Mesmrch on some Standards of Light. 



471 



Any solid substance that would not disintegrate at a temperature of 

 about 1700° C. might, a priori, be chosen as a radiator. It has, how- 

 ever, been shown that most of the oxides when maintained at these 

 high temperatures undergo a change in their emissive properties, and 

 cannot, therefore, be used for the purpose we have in view.* The 

 choice thus seems restricted in practice either to carbon or to one of the 

 metals of the platinum group. 



Before passing on to the experimental part of the work, it may be 

 well to recapitulate the necessary qualifications of a standard of light. 

 The requirements may be briefly summed up under three heads : — 



1. The standard must remain constant. 



The slow periodic variations of the amyl acetate lamp which extend 

 over a period of several months are as much to be avoided as the 

 flickerings of the candle or the hourly changes of the Carcel lamp. 



2. The standard should be reproducible. 



This condition is satisfied when the standards reconstituted by inde- 

 pendent investigators show no measurable variation between one another. 



3. The light emitted should be as nearly as possible of the same 

 spectral composition as that of the chief artificial lights now in common 

 use.t 



On the Intrinsic Brilliancy of the Crater of the Arc. 



For many years it had been noticed that the area of the crater of 

 an electric arc when burning between carbon poles increased about in 

 proportion to the current : also that the light emitted increased in the 

 same ratio as the area of the crater. % These and other facts led to the 

 conclusion that the temperature of the crater remained constant. The 

 hypothesis was put forward that this temperature was the boiling point 

 of carbon, this theory being supported by the experiments made in 

 1892 by J. Violle.§ In the same year it was proposed simultaneously 

 by Swinburne, S. P. Thompson, and Blondel,|| that the crater of the 

 arc should be used as a standard of light, Blondel publishing a series 

 of experiments illustrating the way in which the new standard might 

 be used. 



In 1894 A. Trotter"!! proved that when the arc is not silent the crater 



* Mchols and Crehore, ' Trans, of the Amer. Inst, of Electrical Eng.,' vol. 13, 

 p. 190, 1896. 



f Strictly speaking, it is only when two lights are of the same spectral composi- 

 tion that the ratio of their intensities can be expressed by a single figure. 



X Professor S. P. Thompson's Cantor Lectures, 1895 ; see also " The Electric 

 Arc," by Mrs. Ayrton, ' The Electrician," vol. 34, p. 399, 1895. 



§ ' Journ. de Physique,' 3 ser., vol. 2, p. 545, 1893, and 'Coinptes Eendus,' 1892, 

 p. 1274. 



|i See « Proc. of the Int. Electrical Congress at Chicago,' 1893, pp. 259, 267, 315. 

 T 'Roy. Soc. Proc.,' 1894, vol. 56, p. 262; see also "Effect of Pressure on the 

 Temperature of the Arc," E. Wilson and C. E. Fitzgerald, ' Eoy. Soc. Proc./ 



