interval since the " driving " pendulum was started, the flexure correction is readily 

 calculated. The result of the motion of the pillar is such as to increase the virtual 

 length of the pendulums by a small amount, which gives rise to a small change in the 

 period of vibration As. 



From theoretical considerations, the following formula has been calculated, which 

 holds good after the pendulums have been swinging a few minutes (say, 10-15 minutes) : 



a T 2 m.h.. 



A g _ X 



A.n(tt ) infa 



m., = mass of driven pendulum. 

 M! = mass of driving pendulum. 



T = 0-508=mean time of swing of the two pendulums. 

 t t =tiine in seconds since the release of the driving pendulum, 

 a = amplitude of driven pendulum at time t. 

 A = amplitude of driving pendulum at time t. 



hi and Ji 2 =fhe distances between centre of suspension and centre of gravity of 

 the pendulums. 



The driving pendulum is set swinging at the time t , which is usually an even half- 

 niinute. After about 10 minutes, when the driven pendulum has received a measurable 

 amplitude of vibration, the procedure runs : 



1. At an even minute, observation, through the telescope, of the scale reading 



above and below for driving pendulum. 



2. 20 sees, later, - observation of the scale readings for driven pendulum. 



3. 20 sees, later still, (2) repeated. 



4. 20 sees, later (at the full minute), (1) repeated. 



The mean of 1 and 4 gives a measure of " A," and the mean of 2 and 3 gives a 

 measure of " a " at the mean time t, which is an even half-minute, so that the value 

 of t 1 is given in even minutes. 



A full minute is now allowed to elapse and the observation repeated at the end 

 of this period. In general, six independent series were observed. The value of As 

 is calculated from each and the mean taken as the correction to be applied to the observed 

 time of vibration. 



The value of As is, in general, different in the two directions at right angles to one 

 another. For determining the value in the plane at right angles to the axis of the 

 observing telescope, Nos. 5 and 7 are used on their respective agate planes. In order 

 to determine the correction to be applied to No. 21, which swings in a plane parallel 

 to the line of sight, use is made of the auxiliary pendulum. This has an attachment 

 for varying the period of vibration so that it may be adjusted to isochronism with 

 No. 21. For making this observation, No. 21 is brought forward to the agate plane A 



10 



