CARBON DIOXIDE AT DIFFERENT TEMPERATURES. 



229 



values of Q for the largest, smallest, and intermediate flows respectively. The fifth 

 column gives the observed values of CE/80 for the intermediate flows, reduced to 

 the values they would have had if the experiments which they represent had been 

 performed at exactly the same temperature as those to which the corresponding 

 largest and smallest flows are referred. The sixth column gives the values of CE/S0 

 calculated by using the values of S and h obtained for the corresponding largest and 

 smallest flows. The last column gives the percentage difference between the observed 

 and calculated values of CE/S0 the percentage being reckoned in the largest value of 

 CE/S0 in the particular set of observations in question : 



AIR Temperature. 



STEAM Temperature. 



To the above mean values of the specific heats a few corrections remain to be 

 applied 



(1) The corrections discussed on pp. 211-215. The values of the corresponding 

 rates of flow were practically the same for all the final experiments from which the 

 specific heats were calculated, so that the correction is one which may conveniently 

 be applied to the mean value of the specific heat. Taking the mean value of 

 (0'3) A 2 /JSQ! . JSQ 2 for all the experiments at 20 C., we find that the correction to be 

 applied to the specific heat at this temperature is additive and amounts to 1 '3 parts 

 per 1000. For the experiments at 100 C. the corresponding correction amounts to 

 3 '3 parts per 1000. In the latter case, however, we have the correction mentioned 

 on p. 214 (lines 7 et seq.), which is subtractive and amounts to 6 parts in 10,000, so 

 that the two corrections together amount to 27 parts per 1000. 



(2) A correction for the fact that, in the calibration of the tubes, a linear relation 

 between PV and 6 was assumed to hold over the range of temperature C. to 17 C. 



