ON THE MEASDREMENT OV THE LUNAR DISTURBANCE 01? GRAVITY. 99 



creased by a small leugfcli ca, the increase of tlie linear deflection of the 

 large pendulum is mcal(m + M). If I be the length of cither pendulum, 

 the angular deflection of the larger one is mlajl^m + M), and this is 

 the deflection which would be produced by a horizontal force equal to 

 mlajl{m+M) of gravity. It is clear, then, that by making the inequality 

 between the two weights rii and M very great, and the displacement of the 

 point of suspension very small, we may deflect the large pendulum by as 

 small a quantity as we like. The theory is almost the same if the two 

 pendulums are not of exactly the same length, or if tho length of one of 

 them be varied. 



Now in our application of this principle we did not actually attach the 

 two pendulums together, but we made the little pendulum lean up against 

 the large one ; the theory is obviously just tho same. 



We call the small pendulum ' the disturber,' because its use is to 

 disturb the large pendulum by known forces. A small copper weight for 

 the disturber weighed '732 grammes, and the large pendulum-bob, with its 

 pulley, weighed 4831"5. Therefore the one was 6600 times as massive as 

 the other. The disturber was hung by a platinum wire about xoVcr*'^ of 

 an inch in diameter, which is a good deal thinner than a fine human 

 hair. 



We must now explain how the disturber was suspended, and the 

 method of moving its point of suspension. 



Parallel to the sides of the slot in the slate slab there was riveted a 

 pair of brass rails, one being V-shaped and the other flat ; on these rails 

 there slid a little carriage with three legs, one of which slid on one rail, 

 and the other two on the other. A brass rod with an eyelet-hole at the 

 end was fixed to the centre of the carriage, and was directed downwards 

 so that it passed through the centre of the slot. The slot was directed 

 so that it was perpendicular to the T-piece from which the pendulum hung, 

 and the brass rod of the little carriage was bent and of such length, that 

 when the carriage was pushed on its rails until it was as near the centre of 

 the slab as it would go, the eyelet-hole stood just below the T-piece, and 

 half-way between the two wires. A micrometer screw was clamped to 

 the slab and was arranged for making the carriage traverse known lengths 

 on its rails, and as the wires of the pendulum were in the E. and W. 

 plane, the carriage was caused to travel N. and S. by its micrometer 

 screw. 



One end of the fine platinum wire was fastened to the eyelet, and the 

 other (as above stated) to the small disturbing weight. The platinum 

 wire was of such length that the disturber just reached the pulley by 

 which the big pendulum hung. We found that by pushing the carriage 

 up to the centre, and very slightly tilting it off one rail, we could cause 

 the disturber-weight to rest on either side of the pulley at will. If it 

 was left on the side of the pulley remote from the disturber-carriage, it 

 was in gear, and the traversing of the carriage on its rails would produce 

 a small pressure of the disturber on to the side of the pulley. If it was 

 left on the same side of the pulley as tho disturber-carriage, the two 

 pendulums were quite independent and the disturber was out of gear. 



On making allowance for the diS'erence in length between the pen- 

 dulum and the disturber, and for the manner in which the thrust was 

 delivered at the top of the pendulum, but omitting the corrections for 

 the weights of the suspending wires and for the elasticity of the copper 

 wire, we found that one turn of the micrometer screw should displace the 



H2 



