52 EXPERIMENTS 0^' CHRONOMETERS. 



thofe of the that part of the face of the tooth ncareft (he point; now as re- 

 unJocking and ^,^^^^j ^^^^^^ ^^^j, ^^^^^ ^^ ^^.^jj ^^ fri^ion, 1 lake it that the 

 tne impiiiie. , r ... 



wheel wears more from this caufe (particularly where the ex- 

 ecution has been favourable ior it) tlian either of the other two 

 parts, and therefore raoft time-keepers made upon this conftruc- 

 tion have a greater tendency to gain than they have to lofe. 

 The advantage If it (liould be admitted that percuffion will wear in any ma- 

 th hies wi°h Jhort *^^'^^ degree, I fubmit that it will lead to the difcovering why 

 vibrations de- clocks whofe pendulums make the thorteft vibrations go better 

 fm"ltf °eT *^ ^'^^" others, and alfo the caufe of the great difparity there is 

 cuffion. between the going of the common verge watch and that of table 



and long eight-day clocks; in both clocks and watches con- 

 ftruded with two pallets upon the recoiling principle, each 

 pallet, with the accumulated force it has obtained by the vibra- 

 tion of the balance or pendulum, meets the wheel which is 

 coming a contrary way when it comes into contact, and a per- 

 cuffion takes place; therefore the longer the vibration of the 

 pendulum or balance, the greater muft this percuffion be, 

 and confequentiy the wear alfo: hence may be fuppofed one 

 reafon why fpring or table clocks that make fliorter vibrations 

 go better than verge watches, and long clocks, which make 

 Somcobjeftions fliorter vibrations ftill, go better than either. I readily admit 

 there is one, and perhaps many weighty obje61ions to this rea- 

 foning, viz. that fome clocks upon the recoiling principle are 

 found to go as well or even better than others on the principle 

 of the dead beat, though the execution in each has been of 

 equal excellence. For an anfvver to this objection, I refer the 

 reader to Mr. Cumrning's elaborate treatife upon clock mak- 

 ing; who in fome part of the work, aOTerts, the reafon to be, 

 that the plane of a6lion of the pallet does not fufficiently fub- 

 tend the angle of vibration ; by which means the power applied 

 i;; adminiftered too fuddenly in the dead beat ; to which I add 

 that the percuffion is alfo greater in this cafe in the dead beat 

 than in the recoil, although the pallet does not meet the wheel 

 in oppofition, and the wearing of the teeth of the wheel may 

 be increafed by that means. There will be among many others 

 one very ferious objeflion to my fuppofition that the wearing 

 of the wheel can produce error in along pendulum clock, viz. 

 the power of the pendulum being fo much fuperior to the mo- 

 tive force. It cannot be affeded by it ; to which I ftiall only ob- 

 ferve that the refults in pra6lice are foraetimes very different 

 from theoretical, conclufions. 



The 



