Only under special circumstances can the hybrid character of a magma be definitely determined. Fenner has given adequate reasons for supposing that much of the cone of Katmai was dissolved by rhyolitic magma prior to and during the eruptions of 1912 which led to the formation of the summit caldera. The ejected pumice clearly betrays a dual origin. Dark streaks of andesite stand out conspicuously in the pale rhyolitic matrix, and crystals from the andesite are scattered throughout the rhyolitic glass. According to Fenner, rapid assimilation occurred chiefly, if not entirely, within the actual crater. Had assimilation proceeded slowly and at depth, the hybrid melt might well have lost the recognizable marks of its twofold origin.
If assimilation occurred inside Mount Mazama, the rocks affected were mainly hypersthene andesites of the cone itself and perhaps also some of the underlying olivine-bearing basic andesites. But seven-eighths of the magma erupted during the final eruptions consists of dacite pumice. The remaining eighth consists of basic scoria. Hence, if much of the interior of Mount Mazama was dissolved, the solvent must have been an acid magma.
A considerable interval of rest preceded the culminating explosions. This much is indicated by the violence and scale of the eruptions themselves, and by the fact that trees were growing on the moraines far up the sides of the volcano. During this interval of quiet the process of assimilation may have gone on until not only were the rising magma and its dissolved constituents thoroughly mixed, but the hybrid melt itself may have undergone gravitative differentiation into an acid, dacitic upper portion and a crystal-rich basic fraction below. Hence, perhaps, the absence of semidigested fragments among the ejecta. Positive proof that the magma chamber was enlarged by assimilation before the culminating eruptions began seems to be entirely lacking.
One more problem calls for discussion, namely, the remarkable paucity of large blocks of old lava among the products of these climactic eruptions. Their scarcity among the deposits of the first eruptions, that is, among the pumice fall, may be due simply to the fact that the initial eruptions, though they hurled out several cubic miles of granular pumice, were not powerful enough to dislodge large fragments of solid rock from the walls of the conduit and crater.
The scarcity of large blocks among the deposits of the later nuées ardentes admits of no such ready explanation. Surely avalanches strong enough to rush 40 miles from the source, carrying pumice bombs several feet in diameter, had power to sweep huge blocks of old rock for long distances beyond what is now the rim of the caldera. Thenuée ardente deposits of Mount Pelée and Montserrat are heavily charged with blocks many miles from their point of origin.
