1865.] The Proposed Pendulum Operations for India. 259 



The apparatus was contrived so as to avoid any uncertainty in the 

 centre of oscillation of the pendulum, as well as any eiTor in the 

 measure of its length, by observing the times of vibration of a pen- 

 dulum ball suspended alternately by two wires, whose difference in 

 length was known. 



A toise was set upright on a narrow horizontal plane firmly fixed to 

 a perpendicular iron bar, and the contrivance by which the pendulums 

 were suspended could be placed either on the horizontal plane, or on 

 the top of the toise itself, so that the effective lengths of the wires 

 differed in the two cases by an amount exactly equal to the length of 

 the toise. The wires, which were of steel, were attached to a thin 

 strip of brass which unwound itself over a small cylinder. The pen- 

 dulum, thus suspended, described the curve called the evolute of the 

 circle. At the lower end of the iron bar, there was a micrometer 

 screw for measuring small differences in the height of the ball. 



The system of observation was as follows. At the commencement 

 of a series of coincidences with the longer pendulum, the thermometers 

 attached to the toise were recorded, and the reading of the lower 

 surface of the ball was taken with the micrometer screw ; the pendulum 

 was then set in motion, and after a sufficient number of coincidences 

 had been observed, the readings of the ball and thermometers were 

 again taken. Exactly the same process was then gone through with 

 the shorter pendulum : then from the times of vibration of the two 

 pendulums, whose absolute lengths were unknown, but whose difference 

 in length was accurately known, the length of the seconds pendulum 

 was easily computed.* There were a great many minute details to be 

 attended to, all of which were carried out with the greatest ingenuity 



* Let t x & Z x be times of vibration and length of longer pendulum.. 

 t 2 l 2 „ „ „ shorter „ 



l t — l 2 ■= difference in length = a 

 L = length of seconds' pendulum. 



9 y y 





*, 2 _ 





V 



V 



t 2 a _ 



i 



h 



h 



l 2 



or 



tx 2 



= 





a 







t L 2 



— 



V 2 



Again 



1 



L 



or 



L 





= 





a 















K 



2 _ 



" *! 



a 2 







34 





























