Now, since the appearances at twilight depend on the condition of the sky, it follows that our weather-wisdom, drawn from such appearances, is based upon a philosophical foundation. When there is a blue sky, and after sunset a slight purple in the west, we have reason for expecting fine weather. After rain, detached clouds, colored red and tolerably bright, may rejoice those who anticipate a pic-nic party. If the twilight show a partiality for whitish yellow in its dress, we say that very likely there will be some rain next day; the more that whitish yellow spreads over the sky, the more the chance of water out of it. When the sun is brilliantly white, and sets in a white light, we think of storms; especially so when light high clouds that dull the whole sky become deeper near the horizon. When the color of the twilight is a grayish red, with portions of deep red passing into gray that hide the sun, then be prepared, we say, for wind and rain. The morning signs are different. When it is very red, we expect rain; a gray dawn means fine weather. The difference between a gray dawn and a gray twilight is this--in the morning, grayness depends usually upon low clouds, which melt before the rising sun; but in the evening grayness is caused by high clouds, which continue to grow denser through the night. But if in the morning there be so much vapor as to make a red dawn, it is most probable that thick clouds will be formed out of it in the course of the operations of the coming day.
Refraction of light has a good deal to do also with the twinkling of the stars; though there may go to the explanation of the phenomenon other principles which do not concern our present purpose. The air contains layers of different density, shifting over each other in currents. The fixed stars are, to our eyes, brilliant points of light; their rays broken in passing through these currents, exhibit an agitation which is not shown by the planets. The planets are not points to our sight, nor points to our telescopes; being much nearer, although really smaller, they are to our eyes of a decided, measurable size; so being in greater body, we at most could only see their edges scintillate; and this we can do sometimes through a telescope, but scarcely with the naked eye.