Abstract
(1) Starlings that were feeding nestlings collected mealworms from an artificial feeder where the cumulative number of prey was a logarithmic function of patch residence time (Tp). The patch was placed at various locations between 8 m and 600 m away from the nests. The number of prey swallowed, Tp, and number delivered to the young was recorded in 1364 round trips. (2) Four models of optimal Tp are discussed: (a) Maximize YIELD: energy collected per total round trip time (including time in the nest); (b) Maximize rate of DELIVERY: YIELD minus parent's metabolism; (c) Maximize FAMILY GAIN: DELIVERY minus brood's metabolism, considering the energy cost of begging by the chicks; (d) Maximize EFFICIENCY :YIELD/parent's metabolism. Relevant when there is a cost in spending energy (for example due to constraints in maximum rate of heat dissipation). (3) The results differ significantly from the YIELD and EFFICIENCY models. The FAMILY GAIN model gives the best fit to the data. (4) The variance in Tp reflects the cost of deviations from the optimum. The frequency distribution of observed Tp resembled the predictions of a descriptive model (PROFITABILITY MATCHING) based on proportional alocation of effort to various behavioural options depending on their relative pay-off. (5) Manipulations in brood size resulted in different allocation of food between parent and young but did not affect overall Tp.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
