The occurrence of interstitial granite-glass in all formations of the Columbia River Basalt Group and its petrogenetic implications

Almost all Columbia River Basalt Group (CRBG) flows contain a residuum that consists of two phases, one chlorophaeite-rich and one a granite-glass. The chlorophaeite includes types that vary from irregular aggregates to polycrystalline spherules and drop-shaped inclusions of two major types occurring totally within the unaltered granite-glass. The granite-glass is usually isotropic but may be cryptocrystalline. The average analysis for our samples from Wanapum Basalt flows is SiO (sub 2) , 76.1 percent; Al (sub 2) O (sub 3) , 12.5 percent; FeO(T), 1.7 percent; MgO, 0.6 percent; CaO, 0.4 percent; Na (sub 2) O, 2 percent; K (sub 2) O, 6 percent; and TiO (sub 2) , 0.7 percent. The modal abundance by volume varies from a trace in the Picture Gorge Basalt and 1 percent in some Imnaha Basalt flows to 10 percent in the Grande Ronde Basalt and as much as 24 percent in the Wanapum Basalt. If segregated and accumulated, this glass could yield a potential potassic granite batholith of about 10,000 km (super 3) , comparable in size to the Idaho Batholith to approximately 1 km depth. This glass is compositionally very similar to some of the Tertiary granites of Syke and to Tertiary rhyolites from east Iceland. Separated glasses contain rare-earth elements (REE) that mimic the whole-rock REE except for a substantial negative Eu anomaly. Cs, Rb, Ba, Hf, and Ta are greatly enriched over the whole-rock composition. The glasses must represent the residual liquid from which the fayalitic olivine, augite, andesine, and magnetite of the Wanapum Basalt ferrobasalts have crystallized. The current petrogenetic theory for North Atlantic Tertiary granite occurrences is by derivation from ferrobasalt by fractional crystallization. The CRBG glasses fit this model, except that in this case the rapid eruption of CRBG has precluded the physical separation of the rhyolite component. Our present theory of direct derivation of CRBG from the mantle without significant crustal contamination thus has the corollary that it is also possible to derive large volumes of granite from that same source, given a suitable fractionation process.