Abstract:
Using a data bank containing over 325 major-element analyses of biotite minerals from many localities around the world, it is demonstrated that biotite compositions depend largely upon the nature of magmas from which they have crystallized. Host rocks of the biotite minerals used include alkaline anorogenic (A-type) suites, calc-alkaline orogenic suites, and peraluminous (including S-type) suites. Biotite minerals representing these three magmatic rock groups define three compositionally distinct fields; biotites in anorogenic alkaline suites (field A) are mostly iron-rich, siliceous biotites (near annite), with an average FeO*/MgO ratio of 7.04; those in peraluminous (including S-type) suites (field P) are siderophyllitic in composition and have an average FeO*/MgO ratio of 3.48; whereas biotites in calc-alkaline orogenic suites (field C) are moderately enriched in Mg, with an average FeO*/MgO ratio of 1.76. It should be noted that the average FeO*/MgO ratio in biotite doubles from calc-alkaline through peraluminous to alkaline suites (FeO* = total Fe). Iron enrichment in biotites in the alkaline system occurs at the expense of both Mg and Al via two substitution schemes; Mg-Fe, and 2AI-3Fe2+. The latter substitution, however, does not play a sigificant role during the crystallization of biotites in calc-alkaline or peraluminous magmas, as Fe shows little variation with Al in biotites in both systems. In those two systems, the substitution 3Mg-2Al is vital; it produces siderophyllitic compositions in peraluminous rocks. The contrasting behaviour of Fe and Mg with respect to Al during the crystallization of biotite minerals is governed, in part, by the various physicochemical conditions including the behaviour of volatiles in these orogenic vs. anorogenic magma systems. The distinct biotite compositions obtained in fields A, P, and C for the various magmatic rock groups can be shown to have a theoretical basis. Furthermore, the discriminant boundaries that define these fields have a statistical basis. Statistical linear discriminant function analysis of the data on a three-group basis (A, P, and C) correctly classifies 89–95% of these biotite minerals. Thus, the results of this study demonstrate that the compositions of igneous biotites clearly reflect the nature of their host magmas.