UPON a perusal of the chapter in Dr. Schellen's “Spectrum Analysis,” specially bearing on the above subjects, I have been led to think—firstly, that our present knowledge of these spectra is far from complete; and secondly, that so far as such knowledge extends, it hardly warrants some of the conclusions arrived at in Dr. Schellen's work. To test the question of the aurora, I have collected, chiefly from the pages of NATURE, a set of observations (excluding a few which gave only rough results), and have arranged them under the heads of the several lines, so that these and their characteristics may be seen at a glance, and the observations compared; and from these observations I deduce the following remarks:—