1896. THE HORIZONTAL PENDULUM. 237 



nearly vertical, leans, of course, slightly in the direction towards the 

 centre of gravity of the pendulum. 



From the tube F, a rod, K, projects outwards through the frame, 

 A, and carries a mirror, M, capable of rotating about the vertical rod, N. 

 The mirror is shown in section in Fig. 5, and its plane is perpen- 

 dicular to the axis of the pendulum. As it is situated somewhat 

 above that axis, the centre of gravity of the pendulum is brought 

 back to the axis by adjusting the movable weight, H. 



In the direction of the axis, and about 4^ metres from the 

 pendulum, is placed a petroleum lamp, the light from which emerges 

 through a vertical slit in the case enclosing the lamp. It then passes 

 through a lens in front of the mirror, M, is reflected by the mirror, 

 passes through the lens again, and by its means is brought to a focus 

 on a graduated scale or photographic recording apparatus placed 

 beside the lens. If the pendulum is deflected through any angle, the 

 ray of light reflected by the mirror, M, is turned through twice that 

 angle, and the distance through which the line of light is displaced 

 along the scale is a measure of the pendulum's deflection. 



If the movements of the pendulum are to be recorded con- 

 tinuously by means of photography (12), the scale is replaced by a 

 cylinder with its axis horizontal and at right angles to that of the 

 pendulum. Round this cylinder is wrapped a sheet of sensitive 

 photographic paper on which the light is received after passing 

 through a horizontal slit in the case enclosing the recording 

 apparatus. The image of the source of light would naturally be a 

 vertical line, but a cylindrical lens is placed in front of, and parallel 

 to, the cylinder, and this reduces the line to a bright point of light. 

 The cylinder revolves by clockwork, and the point traces its course as 

 a narrow line' on the sensitive paper. 



To determine the form of this curve, a base-line is provided by 

 means of a mirror, M^ (Fig. 5), immediately below the other. This is 

 attached to the stand, B, of the pendulum, but is at the same time 

 capable of adjustment. The light proceeding from the slit of the 

 lamp falls on both mirrors, M and M^ and, the latter being fixed, the 

 light reflected by it traces a straight line on the sensitive paper. The 

 varying distance of the curved line from the base-line thus aftbrds a 

 measure of the pendulum's deflection. A small screen is placed 

 immediately below the point of light from the fixed mirror, and is 

 raised by the clock once every hour and kept up for five minutes. 

 The base-line being in this way broken at hourly intervals, the time 

 can be determined at which any particular disturbance of the 

 pendulum has taken place. 



The stand, B, of the pendulum rests on three foot-screws equi- 

 distant from one another. One of them is immediately beneath the 

 axis of the pendulum, and the line joining the other two is at right 

 angles to the axis. The screws are carefully worked and of known 

 pitch, and are provided with large graduated heads. If one of the 



