Successful approaches to race determination of unidentified human remains have been developed by anatomists and physical anthropologists, but few quantitative methods are available for distinguishing American Indian crania from those of whites. The leading method in use today is particularly ineffective in its placement of American Indian skulls from the western regions of the United States. Recent development and testing of a new metric method suggests a much more effective technique. The method involves six breadth and projection measurements of the midfacial skeleton, the calculation of three indices, and a simple direct reading of results. The method has the additional advantage of use in the autopsy room with minimal dissection of soft tissue required. Based upon a less extensive test of East Asian and Arctic Mongoloid crania, the method appears to be even more effective in separating them from the sharp featured whites. Larger samples of American blacks and Polynesians are presently under study and these also appear to separate quite readily from whites using the same or similar sectioning values.