Drilling at Obsidian Dome has provided continuous core samples of the distal and proximal portions of Obsidian Dome, its conduit, and an associated feeder dike. Both the dome and conduit are chemically and mineralogically zoned and consist of a finely porphyritic, high-Ba, low-silica rhyolite occurring in the basal portion of the dome and margins of the conduit and a finely porphyritic, low-Ba, higher-silica rhyolite occurring in the upper portion of the dome and center of the conduit. The high-Ba rhyolite contains two distinct phenocrysts assemblages with two distinct compositions and represents mingled magmas. The low-Ba rhyolite in the dome and conduit contains a significantly fewer disequilibrium phenocrysts and is only slightly mingled. The dike, sampled at 600 m depth, as well as a related tephra fall from Obsidian Dome vent, are entirely low-Ba rhyolite that contain no evidence of magma mingling. End-members of the mingled magma, calculated using two different methods, are a 63% silica end-member and a silicic end-member identical in composition to the dike and tephra fall from Obsidian Dome vent. The silicic end-member was the first magma emplaced in the dike and comprised much or all of the first magma vented to the surface during formation of the Obsidian Dome vent when eruption rates were high. Magma mingling of mafic and rhyolite magmas occurred during formation of the conduit. A comparison of the magma types at Obsidian Dome with the other 600-year-old magma types in the Inyo chain demonstrates that the finely porphyritic rhyolites at Obsidian Dome are similar to the finely porphyritic rhyolites at the other domes. All of these originated by mingling of two end-members. The coarsely porphyritic rhyolites that occur at Glass Creek and Deadman domes are distinctly different from the finely porphyritic rhyolites. The two rhyolites can not be related by any fractionation scheme and could not have been in thermal or chemical equilibrium with each other.

These observations require considerable heterogeneity in the Inyo Domes magmatic system. Two possible interpretations of the chemical patterns variation and emplacement sequence are explored. In the first interpretation of chemical patterns observed in the surface and subsurface samples at Obsidian Dome are explained by differential draw-up of magma from a stratified dikelike reservoir. This differential draw-up is the result of changing flow rates during the course of the eruption. When the magma initially vented to the surface during the main part of the explosive phase, flow rates were very high, and the deepest layer of the reservoir was drawn into the conduit where magma mingling occurred. During the subsequent effusive phase, flow rates were low, draw-up depths were reduced, and only magma from the uppermost layer in the reservoir erupted, only slightly contaminated by earlier hybrid magma drawn up. Similar processes most likely occurred at the other Inyo domes; however, the low-Ba rhyolite was less abundant, and highly viscous, coarsely prophyritic rhyolite was near the surface and drawn up as a late pluglike structure. An alternate interpretation is that the dike initially consisted of heterogeneous magma domains and that the zonation observed in the conduit of Obsidian Dome and on the surface of the three Inyo domes is the result of the tendency of less viscous magma to encapsulate more viscous magma during sustained flow in a conduit. ¿ American Geophysical Union 1989