[ 421 ] 



XLII. A Comparison of Platinum Thermometers of different 

 degrees of Purity. By H. M. Tory, M.A., Mathematical 

 Lecturer at McGill College, Montreal*. 



1. rf^HIS investigation was undertaken at the suggestion of 

 J_ Professor Callendar with a view to determining the 

 probable order of accuracy in the determination of high 

 temperatures attainable by the use of ordinary commercial 

 specimens of platinum wire. It was also desired to observe 

 the effect of large variations in the purity of the wire, in 

 order to be able to estimate the probable effect of such small 

 impurities as were likely to occur in pure wires specially 

 selected for pyrometry. Five wires in all were compared 

 over the range 400° to 1000° C. The fundamental coefficients 

 of the wires (which may be taken as an indication of the 

 purity) varied from "003892 to '002340, i. e. by 40 per cent, 

 of the maximum value. It was found, however, that the 

 extreme variations of the temperatures observed, when calcu- 

 lated on the platinum scale by the formula 



^ = 100(R-R°)/(R / -R o ), . . . . (1) 



in which the letters Pt, B/, and 11° stand for the observed 

 resistance at the temperatures pt, 100°, and0° C. respectively, 

 did not in any case exceed 9° C. at 1000° C, or were less than 

 one per cent, on the interval at any point of the range of 

 observation, in spite of the large differences in the samples. 



When the temperatures were reduced by means of the 

 difference-formula 



t-pt = dt(t -100)/10,000, .... (2) 



although the agreement was found to be exceedingly 

 close at temperatures up to 500° C. (if the value of the 

 difference-coefficient d was calculated by assuming the boiling- 

 point of sulphur (S.B.P.) to be 444 0, 53 C, by the method 

 proposed by Callendar and Griffiths), the discrepancies at 

 1000° were still of the order of 5° or 10° C. We must con- 

 clude that this method of reduction by reference to the 

 S.B.P. will not give results of the highest accuracy if 

 applied to impure wires at 1000° C. This point has also been 

 illustrated by Callendar (Phil. Mag. Feb. 1899). It was 

 found, however, that very fair agreement could be obtained 

 at high temperatures by using the freezing-point of silver 

 (AgF.P.) as a secondary fixed point instead of the S.B.P. 

 in the calculation of the difference-coefficient. For two of 

 the wires the simple method of reduction employed by 

 Heycock and Neville (Journ. Chem. Soc. Feb. 1895, and 



* Communicated by the Physical Society : read June 22, 1900. 

 Phil. Mag. S. 5. Vol. 50. No. 305. Oct. 1900. 2 G 



