172 Dr. H. F. Weber on the Specific Heat of 



ture T . It is only necessary to solve, in reference to T, the 

 equation 



W* o =II(T-T ). 



Let G grammes of platinum and G' grammes of another sub- 

 stance K be raised to the same temperature T, and be then cooled 

 to the common final temperature T in the same calorimeter; 

 from the amounts of heat, W P and W K , given out by the sub- 

 stances in cooling, it is possible to calculate the temperature T 

 to which both substances were heated, and also the average spe- 

 cific heat, Ct -tj between the temperatures T and T of the sub- 

 stance by solving the equations 



Wp=G.n(T-T ), 

 from which T— T may be deduced, and 



G'(T-T ) 



= VT.- r 



Pouillet* has carefully determined the amount of heat required 

 to raise a unit weight of platinum from T to T where T =0° and 



T 



T=1200°, and finds this quantity of heat, W T , to be the fol- 

 lowing function of the temperature — 



W^ o = 0-03237 (T - T ) + 0-0000041 (T 2 - T 2 ) . 



As the estimation was made with all care, I adopted it as the 

 basis for my determination of specific heats at high temperatures. 

 The recent researches of Deville and Troost, which show that 

 platinum at high temperature is permeable by gases, and that 

 the air in the air-thermometer used by Pouillet was therefore not 

 entirely enclosed during his experiments, but that a slow inter- 

 diffusion of the air without and within the instrument took place 

 through the platinum partition, have not been overlooked. In- 

 asmuch, however, as the air without and within the thermometer 

 was under the same circumstances of pressure and temperature, 

 and the platinum was separated from the gaseous products of 

 the flame employed by means of an iron muffle, any error which 

 may be introduced must be but infinitesimal f. 



Adopting, then, this determination of Pouillet's as a basis, the 

 estimation of the specific heat of carbon at high temperatures was 

 carried out as follows : — 



* Comptes Rendus, vol. iii. p. 782; and Traite de Physique, 6th edit, 

 vol. i. p. 227. 



t Weinhold's results (Pogg. Ann. vol. cxlix. p. 213) seem to be Dot so 

 trustworthy as Pouillet's ; he employed a very small quantity of platinum 

 (only 9 grms.), while Pouillet used 178 grms. 



