on the Temperature of the Sun. 325 



In 1876 the French Academy of Sciences offered a prize 

 for the best solution of the question ; but although, at the ter- 

 mination of the competition, a prize was awarded to M. Violle, 

 and encouragements to MM. Vicaire and Crova, the Com- 

 mission of the Academy declared that the problem had not 

 been solved *. The Commission pointed out that the chief 

 source of error was the necessity of making use of an un- 

 trustworthy extrapolation in order to deduce the solar constant 

 to the limits of our atmosphere, and that even this was not 

 the least danger which necessarily follows from extending a 

 law of radiation, which is hardly applicable to temperatures 

 between 0° and 300°, to temperatures above the melting-point 

 of platinum. Thus, from almost identical observations on 

 solar radiation, Secchi obtained over 2,000,000° by dedu- 

 cing the temperature from Newton's formula, whilst M. Violle 

 obtained only 1500° by employing that of Dulong and Petit ; 

 nevertheless it has been shown that both the formulas are only 

 applicable when there is only a very small difference between 

 the temperature of the hot body which radiates and that of the 

 cold body which is warmed by the radiation. When the two 

 formulae are applied to the case of a body rendered incandes- 

 cent by the oxyhydrogen blowpipe, the temperature of which 

 is certainly as high as 2000°, Newton's gives 45,900°, which is 

 excessively high, whilst that of Dulong and Petit gives 870°, 

 which is certainly too low. M. Violle, in order to justify the 

 low value given by Dulong and Petit' s formula, attributes the 

 error to the emissive power of the incandescent body ; and, 

 assuming the formula to be exact and applicable even in this 

 case, he deduces from it an exceedingly small value for the 

 emissive power. 



It appears to me that, instead of forcing the formulae to show 

 what they can never do, it would be much better to confront the 

 question directly, and, by means of well-chosen experiments, 

 to ascertain the law according to which the intensity of radia- 

 tion varies when the temperature changes, to determine the 

 emissive power of the bodies on which the experiments are 

 made and under the conditions in which they are at the time 

 of the experiment, and, after having established the formula 

 which expresses the radiation within the limits of the expe- 

 riments which have served to fix it, to determine exactly 

 whether it is applicable to the cases of higher and well-known 

 temperatures. It is only when this correspondence exists that 

 the use of the formula can be further extended, and the tem- 

 peratures of inaccessible and exceedingly hot bodies like the 



* Comptes Rendus, vol. lxxxvi. p. 813 (1877). 

 Phil. Mag. S. 5. Vol. 8. No. 49, Oct. 1879. Z 



