490 Sir David Brewster, in Iteplt^ to M. Melloni, 



merely pronounce tliem to be incorrect, and to have no real 

 bearing, even if correct, upon the views which they are 

 brought forward to overturn. 



M. Melloni asserts that he has repeated my ftindamcntal 

 experiment, which, lie says, " consists, as is well-known, in 

 interposing between the eye and the spectrum a slip of glass, 

 deeply coloured by the oxide of cobalt," &c. &c.; and he goes 

 on to describe the phenomena exhibited by a spectrum formed 

 from the light of a circular apertnre, ten millimetres or four- 

 tenths of an inch in diameter, and by means of an equilateral 

 prism in the position of minimum deviation. I never made 

 such an experiment, and never wotdd have thought of emploi/ing 

 a spectrum thus produced. The spectrum described by Fraun- 

 hofer was obtained from an aperture one-fjtieth of an inch 

 wide; Dr. Wollaston used an aperture one-twetitieth of an inch, 

 while M. Melloni uses oi\Q twenty-fiftieths, — txventy ixmtis wider 

 than Fraunhofer's aperture, and eight times wider than Dr. 

 Wollaston's ! In such a spectrum, therefore, the separation of 

 the colours must have been much less complete than in those 

 studied by Fraunhofer, Wollaston and myself ; &n6. i\\e yellow 

 and the red rays must have invaded the orange, as observed 

 by Melloni. As this commixture of rays did not appear when a 

 narrow stripe* of the prism was used, that is, in his elementary 

 spectrum, as he calls it, he concludes that my results were owing 

 to the use of a prism with a wide surface, and that I had there- 

 fore used a complex spectrum. I should have been ashamed 

 of my inexperience had I used such an aperture and such a 

 spectrum as that which Melloni employed; and I can assure 

 him that he has formed a very low estimate of the time and 

 labour which I have devoted to these researches, and of the 

 manner in which they were conducted. 



Dreading, as he justly does, that his own results might be 



* This narrow stripe, a " little more than a millimetre in width," was 

 formed in a layer of Indian ink on a face of the prism. As the light was 

 incident very obliquely on the prism, a millimetre, or the twenty-fifth of an 

 inch, would be reduced to nearly the fortieth of an inch when multiplied 

 by the cosine of the angle of incidence. Divergent light passing through so 

 narrow an aperture, would produce diffraction-fringes injurious to the di- 

 stinctness and purity of Melloni's elementary spectrum. I have repeated 

 my experiments with spectra much more distinct and pure than any that 

 can be produced by the contrivance of our author, and 1 have obtained 

 precisely the results contained in my original memoir. These spectra were 

 formed by the finest glass prisms, both single and compound, and by prisms 

 of rock salt, so perfectly homogeneous and colourless, that in looking into 

 the prism the substance of the salt is iiivisil)le. 1 have used prisms with 

 refracting angles of all magnitudes, and some of them so great that the blue 

 and violet rays did not emerge from the second surface, and in all the 

 spectra thus produced I have obtained the very same results. 



