134 F. OMORi : 



plication by the writing pointer. On the other hand, if the pul- 

 satory oscillations be due to tiltings of the ground, the recorded 

 amplitude of motion would depend both on the multiplication 

 ratio of the writing pointer and on the proper period of the 

 pendulum. Thus, if n be the multiplication ratio of the 

 pointer ; /, the distance between the end of the strut and the 

 centre of the heavy-bob ; <f, the angle of inclination of the axis of 

 support of the pendulum to the vertical; «, the angular amount of 

 the change of level of the ground in direction normal to the plane 

 of the pendulum ; and >', the motion of the point of the writing 

 pointer, we have — 



7 « 



or r = nla -^^^ , 



-* ft 



in which T is the complete period of vibration of the horizontal 

 pendulum, and Tq its period when the pendulum swings vertically. 

 In our case, the EW and NS component Horizontal Pendulums 

 are exactly similar to each other, except in the values of the T's, 

 which are respectively equal to 28 and 17 seconds (§ 8). The 

 ratio of the range of motion as recorded by these two apparatus 

 would, for a given value of «, be as 28^ : 17^ that is to say, as 

 2.7 : 1. If, therefore, tlie pulsatory oscillations be due to the 

 tiltings of the ground, the range of motion would, allowing for the 

 alternations of maxima and minima, be shown, on the whole, nearly 

 three times greater in the EW component apparatus than in the 

 NS one. Such, however, is not the case, the range of motion 

 being always equal in the simultaneous diagram of the two 

 component pendulums. 



Fig. 12a and fig. 12b are the EW component diagrams of the 

 pulsatory oscillations on the 14th December, 1898, the former 



