Avian herbivory: an experiment, a field test, and an allometric comparison with mammals.

Mechanistic studies on herbivore functional responses have largely taken place in mammals; very little has been done in herbivorous birds so far. Here we aim to fill that gap by experimentally quantifying the (short-term) functional response of a large avian herbivore, the Bewick's Swan (Cygnus columbianus bewickii). We explicitly distinguish between encounter-limited and handling-limited foraging by analyzing the results in the framework of the models of D. E. Spalinger and N. T. Hobbs, originally developed for mammalian herbivory. Bite size in captive swans was experimentally manipulated by varying sward height. The time interval between two bites increased with bite size, which supports the handling-limited model (process 3) and rejects the encounter-limited models (processes 1 and 2). Subsequently, we took the obtained functional response parameters into the field in order to predict, from measurements of sward height, (1) bite sizes, (2) handling times, and (3) short-term intake rates in free-ranging swans. Indeed, for all three variables, the observed values closely matched the experimentally based predictions. Finally, we review functional response parameters available in the literature on avian herbivores and scale them allometrically in relation to mammals. This analysis revealed that maximum bite sizes, and therefore maximum intake rates, in herbivorous birds are smaller than in herbivorous mammals. We hypothesize and provide evidence that birds compensate by longer daily foraging times.