358 Notices refpeSl'ing Ne-.u Boots, 



inflead of an iron tube filled with quickfilvcr, employed one 

 ofglafs, having at the end a bulb like the tube of a ther- 

 mometer. But ftill happier was Graham's idea, to conftru(Sl 

 a pendulum of feveral rods of different metals, fo combined, 

 that the expanfion of the one fhoutd be fully compenfatcd by 

 the contraction of the other; and this gave rife to the grid- 

 iron pendulum. Before Graham, however, Short, Caffini, 

 and Ellicot had entertained a fimilar idea of a compound 

 pendulum ; and the plan of a gridiron pendulum was firll car- 

 ried into execution by Harrifon in the year 1726. The au- 

 thor here gives a defcription of the pendulums propofed by 

 Graham and flarrifon; and mentions the experiments of 

 Berthoud, Greuier, and SvfFert, for improving them. He 

 fpeaks alfo of Ellicot's lever pendulum, Grenier's lever pen- 

 dulum, and the pendulum with the fmall gridiron ; as alfo 

 Eivaz's tubular pendulum ; and the fimpleil compenfation 

 pendulum of all, invented by Faggot, a Swede, about the 

 year 1740, and which Schmidt the watchmaker, of Stettin, 

 has lately employed with much ingenuity. The higheft de- 

 gree of perfection, however, to which watch-making has at- 

 tained, is in the conftruCtion of the nautical time-keepers, 

 for the invention of which large premiums were offered iii 

 England, France, Holland, and Spain. Huyghcns and 

 Sulley made attempts for this purpofe, but they were not 

 attended with fucccfs; and the ingenious propofals of Leib- 

 nitz were not found fufficient. Flarrifon's firft time-keeper, 

 •which he prefented to the Royal Society in 1736, was regu- 

 lated by balancing rods placed crofswife over each other, with 

 circular fprings at the ends, which relted againft two plates, 

 which by the dilatation of the circular fprings, in confe ;uenco 

 of heat, feparated, and on their contraction by cold approached 

 each other. The friftion alfo was leifcncd, and the time- 

 keeper was fufpended like the mariner's compafs. In a voyage 

 of twelve weeks, the error in going amounted only to 36 fe- 

 conds, A fecond one, conitrufted in 1749, which was 

 fmaller and more convenient, furpaiicd the firff. To a 

 third, conftruoled in 1753, he applied a balance with a fpiral 

 fpring, and a compenfation rod of brafs and fteel. In 1761 

 he conftrufted a fourth, which, on a voyage of experiment 

 that laftcd 81 days, erred only i minute 54'- feconds, A 

 fifth, fiuilhed in r764, erred only 54 feconds in fix weeks. 

 Dr. Maitelyne, however, to whom it was aftervi'ards re- 

 ferred for trial, did not give fo favourable a report of 

 it. Berthoud and Lc liny made attempts alfo to confiniCl 

 time-keepers; but the firtt were not fuccefsful. In the year 

 1741 Le Hoy's time-keeper, which in fix weeks erred only 

 half a degree, was preferred to that of Berthoud, which in the 



Jame 



