From the geophysical and physical laboratories at Harvard University have recently come data bearing on the effect of rising temperature on the compressibility of liquids, glasses, and crystalline solids. These new data have compelled the author of the following article to revise an earth-model which, in 1942, he derived from some general assumptions as to the condition of the globe at its beginning. According to the 1942 model the asthenosphere is pictured as chiefly peridotitic and wholly vitreous; according to the revised model the deeper and greater part of the thick, weak earth-shell is pictured as peridotitic and in the form of a two-phase mixture of dominant crystals and subordinate glass or, possibly, a gaseous solution in the critical state. The revised model seems to suggest good explanation for the weakness, high rigidity, and great viscosity of, and high velocities of seismic waves in, the outer half of the earth's silicate mantle.
During the revision there was the incidental discovery of a method of finding, to a close approximation, the compressibility of a crystalline rock from the norm corresponding to the chemical analysis of the rock, and this at pressures ranging from 1 bar to 10,000 bars; with a somewhat greater chance of error the method can be usefully applied up to 30,000 bars.