210 Baron Fourier's Historical Notice 



by observing with precision the changes in the volume of a 

 soHd metal. Graham and Peter I^eroy * had succeeded in 

 procuring pendulums of an invariable length, by the compen- 

 sation of the unequal expansions of two different metals. 

 Harrison is the first, if T am not mistaken, who proposed to 

 employ a plate formed of two others, inequally expansible, 

 and fixed together in all their points. Ingenious applications 

 have been made of this method in measuring the degrees of 

 heat, and in making the vibrations of a balance isochronous, 

 notwithstanding the variations of temperature. This inven- 

 tion has been greatly improved by M. Breguet. He has em- 

 ployed it in making a thermometer much more quick and suit- 

 able than those which had hitherto been used. The compound 

 plate is composed of platina, gold, and silver. Its total thick- 

 ness is only the fiftieth or the hundredth of a line, and it has 

 the form of a spiral. One of its extremities is fixed, and the 

 other, which is free, and of extreme mobility, carries the index 

 which points out the temperature. The sudden and succes- 

 sive variations in the heat of the air show themselves as rapidly 

 as they could be felt by a living being. Effects of this kind, 

 which other thermometers indicate slowly and slightly, thus 

 become instantaneous, and much enlarged. 



Other researches of the same artist have enabled us to mea- 

 sure with extreme precision the duration of phenomena. He 

 introduces, for example, into an astronomical telescope a chro- 

 nometer, whose hands follow the motions of the star in the 

 field of view, and one may count the tenths, and even the 

 hundredths of a second. But what is particularly remarkable 

 in this instrument is the perfect continuity of the motion of 

 the hands. He has also constructed watches, the needle of 

 which marks suddenly, and at pleasure, a visible point upon 



* Why is Leroy mentioned here along with Graham, without the 

 reader being told that Graham is the original inventor of the compensation 

 pendulum ? His mercurial pendulum, the best yet known, was invented 

 before 1726, whereas Leroy 's comparatively clumsy contrivance was, ac- 

 cording to Biot, (TraiU de Physique, i. p. 173,) invented in 1738, and 

 has besides no resemblance whatever to the beautiful method of Graham. 

 The association of Leroy 's name with that of Graham is no more called 

 for than that of Ellicott, Cuming, and Smeaton, who afterwards proposed 

 new methods of compensation. — Ed. 



