and Recording Earthquake-Motions, 207 



I suggested that a conical pendulum would probably be found 

 a good arrangement for registering one component of the 

 motion of the earth during an earthquake. Since that time 

 an instrument on this principle has been made ; and as it has 

 been found to be very suitable for this purpose, a short de- 

 scription of it may be interesting. 



The instrument is shown in plan and elevation in fig. 4. In 

 that drawing P is a post which is intended to be fixed firmly 

 in the ground, and may project above it from a foot to a foot 

 and a half. W, W are weights fixed to the ends of the cross 

 bars b, each pair forming the bob of a conical pendulum. The 

 suspending wires, t, t, of these pendulums are attached to cross 

 pieces, c, c, slotted and made to slide under clamping-screws, s, s. 

 By sliding the cross pieces c, c backwards and. forwards and 

 turning them round to one side or other, the point of suspen- 

 sion can be adjusted until it is vertically above the point p at 

 which the small cross arm a rests. This cross arm a serves 

 the double purpose of causing the bob of the pendulum to 

 move in a circle, if it moves at all, and of forming the short 

 arm of the lever I, which is used for writing a component of 

 the motion of the earth. When the pendulum is adjusted in 

 the way just described the bob is evidently in neutral equili- 

 brium in any position ; and hence any error due to oscillatory 

 motion is avoided. Since, however, it is desirable that the 

 instrument should have a small amount of stability, it may be 

 made to have a long period of oscillation round a definite 

 point by adjusting the point of suspension a little forward, or 

 by fixing the top of the suspending wire and allowing its tor- 

 sional rigidity to control the motion of the bob. 



The only points where friction acts in this machine are p,p'; 

 and since the pressure on these points is small, they can be 

 made very fine and almost frictionless. In this respect the 

 conical-pendulum machine has a decided advantage over most 

 other machines of a similar nature, such as hinged-horizontal- 

 lever machines. The levers 1,1 are arranged so that they 

 come out at an angle of 45° to the direction of the component 

 they are intended to measure, and parallel to each other. This 

 arrangement is adopted for the purpose of getting the record of 

 the two rectangular components side by side. When the record 

 is made in this way, the direction of motion of the earth can 

 be much more easily determined than if they were recorded 

 at different points of the same drum or revolving plate. The 

 question of change of phase-relation between the two compo- 

 nents can also be much more easily investigated with this 

 arrangement. Another advantage which this apparatus pos- 

 sesses, is the very small mass required for the moving parts, 



Q2 



