I studied the seasonal variation in habitat selection, diet, food supply and food selection of the Priolo or Azores bullfinch, Pyrrhula murina. This species and its associated habitat, the Azorean native cloud forest, are poorly known and are threatened by afforestation and the invasion of exotic flora. In addition detailed studies of these aspects are lacking for oceanic island birds. Throughout this study a comparison between the ecology of the Priolo and that of mainland Bullfinches is made. The primary aims of this study were (1) to evaluate the importance of three alternative hypotheses in explaining the present distribution and abundance of this bird: habitat structure and composition, food availability and food preference, and (2) propose management strategies. Point-counts and capture-recapture indicate a population of 60 to 200 pairs. The bird breeds from mid-June to late August. Annual mortality was in the order of 45-50%. Stations were marked every 200 m along tracks covering all vegetation types in the area. These were walked three times a month to record presence or absence of birds per habitat type. The habitat structure and vegetation composition were recorded at these stations and used to explain the number of times a bird was detected at stations using logistic modelling. Diet was assessed from observations of feeding birds and faecal analysis. Food supply was examined from quadrats placed systematically within foraging habitats and from marked plots of winter food stocks. Food preferences were examined by comparing (a) availability and usage of foods in the field, (b) selected and rejected food items and (c) using food choice trials with captive birds. TTie importance of seed phenolics in explaining patterns of preferences for species and individual trees in winter was examined. The distribution of the Priolo was highly associated with native vegetation and its margins all year round. Other habitats were of marginal importance.The precision of habitat selectivity increased in the following direction: summer<autumn<winter. The diet ranged from invertebrates and herbaceous seeds in summer, to seeds of fleshy fruits in autumn, fern sporangia and tree seeds (of Clethra arborea) in winter and, flower buds, fern fronds and moss tips in spring. Introduced species were very important in the diet but in August/September and in April native species were significantly more important. In April, birds were highly dependent on flower buds of Ilex perado. At this time, the abundance of seeds and sporangia was at its lowest. An experiment with enclosures showed that the Priolo foraging pressure significantly reduced the number of flower buds. Food preferences changed throughout the year and seemed a complex function of availability, size, palatability and accessibility of foods. Seeds of fleshy fruits were preferred to those of C. arborea. Seeds of C. arborea were ignored when flower buds reached a length of 2.8-3.0 mm. In autumn, C. arborea was preferred to sporangia of large ferns, but in spring the birds showed the opposite behaviour. Priolos also showed preference for larger items of most winter foods. Phenolic compounds did not explain C. arborea tree preference nor why birds switched from seeds to flower buds in spring. Within a small resource spectrum birds should be less discriminating and energy may be the most important factor in food selection. Overall, the data suggested that a food supply hypothesis was better in explaining patterns of Priolo distribution and abundance than a habitat structure hypothesis but food preference is important for a full understanding of this question. A shortage of preferred foods may occur in late winter. Other available foods may be inadequate. Some implications for the limitation of this population are discussed. The Priolo needs the several mosaics of the native forest to complete its annual cycle. Nowadays, the mature forest mosaic seems to be the least abundant. In terms of conservation it is necessary to control the expansion of exotic plants and, restore and enlarge the area of native forest. It seems especially important to increase the population of flower-bud producing species.