them to glow with the brilliancy which renders them visible. It is now almost twenty years since a marvelous accession to our knowledge of such objects as the Great Nebula in Orion was made by Dr. Huggins. I have used our gas hydrogen as an illustration in describing the character of the nebula, but I have now to add that the presence of hydrogen is no mere fiction but a substantial verity. Truly we here open up one of the most marvelous chapters which science has to disclose. The chemist can analyze the different substances on the earth with his test tubes, and he can tell the elements of which they are composed. But in this old-fashioned chemistry it was at least reasonable for the chemist to demand a portion of the substance he was expected to analyze. Unless he were provided with a sample, how could it be possible for him to grind it up or submit it to the various operations of his laboratory? In these modern days the chemist can perform operations of which his predecessors never even dreamed. No doubt the old method is still used—nay, is indeed at this moment cultivated with greater skill and means than in any previous age—but side by side with the old method, and as an invaluable supplement thereto, the new method of chemical research, called spectrum analysis, has been created, and has already conducted to many profoundly interesting discoveries in the most varied branches of science. In the application of the spectroscopic method it is not indispensably necessary that we actually have a fragment of the substance; all we require is a beam of light which that substance can be made to yield when heated to a sufficiently high temperature. When a beam of the nebular light is transmitted through the prisms, it declares at once that the object from which that light has come is totally different from a star like the sun. Instead of the beautifully colored band, decked in all the glowing hues of the rainbow, the nebular beam is seen to be composed simply of six or seven widely separated strips. It is important to test the character of the light in these strips. Fortunately this can be done in a way that is completely satisfactory. We can produce artificial lights from known sources, and observe them through the spectroscope simultaneously with the light of the nebula.