204 Dr Brewster on a Nezo Analysis of' Solar Light. 



of the red space of an orange tint, that is, containing yellow 

 light. In support of the opinion that there are yellow rays 

 in every part of the red space, I may adduce a casual experi- 

 ment of Sir W. HerschePs, Phil. Transactions, 1800, vol. xc. 

 p. 255, when he had occasion to view the prismatic spectrum 

 reflected from clear turned brass. " The colour of the brass," 

 says he " makes the red rays appear like orange, and the orange 

 colour is likewise different from what it ought to be."" From 

 these observations it follows, that yellow light may be traced 

 through all the coloured spaces except the violet, where I have 

 not yet been able to find it ; but this is not surprising when 

 we consider the great faintness of the violet rays, and the faci- 

 lity with which they are absorbed by media of almost all co- 

 lours. Even the deep blue ammonio-sulphate absorbs almost 

 the whole of the violet space, and the smalt blue glass nearly 

 one-half of it, so that it is extremely difficult to subject it to 

 the partial action of absorbent media. 



It is obvious, even to the eye, that blue light exists in the 

 violet, indigo, blue, and green spaces, which occupy 247 parts 

 out of 860, or more than two-thirds of the whole spectrum. 

 When the most refrangible rays are absorbed by certain thick- 

 nesses of balsam of sulphur, balsam of Peru, pitch, or red mica, 

 the blue mixed with yellow, and forming green, may be traced 

 verv near the line C of Fraunhofer, which is considerably with- 

 in the red space. That the blue extends over the whole red 

 space is proved by the same arguments which we used for yel- 

 low light ; for when the red space is made of an orange tinge 

 by the absorptive action of certain yellow, orange, and green 

 media, this change can be effected only by the absorption of 

 blue light. 



Having thus proved that red, yellow, and blue light exist in 

 almost every part of the spectrum, I shall proceed to corrobo- 

 rate these views by showing that white light may be actually 

 insulated in different parts of it. 



When we look at the spectrum through a particular blue 

 glass of a certain thickness, we insulate the yellow space, the 

 colour of which is a rich gamboge-yellow. By increasing the 

 thickness of the glass, this compound yellow acquires the pale 

 straw-yellow colour of the yellow monochromatic flame pro- 



