3^4 



The Comparison of Ther7nometers. 



The column Pt. Corr. contains the readings of the 

 platinum thermometer reduced to air thermometer degrees 

 by Callendar's formula, viz : — 





where I was 1*57, agreeing with the number found by 

 Callendar and Griffiths. 



The following table contains the results of a comparison 

 between the readings of {a) the Platinum Thermometer 

 {U) the Mercurial Thermometer, {c) the number from 

 Regnault's Tables (calculated by Broch, Trav. Bur. Folds et 

 Mes. /.) 













Error by 



a. 



b. 



c. 



c—b. 



c — a. 



aireet 

 calibration. 





100-17 



100-55 



+ •38 





+ •38 





lOO'OO 



100-398 



+ •398 







95-08 



94-83 



95'i3 



+ -30 



+ •25 



+ •35 



92-30 



92-00 



92-33 



^'ZZ 



+ •30 



"^'ZZ 



86-26 



86-00 



86-25 



+ •25 



+ -26 



+ -30 



78-98 



78-67 



78-92 



+ -25 



+ •31 



+ ^26 



These numbers show that as far as the experiments 

 have been carried the method is accurate within o-°i. 



If a greater degree of accuracy be desired, the coils may 

 be kept at a constant temperature by means of a 

 thermostat, which is the method used by Griffiths, or the 

 measurements may be made with a slide meter bridge, the 

 comparison being made with reference to coils immersed in 

 water. It would also be advantageous to increase the 

 resistance of the thermometer. 



