Considerable attention has been devoted to contaminants (mainly PCBs and mercury) in subsistence foods (particularly fish) from various parts of the world. However, relatively little attention has been devoted to examining mercury levels in a full range of subsistence foods from a particular region. While managers and scientists compute risk based on site-specific data on contaminant levels and consumption rates, a first step in making risk decisions by subsistence peoples is knowledge about the relative levels of mercury in the foods they eat. This study examined levels of mercury in subsistence foods (edible components) from several islands in the western Aleutians of Alaska, including algae (4 species), invertebrates (9 species), fish (15 species) and birds (5 species). Samples were gathered by both subsistence hunters/fishers and by scientists using the same equipment. Another objective was to determine if there were differences in mercury levels in subsistence foods gathered from different Aleutian islands. We tested the null hypotheses that there were no interspecific and interisland differences in mercury levels. Because of variation in distribution and the nature of subsistence hunting and fishing, not all organisms were collected from each of the islands. There were significant and important differences in mercury levels among species, but the locational differences were rather small. There was an order of magnitude difference between algae/some invertebrates and fish/birds. Even within fish, there were significant differences. The highest mean mercury levels were in flathead sole (Hippoglossoides elassodon, 0.277 ppm), yellow irish lord (Hemilepidotus jardani, 0.281 ppm), great sculpin (Myoxocephalus polyacanthocephalus, 0.366 ppm), glaucous-winged gull (Larus glaucescens, 0.329 ppm) and its eggs (0.364 ppm), and pigeon guillemot (Cepphus columba, 0.494 ppm). Mercury levels increased with increasing weight of the organisms for limpets (Tectura scutum), and for 11 of the 15 fish species examined. Nine of the 15 fish species had some samples over the 0.3 ppm level, and 7 of 15 fish had some samples over 0.5 ppm. For birds, 95% of the pigeon guillemot muscle samples were above the 0.3 ppm, and 43% were above 0.5 ppm. While health professionals may argue about the risk and benefits of eating fish, and of eating alternative protein sources, the public should be provided with enough information for them to make informed decisions. This is particularly true for subsistence people who consume large quantities of self-caught foods, particularly for sensitive sub-populations, such as pregnant women. We argue that rather than giving people blanket statements about the health benefits or risks from eating fish, information on mean and maximum mercury levels should also be provided on a wide range of subsistence foods, allowing informed decisions, especially by those most at risk.