Ill OUR MOLTEN GLOBE 49 



should be much less, because the lower level of the crust 

 in contact with the liquid substratum must always be at 

 about the same temperature — that of melting rock. This is 

 found to be the case ; the rate of increase at the St. Gothard 

 tunnel, where the observations were most complete, being 

 1° F. in eighty-eight feet, and the corresponding thickness 

 of the crust thirty-seven miles. This is certainly a 

 remarkable confirmation of the other observations, and of 

 the theory of mountains being supported in approximate 

 equilibrium by means of vast protuberances into the liquid 

 substratum beneath. 



The general result of the whole series of experiments 

 with the pendulum shows that gravity is normal at the 

 sea-level both over land and sea, and thus proves that the 

 surface of the globe is in a state of equilibrium. The 

 measures of the force of gravity over the oceans have been 

 necessarily taken on islands, and have led to a curious 

 discovery. The pendulum experiments on oceanic islands 

 such as the Galapagos, Ascension, St. Helena, Bourbon, 

 Guam, and others, all show an increase in the force of 

 gravity, which, on the average, is very nearly accounted 

 for by the subaqueous mass of land displacing water of 

 less than half the density of rock. Hence it is concluded 

 that these islands or island-mountains do not have " roots " 

 as do those on continents ; and the same thing occurs with 

 isolated volcanoes on continents, the attraction of Fujisan 

 in Japan being exactly that due to its own bulk unaffected 

 by the presence of " roots " projecting into the substratum. 

 This is explained by the fact that volcanic mountains are 

 not produced by compression forcing the crust both down- 

 wards and upwards, as other mountain masses are supposed 

 to have been produced, but are mere heaps of materials 

 derived either from the crust or the substratum, and 

 probably drawn from a considerable area. Hence' they 

 are balanced not by " roots " projecting immediately below 

 them, but by a slight depression or sagging of the crust 

 over a wide area, and thus having little effect on the rate 

 of the pendulum. In the case of the Falkland Islands, 

 however, the force of gravity is less than it ought to be' 

 and this exception affords an interesting confirmation of 



VOL. I. ^ 



