in Ms Paper " On the History of Spectrum Analysis" 359 



hold for dark heat. An explanation of these results requires us 

 to admit that luminous rays shall in all cases produce heat by 

 their absorption, and I have been informed by an eminent phy- 

 sicist of this country that he considers this to be one of the 

 best proofs yet given of the identity between light and heat. 

 Now, if we assume this identity as proved, the fact that heated 

 coloured glasses invariably lose their colour in the fire, seems to 

 afford a perfect experimental proof of the equality between 

 absorption and radiation for every description of heat ; for the 

 conditions of the experiment evidently secure an equality of 

 temperature between the dark body and the transparent one, 

 while the eye becomes the best possible judge of the equality 

 between radiation and absorption for every individual ray. I 

 shall not here enter upon the question to what extent Kirchhoff 

 in his demonstration had previously been forestalled by Provos- 

 taye ; I may remark, however, that the proof of the Heidelberg 

 Professor is so very elaborate that I fear it has found few readers 

 either in his own country or in this, I conceive, besides, that 

 it is an objection to the completeness of this proof that the sub- 

 ject of internal radiation is not investigated; and I think, more- 

 over, that the experimental support which is here derived from 

 light is not so unobjectionable as that afforded by those rays 

 which are universally acknowledged to produce heat by their 

 absorption. 



I confess, however, that in new problems I do not attach the 

 same extreme importance as the Heidelberg Professor to logical 

 completeness of demonstration; for if the object of the investi- 

 gator be to hasten on the progress of human knowledge, when 

 such a one has matured a sufficiently good demonstration which 

 he has well supported by experiments, it is surely neither advi- 

 sable nor is it right that he should any longer defer to publish 

 the results which he has obtained. 



Although I preceded Kirchhoff nearly two years in my demon- 

 stration, I did not hesitate to acknowledge that his solution 

 had been independently obtained; but, as a general principle, I 

 cannot consent to admit that when a man of science has proved 

 a new law and is followed by another who from the same pre- 

 mises deduces the same conclusion, the latter is justified in de- 

 preciating the labours of the former because he conceives that 

 his own solution is more complete. 



Will Kirchhoff himself willingly forego his own claims in 

 favour of any one who shall in future ages devise (if this be 

 possible) a simpler and more convincing demonstration than 

 that which has been given us by the Heidelberg Professor ? I 

 feel, Sir, that, as an historian of science, you will acknowledge 

 the justice of these remarks, and join with me in regretting that 



2B2 



