Abstract
Irvine’s (1965, 1967) papers on the use of spinels as petrogenetic indicator phases have made petrologists more aware of the thermodynamic approach to evaluating spinel compositions. The six-endmember spinel prism plot (Irvine, 1965) and two-parameter projections of it are now common in the literature. However, various researchers have chosen to give different meanings to the established parameters.
Since Irvine’s papers originally were published, a large body of data has accumulated on kimberlitic spinels. The latter tend to be high in Ti, an element not considered in the original spinel prism plot. Attempts have been presented in the literature for direct adaptation of the prism plot by replacing ferric iron endmembers with titanian ones. However, the unique structural formula of titanian spinels changes the configuration of the spinel prism. More importantly, commonly used ratio plots of the form Cr/(Cr+Al) vs. Fe2+/(Fe2++Mg) can be misleading for titanian spinels. If two-parameter plots are desired, it is suggested that those of the form Cr vs. Fe2+ be among the diagrams used to depict compositional differences among and fractionation trends in kimberlitic and other titanian spinels. Such diagrams have been used successfully to distinguish suites of spinels from several kimberlite intrusions in the De Beers Pipe, South Africa.
