520 Sir DAVID BREWSTER on the Lines of the Solar Spectrum, 



principal lines, and their intensity as depending on the thickness 

 of the absorbing medium, and the brightness of the spectrum ; I 

 was able to distinguish all such compounds, by merely looking 

 through them at a well formed spectrum. Even in those cases 

 where the eye could recognise no difference between the colours 

 of two substances that exercised different specific actions upon 

 light, their discrimination was instantly effected by viewing them 

 through a standard coloured medium. 



As some of these bodies attacked the spectrum at two, three, 

 four, and even five or more points at once, it became probable that 

 the number and intensity of such actions depended on the num- 

 ber and nature of the elements which entered into the compo- 

 sition of the body, or, what is nearly the same thing, that it was 

 the sum of all the separate actions of such elements ; and hence 

 the next step in the inquiry was, to determine the action of ele- 

 mentary bodies on the solar spectrum. This inquiry was not li- 

 mited to coloured bodies, for it is quite possible that a body may 

 transmit light perfectly white, and yet exercise a definite action 

 in absorbing various parts of the spectrum. The only physical 

 condition which is necessary in this case is, that the sum of all 

 the rays thus absorbed shall constitute white light. 



The first substances which I examined were Sulphur and 

 Iodine vapour. The Sulphur attacked the violet end of the spec- 

 trum with great force, and, when combined with arsenic, in the 

 form of native orpiment, its absorptive power for the same colours 

 was greatly increased. Even with the thinnest film that I could 

 detach, and not exceeding the two-hundredth part of an inch, the 

 spectrum was, as it were, cut sharply in two near the boundary 

 of the Green and Indigo spaces, and this body possessed the very 

 uncommon property of having nearly the same colour at small as 

 at great thicknesses. By increasing the thickness, the absorption 

 advances almost imperceptibly from the remaining blue border, 

 and if the transparency continued, the transmitted light would 

 certainly become red at great thicknesses, a property which may 



