J. StarTiie Gardner — Underground Heat. 403 



composing their cliffs, corresponding to that of the maiu-land from 

 which they have been severed, but at a much lower level, a result 

 that could only be brought about through subsidence or elevation. 

 To bring forward the overwhelming proofs of this double movement 

 on coast-lines would carry us beyond the limits of this article. 

 During the Glacial period the land was everywhere depressed by 

 the weight of ice, to the extent of 1350 feet in Wales ; and everywhere 

 rose when the weight was removed by the ice melting off again. In 

 like manner I have twice observed in Iceland that lava debouching 

 on to plains has caused them to sink bodily. Darwin's well-known 

 explanation of the formation of coral reefs and atolls requires that 

 subsidence should keep pace exactly with the added weight, by the 

 depression of the rocks or banks on which they are forming. 



It would be physically impossible for such movements to take 

 place if the earth were solid throughout. The fluid or viscous layer 

 which is near to the surface in volcanic regions must extend under- 

 neath all areas where movements of upheaval or subsidence have 

 been observed, and since movements to a greater or less extent seem 

 universal, the only inference is that the crust rests everywhere upon 

 a viscous substratum. Yet we know that the earth as a whole is a 

 rigid body, of the rigidity of steel or glass, and it is believed that its 

 principal mass is kept solid at a temperature beyond the fusing point 

 of rock through the pressure of the external envelope. But if this 

 is so, the pressure must become relaxed as the surface is neared, and 

 at a certain point the rock must obey its impulse and melt, and we 

 must thus find at some depth or otlaer beneath the crust, rendered 

 solid by loss of heat, a layer in a state of fusion. The Eev. Osmond 

 Fisher has shown the high probability that this state of things does 

 actually exist, and many geologists are now embracing the same 

 views. At the same time the plications the crust has undergone in 

 many places, as at Bergen, demonstrate that it is far more flexible 

 than had been supposed. The ceaseless removal of material, by dis- 

 integration and abrasion, from one area of a flexible crust, and its 

 redeposition by various transporting agencies, on another, would be 

 likely to destroy the equilibrium of the liquid layer beneath, and to 

 cause those changes from sea to land revealed in stratified rocks, as 

 well as the slow movements which are even now discernible on every 

 coast. Subsidence under newly transported weight must lead to a 

 compensating elevation elsewhere, and thus, as old lands sink, new 

 ones rise to take their place. This appears a reasonable explanation of 

 the movements that have taken place throughout geological time, and 

 I know of no other that can compete with it. Pushed to its extreme 

 we should find that all oceans are areas of sedimentation, and there- 

 fore of depression, while high lands would be areas of denudation, 

 and therefore of elevation. Further we should expect to find that 

 the viscous layer would seek relief at the nearest point free from 

 sedimentation, and we should thus find our mountain chains running 

 parallel with ancient or modern coast-lines. At first sight this 

 would favour the theory of the permanence of ocean-basins and of 

 continents ; but the sea-bottom is itself traversed by currents which 



