and on the Elementary Colours of the Solar Spectrum, 267 



the different cases we wish to compare with one another, 

 measuring each time the distance from the lamp to the paper. 

 It is then evident that the values sought will be in the inverse 

 ratio of the numbers found. 



This method, invented by Bouguer to determine the relative 

 intensities of different luminous sources, and employed by 

 Draper to measure the quantities of light emitted by a strip 

 of platina brought to different degrees of incandescence, is the 

 only one by which we could hope for a successful result. 

 The method of the equality of shadows, well-known under the 

 name of Rumford's method, would have furnished in the re- 

 searches of the learned American uncertain data, on account 

 of the difficulty of establishing an exact comparison between 

 the accidental green tint introduced into the shadow enlight- 

 ened by the yellow rays of the lamp, and the red light emitted 

 by the ignited metal. 



As to the measures of the radiant heat, they were determined 

 by the aid of the thermo-multiplier ; that admirable instrument 

 which has revealed to science so many new properties of ca- 

 lorific radiations, and which still is rendering eminent services 

 in the hands of able chemists far beyond the Alps. Professor 

 Draper had only to arrange at a certain distance from his strip 

 of platina a thermo-electric pile, and to observe for each phase 

 of incandescence the deviation of the index of the galvanometer 

 to determine the quantities sought. In this manner he ob- 

 tained the numbers contained in the following table, divided 

 into three columns. The first of these columns indicates the 

 temperatui'e for each degree of the scale of dilatation, com- 

 mencing with the point of incandescence : the difference be- 

 tween each of the successive terms of this series is thus con- 

 stant, and equal to 115°. The second and third columns give 

 the corresponding quantities of light and heat. It is almost 

 superfluous to add, that the unity for light is entirely indepen- 

 dent of that for heat, and that the similar independent unities 

 are not referable to the same point of the scale. 



