In the late 1960s Morton and Wingrove established that slag microstructres could be interpreted in terms of the anorthite-wustite-silica ternary phase diagram. About 10 years later Todd and Charles suggested that slags with high alumina content could not be explained by that system. They proposed the use of the wustite-silica-alumina system. Through this and other more recent work which includes our work on archaeological slags from a number of Eurpoean locations it has become apparent that most slags cannot be characterized in terms of one or two simple ternary phase diagrams. Furthermore, simple metallographic examination of slags can be misleading if not complemented by microprobe analysis, particularly when studying slags from a previously unexplored geographical region. Variations in the local raw materials, furnace conditions and slag cooling rates will determine the phase distribution and the slag/metal separation parameters. The proper phase identification and interpretation is of considerable importance not only in proveniencing ore sources but also in estimating the metal production from a site. This paper gives examples of the phases that it is possible to misidentify, due to morphologcial similaritites with well-known slag minerals and the implications of such interpretations.