Constraints on host choice: why do parasitic birds rarely exploit some common potential hosts?

Abstract

1. Why are some common and apparently suitable resources avoided by potential users? This interesting ecological and evolutionary conundrum is vividly illustrated by obligate brood parasites. Parasitic birds lay their eggs into nests of a wide range of host species, including many rare ones, but do not parasitize some commonly co-occurring potential hosts. 2. Attempts to explain the absence of parasitism in common potential hosts are limited and typically focused on single-factor explanations while ignoring other potential factors. We tested why thrushes Turdus spp. are extremely rarely parasitized by common cuckoos Cuculus canorus despite breeding commonly in sympatry and building the most conspicuous nests among forest-breeding passerines. 3. No single examined factor explained cuckoo avoidance of thrushes. Life-history traits of all six European thrush species and the 10 most frequently used cuckoo hosts in Europe were similar except body/egg size, nest design and nestling diet. 4. Experiments (n = 1211) in several populations across Europe showed that host defences at egg-laying and incubation stages did not account for the lack of cuckoo parasitism in thrushes. However, cross-fostering experiments disclosed that various factors during the nestling period prevent cuckoos from successfully parasitizing thrushes. Specifically, in some thrush species, the nest cup design forced cuckoo chicks to compete with host chicks with fatal consequences for the parasite. Other species were reluctant to care even for lone cuckoo chicks. 5. Importantly, in an apparently phylogenetically homogenous group of hosts, there were interspecific differences in factors responsible for the absence of cuckoo parasitism. 6. This study highlights the importance of considering multiple potential factors and their interactions for understanding absence of parasitism in potential hosts of parasitic birds. In the present study, comparative and experimental procedures are integrated, which represent a novel approach that should prove useful for the understanding of interspecific ecological relationships in general.