1 44 F. OMORi : 



motion in tlie two apparatus would be to each other as 784 : 400, 

 or nearly as 2:1. Such, however, is not the case, the range of 

 motion being equal in the two diagrams. Hence T conclude that 

 tbe undulations in question, as probably also most of the move- 

 ments of this earthquake recorded by the Horizontal Pendulums, 

 are horizontal movements^ rather than tiltings, of the ground. 



The slow undulations, which occur in the Horizontal 

 Pendulum records of earthquake disturbances emanating from 

 very distant origins, and whose periods are sometimes as long 

 as 20 seconds or even more, are also probably due to horizontal 

 movements of the ground, being evidently of the same nature 

 as the group of undulations marked a in the diagrams of figs. 

 22 a and 22 b. 



VI. 



27. Earthquake disturbances proceeding from distant origins. 



I have already stated in the preceding section that short- 

 period vibrations occur only in the earUest portions of an earth- 

 quake, these becoming probably dissipated, by the viscosity of the 

 material constituting the earth's crust, more rapidly than the 

 long-period undulations, which alone remain in the later portion 

 of the shock. Hence it is to be expected that the disturbance 

 coming from a very distant source would consist of onlv slow 

 undulations. 



The duration of a great, distant, seismic disturbance amounts 

 not seldom to 24^ or even 3 hours. The following is an example 

 of such an earthquake. 



28. Earthquake of the 17th November, 1898 ; 9. r)4.44p.m. 

 For the diagram see figs. 23 and 24. The total duration 



