Abstract
The presence of small bone fragments in the archaeological record has been frequently used to infer the activities of bone grease and/or soup manufacturing. This article explores the possibility that the activity or activity-complexes responsible for the bone fragments might be manipulated to account for additional archaeological data. Emphasis is placed on building predictive models against which the data from three Northern Plains-Upper Midwest archaeological sites will be compared. Results indicate that such an approach provides more information about the cultural activities being conducted at a site than does the other approach.

This publication has 1 reference indexed in Scilit: