of a few pounds in weight, which strike the earth, we have become quite familiar, while spectrum analysis has carried us a long way toward the conclusion that there is a unity in the material composition of the universe. But, as every thing is in motion, and all the celestial masses are revolving and whirling at great rates of speed, and, as there is much evidence of fractures, collisions, and transformations, it seems impossible that there should not have been going on constant abrasions and comminutions, with the production of all grades of dust down to the most impalpable, as a consequence of the general wear and tear. Such a notion would, of course, have been inadmissible in old times, when people reasoned about the universe from their a priori notions of what it ought to be; and, holding that the heavens are the type of all perfection, they would have been shocked at the notion that the gearing of the spheres could not work without the production of dirt and dust. But we have survived those prejudices, and now search for the celestial débris and waste with just as much interest as we do for more imposing and dignified objects.
It is but a few years since the world was surprised and delighted by that brilliant series of researches made by Prof. Tyndall, on the formation of artificial clouds and artificial sky by the electric light, in his vacuum-tubes. Spaces, washed clean, and apparently pure, were found to be filthy with dust, and, beginning with the thinnest transparent vapors, he was able to develop a succession of the most exquisite cloud-forms definitely related to the colors of the spectrum as the molecules grew in complexity under the chemical transformation. The firmamental blue he found to be caused by impurities in the air, so attenuated as to react only with the finest waves of the ethereal medium. How far down in the scale of minuteness beyond all previous conception the particles are, which reflect the azure light of the sky, may be gathered from the following memorable passage, taken from Dr. Tyndall's "Fragments of Science" (page 148):
Whatever may be the validity of these quantitative speculations, all the lines of investigation seem to converge