foote: luminosity and temperature of metals 323 



The interesting point is that the color temperature of plati- 

 num at the melting point must be 80° and possibly 120° higher 

 than the true temperature. This is at variance with the con- 

 clusions drawn by Paterson and Dudding, 8 who claim that plati- 

 num is practically gray thus requiring that p = and T = T' . 



Conclusion. The various relations between color temperature, 

 apparent temperature, and true temperature are pointed out 

 for the first time. These relations are checked by observations 

 of Hyde, Cady, and Forsythe on tungsten and are found to agree 

 excellently with experiment. Color temperatures open a new 

 field in the subject of optical pyrometry which is certain to prove 

 highly interesting. Color temperature may be measured by the 

 spectrophotometer using the method of logarithmic isochro- 

 matics, or by an ordinary photometer if the observer is skilled 

 in color matching. A still better means perhaps of studying 

 the color temperatures of various materials is by use of a suit- 

 able form of colorimeter. Measurement of the dominant hue 

 of a black body and of metals at various temperatures is a field 

 of investigation as yet untouched. 



PHYSICS. — Luminosity and temperature of metals. Paul D. 

 Foote, Bureau of Standards. 



, Recently Mr. Fairchild and the writer published a paper on 

 the relation between the luminosity and temperature of a black 

 body. 1 In the present note this work is extended to apply to 

 the case of radiation from metals, and from oxides for which 

 the emissivity coefficient can be represented by equation (4) 

 below. Luminosity of a black body is defined as the integral 

 from to °° in respect to d\ of the product of visibility (V) 

 and energy (J) of the radiating source thus: 



(1) L = J* o " VJdx 



Cz 



where V = f (X) and J = i\ X 5 e *d = Wien's law 



8 Proc. Phys. Soc. London, 27: 253. 1915. 



1 Bureau of Standards Sci. Paper No. 270. 1916. 



