invented by Le Roy of Pari't, et 



the prcflure of the main fpring the wheel (fig. 4) is urged from D towards F, but is pre- 

 vented from moving by the pinG. Let the balance be made to vibrate, and the lever F will 

 move through t!ie arc Ff, ftrike the inner extremity of the fpring I, and difplace the pin G. 

 At tills inftant the face E, which may be called the pallet, will have arrived at the pofition e, 

 againft which the tooth of the wheel will fall, and communicate its impulfe through about 

 150 or 16° of the vibration. But F quits the fpring 1 fooner than the wheel quits the 

 pallet E, and confequently the pin G will have returned to its firft ftation before the wheel 

 can have advanced a whole tooth, and the fpring or detent HG will receive the wheel as 

 before, immediately after its efcape from the pallet. The returning vibration of the ba- 

 lance will be made with the piece EL perfedlly at liberty between two teeth of the wheel, 

 as in the (ketch, and the back (Iroke of the lever F againft the tender fpring I, will have no 

 efFeft whatever on the pin G ; this fpring being like the batk fpring of the jacks of the harp- 

 fichord, aftive in one direcSlion only. The third vibration of the balance will unlock the 

 detent as before ; the impulfe will again be given, and the whole procefs will be repeated j 

 and in this manner, the balance, though it may vibrate through the greateft part of the 

 entire circle, will be entirely free of the works, except during the very fmali time of the 

 drop of the wheel*. 



It is hardly neceflary to make any remark on this efcapement. It requires little or no 

 oil ; and when all the parts, particularly the pendulum fpring, are duly adjufted, it is found 

 that a very great variation in the firft mover will remarkably alter the arc of vibration with- 

 out affe£ling the rate. The piece EL might have confifted of a fingle pallet or arm, inftead 

 of a portion of a circle or cylinder ; but fuch a piece would have been rather lefs conve- 

 nient to make in fapphire, or ruby, as in the beft time-pieces, and would alfo have been 

 lefsufeful. For if by any accident or (hock the pin G fhould be difplaced for an inft^it, 

 the wheel D will not run down, bccaufe it will be caught upon the circular furface of EL. 



•The cchappement a detente of Le Roy (Recueil cfes Machines, &c. VII. 385) was different from the above 

 defcribed in feveral particulars not effemial to the general principle. i.The wheel was contrate, and in the 

 drawing has 44. thin teeth. 2. The face E was afforded by an aftual pallet concave towards the tooth. 

 3. The detent was forked, and moved on pivots at the angular point. One leg prefcnted a flop to one of the 

 teeth of the wheel, while the other leg refted againft the periphery of a femi-cylinder upon the verge of the 

 balance. The detent was kept to this pofition by a tender back fpring. 4, In the aftual performance the 

 vibration of the balance carried the femi-cylinder clear of the leg of the detent, which confequently would 

 have quitted the wheel,, if the back fpring had been ftrong enough. Immediately after the arrival at this no- 

 Ction, the pallet ftruck a tooth of the wheel, made it recoil, and delivered the detent, which flew inwardj 

 againft the radius of the femi-cylinder. The wheel being thus fet at liberty, was driven back through an 

 angle of near 20', as I eflimate from the drawing, and then in its return followed the balance through about 60"" 

 or 70", during which period the femi-cylinder recovered the pofition requifite to throw the detent in the way of 

 the fucceeding tooth, and retain the wheel, while the balance continued its vibration, difengaged from every ac- 

 tion but that of the rub of the femi-cylinder againft the tail of the detent. The judicious mechanic will per- 

 ceive that the difpofuion of the parts in the efcapement fig. 4. are much for the better. For (a) the flexures- 

 of fprings afford a lefs variable dedu£%ion from the moving force, than the fridlions of furfaces in the pivots of 

 Le Roy's detent, and his ftmi-cylinder. (b) The force loft in the recoil is greater and more variable than that 

 employed againft the fpiing I in fig. 4. (c)The balance is perfeftly difengaged for a much larger portion of both 

 vibrations in the modern, than in the original efcapement, and (d) the fmall number of teettuin the wheel D' 

 affords an advantage in regulating the figure aod C0iita£b of the pacts. 



It 



