Measurement of chick provisioning in Antarctic Prions Pachyptila desolata using an automated weighing system

The frequency and mass of meals delivered to ten Antarctic Prion Pachyptila desolata chicks nesting at Bird Island, South Georgia were examined using automated weighing systems, placed in the nest chamber of the burrow. The w eighing systems, which consisted of an artificial nest platform mounted on electronic load cells linked to a data logger, recorded the mass of the chi ck every 10 min. Meal delivery was concentrated in the first few hours afte r dark with a peak (45% of meals) in the second hour of darkness. The mean number of meals per night per chick ranged from 0.4 to 1.7 with 80% of chic ks receiving at least one meal per night; there were no nights when none of the chicks was fed. Chick 9, which had the highest feeding frequency, rece ived smaller meals (mean 18 g) than all of the other chicks (mean 37 g). Th e mass of first and second meals delivered in the same night was not differ ent nor was there a relationship between meal mass and feeding frequency (e xcept for chick 9). There was no relationship between meal mass and chick a ge between 5 and 50 days. Meals delivered following a night when a chick re ceived no meal were larger than when a chick had been fed the previous nigh t. The proportion of chicks receiving 0, 1 and 2 meals per night differed f rom the predicted binomial distribution, assuming the independence of feedi ng by both parents, with more chicks receiving one meal per night than expe cted. Comparison with periodic weighing methods indicates that the latter u nderestimates meal size and feeding frequency. The variation within the sma ll sample of chicks studied indicates the need for large sample sizes to es timate population provisioning parameters. However, the weighing system was very successful in providing accurate data on chick provisioning and, with additional development, could provide data on the individual provisioning strategies of adult birds.