TIMEKEEPERS. (jQ 



points where the hours 12 and 6 are marked -upon the dial 

 plate, and that the two side screws ff, stand at those points 

 where the hours c) and 3 are marked. If the timekeeper The rule Iipre 

 should go faster with the hour 12 highest (or vertical) ^^^^l'^'^,ll*}J%e° 

 with the hour 6" highest, screw in the screw k a little at the centre of ara-jifji 

 •■:, and unscrew the opposite screw at the hour 12, the .^|^|l]lf^.|,en ijhl^ 

 c quantity, if it should lose most in that position, do just «/ ^kes the 

 •..:ic contrary. The same rule is to be observed with respect V^'^'^^^^ ^''^''^ 

 to the hours 9 and 3, by the two side screws Jf. It may 

 however happen" that the balance will not preponderate at 

 either of these four points, or that the screws may not be 

 sufficiently powerful to effect the purpose. In this case for 

 the positions 12 and 0, by unscrewing a little one of the balls 

 gg, and screwing in the other, we may succeed ; but this 

 method should not be practised in superior timekeepers, 

 because by so doing, it will occasion one expansion piece to 

 ;ict more than it ought to do, and the other less, and destroy 

 that equality of expansion, of weight, and of distance, which 

 the very word Balance informs us ought to be preserved. To 

 remedy this inconvenience, another method has been contriv- 

 ed, hy which the balance may be rendered of equal weight 

 while the balls, the screws, and every opposite part, arc at 

 equal distances from the centre. Let the balance be made 

 with a light ring xxx (as in Fig.l, PI. II.) within the expansioji 

 pieces. Let there be three light equal weights kkk, which 

 by a screw in each may be fixed upon any part of the ring ; 

 then having adjusted the long and short vibrations, and the, 

 heat and cold, and having the mean time screws at equal 

 distances from the centre, and the balls at equal di;^tances 

 upon the taps (there will be no occasion for side screws), 

 try the timekeeper in different positions, and in a veiy few 

 trials, by n]o\ing the weights upon different parts of the ring 

 the positions may be adjusted very accurately. The weights 

 may be brought all to the same part, and the balance made 

 to preponderate in any given point, aud none of the other 

 adjustments will be affected by it, and the weights, upon 

 whatever parts of the circle they may be, will still remain at 

 an equal distance from the centre. Having adjusted for 

 long and short vibrations, heat and cold, and positions, it 

 remains only to regulate for mean time. Should the time- 

 keeper gain, increase the diameter of the balance by drawing 



cut 



