A PRECIPITATION ORIGIN FOR DIVERSE CRYSTALLOGRAPHICALLY-ORIENTED LAMELLAE IN GARNET

Oriented lamellae of several minerals, including rutile, ilmenite, and apatite, are found in garnet from metamorphic and igneous rocks with extreme temperature or pressure histories. If lamellae formed by exsolution (solid-state precipitation), they are valuable tools for reconstructing precursor garnet chemistry at conditions where the trace substitutions were soluble in the garnet lattice as constrained by experimental work. However, competing hypotheses to exsolution have sparked debate about using lamellae to reconstruct garnet composition. These hypotheses commonly invoke lamellae growth from a fluid of implausible composition (such as precipitation of only oxides); emplacement along intersecting garnet “cleavages” without preserving planar structures (moreover, garnet has no cleavage); or emplacement of minerals far from equilibrium with the rock matrix. Likewise, hypotheses that require epitaxial attachment of oriented lamellae minerals to growing garnet faces have as of yet no support from in situ textural evidence.

We test the hypothesis that crystallographic orientation relationships (COR) in lamellae-host systems are indicative of formation process using new electron-backscatter diffraction (EBSD) data for ~1000 °C granulites from Connecticut (rutile, ilmenite, apatite lamellae) together with EBSD data from the Koralpe region, Austria (rutile, ilmenite, corundum; Griffiths et al. 2016). About 55% of the ilmenite from both localities and 50% of the corundum preferentially align a-axis//<111>_garnet, implying like controls for trigonal minerals. Apatite prefers (>50%) to have its close-packed c-axis//<111>_garnet. More than 30% of the rutile from both localities have c-axis within a 28.5 ± 2.5° cone around <111>_garnet. Additional COR are also observed such that the majority of lamellae have a COR with garnet. Furthermore, one group of rutile has analogous COR to the Widmanstätten pattern (WP) in meteorites; this is relevant because the WP has unequivocal precipitation origins. Shared COR and similar COR proportions between localities, as well as clear COR consistency between lamellae minerals of the same crystal system, are strong evidence of host and lamellae lattice energetics controlling solid-state lamellae growth, rather than other formation mechanisms.