A general view of the chemical diversity of the Hawaiian lavas is offered. They are regarded as derivatives of a primary (olivine-basalt) magma, whose approximate composition is deduced. The differentiation is largely controlled by density. The density of each type of molten, pore-free lava can be estimated closely from its “norm density,” as calculated by the Iddings method, the practical accuracy of which is confirmed by a special test. The actual density of the lava in a volcanic conduit is also a function of the weight percentage of juvenile gas, dissolved and in the bubble phase. This gas plays a double rôle, as flux and as inciter of convection in the lava column. As much as 6 per cent of the primary magma in its original home is assumed to be water in solution.
The genetic importance of this gas is particularly illustrated in the case of Hawaiian trachyte. Its consideration is followed by that of the phonolites which in other islands occur in amounts and structural relations like those characterizing the Hawaiian trachytes. Since the volcanic mechanism working in oceanic areas is probably like that working in continental areas, it becomes necessary to account for the fact that in the latter areas rhyolite so often appears at central vents in volumes and general relations essentially like those exhibited by trachyte and phonolite. The explanation offered goes far toward removing a commonly expressed objection to the theory of crystal-fractionation as the chief cause of chemical diversity in igneous rocks.
The picrite-basalt and chrysophyric basalt are most simply explained by the settling of early-formed olivines in the primary basaltic melt, the material of the crystals and melt being thoroughly mixed by two-phase convection in the lava column. “Andesites” and other “intermediate” lavas are attributed to mixing by the same process; the alkaline and melilite-bearing lavas to reaction of the primary basaltic melt with sedimentary material rich in calcium carbonate.
Comparison is made between the primary magmas under deep ocean and continent respectively; it is found that they do not greatly differ and that a fair interpretation of the facts observed at continental centers of eruption strengthens the set of hypotheses erected to account for both volcanism and petrogenesis in the Hawaiian Islands.
In both oceanic and continental areas the chambers in which difierentiation, differential refusion, and other reactions with crust-rock take place are thought to be abyssoliths, that is, more or less vertical injections of the basaltic, substratum melt into the earth's crust.