ON EARTH TKEMORS. 



305 



frame of the support, and can be rotated about the axis of a rod, N, 

 pai-allel to P Q. The weight of the whole pendulum is 42 grammes, and 

 the distance of the centre of gravity, G, from the axis of rotation is 

 100 mm. 



Immediately beneath the mirror, m, and in the same plane with it, a 

 fixed mirror, m' (fig. 7), is attached to the tripod-stand, and can be 

 rotated about a horizontal and vertical axis from outside the instrument. 



To adjust the pendulum the mirror, m, is fixed on the rod, N, so that 

 its surface is approximately perpendicular to the axis of the ti'iangle. 

 The centre of gravity, G, of the whole pendulum is next brought into 

 that axis by supporting the pendulum at two points of the axis, near 

 L and H, and by displacing the small weight, i. The cylinders, b, b, are 

 then rotated until the steel points are directed towards the agate cups, 

 and are parallel respectively to the lines G p and Q G. This can be done 

 with sufficient accuracy by the eye by applying a cardboard triangle of 

 the form gpq. When the pendulum is suspended the arrangement is 

 therefore such that the pressure is dii-ected perpendicularly to the cups, 

 and there is consequently no tendency to slip. 



Fig. 8. 



M 



Two stops are fixed to the tripod-stand to prevent the pendulum up- 

 setting. One of these stops consists of a tube directed towards the 

 pendulum, and communicating on the outside of the tripod-stand with 

 an indiarubber tube and bellows, so that the pendulum can be made to 

 oscillate by the observer. The other stop is a vertical rod a little longer 

 than the pendulum, and carrying at the upper end a small horizontal 

 knife-edge. On this the pendulum can be suspended in a vertical position 

 by means of the perforation at l, and the knife-edge then agrees nearly 

 with the direction of the true axis of rotation. By finding the period of 

 oscillation in this position the constant required for the reduction of 

 the observations when the axis is in an inclined position can be deter- 

 mined with all desirable accuracy. 



The apparatus for photographic registration is placed about 4^ 

 metres north of the pendulum. The source of light is a petroleum lamp 

 capable of burning fourteen hours, and enclosed in an opaque case. In 

 this there is a movable vertical slit, 20 mm. in length, through which 

 the light passes to the lens ^ of the pendulum apparatus. The lamp-case 



' In the more recent experiments at Strassburg, the plane mirror, M, attached to 

 the pendulum is replaced by a concave mirror, and this renders the use of the lens 

 superfluous. 



1893. X 



