desertion) when laying their egg late relative to the host, which is when they have most to gain by preventing host eggs from hatching sooner than their own. How do hosts defend themselves? Host eggs that were structurally stronger (thicker-shelled and rounder in shape) were more heavily punctured, suggesting they were harder to damage effectively, and hosts that were able to defeat the honeyguide and gain some breeding success in spite of being parasitised had thicker-shelled eggs than those that failed altogether. This suggests that natural selection favours thicker-shelled eggs, and accordingly we found that host species have indeed evolved disproportionately thick eggshells compared to closely related species that are not parasitised by honeyguides.

See also: Read the full paper on the journal website [Open Access]