SHAPE OF THE INTERFERENCE FUNCTION IN A FORAGING VERTEBRATE

1. We provide evidence that interference between overwintering oysterc atchers Haematopus ostralegus feeding on mussels Mytilus edulis is abs ent or has only a negligible effect on intake rate at low competitor d ensities, and only reduces intake rate above a threshold bird density. The threshold for interference did not differ significantly between i ndividual birds and showed no association with their local dominance. The threshold value was c. 50 birds ha(-1) for birds that opened prey by hammering and 150 birds ha(-1) for those that stabbed into mussels. 2. The interference-free intake rate at densities below the threshold for interference differed significantly between individual birds but was not associated with local dominance; nor did it vary either during the tidal exposure period or between autumn and winter. 3. At competi tor densities above the threshold for interference, birds of lower loc al dominance were more susceptible to interference than those of highe r dominance, the magnitude of the difference being smaller in hammerin g birds than in stabbers. For birds of a given local dominance, interf erence was most intense towards the end of winter and, in hammerers, a t the beginning of the exposure period. 4. We conclude that theoretica l models in which interference operates as soon as competitor density starts to increase provide a poor description of this system. Rather, our data support those models in which interference begins only after density has reached a particular, and constant, level.