262 FRAGMENTS OF SCIENCE. 



system of splendidly-colored rings. Precisely the same 

 phenomena are observed when we look at the blue firma- 

 ment in a direction perpendicular to the solar rays. 



We have thus far illuminated our artificial sky with 

 ordinary light. We will now examine the effects produced 

 when the light which illuminates the particles is itself 

 polarized. In front of the electric lamp, and between it 

 and the experimental tube, is placed this fine Nicol's prism, 

 which is sufficiently large to embrace and to polarize the 

 entire beam. The plane of vibration of the light now 

 emergent from the prism, and falling upon the cloud, is 

 vertical ; and we find that this formless aggregate of infini- 

 tesimal particles, without definite structure, is absolutely 

 incompetent to scatter the light upward or downward, 

 while it freely discharges the light horizontally, right and 

 left. I turn the polarizing Nicol so as to render the plane 

 of vibration horizontal ; the cloud now freely scatters the 

 light vertically upward and downward, but it is absolutely 

 incompetent to shed a ray horizontally to the right or left. 



Suppose the atmosphere of our planet to be surrounded 

 by an envelope impervious to light, with an aperture on 

 the sunward side, through which a solar beam could enter. 

 Surrounded on all sides by air not directly illuminated, the 

 track of the sunlight would resemble that of the electric 

 beam in a dark space filled with our incipient cloud. The 

 course of the sunbeam would be blue, and it would dis- 

 charge laterally, in all directions round it, light in precisely 

 the same polarized condition as that discharged from the 

 incipient cloud. In fact, the azure revealed by the sunbeam 

 would be the azure of such a cloud. And if, instead of 

 permitting the ordinary light of the sun to enter the aper- 

 ture, a Nicol's prism were placed there, which should 

 polarize the sunlight on its entrance into our atmosphere, 

 the particles producing the color of the sky would act 

 precisely like those of our incipient cloud. In two directions 



