«6 I'ree EfcapetnentS. > ' 



It is indeed very eafy to obferve, that the piece EL would operate without the detent, 

 though with much fridion during the time of repofe. The tooth of the wheel would in that 

 cafe reft upon its circular face. 



This efcapement is better calculated for a long than a fliort vibration. I have feen it 

 adapted to an half feconds pendulum clock, in the form delineated in fig. 6, PI. III. It 

 kept up a vibration of 8° with a very heavy pendulum, and no great maintaining weight, 

 but I was not informed of the quantities of thefe weights. The part ABDC is fixed to the 

 pendulum rod. BC is a lever moveable on the joint B, but not capable of falling below its 

 horizontal pofition. It terminates in a claw at C. At D is a pallet of agate or hard ftone, 

 having a flat poliflied face. F G H I is a detent moveable on the centre H, but kept in 

 its pofition againft the pin or flop K, by means of the counterpoife G. As it is now repre- 

 fented, a tooth I of the wheel bears againft the end of the lower arm or branch. Suppofc 

 the pendulum to be made to vibrate, and it will pafs clear of the teeth I L ; but when the 

 claw C ftrikes againft the upper end F of the detent, the flope furface of thofe extremities 

 will caufe the lever B C to rife, and the claw in its return will draw F towards the wheel, 

 and confequently difengage I. The tooth E will therefore fall on the pallet D, and impell 

 the pendulum in its vibration ; but before it quits the pallet the claw C will be drawn away 

 from F, and the counterweight G will reftore the original pofition of the wheel, which 

 will receive and hold the tooth M during the whole of the vibration made in the direftion 

 BC. The next vibration in the oppofite diredtion will difplace the detent as before, and 

 by thefe alternations the motions will continue fo long as the maintaining power continues 

 to aft. 



In the two laft efcapements we have feen the variable efFe£ls of the maintaining power 

 ali^ft entirely removed, as far as can be pradically difcerned. Fig. 7. * exhibits the 

 efcapement of Mudge, in which the balance is perfe<31y detached from the train of wheels, 

 except during the extremely (hort interval of ftriking out the parts which ferve the purpofe 

 of detents. O N E B Q_ is the circumference of the balance; vibrating by the a£lion of a 

 fpiral fpring as ufual on its axis C A D H paflfmg through the centre C : the axis is 

 bended into a crank, A X Y D, to make room for the other work. L M, Z W, are two 

 rods fixed to the crank at the points L and Z, parallel to X Y. c d e f r s are fixed parts of 

 the machine. T R is an axis concentric with that of the balance, and carrying an arm 

 G o nearly at right angles to it, and a fmall auxiliary fpring u, which is wound up when- 

 ever the arm G o is moved in the direftion oh. p is a curved pallet fixed to the axis T R, 

 which receives the tooth of the balance wheel near the axis. The tooth, proceeding along 

 the curved furface, by the force of the main fpring turns the axis and its arm Go,. and 

 winds up the fpring u. A fmall projeftion at the extremity of the curved furface of the 

 pallet p prevents the farther progrefs of the tooth, when the arm o G has been turned 

 thfough an arc oh, of about 27O; and, confequently, the fpring u has been wound up 



through the fame angle or arc, Gh= 27°. F S is another axis exaftly Cmilar to TR. 



It canics its arm I o, and fpring v, and the tooth of the balance wheel 1 m winds up the 



• From Atwood's lovcfiigations for determining the Times of Vibration of Watch Balances. Philof. 

 Tranf. 1794. 



fpring 



