On the Maintaining Power in Clocks and JVatches, 429 



j^lcalTranfackions, that he had added the fimple expedient of an horizontal gimbal-hoop to 

 the crofs cirtle which fupported the meridional pivots. Such a circle, divided, inftead of the 

 fixed horizon in his figure, would have rendered it a complete inftrument for general pur- 

 ppfes. It is probable however, that for ordinary naval ufe the fimplcr compaflt;s of the 

 pjefent conflrudlion would be preferable. 



VIII. 



On ike Alttititaining Pa-wer in Clodt and IVatches. 



I 



N a former paper on the compenfations for change of temperature in pendulums, a flight 

 view was taken ot the methods by which a train of wheels might be made to move either 

 uniformly or by equal intervals of progreffion, fo as to afford a mcafure for the lapfe of 

 time. The pendulum and the balance are at prefent the regulators which for juft reafons 

 are preferred to every other contrivance. If we fuppofe a body to be fufpended at the ex- 

 tremity of a ftring, its gravity, as is well known, will caufe the firing to point towards the 

 centre of the earth's gravitation at that part of the furface. If the body be then drawn afide 

 Out of this perpendicular polition, it is equally clear that it will be removed^ further from 

 the earth's centre, towards which, when again fet at liberty, it will fall through the arc of 

 a-circle, which is the only line the ftring will permit it to defcribe. At the inftant the per- 

 pendicular pofition is thus regained; the body will not be at reft, but will poffefs a por- 

 tion of velocity in a direction at right angles to that of gravity. It will not, therefore, be 

 in the power of gravitation to affe£t that velocity; and confequently the body will proceed 

 onwards on the oppofite fide of the perpendicular, in a fecond circular arc, which ele- 

 mentary writers demonftrate to be in theory perfeftly fimilar, and equal to the' arc of de- 

 fcent : — that is to fay, the aflion of gravity will more and more oppofe the afcending mo-- 

 tion, as its direction becomes more remote from the horizontal line, and will at length not 

 only deftroy the motion of afcent, but again generate a falling motion precifcly the fame 

 as the firft. Such a pendulum would, therefore, if no other circumftances prefented them- 

 felves than have yet been ftated, vibrate for ever ; and its vibrations would be performed in 

 times precifely equal. But the truth is, that fuch a body muft in aiflual experiment give 

 motion to the air through which it pafl'es ; and muft alfo overcome- a certain degree of 

 rigidity to which the ftring, or any other fubftitute in nature which might be propofed, is 

 fubje<fl. On thefe accounts, the quantity of motion ni the pendulous body will be con-- 

 tinually diminilhed, and at length become infenfible. It may alfo be eaGly apprehended, that 

 a pendulum, though conftrufted with fuch delicacy as to move for many hours before it 

 comes to apparent reft, could not be of much pra£lical utility without fome ready means to 

 keep an account of its vibrations. For both reafons, therefore, a pendulum without any 

 other addition can fcarccly be ufed as a time-keeper. 



A train of wheels, urged by a weight or fpring, is therefore attached to the pendulum 

 for thefe two purpofes, namely, to maintain its motion by a fmall impulfe given at each 

 vibration, or at moft at each fecond vibration, and to keep account of the number of times. 

 The principle of the balance is, that its fimilar vibrations arc performed in equal times, 

 like thofc of the pendulum. The methods of adapting the train of wheels to one of thefe 

 regulating organs wilt not therefore ciTcntially difkr fiom thole of conneAing. it with the 



other. 



