The predatory behaviour of common kestrels facing various types of prey

Nine rehabilitated adult common kestrels (Falco tinnunculus) were tested for predatory behaviour. Tests were carried out individually on four different prey: two invertebrates (grasshopper and earthworm) and two vertebrates (lizard and laboratory mouse). The birds were offered prey randomly once daily, not necessarily on subsequent days. The latency to attack was similar. In contrast to predation tests with vertebrates, kestrels preying on invertebrates landed mostly (P < 0.01 for earthworm and P < 0.05 for grasshopper) a few centimetres from the prey itself, grasping it after a few steps. Earthworms were gripped almost invariably with the beak, but other prey with the toes. The birds struck the prey with their beak at highly variable rates (P < 0.001): vertebrates mostly with the beak, while earthworms were never struck. The latency to ingest prey varied greatly (P < 0.001) between prey. Captivity had only limited importance; it did not affect the behaviour sequence on any prey, but just induced kestrels in captivity for long time to attack mice with shorter latency and to strike them more with the beak. It is then demonstrated that the common kestrel has a strong behavioural plasticity even within the predatory sequence, which is then not stereotyped, and is able to modify its behaviour patterns in relation to different prey.