BUNSEN AND KIRCHHOFF 329 



Kirchhoff 's law thus stands together with these facts of ex- 

 perience, and with the mass of other experimental facts col- 

 lected together in the theory of heat, and is thus itself as well 

 secured as all these experimental facts themselves. The law 

 states that the emission and absorption of light of definite 

 wave-length are proportional to one another in the case of all 

 bodies at the same temperature, A body - such as sodium 

 vapour, for example - that emits the yellow light cor- 

 responding to a definite position in the spectrum, and so of 

 definite wave-length, will also absorb the same light to a cor- 

 responding degree, and if it does not emit other kinds of 

 light, it will also not absorb them. This means that a flame 

 fed with sodium and placed between a very bright white 

 source of light and the prism apparatus, must likewise 

 cause a dark line to appear in the same position as the bright 

 line given by the flame. Kirchhoff himself observed that 

 this is actually the case. This provided a special experi- 

 mental foundation to the law which in Kirchhoff 's case pre- 

 ceded its proof. 



But at the same time an artificial Fraunhofer line had been 

 produced in the spectrum, at exactly the same position as 

 the line named the D line by Fraunhofer in the spectrum of 



Gesammelte Abhandlungen, p. 573) this remark is of no importance, but 

 the same is not true regarding the fact that all such proofs of natural laws 

 have no more and no less value than that of demon?;trating convincingly 

 a connection between the new laws to be proved and one or more othei 

 laws already regarded as certain, this connection being of such a kind that 

 the new and old laws now stand or fall together, if nature is to remain 

 true to itself (which assumption has always been found to be justified in 

 all non-living matter). In so far as the old laws made use of have been 

 ascertained by experience to hold, that is to say to agree with reality, 

 the same must also be true, in view of the proof, as regards the new law. 

 It is a matter of complete indifference how much or how little calculation 

 is made use of in constructing the connection in question. If there is much 

 calculation, and if at the same time the laws of experience made use of are 

 not very clearly set forth, we get the deceptive appearance of a 'mathemat- 

 ical proof,' which appearance obscures and hides the high character- 

 forming value of scientific investigation, by leaning to the opinion that its 

 true content is only accessible to mathematicians, or that mathematicsof 

 itself leads to knowledge of nature. 



