Eskimos and Aleuts have long shown uncertain anthropological relationships to each other and to Amerindians. Various researchers, using different samples, data, and methods, have derived diverse relationships among them. Despite supervising the collection of anthropometric data during various expeditions around the turn of the century, Franz Boas did not use these data to formulate or support his Eskimo wedge theory. The perception of Eskimo biological uniqueness has persisted (Greenberg et al. 1986; Laughlin et al. 1979; Laughlin and Harper 1988), although it has been challenged by Szathmary and Ossenberg (1978), Szathmary (1979), and Ossenberg (1992). Genetic analyses, whether using nuclear or mitochondrial DNA, have not revealed consistent relationships for various reasons. A widely known multidisciplinary study (Greenberg et al. 1986) has failed to show agreement among data sets. As a result, Boas's theory has not been adequately tested to this day. A modern analysis of anthropometric data collected under Boas casts doubt on an Eskimo wedge and the assumed close relationship between language and biology. Compared with other north Pacific populations, the Siberian, Labrador, and MacKenzie Delta Eskimo samples are anthropometrically closest to northeast Siberians, whereas the Aleuts are closest to some Northwest Coast Amerindians. An analysis with additional data collected by Hrdlicka (1930) reflects ethnohistorical relationships and a geographic pattern of population affinities: The Eskimos of southwestern Alaska are similar to Aleuts and selected Amerindian tribes on the Northwest Coast. Despite linguistic similarities, Eskimo and Aleut populations are diverse, with some being more similar to populations from opposite sides of the north Pacific.