GENERALIZATION'S CONCERNING EARTHQUAKES 47 



fault lines, of which the history and the exact locations 

 of the previous occurrences of shocks are known. 



It is not to be denied that in some regions the 

 rhythm and periodicity of stresses and strain relief 

 has been sufficiently determined to permit some ap- 

 proximately accurate predictions. Cotton has noted 

 that "Hayes and Lawson predicted with some success 

 the Mexican earthquake of 1907 and Omori has shown 

 by his prediction of the Valparaiso and Formosa earth- 

 quakes how, in a very general way, it may be possible 

 within rather wide limits, to approximate both time 

 and place of earthquakes. He has also pointed out 

 that in a number of cases there is an increase of seis- 

 mic activity preceding a disastrous shock. "^ 



DR. G. K. GILBERT'S OPINION 



The words of the late Mr. G. K. Gilbert, one of 

 America's most eminent geologists, upon this subject 

 of earthquake predictions constitute a pertinent and 

 correct summation of the whole subject.- He states 

 that efforts to solve this problem have been connnected 

 with: (1) rhythm, (2) alternation, (3) idea of trigger 

 or starter, (4) prelude. All the above lines of approach 

 are worthy of consideration. The questions of rhythm 

 "are based on two ideas, (a) that any region having 

 a hard shock is immune for a period of years or dec- 

 ades, (b) that after the first destructive shock of any 

 earthquake the danger is over. Neither of these have 

 proven out in practice in all earthquakes." In con- 

 nection with this subject he also says that "It is not 



'Cotton, L., "Earthquake Frequency and Tidal Stresses," Bull, of the 

 Seismological Soc. of America, June-Sept. 1922. 

 -Science, Jan. 22, 1909. 



