On the Distribution of Heat in the Spectrum. 105 



Foucault and Fizeau, in their well-known experiments on the 

 interference of heat, no one has made reference to these lines, 

 and that all experimenters describe the heat- curve as a conti- 

 nuous one (Phil. Mag. April 1872). 



1 may therefore be excused for remarking at this point, not 

 only that the three lines in question were observed by me nearly 

 thirty years ago, but that an engraving of them was published in 

 the Philosophical Magazine, in a memoir announcing the disco- 

 very of fixed lines in the invisible portions of the spectrum 

 (May 1843). It will be seen from an inspection of that engra- 

 ving that the lines are marked a, j3, y. They were impressed 

 on Daguerreotype plates by resorting to the well-known pro- 

 cesses for obtaining photographs of the less-refrangible regions 

 of the spectrum. 



In view of the preceding statements and others that might be 

 given, it may, I think, be affirmed that the general opinion held 

 at the present day as to the constitution of the spectrum is this 

 — that there exists a heat-spectrum in the less-refrangible re- 

 gions, a light-spectrum in the intermediate, and a spectrum 

 producing chemical action in the more refrangible regions. An 

 experimental attempt to correct this view, and to introduce a 

 more accurate interpretation of the constitution of the spectrum, 

 will not be without interest, especially as it is necessarily and 

 directly connected with the important subject of photometry. 

 In this memoir I shall offer some experim.ents and suggestions 

 respecting the heat of the spectrum, and in another, shortly to 

 be published, shall consider the distribution of the so-cailed 

 chemical rays. Among the numerous problems of actino-che- 

 mistry there are none more important than these. 



All the experiments hitherto made on the heat of the spectrum 

 have been on the principle of exposing a thermometer in the 

 differently coloured spaces. Such was Sir W. HerschePs me- 

 thod. Leslie used a differential with small bulbs. Melloni, 

 Mliller, Tyndall, a thermoelectric pile, the form preferred being 

 the linear. This was advanced successively through all the ra- 

 diations, and the deflections of the multiplier noted. 



Is not this method essentially defective ? Does it not neces- 

 sarily lead to incorrect results ? 



" There is an inherent defect in the prismatic spectrum, a 

 defect originating in the very cause which gives rise to that spec- 

 trum itself — unequal refrangibility. Of two groups of rays 

 compared together, one taken in the red the other in the violet 

 region, it is clear that in the same spectrum, from the very 

 circumstance of their greater refrangibility, those in the violet 

 will be relatively more separated from each other than those in 

 the red. The result of this increased separation in the more 



