EXPERIMENTAL K'NOWLEDGE OF THE PROPERTIES OF MATTER. 485 



temperature quoted the metal is very far from its melting or softening 

 point ; there is in fact, as these experiments, taken as a whole, show, a 

 gradual increase in specific heat, which at three or four hundred degrees 

 would remove the metal from the state with respect to specific heat 

 required by Dulong and Petit's law. 



He finds that silver melts at 907°, and that it3 specific heat before and 

 after fusion is the same. 



As in the case of silver, so in the case of iron, Pionchon refers the 

 absorption of heat and almost constant specific heat of iron at two tem- 

 peratures to some change of state in iron at those temperatures ; and he 

 thinks this idea confirmed by changes, at the neighbourhood of those tem- 

 peratures, of the sign of the specific heat of electricity of iron observed by 

 Thomson and by Tait. 



Tin melts at quite a low temperature, about 230° ; 232-7° as found by 

 Person ; ' and may be heated to a high temperature without volatilising. 

 Thus Pionchon was able to make a thorough investigation into the specific 

 heat of liquid tin from 2327° to 1110°. 



This investigation in respect of metals and Weber's on the specific 

 heats of carbon constitute the greatest advance in our knowledge of the 

 variation of specific heat of bodies in the solid state that has hitherto been 

 made ; for hitherto we had to be content with determinations made, as 

 those by Dulong and Petit, and afterwards by Bede, at temperatures up 

 to no higher than 350°. 



The calorimetric method which Pionchon used is founded on Pouillel'j 

 principle ^ of determining the temperature of a ball of platinum by means 

 of an air-thermometer, and the specific heat by letting the ball drop into a 

 water-calorimeter of sufficient depth. A formula was thus obtained by 

 Pouillet applicable to temperatures up to 1200° for the specific heat at any 

 temperature t ; thus total heat-capacity of platinum between 0° and t°, or 



2i = -03237i-|- -000041^2; 

 whence 



c =-03237 + -000082^ 



gives the specific heat at t° by the air-thermometer. 



Violle subsequently, in 1887, adopting Pouillet's method, found spe- 

 cific heats of platinum and other metals, using a more accurate air-ther- 

 mometer than Pouillet's ; ^ his more accurate results for platinum were 



2i=-0317i-t--000006i2; 

 whence 



c=-0317+-000012i, 



which is an equation from which t can be found when c is known. 



Pionchon in this memoir finds the melting-point of silver to be 907° as 

 against the temperature found by Violle, viz., 954°, which is much too 

 high. 



The Specific Heat of Water. 



The specific heat of water is required in the determinations of tempera- 

 tures by the use of the platinum ball and the water calorimeter ; this 

 was found by Regnault* to be l + -00004«-f -0000009^- for temperatures 

 between 0° and 230° by the air-thermometer. 



> Jahresb.f. Cliein. Fittica, 1847. " C. R. 3. 1836, p. 782. 



^ C. R. 1877, p. 543 ; 1878, p. 981 ; 1879, p. 702. 

 < Mem. de. VAcad. 21, p. 729. 



