"Very High Alumina Basalt":

The proposal by Hubbard et al. (1) that "Very High Alumina basalt" ("VHA basalt") may represent an important primary lunar magma type neglects consideration of certain well-characterized high-magnesium highland rocks, among them spinel troctolite from Luna 20 (2) and Apollo 16 samples (3) and dunite from Apollo 17 samples (4), and the possibility that "VHA basalt" may in fact have originated by impact mixing and remelting of these and anorthositic rocks with KREEP (alkalic high-alumina) basalt. We suggest that the latter hypothesis accounts in a more satisfactory manner for many of the properties of "VHA basalt" discussed by Hubbard et al. (1) than their hypothesis that "VHA basalt" is a widespread volcanic rock. Hubbard et al. (1) point out correctly that the compositions of "VHA basalts" cannot be derived by a twocomponent mixing model with KREEP basalt and either anorthosite or "highland basalt" (anorthositic norite or gabbro) as end-members. However, if it is considered that KREEP basalt is mixed with a suite of cumulate rocks consisting of various proportions of plagioclase, low-calcium pyroxene, olivine, and spinel, the major element compositions of "VHA basalts" and some other brecCiated highland rocks can