Sex difference in criteria determining fidelity towards breeding sites in the great cormorant.

1. Many animals choose to breed in sites where they have previously been successful. Such fidelity could arise from the predictability of high quality breeding sites in a temporally stable environment. The quality of a site may be indicated by factors other than an individual's own success, because it may fail as a result of a random event that is unrelated to the intrinsic quality of the site. In particular, prior experience (familiarity) with the breeding area and the performance of neighbours could give complementary information about the quality of the site. 2. We present results from a long-term study of colonial great cormorants (Phalacrocorax carbo sinensis), where movement and reproductive success of individually marked birds within a colony was known. 3. Individuals were more likely to return to the same breeding site if they had been successful the previous season. 4. Fidelity of both males and females increased with increasing level of familiarity with the breeding area. Males were more likely to breed again in the same area, and their fidelity was more dependent on familiarity with the area than female fidelity. 5. The success of breeding sites within the colony was spatially autocorrelated, and cueing on neighbour performance should thus be advantageous. Female fidelity increased with increasing success of neighbouring birds, while male fidelity was unaffected by neighbour success. 6. We suggest that the difference between the sexes in the criteria determining fidelity arises because it is mainly the males that are involved in territorial disputes. They may therefore benefit more than females from knowing their neighbours, and this could override the importance of the intrinsic quality of the breeding area (i.e. reproductive success of neighbours). 7. The conflict over preferred breeding sites that arises within breeding pairs because of this sex difference may be an explanation for the high rate of mate change between years (92.5%) observed in this species