We see the same idea also in the rising and setting sun and moon. On a clear, cloudless night, when nothing seems to interfere with the brightness of the stars, we cannot, by looking upward, perceive any moisture present in the atmosphere; but if we cast our eyes to the horizon, whereby we see through the mass of atmosphere endwise, as it were, and note the appearance of the stars there, or the rising or setting moon, we will see that the atmosphere there gives a redness to the rising body, which it does not have when it has ascended to mid-heaven. On a clear night, which is caused by the presence of the area of high barometer, the moon when in mid-heaven is of a clear, silver-white, and it is the same moon that at the horizon was a deep red. The color of the moon has not changed; it is simply the medium through which it is seen that produces the difference in color. Occasionally, on a clear, bright ("high") night, when the moon is full, prior to rising, when just below the horizon, it will so illuminate this lower strata of atmosphere as to appear like a great fire; the moon rises red, but its deep color gradually fades as it rises, and when well up in the heavens we perceive that this deep coloring was an illusion and merely the influence of its surroundings. I never, though, knew of any one to attempt to account for this by "meteoric dust;" and yet it is an embodiment of the same principle. Place the sun where the moon is, and from its far superior abundance of light we have a much grander display. Under no other conditions or relations of the sun and earth is it possible to have this phenomenon of the delicate red sky but when a positive area of high barometer is passing and extends over us. In order to produce this effect we must have the clear atmosphere of high barometer, when there is a minimum of moisture present. The action of the sun's rays upon this extensive area of slightly moist rarefied air is unconfined by clouds, and reaches far and wide, and produces a delicacy of color which from no other source or condition can be realized. ISAAC P. NOYES. Washington, D. C., 1884. * * * * * A THEORY OF COMETARY PHENOMENA. _To the, Editor of the Scientific American_: The following subject, substantially, was written more than a year ago with a view to its publication. It was not, however, until January of the present year that I sent a brief communication to the _Brooklyn Eagle_, which was published Feb. 3, giving my views in relation to cometary phenomena. With this I might remain satisfied, were it not that the interesting paper by G. D. Hiscox, published in the SCIENTIFIC AMERICAN SUPPLEMENT, Feb. 16, impressed me with the idea that the theory I advanced might assist in explaining others, if brought to the notice of those interested through the columns of your valuable journal.