Decoding dumping ducks



    1. Department of Natural Resources, Cornell University and Cornell Laboratory of Ornithology, 159 Sapsucker Woods Road, Ithaca, New York 14850, USA
    Search for more papers by this author

Janis L. Dickinson, Fax: 607-254-211; E-mail:


Conspecific brood parasitism, where females of the same species lay eggs in each other's nests, is common in waterfowl, and is usually considered costly to host females, which are stuck looking after eggs and chicks that are not their own. However, since female waterfowl often exhibit an unusual propensity to nest near where they were born, there has been some uncertainty over whether, in ducks and geese, laying in nests of conspecifics really is parasitism. Do parasitic and host females tend to be related? And is parasitism actually a form of cooperation in disguise? In a population in Hudson Bay, Andersson & Waldeck (this issue) found that ‘parasitic’ eggs in nests of the common eider, Somateria mollissima sedentaria, are more closely related to host eggs than expected by chance. In fact, host and ‘donor’ eggs are more closely related than are females breeding at neighbouring nests. The Hudson Bay population of common eiders is unusual, because unlike in more benign climates, females do not tend to breed near their natal nest. Spatial proximity alone cannot account for the high relatedness between host eggs and ‘dumped’ or donor eggs. Instead, the high relatedness values are probably the result of active recognition, where females favour kin, either when dumping or accepting eggs. These new data, along with evidence indicating that the donor lays the first egg in the nest nearly half the time, suggest that what appears to be parasitism in common eiders may be a form of kin-based cooperation.

At first, it seems obvious that a female common eider (Fig. 1) will gain by getting another female to look after her eggs, but when the host and donor are related, the evolutionary predictions, based on fitness outcomes, are not so simple. For a given degree of relatedness, it is important to consider both direct fitness, arising through an individual's own offspring production, and indirect fitness, arising through actions that influence offspring production by relatives. When close kin meet on the breeding grounds, their interactions can be either cooperative or competitive: (i) increasing average per capita reproductive success, as occurs with local resource enhancement (Gowaty 1993) and group augmentation (Kokko et al. 2001), or (ii) reducing average per capita reproductive success, as occurs with local resource competition (Clarke 1978). Because the net inclusive fitness benefits are interrelated, the resulting evolutionary outcome is complicated and, potentially, counterintuitive. The reason is: when the host and donor are relatives, each bears costs and benefits that are intricately tied to the fitness of the other.

Figure 1.

Incubating female common eider (Somateria mollissima). Photo credit: Malte Andersson.

Take, for example, a host female that receives two eggs from a sibling donor that would otherwise not reproduce. This host female could experience an indirect fitness gain by contributing an extra set of gene copies to future generations. Two nieces are equivalent to producing an extra offspring. Providing the host does not suffer an appreciable loss of direct fitness by accepting two extra eggs, the inclusive fitness balance sheet would show both donor and host in the black. Less intuitive is the possibility that donor females suffer a loss in inclusive fitness by laying eggs in the nests of relatives. For example, if there is a substantial negative impact of egg addition on the host's direct fitness, this will reduce the net inclusive fitness benefit for the donor, and her inclusive fitness tally could suddenly be in the red. Such a negative impact on host fitness would not matter for an unrelated female, who would benefit from parasitizing as long as her offspring's survival outweighed the current and future costs of producing and laying the eggs.

Only with the recent capacity to determine kinship in the wild have we been able to determine the relatedness of parasitic eggs to the females that are incubating them. Fingerprints from egg albumen proteins permit nondestructive assessment of relatedness early in the egg stage and this is critical for studies of egg dumping, because differential survival of host and donor eggs is of great interest. We do not yet have the information needed for rigorous tallies of host and donor fitness or even egg survival patterns in the common eider, and this information is critical to understanding the nature of the donor–host relationship and whether it is parasitic or cooperative. The possibilities are many. For example, extra eggs might be costly to incubate, but they may also increase host fitness due to the positive effects of group size on brood survival. To understand the nature of the interaction between egg donors and hosts, both the genetic relationships and the inclusive fitness effects of adding eggs must be fully understood.

Natal philopatry, where the young stick around and breed near where they were born, increases the likelihood of future interaction with close relatives, providing opportunities for kin to influence each other's fitness (Ekman et al. 2004). In most birds, the tendency to breed near home is restricted to one sex, typically males. In waterfowl, this pattern is reversed and females more commonly breed near their birth site. Because brood mates disperse as juveniles and do not remain together, as evidenced by the unusual lack of spatial genetic structure within this particular common eider population, simple proximity cannot explain the relatively high degree of relatedness between donor and host eggs (0.12–0.14, compared to 0.5 for a full-sib). While females must be somewhat philopatric for the eggs to be as closely related as they are, related females did not tend to nest near each other, so simple proximity is insufficient to explain Andersson & Waldeck's (2007) results. The evidence suggests an active preference in which donors prefer to lay in the nests of relatives and/or hosts preferentially accept eggs from kin.

Recent models indicate that when egg dumping is costly for the host, donors should avoid laying eggs in nests of relatives unless hosts also discriminate and avoid ejecting eggs of kin (Lopez-Sepulchre & Kokko 2002). This study provides surprising evidence of discriminatory ability, but it is not yet clear whether discrimination involves the donor, the host, or both. If common eiders possess the ability to discriminate kin without prior experience, this would not be entirely surprising. In mating aggregations known as leks, groups of males gather in what is often a winner-takes-all competition for copulations with females. In some lekking species, like the peacock, Pavo cristatus, males apparently recognize and associate with relatives even without prior exposure to related individuals (Petrie et al. 1999). This indicates that they use a self-referent phenotype matching mechanism. Does such a mechanism also operate in common eiders? Or does recognition occur mainly among individuals that were brood mates and thus have prior experience with each other? This has yet to be determined, but what is clear is that birds see greater genetic structure within their populations and neighbourhoods than is generally appreciated.

The most surprising result from this study is the finding that donor females often lay the first egg in the nest. Preliminary data indicate that when donor females are first to lay they tend to be closely related to hosts (r = 0.21 vs. 0.05, NS). This certainly bears further investigation. Is egg-laying sometimes a joint endeavour between sisters that culminates in abandonment by one female, typically the first one to lay? How is laying synchronized? And are there multiple strategies along the continuum from cooperation to parasitism? It is even possible that cooperation and competition take the form of a prisoner's dilemma where the superior strategy is to lay the first egg in the nest and abandon (Koenig et al. 1995). As often is the case with intriguing patterns divined from fingerprinting techniques, the most interesting results will ultimately come from coupling the genetic patterns with detailed information on behaviour and demography.

Janis Dickinson is a behavioural ecologist who has used long-term demography and molecular techniques to explore mating systems, benefits of helping, and behavioural decision-making in birds, primarily western bluebirds. She also directs the citizen science program at Cornell Lab of Ornithology, which addresses changing landscapes, bird distributions and abundance, demographics, and behavioural patterns at the continental scale.