Differences in habitat selection, diet and behaviour of resident and wintering Red Kites Milvus milvus were studied in Doiiana National Park, southwest Spain. Adult resident Red Kites roosted at their nests, while immature residents and wintering birds gathered at communal roosts. Individuals remained on average for two consecutive nights (range 1-8) at the same communal roost. Wintering kites spent significantly more time foraging gregariously than residents. The use of the marsh was greater by wintering kites while residents more often used the forest. Wintering birds consumed more goose carrion than the residents. Wintering kites had larger core areas, moved farther from their roost sites to feed and changed foraging areas more frequently than residents. Among residents, adult females had the smallest home ranges (core area, distances travelled and time spent flying). We suggest that Red Kites wintering in Doiiana occupy the marsh because it provides high food availability, and is vacated by Black Kites M . migrans which exploit this profitable habitat during the breeding season.

T h e behaviour of five broods of radio-tagged Spanish Imperial Eagles Aquila helium udalberti was studicd in Dofiana National Park, Spain during the post-fledging period. T h e distance between perching sites and nest, the mean flight duration and distance, the percentage o f time spent flying and the home range all increased exponentially with age. As the young got older, the parents spent less time in their vicinity. Young were not seen hunting, but depended on their parents for food. They begged and chased their parents throughout the post-fledging period, with hrgher intensity at the end. Nevertheless, the adults became progressively more reluctant to feed them, as reflected in the dccrease in feeding frequency and in the number of approaching flights towards the young. At the end of the post-fledging period, adults often performed aerial displays and frequently chased their offspring. T h e age of independence of the different young studied varied between 123 and 145 days. T h e correlations between individual independence and the dates when the young were last fed by their parents, and when the highest intensity of parental aggressive behaviour occurred, were higher than correlations with the variables related to the maturation of flying. Therefore, it is suggested that parental 'meanness' and aggressive behaviour may be the factors determining the date of juvenile independence and dispersal from the home territory.

Social feeding strategies of wintering red kites are analyzed in relation to age, food, roost‐sites and differences from kite residents. Whereas young and adult wintering kites gathered at roost sites almost daily, adult residents did not, and immature residents only occasionally. Kites using roost sites feed more often on prey prelocated by others, while lone roosters also forage and discover food alone. After finding food, kites tend to shift to a new roost site and foraging area. Two details of the ‘information centre’ hypothesis are confirmed in our study: carcasses are unpredictably found patches, divisible between several individuals. But carcasses disappeared fast in the study area, and no increase with time in the number of birds consuming a carcass was observed, so that information transmission was unconfirmed. When kites leave the roost in groups no leader is detectable. It seems that other types of social foraging are operating, and the model best matching our results is network foraging.