392 Dr. J. Joly. On the Specific Heats [Apr. 



II. "On the Specific Heats of Gases at Constant Volui 

 Part III. The Specific Heat of Carbon Dioxide as a Fn 

 tion of Temperature." By J. JOLY, M.A., Sc.D., F.l 

 Received March 9, 1894. 



(Abstract.) 



In order to investigate the question of the variation of the 

 heat of carbon dioxide with temperature, a steam calorimeter 

 constructed having double walls of thin brass, between which 

 vapour of a liquid boiling under atmospheric pressure could 

 circulated. The vessels used in the experiments were hung in 

 closed inner chamber. Into this chamber steam could be admit 

 .after the temperature had become stationary and the same as that 

 the jacketting vapour. In this way the initial temperature could 

 varied. 



Experiments at various densities and over four intervals of 

 perature were carried out. The densities chosen were 

 0-0800; 0-1240; 0-1800, and 0'1973. The intervals of temperat 

 over which the gas at each density was investigated were : 

 temperature to 100 ; 35 C. (boiling point of ether) to 100 ; 

 (boiling point of acetone) to 100, and 78 (boiling point of eth 

 alcohol) to 100. 



The results are plotted on 5 equi-density lines, and it was fom 

 most intelligible to plot the precipitation due to the calori 

 capacity of the gas between ti and 100 against the initial temper 

 ture ti in each case. If the initial temperatures are measured 

 abscissas, starting from G. at the origin, and the precipitations a 

 set off as ordinates, the 5 equi-density lines so determined slo 

 downwards to the right (the slope increasing with the densi 

 passing, necessarily, through the point 100 C. on the axis of tern 

 ature in each case. If the specific heat is invariable these are ri 

 lines. This proves to be sensibly the case for the lines p = O" 1 

 and p = 0'0800 ; those of lowest density. 



The next line, 0'124, is nearly rectilinear over the higher 

 but pursued from right to left it rises markedly at the lower 

 peratures ; thus indicating that the specific heat at constant vol 

 falls in value with increasing temperature. The line p = 0*1 

 and the one close above it, />= 0*1973, show this variation 

 markedly. Their variation below the critical temperature is com 

 cated by the presence of liquid. 



The following table contains the principal results. The col 

 headed w contains the weight in grams of the precipitation due to 



