We studied the foraging behavior of the Double-crested Cormorant (Phalacrocorax auritus) on Oneida Lake, New York, by monitoring the activities of 27 radio-tagged birds in July and August of 1999 and 2000. A total of 224 locations were obtained of cormorants actively diving, and presumed foraging, at the time of detection. A geographic information system was used to examine foraging distances from the nesting island, the water depth and type of substrate at preferred foraging sites, and to estimate kernel home ranges for analysis of individual foraging site fidelity. An explanatory model was developed to determine parameters affecting the distance to cormorant foraging sites. The mean distance to foraging locations of tagged cormorants from the colony site was 2,920 m (SE ?? 180 m, max = 14,190 m), and 52% of the locations were within 2,000 m of the nesting island. No cormorant was observed making daily foraging trips to outside water bodies. Mean foraging distance was greater during morning than in the afternoon, and there was a significant effect of the time of day on distance. There was no significant effect of sex date, a seasonal measure on distance to foraging location. Individual cormorants exhibited fidelity to specific foraging sites. Most cormorants foraged in close proximity to the nesting island much of the time, while those detected further from the island tended to return repeatedly to the same locations. Ninety percent of the foraging locations were in water depths ???7.5 m, and most were in water 2.5-5 m deep. Compositional analysis of habitat use revealed a preference for these depths, along with substrates of cobble with rubble, and silt with clay.