COMPARISON OF PLATINUM AND GAS THERMOMETERS. 



89 



First Comparison*. (Platinum thermometers, K.8 and K.9, with " verre dur " 

 gas thermometer.) The comparisons numbered 70 to 72 in the table for K.8, and 

 25 to 32 in that for K.9 were made during the summer of 1898 with the 

 thermometric reservoir of " verre dur." The following table shows the sequence of 

 the various operations. 



The comparisons all being subsequent to the heating of the reservoir to 500, the 

 nitrogen temperatures have been calculated, assximing for the initial pressure the 

 mean of the observations made after the comparisons, viz. : 



H = 534-305 millims. 



It will be noticed that this value differs by T42 millims.* from the initial pressure 

 observed before the comparisons given previously ; this increase is obviously due to 

 the contraction of the reservoir by the annealing effect of the high temperatures. As 

 the reservoir had been heated to about 500 before the observations, we concluded 

 that the contraction was produced entirely before the first measurements. 



* 



XXXIX. DETERMINATION OF THE CONSTANTS OF THE NEW GAS THERMOMETER 



WITH PORCELAIN RESERVOIR. 



We have already described the porcelain reservoir and the way it is connected to 

 the manometer tube. It remains to indicate the method by which we have measured 



* This variation of pressure corresponds to 0"'7 C. 

 VOL. CXCIV. A. N 



