174 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1940 



scale. However, the experimental difficulties incident to the practical 

 realization of the thermodynamic scale by means of the gas ther- 

 mometer and the consequent discrepancies that arose in the tem- 

 perature scales used by different nations led the national laboratories 

 of Germany, Great Britain, and the United States in 1911 to under- 

 take the unification of their temperature scales. This culminated a 

 quarter of a century later in the adoption by the International 

 Committee of Weights and Measures of the International Tempera- 

 ture Scale (5). 



This scale conformed with the thermodynamic scale as closely as 

 knowledge permitted at the time, and was designed to be definite, 

 conveniently and accurately reproducible, and to provide means for 

 uniquely determining any temperature within the range of the 

 scale. It is based upon six fixed and reproducible equilibrium tem- 

 peratures to which numerical values are assigned; namely, the oxy- 

 gen point (-183° C), the ice point (0° C), the steam point (100°), 

 the sulfur point (444°), the silver point (960.5°), and the gold 

 point (1063°). Intermediate temperatures are determined by means 

 of interpolation instruments calibrated according to a specified pro- 

 cedure at the fixed temperatures. 



From the oxygen point to 660° C. the temperature t is deduced 

 from the resistance of standard platinum resistance-thermometers; 

 from 660° C. to the gold point, by means of a standard platinum 

 versus platinum-rhodium thermocouple; and above the gold point, 

 from Wien's law by means of the intensity of blackbody radiation, 

 measured by an optical pyrometer. 



The International Temperature Scale was adopted with the under- 

 standing that it was susceptible to revision and amendment. The 

 first meeting for the consideration of possible revisions will be called 

 in Paris in June 1939. 



There is great need for the official extension of the scale from the 

 oxygen point down to the triple point of normal hydrogen 

 (—260° C). At present the National Bureau of Standards main- 

 tains the scale in this region by means of a group of well-seasoned 

 resistance thermometers. Hoge and Brickwedde (6) have shown the 

 feasibility of establishing a number of fixed points m this part of 

 the scale, represented by the equilibrium states of coexisting phases 

 of pure substances. With these points once established, resistance- 

 thermometers need be used only for interpolation. 



Another matter that requires consideration is the value of C2 in 

 the radiation equation. The value now specified in defining the 

 International Scale is 6?2= 1.432 cm. degrees. More recent determi- 

 nations indicate that this value should be increased to 1.436 cm. 

 degrees. This would decrease the present value of the platinum 

 point by about 3° C. 



