158 LIGHT AND PROTOPLASM [Cn. VII 



Finally, a fourth and decidedly practical way of obtaining 

 pure colors is by the use of transparent plates of colored glass or 

 other transparent solids. It is very difficult to get monochro- 

 matic glasses of certain colors in the market. A pure red is 

 easily obtainable ; the blue is apt to contain some red also ; and 

 the green, both blue and yellow. Lord RAYLEIGH ('81, p. 64) 

 has used " films of gelatine or of collodion, spread upon glass 

 and impregnated with various dyes, gelatine being chosen when 

 the dye is soluble in water and collodion when the dye is soluble 

 in alcohol." This method seems to me to be of wide applica- 

 bility in our light experiments. For solid media are, after all, 

 far less troublesome than fluids, vapors, or spectra ; and con- 

 venience is one of the most valuable qualities of .a method. 



A brief statement must be made concerning the physical 

 properties of the different light waves. An inspection of any 

 prismatic solar spectrum shows that certain parts are brighter 

 to our eyes than others, and a thermometer placed in different 

 parts of the spectrum indicates a higher temperature towards 

 the red end. Curves are given in Fig. 42 which show the 

 relative warmth of different parts of the visible spectrum both 

 when the spectrum is a normal one (i.e. such as is given by a 

 diffraction grating, where all rays differing in wave length by 

 0.1 /A are equally distant) and when it is prismatic (in which 

 there is a crowding of rays at the red end). Curves of relative 

 brightness and of relative chemical (actinic) activity, so far as 

 can be judged from the union of chlorine and hydrogen, are 

 also given, for the prismatic spectrum. Being laid off on the 

 " normal " scale the curves last mentioned are somewhat dis- 

 torted. From these curves it appears that the brightest part 

 of the spectrum lies between lines D and E, at X=0.59/i;* the 

 warmest part is, in the normal spectrum, near X=0.60^,, but 

 in the prismatic spectrum, beyond the visible red, at about 

 X = 1.00/i. Finally, the chemical activity of the rays increases 

 towards the blue end of the spectrum, but the relative activity 

 is different for the different substances acted upon. Measured 

 by their ability to unite chlorine and hydrogen, the rays having 



* MENGARINI ('89, p. 135) finds the point of maximum brightness to lie at 

 about 0.57 /u. 



