UPHEAVAL AND SUBSIDENCE. 



71 



substautially nil; for the ideal perfect fluid is the one which ofl^ers no resist- 

 ance to a shear. Of course there is no such substance as a perfect fluid ; but 

 it is manifest that such a liquid as water under these circumstances would 

 move with very little friction and exert only an excessively minute lateral 

 pressure on the mass below B. Thus, if the earth's interior were of water- 

 like fluidity, the mass under B would sink in a substantially vertical direc- 

 tion. 



On the other hand, if the earth's interior were a viscous substance, like 

 asphalt, the transfer of material from below ^ to ^1 could not occur without 

 overcoming great resistance. The viscous material would thus drag the 

 mass underlying B with it towards J. to a certain extent, and B would no 

 longer tend to sink in a substantially vertical direction, but in a line of least 

 resistance, inclining towards A. The angle which this line would make with 

 the vertical would increase with the viscosity up to a certain limit. 



It is important and very easy to determine a limit which this angle cannot 

 surpass. The viscous resistance is a horizontal force due to a vertical press- 

 ure. Now this resistance certainly cannot exceed the pressure which excites 

 it; in fact, it must always fall at least a little short of the pressure. But if 

 one imagines the horizontal traction quite equal to the vertical pressure, the 

 mass under B will move under the influence of two equal forces, one vertical 

 and the other horizontal, and will therefore tend to sink at an angle of 45°. 

 Thus, under no circumstances can the center of inertia of the mass at B tend 

 to diverge from a vertical line more than 45°. 



Experiments on Subsidenee. — On the basis of these deductions I have made 

 some rough experiments. Fancy a vessel containing some such substance 



Figure 10— Experiment on Subsidence. 



as hard tar with a level surface carrying a uniformly distributed load, I. 

 and then let half of the load be removed from one half of the surface to the 

 other. Then the conditions will be those represented in the diagram, figure 

 10. The load should now tend to sink into the mass in a slanting direction, 

 if the argument presented above is correct, instead of subsiding vertically, as 

 it would do in a perfect fluid. On trying this experiment with oiled asphalt 



