104 FRAGMENTS OF SCIENCE. 



air and subjected to the action of a solar or an electric beam, 

 are thus decomposed, the products of decomposition hang- 

 ing as liquid or solid particles in the beam which generates 

 them. And here I must appeal to the inner vision already 

 spoken of. Remembering the different sizes of the waves 

 of light, it is not difficult to see that our minute particles 

 are larger with respect to some waves than to others. In 

 the case of water, for example, a pebble will intercept and 

 reflect a larger fractional part of a ripple than of a larger 

 wave. We have now to imagine light-undulations of dif- 

 ferent dimensions, but all exceedingly minute, passing- 

 through air laden with extremely small particles. It is 

 plain that such particles, though scattering portions of all 

 the waves, will exert their most conspicuous action upon 

 the smallest ones; and that the color-sensation answering 

 to the smallest waves in other words, the color blue 

 will be predominant in the scattered light. This harmon- 

 izes perfectly with what we observe in the firmament. The 

 sky is blue, but the blue is not pure. On looking at the 

 sky through a spectroscope, we observe all the colors of the 

 spectrum; blue is merely the predominant color. By means 

 of our artificial skies we can take, as it were, the firmament 

 in our hands and examine it at our leisure. Like the nat- 

 ural sky, the artificial one shows all the colors of the spec- 

 trum, but blue in excess. Mixing very small quantities of 

 vapor with air, and bringing the decomposingluminous beam 

 into action, we produce particles too small to shed any sen- 

 sible light, but which may, and doubtless do, exert an 

 action on the ultra-violet waves of the spectrum. We can 

 watch these particles, or rather the space they occupy, till 

 they grow to a size able to yield the firrnamental azure. 

 As the particles grow larger under the continued action of 

 the light, the azure becomes less deep; while later on a 

 milkiness, such as we often observe in nature, takes the 

 place of the purer blue. Finally the particles become 

 large enough to reflect all the light-waves, and then the 

 suspended " actinic cloud" diffuses white light. 



It must occur to the reader that even in the absence of 

 definite clouds there are considerable variations in the hue 

 of the firmament. Everybody knows, moreover, that as 

 the sky bends toward the horizon, the purer blue is im- 

 paired. To measure the intensity of the color De Saussure 

 invented a cyanometer, and Humboldt has given us a 



