Whether species whose markings mimic those of less palatable species are mutualistic or parasitic has divided theoreticians and field ecologists for more than 100 years. Now a group of experimentalists has gone some way to settling the debate (see page 64).

Two types of mimicry exist in nature. Müllerian mimics are unpalatable species that evolve to resemble another unpalatable, model species. Such mimics share the cost of educating predators and so, the thinking goes, have a mutualistic relationship with their model. But batesian mimics — which are completely palatable and edible — receive all the benefit of their model's protection without making any contribution. These mimics are therefore viewed as having a parasitic relationship with their model species.

Debate has centred on situations in which a müllerian mimic is unequally defended — that is, its chemical defence renders it only slightly unpalatable compared with its highly unpalatable model. Theoreticians, including Michael Speed of the University of Liverpool, UK, proposed that such a situation would be quasi-batesian and would degrade the model species' protection, because a predator would take longer to learn that the markings meant 'not tasty'. But field workers believed that any level of defence in a müllerian mimic would be mutualistic.

Speed and his doctoral student, Hannah Rowland, took advantage of a 'novel world' set-up devised by their collaborators, Johanna Mappes and Leena Lindström at the University of Jyväskylä in Finland. Wild-caught great tits (Parus major) were pre-trained to open a paper-wrapped almond piece as a new type of prey. Then, in the novel-world aviary, Rowland and her Finland-based co-author Eira Ihalainen presented naive birds with several scenarios that represented batesian and differently defended müllerian mimics of a very unpalatable model prey — a piece of almond soaked in the bitter-tasting antimalarial drug chloroquine.

The students could test various factors individually, such as a mimic's level of defence (none, moderate or high), the quality of the mimicry (perfect or imperfect) and the number of mimics in relation to the model prey. In addition to the model and mimic prey items, the birds could choose a completely palatable prey item camouflaged against the background of the aviary.

In the unequally defended müllerian situation, Rowland and her colleagues observed an overall mutualistic relationship. In other words, Rowland says, “everyone's mortality — both the mimic and the model — gained from association with the model's markings”. Under these conditions there was no quasi-batesian parasitic relationship.

The authors found that introducing a batesian mimic to the system didn't increase the model's mortality. In fact, the model's mortality stayed flat with increasing numbers of edible mimics. This indicates that the dilution effect of adding more total prey to the system overwhelmed any parasitic relationship from the batesian mimic.

“In a forest, you could have a situation where a highly defended butterfly has a batesian mimic migrate into the area, and although it might affect the community's risk of predation, it wouldn't have a massive impact on the species survival,” says Rowland. She thinks this work “resolves a certain amount of the debate”. She adds that the novel-world aviary provides an opportunity to put other theoretical models of ecological interactions to the test.