Movements of animals provisioning offspring by central place foraging extend from short, highly local trips where food is brought back essentially unchanged from its normal condition to extensive interseasonal movement where the offspring are nourished from body reserves built up during the adult's absence from the breeding site. Here, appropriate strategies for maximizing lifetime reproductive success depend on the abundance and location of prey in relation to breeding sites and the energetics and speed of travel of the animal. Magellanic Penguins Spheniscus magellanicus undertake central place movements that are particularly variable during the incubation period; trips may last from a single day to over three weeks depending on colony locality. We reasoned that site-specific variability in prey distribution and abundance is responsible for this. Remote-sensing systems attached to 92 penguins from six different colonies over the species distributional range over the Patagonian Shelf were used to determine space use and foraging patterns in an attempt to understand the observed patterns. Birds in the north and south of the latitudinal range were essentially monophagic, feeding primarily on anchovies Engraulis anchoita and sprats Sprattus fuegensis, respectively, both species that are to be found relatively close to the colonies. Penguins in the center of the distributional range, where these pelagic school fish prey are essentially absent at that time of the year, traveled either north or south, to the same regions utilized by their conspecifics, presumably to exploit the same prey. A simple model is used to clarify patterns and can be used to predict which movement strategy is likely to be best according to colony location. During chick rearing, southerly movement of anchovies and northerly movement of sprats mean that Magellanic Penguins in the center of the distributional range may benefit, although the abundance of these fish is considered to be less than that closer to the Magellanic Penguin range limits. The extensive time involved in the foraging trips during incubation coupled with the postulated poorer prey conditions during the chick-rearing phase may help explain why Magellanic Penguin colony sizes in the center of the range are not elevated.