If, on the other hand, we consider the outer solid shell to be of great thickness, we are met by the difficulty of accounting for the upheaval of liquid matter through such vast thicknesses of a solid shell. The differences in character of lavas extruded from closely adjoining volcanic districts seem equally difficult of explanation on any theory of a central, fluid nucleus and a solid, outer shell. Nor is the distribution of heat within the earth's crust so uniform as might be anticipated, if the source of that heat be a great central mass of highly heated materials. Under these circumstances, geologists and physicists have enquired whether any other conditions can be imagined as existing in the earth's interior, which would better account for the observed phenomena than does the hypothesis of a liquid nucleus and a solid outer shell. Two such alternative hypotheses have been suggested. Mr. Hopkins, adopting the theory that the earth has solidified both at the centre and its outer surface, endeavoured to explain the occurrence of volcanoes and earthquakes by supposing that cavities of liquid material have been left between the solid nucleus and the solid shell, and these cavities full of liquid material constitute the sources from which the existing volcanoes of the globe draw their supplies. But this hypothesis is found to be beset with many difficulties when we attempt to apply it to the explanation of the phenomena of volcanic action. It entirely fails, among other things, to account for the remarkable fact that during past geological periods the scene of volcanic action has been continually shifting over the surface of the earth, so that there is probably no considerable area of our globe which has not at one time or other been invaded by the volcanic forces.