Bulk chemical characterizaton with techniques such as neutron activation analysis (NAA) has been used for many years to identify sources of archaeological artifacts by comparing bulk chemical profiles of archaeological artifacts to profiles determined for samples of raw material for source identification. The chemical approach to determining the provenance of archaeological pottery is complicated by the paste preparation activities of potters and by post-depositional processes. In a ceramic provenance investigation, the investigator typically identifies a series of relatively homogeneous subgroups, which most often are assumed to represent geographically restricted raw material sources. Although the source-specific interpretation is often the most parsimonious explanation of compositional patterning, it is really only the first of several competing hypotheses which, singly or in combination, might account for patterns of similarity and difference in a compositional data set. In this case, two white firing pottery types (SAWH and Ivory) made during the Late and Terminal Formative periods (300 B.C. - 200 A.D.) in southeastern Mesoamerica (Guatemala) were found to be chemically distinct.