Sirius, as will be presently seen, belongs to the division of stars whose spectrum indicates a very high temperature, and in this case, as in what follows, we may remark (to use in part Mr. Lockyer’s words) that one of the most important distinctions between the stars in the heavens is one not depending upon their mass or upon anything of that kind, but upon conditions which make their spectra differ, just in the way that in our laboratories the spectrum of one and the same body will differ at different temperatures. What these absolutely are in the case of the stars, we may not know; but placing them in their most probable relative order, we have taken as an instance of the second class, or lower-temperature stage, our own sun. The impossibility of giving a just notion of its real complexity may be understood, when we state that in the recent magnificent photographs by Professor Rowland, a part alone of this spectrum occupies something like fifty times the space here given to the whole, so that, crowded with lines as this appears, scarcely one in fifty of those actually visible can be given in it. Without trying to understand all these now, let us notice only the identity of two or three of its principal elements with those found in other stars, as shown by the corresponding identity of some leading lines. Thus, C and F (with others) are known to be caused by hydrogen; D, by sodium; _b_, by magnesium; while fainter lines are given by iron and by other substances. These elements can be traced by their lines in most of the different star-spectra on this plate, and all those named are constituents of our own frames. The hydrogen lines are not quite accurately shown in the plate from which our engraving is made, those in Sirius, for instance, being really wider by comparison than they are here given; and we may observe in this connection, that by the particular appearance such lines wear in the spectrum itself we can obtain some notion of the _mass_ of a star, as well as of its chemical constitution. We can compare the essential characteristics of such bodies, then, without reference to their apparent size, or as though they were all equally remote; and it is a striking thought, that when we thus rise to an impartial contemplation of the whole stellar universe, our sun, whose least ray makes the whole host of stars disappear, is found to be not only itself a star, but by comparison a small one,--one at least which is more probably below than above the average individual of its class, while some, such as Sirius, are not impossibly hundreds of times its size.