224 ELECTRICAL MEASUREMENTS 



resistance. This is facilitated by the linear relation between 

 temperature and resistance; for, assume that this relation holds 

 for all temperature intervals, and prolong the line connecting 

 temperature and resistance downward. It will cut the axis of 

 temperatures to the left of the origin at a temperature -- T. 

 Now let R tl and R tt be the resistances of the windings at tem- 

 peratures ti and tz, then 



R tt R tl _ Rt l 



or 



t 2 - tl = t$*J*J [T + t,]- (27) 



Again from the above, 



7^ = 1 ~T~ } 



The quantity T is usually called the " inferred absolute zero.' 

 It may be calculated from the data given in the table. For 

 example, 



D r> f -t i [j OftT\ 



Kt -tC20\J- I C*20| J\) 



if the resistance becomes zero and 100 per cent, conductivity be 

 taken. 



t - 20 = - = - n n * = - 254.5; 

 a 20 0.00393 



.'. T = - 234.5. 



These values are entered in the last column of the table on 

 page 223. When used in the above formulae the minus sign is 

 omitted. 



The Resistance Pyrometer. 13 Following a suggestion made 

 in 1871 by W. Siemens, the variation of the electrical resist- 

 ance of platinum with temperature is utilized in pyrometry. 

 The first experiments in this direction were not successful, and 

 in 1874 the British Associaton report on the instruments sub- 

 mitted was not favorable, it being found that after exposure to a 

 high temperature the platinum coils did not return to their 

 original resistances and that the changes were progressive. In 

 1886 Callendar proved that these changes were due to the ab- 

 sorption by the platinum of silica from the coil support and of 



