﻿Vol. 66.~\ AN EARTHQUAKE MODEL. 351 



I have not attempted in the model to illustrate movement 

 of more than one description at the same time. The same model 

 may, however, be employed to illustrate an earthquake connected 

 with a fault in which the vertical movement is the more important, 

 whether it be an incident of the folding due to the compression of 

 the earth's crust, or of the rise of mountain-ranges as a result of 

 the removal of material from their summits and slopes and its ac- 

 cumulation on the adjoining plains or sea-bed. 



"We are told that, after the earthquake of 1822, the shore-line of 

 Chile was found to be raised. This probably represents a rise by 

 its own elasticity of the tract under strain on the landward side of 

 the fault when released by fracture. The temporary withdrawal 

 of the sea, which so often accompanies an earthquake on that coast, 

 has been attributed to a depression of the sea-bottom. This may 

 represent the corresponding movement in the contrary direction on 

 the other— seaward — side of the fault. 



There is another point of some importance which is illustrated by 

 the model. When the left-hand frame f t is moved upwards by 

 means of the screw, the strips of steel are not only bent but are 

 subjected to tensional force, so that it would be impossible to keep 

 the vertical rods together if it were not for the lateral play allowed 

 to the right-hand frame f 2 , which permits of its approximating to 

 the other. In Nature, the distortion giving rise to such a tensile 

 force is not so marked ; for the gradual movement of one portion 

 of the earth's crust relatively to the other is usually small, com- 

 pared with the width of the tract of ground between them that 

 suffers deformation : but it must have an important influence in 

 determining the occurrence of fracture in a region of strain, as 

 rocks offer but feeble resistance to tension. In areas where the 

 earth's crust is in a state of tension and normal faulting occurs, 

 the tendency to fracture will be greater and faults more numerous. 

 On the other hand, in regions characterized by reversed faults 

 where tangential compression is present, this will play the same 

 part as the movement of the right-hand frame, and tend to 

 diminish the tendency to fracture, so that it will only be when 

 movements are very considerable or take place with comparative 

 rapidity that faults will result : in other cases they will be re- 

 placed by folds. The same region may, of course, be at one 

 time affected by compression and at another by tension. 



On these principles it is easy to understand why normal faults 

 should be more frequent than reversed faults, despite the fact that 

 we have every reason to believe that conditions involving pressure 

 more commonly prevail than those resulting in tension. 



Discussion. 



Mr. R. D. Oldham said that the model was certainly very in- 

 structive, so far as it reproduced the effects observed after the 

 Californian earthquake ; but, with regard to the cause of the earth- 

 quake so little was known, and the impossibility of reproducing the 



