[ 292 ] 



XXXVIII. Determination of the Acceleration of Gravity for 

 Tokio, Japan. By W. E. Ayrton and John Perry*. 



AS no experiments had, as far as we were aware, been made 

 to determine the value of g in Japan, it appeared to us 

 desirable that the value should be accurately measured, at any 

 rate for our own college in the capital, Tokio. Consequently 

 in 1877 an elaborate series of experiments was carried out by 

 some of the students under our supervision. The method first 

 employed consisted in experimentally finding two parallel axes 

 in a pendulum on opposite sides of the centre of gravity, and 

 in a plane with it, such that the times of oscillation about 

 either axis would be the same. The distance, then, between 

 these axes experimentally found would, as is well known, be 

 the length of the equivalent simple pendulum, from which g 

 could be calculated by the formula 



=*\/l 



V 9 



Two Kater's pendulums were employed — one made by 

 Messrs. Elliott, and the other by Messrs. Negretti and Zambra. 

 Borda's method of coincidence was employed ; that is, one of 

 the Kater s pendulums was suspended exactly in front of the 

 pendulum of a clock, consisting of a wooden rod carrying a 

 brass bob and beating approximately seconds. One observer 

 watched the two pendulums, vibrating one in front of the 

 other, through a telescope some ten feet away; and at the in- 

 stant a pointer attached to one of the pendulums exactly coin- 

 cided with a line drawn on the other (the coincidence taking 

 place in the axis of the telescope as observed by the cross- wires), 

 a signal was given, and the time of the clock noted by two 

 independent observers. A very large number of successive 

 coincidences was in this way observed, then the Kater's pen- 

 dulum inverted, swung on the other knife-edge, and the same 

 thing repeated. From these experiments the exact time of 

 vibration of the Kater's pendulum about either knife-edge 

 was ascertained, and one or both knife-edges moved to di- 

 minish the difference in the time of vibration, and the whole 

 experiment repeated. 



But although these observations, first with one of the 

 Kater's pendulums and then with the other, were continued 

 for some months, many thousands of vibrations being observed 

 by the students, who worked at these experiments in their 

 usual most praiseworthy way, the results were always unsatis- 

 factory. For an approximate value of g for any part of the 



* Communicated by the Physical Society. 



