270 M. Melloni o« the Radiations of Incandescent Bodies, 



I shall here observe frankly, that the optical complica- 

 tion (alluded to in this criticism) does take place, as may be 

 clearly proved; it is this very circumstance which led me to 

 detect the errors which had been committed in the determina- 

 tion of the maximum of temperature in spectra arising from 

 different substances. That there may remain no doubt on 

 this point, I will transcribe from ray memoir, communicated 

 to the Royal Academy on November 24, ISl'S, the part that 

 relates to tliis subject. 



" Let one of the three surfaces of an ordinary glass prism 

 be covered with a layer of India ink ; let it dry, and then 

 divide it into three equal portions at right angles to its axis. 

 Remove with a penknife the ink from the middle portion, and 

 also a band four or five millimetres wide on the sides of the 

 two lateral compartments, so that these two bands from which 

 the ink has been removed, may be on opposite sides, and 

 form by their junction with the central force a kind of Z. It 

 will be understood, that a solar beam issuing from a prism 

 thus arranged will produce three coloured images side by 

 side; the middle one is very luminous, —it arises from the 

 part of the prism from which all the ink is removed ; the two 

 others, which are much paler, arise from the lateral bands. 

 It will also be perceived, that the middle image or spectrum has 

 each of its extremities on the boundary of one of the extremities 

 of the lateral spectra ; and that when, for example, its red 

 extremity is in the same line with the red extremity of the left 

 spectrum, its violet extremity will be upon the same line as 

 the violet extremity of the right spectrum, and vice versa. As 

 to the other two extremities of the lateral spectra, they will 

 not be found corresponding to the extremities of the central 

 spectrum, but to some one of the interior colours, and they 

 will evidently be more distant as the width of the uncovered 

 bands is less in proportion to the width of the prism. In one 

 of my experiments made with an equilateral prism of crown 

 glass, the width of which was twenty-four millimetres, and that 

 of the lateral bands five, I found, at a distance of two metres, 

 that the red extremity of the left spectrum was upon the same 

 line as the upper part of the yellow belonging to the central 

 image, and the violet extremity of the right spectrum was 



and the results are precisely the same when the refracting face of the prism 

 is reduced to the smallest possible dimensions." 



" My analysis of the spectrum by absorption does not therefore indirectly 

 controvert the principle, that ' to a particular colour there ever belongs a 

 particular wave-len^'th, and to a particular wave-length there ever belongs 

 a particular colour,' as Dr. Draper states, in theoretical language, the well- 

 known proposition of Sir Isaac Newton ; but it direct/// controverts it, and 

 absolutely overturns iV."— Phil. Mag., June 1847, p. 462. (Note of M. de laR.) 



