106 REPOKT— 1881. 



mirror is hung from a silk fibre passing throngli botli loops, the weight of 

 the mirror is sufficient to pull each loop taut. 



A single silk fibre was threaded through the eyelet-hole at the end of 

 the blunt point of the pendulum-bob, and tied in such a way that there 

 was no loose end projecting so as to foul the other side of the bifilar 

 suspension. The other end of the silk fibi-e was knotted to a piece of 

 sewing silk on which a needle was threaded. 



The pendulum was then hung from the cap by its wire loop, outside 

 the copper tube, and the silk fibre with the sewing silk and needle 

 attached dangled down at the bottom. The cap, with the pendulum 

 attached thereto, was then hauled up and carefully let down into the 

 copper tube. The sewing silk, fibre, and blunt end came out through the 

 hole in the diaphragm. 



"We then sewed with the needle through the two loops on the margin 

 of the mirror, and then through the eyelet-hole in the little horizontal 

 arm. The silk was pulled taut, and the end fastened ofi" on to the little 

 vertical rod, from which the horizontal arm projects. 



The mirror then hangs with one part of the silk attached to the 

 pendulum-bob and the other to the horizontal arm. 



The two parts of the silk are inclined to one another at a considerable 

 angle, so that the free period of the mirror is short, but the upper parts 

 of the silk stand very close to oiie another. The mirror-box encloses the 

 mirror and makes the copper tube watertight. 



There is another pai't of the apparatus which has not yet been ex- 

 plained, namely, the disturber. This part of the instrument was in reality 

 arranged before the mirror was hung. 



We shall not give a full account of the disturber, because it does not 

 seem to work very satisfactorily. 



In the form of disturber which we now use the variation of horizontal 

 thrust is produced hj variation in the length of the disturbing pendulum, 

 instead of by variation of the point of support as in the previous experi- 

 ment. It was not easy to vary the point of support when the pendulum 

 is hung in a tube which nearly fits it. 



The disturber-weight is a small lump of copper, and it hangs by fine 

 sewing silk. The silk is threaded through the eyelet in the horizontal 

 arm which forms part of the upper plug of the pendulum ; thus the 

 disturber- weight is to the N. of the pendulum. The silk after passing 

 between the wires suppoi-ting the pendulum has its other end attached to 

 the cap at the top at a point to the S. of the centre of the cap. Thus 

 the silk is slightly inclined to the plane through the wires. The arrange- 

 ment for varying the length of the disturbing pendulum will not be 

 explained in detail, but it may suffice to say that it is produced by a third 

 weight, which we call the ' guide weight,' which may be hauled up or let 

 down in an approximately vertical line. This guide weight determines by 

 its position how much of the upper part of the silk of the disturber shall 

 be cut off", so as not to form a part of the free cord by which the disturb- 

 ing weight hangs. 



The guide weight may be raised or lowered by cords. which pass 

 through the cap. If the apparatus were to work properly a given amount 

 of displacement of the guide weight should produce a calculable horizontal 

 thrust on the pendulum. The whole of the arrangements for the disturber 

 could be made outside the copper tube, so that the pendulum was lowered 

 into the tube with the disturber attached thereto. 



