452 mi. CHARLES H. LEES ON THE KFFI-XT <>F TKMI'KIIATURE ON THK 



To carry out the conversion on the above assumption we have, since 



dx dx 

 from which the following table has been calculated 



-,,)}. 



Measurement of Temperature. 



The most important of the temperature observations was that of the difference of 

 temperature between the two platinum spirals. 



Since the copper wires connecting the two spirals to the resistance bridge were 

 of the same length, the resistance it was necessary to place in series with one 

 of the coils when, after an adjustment to a balance against two equal resistances, the 

 other coil was heated, was equal to the difference of the resistance of the two coils. 

 If the coils had been of exactly the same resistance when at the same temperature, 

 the difference of mean temperature of the coils would have been proportional to this 

 observed difference of resistance. As they were not quite equal, a second observation 

 was taken with the coil formerly cold now heated and the other cold. This is better 

 than taking an observation of the difference of resistance of the two coils when at the 

 same temperature, as it eliminates at the same time errors due to want of symmetry 

 in the positions of the coils and heaters in the cylinder of material tested. 



